Handbook of Exploration and Environmental Geochemistry
VOLUME 8 Life Cycle of the Phosphoria Formation: From Deposition to the Post-Mining Environment
Handbook of Exploration and Environmental Geochemistry
VOLUME 8 Life Cycle of the Phosphoria Formation: From Deposition to the Post-Mining Environment
H A N D B O O K OF E X P L O R A T I O N A N D E N V I R O N M E N T A L GEOCHEMISTRY M. HALE (Editor) 1. 2. 3. 4. 5. 6. 7. 8.
ANALYTICAL METHODS IN GEOCHEMICAL PROSPECTING STASTISTICS AND DATA ANALYSIS IN GEOCHEMICAL PROSPECTING ROCK GEOCHEMISTRY IN MINERAL EXPLORATION REGOLITH EXPLORATION GEOCHEMISTRY IN TROPICAL AND SUB-TROPICAL TERRAINS REGOLITH EXPLORATION GEOCHEMISTRY IN ARCTIC AND TEMPERATE TERRAINS DRAINAGE GEOCHEMISTRY GEOCHEMICAL REMOTE SENSING OF THE SUB-SURFACE LIFE CYCLE OF THE PHOSPHORIA FORMATION: FROM DEPOSITION TO THE POST-MINING ENVIRONMENT
Handbook of Exploration and Environmental Geochemistry
VOLUME 8 Life Cycle of the Phosphoria Formation: From Deposition to the Post-Mining Environment
Edited by JAMES R. HEIN
US Geological Survey Menlo Park, CA, USA
2004 ELSEVIER Amsterdam 9B o s t o n 9H e i d e l b e r g Paris 9S a n D i e g o 9S a n F r a n c i s c o
9L o n d o n 9N e w Y o r k 9S i n g a p o r e 9S y d n e y
9O x f o r d 9T o k y o
E L S E V I E R B.V. S a r a B u r g e r h a r t s t r a a t 25 P.O. B o x 2 1 1 , 1000 A E A m s t e r d a m ,
92004 US Geological
The Netherlands
Survey. All rights reserved.
This work is protected under copyright by Elsevier Science, and the following terms and conditions apply to its use: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier's Science & Technology Rights Department in Oxford, UK: phone: (144) 1865 843830, fax: (144) 1865 853333, e-mail:
[email protected]. You may also complete your request online via the Elsevier Science homepage (http://www.elsevier.com), by selecting 'Customer Support' and then 'Obtaining Permissions'. In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: (11)(978)7508400, fax: (11)(978)7504744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London WlP 0LP, UK; phone: (144) 207 631 5555; fax: (144) 207 631 5500. Other countries may have a local reprographic rights agency for payments. Derivative Works Tables of contents may be reproduced for internal circulation, but permission of Elsevier Science is required for external resale or distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and translations. Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier's Science & Technology Rights Department, at the phone, fax and e-mail addresses noted above. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made.
First e d i t i o n 2004 L i b r a r y o f C o n g r e s s C a t a l o g i n g in P u b l i c a t i o n D a t a A c a t a l o g r e c o r d f r o m t h e L i b r a r y o f C o n g r e s s h a s b e e n a p p l i e d for. B r i t i s h L i b r a r y C a t a l o g u i n g in P u b l i c a t i o n D a t a A c a t a l o g u e r e c o r d f r o m t h e B r i t i s h L i b r a r y h a s b e e n a p p l i e d for.
ISBN: 0-444-51132-6
oo The p a p e r used in this p u b l i c a t i o n m e e t s t h e r e q u i r e m e n t s of A N S l / N I S O Z39.48-1992 ( P e r m a n e n c e of Paper). Printed in T h e N e t h e r l a n d s .
V
PREFACE
US Geological Survey (USGS) scientists have studied the Phosphoria Formation in the Western United States Phosphate Field throughout much of the twentieth century (see Chapter 2). In response to a request by the US Bureau of Land Management (BLM), a new series of geologic, geoenvironmental, and resource studies was initiated in 1997. This program followed three earlier USGS field programs that took place in 1909-1916, 1941-1944, and 1947-1952. Follow-up work to each of those programs continued for many years after conclusion of the main phase of field work. This latest program (1997-2002) consisted of integrated, multiagency, multidisciplinary research with emphasis in four areas: (a) geological and geochemical baseline characterization of the Meade Peak Phosphatic Shale Member and related rocks of the Permian Phosphoria Formation, headed by R.I. Grauch; (b) delineation, assessment, and spatial analysis of phosphate resources and lands disturbed by mining, headed by P.R. Moyle; (c) contaminant residence, reaction pathways, and environmental fate associated with the occurrence, development, and use of phosphate rock, headed by J.R. Herring; and (d) depositional origin and evolution of the Phosphoria Formation and geoenvironmental and deposit modeling, headed by G.J. Orris. The overriding objective of this latest research program was science in support of land management. To carry out these studies, the USGS formed cooperative research relationships with the BLM and the US Forest Service (USFS), which are responsible for land management and resource conservation on public lands, and with five private companies currently leasing or developing phosphate resources in Southeast Idaho. Four operating phosphate mines exist in Southeast Idaho (Dry Valley, Smoky Canyon, Rasmussen Ridge, and Enoch Valley mines) and one in northern Utah (Vernal Mine; see Chapter 3). In addition, 12 inactive mines exist in Southeast Idaho and leases have been or are in the process of being developed for several new mines. The Western Phosphate Field encompasses an area of about 350,000 sq. km in adjacent parts of Idaho, Utah, Montana, Wyoming, Nevada, and Colorado in the northern Rocky Mountains. The thick, high-grade phosphate deposits in the Meade Peak Member of the Phosphoria Formation constitute an important economic resource providing about 12-14% of total United States production (see Chapter 3). The remaining phosphates in the Western Field constitute about 3% of the world reserves and 30% of United States reserves. Phosphorus is an essential nutrient for life and phosphate is essential for the production of many commodities used by modem societies. The principal use of phosphate is as fertilizer. However, products derived from phosphate are also used in other industrial applications,
Preface
VI
such as fire retardants, detergents and other cleaning supplies, pharmaceuticals, food and beverages, feed, herbicides, and water softeners (see Chapter 22). Many large sedimentary phosphate deposits are hosted by black shale. Black shales are well known to host a wide variety of trace elements, some of economic interest and some of environmental concern. Many of those trace elements may be environmentally sensitive and can be released during mining and weathering of waste rock. In the Western Phosphate Field, phosphate is mined from two high-grade zones in the Meade Peak Member. The ore zones enclose a middle waste zone about 25-30 m thick composed predominantly of low-grade phosphatic black shale. This waste rock is placed in waste piles along with unmineralized rock that is removed to expose phosphate-bearing strata rich enough to be mined. Leaching of the waste rock has released several potentially toxic elements into the environment. Selenium has been the most detrimental element released in terms of its affect on livestock and wildlife. Selenium is a nutrient in low concentrations and a toxicant in only slightly higher concentrations. An important part of our 1997-2002 study, and nine of the 22 chapters (11-19) in this book, concern the distributions of selenium in rocks, soils, plants, animals, and water and its effects on the environment. Other elements of potential environmental concern in the Western Phosphate Field include chromium, copper, molybdenum, vanadium, uranium, and zinc. In fact, the concentrations of vanadium and uranium are high enough that vanadium was recovered as a byproduct of phosphate mining from the early 1940s until 1999; geochemical exploration for uranium occurred during the 1947-1952 USGS field program. Our five-year program has tied together geology, geochemistry, water resources, and biology into an integrative approach to understand an important United States phosphate resource and the consequences of recovery of that phosphate. This type of approach and the knowledge gained are essential prerequisites required for the mining of ores that are essential to the functioning of modern society in an environmentally sound way.
ACKNOWLEDGEMENTS I would like to acknowledge and thank Brandie Mclntyre for invaluable help in processing and standardizing all the manuscripts that compose this 22-chapter book. We would like to thank the mining companies for access to the mines for sampling, especially Nu-West, Rasmussen ridge Mine, J.R. Simplot Co., Smoky Canyon Mine, and P4 Production LLC, Enoch Valley Mine, as well as Rhodia Inc. and Astaris Production LLC. The following scientists provided excellent and timely reviews of one or more chapters in this book: 9 9 9 9 9
Michael C. Amacher, USDA, Forest Service, Logan, UT; John A. Barron, USGS, Menlo Park, CA; Paul Belasky, Ohlone College, Fremont, CA; John M. Besser, USGS, Columbia, MO; Arthur A. Bookstrom, USGS, Spokane, WA;
Preface 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9
VII
Steven E. Box, USGS, Spokane, WA; George N. Breit, USGS, Denver, CO; George A. Desborough, USGS, Denver, CO; Robert E. Garrison, University of California, Santa Cruz, CA; Richard I. Grauch, USGS, Denver, CO; Philip L. Hageman, USGS, Denver, CO; Steven J. Hamilton, USGS, Yankton, SD; Peter W. Harben, Industrial Minerals Consultants, Las Cruces, NM; James R. Hein, USGS, Menlo Park, CA; Stephen M. Jasinski, USGS, Reston, VA; Margaret A. Keller, USGS, Menlo Park, CA; Lisa B. Kirk, Maxim Technologies Inc., Bozeman, MT; Andrea Koschinsky, International University Bremen, Germany; Randolph A. Koski, USGS, Menlo Park, CA; Joel S. Leventhal, USGS, Denver, CO; Dean A. Martens, USDA, Tucson, AZ; Greg M611er, University of Idaho, Moscow, ID; Philip R. Moyle, USGS, Spokane, WA; Joyce A. Ober, USGS, Reston, VA; Peter Oberlindacher, USBLM, Boise, ID; John C. Mars, USGS, Reston, VA; Robert B. Perkins, Portland State University, Portland, OR; David Z. Piper, USGS, Menlo Park, CA; Theresa S. Presser, USGS, Menlo Park, CA; Robert Rosenbauer, USGS, Menlo Park, CA; Richard E Sanzolone, USGS, Denver, CO; Calvin H. Stevens, California State University, San Jose, CA; Peter W. Swarzenski, USGS, St. Petersburg, FL; Marc A. Sylvester, USGS, Menlo Park, CA; George E Vance, University of Wyoming, Laramie, WY; Florence L. Wong, USGS, Menlo Park, CA; Robert A. Zielinski, USGS, Denver, CO. JAMES R. HEIN, Menlo Park, CA February 2003
VIII
LIST OF CONTRIBUTORS
Michael C. Amacher received a BSc in chemistry, a MSc in chemistry, and a PhD in soil
chemistry from Pennsylvania State University. He joined the USDA Forest Service in 1989. He works on soil analysis methods, trace element biogeochemistry in soils, reactivity, and transport of metals in soils, and the status and trend of forest soil quality indicators (USDA, Forest Service, Logan, UT). Kenneth J Bird joined the US Geological Survey in 1974. His research focused initially on
the petroleum potential of carbonate rocks in northern Alaska. With broadening interests, primarily in stratigraphy and sedimentology, he became extensively involved in research and petroleum resource assessment activities in Alaska and elsewhere in the United States. Currently he is the leader of a relatively large, multidisciplinary team conducting petroleum geologic research and new assessments of undiscovered oil and gas resources in Alaska (USGS, Menlo Park, CA). James R. Budahn received a BSc in chemistry from Southwest State University, Marshall,
Minnesota and a MSc in chemistry from Oregon State University, Corvallis, Oregon. He joined the US Geological Survey in 1979. From 1979, he has worked on INAA and gamma-ray spectrometry methodologies and applications (USGS, Denver, CO). Kevin J. Buhl received a BA in biology from St. Mary's College, Winona, Minnesota and
an MA in biology from the University of South Dakota, Vermillion. He joined the US Fish and Wildlife Service in 1979. He has conducted laboratory and field studies of environmental contaminant problems in a variety of aquatic ecosystem types. He has published 41 scientific papers and reports (USGS, Yankton, SD). Steven J. Detwiler did his doctoral research at the University of California Davis on sele-
nium toxicokinetics in avian eggs. His work focused upon detectable inter-specific differences to help elucidate Se ecotoxicology, and potential underlying toxicodynamic mechanisms. Steven worked as part of the Interagency San Joaquin Valley Drainage Program spawned by the discoveries at Kesterson in the 1980s, and currently works for the US Fish and Wildlife Service conducting primarily Se-related biomonitoring and risk assessment (USFWS, Sacramento, CA).
List of contributors
IX
James G. Evans received a BSc in geology in 1960 from the University of Massachusetts at Amherst and a PhD (structural geology and igneous and metamorphic petrology) in 1966 from the University of California (Los Angeles). His dissertation comprised the first structural analysis of a segment of the San Andreas fault zone. Since 1967, he has worked for the US Geological Survey on projects with a mineral resources focus. An early assignment included detailed mapping of the Lynn Window, site of the Carlin gold mine. Data gathered during that study was the basis of the first structural analysis of a segment of the Roberts Mountains thrust. Most of his work concentrated on resource assessments and commodity studies (especially Au, Cr, P). Data from these studies resulted in structural analyses of the widespread Josephine Peridotite in NW California and SW Oregon, Belt-age rocks in northeast Washington, and most recently of the Meade thrust plate in Southeast Idaho. He presently works on garnet deposits (abrasives) in northern Idaho and Great Basin studies in Nevada and southern Oregon (USGS, 904 W. Riverside Ave., Rm 202, Spokane, WA, 99201;
[email protected]). Carlotta B. Chernoff received a BSc in geophysics and MA in geology from The University of Texas at Austin, and a PhD in geology from The University of Arizona. She studies the processes controlling the incorporation of metals into sedimentary rocks and the mechanisms of chemical exchange and mass transfer that occur during diagenesis and metamorphism of those rocks. Since completing her PhD in 2002, she has been working as an exploration geologist for Conoco Phillips in Houston, Texas (ConocoPhillps, Houston, TX). Andrea L. Foster received a BSc in geology from Indiana University (Bloomington) and a PhD in geochemistry from Stanford University (Stanford, CA). She joined the US Geological Survey in 1999. Her work has focused primarily on the application of synchrotron-based spectroscopic techniques to identify the forms of potentially toxic metals such as As, Cd, Cr, Se, and Hg in solid materials derived from recent and historical mining activities (USGS, Menlo Park, CA). Richard I. Grauch received an AB from Franklin and Marshall College and a PhD from the University of Pennsylvania. He was a postdoctoral researcher at SUNY, Binghamton, and a postdoctoral scholar at the University of California (Los Angeles) before consulting for the Ministerio de Minas e Hidrocarburos in Venezuela. He joined the US Geological Survey in 1974 where he has worn several different administrative hats, but prefers research which has focused on the genesis of a variety of ore deposits, mostly unconventional types. Recent work focuses on detailed petrologic and geochemical studies that include (a) natural and anthropogenic sources of environmental contaminants and (b) experimental work on element partitioning in the system shale-brine-vapor under diagenetic P - T conditions (USGS, DFC, MS 973, Box 25046, Denver, CO, 80225-0046;
[email protected]). Mickey E. Gunter received a BSc (geology and mathematics) from Southern Illinois University, Carbondale, and a MSc and PhD (geological sciences) from Virginia Tech,
X
List of contributors
Blacksburg, Virginia. He joined the Department of Geological Sciences, University of Idaho in 1989 and is currently a Professor of Mineralogy. Gunter's research involves crystal chemical and structural characterization of natural and cation-exchange zeolites, phosphates, and new minerals. He is also interested in studying the directional dependence of the physical properties of minerals and health affects of minerals, both positive and negative. He continues to pursue research and development of new methods in optical mineralogy (University of Idaho, Moscow, ID). Jeffery O. Hall received a BSc in agriculture economics and a DVM from Oklahoma State
University, and a PhD in toxicology from the University of Illinois and is a Diplomat American Board of Veterinary Toxicology. He joined the Utah Veterinary Diagnostic Laboratory/Utah State University in 1996. He works in diagnostic toxicology, with mineral toxicoses and interactions, and with poisonous plants (Utah State University, Logan, UT). Steven J. Hamilton received a BSc (Wildlife Management) from Humboldt State
University, Arcata, California, and MSc and PhD (Fisheries and Wildlife) from the University of Missouri, Columbia. He joined the US Fish and Wildlife Service in 1975. He works on various aspects of toxicology in aquatic ecosystems and since 1984 has focused primarily on selenium toxicology. He has published 102 scientific papers, reports, and book chapters (USGS, 31247 436th Ave., Yankton, SD, 57078-6364; steve_hamilton@ usgs.gov). Mark A. Hardy received a BSc (water science/limnology) from the University of
Wisconsin, Stevens Point, Wisconsin. He then joined the US Geological Survey in 1975. In addition to performing numerous assessments of water quality in streams, ground water, and wetlands across the United States, he has been extensively involved with the standardization of data-collection methods and the development of instrumentation for national water-quality programs. He currently evaluates data-quality needs, project designs, data-collection methods, and data interpretations for water-quality studies (USGS, Boise, ID). James R. Hein received a BSc (geology) from Oregon State University and a PhD (Earth
Sciences) from the University of California (Santa Cruz) before joining the US Geological Survey in 1974. Hein has spent much of his career studying marine mineral deposits in the modern ocean basins and analogs in the geologic record. Mineral deposit types studied include phosphorite, hydrogenetic ferromanganese crusts, diagenetic-hydrogenetic ferromanganese nodules, hydrothermal manganese deposits, barite, ironstones, and polymetallic sulfides. Hein is also applying geochemical and isotopic proxies in marine chemical sediments to the study of paleoceanography. Hein has edited or co-edited seven books, including this one (USGS, MS 999, 345 Middlefield Rd., Menlo Park, CA, 94025;
[email protected]). James R. Herring received a PhD in Earth Science from Scripps Institution of
Oceanography, University of California, San Diego. He joined the US Geological Survey
List of contributors
XI
in 1978 and has conducted a variety of geochemical and mineral deposit studies (USGS, DFC, MS 973, Box 25046, Denver, CO, 80225-0046;
[email protected]). Stephen M. Jasinsld received a BSc in geology from the University of Pittsburgh and worked for the US Bureau of Mines from 1987 until 1996 as a mineral commodity specialist in the nonferrous metals section. He joined the US Geological Survey in 1996 and has been the mineral commodity specialist for phosphate rock and peat since 1997 (USGS, 983 National Center, Reston, VA, 20192;
[email protected]). E.A. Johnson received a PhD from Rice University and is a sedimentologist-stratigrapher with the US Geological Survey. He spent most of his career conducting research in support of regional framework studies of energy resource-bearing sedimentary systems by constructing depositional models as a component of basin analysis. His research has been both domestic (Colorado, Wyoming, Nevada) and foreign (China, Pakistan, Kyrgyzstan, Armenia, China). Currently, he is Acting Associate Team Scientist for the Central Energy Team, and participates in national resource assessments of oil and gas and coal (USGS, Denver, CO). Andrew C. Knudsen received a BA in geology from Hamilton College and a PhD in geology from the University of Idaho. He has worked as a Post-Doctoral Research Fellow in the Environmental Soil Chemistry lab at the University Idaho. His research has focused on environmental mineralogy and geochemistry in mining-impacted sites. He is now a professor in the Geology Department, Lawrence University, Appleton, WI, 54911 (
[email protected]). Paul J. Lamothe received a BSc (chemistry) from the University of San Francisco, California and a PhD (chemistry), from Marquette University, Milwaukee, Wisconsin. He was a research chemist with the US Environmental Protection Agency prior to joining the US Geological Survey in 1976. He is editor of "The Geoanalyst" and he is a member of the council of the International Association of Geoanalysts. His research interests include atomic spectroscopy and trace-element geochemistry (USGS, DFC, MS 973, Box 25046, Denver, CO, 80225-0046;
[email protected]). William H. Lee received a BA (geology) from the University of Colorado and pursued graduate studies at the Colorado School of Mines before joining the US Geological Survey in 1961. From 1961 to 1972, he worked on the safety of underground nuclear testing in the Special Projects Branch. From 1972 to 1982, he worked in economic geology and mineral resource management in the western United States. From 1982 to 1983, he worked for the Minerals Management Service in economic mineral resources and from 1983 to present he has worked for the Bureau of Land Management in Wyoming, Washington DC, and Idaho as a senior minerals specialist (USBLM, Boise, ID). Cheryl L. Mackowiak received a BSc (plant and soil science) and MSc (plant and soil science) from Southern Illinois University. She was a research horticulturist at NASA's
XII
List of contributors
Advanced Life Support program, Kennedy Space Center, Florida, before completing a PhD (plant nutrition and soil fertility) at Utah State University. She subsequently joined the USDA Forest Service as a postdoctoral soil scientist in 2002. She works on trace element biogeochemistry, transport, and plant bioavailability (USDA, Forest Service, 860 N. 1200 E., Logan, UT, 84321-5700;
[email protected]). Brandie R. Mclntyre received a BSc (geology) from California State University, Fresno and joined the US Geological Survey in 2001. She is currently working on a graduate degree at San Jose State University and is doing research on marine mineral deposits and environmental geochemistry (USGS, Menlo Park, CA). Phillip R. Moyle received a BSc (geology) from the University of California at Davis and has worked in mineral resources, especially industrial minerals, for over 25 years. Most of his resource assessment and deposit investigations, including extensive underground work, were conducted from 1979 to 1996 while with the US Bureau of Mines. He also specialized in shallow geophysical methods while a member of the Bureau's mine waste site environmental characterization team and worked to develop real-time, multidisciplinary, siteinvestigation techniques. Since joining the US Geological Survey in 1997, he has studied phosphate, aggregate, diatomite, and garnet deposits in the western United States. He also teaches a mine safety course on "health risks associated with abandoned mine and mill sites" for the US Forest Service National Mineral Training Center (USGS, 904 W. Riverside Ave., Rm 202, Spokane, WA, 99201;
[email protected]). Benita L. Murchey received a BA in biology from Rice University and a PhD in geology from University of California at Santa Cruz. She joined the US Geological Survey in 1978. Her work has focused on the stratigraphic and paleogeographic distribution of siliceous microfossils in Paleozoic and Mesozoic marine basins of western North America. The depositional and tectonic histories of basins receiving biosiliceous sediments have been a particular interest as well (USGS, MS 973,345 Middlefield Rd., Menlo Park, CA, 94025;
[email protected]). Greta J. Orris received a BA and a MSc in geology and PhD in mineral economics. She has spent much of her career as a research geologist for the US Geological Survey. As an industrial minerals expert, much of her research has focused on the development of deposit models, numerical modeling of deposit distribution, economic modeling, and development of resource appraisal techniques. These skills have allowed her participation in a wide variety of domestic and international research efforts (USGS, 520 N. Park Ave., Tucson, AZ, 85719;
[email protected]). Robert B. PerMns received a BSc (geology) from Morehead State University (Morehead, Kentucky) and a MSc (geology) from Eastern Kentucky University (Richmond). He then became an environmental consulting geologist prior to completing a PhD (environmental
List of contributors
XIII
geology) at Portland State University, Oregon. The research presented in this volume was completed as part of a US Geological Survey Mendenhall Postdoctoral Fellowship. He is now a professor in the Department of Geology, Portland State University, P.O. Box 751, Portland, OR, 97207-0751 (
[email protected]).
David Z. Piper received a BSc from the University of Kentucky, a MSc from Syracuse University, and PhD from Scripps Institute of Oceanography before joining the US Geological Survey in 1975. His research focused initially on the geochemistry of ferromanganese deposits in the Pacific Ocean. He then moved on land to examine phosphaticenriched black shales of Paleozoic to Holocene age. Currently, he is involved in the examination of the geochemistry of soils and stream sediments in the conterminous States, while still maintaining an interest in unraveling the depositional environments of black shales (USGS, Menlo Park, CA). Theresa Presser is a chemist with the US Geological Survey. She became involved in selenium issues in 1983 during the investigation of environmental damage at Kesterson National Wildlife Refuge, California. Her biogeochemical model describing the kesterson effect and the kesterson, a unit of measure of hazard to wildlife, has contributed to the overall understanding of selenium's origins, exposures, and risk. She authored, among other articles, chapters in Selenium in the Environment and Environmental Chemistry of Selenium. She recently collaborated on a selenium model for the San Francisco Bay-Delta Estuary to help resolve issues of water management and ecological effects (USGS, MS 435, 345 Middlefield Rd., Menlo Park, CA, 94025;
[email protected]). Joseph P Skorupa has been conducting field research on the ecotoxicology of selenium since 1987 when he first joined the US Fish and Wildlife Service as an avian ecologist. His research has focused on California, but has also included research in Nevada, Wyoming, and southeast Idaho. From 1992-1998 he served as the US Fish and Wildlife Service's technical lead for the National Irrigation Water Quality Program's data synthesis project, a detailed examination of selenium ecological risk across the entire western United States. Currently, he is serving as the Clean Water Act biologist in one of the national offices (USFWS, Arlington, VA). Lisa L. Stillings received a BSc (geology) from Allegheny College, Pennsylvania, a MSc (hydrogeology) from Kent State University, Ohio, and a PhD (geochemistry) from Pennsylvania State University, University Park. She joined the US Geological Survey in 1998 and works on mineral weathering and metal cycling in mining-impacted environments (USGS, MS 176, University of Nevada, Reno, NV, 89557;
[email protected]). Robert A. Zielinski received a BSc (chemistry) from Rutgers University, New Brunswick, New Jersey and PhD (geochemistry) from MIT, Cambridge, Massachusetts, before joining the US Geological Survey as a postdoctoral researcher in 1972. He was appointed as a
XlV
List of contributors
research chemist in 1974. His work for the US Geological Survey utilizes chemical, isotopic, and instrumental measurements as well as laboratory-based selective extractions and process simulations. His primary area of interest is the study of natural processes that influence the mobility of trace elements and radionuclides of environmental and human health concern (USGS, DFC, MS 973, Box 25046, Denver, CO, 80225-0046; rzielinski@ usgs.gov).
XV
CONTENTS
Preface ........................................................................................................................................ V List o f Contributors ................................................................................................................ VIII PART I. I N T R O D U C T I O N
Chapter 1. The Permian Earth .................................................................................................... 3 J.R. Hein Introduction ................................................................................................................ 3 Geology, plate tectonics, paleogeography .................................................................. 3 Climate ...................................................................................................................... 6 Oceanography ............................................................................................................ 7 Western North American margin and the Phosphoria sea ...................................... 10 End o f Permian ........................................................................................................ 12
Chapter 2. Evolution of thought concerning the origin of the Phosphoria Formation, Western US Phosphate Field .................................................................................. 19 J.R. Hein, R.B. Perkins and B.R. Mclntyre Abstract .................................................................................................................... 19 Introduction .............................................................................................................. 20 Delineation o f western phosphate lands: Pre- 1940s ................................................ 21 Geochemical exploration, P, U, and V: The 1940s-1950s ...................................... 24 Depositional environments and transgressive-regressive cycles: The 1960s-1970s .................................................................................................... 27 Paleogeography, phosphogenesis, and sequence stratigraphy: 1980-2002 ............ 30 Paleogeography o f the Phosphoria basin ............................................................ 30 Phosphogenesis .................................................................................................... 32 Eustatic changes and sequence stratigraphy ........................................................ 34 Outstanding issues .................................................................................................. 36 PART II. R E G I O N A L S T U D I E S
Chapter 3. The history of production of the Western Phosphate Field .................................... 45 S.M. Jasinski, W.H. Lee and J.D. Causey Abstract .................................................................................................................... 45 Introduction .............................................................................................................. 45 Early scientific surveys ............................................................................................ 48
XVI
Contents Laws associated with US phosphate exploration and mining ................................ Idaho ........................................................................................................................ Montana .................................................................................................................. Utah .......................................................................................................................... Wyoming .................................................................................................................. Mining in the Western Phosphate Field in the twenty-first century ........................
49 52 55 55 58 59
Chapter 4. The Meade Peak Member of the Phosphoria Formation." Temporal and spatial variations in sediment geochemistry ..........................................................73 R.B. Perkins and D.Z. Piper Abstract .................................................................................................................... 73 Introduction .............................................................................................................. 74 Oceanographic setting of the Phosphoria basin .................................................. 74 Origin of Meade Peak sediments ........................................................................ 76 Methods and data evaluation .................................................................................. 77 Element associations ................................................................................................ 80 Temporal variations in terrigenous elements .......................................................... 80 Major-element oxides .......................................................................................... 83 Minor and trace elements .................................................................................... 89 Spatial variations in terrigenous elements .............................................................. 90 Temporal variations in marine elements .................................................................. 90 Spatial variations in marine elements ...................................................................... 99 Conclusions ............................................................................................................ 103
Chapter 5. Regional analysis of spiculite faunas in the Permian Phosphoria basin: Implication for paleoceanography ........................................................................ 111 B.L. Murchey Abstract .................................................................................................................. Introduction ............................................................................................................ Background and previous studies .......................................................................... Methods .................................................................................................................. Identification of sponge spicule morphotypes .................................................. Quantitative comparison of sponge spicules to radiolarians ............................ Results .................................................................................................................... Eastern belt: Rhax-bearing, demosponge-dominated spiculite assemblages ........................................................................................................ Rex Chert of the Phosphoria Formation in southeastern Idaho, central basin (Table 5-II; Sample 1 - 3 ) .......................................................... Black chert, northeastern Nevada, southwestern margin of Phosphoria Basin and inferred Antler high (Table 5-II; Sample 4-7) .............................. Edna Mountain Formation of Nevada, overlap sequence deposited on the Antler high (Table 5-II; Samples 8-12) ........................................................ Spiculitic black chert, Havallah assemblage, Nevada eastern basin margin facies (Table 5-II; Samples 1 3 - 2 4 ) .................................................. Central belt mixed choristid demosponge-hexactinellid sponge assemblages associated with (ruzhencevispongacid) radiolarians ..........................................
111 112 114 117 117 118 119 119 119 122 122 123 124
Contents
XVII Western belt: Radiolarian-dominated assemblages with or without a minor component o f hexactinellid sponge spicules .................................................... N o r t h e m Basins ................................................................................................ N o r t h w i n d Ridges, Chukchi Sea ...................................................................... A n g a y u c h a m terrane and Northern Brooks Range ................................................ Discussion and conclusions ..................................................................................
125
125 125 126 126
Chapter 6. Strain distribution and structural evolution of the Meade plate, southeastern Idaho ................................................................................................ 137 J.G. Evans Abstract .................................................................................................................. Introduction ............................................................................................................ Depth o f burial ...................................................................................................... Thermal history ...................................................................................................... Petroleum generation and very low-grade m e t a m o r p h i s m ................................ Low-grade m e t a m o r p h i s m ................................................................................ Structure ................................................................................................................ Pretectonic structures ........................................................................................ Syntectonic structures ........................................................................................ Orogenic and structural terminology ................................................................ Timing o f thrusting in southeastern Idaho ........................................................ Shortening o f the Meade and other plates ........................................................ Direction o f tectonic transport .......................................................................... Thickness o f the Meade thrust plate and topology o f the Meade thrust .......... Style o f deformation o f the Meade thrust and plate .......................................... Shortening and extension implied by folding and faulting in the Meade plate Estimates o f shortening and compression directions from other data .............. Variable displacement ........................................................................................ Conclusions ............................................................................................................
137 137 140 141 141 142 143 143 144 144 145
146 147 147 151 152
157 158 161
PART III: G E O L O G I C A L A N D G E O C H E M I C A L STUDIES IN S O U T H E A S T I D A H O
Chapter 7. The effects of weathering on the mineralogy of the Phosphoria Formation, southeast Idaho .................................................................................................... 169 A.C. Knudsen and M.E. Gunter Abstract .................................................................................................................. Introduction ............................................................................................................ Carbonate fluorapatite ............................................................................................ Methods .................................................................................................................. Sampling and sample preparation ...................................................................... X R D analysis and Rietveld refinement ............................................................ CO 2- substitution in CFA .................................................................................. Statistical analyses ............................................................................................ Results .................................................................................................................... Bulk mineralogy ................................................................................................
169 169 171 172 172 172 173 173 174 174
Contents
XVIII
N o n d i f f r a c t i n g c o m p o n e n t ................................................................................ C a r b o n a t e substitution in fluorapatite ................................................................ D i s c u s s i o n .............................................................................................................. W e a t h e r i n g ........................................................................................................ C a r b o n a t e substitution in C F A ..........................................................................
177 180 180 180 182
C o n c l u s i o n s ............................................................................................................ 185
Chapter 8. Petrogenesis and mineralogic residence of selected elements in the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation, southeast Idaho .................................................................................. 189 R.I. G r a u c h , G.A. D e s b o r o u g h , G.P. M e e k e r , A.L. Foster, R.G. Tysdal, J.R. H e r r i n g , H.A. L o w e r s , B.A. Ball, R.A. Z i e l i n s k i and E.A. J o h n s o n A b s t r a c t .................................................................................................................. I n t r o d u c t i o n ............................................................................................................ G e o l o g i c setting and h i s t o r y .................................................................................. A p p r o a c h and m e t h o d o l o g y .................................................................................. R e s u l t s and d i s c u s s i o n .......................................................................................... Detrital a s s e m b l a g e ............................................................................................ A u t h i g e n i c / d i a g e n e t i c a s s e m b l a g e ....................................................................
189 190 192 193 197 197 201
P h o s p h a t e ...................................................................................................... 201 Silicates and c a r b o n a t e s ................................................................................ 2 0 4 S u l f i d e s .......................................................................................................... O t h e r m i n e r a l s ................................................................................................ E p i g e n e t i c / s u p e r g e n e a s s e m b l a g e ...................................................................... Silicates .......................................................................................................... H a l i d e s .......................................................................................................... O t h e r M i n e r a l s ..............................................................................................
206 212 212 212 213 213
S e l e n i d e (CuxSey) .......................................................................................... 213 N a t i v e e l e m e n t s .............................................................................................. 213 O x i d e s ............................................................................................................ 217 P h o s p h a t e s ...................................................................................................... 217 Sulfates .......................................................................................................... 217 S u l f i d e s .......................................................................................................... 218 C o n c e p t u a l m o d e l .................................................................................................. 218
Chapter 9. Weathering of the Meade Peak Phosphatic Shale Member, Phosphoria Formation: Observations based on uranium and its decay products .................. 227 R.A. Zielinski, J.R. B u d a h n , R.I. G r a u c h , J.B. P ac e s and K.R. S i m m o n s A b s t r a c t .................................................................................................................. 227 I n t r o d u c t i o n ............................................................................................................ S a m p l e c o l l e c t i o n and d e s c r i p t i o n ........................................................................ A n a l y t i c a l m e t h o d s ................................................................................................ G a m m a - r a y s p e c t r o m e t r y .................................................................................. F i s s i o n - t r a c k r a d i o g r a p h y .................................................................................. Selective extraction e x p e r i m e n t s ........................................................................ M a s s s p e c t r o m e t r y ............................................................................................
227 228 230 230 232 232 232
O t h e r a n a l y s e s .................................................................................................... 233
Contents
XIX Results and Discussion .......................................................................................... 233 A m o u n t o f extractable uranium and comparison to other elements .................. 233 Microdistribution o f uranium in phosphorite .................................................... 234 Disequilibria in the 238U decay-series determined by g a m m a - r a y spectrometry ...................................................................................................... 240 Disequilibria in the 238Udecay-series determined by mass spectrometry ........ 243 Conclusions ............................................................................................................ 246
Chapter 10. Mineral affinities and distribution of selenium and other trace elements in black shale and phosphorite o f the Phosphoria Formation .............................. 251 R.B. Perkins and A.L. Foster Abstract ................................................................................................................ Introduction .......................................................................................................... Methods ................................................................................................................ Sample selection and preparation .................................................................... Solid characterization o f selected samples ...................................................... Sequential extractions ...................................................................................... Sequential-extraction techniques ................................................................ Analyses o f extracts .................................................................................... Initial and residual solids characterization .................................................. Results .................................................................................................................. Solid characterization ...................................................................................... Scanning electron microscopy .................................................................... Quantitative analyses using the electron microprobe .................................. X-ray absorption spectroscopy .................................................................... Sequential extractions ...................................................................................... Reference materials ...................................................................................... Samples ........................................................................................................ Discussion ............................................................................................................ Conclusions ..........................................................................................................
251 252 252 252 256 257 257 259 260 260 260 260 264 268 270 270 273 287 291
P A R T IV. G E O E N V I R O N M E N T A L S T U D I E S
Chapter 11. The Phosphoria Formation: A Model for forecasting global selenium sources to the environment ................................................................................299 T.S. Presser, D.Z. Piper, K.J. Bird, J.P. Skorupa, S.J. Hamilton, S.J. Detwiler and M.A. H u e b n e r Abstract ................................................................................................................ 299 Introduction .......................................................................................................... 300 Methods and sources o f data .............................................................................. 300 Selenium guidelines ........................................................................................ 300 Western US (Colorado River watersheds and San Joaquin Valley and San Francisco Bay-Delta Estuary, California) ................................................ 301 Idaho ................................................................................................................ 301 Global distribution o f phosphate deposits and petroleum basins .................... 302 Conceptual model ................................................................................................ 302
XX
Contents S e l e n i u m b i o c h e m i s t r y a n d g u i d e l i n e s ............................................................ 303 S e l e n i u m o c e a n c h e m i s t r y .............................................................................. 305 F i e ld c a s e - s t u d i e s a n d e n v i r o n m e n t a l s e l e n i u m c o n c e n t r a t i o n s ...................... 305 I d a h o c a s e - s t u d y .................................................................................................. 308 P h o s p h a t e p r o d u c t i o n a n d shale e x p o s u r e s .................................................... 308 G e o c h e m i c a l m e c h a n i s m o f d i s p e r s a l a n d s e l e n i u m d i s c h a r g e s .................... 308 B i o l o g i c a l r e a c t i o n s a n d s e l e n i u m c o n c e n t r a t i o n s in b i o t a ............................ 310 Plants, invertebrates, a n d fish ...................................................................... 310 B i r d s a n d m a m m a l s .................................................................................... 311 G l o b a l o c c u r r e n c e o f p h o s p h o r i t e s a n d p e t r o l e u m .............................................. 313 P r e d i c t i o n o f s e l e n i u m s o u r c e s ........................................................................ 313 C o m m o d i t i e s a n d e x p l o r a t i o n .......................................................................... 315 C o n c l u s i o n s .......................................................................................................... 315
Chapter 12. Lithogeochemistry of the Meade Peak Phosphatic Shale Member of the Phosphoria Formation, southeast Idaho ............................................................ 321 J.R. H e r r i n g and R.I. G r a u c h A b s t r a c t ................................................................................................................ 321 I n t r o d u c t i o n .......................................................................................................... 322 P u r p o s e and b a c k g r o u n d .................................................................................. 322 M e a s u r e d sections ............................................................................................ M e t h o d s ................................................................................................................ S a m p l i n g .......................................................................................................... A n a l y s e s .......................................................................................................... R e s u l t s .................................................................................................................. C o m p o s i t i o n a l a v e r a g e s o f e l e m e n t s in the M e a d e Peak ................................
324 327 327 328 329 329
G e o m e t r i c m e a n s and d e v i a t i o n s .................................................................... L i t h o l o g i c c h a r a c t e r i z a t i o n .............................................................................. Trace e l e m e n t s ................................................................................................ I n d i v i d u a l rock s a m p l e s ..................................................................................
333 334 338 338
C l o s e - s p a c e d c h a n n e l s a m p l e s , S e c t i o n Z ...................................................... 340 C o m p o s i t i o n a l c h a n g e s due to w e a t h e r i n g , E n o c h Valley c h a n n e l s a m p l e s ............................................................................................................ 340 W e a t h e r i n g and other alteration .......................................................................... 343 T r a c e - e l e m e n t a s s o c i a t i o n s as a f u n c t i o n o f alteration .................................... 345 I n d i v i d u a l trace e l e m e n t s ................................................................................ 354 S e l e n i u m ...................................................................................................... U r a n i u m ...................................................................................................... V a n a d i u m .................................................................................................... O t h e r G e o e n v i r o n m e n t a l l y s i g n i f i c a n t trace e l e m e n t s ........................................ Silver ............................................................................................................ A r s e n i c ........................................................................................................ C a d m i u m ...................................................................................................... T h a l l i u m ...................................................................................................... A l t e r a t i o n m o d e l .................................................................................................. C o n c l u s i o n s ..........................................................................................................
356 356 358 358 358 358 358 359 359 363
Contents
XXI
Chapter 13. Rock leachate geochemistry of the Meade Peak Phosphatic Shale Member of the Phosphoria Formation, southeast Idaho ............................................................ 367 J.R. H e r r i n g A b s t r a c t ................................................................................................................
367
I n t r o d u c t i o n .......................................................................................................... 368 B a c k g r o u n d ...................................................................................................... 368 R e l e v a n c e o f l e a c h a t e e x p e r i m e n t s .................................................................. 368 R o c k s a m p l e s .................................................................................................. 369 M e t h o d s ................................................................................................................
370
S a m p l e p r e p a r a t i o n .......................................................................................... 3 7 0 A n a l y s e s .......................................................................................................... 371 R e s u l t s ..................................................................................................................
372
D i s c u s s i o n ............................................................................................................ 373 C o r r e l a t i o n a n a l y s i s ........................................................................................ 373 C o r r e l a t i o n s w i t h m a j o r c o m p o n e n t s .............................................................. 379 C o r r e l a t i o n w i t h m i n e r a l o g y ............................................................................ 379 F a c t o r a n a l y s i s .................................................................................................. 380 C o r r e l a t i o n w i t h b u l k c h e m i s t r y ...................................................................... 382 L e a c h a t e c o n d i t i o n v a r i a t i o n s .......................................................................... 383 Particle size .................................................................................................. 383 L e a c h a t e t i m e .............................................................................................. 383 H i g h l y a l t e r e d r o c k s .................................................................................... 392 L e s s - a l t e r e d r o c k s ........................................................................................ 393 L e a s t - a l t e r e d r o c k s ...................................................................................... 393 A n o x i c c o n d i t i o n s ........................................................................................ 394 F r e e z e - t h a w effects ...................................................................................... 395 M u l t i p l e l e a c h i n g ........................................................................................ 396 C o n c l u s i o n s .......................................................................................................... 396
Chapter 14. Rex Chert Member of the Permian Phosphoria Formation." Composition, with emphasis on elements of environmental concern .............................................. 399 J.R. Hein, B.R. M c l n t y r e , R.B. Perkins, D.Z. P i p e r a n d J.G. E v a n s A b s t r a c t ................................................................................................................ 399 I n t r o d u c t i o n .......................................................................................................... 4 0 0 P r e v i o u s studies .................................................................................................. 401 M e t h o d s ................................................................................................................ 402 Field s a m p l i n g .................................................................................................. 402 R o c k s a m p l e p r e p a r a t i o n ................................................................................ 405 G e o c h e m i c a l a n a l y s e s ...................................................................................... 405 Statistical a n a l y s e s .......................................................................................... 405 M i n e r a l o g i c a l a n a l y s i s .................................................................................... 4 0 6 R e s u l t s .................................................................................................................. 4 0 6 L i t h o s t r a t i g r a p h y .............................................................................................. 4 0 6 P e t r o g r a p h y ...................................................................................................... 4 1 0 M i n e r a l o g y ...................................................................................................... 411 C h e m i c a l c o m p o s i t i o n .................................................................................... 411
XXII
Contents S t r a t i g r a p h i c c h a n g e s in c h e m i c a l c o m p o s i t i o n .............................................. 4 1 9 P h a s e a s s o c i a t i o n s o f e l e m e n t s ........................................................................ 4 1 9 D i s c u s s i o n a n d c o n c l u s i o n s : E n v i r o n m e n t a l l y sensitive e l e m e n t s ...................... 4 2 4
Chapter 15. Gaseous selenium and other elements in near-surface atmospheric samples, southeast Idaho .................................................................................................. 4 2 7 P.J. L a m o t h e and J.R. H e r r i n g A b s t r a c t ................................................................................................................ I n t r o d u c t i o n .......................................................................................................... L o c a t i o n .............................................................................................................. S t u d y d e s i g n ........................................................................................................ S a m p l e c o l l e c t i o n a n d a n a l y s i s ............................................................................ R e s u l t s a n d d i s c u s s i o n ........................................................................................ C o n c l u s i o n s ..........................................................................................................
427 427 428 428 430 432 433
Chapter 16. Selenium loading through the Blackfoot River watershed." Linking sources to ecosystems ...................................................................................................... 437 T.S. Presser, M. Hardy, M . A . H u e b n e r a n d P.J. L a m o t h e A b s t r a c t ................................................................................................................ I n t r o d u c t i o n .......................................................................................................... Site location and d e s c r i p t i o n ................................................................................ M e t h o d s ................................................................................................................ F l o w ................................................................................................................ S a m p l e collection ............................................................................................ W a t e r ............................................................................................................ S e d i m e n t ...................................................................................................... A n a l y s i s ............................................................................................................ R e s ults .................................................................................................................. Q u a l i t y a s s u r a n c e and q u a l i t y control ............................................................ R e g i o n a l w a t e r - d i s c h a r g e and s e l e n i u m c o n c e n t r a t i o n , speciation,
437 438 441 444 444 444 444 445 446 446 446
and l o a d i n g ...................................................................................................... 4 4 8 H y d r o l o g i c c o n d i t i o n s .................................................................................. 448 S e l e n i u m c o n c e n t r a t i o n s .............................................................................. 453 S e l e n i u m s p e c i a t i o n - s u p p l e m e n t a l data ...................................................... 4 5 4 S u s p e n d e d s e d i m e n t - s u p p l e m e n t a l data .................................................. 455 D i s s o l v e d s e l e n i u m load c a l c u l a t i o n s .......................................................... 455 S e l e n i u m load forecasts for a v e r a g e and wet y e a r s - s u p p l e m e n t a l data .... 4 5 6 R e g i o n a l s e l e n i u m r e s e r v o i r ................................................................................ 457 G e o h y d r o l o g i c b a l a n c e .................................................................................... 4 5 7 S e l e n i u m sources and source d r a i n a g e ............................................................ 4 5 8 C o n c l u s i o n s .......................................................................................................... 461
Chapter 17. Selenium attenuation in a wetland formed from mine drainage in the Phosphoria Formation, southeast Idaho ................................................................................ 4 6 7 L.L. Stillings a n d M.C. A m a c h e r A b s t r a c t ................................................................................................................ 4 6 7 I n t r o d u c t i o n .......................................................................................................... 468
Contents
XXIII M e t h o d s ................................................................................................................ 469 Site .................................................................................................................. 469 C o l l e c t i o n and analytical m e t h o d s .................................................................. 469 Surface waters .............................................................................................. 469 S e d i m e n t s .................................................................................................... 472 Results .................................................................................................................. 473 Water samples .................................................................................................. 473 S e d i m e n t samples ............................................................................................ 473 D i s c u s s i o n ............................................................................................................ 474 C o n c l u s i o n s .......................................................................................................... 480
Chapter 18. Selenium and other trace elements in water, sediment, aquatic plants, aquatic invertebrates, and fish from streams in SE Idaho near phosphate mining ................................................................................................ 483 S.J. H a m i l t o n , K.J. Buhl and P.J. L a m o t h e A b s t r a c t ................................................................................................................ 483 I n t r o d u c t i o n .......................................................................................................... 483 M e t h o d s and materials ........................................................................................ 484 Collection site description .............................................................................. 484 Sample collection ............................................................................................ 487 Water quality analyses and flow m e a s u r e m e n t ................................................ 488 E l e m e n t analysis .............................................................................................. 489 Statistical analyses .......................................................................................... 489 Results and discussion ........................................................................................ 490 Quality assurance/quality control o f c h e m i c a l analyses .................................. 490 Water ................................................................................................................ 491 S e l e n i u m ...................................................................................................... 491 O t h e r e l e m e n t s ............................................................................................ 493 C o m p a r i s o n to other Idaho data .................................................................. 494 S e d i m e n t .......................................................................................................... 495 S e l e n i u m ...................................................................................................... 495 O t h e r e l e m e n t s ............................................................................................ 498 C o m p a r i s o n to other Idaho data .................................................................. 500 Aquatic plants .................................................................................................. 501 S e l e n i u m ...................................................................................................... 501 O t h e r e l e m e n t s ............................................................................................ 501 C o m p a r i s o n to other Idaho data .................................................................. 503 Aquatic invertebrates ...................................................................................... 504 S e l e n i u m ...................................................................................................... 504 O t h e r e l e m e n t s ............................................................................................ 505 C o m p a r i s o n to other Idaho data .................................................................. 506 Fish .................................................................................................................. 506 S e l e n i u m ...................................................................................................... 506 O t h e r elements ............................................................................................ 508 C o m p a r i s o n to other I d a h o data .................................................................. 513 O t h e r considerations .................................................................................... 515 H a z a r d a s s e s s m e n t .............................................................................................. 515
Contents
XXIV
Chapter 19. Uptake of selenium and other contaminant elements into plants and implications for grazing animals in southeast Idaho ........................................ 527 C.L. M a c k o w i a k , M . C . A m a c h e r , J.O. H a l l a n d J.R. H e r r i n g A b s t r a c t ................................................................................................................ 527 I n t r o d u c t i o n .......................................................................................................... 528 L o c a t i o n s a n d g e n e r a l g e o l o g y ........................................................................ 528 Se a n d o t h e r trace e l e m e n t s - e n v i r o n m e n t a l c o n c e r n s .................................. 528 Plants for a s s e s s i n g t r a c e - e l e m e n t m o b i l i t y .................................................... 529 M e t h o d s ................................................................................................................ 531 E x p e r i m e n t a l d e s i g n ........................................................................................ 531 P l a n t s a m p l i n g a n d p r e p a r a t i o n ...................................................................... 531 A n a l y s e s .......................................................................................................... Statistical a n a l y s e s .......................................................................................... R e s u l t s and d i s c u s s i o n ........................................................................................ S e l e n i u m in v e g e t a t i o n .................................................................................... G e o g r a p h i c effects ...................................................................................... G e o l o g i c effects .......................................................................................... R e d o x effects .............................................................................................. Ve g e t a t i o n effects ........................................................................................ O t h e r trace e l e m e n t s in v e g e t a t i o n ..................................................................
533 533 534 534 534 535 535 539 542
C a d m i u m ...................................................................................................... 542 C h r o m i u m .................................................................................................... 544 C o p p e r .......................................................................................................... M a n g a n e s e .................................................................................................. M o l y b d e n u m ................................................................................................ N i c k e l ..........................................................................................................
544 544 545 545
Z i n c .............................................................................................................. 545 L i v e s t o c k / w i l d l i f e r e s p o n s e to s e l e n i u m a n d other trace e l e m e n t s ................ 546 S e l e n i u m ...................................................................................................... 546 M o l y b d e n u m ................................................................................................ 548 O t h e r trace e l e m e n t s .................................................................................... L i v e s t o c k p r o t e c t i o n ........................................................................................ P h y s i c a l m a n i p u l a t i o n s ................................................................................ C h e m i c a l m a n i p u l a t i o n s ..............................................................................
549 549 549 550
Vegetation m a n i p u l a t i o n s ............................................................................ 550 C o n c l u s i o n s .......................................................................................................... 551 P A R T V. M O D E L I N G S T U D I E S
Chapter 20. Review of world sedimentary phosphate deposits and occurrences .................. 559 G.J. Orris and C.B. C h e r n o f f I n t r o d u c t i o n .......................................................................................................... 559 D a t a ...................................................................................................................... A c q u i s i t i o n ...................................................................................................... D e s c r i p t i o n ...................................................................................................... S e d i m e n t a r y p h o s p h a t e deposits ..........................................................................
559 559 561 562
Contents
XXV Marine sedimentary phosphate deposits .......................................................... Active margin basin and epicontinental sea deposits .................................. Shelf and platform deposits ........................................................................ Other marine deposits .................................................................................. Insular phosphate deposits .............................................................................. Weathering-related residual and infiltration deposits .................................. Guano and guano-related deposits .............................................................. Formation and distribution o f deposits ................................................................ Conclusions ..........................................................................................................
563 563 563 563 564 564 565 565 571
Chapter 21. Western Phosphate F i e l d - Depositional and economic deposit models .......... 575 P.R. Moyle and D.Z. Piper Abstract ................................................................................................................ 575 Introduction .......................................................................................................... 576 H y d r o g r a p h y o f the Phosphoria s e a - a depositional model .............................. 576 Non-marine sediment fraction ........................................................................ 578 Marine sediment fraction ................................................................................ 580 Character and controls o f phosphate r e s o u r c e s - an economic model .............. 586 Geological setting ............................................................................................ 586 Geological attributes related to mining .......................................................... 587 Mining characteristics and specifications ........................................................ 589 Weathering ...................................................................................................... 590 Resources and reserves .................................................................................... 592 Conclusions .......................................................................................................... 593
Chapter 22. Societal relevance, processing, and material flow o f western phosphate Refreshments, fertilizer, and weed killer ............................................................ 599 S.M. Jasinski Abstract ................................................................................................................ Introduction .......................................................................................................... Utilization and societal relevance o f phosphate .................................................. Mining methods .................................................................................................. Processed products .............................................................................................. Phosphoric acid production ............................................................................ Elemental phosphorus production .................................................................. End products ........................................................................................................ Fertilizer products ............................................................................................ Major non-fertilizer applications .................................................................... Material flow in the environment ........................................................................
599 599 600 602 604 604 604 607 607 607 608
Appendix CD: Table o f world sedimentary phosphate deposits (with appendices for chapters 12, 13, 18, and 19) .......................................................................... 611 G.J. Orris and C.B. Chernoff
Author Index ............................................................................................................................ Subject Index ..........................................................................................................................
613 621
This Page Intentionally Left Blank
PART I.
INTRODUCTION
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
Chapter 1
THE PERMIAN EARTH J.R. H E I N
INTRODUCTION The Permian (about 290-250 Ma) was a time of immense global change and unique paleogeographic configurations. Climate conditions evolved from a glacial icehouse Earth to a hothouse Earth. Land masses were assembled into one great continent of Pangea that extended from pole to pole and the mega-ocean Panthalassa dominated the Earth's surface. Ocean chemistry changed dramatically. For example, the ratio of the isotopes of strontium dissolved in ocean water plunged sharply, equivalent in magnitude (not absolute values) to the marked rise in that ratio during the Cenozoic. Climatic and oceanic conditions were favorable for the formation of vast quantities of energy resources and mineral deposits, such as petroleum, coal, phosphorite, and evaporites. During the Late Permian, central northern Pangea was rocked by the outpouring of voluminous volcanic eruptions that produced the Siberian traps. The Permian ended with the greatest mass extinction of biota recorded in Earth history. Because of these dramatic events that characterized the Permian, many books have been produced over the past decade describing this remarkable geologic Period (e.g. Erwin, 1993; Scholle et al., 1995a,b; Shi et al., 1998; Yin et al., 2000; Shi and Metcalfe, 2002). Here I provide but a very brief summary of the Permian to set the stage for later chapters in this book. The International Union of Geological Science's (IUGS) International Commission on Stratigraphy (ICS) has adopted a subdivision of the Permian that includes three Epochs that are divided into nine Stages, that is four Stages, three Stages, and two Stages for the Cisuralian, Guadalupian, and Lopingian Epochs, respectively (Fig. 1-1; http://www.micropress.org/stratigraphy). According to the IUGS-ICS, the Permian began about 292 Ma ago and ended about 251 __+3.6 Ma.
GEOLOGY, PLATE TECTONICS AND PALEOGEOGRAPHY Through movement of Earth's tectonic plates, the land masses had largely amalgamated into the supercontinent of Pangea by the Early Permian. Pangea during that time consisted of an arcuate western subduction margin with several huge embayments (Fig. 1-2). Gondwana, Laurasia, and Siberia were colliding at that time to form the western part of Pangea.
1 --
I
Harland e t a .
-
248_+10My
Marine Series and suws
(
SW North America
I
Ural Region Russia
1
NW Europe
1
Tethyan Asia SW
I
China
I
Age Ma
Tatarian
-
-?-
-
Kazanian
Ufimian
z - 258f12My 9
3w
Kungurian
a
-
263illMy
-
Artinskian
-
268t6My
-
Sakrnarian
---Asselian
- 286SMy
-
CARBONIFEROUS
Fig. 1-1. International Commission on Stratigraphy's (ICS) recommended Periods and Stages of the Permian, compared with that of Harland et al. (1990) and Stages used in various parts of the world as presented by Ross and Ross (1995). The bold ages for Stage boundaries are those approved by the ICS and the remainder of the ages were proposed by Wardlaw (1 999).
b
3
2
$.
I:: Summer~nlyupwelling
I (~2OOOm) IL a w h d (O-ZfMm) ~ Mountains Summer & winter upwelling I (1000-2000 m) l Shelf (-2OMm) Uplenda (20(t1000 m) I Deep Ocean (c-200m) Pre-accreted Terranes
Fig. 1-2. Paleogeographic continental reconstruction of the Roadian-Wordian Earth and location of the Phosphoria sea (base map from Ziegler et al. (1997); surface water currents, climatic provinces, and pre-accreted terranes from Mei and Henderson (2001); and upwelling zones from Parrish (1982).
6
JR. Hein
An irregular eastern margin of Pangea semi-enclosed the wide Paleotethys sea and was characterized by a complex array of continental fragments. Those continental fragments of Cimmeria were being tiffed from northern Gondwana to the north through the Paleotethys (Scotese and Langford, 1995). The Paleotethys covered what is now most of southern and central Europe. Subduction characterized much of the eastern margin of Pangea and the northern margin of the Paleotethys (Scotese and Langford, 1995). The Paleotethys contracted during the Permian and closed completely near the end of the Triassic. Shallow, salty inland seas (e.g. Zechstein sea) covered parts of what is now northern Europe and very large freshwater lakes formed in the southern hemisphere after the south polar ice cap melted. Pangea moved north during the Permian. Most of the southem hemisphere south of 60 ~ was covered by glacial ice during the early Permian, which had disappeared by late-early Permian (Ziegler et al., 1997). Mountain ranges bordered the southern, southwestern, and northeastern margins of Pangea, with the extensive northeast-southwest trending Central Pangean Mountains (Appalachian-Mauretanide-Variscan orogenic belts) dividing the continent (Fig. 1-2; Hatcher et al., 1989; Ziegler et al., 1997). An island arc bordered the northwest continental margin, which formed the western margin of the Phosphoria sea.
CLIMATE The formation of Pangea had profound effects on continental climates. Temperatures in continental interiors increased, thereby gradually increasing the size and aridity of deserts and decreasing the size of the south polar ice cap and extent of northern sea ice. Seasonal fluctuations in climate increased dramatically as did the number of great storms (e.g. Parrish et al., 1986; Crowley et al., 1989; Kutzbach and Gallimore, 1989; Barron and Fawcett, 1995; Parrish, 1995). The general increase in aridity and decrease in humidity are reflected in the expansion of eolian and evaporite deposits and decrease in coal formation through the Permian. Low latitudes in Pangea were characterized by a monsoonal climate in the Early Permian, which gave way to increased aridity in the Late Permian (Parrish, 1995). Global warming may have been caused by a nearly twofold increase in atmospheric carbon dioxide levels, from levels somewhat less than modem values to levels no more than twice that of present day (Berner, 1991). Kutzbach and Gallimore (1989) modeled a fourfold increase in carbon dioxide levels for the Permian, which resulted in minor to moderate changes in precipitation (8% increase) and mean surface temperature (3.5~ increase). More recent studies have indicated that the rise in carbon dioxide levels in the atmosphere through the Permian may have been much greater, with concomitant increases in temperatures (Berner, 1994; Ekart et al., 1999; Ghosh et al., 2001). Temperatures may have gotten hot enough near the end of the Permian to have caused stress to global ecosystems. Atmospheric oxygen contents plunged from the Phanerozoic high in the Carboniferous of about 40% 02 to about 20% 02 at the end of the Permian, essentially equivalent to the present-day 21% 02 (Berner et al., 2000). Siberia and continental fragments of Cimmeria in the Paleotethys may have been the only parts of Pangea with a positive moisture balance,
The Permian Earth
7
with much of the remainder of interior Pangea within 45 ~ latitude of the equator being arid (Ziegler, 1990; Barron and Fawcett, 1995). Winter storm tracts may have occurred mostly between 40 ~ and 70 ~ south latitudes (June-August) and 30 ~ and 50 ~ north latitudes (December-February; Barron and Fawcett, 1995). The northern hemisphere storm track only impinged on the continent of Pangea in the northeast, otherwise was confined to Panthalassa. The southern hemisphere storm tract may have been accompanied by a small, secondary track in the northern hemisphere, offshore northeastern Pangea.
OCEANOGRAPHY Little is known about Panthalassa except along its margins with Pangea. General circulation modeling, however, indicates that surface-water currents were likely rather simple for this mega-ocean that spanned about 300 ~ of longitude at its greatest extent. Equatorial Panthalassa was likely characterized by a westward-flowing surface-current system, that entered the Paleotethys sea on encountering eastern Pangea (Belasky et al., 2002). After transiting the Paleotethys sea, the current bifurcated into coast-parallel northto-east and south-to-east currents (Kennett, 1982; Kutzbach et al., 1990). In the northern and southern hemispheres, Panthalassa may have been characterized by subpolar cyclonic and subtropical anticyclonic circulation cells (Kutzbach et al., 1990). High-latitude surface waters and deep waters were likely warmer than they are today, whereas the equator-to-pole thermal gradient was less or about the same as it is today (e.g. Ziegler et al., 1997; Ziegler, 1990; Belasky, 1994). Thermohaline circulation might have been driven by sinking of Panthalassa cool polar water by melting of seasonal sea ice, and secondarily from sinking of warm, saline Paleotethys water (Kutzbach et al., 1990; Beauchamp and Baud, 2002). Thermohaline circulation may have ceased during the Lopingian. Identification of zones of upwelling is important in terms of mapping potential regions of petroleum source beds and phosphorites (Parrish, 1982). Upwelling of deep cold marine waters to the surface redistributes nutrients and reflects specific conditions of atmospheric circulation. The Late Permian Earth displayed zones of upwelling distributed much as they are today, that is off the west coast of continents. Modeling of atmospheric circulation and pressure cells for the Late Permian indicate that upwelling occurred along most of the western margin of Pangea from about 35~ to 35~ latitudes (Parrish, 1982). In addition, local areas of upwelling may have occurred along the margin of the Paleotethys sea, especially along the northeast, west, and southeast margins (Fig. 1-2). One of the many controversies in Permian studies is whether Panthalassa was stratified and anoxic, and if so, at what water depths and when. An anoxic Panthalassa has been implicated in the great Permian-Triassic biotic extinctions (see below; Erwin, 1993). Based on analysis of redox sensitive elements in what are interpreted to be open-ocean deep-water cherts from Japan, Kato et al. (2002) determined that the deep-waters of Panthalassa were anoxic to suboxic for nearly l0 Ma prior to the Permian-Triassic boundary. Also based on the composition of cherts from Japan and British Columbia,
8
JR. Hein
The Permian Earth
9
Isozaki (1997) suggested that deep-water anoxia was maintained for 20 Ma and that the entire water column was anoxic for 10 Ma. Musashi et al. (2001) measured a significant drop in 813C values for what are interpreted to be open-ocean shallow-water carbonates and organic matter, similar to the ~il3c drop found in continental margin rocks of Late Permian age. This global drop in gl3c values indicates a significant input of 12C into the oceans and atmosphere in the Late Permian, perhaps by dissociation of huge quantities of gas hydrates. On the other hand, modeling of Panthalassa indicates that it would be difficult to sustain sulfate-reducing anoxic condition in the deep oceans for any significant length of time (Zhang et al., 2001). Hotinski et al. (2001) add that oceanic anoxia would markedly reduce upwelling and consequently productivity; yet the accumulation of carbon dioxide and hydrogen sulfide that may have promoted mass extinction would have required considerable increases in nutrient supply. The chemical composition of Panthalassa ocean water may have changed markedly through the Permian (Fig. 1-3). Ocean water had a relatively high Mg2+/Ca 2+ ratio (>2.1, maximum --5) and relatively high Na + concentration during the Permian, although that ratio fluctuated dramatically (Hardie, 1996; Lowenstein et al., 2001; Dickson, 2002; Horita et al., 2002). During the whole of the Permian, non-skeletal aragonite and high-Mg calcite precipitated from Panthalassa ocean water, rather than calcite, as determined by a Mg 2+ / Ca 2§ ratio of >2; evaporite basins produced predominantly mixed MgSOa-plus KCl-type potash deposits, rather than solely one type or the other (Hardie, 1996). The precipitation of aragonite from high Mg2+/Ca 2+ ocean water resulted in low ratios of Sr2+/Ca 2+ in Permian ocean water (Steuber and Veizer, 2002). These variations in ocean-water chemistry are thought to have resulted from variations in the production of ocean-floor basalt (and associated hydrothermal products) relative to the input of materials from rivers. The production rate of oceanic crust was among the lowest rates for the Phanerozoic, comparable to present-day rates (Gaffin, 1987). The isotopic compositions of Panthalassa ocean water changed markedly through the Permian, during which C, S, and Sr isotopes reached their most extreme Phanerozoic values (Scholle, 1995). Sulfur isotopes of ocean-water sulfate gradually decreased from about 30%0 ~34ScDT in the Cambrian to the lowest Phanerozoic value of 10%o in the Permian. ~34S then increased sharply near the Permian-Triassic boundary to about 30%0, dropped just as sharply in the Early Triassic to about 12%o and then slowly climbed to its present-day value of 21%0 (Scholle, 1995; Strauss, 1997). Global changes in the isotopic
Fig. 1-3. Geologic, oceanographic, and biotic changes that occurred during the Permian; shaded horizontal band marks time of deposition of the Phosphoria Formation. Magnetic data from Haag and Heller (1991); climate change based on conodonts from Mei et al. (2002); tectonic-volcanic events from Holser and Magaritz (1987), Lo et al. (2002), and Reichow et al. (2002); third-order eustatic curve from Ross and Ross (1995); ocean-crust production (on the scale of the Phanerozoic, the whole range of Permian production rates is low) from Gaffin (1987); number of extant genera from Holser and Magaritz (1987); ~13CpDB from Scholle (1995) and Kakuwa (1996); ~34S from Scholle (1995) and Strauss (1997); 87Sr/86Sr from Holser and Magaritz (1987) and Denison and Koepnick (1995); Mg/Ca molar ratio of ocean water from Hardie (1996).
10
JR. Hein
composition of ocean-water sulfate is generally considered to result from relative changes in the rates of formation of reduced and oxidized sulfur deposits, sulfides, and sulfates (e.g. Strauss, 1997). Because of changes in the amount of biomass, it would be expected that ~34Sof ocean-water sulfate should decrease during major extinction and increase during the subsequent radiation. The dating of Permian-Triassic sections where ocean-water sulfate isotopic values can be determined are not dated well enough to pinpoint where the spike in the curve occurs. Sulfur isotopes did decrease prior to the main extinction event(s) and increased sometime after the major spike near the Permian-Triassic boundary. During the Permian, C isotopes (813C) in carbonates and organic matter varied in tandem and decreased slightly, increased slightly, or remained constant in different regions up to the Late Permian. Then in all sections, 813C values plunged dramatically (negative excursion) near the Permian-Triassic boundary (e.g. Scholle, 1995; Musashi et al., 2001). This negative excursion has been recognized for many years and can be found at times of other mass extinctions in the geologic record. A variety of explanations have been put forth to explain the negative excursions, with early explanations involving changes in the rate of burial and storage of organic carbon. More recent explanations have included dramatic inputs of 12C-rich material into the ocean-atmosphere system, such as by dissociation of gas hydrates, volcanic eruptions, overturn of a stagnant ocean, etc. (see End of Permian section below). Likewise, Sr isotopes (87Sr/86Sr) show their lowest Phanerozoic ratio near the Permian-Triassic boundary. Sr isotopes show a steep decline from about 0.7083 at the beginning of the Permian to about 0.7067 in the Late Permian (Denison and Koepnick, 1995). This dramatic drop is equivalent in magnitude to the dramatic increase in the ratio during the Cenozoic. Changes in the Sr isotope ratio of ocean water is usually attributed to changes in the ratio of input of low-ratio Sr from mafic crustal sources relative to highratio Sr from old continental cratonic crust. Although production of oceanic crust was low during the Permian in general, and decreased during the first half of the Permian, it increased during the last half and, combined with extensive production of flood basalts, likely contributed to the decline in the oceanic Sr isotope ratio. Perhaps more important was the increasing aridity of Pangea through the Permian and the accompanying increase in internal drainage and sedimentation, which inhibited the input of old continental material to Panthalassa (Denison and Koepnick, 1995).
WESTERN NORTH AMERICAN MARGIN AND THE PHOSPHORIA SEA The Phosphoria Formation was deposited off the western margin of North America (northwest margin of Pangea) predominantly during the Roadian and Wordian Stages (Wardlaw et al., 1995). That margin consisted of islands to the west, a marginal sea with a wide, shallow-water eastern margin, and wide coastal plain on which sand dune fields and evaporite basins were common (Fig. 1-4). Terranes that would later be accreted to the northwest Pangean continental margin existed farther to the west (Belasky et al., 2002). The shallow marginal sea was the location of deposition of the Phosphoria Formation.
The Permian Earth
11
Fig. 1-4. Physiographic map of northwest Pangea during the early Guadalupian and location of the Phosphoria sea (map from the web site of R. Blakey, Northern Arizona University and used with his permission: http://jan.ucc.nau.edu/--rcb7/RCB.html). During the Late Permian, the Ancestral Rocky Mountains were subdued and the Front Range and Uncompahgre uplifts were the only important sediment sources (Burchfiel et al., 1992). Late-Permian and Triassic subduction-related mountain building is reflected in the Sonoma orogeny. Remnants of that arc can be traced from California into southern British Columbia (Burchfiel et al., 1992; Miller et al., 1992). Northwest Pangea was dominated by high pressure in both the summer and winter seasons and winter storm tracts were well to the north of the Phosphoria basin (Barron and Fawcett, 1995; Parrish, 1995). The entire area of the Phosphoria basin was a region of moderate to intense upwelling during both the summer and winter. Upwelling brought cold-nutrient-rich waters to the surface, which promoted productivity, leading to the accumulation of organic carbon-rich sediments on the seafloor. Organic matter was the source of the phosphorous for the phosphorites (see Moyle and Piper, Chapter 21). Upwelling was created by the equator-directed surface current that flowed along the continental margin (Fig. 1-2). This situation is comparable to that of the California margin today. Upwelling would have been promoted by westerly winds at temperate latitudes. A decline of upwelling related to a number of oceanographic and atmospheric changes during the Late Permian terminated the deposition of high-grade phosphorite. However, minor episodes of phosphogenesis producing low-grade or localized deposits occurred into the Triassic within the Phosphoria sea and elsewhere adjacent to Pangea during the
12
JR. Hein
Permian, especially around the margins of the Paleotethys sea (Herring, 1995). However, about 85% of Permian phosphorites formed in the Phosphoria sea. High-grade phosphorite, deposited within the Phosphoria sea during the Roadian (Meade Peak Member) and Wordian (Retort Member), formed under cool-water conditions during maximum transgression, in contrast to warm-water conditions that characterized deposition of carbonates below the Meade Peak Member of the Phosphoria Formation (Wardlaw et al., 1995; see also Murchey, Chapter 5). The Phosphoria sea was characterized by a low influx of terrigenous sediment, suboxic bottom-water conditions during phosphorite deposition, and moderate levels of primary productivity in the photic zone (Cook, 1968; Piper, 2001; see also Chapters 4, 5, and 21). Sedimentary structures indicate that bottom waters were relatively calm compared to those typically found in open-shelf environments. Estimates of the maximum water depth of the Phosphoria sea have varied widely, up to 1000 m (e.g. McKelvey et al., 1959), but likely had a maximum depth range of 300-500 m.
END OF PERMIAN The cause(s) of the end of Permian mass extinction is difficult to assess for many reasons, not the least of which is the poor preservation of Late Permian and Early Triassic strata resulting from marine regression and reflected by a global hiatus. Raup (1979) and Sepkoski (1989) suggested that up to 96% (90% commonly cited figure) of marine species became extinct during the latest Permian. A dominant change from Late Permian to Early Triassic was the replacement of marine sessile filter-feeding epifauna by mobile infauna (Erwin, 1993). Reef communities, plankton, and invertebrates with a planktonic larval stage were especially devastated. The amount of loss of land fauna and flora is less clear because of poor preservation of terrestrial strata, but up to 70% of vertebrate species may not have survived the Late Permian. Terrestrial organisms clearly declined, but correlation of those declines with those of marine organisms has generally not been established (Erwin, 1993), except in a section in Greenland (Twitchett et al., 2001). A whole range of causes or mitigating circumstances for the mass extinction has been put forward: (a) Extensive volcanism, especially the west Siberian Traps (e.g. Renne and Basu, 1991); (b) Bolide impact (e.g. Becker et al., 2001); (c) Tectonic events (Holser and Magaritz, 1987); (d) Magnetic reversal, end of the Kiaman reversed superchron (Fig. 1-3; Haag and Heller, 1991); (e) Profound marine regression (Erwin, 1993); (f) Severe environmental changes (e.g. climatic cooling; Stanley, 1988); (g) Dramatic decrease in atmospheric and soil oxygen content (Sheldon and Retallack, 2002); (h) Development of oceanic anoxia (e.g. Wignall and Hallam, 1992; Hotinski et al., 2001); (i) Development of extensive evaporite basins and brackish oceans (e.g. Fischer, 1964; Stevens, 1977); (j) Massive release of methane from gas hydrates (e.g. Erwin, 1993); and (k) Changes in marine hot-spot systems and spreading ridges (Fig. 1-3; Holser and Magaritz, 1987). The final answer for the cause(s) of this enormous mass extinction is still pending. It is likely that the end of Permian extinctions developed over a million years or so. Also, this relatively slow progression may have been finalized by a devastating blow from eruption
The Permian Earth
13
of the Siberian Traps (Bowring et al., 1998), overturn of an anoxic ocean causing carbon dioxide poisoning of shallow-water biota (Wignall and Twitchett, 1996; Knoll et al., 1996), and/or a bolide collision with the Earth at that time (Becker et al., 2001), which may in fact have initiated the Siberian Traps volcanism (Jones et al., 2002). It is likely that a host of events and processes conspired in this greatest mass extinction, including climate change due to the evolution of Pangea and major volcanic events; a disruption of nutrient supplies and loss of habitat caused by major oceanic regression, and decreased upwelling resulting from changes in ocean stratification. The potential for considerable climate change is supported by recent work that has demonstrated that the west Siberian flood basalts were twice as voluminous as had been previously thought for this largest of known Large Igneous Provinces (LIP) (Reichow et al., 2002). Its potential to have influenced climate change is compelling. In addition, the Emeishan flood basalts in south China were erupted at the same time or slightly before the Siberian Traps. Significantly, the Emeishan basalts were erupted through marine limestones that likely triggered massive releases of carbon dioxide and methane (Lo et al., 2002). These two huge volcanic events could have produced a severe but short-lived cooling event (caused by volcanic dust, aerosols, sulfur dioxide- with acid rain) followed immediately by global warming (caused by CO2, CH4; in part from dissociation of gas hydrates), which in concert may have orchestrated this unparalleled mass extinction.
ACKNOWLEDGEMENTS Brandie McIntyre provided technical support. Paul Belasky, Ohlone College, and Calvin Stevens, San Jose State University, provided very helpful reviews. I thank R. Blakey, Northern Arizona University, and A.M. Ziegler, University of Chicago, for allowing the use of their paleogeographic maps.
REFERENCES Barron, E.J. and Fawcett, P.J., 1995. The climate of Pangaea: A review of climate model simulations of the Permian. In: P.A. Scholle, T.M. Peryt and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea l: Paleogeography, Paleoclimates, Stratigraphy. Springer-Verlag, Berlin, pp. 37-52. Beauchamp, B. and Baud, A., 2002. Growth and demise of Permian biogenic chert along northwest Pangea: evidence for end-Permian collapse of thermohaline circulation. Palaeogeogr. Palaeoclimatol. Palaeoecol., 184: 37-63. Becker, L., Poreda, R.J., Hunt, A.G., Bunch, T.E. and Rampino, M., 2001. Impact event at the Permian-Triassic boundary: evidence from Extraterrestrial noble gases in fullerenes. Science, 291: 1530-1533. Belasky, E, 1994. Biogeography of Permian corals and the determination of longitude in tectonic reconstructions of the paleoPacific region. Can. Soc. Pet. Geologists Memoir, 17; 621-646.
14
JR. Hein
Belasky, P., Stevens, C.H. and Hanger, R.A., 2002. Early Permian location of western North American terranes based on brachiopod, fusulinid, and coral biogeography. Palaeogeogr. Palaeoclimatol. Palaeoecol., 179: 245-266. Berner, R.A., 1991. A model for atmosphere CO2 over Phanerozoic time. Am. J. Sci., 291: 339-376. Berner, R.A., 1994. GEOCARB II: A revised model of atmospheric CO2 over Phanerozoic time. Am. J. Sci., 294: 56-91. Berner, R.A., Petsch, S.T., Lake, J.A., Beerling, D.J., Popp, B.N., Lane, R.S., Laws, E.A., Westley, M.B., Cassar, N., Woodward, El. and Quick, W.P., 2000. Isotope fractionation and atmospheric oxygen: implications for Phanerozoic 02 evolution. Science, 287:1630-1633. Bowring, S.A., Erwin, D.H., Jin, Y.G., Martin, M.W., Davidek, K. and Wang, W., 1998. U/Pb zircon geochronology and tempo of the end-Permian mass extinction. Science, 280:1039-1045. Burchfiel, B.C., Cowan, D.S. and Davis, G.A., 1992. Tectonic overview of the Cordilleran orogen in the western United States. In: B.C. Burchfiel, P.W. Lipman and M.L. Zoback (eds.), The Cordilleran Orogen: Conterminous US. Geology of North America vol. G-3, Geological Society of America, Boulder, CO, pp. 407-480. Cook, P.J., 1968. The petrology and geochemistry of the Meade Peak Member of the Phosphoria Formation. Unpublished Ph.D. thesis, University of Colorado, Boulder, CO, 204 pp. Crowley, T.J., Hyde, W.T. and Short, D.S., 1989. Seasonal cycle variations on the supercontinent of Pangea. Geology, 17: 457-460. Denison, R.E. and Koepnick, R.B., 1995. Variation in 87Sr/86Sr of Permian seawater: an overview. In: P.A. Scholle, T.M. Peryt and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea 1: Paleogeography, Paleoclimates, Stratigraphy. Springer-Verlag, Berlin, pp. 124-132. Dickson, J.A.D., 2002. Fossil echinoderms as monitor of the Mg/Ca ratio of Phanerozoic oceans. Science, 298: 1222-1224. Ekart, D.D., Cerling, T.E., Montanez, I.P. and Tabor, N.J., 1999. A 400 million year carbon isotope record of pedogenic carbonate: implications for paleoatmospheric carbon dioxide. Am. J. Sci., 299: 805-827. Erwin, D.H., 1993. The Great Paleozoic Crisis: Life and Death in the Permian. Columbia University Press, New York, 327 pp. Fischer, A.G., 1964. Brackish oceans as the cause of the Permo-Triassic marine faunal crisis. In: A.E.M. Nairn (ed.), Problems in Palaeoclimatology. Interscience, London, pp. 566-574. Gaffin, S., 1987. Ridge volume dependence on seafloor generation rate and inversion using long term sealevel change. Am. J. Sci., 287:596-611. Ghosh, P., Ghosh, P. and Bhattacharya, S.K., 2001. CO2 levels in the Late Palaeozoic and Mesozoic atmosphere from soil carbonate and organic matter, Satpura basin, Central India. Palaeogeogr. Palaeoclimatol. Palaeoecol., 170:219-236. Haag, M. and Heller, E, 1991. Late Permian to Early Triassic magnetostratigraphy. Earth Planet. Sci. Lett., 107: 42-54. Hardie, L.A., 1996. Secular variation in seawater chemistry: an explanation for the coupled secular variation in the mineralogies of marine limestones and potash evaporites over the past 600 m.y. Geology, 24: 279-283. Harland, W.B., Armstrong, R.L., Cox, A.V., Craig, L.E., Smith, A.G. and Smith, D.G., 1990. A Geologic Time Scale 1989. Cambridge University Press, Cambridge, 263 pp. Hatcher, R.D. Jr., Thomas, W.A., Geiser, P.A., Snoke, A.W., Mosher, S. and Wiltschko, D.V., 1989. Alleghanian orogen. In: R.D. Hatcher Jr., W.A. Thomas and G.W. Viele (eds.), The
The Permian Earth
15
Appalachian-Ouachita Orogen in the United States. Geology of North America vol. F-2, Geological Society of America, Boulder, CO, pp. 233-318. Herring, J.R., 1995. Permian phosphorites: A paradox of phosphogenesis. In: EA. Scholle, T.M. Peryt and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea 2: Sedimentary Basins and Economic Resources. Springer-Verlag, Berlin, pp. 292-312. Holser, W.T. and Magaritz, M., 1987. Events near the Permian-Triassic boundary. Mod. Geol., 1 l: 155-180. Horita, J., Zimmermann, H. and Holland, H.D., 2002. Chemical evolution of seawater during the Phanerozoic: implications from the record of marine evaporites. Geochim. Cosmochim. Acta, 66: 3733-3756. Hotinski, R.M., Bice, K.L., Kump, L.R., Najjar, R.G. and Arthur, M.A., 2001. Ocean stagnation and end-Permian anoxia. Geology, 29: 7-10. Isozaki, Y., 1997. Permo-Triassic boundary superanoxia and stratified superocean: Records from lost deep sea. Science, 276: 235-238. Jones, A.P., Price, G.D., Price, N.J., DeCarli, P.S. and Clegg, R.A., 2002. Impact induced melting and the development of large igneous provinces. Earth Planet. Sci. Lett., 202:551-561. Kakuwa, Y., 1996. Permian-Triassic mass extinction event recorded in bedded chert sequence in southwest Japan. Palaeogeogr. Palaeoclimatol. Palaeoecol., 121: 35-51. Kato, Y., Nakao, K. and Isozaki, Y., 2002. Geochemistry of Late Permian to Early Triassic pelagic cherts from southwest Japan: implications for an oceanic redox change. Chem. Geol., 182: 15-34. Kennett, J., 1982. Marine Geology. Prentice-Hall, Englewood Cliffs, N J, 813 pp. Knoll, A.H., Bambach, R.K., Canfield, D.E. and Grotzinger, J.P., 1996. Comparative earth history and Late Permian mass extinction. Science, 273: 452-457. Kutzbach, J.E. and Gallimore, R.G., 1989. Pangean climates: megamonsoons of the megacontinent. J. Geophys. Res., 94: 3341-3357. Kutzbach, J.E., Guetter, EJ. and Washington, W.M., 1990. Simulated circulation of an idealized ocean for Pangaean time. Paleoceanography, 5: 299-317. Lo, C.-H., Chung, S.-L., Lee, T.-Y. and Wu, G., 2002. Age of the Emeishan flood magmatism and relations to Permian-Triassic boundary events. Earth Planet. Sci. Lett., 198: 449-458. Lowenstein, T.K., Timofeeff, M.N., Brennan, S.T., Hardie, L.A. and Demicco, R.V., 2001. Oscillations in Phanerozoic seawater chemistry: evidence from fluid inclusions. Science, 294: 1086-1088. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M. and Swanson, R.W., 1959. The Phosphoria, Park City, and Shedhorn Formations in the western phosphate field. US Geological Survey Professional Paper 313-A, 47 pp. Mei, S. and Henderson, C.M., 2001. Evolution of Permian conodont provincialism and its significance in global correlation and paleoclimate implication. Palaeogeogr. Palaeoclimatol. Palaeoecol., 170: 237-260. Mei, S., Henderson, C.M. and Wardlaw, B.R., 2002. Evolution and distribution of the conodonts Sweetognathus and Iranognathus and related genera during the Permian, and their implication for climate change. Palaeogeogr. Palaeoclimatol. Palaeoecol., 180:57-91. Miller, E.L., Miller, M.M., Stevens, C.H., Wright, J.E. and Madrid, R., 1992. Late Paleozoic paleogeographic and tectonic evolution of the western U.S. Cordillera. In: B.C. Burchfiel, P.W. Lipman, and M.L. Zoback (eds.), The Cordilleran Orogen: Conterminous US. Geology of North America vol. G-3, Geological Society of America, Boulder, CO, pp. 57-106.
16
JR. Hein
Musashi, M., Isozaki, Y., Koike, T. and Kreulen, R., 2001. Stable carbon isotope signature in mid-Panthalassa shallow-water carbonates across the Permo-Triassic boundary: evidence for 13C-depleted superocean. Earth Planet. Sci. Lea., 191: 9-20. Parrish, J.T., 1982. Upwelling and petroleum source beds, with reference to Paleozoic. Am. Association of Petroleum Geologists Bulletin, 66: 750-774. Parrish, J.T., 1995. Geologic evidence of Permian climate. P.A. Scholle, T.M. Peryt and D.S. UlmerScholle (eds.), The Permian of Northern Pangea 1: Paleogeography, Paleoclimates, Stratigraphy. Springer-Verlag, Berlin, pp. 53-61. Parrish, J.M., Parrish, J.T. and Ziegler, A.M., 1986. Permian-Triassic paleogeography and paleoclimatology and implications for Therapsid distribution. In: N. Hotton III, P.D. MacLean, J.J. Roth and E.C. Roth (eds.), The Ecology and Biology of Mammal-like Reptiles. Smithsonian Institute Press, Washington DC, pp. 109-131. Piper, D.Z., 2001. Marine chemistry of the Permian Phosphoria Formation and basin, southeast Idaho. Econ. Geol., 96: 599-620. Raup, D.M., 1979. Size of the Permo-Triassic bottleneck and its evolutionary implications. Science, 206:217-218. Reichow, M.K., Saunders, A.D., White, R.V., Pringle, M.S., Al'Mukhamedov, A.I., Medvedev, A.I. and Kirda, N.P., 2002. 4~ dates from the west Siberian basin: Siberian flood basalt province doubled. Science, 296:1846-1849. Renne, P.R. and Basu, A.R., 1991. Rapid eruption of the Siberian Traps flood basalts at the PermoTriassic boundary. Science, 253:176-179. Ross, C.A. and Ross, J.R.P., 1995. Permian sequence stratigraphy. In: P.A. Scholle, T.M. Peryt and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea 1: Paleogeography, Paleoclimates, Stratigraphy. Springer-Verlag, Berlin, pp. 98-123. Scholle, EA., 1995. Carbon and sulfur isotope stratigraphy of the Permian and adjacent intervals. In: P.A. Scholle, T.M. Peryt and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea 1: Paleogeography, Paleoclimates, Stratigraphy. Springer-Verlag, Berlin, pp. 133-149. Scholle, P.A., Peryt, T.M. and Ulmer-Scholle, D.S. (eds.), 1995a. The Permian of Northern Pangea 1: Paleogeography, Paleoclimates, Stratigraphy. Springer-Verlag, Berlin, 261 pp. Scholle, P.A., Peryt, T.M. and Ulmer-Scholle, D.S. (eds.), 1995b. The Permian of Northern Pangea 2: Sedimentary Basins and Economic Resources. Springer-Verlag, Berlin, 319 pp. Scotese, C.R. and Langford, R.P., 1995. Pangea and the paleogeography of the Permian. In: P.A. Scholle, T.M. Peryt and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea 1: Paleogeography, Paleoclimates, Stratigraphy. Springer-Verlag, Berlin, pp. 3-19. Sepkoski, J.J. Jr., 1989. Periodicity in extinction and the problem of catastrophism in the history of life. J. Geol. Soc. Lond., 146: 7-19. Sheldon, N.D. and Retallack, G.J., 2002. Low oxygen levels in earliest Triassic soils. Geology, 30: 919-922. Shi, G.R. and Metcalfe, I. (eds.), 2002. Permian of southeast Asia. Special Issue, J. Asian Earth Sci., 20: 549-774. Shi, G.R., Archbold, N.W. and Grover, M. (eds.), 1998. Permian of Eastern Tethys: Biostratigraphy, Palaeogeography and Resources. Proc. Royal Soc. Victoria, vol. 110(1/2), Melbourne, Australia, pp. 480 pp. Stanley, S.M., 1988. Paleozoic mass extinctions: shared patterns suggest global cooling as a common cause. American Journal of Science, 288: 334-352. Steuber, T. and Veizer, J., 2002. Phanerozoic record of plate tectonic control of seawater chemistry and carbonate sedimentation. Geology, 30:1123-1126.
The Permian Earth
17
Stevens, C.H., 1977. Was development of brackish oceans a factor in Permian extinctions? Geol. Soc. Am. Bull., 88, 133-138. Strauss, H., 1997. The isotopic composition of sedimentary sulfur through time. Palaeogeogr. Palaeoclimatol. Palaeoecol. 132:97-118. Twitchett, R.J., Looy, C.V., Morante, R., Visscher, H. and Wignall, P.B., 2001. Rapid and synchronous collapse of marine and terrestrial ecosystems during the end-Permian biotic crisis. Geology, 29:351-354. Wardlaw, B.R., 1999. Notes from the SPS chair. Permophiles issue #35: 1-4. Wardlaw, B.R., Snyder, W.S., Spinosa, C. and Gallegos, D.M., 1995. Permian of the western United States. In: P.A. Scholle, T.M. Peryt and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea 2: Sedimentary Basins and Economic Resources. Springer-Verlag, Berlin, pp. 23-40. Wignall, EB. and Hallam, A., 1992. Anoxia as a cause of the Permian/Triassic mass extinction: Facies evidence from northern Italy and the westem United States. Palaeogeogr. Palaeoclimatol. Palaeoecol., 93:21-46. Wignall, P.B. and Twitchett, R.J., 1996. Oceanic anoxia and the end Permian mass extinction. Science, 272:1155-1158. Yin, H., Dickins, J.M., Shi, G.R. and Tong, J. (eds.), 2000. Permian-Triassic Evolution of Tethys and Western Circum-Pacific. Developments in Palaeontology and Stratigraphy 18, Elsevier, Amsterdam, 392 pp. Zhang, R., Follows, M.J., Grotzinger, J.P. and Marshall, J., 2001. Could the Late Permian deep ocean have been anoxic? Paleoceanography, 16:317-329. Ziegler, A.M., 1990. Phytogeographic patterns and continental configurations during the Permian Period. In: W.S. McKerrow and S.R. Scotese (eds.), Palaeozoic Palaeogeography and Biogeography. Geol. Soc. Lond. Memoir 12, pp. 363-379. Ziegler, A.M., Hulver, M.L. and Rowley, D.B., 1997. Permian world topography and climate. In: I.P. Martini (ed.), Late Glacial and Postglacial Environmental C h a n g e s - Quaternary, Carboniferous-Permian and Proterozoic. Oxford University Press, New York, pp. I I l-146.
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
19
Chapter 2
EVOLUTION OF T H O U G H T C O N C E R N I N G THE O R I G I N OF THE PHOSPHORIA FORMATION, WESTERN US PHOSPHATE FIELD J.R. HEIN, R.B. PERKINS and B.R. MclNTYRE
ABSTRACT The Phosphoria Formation has been the subject of intensive study for nearly 100 years. Most work during the first half of the twentieth century on the Phosphoria Formation fell under three US Geological Survey (USGS) programs. The first program, from 1909 to 1916, was concerned with mapping the extent of phosphate rock within the western United States, the so-called Western Phosphate Field. Many of the basic features concerning the distribution, structure, and composition of the Phosphoria were determined during that time. The work of Blackwelder was especially critical in delineating geochemical conditions of the depositional basin, the Phosphoria sea. An important process in the formation of the Phosphoria Formation, coastal upwelling, was not understood in those earlier works. A detailed understanding of the processes involved in coastal upwelling was first developed by Kazakov in the late 1930s and those processes were later applied to the Phosphoria Formation in the late 1940s by McKelvey and co-workers. The second USGS program began in 1941 and involved detailed sampling and analyses of Phosphoria rocks to delineate vanadium-rich zones, which were discovered during the earlier USGS program. This sampling and analysis program also identified zinc- and uranium-rich strata within the Phosphoria Formation. The discovery of uranium, along with an increasing demand for phosphate, led to the third large USGS program headed by McKelvey, which began in 1947. From 1947 to 1952, hundreds of stratigraphic sections were measured, described, and sampled from 200 locations in Montana, Wyoming, Idaho, and Utah. From this work, a comprehensive genetic model was developed for the sedimentary and oceanographic conditions that promoted phosphorite deposition. Delineation of regional facies changes, thickness changes, and chemical and petrologic changes also provided criteria for regional correlations of rock units. The cyclic nature of deposition of Phosphoria Formation strata was emphasized. Succeeding studies in the 1960s and 1970s established the characteristics of regional and local depositional environments and paleogeographic and paleoclimatic conditions of the Phosphoria sea and surrounding areas. The Phosphoria Formation was placed into a more restrictive Middle Permian time frame than had been used by previous workers. From the 1980s to the present, investigations focused on the details of phosphogenesis and the use of sequence stratigraphy to understand the role that sea level played in the evolution of the Phosphoria sea and in the cyclic deposition of the Phosphoria sequence.
20
J R . Hein et al.
INTRODUCTION The Middle Permian Phosphoria Formation comprises one of the largest resources of phosphate rock in the world and has been mined for nearly a century (Jasinski et al., Chapter 3). Phosphoria Formation strata are found throughout a region covering 350,000 k n l 2 in Idaho, Wyoming, Montana, Utah, and Nevada in the so-called Western US Phosphate Field (Fig. 2-1). This formation has been studied systematically by the US Geological Survey (USGS) during three major field mapping and sampling programs in 1909-1916, 1941-1944, and 1947-1952; and during follow-up work, which continues to the present day. USGS scientists have returned to the Phosphoria off-and-on for much of the past half century in an effort to better understand this remarkable accumulation of phosphate. The book that includes this paper is a product of the latest USGS effort (1997-2002) to study the geology and resources, in addition to environmental aspects concerned with mining of the deposit. The Phosphoria Formation was named by Richards and Mansfield (1912) for sections they studied in southeast Idaho. It is our primary purpose here to trace the evolution of thought concerning the origin of this world-class phosphate-ore deposit, especially with respect to the depositional environment, nature of the Phosphoria sea, and conditions that promoted accumulation of huge quantities of phosphate. Although many of the early studies were solely descriptive works, they set the background for work that came later and therefore are briefly discussed here.
na
Western
'
Phosphate Field
.~!
9
~L
.,,.
"
,..
\, ~,.~.,,
,.~
-
..
Nevada
.t;
% ,~.- ..-~
J
9, / U,a.
/
i/Colorado
.
1.,I
Fig. 2-1. Outline of the Western Phosphate Field, a 350,000 km2 area in the northern Rocky Mountains; exposures of the Phosphoria Formation are indicated in black.
Origin of the Phosphoria Formation, Western US Phosphate Field
21
DELINEATION OF WESTERN PHOSPHATE LANDS: PRE-1940s USGS geologists started mapping the Western Phosphate Field in 1909, shortly after the first mining of phosphate began in 1908. This initial USGS field program was conducted between 1909 and 1916 and involved geologic mapping and chemical and paleontological analyses of collected rocks. The objective was to identify phosphate and nonphosphate lands in the Western Phosphate Field as mandated by the 1908 land withdrawal order of the Secretary of the Interior (see Jasinski et al., Chapter 3). That survey not only identified rocks rich in phosphate, but also found rocks with high concentrations of vanadium, nickel, and molybdenum. The second large-scale USGS effort was initiated in 1941 because of the vanadium-rich rocks discovered during the 1909-1916 survey. The earliest works that resulted from that first project were published in 1910 (Blackwelder, 1910; Gale, 1910; Gale and Richards, 1910; Girty, 1910; Richards and Mansfield, 1910) and provided field, petrographic, and paleontological analyses of Phosphoria Formation rocks in Idaho, Wyoming, Montana, and Utah. At that time, the Phosphoria Formation was considered to be part of the Park City Formation of Carboniferous age. The definitive and comprehensive work derived from the 1909-1916 field program was by Mansfield (1927), which addressed the entire geologic history of southeast Idaho. Another outcome of this early work in the Western Phosphate Field was assignment of names to strata in Idaho, Utah, and Wyoming, specifically the Phosphoria Formation, the Rex Chert Member of the Phosphoria Formation, and the Wells Formation, which underlies the Phosphoria Formation in Idaho (Richards and Mansfield, 1912). Although this 1912 paper was the first to publish those names, Gale selected the name for the Rex Chert after Rex Peak studied during his field work in the Crawford Mountains in Utah. The Phosphoria Formation, named after Phosphoria Gulch near Georgetown, Idaho, was considered to be coeval with the upper part of the Park City Formation, and therefore of Pennsylvanian age. By 1914, the Phosphoria was provisionally assigned a Permian age (Richards and Mansfield, 1914) and the name Phosphoria Formation was applied to rocks in Montana by 1916 (Pardee, 1917). Most early researchers correctly considered the bedded phosphorites of the Phosphoria to be of primary sedimentary origin. They, however, were divided into two camps: (a) those who thought that organisms played a major role and (b) those who thought that inorganic chemical and physical processes were dominant. Richards and Mansfield (1910) considered that the phosphorites formed predominantly by chemical and physical processes rather than by organisms. They stated that the CO2 content of the atmosphere was greater than it is at present (1910), but did not provide evidence. They also calculated that the Western Phosphate Field is the largest accumulation of phosphorite in the world. For a contrasting view, Richards and Mansfield referenced a G.A. Koenig (unpublished) who hypothesized that the phosphate was secreted by extremely prolific protozoa that accumulated rapidly in great quantities. Along those same lines, Breger (1911) reasoned that the bitumen and phosphate in the phosphorites and black shales shared a common origin. Breger envisioned a submarine ooze composed of microorganisms living on the seafloor, most likely bacteria, that extracted phosphate and calcium from seawater and deposited
22
J R . Hein et al.
calcium phosphate. He thought that distillation of that ooze upon burial produced the petroleum associated with the Western Phosphate Field. Blackwelder (1916) developed a model for global phosphorus cycling that started with formation of igneous phosphorus, weathering of that phosphorus and its delivery to soils, plants, animals, rivers, and eventually to the oceans where it is used by marine biota. He correctly outlined many of the basic chemical and biochemical processes that likely occurred in the Phosphoria sea during accumulation of Phosphoria phosphorites. He understood that phosphorus accumulates in tissue, bone, teeth, and shells of marine organism, but dismissed earlier ideas that large phosphate deposits could form by accumulation of the hard parts of marine organisms supplied during one or more episodes of mass mortality. This is curious, however, because many papers produced during the next half century ascribed to Blackwelder that mass mortality of organisms was the source of phosphorus for marine phosphorites, an idea that he rejected. He also rejected fecal pellets as being the sole source of phosphorus for such deposits. From marine biota, the phosphorus is transferred via birds to form guano deposits, or via the death of the biota and their sinking to the seafloor where they provided the organic matter that would release the phosphorus needed to form primary sedimentary phosphorites (Blackwelder, 1916). Blackwelder understood that the Phosphoria seafloor was suboxic or anoxic (he emphasized anoxic conditions) and that phosphorus, hydrocarbons, nitrogen, ammonia, and hydrogen sulfide were released during bacterial degradation of organic matter on the seafloor. He concluded that oxygen deficiency was the key factor that allowed accumulation of organic matter on the seafloor. He surmised that bottom-water circulation was weak and that semi-closed basins prevented ventilation of the oxygen-deficient bottom waters. The phosphoric acid produced by bacterial decay reacted primarily with carbonates in the presence of ammonia to produce calcium phosphates. Calcium phosphate precipitated as cement in near seafloor sediment, replaced some sediment grains, and formed pellets and ooids in some places. He believed that the phosphorus for even the largest phosphorite deposits could be supplied by normal seawater during a relatively short period of time. Blackwelder's observations established a reasonably accurate scenario for the origin of Phosphoria phosphates. What he did not address was the oceanographic environment in which large quantities or organic matter were produced and accumulated. A succinct presentation of that aspect would not come for another 20 years (see below; Kazakov, 1937). Pardee (1917) added another dimension by bringing climate into the picture. He suggested that development of biogenic (coralline-type) carbonates and the precipitation of calcium carbonate were inhibited by cold oceanic waters produced by glaciation. He believed that the Phosphoria sea (called Carboniferous sea) was of moderate depth and high productivity, which produced the organic-rich sediments associated with the Phosphoria. Decay of organic matter on the seafloor produced CO2, which further inhibited the formation of carbonates, but not phosphates. The phosphates precipitated from seawater at an ordinary rate, but because of the lack of carbonate deposition, accumulated in relatively pure form.
Origin of the Phosphoria Formation, Western US Phosphate Field
23
Mansfield (1918) added several ideas to the growing list based on field evidence, petrographic observations, earlier twentieth century work on the formation of aragonite ooids, and the earlier proposed ideas about the Phosphoria. He suggested that the Phosphoria sea was closed to the east, west, and south, but open to the north and northwest. Further, he proposed that an important factor in the deposition of the phosphorite was the lack of detrital input, which resulted from the low relief of the continental margin east of the Phosphoria sea. A condensed section formed by slow precipitation unaffected by the input of much terrigenous debris. Less insightful was his proposal that the Phosphoria ooids formed originally as aragonite ooids under the influence of denitrifying bacteria in an environment similar to the modern Bahamas, that is, shallow, warm waters. Those ooids were then replaced by phosphates when seawater cooled by a change in currents, or perhaps by Pardee's glaciation. He envisioned that productivity increased and organic matter accumulated and decayed on the seafloor via Blackwelder's mechanism. The temperature change was considered necessary to cause the death of great numbers of organisms. Mansfield (1927) later abandoned the idea of replaced aragonite ooids and agreed that the phosphate ooids in the Phosphoria were original precipitates. In a still later paper, Mansfield (193 l) proposed that the vast quantity of fluorine, estimated to be 5.4 x 108 metric tons in the phosphorites, was derived from considerable volcanic activity in the vicinity of the Phosphoria sea, an incorrect idea (see discussion of McKelvey et al., 1953), that he continued to develop (Mansfield, 1940). Condit et al. (1928) went a different direction and proposed a shallow-water restricted basin for much of the Phosphoria. They proposed a beach origin for the lowermost bed composed of abundant shell fragments and other bio-debris (the so-called fish-scale bed); a mudflat (occasionally subaerially exposed) origin for phosphorite pebbles; and evaporative basin origin for phosphorite desiccation conglomerates. They suggested that the organic matter in the Phosphoria was terrestrial plant debris. Decay of that plant debris created toxic bottom waters where hydrogen sulfide combined with carbon dioxide to produce a solvent for phosphatic skeletal debris, which then precipitated as acid phosphate salts on the seafloor. Branson (1930), based on regional fossil assemblages, found that the Phosphoria sea extended into Nevada and that the sea was open to the north and northwest based on similarities of Phosphoria and Alaskan Artinskian fauna. Based on lithologic correlations, he proposed that phosphate rocks in Alberta were deposited in a connecting sea. Further, he suggested that rocks in what is now known as the Permian basin in the southwest US were coeval in part with the Phosphoria Formation, but the faunas were different enough to have been deposited in different arms of the Pennsylvanian-Permian sea. An important piece of the puzzle missing from these early scenarios was the mechanism by which primary productivity was maintained. That mechanism is upwelling and was first articulated by Kazakov in 1937 at the International Geological Congress in Leningrad, when many of the salient conditions that promote the formation of marine phosphorites were proposed. Though he did not discuss the Phosphoria Formation per se, his ideas, with minor modifications, were applied to the Phosphoria Formation by McKelvey (1946b) and McKelvey et al. (1953) and are still accepted in large part today (see below).
24
J R . Hein et al.
GEOCHEMICAL EXPLORATION, P, U, AND V: THE 1940s-1950s Publications and new ideas about the Phosphoria in the 1940s were few. Keller (1941) provided excellent petrographic descriptions of the Rex Chert, but incorrectly surmised that the silica was derived from precipitation of silica gel on the seafloor. A second important USGS Phosphoria program (1941-1944)was part of the Strategic Minerals Program and involved detailed sampling and chemical analyses of Phosphoria rocks. That program, headed by W.W. Ruby, was undertaken to delineate vanadium-rich zones in the Phosphoria Formation, originally identified in the 1909-1916 project. The vanadium data were included in McKelvey's (1946a) PhD dissertation. That early 1940s survey not only delineated vanadiferous zones, but also identified zinc- and uranium-rich strata. This discovery of uranium, along with an increasing demand for phosphate, led to the third USGS program headed by McKelvey in 1947 (see below). Other works in the 1940s included studies by Gardner (1944) of regional thickness variations of the Phosphoria and the contribution of thrust-plate structures to those thickness variations. He showed that in places the Phosphoria has been tectonically thickened up to fivefold. Deiss (1949) was the first to describe discrete ore zones in the Meade Peak Phosphatic Shale Member (called phosphatic shale member in the 1940s) in southeast Idaho. He described a lower phosphate ore, middle medium- to low-grade ore (now called middle waste zone, which is not mined), and upper ore, the subdivision of the Mead Peak Member used today. Deiss also proposed an origin for carbonate nodules in the phosphatic shale member that involved formation from ground waters after deposition of the sediment and uplift of the rocks. It is now known that the nodules are early diagenetic (pre-compaction) deposits. In the 1950s, considerable progress was made in understanding the nature of the Phosphoria sea and adjacent areas of northwest Pangea (see Hein, Chapter 1). McKelvey et al. (1952, 1953) determined the basic oceanographic and sedimentary conditions for deposition of phosphorite of the Phosphoria Formation. They based their conclusions on the theory developed by Kazakov (1937, 1938), the basic tenets of which follow. Marine phosphorites are a direct result of the process of coastal upwelling. The phosphorites generally form during transgressions and in areas of little input of terrigenous debris (very low sedimentation rates). The phosphorites form on the shelf with the adjacent offshore basin having an open connection to the ocean. The thickness of the phosphorites and the amount of P205 increase seaward. Phosphate was precipitated at the seafloor generally by inorganic processes at water depths of 50-200 m. The pH and temperature of upwelled water increased along its path thereby promoting the precipitation of phosphate. McKelvey and coworkers modified these ideas to fit field and analytical data that they had collected for the Phosphoria. They thought that the water depth was more likely between 200 and 1000 m, that the Phosphoria basin was at least partly closed to the open ocean, and that phosphorites were deposited on three sides of the basin, as opposed to one side as suggested by Kazakov. Significantly, McKelvey and coworkers suggested that the phosphorus for the phosphorites was derived from the bacterial decay of planktonic organic matter produced in the zone of upwelling, which is an additional source of phosphorus from the
Origin of the Phosphoria Formation, Western US Phosphate Field
25
dissolved phosphate in the cold, upwelled waters. In addition, McKelvey and coworkers calculated the mass balance for phosphorus and determined that the Phosphoria Formation contains more than five times the amount of phosphorus dissolved in the modern ocean. They also calculated that the amount of phosphorus in the Phosphoria Formation is not unusual in the context of ancient continental-margin upwelling systems where sedimentation rates were high and the phosphorus was dispersed through a thick stratigraphic section of siliciclastic rocks. Finally, McKelvey and coworkers surmised that the large amounts of fluorine, chromium, vanadium, rare-earth elements, selenium, and other elements in the Phosphoria were derived from seawater, either directly through precipitation or sorption, or indirectly through alteration of biogenic material, and these conclusions are supported by the recent work of Piper (2001). Most of these ideas are accepted today, although the water depth of the Phosphoria sea is still a matter of discussion. Two seminal papers on the regional characteristics of the Phosphoria Formation and other Permian rocks in the western United States (McKelvey et al., 1956, 1959) subdivided the Permian rocks into the classification used today. These papers were based on the measurement and description of hundreds of sections from 200 locations in Montana, Wyoming, Idaho, and Utah. Field work was carried out during 1947 through 1952 and data were published in a series of USGS Circulars (208-211,260, 262, 301-307, and 324-327) in 1953 and 1954. These circulars contain a representative stratigraphic section from each region and chemical data for rocks through each measured section. This extensive mapping effort provided an understanding of the regional facies changes, thickness changes, and chemical and petrologic changes sufficient enough to define the basic characteristics of the Phosphoria sea and provide criteria for the correlation of rock units regionally (Fig. 2-2 and Table 2-I). McKelvey et al. (1956, 1959) divided the Permian rocks into the Park City, Phosphoria, and Shedhorn Sandstone Formations. The Park City Formation was subdivided into the Grandeur, Franson, and Ervay Members. TABLE 2-! Permian stratigraphy as proposed by McKelvey et al. (1956, 1959) Formation
Member
Shedhorn Sandstone
Upper Lower
Phosphoria
Tosi Chert Retort Phosphatic Shale Cherty Shale Rex Chert Meade Peak Phosphatic Shale Lower Chert
Park City
Ervay Carbonate Rock Franson Grandeur
Southern Idaho
Western Wyomlng
'
$a 9 C
g
P
-0,a,
a C
8
Central Wyoming m 5 a,
gF $9
,-a. 2
wz
V)
5 $E b aE
c m
9
+
1
V)
E2
c
?a,
--=
Southeastern Wyomlng
a,
U E
m
u
22
.g '5 c
E
02 E V)
E'
LL
Explanation
Phosphorla Formation of PermIan age
Park City Format~onand its equ~valentsof PermIan age
Chugwater Formation and equivalant of Permian
Vertical scale greatly exagerated
0
I 0
50 km I
150 km
I 50 miles Horizontal scale
Fig. 2-2. Composite section across the Phosphoria basin, modified from McKelvey et al. (1956, 1959).
I
100 miles
Origin of the Phosphoria Formation, Western US Phosphate Field
27
The Phosphoria Formation was subdivided into six members: (a) Lower Chert, (b) Meade Peak Phosphatic Shale, (c) Rex Chert, (d) Cherty Shale, (e) Retort Phosphatic Shale, and (f) Tosi Chert (Table 2-I). In places, some Phosphoria members are lateral equivalents, such as the Tosi Chert and the upper Retort, the Retort and the Cherty Shale, and the Lower Chert and the Meade Peak. Where these members interfinger, they are called tongues rather than members. Only the Meade Peak, Rex Chert, and Cherty Shale Members are present to any significant extent in southeast Idaho, the focus of this volume. The Shedhorn Sandstone Formation was subdivided into Upper and Lower Members. Regional facies changes in northwestern Wyoming and petrographic analyses led Sheldon (1957) to propose cyclic deposition of Phosphoria rocks starting with carbonate then chert then phosphorite and back to chert followed by carbonate. In addition, subcycles within the chert and phosphorite cycles were proposed. He related these cycles to two transgressive-regressive sea-level cycles where phosphorite was associated with transgression and carbonate with regression. Sheldon proposed that changes of seawater pH and Eh were associated with these cycles and reflected the characteristics of the extant upwelling regime. Sheldon incorrectly associated the transgressive-regressive sea-level changes predominantly with local tectonism rather than eustatic sea-level changes. He correctly determined that the silica for bedded and nodular Phosphoria cherts was derived from dissolution of biogenic silica, mostly sponge spicules. He also determined that the phosphorites are diagenetic deposits, but incorrectly thought that phosphatization was post-compaction. He suggested that the deeper basin was characterized by subdued bottomcurrent activity, an idea initially proposed by Blackwelder (1916) and accepted today. One other important conclusion published in the 1950s was that the source of sand in the Phosphoria Formation had eastern, western, and possibly northern sources, whereas the silt and clay were derived only from western and northwestern sources (Cressman, 1955). Cressman also correctly determined that the Phosphoria chert was primarily a product of the diagenesis of siliceous sponge spicules. Each of the three major USGS Phosphoria programs had components of geochemical exploration for phosphate, the latter two for vanadium, and the last one for uranium. The first program identified vanadium-rich strata, which led to the 1940s vanadium program that supported the WWII effort. That program in turn identified zinc- and uranium-rich zones that led directly to the work begun in 1947 that was partly funded by the US Atomic Energy Commission because of their interest in uranium resources.
DEPOSITIONAL ENVIRONMENTS AND TRANSGRESSIVE-REGRESSIVE CYCLES: THE 1960s-1970s The 1960s and 1970s saw a broadening of investigations of the Phosphoria Formation with the completion of several PhD and MA dissertations regarding these rocks. In addition, the USGS efforts continued with follow-up work to the program that was started in 1947. Based on detailed study of drill core and outcrop facies and thickness changes and petrography, Campbell (1962) described the eastern margin of the Phosphoria sea in the
28
J R . Hein et al.
region of the Big Horn Basin, western Wyoming. Permian rocks in the Big Horn Basin reflect two transgressive-regressive cycles. He described a broad carbonate-producing shelf that did not attain depths greater than about 10 m for a distance of up to 110 km offshore. Marine evaporite basins occurred to the east and still farther east were mudflats. The elongate evaporite basins extended into the continental margin and had physical barriers during the first cycle and dynamic barriers during the second. Campbell proposed that the net flow of surface water was toward the coast due to evaporation in near-shore areas and that dense saline waters sank and flowed offshore, perhaps dolomitizing the carbonate sediments on the shelf in the process. The climate was thought to be semi-arid to arid and subtropical to tropical. The coastal plain was nearly flat with no more than about 60 m relief that become more subdued with time. Flash floods and ephemeral rivers produced localized sand bodies and deltas. Phosphorites formed farther to the west, outside his study area, where temperatures of cold upwelling waters were < 15~ By the time these waters reached the shoreline, they may have been more than 30~ Sheldon (1963) expanded his earlier work on cyclic sedimentation of Permian rocks, with the two transgressive-regressive cycles each represented by 11 lithologic units. Distribution and thickness maps showed that the Phosphoria facies of the second cycle extended farther north and east than it did during the first cycle, although the first cycle was much thicker in the southern part of the region. Sheldon detailed how transgressive overlap and regressive offlap sequences manifested in vertical sequences and areal distributions, and how those related to energy and mineral resources (Fig. 2-3). He also described how the distributions of minerals such as pyrite, anhydrite, glauconite, fluorite, and apatite, as well as organic matter were determined by the transgressive-regressive cycles. Kazakov's (1937) upwelling ideas were supported by inferring dominant south-flowing currents through the
East
West Southeast Idaho
Central Wyoming
9
source beds
i reservoir beds
D sealing beds
sea level
o0e,,,n0
-
~~r162 ~.~--'~
,,
,oO ~ ~ " ~ Cher ^ . . .
~
1l
~ I
~ ~
1
Limestone
"
... - ~
Dolomite
i Algal
edbed
S~~
Anhydrite
Dolomite
ark Phosphatic Shale, Phosphorite and Dolomite P
V, U, Cr
Fossiliferous Pelletal .
Foss~hferous
"
7,--
I I
Fig. 2-3. Schematic pattern of sedimentation, currents, petroleum generation, and metal accumulation in the Phosphoria sea (modified from Sheldon, 1963).
Origin of the Phosphoria Formation, Western US Phosphate Field
29
Phosphoria sea and westward-directed winds, both essential in supporting coastal upwelling. This latter inference was derived primarily from the fact that little volcanic debris is found in the Phosphoria Formation even though coeval volcanism occurred to the west. He also inferred upwelling of cold water from the fact that corals are rare in Permian rocks of the Western Phosphate Field. The vertical zoning of the cycles indicates that the transgressions were more rapid that the regressions. Sheldon persisted in the idea that the sea-level changes were dominantly the result of regional tectonism. From an extensive study of Permian rocks in southwest Montana, Cressman and Swanson (1964) came to much the same conclusions as Sheldon (1963). Cressman and Swanson thought that upwelling in Meade Peak and Retort times was promoted by the submergence of a ridge located along the present Montana-Idaho border, which allowed free access of a southward-flowing current to the Phosphoria basin. An analysis of 1509 fossil collections from the Shedhorn, Phosphoria, and Park City Formations led Yochelson (1968) to many of the same conclusions that Cressman, Sheldon, and coworkers had made based on paleogeography, lithology, and physical stratigraphy. Yochelson did differ with the estimate of the maximum water depth for the Phosphorita sea, which he suggested was about 90 m rather than the 200-1000 m postulated by McKelvey et al. (1953). He also emphasized the low diversity of fauna in these rocks, a characteristic that is generally associated with environments that are less than optimal for life. He suggested that the Phosphoria sea had generally normal salinity throughout most of its history, but may have been hypersaline in some near-shore environments. Near-shore environments generally had quiet bottom waters, intermediate-depth environments had somewhat more vigorous bottom waters, and deepest-water environments again had quiet bottom waters. Perhaps most importantly, Yochelson correctly indicated that these formations were more restricted in time than formerly thought. They were likely deposited only during the Middle Permian, during the late Leonardian through the late Wordian. Previous papers had suggested that these rocks were deposited during half of the Permian or even the entire Permian. Yochelson (1968) mentioned that carbonates in the Rex Chert might represent bank deposits, an idea that was more fully developed by Brittenham (1976). Brittenham based his results on field studies and on the fossil identifications provided by Yochelson. He suggested that the banks developed on the outer-shelf margin during transgression when open circulation was established on the shelf. Carbonate banks started with the build-up of ramose bryozoa, and brachiopods in areas of greatest circulation. Crinoids and rugose corals were also common components of the banks in places. These small mounds developed into larger banks up to several kilometers long that comprised a variety of biofacies and sedimentary structures. Shallow (<30 m), inter-bank areas were muds composed of siliceous sponge spicules. During the final stage of bank development, foraminifers, algae, mollusks, and brachiopods inhabited carbonate shoals that were flanked by bars composed of winnowed sponge spicules (Brittenham, 1976). Murata et al. (1976) concluded that dolomite in the Phosphoria formed by bacterial degradation of organic matter in the diagenetic zone of bacterial sulfate reduction, based on carbon and oxygen isotopic analyses. This indicates that the sediment was anoxic, and
30
JR. Hein et al.
although it does not indicate the redox state of bottom water, they implied that it was also anoxic. They also showed that Park City dolomites (except Ervay) have isotopic compositions of normal marine carbonates, but a few of the easternmost occurrences show oxygen-isotope compositions indicative of an evaporative tidal-fiat or sabkha environment. Dolomite in the Ervay Member and Shedhorn Sandstone formed in near-shore environments by replacement of limestone and their oxygen isotope compositions reflect the influx of freshwater during their formation. Sheldon (1972) suggested that tongues of the Shedhorn Sandstone formed a complex of barrier islands that created a lagoon to the east. Some dolomite in the Ervay Member and Shedhorn Sandstone would have formed in such a lagoon. Wardlaw et al. (1979) proposed a northwest-southeast trending submarine high, located in present central Utah-northeastern Nevada, that divided the Phosphoria sea into northern and southern basins. A proposed western highland would have provided siliciclastic detritus to the Phosphoria sea. They based their findings on regional biostratigraphic studies of Permian conodonts and brachiopods. Using similar data, Wardlaw (1979) tracked the progress of the second-cycle transgression from northwest to southeast and east during which the Retort Phosphatic Shale Member was deposited. That marine transgression was time-transgressive through the Wordian (Guadalupian).
PALEOGEOGRAPHY, PHOSPHOGENESIS, AND SEQUENCE STRATIGRAPHY: 1980-2002 The 1980s brought a surge of new studies on both phosphorites and the Phosphoria Formation. Many of the studies were initiated in response to a renewed interest in petroleum source rocks and oil shale, while others followed new developments in geochemistry and stratigraphy. New ideas and refinements on old themes were offered in an attempt to address the key questions concerning the processes responsible for the concentration of phosphate and associated metals in the Phosphoria. Detailed and continued efforts by a wide range of workers provided an increasingly complete description of Permian paleogeography along the northwestern margin of Pangea as well as more sophisticated models of phosphate and carbonate deposition.
Paleogeography o f the Phosphoria basin Maughan (1979, 1980, 1984), Peterson (1980a, b, 1984), and Tisoncik (1984) interpreted the regional paleogeography of the Phosphoria basin. Maughan (1979, 1980) described the Phosphoria depositional environment as the foreslope of a foreland basin that he called the Sublett basin. His isopach maps showed that the basin encompassed about 400,000 km 2 and had a roughly northwest-southeast trend. Both Maughan and Peterson recognized that the remnant Antler highlands was the southwestern limit of the basin. Maughan (1980, 1983, 1984) also identified the Milk River uplift in Montana as the source
Origin of the Phosphoria Formation, Western US Phosphate Field
31
of the Shedhorn Sandstone along the northeast margin of the basin and suggested that the ancestral Uncompahgre uplift in present western Colorado and eastern Utah was the source of similar sands along the southeast margin of the basin. Peterson (1984) and Tisoncik (1984) emphasized paleostructural highs and lows such as the Bannock high and the Green River trough in explaining thinning of regionally extensive units. Both Maughan (1979, 1980, 1984) and Peterson (1980b, 1984) recognized the possible existence of an island-arc system along the northwest boundary of the basin. Darby et al. (2000) cited a correspondence in ages of zircons from Triassic strata of the Black Rock Terrane in northwest Nevada with those from basement provinces and off-shelf deposits further to the east as evidence for a volcanic-arc system proximal to the Wyoming shelf margin in Late Permian time. Wardlaw and Collinson (1986) described the Phosphoria basin as an interior sag basin and the eastern and southern margins of the basin as a carbonate ramp. This latter term denotes an overall slope angle of less than 1~ and the lack of a shelf-edge build-up or a prominent shelf-slope break (Ahr, 1973). The distinction between a conventional rimmed shelf and a ramp has important implications in terms of the sedimentary patterns that developed during sea-level changes, wave-energy dampening, sediment mixing, and the gradual warming of impinging bottom waters (Burchette and Wright, 1992; Hiatt, 1997; Hiatt and Budd, 2001). The previous recognition of a significant slope break near the Wyoming-Idaho border were based, in part, on rapid westward thickening of Paleozoic sediments. However, the degree of thickening may have been overestimated by early workers due to crustal thickening resulting from Sevier thrusting and the lack of accurate palinspastic reconstruction (Peterson, 1984). The regional occurrence of marker beds in the Phosphoria Formation (e.g. the 10 cm thick basal fish-scale bed) precludes large-scale paleorelief or shelf-slope break. Furthermore, basin dimensions constrain the overall slope angle to far less than 1~ (Hiatt, 1997). These lines of evidence led many workers in the 1990s to recognize the applicability of the ramp model to deposition of the Phosphoria and Park City Formations. In a study of the Park City Formation in Utah and Wyoming, Whalen (1996) suggested that the ramp varied from a distally steepened low-gradient ramp in the south to a homoclinal ramp with basins to the north. Wardlaw and Collinson (1986) suggested that phosphorite deposition occurred primarily in deep-ramp facies under low energy and slow terrestrial-sediment accumulation. They differentiated three coeval depositional lobes for the Meade Peak, located in northern Utah and Nevada, southeastern Idaho and western Wyoming, and southwestern Montana, whereas deposition of the Retort Phosphatic Shale Member was limited only to the two northern areas. Martindale (1986) identified a tongue of the Meade Peak in the Leach Mountains of Nevada along the southwest margin of the Phosphoria basin. He provided a range of fossil and petrologic evidence pointing to deposition in an intertidal to shallow-subtidal environment. Other workers (Clark, 1994; Whalen, 1996; Hiatt, 1997; Hendrix and Byers, 2000) provided evidence for phosphate formation throughout a wide range of depositional settings, from basinal or outer ramp to subtidal and peritidal environments. Paleoclimatic and atmospheric circulation computer modeling by Kutzbach and Ziegler (1993) and Golonka et al. (1994) indicate high mean surface temperatures (>125 ~ and
32
JR. Hein et al.
a prevailing northerly wind direction in the region of the Phosphoria basin during the Permian (see Hein, Chapter 1). Carroll et al. (1998) argued that terrigenous sediment in the Meade Peak Member was transported by northerly winds to the Phosphoria basin from present central Montana, citing as evidence well-sorted grain-size distributions, a scarcity of clay minerals, and the planar-parallel fabric of siltstone beds. These indications of a predominant northerly wind direction support oceanic upwelling via offshore Eckman transport of surface waters. Wardlaw (1980) and Wardlaw and Collinson (1984) suggested that upwelling water in the area of maximum phosphate deposition was relatively cold, based on the presence of presumed cold-water conodonts and brachiopods. Hiatt (1997) questioned that interpretation, citing a greater abundance of other species that have widespread occurrence in low-paleolatitude sections of western North American and the possibility that the initial associations of the specific fauna referenced by Wardlaw and Collinson with Arctic environments were based on inaccurate paleogeographic locations. Piper and Kolodny (1987) measured phosphate (PO 3 - ) 618OsMow values of between + 14.9 and + 17.5%0, from which they calculated water temperatures of 34-40~ similar to the summer surface-water temperatures for the Persian Gulf today. Hiatt and Budd (2001) reported an even larger range of phosphate 61SOsMow values, from 13.7 to 20.2%0, from which they calculated temperatures of 14-26~ for western sections, presumably representing temperate upwelling waters, and 34-42~ in eastern sections, representing warming of water across the shallow paleoshelf. Hiatt (1997) and Hiatt and Budd (2001) suggested that a drop in the oxygen-carrying capacity of impinging waters resulting from this warming helped maintain dysoxic conditions that promoted phosphogenesis in shallow-water environments.
Phosphogenesis
Several important papers, first presented at the 1978 International Congress on Sedimentology, were published by the Society of Economic Paleontologists and Mineralogists (SEPM) in 1980 in a collection entitled Marine Phosphorites- Geochemistry, Occurrence, Genesis. In this volume, Bentor (1980) attempted to discredit the hypothesis that phosphates were inorganically precipitated from bottom waters, arguing that direct precipitation was inconsistent with measurements of P concentrations in modern upwelling systems and carbonate fluorapatite (CFA) solubility. Bentor pointed out that plankton concentrate P from seawater about 250,000-fold, which can account for most of the phosphorus locked up in ancient phosphorites. He also suggested that phosphorite accumulation was largely dependent on the lack of dilution by other materials, especially carbonates, which are largely unstable in porewaters with a pH <7.5. In the same volume, Sheldon (1980) argued that the phosphorus content of the ocean varied during geologic time and that direct precipitation of phosphate from seawater had occurred at times. He cited the presence of extensive beds of oolitic CFA, up to a few tens of centimeters thick, and biostromes of phosphorite in the Phosphoria Formation as evidence for direct precipitation. Martindale (1986) also reported evidence for phosphate
Origin of the Phosphoria Formation, Western US Phosphate Field
33
precipitation at or just below the sediment-water interface that included phosphatic enrichment of rip-up clasts relative to their matrix, and oolitic packstones presumably formed by agitation at the surface and accompanying accretion of concentric layers in the ooids. Evidence for CFA precipitation from interstitial waters of organic matter-rich sediment as a predominant mechanism for formation of the phosphorite came from stable isotope studies. Nathan and Nielson (1980) reported sulfate t~34ScDTvalues of 15.0-19.7%o in CFA from the Meade Peak Member. Because their measured values were significantly heavier than those that define the established Permian seawater curve and coeval gypsum, they concluded that the apatite formed in a partly closed system in which sediments rich in organic matter underwent extensive bacterial sulfate reduction. A subsequent study by Kuniyoshi and Sakai (1987) found even greater 348 enrichment of CFA (t~34ScDT 18.2--27.3%0) relative to Permian seawater and evaporites from Retort Member samples. However, Piper and Kolodny (1987) measured t~34Svalues in Phosphoria CFA samples that are close to Permian seawater values of 12-15%o, suggesting major CFA precipitation at or very near the seafloor. CFA carbonate 613CpDBvalues of -3.1 to -10.8%o, values lying between Permian inorganic carbonate (+2%o) and organic carbon (-26%0) were reported for Phosphoria samples by McArthur et al. (1986), Piper and Kolodny (1987), and Kuniyoshi and Sakai (1987). These intermediate values suggest derivation of the carbonate from both seawater bicarbonate and decay of organic matter. Such interpretations are consistent with phosphate precipitation in modern environments (e.g. Jahnke et al., 1983; Froelich et al., 1988). By 1980, most workers generally accepted the idea of planktonic uptake and concentration of phosphorus in upwelling zones as the primary source of phosphorite phosphorus. However, many continued to insist that mechanisms other than upwelling must have been at work to augment primary productivity and concentrate phosphate and metals in the Phosphoria sea. Peterson (1980a, b, 1984) hypothesized that phosphate deposition occurred only during periods of sea-level lows (during regression and early transgression) when the basin experienced restricted circulation and increased salinity, leading to mass die-offs and anoxic bottom-water conditions. Maughan (1983, 1984, 1994) discussed previously proposed ideas regarding the concentration of phosphorus and various trace metals in the Phosphoria sea via reflux circulation from hypersaline lagoons to the east, volcanic inputs from the west, and other means for extraordinary metallic ion enrichment. Hite (1978), Stephens (1998), and Stephens and Carroll (1999) proposed that the evaporative basins of eastern Wyoming provided highly saline waters to the Phosphoria sea, leading to salinity stratification, increased nutrient levels, and preservation of organic residues and phosphate in sediments. Dahl et al. (1993) and Stephens and Carroll (1999) cited elevated concentrations of the biomarker gammacerane in Meade Peak organic matter as evidence for the high salinity. Gammacerane is a derivative of tetrahymenol, which is reportedly synthesized by marine ciliates that grow in hypersaline waters (Harvey and McManus, 1991). However, Piper and Link (2002) countered this idea, arguing that the 11 samples analyzed by Dahl et al. (1993) and Stephens and Carroll (1999) were collected from sites located at the margin of the Phosphoria basin and, furthermore, that their reported gammacerane to hopane ratios overlapped with values from samples known to
34
JR. Hein et al.
have accumulated under non-hypersaline conditions. Piper and Link (2002) argued that the presence of the broad, intervening carbonate sill should have restricted water exchange between the Phosphoria basin and the Goose Egg basin to the east. They also used massbalance models of phosphate, salt, and oxygen circulation to show that any exchange that occurred must have been minor. They argued, as did Calvert (1976), that the phosphate and metal contents of the Phosphoria resulted solely from upwelling and further estimated that the rate of upwelling and primary production in the Phosphoria sea was within the range found in modern oceans. Filippelli and Delaney (1992) also concluded that neither phosphorus accumulation nor burial rates in the Phosphoria sea were particularly anomalous in comparison to the Miocene Monterey Formation of California or the modem phosphogenic province of the Peruvian margin. Piper and Link (2002) concluded that the high concentrations of phosphate and metals in the Phosphoria Formation are principally the results of the low influx of diluting terrigenous sediments and the sorption or precipitation of metals under denitrifying bottom-water conditions (see Moyle and Piper, Chapter 21).
Eustatic changes and sequence stratigraphy
A key area of investigation in the 1980s and especially the 1990s was the relationship of eustatic sea-level changes to the cyclic deposition of units in the Phosphoria and intertonguing Park City Formations. Peterson (1980a,b) recognized three major depositional cycles in the Phosphoria and Park City Formations, each bounded by a regional disconformity. Each cycle is characterized, to varying degrees, by a basal phosphatic unit that grades upward into chert beds and finally fossiliferous shelf carbonates at the top. The bottom cycle consists solely of the Grandeur Member of the Park City Formation; the second cycle is represented by the Meade Peak and Rex Chert Members of the Phosphoria and the overlying Franson Member of the Park City Formation; and the third (upper) cycle includes the Retort and Tosi Members of the Phosphoria Formation and the Ervay Member of the Park City Formation. Wardlaw (1980) presented brachiopod and conodont evidence to divide the Phosphoria and Park City rocks into seven discrete biostratigraphic units of Middle and Late Permian age. He further delineated two major phases of transgression, assigning the Meade Peak Member to the period of maximum rise during the first transgressive phase in late Roadian time. The Rex Chert Member, lower Shedhorn Sandstone, and Franson Member were assigned to the early Wordian stage. Wardlaw (1980) suggested that these transgressive phases were eustatic because they appeared synchronous with other Permian transgressions on the North American continent. Sheldon (1980, 1984) attempted to link episodes of upwelling and phosphogenesis in the geologic past with perturbations in global ocean circulation. Lacking accurate age controis on Phosphoria deposition, he linked the accumulation of the Phosphoria phosphorite deposits with a transition from high sea-level, warm oceans to low sea-level, cold oceans associated with the onset of glaciation. Wardlaw and Collinson (1986) argued against glacially driven sea-level change based on a lack of evidence of glaciation in parts of
Origin of the Phosphoria Formation, Western US Phosphate Field
35
Antarctica, the presence of Middle and Late Permian coals throughout much of Gondwanaland, and the possibility that previous studies that suggested widespread Middle Permian glaciation may have been based on localized, physiographically driven glaciation events. Ross and Ross (1987a, b, 1994, 1995) provided crucial biostratigraphic correlations among globally distributed Late Paleozoic strata that defined a series of marine transgressiveregressive cycles occurring at intervals of 1-4 Ma and having typical sea-level fluctuations on the order of 10-200 m. The disconformities formed during the regressive phases of these third-order cycles define the fundamental sequence stratigraphic depositional units or sequences. Although sequence stratigraphy was derived largely from geophysical interpretations of the geometry of large-scale siliciclastic deposits (Mitchum et al., 1977; Vail et al., 1977; Van Wagoner et al., 1990), workers soon started to apply the basic concepts and terminology to carbonate systems. A series of dissertations from the early to middle 1990s dealt specifically with the sequence stratigraphy of the Phosphoria Formation and related units. Thornburg (1990) examined the sedimentology of the Shedhorn Sandstone; Whalen (1993, 1996) performed a regional analysis of facies patterns in the Park City Formation; Clark (1994) investigated the stratigraphy of the Ervay sequence that included the Retort and Tosi Members of the Phosphoria Formation and the Ervay Member of the Park City Formation; and Hiatt (1997) examined depositional cycles and geochemistry of the upper and lower tongues of the Phosphoria. There is now a broad consensus for a basic framework of three disconformity-bounded sequences in the inter-tonguing Phosphoria and Park City Formations (e.g. Inden and Coalson, 1996; Peterson, 1980a,b; Whalen, 1996). These Grandeur, Franson, and Ervay cycles are bounded by disconformities that may be either conformable or unconformable, depending on the stratigraphic position with respect to the basin margin. Although most workers recognized the lower Meade Peak and Retort Members as representing transgressive phases, there is less agreement among workers with respect to specification of systems tracts and maximum flooding surfaces within each stratigraphic sequence. This confusion may be partly due to higher-order (fourth and fiith order) transgressive-regressive cycles represented in these units. Whalen (1996), working in Utah and western Wyoming, interpreted the entire Meade Peak Member as representing an overall transgressive system tract and Franson carbonates as recording highstand conditions punctuated by several shoaling-upward cycles. Inden and Coalson (1996) found up to six higher order subcycles within the upper Ervay sequence in the Bighorn Basin. They attributed the Retort and Tosi units to deposition during transgressions and the carbonate units of the Ervay to deposition during highstand conditions. Hiatt (1997) concluded that the maximum flooding surfaces in the Franson and Ervay cycles occur at the contact between the Phosphatic Shale members (the Meade Peak and Retort, respectively) and the overlying Rex and Tosi Chert Members. He also suggested that the lower Meade Peak and Retort Members represent lowstand conditions and that significant transgression did not occur until well into deposition of the Meade Peak in basinal environments. His interpretation is based in part on a biostratigraphic change occurring in the lower half of the Meade Peak Member, which
36
J R . Hein et al.
is recognized as marking the transition from Cathedralian (Kungurian)- to Roadian-age rocks and the onset of a third-order eustatic change. Hendrix and Byers (2000), working in the Uinta Mountains of Utah, suggested that the Meade Peak Member represents transgressive and highstand tracts and placed the surface of maximum flooding within the middle of the unit, based on the presence of laminated organic-rich phosphatic dolomite in basinal sections and granular phosphorite associated with green shale in the shoreward sections. This is analogous to the interpretation of McKelvey et al. (1959) that the middle Meade Peak represented deepest water conditions. Hendrix and Byers (2000) called upon shoaling conditions for deposition of the Franson Member, leading to subaerial exposure. These various interpretations indicate that significant phosphogenesis may have occurred not only during transgressive phases, but also during highstand and late lowstand stages. The precipitation of phosphate under varying eustatic conditions and in a wide range of environments suggests that phosphogenesis is the result of a complex interaction of different processes and that no single model for phosphogenesis is likely applicable to shelf-type phosphorites in general or to the Phosphoria Formation as a whole.
OUTSTANDING ISSUES A substantial understanding of the origin and evolution of the Phosphoria Formation has been developed over the past 100 years. However, key outstanding issues await further results, a few of which are listed here. The absolute ages of the Permian time intervals during which the Phosphoria Formation and its various members were deposited are imprecisely known. This may be resolved by analyzing the crown of unaltered conodonts for their strontium isotope composition. Those strontium-isotope records can be compared with the very steep strontium-isotope seawater curve for the Permian, which should pinpoint the absolute ages of the Phosphoria members and, therefore, the amount of time represented by each unit. (2) The relationships of tectonics to various phases of mineralization are still not well understood (see Grauch et al., Chapter 8). This may be resolved by dating diagenetic, authigenic, hydrothermal, and metamorphic minerals by radiometric and various other techniques combined with interpretation of replacement and cross-cutting features. (3) Related to point (2) is the unknown association of many elements with mineral and other phases and the redox forms of those elements that were mobilized and reprecipitated during various periods in the history of the Phosphoria Formation. These relationships can be resolved by synchrotron radiation (XANES, EXAFS) and similar techniques, microprobe studies, and sequential-leaching procedures (see Perkins and Foster, Chapter 10). (4) There is still little evidence as to the mechanisms of CFA pellet and ooid growth, their concentration on the seafloor, and the time frame of growth. Given the wide range of environments in which pellets and ooids formed within the Phosphoria, it is doubtful
(1)
Origin of the Phosphoria Formation, Western US Phosphate Field
(5)
(6)
(7)
37
that any one model will be completely inclusive, but it must consider combinations of agitation of bottom waters and sediment, fluctuating redox boundaries, and winnowing and transport. These issues may be addressed by measuring stable-isotope ratios, REEs, and trace elements across individual grains. The cause for relatively low trace-element contents in phosphorites in deeper-water facies of the Phosphoria has not been established. Did varying plankton populations and changes in the locus of upwelling play important roles? Outstanding oceanographic and paleogeographic questions include the pattern of water circulation in the Phosphoria sea and the nature of the western boundary of the Phosphoria sea. Did a volcanic arc exist there and if so, how far to the west was it located? The big picture concerning sequence stratigraphy has been established, but there are still lingering questions about where the highstand horizon is l o c a t e d - the middle or top of the Meade Peak Member? Also, should the lower Meade Peak be considered part of the transgressive or lowstand tract? Higher (fourth and fifth) order fluctuations and parasequences need to be identified from different parts of the Phosphoria basin.
REFERENCES Ahr, W.M., 1973. Carbonate ramp; alternative to the shelf model. Trans. Gulf Coast Assoc. Geol. Soc., 23:221-225. Bentor, Y.K., 1980. Phosphorites - the unsolved problem. In: Y.K. Bentor (ed.), Marine Phosphorites- Geochemistry, Occurrence, and Genesis. Society of Economic Paleontologists and Mineralogists Special Publication 29, Tulsa, OK, pp. 3-18. Blackwelder, E., 1910. A reconnaissance of the phosphate deposits in western Wyoming. US Geological Survey Contributions to Economic Geology 470, pp. 452-483. Blackwelder, E., 1916. The geologic role of phosphorus. Am. J. Sci., XLII(250): 285-298. Branson, C.C., 1930. Paleontology and stratigraphy of the Phosphoria Formation. The University of Missouri Studies: A Quarterly of Research, vol. V, no. 2, Columbia, 99 pp. Breger, C.L., 1911. Origin of Lander oil and western phosphate. Mining Engi. World, 35:631-633. Brittenham, M.D., 1976. Permian Phosphoria carbonate banks, Idaho-Wyoming thrust belt. In: Geology of the Cordilleran Hingeline - A Symposium. Rocky Mountain Association of Geologists, pp. 173-191. Burchette, T.P. and Wright, V.P., 1992. Carbonate ramp depositional systems. Sediment. Geol., 79: 3-57. Calvert, S.E., 1976. The mineralogy and geochemistry of near-shore sediments. In: J.P.Riley and R. Chestor (eds.), Chemical Oceanography. Academic Press, London, pp. 187-280. Campbell, C.V., 1962. Depositional environments of Phosphoria Formation (Permian) in southeastern Bighorn Basin, Wyoming. Bull. Am. Assoc. Pet. Geologists, 46: 478-503. Carroll, A.R., Stephens, N.P., Hendrix, M.S. and Glenn, C.R., 1998. Eolian-derived siltstone in the Upper Permian Phosphoria Formation: implications for marine upwelling. Geology, 26: 1023-1026. Clark, W.J., 1994. The Ervay cycle of the Permian Phosphoria and Park City formations, Bighorn Basin, Wyoming; sequence stratigraphy, facies, and porosity trends. PhD Dissertation, University of Colorado, Boulder CO, 606 pp.
38
J R . Hein et al.
Condit, D.D., Finch, E.H. and Pardee, J.T., 1928. Phosphate rock in the Three Forks-Yellowstone Park region, Montana. US Geological Survey, Bulletin, 795, pp. 147-209. Cressman, E.R., 1955. Physical stratigraphy of the Phosphoria Formation in part of southwestern Montana. US Geological Survey, Bulletin, 1027-A, 31 pp. Cressman, E.R. and Swanson, R.W., 1964. Stratigraphy and petrology of the Permian rocks of southwestern Montana. US Geological Survey, Professional Paper, 313-C, 569 pp. Dahl, J., Moldowan, J.M. and Sundaraman, P., 1993. Relationship of biomarker distribution to depositional environment: Phosphoria Formation, Montana, USA. Organic Geochem., 20: 1001-1017. Darby B.J., Wyld, S.J. and Gehrels, G.E., 2000. Provenance and paleogeography of the Black Rock Terrane, northwestern Nevada; implications of U-Pb detrital zircon geochronology. In: M.J. Soreghan and G.E. Gehrels (eds.), Paleozoic and Triassic Paleogeography and Tectonics of Western Nevada and Northern California. Geol. Soc. Am. Spec. Pap., pp. 77-87. Deiss, C., 1949. Phosphate deposits of the Deer Creek-Well Canyon area Caribou County, Idaho. US Geological Survey, Bulletin, 955-C, 101 pp. Filippelli, G.M. and Delaney, M.L., 1992. Similar phosphorus fluxes in ancient phosphorite deposits and a modern phosphogenic environment. Geology, 20:709-712. Froelich, P.N., Arthur, M.A., Burnette, W.C., Deakin, M., Hensley, V., Jahnke, R.A., Kaul, L., Kim, K.K., Roe, K., Soutar, A. and Vathakanon, C., 1988. Early diagenesis of organic matter in Peru continental margin sediments: phosphorite precipitation. Mar. Geol., 80: 309-343. Gale, H.S., 1910. Rock phosphate near Melrose, Montana. US Geological Survey Contributions to Economic Geology 470, pp. 440-451. Gale, H.S. and Richards, R.W., 1910. Preliminary report on the phosphate deposits in southeastern Idaho and adjacent parts of Wyoming and Utah. US Geological Survey, Bulletin, 430, pp. 441-447. Gardner, L.S., 1944. Phosphate deposits of the Teton basin area, Idaho and Wyoming. US Geological Survey, Bulletin, 9444-A, 36 pp. Girty, G.H., 1910. The fauna of the phosphate beds of the Park City formation of Idaho, Utah, and Wyoming. US Geological Survey, Bulletin, 436, 82 pp. Golonka, J., Ross, M.I. and Scotese, C.R., 1994. Phanerozoic paleogeographic and paleoclimatic modeling maps. In: A.E Embry, B. Beauchamp and D.J. Glass (eds.), Pangea: Global Environments and Resources. Canadian Society of Petroleum Geologists, Calgary, Memoir 17, pp. 1-47. Harvey, H.R. and McManus, G.B., 1991. Marine ciliates as a widespread source of tetrahymanol and hopan-3B-ol in sediments. Geochim. Cosmochim. Acta, 55: 3387-3390. Hendrix, M.S. and Byers, C.W., 2000. Stratigraphy and sedimentology of Permian strata, Uinta Mountains, Utah; allostratigraphic controls on the accumulation of economic phosphate. In: C.R.Glenn, L. Pr6v6t-Lucas, and J. Lucas (eds.), Marine Authigenesis: From Global to Microbial. Society for Sedimentary Geology (SEPM), Denver, Special Publication 66, pp. 349-367. Hiatt, E.E., 1997. A paleoceanographic model for oceanic upwelling in a late Paleozoic epicontinental sea: a chemostratigraphic analysis of the Permian Phosphoria Formation. PhD dissertation, University of Colorado, Boulder, 294 pp. Hiatt, E.E. and Budd, D.A., 2001. Sedimentary phosphate formation in warm shallow waters: new insights into the palaeoceanography of the Permian Phosphoria Sea from analysis of phosphate oxygen isotopes. Sediment. Geol., 145:119-133. Hite, R.J., 1978. Possible genetic relationships between evaporites, phosphorites and iron-rich sediments. Mountain Geologist, 15: 97-107.
Origin of the Phosphoria Formation, Western US Phosphate Field
39
Inden, R.E and Coalson, E.B., 1996. Phosphoria Formation (Permian) cycles in the Bighorn Basin, Wyoming, with emphasis on the Ervay Member. In: M.W Longman and M.D. Sonnenfeld (eds.), Paleozoic Systems of the Rocky Mountain Region. Rocky Mountain Region of the Society of Economic Paleontologists and Mineralogists, Denver, pp. 379-404. Jahnke, R.A., Emerson, S.R., Roe, K.K., and Burnett, W.C., 1983. The present day formation of apatite in Mexican continental margin sediments. Geochim. Cosmochim. Acta, 47: 259-266. Kazakov, A.V., 1937. The phosphorite facies and the genesis of phosphorites. In: Geological Investigations of Agricultural Ores. Transactions of the Science Institute of Fertilizers and Insecto-Fungicides 142 (publication for the 17th Session of the International Geological Congress, Leningrad), pp. 95-113. Kazakov, A.V., 1938. The phosphorite facies and the genesis of natural phosphorites. Sovetskaya Geologiya, 8: 33-47. Keller, W.D., 1941. Petrography and origin of the Rex Chert. Geol. Soc. Am. Bull., 52: 1279-1298. Kuniyoshi, S. and Sakai, H., 1987. Sulfur and carbon isotopes in phosphorites of the Retort Phosphatic Shale Member of the Phosphoria Formation, Hawley Creek Area, Idaho. US Geological Survey, Open-File Report, 87-0100. Kutzbach, J.E. and Ziegler, A.M., 1993. Simulation of Late Permian climate and biomes with an atmosphere-ocean model: comparison with observations. Royal Soc. Lond. Philos. Trans. B, 341: 327-340. Mansfield, G.R., 1918. Origin of the western phosphates of the United States. Am. J. Sci., XLVI(274): 591-598. Mansfield, G.R., 1927. Geography, geology, and mineral resources of part of southeastern Idaho. US Geological Survey, Professional Paper, 152, 453 pp. Mansfield, G.R., 1931. Some problems of the Rocky Mountain phosphate field. Econ. Geol., XXVI: 353-374. Mansfield, G.R., 1940. The role of fluorine in phosphate deposition. Am. J. Sci., 238: 863-879. Martindale, S.G., 1986. Depositional environments and phosphatization of the Meade Peak Phosphatic Shale Tongue of the Phosphoria Formation, Leach Mountains, Nevada. Contributions to Geology, University of Wyoming, 24: 143-156. Maughan, E.K., 1979. Petroleum source rock evaluation of the Permian Park City Group in the northeastern Great Basin, Utah, Nevada, and Idaho. In: Great Basin Guidebook, Rocky Mountain Association of Geologists, Utah Geological Association, pp. 523-530. Maughan, E.K., 1980. Relation of phosphorite, organic carbon, and hydrocarbons in the Permian Phosphoria Formation, western United States of America. In: Geologie Comparee des Gisements de Phosphates et de Petrole, Orleans, pp. 64-91. Maughan, E.K., 1983. Geological setting and geochemistry of oil shales in the Permian Phosphoria Formation. In: EP. Miknis and J.E McKay (eds.), Geochemistry and Chemistry of Oil Shales. American Chemical Society, Washington, pp. 199-224. Maughan, E.K., 1984. Geological setting and some geochemistry of petroleum source rocks in the Permian Phosphoria Formation. In: J. Woodward, EE Meissner, and J.L. Clayton (eds.), Hydrocarbon Source Rocks of the Greater Rocky Mountain Region. Rocky Mountain Association of Geologists, Denver, pp. 281-294. Maughan, E.K., 1994. Phosphoria Formation (Permian) and its resource significance in the Western Interior, USA. In: A.E Embry, B. Beauchamp, and D.J. Glass (eds.), Pangea: Global Environments and Resources. Canadian Society of Petroleum Geologists, Calgary, pp. 479-495.
40
J R . Hein et al.
McArthur, J.M., Benmore, R.A., Coleman, M.L., Soldi, C., Yeh, H.W. and O'Brien, G.W., 1986. Stable isotopic characterization of francolite formation. Earth Planet. Sci. Lett., 77: 20-34. McKelvey, V.E., 1946a. Stratigraphy of the Phosphatic Shale Member of the Phosphoria Formation in western Wyoming, southeastern Idaho, and northern Utah. PhD dissertation, University of Wisconsin, Madison, WI, 152 pp. McKelvey, V.E., 1946b. Stratigraphy of the Phosphatic Shale Member of the Phosphoria Formation in western Wyoming, southeastern Idaho, and northern Utah. US Geological Survey, Open-File Report, 162 pp. McKelvey, V.E., Swanson, R.W. and Sheldon, R.P., 1952. The Permian phosphorite deposits of western United States. US Geological Survey, Open-File Report, 176, 30 pp. McKelvey, V.E., Swanson, R.W. and Sheldon, R.P., 1953. The Permian phosphorite deposits of western United States. Comptes Rendus, 19th International Geological Congress, section XI, pp. 45-64. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M. and Swanson, R.W., 1956. Summary description of Phosphoria, Park City, and Shedhorn Formations in western phosphate field. Am. Assoc. Pet. Geol. Bull., 40: 2826-2863. McKelvey, V.E., Williams, J.S., Sheldon, R.P, Cressman, E.R., Cheney, T.M., and Swanson, R.W, 1959. The Phosphoria, Park City and Shedhorn Formations in western phosphate field. US Geological Survey, Professional Paper, 313-A, 47 pp. Mitchum, R.M., Vail, P.R. and Thompson, S., 1977. Seismic stratigraphy and global changes in sea level, Part 2: The depositional sequence as a basic unit for stratigraphic analyses. In: C.E. Payton (ed.), Seismic Stratigraphy: Applications to Hydrocarbon Exploration. American Association of Petroleum Geologists Memoir 26, Tulsa, OK, pp. 53-62. Murata, K.J., Friedman, I. and Gulbrandsen, R.A., 1976. Geochemistry of carbonate rocks in Phosphoria and related formations of the wcstern phosphate field. US Geological Survey, Professional Paper, 800-D, pp. D 103-D 110. Nathan, Y. and Nielsen, H., 1980. Sulfur isotopes in phosphorites. In: Y. K. Bentor (ed.), Marine Phosphorites- Geochemistry, Occurrence, Genesis. Society of Economic Paleontologists and Mineralogists Special Publication 29, Tulsa, OK, pp. 73-78. Pardee, J.T., 1917. The Garrison and Philipsburg phosphate fields, Montana. US Geological Survey, Bulletin, 640, Contributions to Economic Geology 1916, Part 1, pp. 195-228. Peterson, J.A., 1980a. Depositional history and petroleum geology of the Permian Phosphoria, Park City, and Shedhorn formations, Wyoming and southeastern Idaho. US Geological Survey, Open File Report, 80-667, 43 pp. Peterson, J.A., 1980b. Permian paleogeography and sedimentary provinces, west central United States. In: Paleozoic Paleoceanography of West-Central United States Symposium. Soc. Econ. Paleo. Min., Denver CO, pp. 271-292. Peterson, J.A., 1984. Permian stratigraphy, sedimentary facies, and petroleum geology, Wyoming and adjacent area. In: J. Goolsby and D. Morton (eds.), The Permian and Pennsylvanian Geology of Wyoming. Wyoming Geological Association Guidebook 35, pp. 2 5 - 6 4 . Piper, D.Z. and Kolodny, Y., 1987. The stable isotopic composition of a phosphorite deposit, 8 13C, 34S, and 8 180. Deep-Sea Res. Part A: Oceanogr. Res. Pap., 34:897-911. Piper, D.Z., 2001. Marine chemistry of the Permian Phosphoria Formation and basin, southeast Idaho. Econ. Geol., 96: 599-620. Piper, D.Z. and Link, P.K., 2002. An upwelling model for the Phosphoria Sea - a Permian, oceanmargin basin in the northwest United States. Am. Assoc. Pet. Geol. Bull., 86:1217-1235. Richards, R.W. and Mansfield, G.R., 1910. Preliminary report on a portion of the Idaho Phosphate Reserve. US Geological Survey Contributions to Economic Geology 470, pp. 371-439.
Origin of the Phosphoria Formation, Western US Phosphate Field
41
Richards, R.W. and Mansfield, G.R., 1912. The Bannock overthrust. J. Geol., 20: 683-689. Richards, R.W. and Mansfield, G.R., 1914. Geology of the phosphate deposits northeast of Georgetown, Idaho. US Geological Survey, Bulletin, 577, 76 pp. Ross, C.A. and Ross, J.R.P., 1987a. Biostratigraphic zonation of Late Paleozoic depositional sequences. In: C.A. Ross and D. Homan (eds.), Timing and Depositional History of Eustatic Sequences: Constraints on Seismic Stratigraphy. Cushman Foundation for Foraminiferal Research Special Publication 24, Ithaca, NY, pp. 151-168. Ross, C.A. and Ross, J.R.P., 1987b. Late Paleozoic accreted terranes of western North America. In: C.H. Stevens (ed.), Pre-Jurassic Rocks in Western North American Suspect Terranes. Pacific Section of Society of Economic Paleontologists and Mineralogists, Los Angeles, pp. 7-22. Ross, C.A. and Ross, J.R.P., 1994. Permian sequence stratigraphy and fossil zonation, In: B. Beauchamp, A.E Embry and D.J. Glass (eds.), Pangea: Global Environments and Resources. Canadian Society of Petroleum Geologists Memoir 17, Calgary, pp. 219-231. Ross, C.A. and Ross, J.R.P., 1995. Permian sequence stratigraphy. In: P.A. Scholle, T.M. Peryt and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea: vol. I, Paleogeography, Paleoclimates, Stratigraphy. Springer-Verlag, Berlin, pp. 98-123. Sheldon, R.P., 1957. Physical stratigraphy of the Phosphoria Formation in northwestern Wyoming. US Geological Survey Bulletin, 1042-E, 185 pp. Sheldon, R.P., 1963. Physical stratigraphy and mineral resources of Permian rocks in western Wyoming. US Geological Survey, Professional Paper, 313-B, 273 pp. Sheldon, R.P., 1972. Phosphate deposition seaward of barrier islands at edge of Phosphoria sea in northwestern Wyoming. Abstract, Am. Assoc. Pet. Geol. Bull., 56: 653. Sheldon, R.P., 1980. Episodicity of phosphate deposition and deep ocean circulation- a hypothesis. In: Y. K. Bentor (ed.), Marine Phosphorites - Geochemistry, Occurrence, and Genesis: Society of Economic Paleontologists and Mineralogists Special Publication 29, Tulsa, OK, pp. 239-247. Sheldon, R.P., 1984. Polar glacial control on sedimentation of Permian phosphorites of the Rocky Mountains, USA. In: Proceedings of the 27th International Geological Congress, Moscow, pp. 223-243. Stephens, N.P., 1998. Salinity stratification and upwelling during organic carbon deposition, and related stratigraphy in Permian Phosphoria Formation, Utah, Wyoming, and Idaho, USA. Masters Dissertation, University of Wisconsin, Madison, 181 pp. Stephens, N.P. and Carroll, A.R., 1999. Salinity stratification in the Permian Phosphoria Sea: a proposed paleoceanographic model. Geology, 27: 899-902. Thornburg, J.M.B., 1990. Petrography and sedimentology of a phosphatic shelf deposit: the Permian Shedhorn Sandstone and associated rocks in southwest Montana and northwest Yellowstone National Park. PhD Dissertation, University of Colorado, Boulder, 719 pp. Tisoncik, D.D., 1984. Regional lithostratigraphy of the Phosphoria Formation in the Overthrust Belt of Wyoming, Utah and Idaho. In: J. Woodward, EE Meissner and J.L. Clayton (eds.), Hydrocarbon Source Rocks of the Greater Rocky Mountain Region. Rocky Mountain Association of Geologists, Denver CO, pp. 295-320. Vail, P.R., Mitchum, R.M. and Thompson, S., 1977. Global cycles of relative changes in sea level. In: C.E. Payton (ed.), American Association of Petroleum Geologists Memoir 26, Tulsa, OK, pp. 83-97. Van Wagoner, J.C., Mitchum, R.M., Campton, K.M. and Rahmanian, V.D., 1990. Siliciclastic sequence stratigraphy in well logs, cores, and outcrops: concepts for high-resolution correlation of time and facies. Methods in Exploration, vol. 7. Am. Assoc. Pet. Geol., Tulsa, OK, 55 pp.
42
J.R. Hein et al.
Wardlaw, B.R., 1979. Transgression of the Retort Phosphatic Shale Member of the Phosphoria Formation (Permian) in Idaho, Montana, Utah, and Wyoming. US Geological Survey, Professional Paper, 1163-A, pp. 1-4. Wardlaw, B.R., 1980. Middle-Late Permian paleogeography of Idaho, Montana, Nevada, Utah, and Wyoming. In: Paleozoic Paleoceanography of West-Central United States Symposium. Society of Economic Paleontologists and Mineralogists, Denver, CO, pp. 353-361. Wardlaw, B.R., Collinson, J.W. and Maughan, E.K., 1979. Stratigraphy of Park City Group equivalents (Permian) in southern Idaho, northeastern Nevada, and northwestern Utah. US Geological Survey, Professional Paper, 1163-C, pp. 9-16. Wardlaw, B.R. and Collinson, J.W., 1984. Conodont paleoecology of the Permian Phosphoria Formation and related rocks of Wyoming and adjacent areas. In: D.L. Clark (ed.), Conodont Biofacies and Provincialism. Geological Society of America Special Paper 196, pp. 263-281. Wardlaw, B.R. and Collinson, J.W., 1986. Paleontology and deposition of the Phosphoria Formation. In: D.W. Boyd and J. A. Lillegraven (eds.), Western Phosphate Deposits. Contributions to Geology, University of Wyoming, Laramie, pp. 107-142. Wardlaw, B.R., Snyder, W.S., Spinosa, C. and Gallegos, D.M., 1995. Permian of the Western United States. In: P.A. Scholle, T.M. Peryt, and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea, vol. II, Sedimentary Basins and Economic Resources: Springer-Verlag, Berlin, pp. 23-40. Whalen, M.T., 1993. Depositional and diagenetic history of the Park City Formation (Permian) carbonate ramp, northwestern USA: implications for eastern ocean basin carbonate platform development. PhD dissertation, Syracuse University, 270 pp. Whalen, M.T., 1996. Facies architecture of the Permian Park City Formation, Utah and Wyoming: implications for the paleogeography and oceanographic setting of western Pangea. In: M. W. Longman and M.D. Sonnenfeld (eds.), Paleozoic Systems of the Rocky Mountain Region. Society for Sedimentary Geology (SEPM), Denver, CO, pp. 355-378. Yochelson, E.L., 1968. Biostratigraphy of the Phosphoria, Park City, and Shedhorn Formations. US Geological Survey, Professional Paper, 313-D, pp. 571-660.
PART II.
R E G I O N A L STUDIES
This Page Intentionally Left Blank
Li/e Cycle of the Pho~phoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exph)ration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
45
Chapter 3
THE HISTORY OF P R O D U C T I O N OF THE W E S T E R N P H O S P H A T E FIELD
S.M. JASINSKI, W.H. LEE and J.D. CAUSEY
ABSTRACT The Western Phosphate Field of the United States contains one of the largest resources of phosphate rock in the world and it has been mined for nearly a century. Since the opening of the first mine in 1906, 229 million metric tons of marketable phosphate rock have been produced from 70 mines in the four states that comprise the Western Phosphate Field. Of these 70 mines, 49 were underground, 17 were surface, and seven used both methods; however, since 1993, all production has been from surface mines. Early scientific studies and changes in US mining laws have contributed to the exploration and development of this valuable resource. Cumulative production from the Western Phosphate Field represented 12% of the total phosphate rock produced in the United States and its end uses are evenly divided between fertilizer and industrial applications. Idaho has been the most significant producing state followed by Montana, Utah, and Wyoming. Currently, mining occurs only in Idaho and Utah at an average rate of 5 million tons per year.
INTRODUCTION Phosphorite deposits occur in the Permian marine Phosphoria Formation in Idaho, Montana, Utah and Wyoming (Service and Popoff, 1964). The deposits are found over an area of 350,000 km 2 and represent one of the largest available resources of phosphate rock in the world (Fig. 3-1). Phosphate has been mined and processed in the Western Phosphate Field for nearly 100 years. Discovered in Cache County, Utah, in 1889, exploration and development of phosphate commenced in 1904. Production in Idaho was first reported in 1906 from a mine near Montpelier. Production in Wyoming and Utah was first recorded in 1907 and production in Montana began in 1929. However, significant commercial development did not start until around 1950. Phosphate production rates rose slowly during the first half of the twentieth century. Mining was predominantly by underground methods until the 1940s when the transition to open-pit mining in Idaho and Wyoming significantly increased production rates and reduced unit cost. Montana and Utah experienced steady growth similar to that in Idaho
46
S.M. Jasinsla', W.H. Lee and J.D. Causey
120~
115~
I
I
110~ 0"W
105~ 0"W
I
100~ 0"W
I
I
--.__..__
Montana 45~
45o0'0"N Idaho /~ I B ~ a h
yoming
Cheyenne 40~
/ Utah
I
115~ 0"W
Denver
40o0'0,,N
Colorado
I
110~
I
105~ 0"W
Fig. 3-1. Extent of the Western Phosphate Field where mining has taken place.
from 1950 to 1968, but in contrast, deposits in those states were extracted by underground mining. By the end of the 1960s, production in the region dropped and many mines closed because of rising mining costs and competition from lower-cost producers in Florida and North Carolina. Western United States phosphate production increased from about 716,000 metric tons, or 7.5% of US production in 1948, to a high, in terms of tonnage, of about seven million metric tons, or 17.2% of US production in 1994 (Fig. 3-2). Data presented in Fig. 3-2 was collected through confidential surveys of the producers. Phosphate-rock production is separated into two categories; crude ore, which represents the mined product prior to beneficiation, and marketable, which is the final phosphate-rock product. The amount of crude ore versus marketable product is relatively low for the western United States because phosphate rock used in elemental phosphorus production, which is predominately in the west, is used in its mined form, whereas phosphate rock used for phosphoric acid is concentrated by flotation and/or calcining to increase the P205 content. From 1948 to 2000, the use of phosphate rock from the Western Phosphate Field was equally distributed between elemental phosphorus and fertilizer use. This includes exports, which primarily went to Canada for fertilizer (Fig. 3-3).
The history of production of the Western Phosphate Field
47
Fig. 3-2. US production of marketable phosphate rock.
Fig. 3-3. Western phosphate rock use distribution, 1948-2000: (A) summary distribution; (B) annual distribution.
48
S.M. Jasins~', W.H. Lee and J.D. Causey
Idaho has been the major producer in the region, followed by Montana, Utah, and Wyoming. Currently, only four mines in Idaho and one in Utah are operating. Production ceased in Wyoming in 1978 due to depletion of reserves and in Montana in 1993 because of the high cost of underground mining. Combined annual production of the five open-pit mines currently operating in the Western Phosphate Field ranges from five to six million metric tons of phosphate rock, about 12-14% of total US production. All five active mines in the region recover thick, high-grade units from the Meade Peak Phosphatic Shale Member of the Phosphoria Formation. The Western Phosphate Field, with more than 30% of domestic reserves, will likely continue to be a significant producer of phosphate rock for many decades. The last operating elemental phosphorus plant in North America is located in Soda Springs, Idaho (Vranes, 1999). High electricity costs, environmental problems, and competition from purified wet process acid have led to the closure of other elemental phosphorus plants.
EARLY SCIENTIFIC SURVEYS The first coordinated, scientific expedition, the Geological and Geographical Survey of the Territories, visited southeastern Idaho during 1871-1877. Ferdinand Vandeveer Hayden led a group of geologists, paleontologists, mineralogists, topographers, artists, and photographers, popularly known as the Hayden Survey, in exploring, mapping, and documenting this part of the West. One of the geologists, A.C. Peale, documented his many findings of the geology and minerals of southeast Idaho in the annual reports of the Survey to Congress (Peale, 1879). Thus, the Hayden Survey established the basic geologic framework for southeast Idaho. Formations were described, geographic characteristics were identified, and some mineral deposits were discovered. However, even the scientific experts of this formidable expedition failed to recognize the one mineral that was to make southeast Idaho famous. With the passage of Federal mining laws, particularly the General Mining Law of 1872, mining claims were taken throughout the western United States including various sites in southeastern Idaho. Mining claims were even filed on phosphate rock, but not for phosphate. Those early claims were located mistakenly for either copper glance (chalcocite) or coal because of the black, earthy nature of the phosphate. No one suspected that the black rock was a valuable phosphate deposit. There is some controversy about when the presence of phosphate rock was finally recognized. Albert Richter, a prospector from Salt Lake City, claimed to have recognized phosphate deposits northeast of Ogden, Utah, as early as 1889. In 1897, R.A. Pidcock found some older workings on a soft, black formation in Rich County, Utah (Jones, 1907; Hansen, 1964). Pidcock located a number of claims and sampled the "ore." The samples were analyzed and shown to contain no gold or silver, but did contain phosphate in large amounts (32%). This was the first specific documentation of phosphate rock. Charles Colcock Jones, a mining engineer with the Mountain Copper Company, Ltd., examined Pidcock's discovery in 1903. Jones recognized other phosphate deposits scattered throughout that part of Utah and in 1903 heard of a deposit being worked for coal
The history of production of the Western Phosphate Field
49
Fig. 3-4. Armstrong Shuveloader, Conda Mine, circa 1922. Photo from the Idaho State Historical Archives.
near Montpelier, Idaho. Upon examination of a 76 m shaft found on the property, Jones determined that the "coal" was really the same type of phosphate ore he had observed in Utah. In subsequent years, Jones described phosphate deposits near Hot Springs, Soda Springs, Bennington, and Bloomington, Idaho; Cokeville and Sage Station, Wyoming; and other areas throughout southeastern Idaho, southwestern Wyoming, and northern Utah (Jones, 1907, 1913) (Fig. 3-4).
LAWS ASSOCIATED WITH US PHOSPHATE EXPLORATION AND MINING From 1890 to about 1915, a series of legislative actions were undertaken to protect and conserve the resources on public lands, particularly resources in the west. Laws enacted included the designation of forest reserves (1897), the Reclamation Act of 1902, the creation of the first National Monuments (1906), the Antiquities Act of 1906, the Enlarged Homestead Act of 1909 (reserved minerals to the government), the Pickett Act of 1910 (giving Executive Branch the power to withdraw public lands from entry), the Stock-Raising Homestead Act of 1916 (reserving mineral fights), and the mineral leasing acts of 1917 and 1920 (Table 3-I).
TABLE 3-1 Historic phosphate legislation, litigation, and policy (from Lee, 2000) Citation
Remarks
Date
Action
Purpose
December, 1908
Secretarial withdrawal
Withdrew phosphate-bearing lands in west from entry; eliminated ability to locate claims; required land classification for phosphate; grandfathered existing claims
Gave responsibility for phosphate land classification to U.S. Geological Survey
June 3, 1909
Secretarial policy
Claims could be patented under both lode and placer provisions; either patent would be valid; established "first in time, first in right"
Formally stated by First Assistant Secretary in a letter to Judge E.B. Crithlow
June 25, 1910
Act
The Pickett Act, giving the President authority to withdraw federal lands, was signed into law
36 Stat. 847
The President formalized the earlier Secretarial phosphate withdrawals & added lands
January 15, 1912
Litigation
First 2 litigations to address phosphate lode vs. placer; judge decreed in favor of lode claimants in both civil actions; appealed; Appeals Court affirmed lower court decision
Civ. Nos. 568, 569, D. Wyo.; 201 F. 830
Placer locations were located 2 years before lode locations; extensive discussion as to past case law & definitions of lode & placer contained in Appeals Court decision
September 3, 1912
Litigation
Placer claims located in 1904, 1905; lode claims located in 1907; District Court determined "first in time, first in right"; appealed to 9th Circuit Court, lower Court reversed and found for lode claimants
December 12, 1912
Public Land Decision
Secretarial decision that phosphate claims were to be patented under lode provisions of the Mining Law
Lengthy discussion of character of phosphate deposit &jurisdictional responsibilities contained in Appeals Court decision 41 L.D. 403
Case attempted to define what constitutes a lode; provided several supportive citations
July 18, 1914
Secretarial Policy
Restricted acceptance of locations for phosphate to lode claims only; required placers locators to re-form claims to lodes
Policy first issued by First Assistant Secretary in a letter to the Chairman of House Committee on Public Lands
January 11, 1915
Act
Validated all existing phosphate placer claims
38 Stat. 792b
March 3 1, 1915
Secretarial Policy
Issued authorization to process patents for phosphate placer locations under placer rules of the Mining Law
N/A
February 25, 1920
Act
Congress passed Mineral Leasing Act and removed phosphate (and other minerals) from the provisions of the Mining Law; created complete leasing system; grandfathered existing phosphate lode and placer claims until final disposition, patent, or relinquishment
41 Stat. 437
3 (D
E.
$% P d a
Policy brought the contest between phosphate lode & placer claimants to a close This law remains in effect.
83 3
% 5 6
$s 3
= b o fi
3
&
52
S.M. Jasinski, W.H. Lee and J.D. Causey
On December 9, 1908, the Secretary of the Interior issued a Secretarial Order that created a "temporary" phosphate reserve of 18,400 km 2 in Idaho, Utah, and Wyoming. These lands were identified by the USGS as either containing or having the potential for phosphate deposits. Three additional Secretarial withdrawal actions in 1908 and 1909 added approximately 18,600 km 2 to the phosphate reserve, bringing the total to about 20,200 km 2. The underlying reasons for these withdrawal actions were (a) to protect the phosphate resource from appropriation under the Homestead laws and the Mining Law of 1872, and (b) to give the USGS time to classify the withdrawn lands as either "mineral" or "non-mineral" in character. The Land Classification Board formed by the USGS on December 18, 1908 was charged with the responsibility to classify the lands and make appropriate recommendations of revocation and restoration to the General Land Office. The board initially recommended that approximately 94,200 km 2 in Idaho, Utah, and Wyoming be restored without classification, as the land did not have potential for phosphate resources. Under the Pickett Act (36 Stat. 847), as amended, Presidents Taft and Wilson withdrew approximately 10,500 km 2 in Idaho, Utah, and Wyoming (1910-1917) and formally created the Western Phosphate Reserve. This withdrawal formalized the lands that remained temporarily withdrawn under the existing Secretarial orders. The USGS was again charged to investigate the lands for phosphate resources, to classify the lands as to their phosphate content, and to recommend specific restorations to the General Land Office. Starting about 1909 and extending to 1914, USGS geologists conducted extensive and detailed studies of the phosphate rock deposits of the withdrawn area (Mansfield, 1927). The Phosphoria Formation (Richards and Mansfield, 1912) was determined to be widespread and relatively consistent in its phosphate content and quality. The Mineral Leasing Act of February 25, 1920 (41 Stat. 437) ended the acquisition of phosphate through the Mining Law and rendered moot the need for phosphate withdrawal and classification actions. Beginning in about 1942, a renewed interest in the reserve started, not for phosphorous, but for associated vanadium, and, in 1947, for the uranium associated with the phosphate ore (Montgomery and Cheney, 1967). Because of this vanadium-uranium interest in the 1940s and 1950s, additional classifications of the phosphate withdrawals as mineral or non-mineral were made with resulting restorations of some of the withdrawn lands. The last government sponsored investigation for phosphate classifications began in the 1960s and continued into the 1980s. Under this program, Known Phosphate Leasing Areas (KPLA) were defined. KPLAs are areas where the phosphate resource is available only through the competitive leasing provisions of the Mineral Leasing Act.
IDAHO Production was first reported in 1907 from the Waterloo mine near Montpelier. Phosphate production rates rose slowly in the first half of the twentieth century. The phosphate industry grew steadily from 1948 to 1968, during which time the reserves were quantified and demand grew for phosphates.
The history of production of the Western Phosphate Field
53
Since 1907, 30 mines have reported production of phosphate rock in Idaho, 10 mines were strictly underground operations, 15 were open-pit, and five used both mining methods (Fig. 3-5; Appendix A). By 1960, all underground mining had ceased. The major mines that operated prior to 1960 were the Waterloo Mine, reopened as an open pit operation, the Gay Mine on the Fort Hall Indian reservation in Bingham County, and the Conda mine, which began surface operation in 1955. Before 1948, all phosphate rock was used either for direct application as a soil amendment or shipped to the Anaconda Co. fertilizer plant in Anaconda, Montana, which was
11 3~
11 2~
i
Xx
111 ~
i
CLARK
--
~
S
.
.
.
.
.
FREMONT
Il
_
~
JEFFERSON
z o ,~.
_____q O
x
6
7
x 8
10 12 (9 x x l 1 x x13 16 14 15 x x
IdahoFalls 9
x
20
X17 x
,,
X
X18
19
Inset
--~
--~
3
BONNEVILLE
I
A BINGHAM
x5 z
(
~
Pocatello
~r ~
x xx
See Inset
•215 •
•
~ ~
21 x --~ z,~ NKLIN
I1 I
113~
IF~AHO UTAH
I
112~
(L
x
A
"-" 22~,._~ x2324
29
~ 25
x
BEAR LAKE
z
I 111~
Fig. 3-5. Phosphate-rock mines in Idaho, refer to Appendix A for corresponding mine names.
54
S.M. Jasinski, W.H. Lee and J.D. Causey
opened in 1919 for conversion into phosphoric acid for fertilizer use. Elemental phosphorus production was the driving force for the opening of new mines in the 1950s. From 1949 to 1959, four plants were constructed, three in Idaho and one in Montana, although one plant operated only for four years (Table 3-11). The Anaconda fertilizer plant was sold to TABLE 3-II Elemental phosphorus plants in the Western Phosphate Field Company
Location; County and State
Year opened
Year closed
Annual Production capacity (metric tons)
FMC Corp.
Bannock, Idaho
1949
2001
127,000
Central Farmers Fertilizer Co.
Bear Lake, Idaho
1959
1963
22,700
Monsanto Co.
Caribou, Idaho
1952
Active
109,000
Rhodia Inc.
Silver Bow, Mont.
1951
1995
34,000
Data from Service and Popoff (1964); Jasinski (2001).
Fig. 3-6. Truck-shovel operations, Enoch Valley Mine, August 1997. Photo by Peter Oberlindacher, US Bureau of Land Management.
The history of production of the Western Phosphate Field
55
J.R. Simplot Co. in 1960 and was relocated to Pocatello. The same year two other plants were opened in Conda and Kellogg and western production reached 19% of the US total (Fig. 3-2). Ore from two of the four active mines in the State is used to produce wet process phosphoric acid that is the basic component of phosphatic fertilizers. One mine supplies the only elemental phosphorus plant in the United States. Ore from the other mine, which until 2001 supplied a now-closed remaining elemental phosphorus plant, is used to produce purified phosphoric acid for industrial applications.
MONTANA Phosphate rock was first mined in Montana in 1921 and mining continued until 1993. During this period, 23 underground and one open-pit mine operated in the State (Fig. 3-7; Appendix B). Commercial production did not begin until 1929 when Montana Phosphate Co., a subsidiary of what is now Cominco Ltd., started development of the Anderson Mine in Powell County (Popoff and Service, 1965). Montana is second to Idaho in total phosphate rock produced in the Western Phosphate Field. The phosphate deposits in Montana were not suited to surface mining, unlike those in the other western States. Underground phosphate mining was similar to hard-rock mining in its exploration, development, and methods. The cost associated with underground mining and declining reserves resulted in the closure of the last underground phosphate mine in the country in 1993 (Llewellyn, 1994). The Montana part of the Western Phosphate Field, located in the southwestern part of the State, was divided into 10 separate districts. Most production occurred in the Garrison, Melrose, Maxville-Philipsburg, and Centennial Range districts (Fig. 3-8). From 1929 to 1950 and from 1969 to 1993, all production was from the Garrison District in Powell County. Until 1952, all phosphate rock produced in the state was shipped to the Cominco Trail and Kimberley, B.C., Canada, fertilizer plants for processing into phosphoric acid. A large percentage of the P205 contained in the exported rock was shipped back to the United States in the form of fertilizer. In 1952, Stauffer Chemical Co. opened an elemental phosphorus production facility in Silver Bow, near Butte. Rhone Poulenc, S.A. acquired the assets of Stauffer in the 1980s and operated the facility until it was closed in 1995 for economic reasons. The plant was supplied by underground mines in the Silver Bow area until 1967, when Stauffer began using lower-cost ore from surface mines in Idaho (Carillo et al., 1968).
UTAH As discussed in a previous section, the discovery of phosphate rock in the western United States was credited to Albert Richter, who found deposits in Cache County in 1889. Actual development of phosphate-rock mines did not begin until 1904 and phosphate rock
56
S.M. ,lasinski, W.H. Lee and .i.D. Causey 114~ I
112~ I
MISSOULA oNlissoula
LEWIS & CLARK POWELL
15 1 16X XX GRANITE
X"
X18
JEFFERSON
RAVALLI DEER LODGE
BOW ~
.•3•LVER ~3utte 19
~..
MADISON
BEAVERHEAD 23 x
\ 10 X 9 11 X
12 X
x
7
5
X 8
[1
6x X
X
Inset I
114~
I 112~
Fig. 3-7. Phosphate-rock mines in Montana, refer to Appendix B for corresponding mine names.
was first produced in 1907 from the Arickaree Mine in the Crawford Mountains of Rich County. The mine operated intermittently until 1920 and then was inactive until 1953. Between 1941 and 1952, several small mines operated infrequently producing less than 3000 metric tons annually of phosphate rock. In 1953, the San Francisco Chemical Co. began development of six underground mines in the Crawford Mountains (Coffman and Service, 1967). All were closed by 1960 except the Cherokee Mine, which operated until 1977 (Stowasser, 1980). Several other companies had mines in the Crawford Mountains in
The history of production of the Western Phosphate Field
57
112~
114~ !
I
SANDERS
CASCADE LEWISANDCLARK
JEFFERSON
Melrose
•r
BEAVERHEAD
I MADISON
.,,..,..,..,.,..~~ [ L - - c e net na !
114~
ange
!
112~
Fig. 3-8. Mining districts in Montana. the late 1950s, which were all closed by 1966. Since 1907, 14 underground, and one surface mine produced phosphate rock in the state (Fig. 3-9; Appendix C). The only currently operating phosphate-rock mine, the Vernal Mine (Little Brush Creek), commenced operation in 1961. This was the first surface operation in the State (Howes, 1962).
58
S.M. Jasinsla', W.H. Lee and J.D. Causey 112~ i IDAHO
111 ~ i
110~ i
109~
I j
~z,
3~ X5 8xX6
See Inse
/x14
RICH
X9
R,cH
xl0
-r
X11
DAVIS Jxk ~'~'~ ~
WYOMING UTAH
/
s3
,. I
/
X15
Provo
WASATCH
DUCHESNE
UINTAH
_ zo
112"W
11 "W
1l()"W
109~
Fig. 3-9. Phosphate-rock mines in Utah, refer to Appendix C for corresponding mine names. WYOMING Accessible deposits of phosphate rock were identified in the Sublette Range in Wyoming and the Crawford Mountains of Northern Utah shortly after the discovery of phosphate in Utah. In 1906, the Union Phosphate Co. began development of phosphate deposits near Cokeville, Wyoming. Production was first reported in 1907 and continued intermittently until 1931 with more than 25,000 metric tons of phosphate rock produced from the mine during these 25 years. The company also attempted to develop an underground mine in the Beckwith Hills area during the same period. The US Phosphate Co. of Salt Lake City produced ground phosphate rock for direct application to soil from an underground mine in the York Canyon area from 1913 to 1917 (Coffman and Service, 1967). From 1907 to 1978, a total of four mines, two underground, one surface, and one that used both methods, operated in the state (Fig. 3-10; Appendix D). In 1947, the San Francisco Chemical Co. opened the Leefe surface mine in Wyoming near Sage Junction. A processing facility was constructed on site in 1958 and the marketable product was shipped to phosphoric acid producers in the western United States and Canada. The Leefe mine continuously operated until 1978 when it closed because of depleted reserves and economic conditions. The South Mountain Mine located near Kemmerer operated from 1947 to 1951, producing a relatively small amount of phosphate rock for direct soil application. Several other deposits were explored in the State before 1970, however, none produced.
The
history of production of the 111 ~
Western
Phosphate Field
110~
I
59
109~
I
108~
I
I _
i
z o
I
HOT SPRINGS TETON
~--L---L__L__ ~
FREMONT
Z
SUBLE'FI'E m
o
1 2 LINCOLN z o t'xl
-r"
<_ SWEETWATER Rock Green
Evanston
Springs
River
UINTA
z o
WYOMING
,r.-
COLORADO I
111 ~
I
110~
I
109~
I
108~
Fig. 3-10. Phosphate-rock mines in Wyoming, refer to Appendix D for corresponding mine names.
MINING IN THE WESTERN PHOSPHATE FIELD IN THE TWENTY-FIRST CENTURY Currently, active mining areas in Idaho and Utah are not expected to change much from the current situation. Several new mines located near existing mines are planned in Idaho to replace existing mines when reserves have been exhausted. In Utah, the Vernal Mine has more than 100 years of reserves at current mining levels and likely will produce at its current rate for the near future. In the Vernal area, an Illinois company plans to open a new
60
S.M. Jasinski, W.H. Lee and J.D. Causey
mine in the Ashley Creek deposit. However, as of late 2002, the project had yet to receive financial support and no opening date had been set (Jasinski, 2001). It is highly unlikely that phosphate-rock mining will restart in Montana or Wyoming, except in the event of significant shortage of phosphate rock. Deposits in Montana require high-cost underground mining, and deposits in Wyoming have been nearly exhausted. Production rates in the Western Phosphate Field may slowly increase over the next 20 years with the expansion of purified phosphoric acid facilities and great demand for fertilizers in the western United States and Canada.
ACKNOWLEDGEMENTS Reviewers provided helpful comments and suggestions on this chapter.
REFERENCES Carillo, EV., Collins, R.P. and Hale, W.N., 1968. The Mineral Industry of Montana, 1967. US Bureau of Mines, Minerals Yearbook, vol. 3, Area Reports - Domestic: 485-503. Coffman, J.S. and Service, A.L., 1967. An evaluation of the western phosphate industry and its resources. Part 4, Wyoming and Utah. US Bureau of Mines Report of Investigations, 6934, 158 pp. Crowley, EA., 1962. Phosphate rock. Montana Bureau of Mines and Geology, Special Publication, 12, May, 8 pp. Hansen, O.T., 1964. A history of the phosphate industry in Idaho. 64th Annual Report of the Mining Industry of Idaho for 1963-1964, 192 pp. Howes, M.H., 1962. The Mineral Industry of Utah. Minerals yearbook vol. 2, 1961, 9.28. Jasinski, S.M., 2001. Phosphate rock- 2000 annual review. US Geological Survey, Mineral Industry Surveys, August, 13 pp. Jones, C.C., 1907. Phosphate rock in Utah, Idaho and Wyoming. Eng. Min. J., 83(20): 953-955. Jones, C.C., 1913. The discovery and opening of a new phosphate field in the United States. Amer. Inst. Min. Eng., Bulletin 82, October, 2411-2435. Lee, W.H., 2000. A history of phosphate mining in southeastern Idaho. US Geological Survey, Openfile Report, 00-425,242 pp. Llewellyn, T.O., 1994. Phosphate rock annual report, 1993. US Bureau of Mines, Minerals Yearbook, vol. l, 20 pp. Mansfield, G.R., 1927. Geography, geology, and mineral resources of part of southeastern Idaho, with descriptions of Carboniferous and Triassic fossils by G.H. Girty. US Geological Survey, Professional Paper, 152, 453 pp. Montgomery, K.M. and Cheney, T.M., 1967. Geology of the Stewart Flat quadrangle, Caribou County, Idaho. US Geological Survey, Bulletin, 1217, 63 pp. Peale, A.C., 1879. Report on the geology of the Green River District. US Geological and Geographical Survey of the Territories, 1l th Annual Report (Hayden), pp. 509-646. Popoff, C.C. and Service, A.L., 1965. An evaluation of the western phosphate industry and its resources; Part 2, Montana. US Bureau of Mines Report of Investigations 661 l, 146 pp.
The history of production of the Western Phosphate Field
61
Richards, R.W. and Mansfield, G.R., 1912. The Bannock overthrust, a major fault in southeastern Idaho and northeastern Utah. J. Geol., 20:681-709. Service, A.L. and Popoff, C.C., 1964. An evaluation of the western phosphate industry and its resources; Part 1, Introductory review. US Bureau of Mines Report of Investigations 6485, 86 pp. Service, A.L., 1966. An evaluation of the western phosphate industry and its resources; Part 3. Idaho. US Bureau of Mines Report of Investigations, 6801, 201 pp. Stowasser, W.E, 1980. Phosphate rock. In: 1977 Minerals Yearbook, vol. 1, US Bureau of Mines, 24 pp. Vranes, R., 1999. Information technology aids the Soda Springs phosphate mine. Mining Eng., 51:15-20.
APPENDIX A Summary of historic and active phosphate-rock mines in Idaho (from Service, 1966; Lee, 2000) Fig. 3-5 number
Mine name
Last owner of record
Underground1 open-pit
Date opened
First production date
Date closed
Brief history
25
Waterloo
ID Fish & GameBear Lake County
Both
1907
1909
Underground mine with some surface workings by SFCC' 1907-1 9 18; reopened 1919; suspended 1920; permanently closed as underground mine 1929; reopened as surface mine 1945-1 958; briefly reopened early and permanently closed 1960; property donated to ID Fish & Game Dept. 1971
30
Hot Springs
Rhodia Inc.
Underground
1908
NIA
Developed by Union Phosphate Co. 1908-1 920, second underground development 1954-1 956
26
Paris Canyon Earth Science Inc.
Underground
191 7
1917
Initial exploration 1913; operated intermittently from 1917-1926, acquired by Earth Science Inc. 1973
21
Rattlesnake Canyon
U.S. Govt.
Underground
1920
1920
Shipped small experimental tonnage to fertilizer plant in NE 1920, no further activity
29
Bear Lake
Federal Minerals
Underground
1919
1921
Developed 1919; operated intermittently 1921- 193 1, under several owners
14
Conda
J.R. Simplot Co.
Both
1920
1921
Claims located 1906; operated by Anaconda Copper Mining Co.; underground mine 1920-1956; open-pit mining 1952-1 984, operated by Simplot 1960-1984
24
Home Canyon
SFCC
Underground
1920
1920
Lode patents owned by SFCC, leased to American Phosphate Co. 192&1925; ore shipped at 5 8 4 5 metric tonlday; idle; SFCC reentered old workings and then operations abandoned 1953
27
Consolidated Earth Science Inc.
Underground
1920
1930
Lode claims located 1908; operated by Solar Development Co. 1930-1932; acquired by Earth Science Inc. 1973
Both
1909
1941
Mining claims located 1907-1 9 12; 3 short adits opened 1909; claims relinquished and land patented 1914; underground mine opened 1920, no production, lease to Teton Phosphate Co. t 1940; operated open pit operation intermittently 1941-1943
Bennington Canyon
Teton Phosphate Co.
Gay
J.R. Simplot Co.
Located on Fort Hall Indian Reservation; tribal and allottee leases acquired and mine opened 1946; first open-pit mine to produce Federal phosphate; joint ownership Simplot and FMC 19561993
Ballard
Monsanto Co.
Federal lease to Simplot 1948; lease to Monsanto Co. 1951; enlarged 1951, 1955; closed 1969; leases relinquished 1984
Maybe Canyon
Agriurn Inc.
Both
1951
1951
Underground mining 1951-1965, intermittent open-pit mining 1965-1995; inactive status 1995
Blackfoot Narrows (Terteling Pit, Wooley Valley Unit #I)
Rhodia Inc.
Open-pit
1955
1955
Federal lease to Terteling Co. operated 1955-1956; leased to Stauffer Chemical Co. and operated 196749
Fall Creek
E.A. Rex Mining Corp.
Underground
1955
1955
Federal lease to E.A. Rasmussen 1959; lease to E.A. Rex Mining Corp. 1960; underground workings abandoned 1964; lease relinquished 1997
Centennial
J.R.Simplot
Federal leases to Simplot 1953, 1954; operated 19561959; some drilling exploration conducted 1960
Co. Georgetown Agrium Inc. Canyon Rhodia Inc. Diamond Gulch
Both Open-pit
1909; 1958 1960
1959 1960
Placer mining claims located 19061907; underground exploration 1909; open-pit mining 1958-1 964 Federal lease issued to SFCC 1957; operated briefly 1960; reclamation 1961-1 962; lease relinquished, 1993 Continued
APPENDIX A Continued Fig. 3-5 number
Date opened
First production date
Date closed
Brief history
Mine name
Last owner of record
Underground1 open-pit
Mountain Fuel
Nu-West Mining Inc.
Open-pit
Federal lease to Mountain Fuel Supply Co. 1963; operated 19661967 and 1985-1993
Mill Canyon Rhodia Inc. (Wooley Valley Unit #4)
Open-pit
Federal lease to the Terteling Co. 1953; lease to Stauffer Chemical Co. 1967; production 1969-1 974, Rhodia (formerly Rhone Poulenc Basic Chemicals) acquired Stauffer, 1980
Taylor Creek
Monida Resources, Inc.
Underground
Federal prospecting permit to Northern Investment Co. 1967; small amount produced, lease to Monida Resources Inc. 1994
Henry
Monsanto Co.
Open-pit
Federal leases to Monsanto 1960, 1965; South Henry pit closed 1980; Central Henry pit closed 1985; North Henry pit closed 1989; leases relinquished 1993.
Bloomington Earth Science Underground Canyon Inc.
Underground mining Wyodak Coal 1942-1 943; Federal lease to Ruby Co. (Simplot) 1962; lease to Earth Science Inc. 1973; new underground mining 1973-1975
Pritchard Creek
E.A. Rex Underground Mining Corp.
Opened 1975; underground workings terminated 1976; lease relinquished 1997
Little Long Valley (Wooley Valley Unit #3)
Rhodia Inc.
Federal lease to Terteling 1953; leases to Stauffer 1967, 1969; operated 1 9 7 6 1989 and reclamation completed 1989; leases to Rhodia, Inc. 1989
Open-pit
12
17
20
7
8
15
Lanes Creek
J.R. Simplot Co.
Open-pit
Champ
Agrium Inc.
Open-pit
Smoky Canyon
Enoch Valley
J.R. Simplot Co.
P4 Prod., LLC
Rasmussen Ridge
Agrium Inc.
Dry Valley
Astaris Production, LLC
'SFCC, San Francisco Chemical Co.
Open-pit
Open-pit
Open-pit
Open-pit
1978
1982
1984
1989
1990
1992
1978
1982
1984
1990
1991
1992
1989
1986
Open
Open
Open
Open
Located on private surfacelprivate minerals land; opened 1978; suspended 1988, reclamation of waste dump 1998 Federal lease to F.P. Champ 1954; started, 1982; all recoverable ore mined out 1985; reclamation completed I986
3 2
$
%
3Q kg.
Federal lease to Ruby Co. (Simplot) 1962; lease to J.D. Archer 1969; lease to Alumet Co, 1975; construction started 1982; leases to Simplot 1983; 1991 sluny pipeline completed 1983; brought online 1984; mine still in active production
1
Federal lease to Ruby Co. (Simplot) 1963; leases to FMC 1964; 1968, leases to Monsanto 1981, 1990; leases to P4 Productions, LLC 1997; mine still in active production, scheduled for closure in 2003
2 o
Federal leases to Terteling Co. 1953, 1957; leases to Stauffer 1967; leases to Rhone-Poulenc 1986; mining initiated 199 1; leases to Rhodia 1998; leases to Agrium 1998; mine still in active production Federal lease to Simplot 1951; lease to Monsanto 1956; leases to FMC 1967, 1968, 1981; excavation 1992; leases to Astaris Productions, LLC 200 1; mine still in active production
3
r@
82 2%
3
&
APPENDIX B Summary of historic phosphate-rock mines in Montana (from Crowley, 1962; Popoff and Service, 1965) Fig. 3-7 number
Mine name
Last owner of record
Underground/ open-pit
Date opened
First production date
Date closed
Brief history
10, 11
AndersonBrock Creek
Cominco American
Underground Open Pit
1929 1955
1929 1955
1976 1961
First mine in MT, deepest underground phosphate rock mine in world, Brock section opened 1955, continued in area after 1976 at Warm Springs mine, open pit mining 1955-1961 in area of high-grade ore discarded from underground mine
16
Maxville (Skeels)
Leo H. Skeels
Underground
1929
1929
1948
Developed by Washington Phosphate and Silver Co. 1929-1 932, Soluble Phosphates Inc. produced small quantity from 1 9 4 6 1948
2
Jack Pine (Trout Creek)
Russell Luke
Underground
1930
1930
1932
Northwestern Improvement Co conducted original exploration 1930-1 932, leased to R. Luke 1949-1 960, 1962 lease to Bunker Hill Mining Co. 196G1962, no further activity recorded
3
Little Blackfoot River
Northwestern Improvement Co.
Underground
1930
1930
1932
Exploration only, ore shipped to Trail, B.C., Canada for testing 1930
14
Douglas Creek
Cominco American
Underground
1931 1959
1932 1963
1945 1968
Developed by NW Improvement Co. 1931, produced 10,000 tlyr 1943-1 945, lease to Montana Phosphate Products (later Cominco) 1958, began development 1959, operated 1963-1968
Omar Edgar
Underground
1932
Dog Creek
NW Improvement Co.
Underground
Graveley
ComincoAmerican
Luke
1932
1952
Operated intermittently, 1932-1 952, for direct application to soil, shipped ore to Trail, B.C., Canada for fertilizer use and to Silver Bow, MT for phosphorus production, 1952
Ca. 1932 Ca. 1932
Ca. 1932
Small quantity of ore produced 1932
Underground
1940
1940
1956
Peak production 345 metric tonslday, operated continuously until reserves depleted 1956
Cominco American
Underground
1943
1943
1966
Explored by R. Luke 1932, purchased by Montana Phosphate Products, Inc. (subsidiary of Cominco) 1943
Relyea
G. Relyea
Underground
1944
1944
1969
Only mine to remain independently owned in Montana, ore used for both phosphoric acid elemental phosphorus production
Dissett (Red Hill)
James Dissett, operated by Manganese Products Inc.
Underground
1946
1946
1948
Initial exploration 1920s, small amount of ore shipped to Japan 1946
Maiden Rock
Stauffer Chemical Co.
Underground
1946
1947
1967
Developed by Victor Chemical Works, 1946, supplied Silver Bow plant, produced up to 680 metric tonslday
Moonlight (Sunlight)
Bunker Hill Mining Co.
Underground
1946
1946
1953
Moonlight Mining Co. shipped ore Japan and Philippines 1 9 4 6 1953, Sunlight Mining produced small amount 1953, leased by Bunker Hill 1961, no further activity
Canyon Creek
Stauffer Chemical Co.
Underground
1948
1948
1967
Supplied Silver Bow plant 1948-1 967, capacity 600 metric tonstday Continued
APPENDIX B Continued Fig. 3-7 number
Date opened
First production date
Date closed
Brief history
Mine name
Last owner of record
Underground1 open-pit
Canyon Camp (Williams)
Mountain Meadows Phosphate Mining Co.
Underground
Discovered by USGS 1928, subleased to Williams Phosphate Co. 1953-1955, no further activity
Tucker Creek
P Antonioli & C.C. Martin
Underground
One shipment of ore sent to Silver Bow 1953, no further activity
Centennial
J.R. Simplot
Underground
Produced acid-grade rock 1 9 5 6 1958, closed because of difficult mining conditions and varying ore grades
Bishop
Anaconda
Underground
Produced furnace-grade ore, high F content
Gimlet
ComincoAmerican
Underground
Discovered 1952, produced 15,000 metric tonslyr 1961-1967, shipped to Trail, B.C., Canada for fertilizer production
LeMarche Gulch Quartz Hill
Stauffer Chemical Co. Stauffer
Underground
Supplied Silver Bow plant
Underground
Supplied Silver Bow plant
A.G. Jackson
Underground
Ore shipped regionally for soil application
Underground
Extension of Brock Mine, last underground mine in United States, closed 1993
Jackson
Warm Springs Cominco Creek American
' ~ x a c location t is unknown.
APPENDIX C Summary of historic and active phosphate-rock mines in Utah (from Coffman and Service, 1967; Stowasser, 1980) Fig. 3-9 number 2
Mine name
Arickaree
Last owner of record SFCC'
Underground1 open-pit Underground
Date opened
3 .=I
First production date
Date closed
Brief history
1907 1953
1907 1953
1920 1958
First mine in UT, operated 1907-1920; reopened 1953, closed permanently 1958; phosphate rock shipped to Leefe WY
2a
9
14
Woodruff Canyon
SFCC
Underground
194 1
1941
1948
Operated intermittently 1941-1 948
2
Pearl
Pearl Phosphate Co.
Underground
1950
1950
1953
Produced small amount for animal feed supplements 1950-1 953
16
Garfield Chemical & Mfg. Co. SFCC
Underground
1953
1953
1954
Supplied small phosphoric acid and fertilizer plant 1953-1954, closed for economic reasons
1
Little Diamond Creek Mandan
Underground
1953
1954
1956
Phosphate rock shipped to Leefe, WY
3
Pawnee
SFCC
Underground
1953
1954
1956
Phosphate rock shipped to Leefe, WY
5
Emma
SFCC
Underground
1954
1954
1956
Phosphate rock shipped to Leefe, WY
11
Rex Peak
J.R.Simplot
Underground
1954
1954
1955
Phosphate rock shipped to Victor Co, Silver Bow plant 19541955, Victor switched to mines in Idaho and Montana 1956 Phosphate rock shipped to Leefe, WY
Co. 4
Sioux
SFCC
Underground
1954
1954
1956
6
Tuscarora
SFCC
Underground
1954
1954
1956
Phosphate rock shipped to Leefe, WY
8
Cherokee
SFCC
Underground
1955
1957
1977
Only mine in area to operate after 1960, produced about 400,000 metric tons Iyr, ore shipped to Leefe, WY Continued
B
3Z 83 s
2 -zh is 3
s
Q\ \O
APPENDIX C Continued Fig. 3-9 number
Mine name
Last owner of record
Underground1 open-plt
Date opened
First production date
Date closed
Brief history
2
Bradley
SFCC
Underground
1957
1957
1960
Phosphate rock shipped to Leefe, WY
15
Vernal (Little Brush Creek)
SF Phosphates, Ltd. Co.
Open pit
1960
1961
Active
Discovered 1915, purchased by SFCC from Humphrey's Phosphate Co. 1956, sold to Chevron 1981, sold to SF Phosphates 1993, only active mine in Utah
9
Benjamin Jeffs
SFCC FMC Corp.
Underground Underground
1964 1964
1964 1964
1966 1966
Phosphate rock shipped to Leefe, WY Supplied FMC phosphorus plant in Pocatello, ID
2
'SFCC, San Francisco Chemical Co. 2Exact location is unknown.
APPENDIX D Summary of historic phosphate-rock mines in Wyoming (from Coffman and Service, 1967) Fig. 3-10 number
Mine name
Last owner of record
Underground1 open-pit
Date opened
First production date
Date closed
Brief history
2
Cokeville
Cokeville Phosphate Co.Nnion Phosphate Co.
Underground
1906
1907
1931
Developed by Union Phosphate Co. 1906, production began 1907, continued intermittently until 1931, reorganized as Cokeville Phosphate Co. in 1920s, no further activity reported
1
York Canyon
U.S. Phosphate Co.
Underground
1913
1913
1917
4
South Mountain
Phosphate Mines Inc., Kemmerer Mines Inc.
Both
1947
1947
1951
Operated 1913-1 9 17, produced ground phosphate rock for direct soil application, no further activity reported Operated intermittently, 1947-1 95 1, sold phosphate rock for direct application
3
Leefe
Stauffer Chemical Co.
Open Pit
1948
1948
1978
'SFCC, San Francisco Chemical Co.
First open-pit mine in Wyoming, developed by SFCC', 1948, purchased by Stauffer in 1960s, ore shipped to ID & BC for phosphoric acid production, closed because of depleted reserves 1978
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation." From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
73
Chapter 4
THE M E A D E PEAK M E M B E R OF THE P H O S P H O R I A FORMATION"
T E M P O R A L AND SPATIAL VARIATIONS IN SEDIMENT G E O C H E M I S T R Y R.B. PERKINS and D.Z. PIPER
ABSTRACT Variations in the geochemistry of rocks from the Meade Peak Member of the Phosphoria Formation were examined using ratios of elements associated with either the terrigenous or marine sediment fractions. Inter-element relationships in the terrigenous fraction appear useful for chemo-stratigraphic correlation. A sharp decrease upsection in K20/A1203 ratios occurs in the lower half of all but the most northeasterly section, wherein an offset is still evident in average and minimum values. These offsets correspond closely to the lower Guadalupian Series boundary as defined by conodont zonations, coincident with a change from major low-stand to transgressive conditions. The offsets are possibly the result of increased transport distances or flooding of source areas related to transgression of the Phosphoria sea on the Wyoming shelf. A series of intervals displaying high FezO3/AI203, Ba/AI203 and Sc/A1203 ratios occur in the upper beds of the easternmost sections. The intervals do not appear to reflect amplified marine signals, but rather the introduction of terrigenous sediment from a secondary source, or, simply, reworking of sediments under higher energy conditions. The westernmost section, presumably representing the deepest parts of the Phosphoria basin, contains intervals with high Ba/AI203. We suggest these horizons represent periods of low sediment accumulation during maximum flooding and high-stand conditions. Inter-element relationships in the marine-derived sediment fraction indicate that bottom waters of the Phosphoria basin were dominantly denitrifying (suboxic). Ratios of Cd and Mo to Zn and Cu closely approach those in modern plankton in most of the sections, implying a major biogenic source for these elements. Exceptions occur throughout the westernmost (distal) section, possibly due to changes in the dominant plankton populations and relative nutrient uptakes, and in the upper part of the most northeasterly (shoreward, ramp) section, which we suggest is due to increased oxygen levels. Relatively thick phosphatic layers occur in basinal areas due largely to lack of terrigenous dilution during deposition. These basinal deposits appear to have lower concentrations of many trace elements than more shoreward deposits. This may reflect deposition away from areas of peak primary production. Alternatively, biogenic detritus in these areas may have been derived from differing populations of primary producers with differing
74
R.B. Perkins and D.Z. Piper
nutrient requirements. Both mid-shelf (middle ramp) and marginal environments were sites of accumulation of rich phosphatic units with high concentrations of trace elements. Deposits from marginal areas have the most varied geochemistry, largely because they experienced greater variability in terrigenous sediment influx. Even moderate changes in sea level may have dramatically altered energy levels, sediment mixing, and the amount of organic detritus reaching the sediment surface in these shallower marginal areas.
INTRODUCTION The Permian Phosphoria Formation is a marine sedimentary phosphate deposit that extends over several 100,000 km 2 in southeast Idaho, southwest Montana, northern Utah and western Wyoming (McKelvey et al., 1959). The richest phosphorite accumulations are found in the Meade Peak Phosphatic Shale Member in southern Idaho and westernmost Wyoming. The phosphorite beds in the Meade Peak Member are interbedded with limestone, dolostone, mudstone, and chert. The geochemistry of individual sections of the Meade Peak Member have previously been studied and used to define geochemical conditions of the environment of deposition and to model the exchange of seawater between the open ocean and the Phosphoria basin (Piper, 2001; Piper and Link, 2002). However, temporal variations in geochemical properties have not been examined, primarily due to a lack of firm control on the rate of accumulation of the deposit. Similarly, spatial variations are difficult to interpret because of the lack of fine-scale correlations of lithologic units among sites across the basin. Despite these limitations, ratios of elements derived from the two separate sourcesterrigenous and m a r i n e - may provide an independent means of correlating lithologic changes among sites (i.e. across the basin) and allow examination of temporal variations in the sediment geochemistry for individual sites. Changes in ratios of elements interpreted as having had a terrigenous source should be related to changes in sediment provenance, whereas changes in ratios of seawater-derived elements should be related to changes in primary productivity and the level of bacterial respiration, or redox condition, in the water column.
Oceanographic setting o f the Phosphoria basin The Phosphoria Formation accumulated in a terrigenous-sediment-starved basin along the outer margin of the Wyoming-Idaho epicratonic platform after the final phases of Pennsylvanian and Early Permian Ancestral Rocky Mountain tectonism (McKelvey et al., 1959; Peterson, 1980a,b; Maughan, 1984, 1994; Tisoncik, 1984). The depocenter of the Phosphoria sea, an interior sag basin (Wardlaw and Collinson, 1986), was located between approximately 5~ and 25~ latitude (Fig. 4-1). The basin was bounded to the north by the Milk River uplift in western Montana, to the south by the Confusion shelf and Front Range uplift, and to the east by the broad, evaporative Goose Egg basin in eastern Wyoming and
The Meade Peak Member of the Phosphoria Formation
75
Fig. 4-1. Map of the northwest United States showing the extent of the Phosphoria Formation and equivalent Permian units and locations of rock-sections studied. The eastern and northern extent of the Phosphoria basin is shown by bold line. Major tectonic features include the remnant Antler Highlands to the west, the Ancestral Front Range to the southeast, and the Milk River uplift to the north. The Phosphoria basin is separated from shallow, evaporite basins in eastern Wyoming, North and South Dakota, and Nebraska by a carbonate platform.
western North Dakota and Nebraska. An island chain has been proposed to have lain to the west of the Phosphoria basin, a remnant of the Antler or Humboldt highland (Maughan, 1984; Geslin, 1998). The palinspastic width of the Phosphoria basin was approximately 350-450 km (Peterson, 1980b). The paleoshelf on which the Phosphoria Formation was deposited was nearly flat, as suggested by the regional occurrence of marker beds (e.g. the basal "fish-scale bed"). Using values of ~400 km for basin width and 400 m for maximum depth in the western margin of the basin (McKelvey et al., 1959; Maughan, 1994; Piper and Isaacs, 2001), the calculated shelf slope would have been 0.06 ~ Even with extreme values (e.g. 150 km width and 1000 m depth), the slope would still have been much less than 0.5 ~ Therefore, a carbonate ramp model (Burchette and Wright, 1992) would appear more appropriate for the Phosphoria basin than would the conventional paleoshelf model with a marked slope break. A carbonate ramp may be divided, as shown in Fig. 4-2, into the inner ramp, defined as lying above the normal fair-weather wave base, the middle ramp, defined as lying between normal-storm and fair-weather wave bases, and the outer ramp and basinal environments, lying below the normal-storm wave base (Burchette and Wright, 1992). A ramp setting has important ramifications with respect to stratigraphic relationships, facies transitions, energy regimes, and the location of most intense upwelling, which, ultimately, exerted a strong influence on cycling of phosphorus and other nutrients. For example, upwelling along the narrow, deep Peru shelf results in a narrow, near-shore zone of upwelling whereas upwelling along the much broader and more shallow shelf off northwestern Africa results in upwelling far (~50 km) offshore (Codispoti, 1983).
76
R.B. Perkins and D.Z. Piper
The Phosphoria Formation and the correlative Park City Group contain several members that have been interpreted as reflecting three eustatic transgressions onto and regressions off of the Wyoming craton (Peterson, 1980a; Inden and Coalson, 1996; Whalen, 1996). Rocks from the base of the Meade Peak Member to the top of the Franson Member of the Park City Group are thought to represent the second major cycle of marine transgression and regression (Peterson, 1980a,b).
Origin o f Meade Peak sediments The Phosphoria Formation consists of organic carbon-rich silty mudstone, chert, carbonates, and phosphorite. Ratios among major-element oxides (Fe203, K20 , TiO2, Na20, and A1203) are similar to world shale average (WSA) values, suggesting a terrigenous source for the siliclastic components (Medrano and Piper, 1995). The well-sorted grainsize distributions, sedimentary structures indicative of grains settling from suspension, and paleogeographical considerations suggest that this terrigenous component was largely wind transported from the north and northeast (Poole, 1964; Carroll et al., 1998). The marine component includes residual organic matter, calcite and dolomite, biogenic SiO 2 (opal-A, now quartz), PO43- (now carbonate fluorapatite), and trace elements, commonly present as sulfides. The marine component may be divided into two source fractions- biogenic and hydrogenous fractions. Phosphate, an essential nutrient of marine algae, and other nutrients rain onto the sea floor of modern oceans from the photic zone, predominantly as phytoplankton debris (Schuffert et al., 1994). The release of phosphate to sediment pore water via bacterial oxidation of organic matter, containing approximately 2.7 wt. % of PO 3-, leads to the precipitation of apatite in the upper few centimeters of sediment (Kolodny, 1981; Schuffert et al., 1994). The approximate mean apatite accumulation rate of 10 gm -2 yr -l in the Meade Peak Member, which has an approximate PO43- content of 20%, corresponded to a PO 3- accumulation rate of 15.1 mgm -2 day -I (Piper, 2001). This translates into a rain rate of organic carbon onto the sea floor of 0.2 gm -2 day -! assuming a Redfield stoichiometry (Redfield et al., 1963). Sediment traps deployed in ocean margin environments have collected as much as 0.3 g m -2 day -~ organic carbon (Tsunogai and Noriki, 1987), but commonly collect much less (Thunell, 1998; Miiller-Karger et al., 2001). Assuming that 0.2 gm -2 day -I of organic carbon represented 25% of primary productivity, likely a maximum value (Suess, 1980; Sarnthein et al., 1988), our calculated accumulation rate of apatite would translate to a primary productivity rate of 0.8 gm -2 day -1 organic carbon. This is similar to the mean primary productivity rates of 0.5 to 1.0 gm -2 day-l in modem ocean-margin basins such as the Gulf of California, the Gulf of Mexico, and the Sea of Japan (Berger et al., 1988). Sediment-trap experiments of modern continental-shelf environments clearly show that only about 5-10% of primary productivity settles out of the photic zone to a depth corresponding to the depth of the Phosphoria sea. However, such studies represent an instant in time. At the other extreme, the ocean floor captures 100% of the phosphate that enters via rivers (i.e. the open ocean is in steady state). This is achieved by recycling of phosphate
The Meade Peak Member of the Phosphoria Formation
77
and the other nutrients many tens of times between the photic zone and deeper water (Broecker and Peng, 1982). Thus, the seemingly high percentage of phosphate surviving to the sea floor of the Phosphoria basin could have been achieved similarly, by its recycling through the photic zone. The much greater width of the Phosphoria basin of 350-450 km, vs. 15-20 km for the modern shelf environments that exhibit phosphogenesis, extended the residence time of water in the basin to approximately 4.5 year. The extended residence time of water would have promoted the cycling of PO 3- several times between the bottom water and the photic zone. The level of primary productivity and geochemistry of the bottom water required a rather continuous and substantial import of PO 3- into the basin from the open ocean. The hydrography of upwelling environments in the ocean today requires this input to have been between approximately 100- to 250-m depth. Its flux into the basin determined the rate of upwelling of water into the photic zone of approximately 7.8 L cm -2 year-1. Two major differences between the Phosphoria sea and many modern ocean-margin environments are: (a) the low mean accumulation rate of terrigenous debris of about 0.95 mg cm -2 year -l, comparable to its rate in the open ocean (Piper, 2001), and (b) tranquil bottomwater conditions throughout much of the area where phosphatic mudstone was deposited (Cook, 1968; Carroll et al., 1998). PO43- was efficiently retained in the sediment as apatite, similar to its level of retention in sediment-starved environments of the ocean today (Filippelli, 1997). Its high concentration and those of the trace elements reflect the low level of dilution by terrigenous debris. Although the environment may be difficult to envision because no modem ocean-margin sea represents an exact analog (Bentor, 1980), the bathymetry, hydrography, seawater chemistry, level of primary productivity, and terrigenous accumulation rate of the Phosphoria sea are nonetheless present in separate areas of the modern ocean.
METHODS AND DATA EVALUATION Rock samples were obtained from six sections spanning the Meade Peak Member in south-central and south-eastern Idaho and western Wyoming (Table 4-I; Figs 4-1 and 4-2). Samples from the Enoch Valley Mine, Fontanelle Creek, and Mud Spring sections were collected from surface outcrops, open-pit mine faces, or trenches (Sheldon et al., 1953, 1954; Smart et al., 1954; Sheldon, 1963; Piper et al., 2000). Samples from the Lakeridge section are from a core collected from an exploratory well drilled to >4200 m (Sheldon, 1963; Murata et al., 1972). Samples from the Hot Springs Mine were collected from a mine adit (Gulbrandsen, 1979). Samples in all sections were collected at relatively close spacing (e.g. as close as 1 cm but averaging 71 cm in the Enoch Valley, Lakeridge, and Hot Springs sections, 106 cm in the Mud Springs section, and 311 cm in the Wheat Creek and Fontanelle Creek sections) over an average section thickness of 35 m (Table 4-I). The six sections vary in degree of weathering. The Lakeridge core is likely unweathered, the Hot Springs section appears minimally weathered as evidenced by the dark color, induration, and abundance of sulfide minerals, and the other sections are likely weathered
78
R.B. Perla'ns and D.Z. Piper
TABLE 4-1 Summary of locations and number of samples collected for six sections of the Meade Peak Member of the Phosphoria Formation Section
Approximate location UTM Zone: E, N (m)
Enoch Valley
12: 465000, 4749000 [NW _ sec 16, T6S, R43E, Caribou Co., ID; 42~ , N, 111 ~ 24.745'W (Piper, 1999)]
Fontanelle Creek
12: 538000, 4660000 [sec 35, T27N, R116W, Lincoln Co., WY (Sheldon et al., 1954)]
Hot Springs
12: 459000, 4675000 [42 ~ 13.109' N, 111 ~ 24.745' W (Piper et al., 2000)]
Lakeridge
12: 544000, 4704000 [NW _ sec. 19, T29N, R114W, Sublette Co., WY (Sheldon, 1963)]
Mud Spring
12: 33000, 4695000 [42023.5 ' N, 113~ 4' W; Fig. 1 (Medrano and Piper, 1995)]
Wheat Creek
12: 537000, 4650000 [sec. 4, T23N, R116W, Lincoln Co., WY (Sheldon et al., 1954)]
Meade Peak thickness (m) l
No. of samples
Missing analytes 2 Incomplete Th
Source refs. 3
52.0
105
26.5 w Rex Chert
23
53.0
75
25.5 w Rex Chert
36
25.5
25
No F, Se, U
2
26.9
17
No E As, Se, Th, U
2
1
Incomplete Ag, No F, Se, Th, U Incomplete Co
Incomplete SiO2, No Ag, As, Th, U; Se measured but not reported.
3
2
IThickness to nearest 0.5 m measured from base of Meade Peak; Lakeridge and Fontanelle Creek sections include transition zone with Rex Chert. 2See Table 2 and text for full analyte suite. 3References for analytical data: (1) Piper (1999); (2) Medrano and Piper (1995); (3) Piper et al. (2000). to varying degrees. However, none of the sections is extremely weathered, as evidenced by high trace-element and organic-carbon contents (e.g. the mean total organic carbon (TOC) content in the Enoch Valley section is 2.8%). Encapsulation o f sulfides and organic matter by more resistant phases, particularly apatite, may also help to maintain primary elemental
The Meade Peak Member of the Phosphoria Formation
79
Fig. 4-2. (Upper) Diagram showing the estimated positions of the sections used in this study along a shallow dipping carbonate ramp and (Lower) the generalized lithologic relationships of the intertonguing Park City Formation (white), Phosphoria Formation (shaded, excluding the red beds) and the Goose Egg Formation (red beds) (modified from Maughan, 1994; Burchette and Wright, 1992).
relationships. Even moderate weathering is unlikely to impact the terrigenous fraction (Littke et al., 1991). Aliquots of powders (200 mesh) were dried at 60~ Major-element oxides and trace elements were measured by inductively coupled plasma atomic emission spectroscopy (ICP-AES) or inductively coupled plasma mass spectroscopy (ICP-MS) after acid digestion (Lichte et al., 1987; Briggs, 1996; Briggs and Meier, 1999). Some samples were analyzed by X-ray fluorescence spectroscopy (XRF; Taggart et al., 1987). Selenium and As were measured by hydride generation- atomic absorption spectrometry (Hageman and Welsch, 1996). Total S and C, evolved as oxides during combustion, were measured by absorption of infrared radiation (Jackson et al., 1987; Curry, 1996a,b). Carbonate C was measured as CO2 by coulometric titration (Jackson et al., 1987; Papp et al., 1996). Fluorine was determined by ion-selective electrode following LiBO2 fusion and HNO3
80
R.B. Perkins and D.Z. Piper
digestion (Bodkin, 1977; Cremer et al., 1984). Data from each section, including calculated mineral fractions as discussed below, were presented in Medrano and Piper (1995), Piper (1999), and Piper et al. (2000). Limits of detection, precision, and accuracy of each analytical technique have been determined by repeat analyses of rock standards (Baedecker, 1987) and analyses by differing techniques (Medrano and Piper, 1995; Piper et al., 2000).
ELEMENT ASSOCIATIONS Summary statistics for all analytes determined in all samples are provided in Table 4-11. Concentrations of Fe203, K20, SiO2, TiO2, Ba, Ga, Sc, and Th show strong to moderate correlation with A1203 (Fig. 4-3; Ga and Th not shown) and are assumed to be hosted primarily in the terrigenous component. Principal component analysis was also performed on log-standardized data for all 273 samples, resulting in factor scores of > 0.90 for A1203, K20, and TiO2, >0.80 for Fe203, and >0.70 for Ba and Sc on the same factor. These results indicate that most of the variability in the concentration of these elements is controlled by the same factor, which we interpret as the terrigenous component. SiO2 and Th were not included in the factor analysis due to lack of data for some sections. The correlation of K20 and A1203 is especially consistent in all samples in all sections (Fig. 4-3). The concentration of Fe relative to A1 is uniformly higher and more scattered (R2= 0.27) in the Lakeridge samples than in other sections; the intercept is higher as well (2.3% Fe203). The correlation of SiO2 and A1203 is particularly poor in the Enoch Valley section, likely because the SiO2 contents were determined via residuals from the calculated normative components. The slopes of TiO2/A1203 plots vary considerably from section to section, although the intercepts for each section are near zero (<0.03% TiO2). We suggest that the variations in TiO2/A1203 plots are due partly to the use of different analytical techniques for TiO2 determinations. Data for samples analyzed by XRF after fusing with lithium borate plot closer to the WSA (e.g. Mud Springs samples) whereas data for samples analyzed by ICP methods following strong acid digestions plot lower than the WSA trend (e.g. Enoch Valley, Hot Springs), suggesting incomplete dissolution of Ti-bearing phases using this method. The Ba/A1203 ratios show divergence in the Mud Spring section with some samples having very high (> 1000 ppm) Ba concentrations. Barium concentrations were measured exclusively by ICP methods. Thorium concentrations determined by ICP-AES were below detection limits in most samples. Thorium concentrations in the Hot Springs section were also measured by neutron activation analysis (NAA) and those values are moderately well correlated with A1203 (R2 = 0.65).
TEMPORAL VARIATIONS IN TERRIGENOUS ELEMENTS Given the variable lithology in the Meade Peak Member, there is no reason to expect that sedimentation rates in the Phosphoria basin were constant through time or that equal
TABLE 4-11 Summary statistics for all analytes from samples collected from six sections of the Meade Peak Member (Oxides, F, S, and C in wt. %; all others in mg k g ' ) Mean
Standard Deviation
Standard Error
Count
0.23 0.87 0.084 0.11 0.057 0.28 0.003 0.033 0.71 1.3 0.19 0.008 0.18 0.23 0.077 0.32 2.18 24 0.11 5.3 0.68 63 4.5 0.42
279 279 216 279 279 279 279 279 279 257 279 279 279 279 279 225 209 279 30 278 212 279 279 278
Minimum
Maximum
# Missing
2
Ff 2
x.
%
(h
A1203
CaO F Fez03 K2O MgO MnO NazO p205 Si02
so3 TiOz C, Inorganic C, Organic S Ag As Ba Be Cd Co Cr Cu Ga
5.7 22 1.4 2.2 1.5 2.8 0.02 0.82 13 26 2.7 0.16 1.9 3.7 1.08 5.1 26 228 1.7 59 9.2 1038 86 9.3
3.9 15 1.2 1.9 0.96 4.7 0.04 0.55 12 21 3.2 0.13 3.1 3.8 1.3 4.9 31 415 0.60 88 9.9 1064 75 7.0
<0.09 0.35 0.02 0.07 0<0.1 0.07 <0.01 0.04 0.05
20.8 5 1.8 5.81 22.8 4.2 22 0.52 3.O 41 96 24 0.57 12.7 25 9.8 36 400 4860 3 590 108 10,000 540 41
1 64 1 1 1 I 1 l
23
9
5 z $
+g % 6' a
3
g.
1 1 1 1 55 71 1 250 2 68 1 1 2 Continued
2
TABLE 4-11 Continued Mean
Standard Deviation
Standard Error
Count
Minimum
Maximum
# Missing
The Meade Peak Member of the Phosphoria Formation
83
Fig. 4-3. Bivariate scattergrams of selected oxides and elements vs. AI203 in six sections of the Meade Peak Member. Lines represent WSA ratios. SiO 2 values for Enoch Valley section are determined from residuals in calculated normative components.
stratigraphic intervals between samples translate to equivalent time periods. Nonetheless, we can still examine the variation in element ratios throughout a given section to determine geochemical zonations within the depositional sequence. Changes in geochemical signatures at zone boundaries, if not related to diagenetic processes, presumably represent changes in the depositional environment, or, for the terrigenous component, in the source or transport processes. In order to compare geochemical data among samples and sections, we standardize a given parameter to account for variability in component fractions within and between the different rock types. The simplest method to account for the terrigenous fraction is to utilize element-oxide to alumina ratios for those element species that are strongly associated with A1203 and whose ratios approach those of the WSA.
Major-element oxides Average K20/A1203 ratios (Fig. 4-4) decrease upsection at all sites, from slightly above the WSA value of 0.26 to this value or slightly below (see Table 4-III for WSA concentrations). The decrease occurs as a discrete offset, approximately 10-15 m above the base of
84
R.B. Perkins a n d D.Z. P i p e r
Hot Springs
EnochValley
Mud Spring
.... ..-. 2 5 0 0
5000
5000
~- :~ .o_
2000
4000
4000
~
1500
3000
3000
Eo
2000
2000
= Eo O
1000
1000
(1) .>-~ ~> lOOO 500
0
0 0
0.1
0.2
0.3
0.1
WheatCreek i ....
,i . . . .
0.2
0.3
0.4
i ....
.
,.
FontanelleCreek
.
.
.
.
.
.
2500'
25s 0 .
~ ~; E
2000
2000.
20s 0
N ~ 1500 ..Q
1500.
15s 0 -
1000
10C O -
8~
,
i ~{-'-
soo
.....p~..,.
500
0 ..... 0
i . . . . ,, . . . . . 0.1 0.2 0.3 0.4 K20/AI203
0.1
0.2
0.3
K20/AI203
0.4
|
.
~1.
.....
i
5C
0 0
0.3
Lakeridge
,,
2500
E~
0.2
K20/AI203
.-.
2 o >~ 1 0 0 0 "~
,.
0.1
K20/AI203
K20/AI203
....
....
0 0
0.4
1.
0.4
0
0.1
0.2
0.3
0.4
K20/AI203
Fig. 4-4. Plots of K20/AI203 ratios with depth in six sections of the Meade Peak Member of the Phosphoria Formation. Arrows indicate horizons at which the ratios show discrete offsets. Vertical lines represent WSA value. Note differences in y-axis scales.
these sections. The differences in mean K20/AI203 values above and below this offset are statistically significant at the 95% confidence level in all but the Lakeridge section as determined by Student's t-tests. The offset is apparent only in minima values in the Lakeridge section. A "split-moving window" method (Webster, 1973) was used to define the boundaries of various zones based on K20/AI203 values using the generalized distance (D2), calculated as: D2 (XI-X2) 2
where X! is the mean K20/AI203 value of the sample population segment from x i to x i Jr" h having variance s 12 and X2 is the mean of the segment from xi to xi - h having variance s22. Sample window sizes (h) of 8 or 10 were used in the analyses. Zonations in the Fontanelle and Wheat Creek sections could not be determined by this method due to the low number of samples. Although the zone boundaries appear to coincide with lithologic changes (Fig. 4-5), there is no consistent pattern with respect to the nature of such changes. Much greater offsets in K20/A1203 ratios are observed in the Lakeridge section at sequence boundaries between the basal Meade Peak and the underlying Grandeur Member of the
85
The Meade Peak Member of the Phosphoria Formation
TABLE 4-Ili World shale average element composition and concentrations of trace elements in modern marine plankton and seawater World shale average 1
Major oxides A1203 15.14 Fe203 6.844 MgO 2.494 CaO 3.094 Na20 1.34 K20 3.194 P205 0.164 SiO2 58.44 TiO2 0.774 Selected minor elements Ba 5465 Cd 0.85 Cr 835 Cu 355 Mo 2.0 s Ni 42 (65) s Se 1.0 (1.5) 5 V 98-2600 (160) 5 Zn 1005
Deep seawaterz (ppb)
_
0.10 0.22 0.18 10.6 0.59 0.13 1.80 0.52
Marine plankton 3 (ppm)
m
12 2.0 11 2 7.5 3.0 3.0 110
IWorld shale average values: major-element oxides from Turekian and Wedepohl (1961); minor elements from Wedepohl (1969-1978). Values in parenthesis are alternate values used in calculation of the marine element fractions (e.g. Piper, 1999). 2Deep seawater values from Boyle et al. (1976, 1977) and Bruland and Franks (1983). 3Marine plankton values from Martin and Knauer (1973), Collier (1985) and Brumsack (1986). 4Values in wt %. 5Values in ppm.
Park City Group and between the Franson Member of the Park City and the overlying Retort Phosphatic Shale Member (not shown here). However, these boundaries represent regional unconformities accompanied by scouring and reworking of exposed sediments during lowstand conditions (Clark, 1994; Whalen, 1996; Hen&ix and Byers, 2000) and
86
cm above 5000 base of Meade Peak
R.B. Perkins and D.Z. Piper
Enoch Valley
Hot Springs
F~.S.] Chert
[
]Carbonate
[i
I Mudstone
Phosphorite
4000
3000 Lakeridge
2000
1000
Fig. 4-5. Stratigraphic cross-section through the Meade Peak Member from four sections that transect the Phosphoria basin west to east from south-central Idaho to western Wyoming. Connecting lines show offset horizon observed in K20/AI203 ratios.
such large offsets may reflect the introduction of new sediment sources. Though less pronounced, the offsets within the Meade Peak nonetheless represent significant shifts that possibly reflect changes in sediment sources. A decrease in the K20/A1203 content of terrigenous siliclastic material could signify increased subaerial weathering of rocks in source areas, as K20 tends to be lost and AI203 enriched during weathering of continental rocks. However, it seems unlikely that an increased rate of chemical weathering of source areas would have taken place during deposition of the Meade Peak Member as arid conditions were prevalent in northwestern Pangea at that time (McKelvey et al., 1953; Cressman and Swanson, 1964; Isbell and Cuneo, 1996). Carroll et al. (1998) proposed that the terrigenous component of the Meade Peak Member is eolian silt transported from the Milk River uplift area of central Montana. They based their conclusion on grain sizes and planar-parallel fabric observed in the siltstone beds, a lack of evidence of gravity transport, and the Permian climate and physiography of
The Meade Peak Member of the Phosphoria Formation
87
the region. A source area to the northeast is consistent with an earlier study by Maughan (1966) of Permian eolian deposits in Colorado, Utah, and Wyoming that indicated the region experienced predominantly northeasterly winds during the Permian. Assuming eolian transport, reduction in the K20/A1203 values could be related to a change in atmospheric circulation. Alternative explanations are that the reduction in K20/A1203 values is related to (a) inundation of particular, K-rich (mica-rich?) source materials, or, (b) to an increased distance between a given basin location and available sediment source areas in the Milk River uplift, due to transgression of the Phosphoria sea on the Wyoming shell An increased transport distance may have led to loss of larger, heavier particles (Tsoar and Pye, 1987) that are likely to have included a greater fraction of primary mineral crystals, including feldspars. A net increase in finer material, including aluminum-rich, relatively K-poor clay minerals, would then make up a proportionally greater amount of the terrigenous component arriving at a given location, either by wind or, for clay minerals with slow settling rates, ocean-current transport. The latter scenario seems consistent with variations in mineralogy, determined by Knudsen et al. (2001), of continuous channel samples collected in the "J" section of the Enoch Valley mine. They found a significant decrease in orthoclase (from 2.7 wt. % in the lower waste and ore units to 1.0 wt. % in the upper waste and ore units) and a concomitant increase in both muscovite and illite (4.4-8.1 wt. % and 0.4-1.5 wt. %, respectively). Inundation of source areas and increased transport distances are consistent with previous stratigraphic studies (McKelvey et al., 1959; Peterson, 1980; Whalen, 1996; Hendrix and Byers, 2000), which have interpreted the Meade Peak Member as representing a marine transgression. Faunal evidence also has established that the Meade Peak was deposited during the Early Guadalupian (Roadian) third-order eustatic sea-level rise (Behnken et al., 1986; Ross and Ross, 1987, 1995). Interestingly, the position of the offset in K20/A1203 ratios (Fig. 4-5) corresponds closely to the boundary between the lowstand systems tract and the overlying transgressive systems tracts proposed by Hiatt (1997) and lying a few meters above a widespread carbonate marker bed (False Cap Limestone). Fe203/AI203 values plot slightly below the WSA value (0.45) throughout most sections (Fig. 4-6). However, Fe203/A1203 ratios within the upper parts of the eastern sections (Lakeridge and Fontanelle Creek) increase sharply, exceeding the WSA ratio by an order of magnitude. These prominent Fe203/A1203 peaks, occurring between 14 and 25 m above the base of the Meade Peak, correspond to a relative decrease of A1. A decoupling of Fe and A1 oxides is reflected in their poor correlation (R 2 = 0.27) in Lakeridge samples. Plots of the ratios of other elements to alumina, including Ba, Sc (Fig. 4-7), and Li show very similar trends for these eastern sections. These relatively iron-rich, aluminum-poor horizons occur just below and within a transition sequence consisting of inter-layered beds of phosphorite (upper Meade Peak) and Rex Chert (beds m45 through r54). Fe203/A1203 ratios show a moderate correlation (R 2 = 0.55) with biogenic silica content, as calculated by Medrano and Piper (1995). This correlation could reflect decreased terrigenous input, which may be important for colonization of sponges, the source of silica. A lower input rate of terrigeneous material would
R.B. Perkins and D.Z. Piper
88 Mud Spring
~oo ~ ...... ~ ~
.
_
3
E
"6 15oo .
5000
4000
4000
3000
3000
2000
2000
1000
1000
:
I
~
' ,
>~ lOOO
-~o E ~
5000
.
~ 2000 ~ -8
Hot Springs
Enoch Valley
. . . . . . . . . . . . .
i]
500
,
i
0
. . . . . . . . . .
0.01
0.1 I Oxide ratios
0 0.01
10
0.1 1 Oxide ratios
Wheat Creek
.....
2500
,. . . . . . . . . . . . . .
10
AI2031 203
0 0.01
0.1 1 Oxide ratios
Fontanelle Creek
Lakeridge
........
2500
10
2500
,,,_, ~o
2000
tI
-8 "6 15oo .--..Q
~ ~ lOOO 500
.
9
i
m
0
0.01
1500
"
1000
i
1500
'
9
E ~
2000
ooo
1000
500
5OO
.
0.1 I Oxide ratios
0 0.01
10
. . . . . . . ,'!-. . . . . . . . . 0.1 1 10 Oxide ratios
0 0.01
0.1 1 Oxide ratios
10
Fig. 4-6. Comparison of Fe203/AI203 and TiOE/AI203 ratios with depth in six sections of the Meade Peak Member. The vertical lines represent WSA ratios. Note differences in y-axis scales.
Mud Spring
Hot Springs
Enoch Valley
2500.
sooo.
" .... /'i
.........
g' ............
.
5000 oo~
2000 9
.o_ "6 1500(1.) 1000-
9
'~-
.ooo
i
~
,ooo
3000
,ooo
2000
1000 O" . . . . . . . . . . . 0.1
1
: ....
10 100 (ppm/%)
"
1000
---~= =='7 1000
0
F
4000
0.1
1
10 100 (ppm/%)
1000
o~
b
0.1
1
Fontanelle Creek
Wheat Creek
o
I
a
10 100 (ppm/%)
1000
Lakeridge
2500.
~ 20oo. 9
-
1500.
a~ 1000 9 E~
0 O.
2000
i:
,~oo
~ ,ooo
.-.,.., . o
8 Eo 5oo.
i
L
i
1
10 100 (ppm/%)
1000
i
0.1
~
"'-'I
~ ]
J
,soo
~
500 o
.........................
L
~_~
,ooo =
..............
1
=
31......
10 100 (ppm/%)
o
1000
0.1
.....
~ . . ,
1
10 100 (ppm/%)
1000
Fig. 4-7. Plots of Ba/AI203 and Sc/Al203 ratios with depth in six sections of the Meade Peak Member. The vertical lines represent WSA values. Note differences in y-axis scales.
The Meade Peak Member of the Phosphoria Formation
89
augment the relative proportion of marine-derived Fe that would otherwise be masked. However, the relatively conservative nature of some of the other elements for which similar peaks were observed, such as Sc and Li, suggests these peaks represent some process other than an increased marine input or a constant marine input amplified by decreased terrigenous sediment. The relatively alumina-poor intervals may reflect a shift in the source of sediment deposited along the eastern margin of the Phosphoria sea. Late Permian sandstones with high iron content (red beds) exist in the Goose Egg basin in southeastern Wyoming, western South Dakota and Nebraska, and Colorado (McKelvey et al., 1959; Maughan, 1994). Iron-oxide-rich particulates or iron-coated sand grains may have been dispersed to the eastern margins of the Phosphoria basin by near-surface waters following a rise in sea level and inundation of barriers (carbonate platforms) that had previously trapped material in lagoonal environments or by subsequent shoaling during highstand conditions. Sediments transported by either process would likely have been restricted to the more shoreward regions. The absence of repeated or consistent peaks in terrigenous element to alumina ratios in other sections, as well as the proximity of the Lakeridge and Fontanelle sections to the evaporative Goose Egg basin, is consistent with this scenario. Increased current strengths are also evidenced by the increased abundance of bioclasts and chaotic bedding observed in some beds (m45, r50, and r54, Lakeridge section; stratigraphic designations from Sheldon, 1963). The high values of terrigenous element to alumina ratios in the eastern sections, therefore, may simply represent reworking of the same terrigenous sediment by increased current activity and bottom turbulence. Such reworking may have promoted loss of Al-rich fines and retention of heavier minerals that could preferentially host elements such as Fe, Ba, Sc, and Li. Interpretation of the TiO2/AI203 ratio is limited by the fact that the TiO2 concentrations are only slightly above analytical detection limits. There is no discernable trend in the TiOz/AI203 ratio with depth other than the positive spikes in the upper part of the Fontanelle Creek section. These spikes may represent sediment reworking. Variations in the SIO2/A1203 and Na/AI203 ratios with depth were not analyzed due to the large degree of scatter in the plots (e.g. SiO2/AIzO3 plot in Fig. 4-3), resulting, in part, from a substantial and variable amount of biogenic silica.
M i n o r and trace elements
Ratios of Ba and Sc to alumina (Fig. 4-7) lie within a factor of five of the WSA ratio throughout the lower and middle parts of all the sections except in the basal Lakeridge section. There is a consistent and marked increase in Ba/A1203 and Sc/AI203 in the upper 5-10 m of the Meade Peak in the easternmost sections and an increase in Ba/AI203 ratios in the upper part of the westernmost (Mud Spring) section. The increases in Ba/AI203 values in the basal Lakeridge section and Ba/AI203 and Sc/AI203 values at the top of both the Lakeridge and Fontanelle sections are due to lower AI203 contents. This suggests either a secondary source of terrigenous sediment, reworking of the terrigenous component and retention of heavier minerals such as barite, or an overall decreased terrigenous input and
90
R.B. Perkins and D.Z. Piper
the resulting augmentation of marine Ba and Sc signals, as discussed for Fe. The 10-fold increase in Ba/A1203 values in the upper part of the Mud Spring section is largely due to a marked increase in Ba concentrations ( > 1000 ppm; Fig. 4-3). Enriched intervals of barite in sediments from the Sea of Japan have been attributed to remobilization of Ba and precipitation of barite at a sulfate-reduction boundary temporarily fixed by cessation of sedimentation (Torres et al., 1996). The same scenario is offered by Breheret and Brumsack (2000) to explain barite nodules in mid-Cretaceous marls in the Vocontion Trough of southeastern France. The increase in Ba/A1203 ratios in the upper Mud Spring section is therefore consistent with a decrease in sedimentation rate. Decreases in sedimentation may be attributed to periods of relatively rapid sea-level rise (maximum flooding events), which increase the distance between source and depositional site and tend to trap sediments near shore.
SPATIAL VARIATIONS IN TERRIGENOUS ELEMENTS Assuming the previously noted offsets in K20/A1203 ratios (Fig. 4-4) represent variations in transport processes or terrigenous sources, they can be used as chemostratigraphic horizons (Fig. 4-5). If so, the Enoch Valley, Hot Springs and Wheat Creek sections have nearly correlative thickness in the lower part of the Meade Peak Member, while the lower Mud Spring, Lakeridge, and Fontanelle Creek sections are condensed. The Mud Spring section would be expected to be the most condensed given that it is the most distal site and, therefore, the one receiving the lowest rate of terrigenous influx. Low sedimentation rates in the upper Mud Spring section are suggested by the high Ba concentrations as previously discussed. The Lakeridge and Fontanelle Creek sections, lying near the basin margins, may be condensed due to low accommodation space during deposition of the basal Meade Peak.
TEMPORAL VARIATIONS IN MARINE ELEMENTS In order to evaluate temporal and spatial variations in the marine elements, their measured bulk concentrations are first corrected for the terrigenous portion using the equation: M = T - D x (% terrigenous fraction) where M is the marine-derived concentration, T the total bulk element concentration, and D x (% terrigenous fraction) is the portion of the total concentration contributed by terrigenous material. D may be determined by extrapolation of the minima line of bulk element versus detritus plots to 100% detritus, from the analyses of a single sample having >95% terrigenous fraction and assuming a constant minor-element concentration in this fraction (Medrano and Piper, 1995), or by using WSA values. Marine components (apatite, carbonates, organic matter, biogenic silica, and other precipitates) are assumed the primary host of elements other than A1, Fe, K, Na, Ti, Sc, and Th.
The Meade Peak Member of the Phosphoria Formation
91
It is important to remember that the present distribution of elements among mineral phases may not reflect their primary sources. This is particularly true for organic-rich rocks such as those that make up the Meade Peak Member. Previous interpretations of marine sediment fractions based on bulk chemistry (Medrano and Piper, 1995; Piper, 1999; Piper et al., 2000) assumed that the net element mass present in the primary sediments, excluding hydrocarbons, remained within the limits of the bulk sample volume. That is, although diagenesis as well as thermal maturation of organics has taken place (perhaps during multiple episodes) and various elements contained therein have been mobilized, the elements were effectively captured by existing or authigenic phases within transport distances on the order of the bulk chemical sample size (i.e. centimeter scale). While this is a valid assumption in models of early diagenetic phosphogenesis, the same principle must be held with respect to the various metals (e.g. Cu, Zn, Se) for their entire post-depositional history, excluding oxidative weathering at the surface. Hiatt (1997) provided some evidence for immobility of elements during deep burial. He examined selected geochemical parameters at sites previously shown through other indicators (e.g. conodont color index, organic chemistry) to have undergone variable degrees of heating. No trends consistent with expected element losses due to heating were found. Marine trace elements are analyzed in a manner analogous to that used for the terrigenous elements, except that Cu and Zn are used as key elements representing the biogenic fraction. These elements have previously been identified as being largely or soley biogenic (Piper, 1999, 2001). Cu/Zn ratios in all sections closely approximate those found in modern marine plankton (Martin and Knauer, 1973; Collier and Edmond, 1984), except in the upper Lakeridge section (Fig. 4-8). Ratios of other elements may also be compared to ratios in modern plankton and deep seawater. We arbitrarily allow for a fivefold variation from the representative ratios for plankton (Table 4-III). The elements having metal to Cu and Zn ratios more than five times that of modern plankton are assumed to have had significant hydrogenous input in addition to their biogenic and terrigenous sources. Cadmium and Mo were also identified by Piper (2001 ) as being largely or soley biogenic. However, there is relatively little difference between the ratios of Cd to Cu or Zn in modern ocean-bottom waters and phytoplankton. Mo/Zn ratios, however, vary by three orders of magnitude between bottom waters and phytoplankton. Mo/Zn ratios in the Meade Peak closely approximate the value (0.02) found in modern plankton, further supporting the interpretation that these elements are largely biogenic (Fig. 4-8). Molybdenite (MoS2) would be expected to precipitate under strongly sulfate reducing conditions (~Eh = -0.175 V). Lack of Mo enrichment over phytoplankton values is a key indicator that such conditions were not present in the bottom waters of the Phosphoria basin. One apparent exception occurs in the upper Lakeridge section where Mo appears enriched relative to Zn (ratios of 0.1-1.0). However, Mo/Cu ratios do not show such a shift (Fig. 4-9). Changes in elemental ratios in the Lakeridge section are discussed below. Chromium, V, and rare-earth elements (REEs) are predominantly hydrogenous elements (Piper, 1999). Chromium is the most obvious example of an element for which the metal to Cu and Zn ratios plot well above both the average modern planktonic and seawater ratios (Fig. 4-9). Reduction of Cr(VI) to Cr(III) (Emerson et al., 1979; Murray et al.,
92
R.B. Perkins and D.Z. Piper Cu/Zn
Mo/Zn Fontanelle Creek
Lakeridge 2500
2500
E2000
2000
~
1500
1500
~
100o
1000
Lakeridge
~
Fontanelle Creek
_5 ,,,,llSOOtl "I "
:I
E 500
~
o
01
1
10
0.01
Hot
Springs 50C
- ' ~
:|
3000
~ooo _ . . . . ~
~1ooo<~__~ 0
0.01
~176176 i l
:
0.1
~--
10
0.01
0.1
1
Hot
5000 ~
"-~ 40C
4000
_30C
3000
~ 20r
10
Springs
...............
2000 [ r ~
E
~ 10C
:
........
1
p
:
~4000'
0.1
Enoch Valley
000
g
.
0.01
1
0 9
Mud Spring
0.1 Wheat
0.1
:...... !
1
10
0 9
Mud Spring
Creek
. . . . . . . . . . .
2500
2500
0.01
1
w
2500
!
9
0.1
1
10
Wheat
Creek
2500 -
|
o2000
. 9 ~ .......i
1500
1500
i
.
2000 "
~2000-
i
/
/
, o E
|
E .
|
2000
|
1500 9
1500 "
1000
1000 -
_
)
1000
1000 E E
= O
|
|
.
= o
500.
500
E E
|
9
500,
500
9
i
0
0.01
0
O.
1
O.Ol
0
"
o.
1
0.01
~
..............
0.1
1
i.
10
0 ~ .............. o.oi o.1 I 10
Fig. 4.8. Plots of marine Cu/Zn and Mo/Zn ratios with depth in six sections of the Meade Peak Member. Solid vertical bars represent average ratios in modern plankton and dashed bars represent modern deep-seawater ratios.
1983) and precipitation as a chromium oxide (Cr203) or hydroxide (Cr(OH)3) from seawater implies bottom-water redox potential at or below upper denitrifying conditions (Eh < 0.55 V). Precipitation of Cu2S should occur whenever free HS- is available. Values
of Cr/Cu ratios that are significantly higher than biogenic values and, for the Meade Peak Member, even higher than seawater values (Fig. 4-9) imply that Cu was not precipitated as
93
The Meade Peak Member of the Phosphoria Formation Cr/Cu
Mo/Cu Lakeridge ...... , .............. . ...... ). ,..
2500 E ~o2000
|..
=
0
15oo
500
|.. |
........2 1 ........ 10 100 0.1
.~ ~" 4000
~
" ~
~
5500. 000005.00000
"
.... ,-;"
_~.__ ~
,
E E
= 500. O
9
0 0.
0 ...................... 1 10 100 0.01 0.1 Hot Springs ...... ~
Enoch Valley 5000 : ~ ' " "
'
1000.
0 0 0
500
500 0 0.01
.......
|
,,_..
1000 E E
Fontanelle Creek
2500 ~2000
000
| | | |
c ._ 1500
Lakeridge
Fontanelle Creek 500 . " ...........................
1
10
100
10
Enoch Valley |
Hot Springs ...... ..
000
O00
4000
000
000
3000
000
000
000
000"
5000
100
',
~
'
"
'
F~ 3000
,-
~ 2000 2 ~-,~ E= 1000 ~
,. ,. "
',
_ 0
|
"~
1000
.......
-
I
~
I
10
100
0 0.01 0.1
1
..... 10 100
9
2500
I
~ 10
......... 100
Mud Spring ...... ......, ....... . .....
2500
g
2000!
._.2000
.
c
9 0.1
9
9
~o2000
~|
0
Wheat Creek
Mud Spring , ....., ...... , ......
.
000
000
i.
0.01 0.1
2500
| 2000
0 0.1
2500
1
............ 10 100
Wheat Creek ...... . ...... . .............
2000
9
1500
1500
c ~ 1500..
1500
1ooo
1000-
~ 1000.
1000 -
500
500
E E 0= 500-
500 -
!
2 E
0 0.01
O.
1
10 100
-
0 0.01 0.1
1
10
100
0
0.1
....... ' . . . . . . . . . . . . . . . . . . . .
1
10
100
0 ................. 0. 1 10 100
Fig. 4-9. Plots of marine Mo/Cu and Cr/Cu ratios with depth in six sections of the Meade Peak Member. Solid vertical bars represent average ratios in modern plankton and dashed bars represent modern deep-seawater ratios.
sulfides. This fact precludes sulfate-reducing conditions. The consistency of Cr/Cu ratios with depth thus provides both upper and lower limits with respect to redox conditions. V/Cu ratios of more than fivefold above planktonic ratios occur chiefly in phosphorite ore zones in the central and lower parts of the eastern sections (Fig. 4-10). V/Cu ratios elsewhere in these sections and throughout the western (Mud Spring) section tend to lie between seawater and planktonic values. Reduction of V as H2V04 and precipitation of
R.B. Perkins and D.Z. Piper
94
V/Cr
V/Cu
"'E2500 - ~
Lakeridge
Fontanelle Creek
Lakeridge
,"
.... " ..... 9" i " i i
25C0
.......
Fontanelle Creek
2500.
".......... ]-
2500
t ,
2000
~2000
20C0
~ 2000.
c o 1500
15C0
--~ 1500. z
*
1500
'= 1000
10C0
1000.
7
1000
...........
, ......
-5 E
= O
= O
5(30
500 0 ..... ~ 0.1 1
" 10
100
0 -0.1
1
10
500.
500-
0 ................... 0.01 0.1 1 10
100
Enoch Valley
Hot Springs
Enoch Valley
5000
50C 0
4000
40C 0
3000
300 0
2000
200 0
o 1000
100 0
0 .................... 0.01 0.1 1 10
0
0
0
Hot Springs
~
5000 ~
'
"
~
g
0 ........ 0.1 1
i
10
100
0 0.1
i I4~176176lit
000F- i i i 000L_ I i i
_
1
10
0.01
100
Wheat Creek
Mud Spring
2500
2500-
g o 2000 ,._..
2000-
o 1500
1500"
. . . . . . . . . . . . . ;,,,...........
'
0.1
1
10
1000'
E E
= o
500'
500
0 0.
1
10
100
2500-
~c2000 ]
2000-
E
x.~ 1500]
,
,ili
1
10
1500.
,
._~
1000.
"5 E E ~
:
' ,
0
0 - . . . . . . . . . . . . . . . . . . . . . . . . . 1 10 100 0.01 0. 0.1
0.1
Wheat Creek
Mud Spring
...-..2500] . . . . . . . . . . . . . . . . .
~
N 1000
0.01
I..... ',_......
1
10
500. 0 .- . . . . . . . . . . . 0.( 1 0.1 1
L ...... 10
Fig. 4-10. Plots of marine V/Cu and V/Cr ratios with depth in six sections of the Meade Peak Member. Solid vertical bars represent average ratios in modern plankton and dashed bars represent modern deep-seawater ratios. V205 occurs under lower redox potentials than does precipitation of Cr(OH)3, but still within the lower range of denitrifying (nitrate-to-Fe(III)-reducing) conditions. These plots suggest that such conditions (i.e. Eh----0.04 V) were present during deposition of the phosphate-rich zones in the central sections and in the lower, organic-rich part of the Lakeridge section. The offsets to lower V/Cu values in the upper half of the Lakeridge and Fontanelle sections and in the middle-waste zones of the central sections suggest increased
95
The Meade Peak Member of the Phosphoria Formation
oxygen levels on average, which would decrease the total amount of V precipitated and incorporated in sediments over time. Further constraint of redox conditions is provided by plots of V/Cr ratios (Fig. 4-10). As both V and Cr precipitate from seawater under the lowermost denitrifying conditions, they should plot around seawater values if such conditions were maintained sufficiently long. The plots show that V/Cr ratios only approach seawater values in the lowermost central and eastern sections and in the upper-ore zone of the Hot Springs sections. The offset to lower V/Cr values in the upper part of the two eastem sections again implies a shift to higher redox conditions with insignificant input of V via precipitation from seawater. REE concentrations throughout the Meade Peak Member, normalized to WSA values, show negative Ce anomalies that become less pronounced with increasing Ce-bearing terrigenous content (Fig. 4-11). The negative Ce anomalies, in addition to the persistent enrichment of Cr, also imply denitrifying conditions were maintained in bottom waters as precipitation of insoluble Ce oxide or hydroxide would have occurred under higher redox potentials (de Baar et al., 1988), imparting a positive Ce anomaly. These data indicate that the bottom water of the Phosphoria sea was dominantly denitrifying (dysoxic to anoxic; Allison et al., 1995) but varied in intensity through time. There is no evidence that bottom waters were sulfate reducing (euxinic). Although sulfate-reducing conditions in bottom waters may have been approached at brief intervals during the deposition of the phosphorite units, as suggested by the maxima in V/Cu and V/Cr ratios,
Lakeridge '
I
'
I
'
I
'
I
'
I
'
Depth < ~10 03 o
I
'
I
'
I
'
= 19.5
I
'
I
'
I
A 2.2
A_~mabove base):--.~-14.1
'
I
I
- -~-0.5
'
I
'
0.1 -.~__n
Eo 1 i
l
O l
i
i
I
,
I
~
I
,
I
,
I
,
I
,
I
,
I
,
I
i
I
,
I
La Ce Pc Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Mud Spring '
I
'
I
'
I
'
I
'
I
'
I
'
I
'
Depth - 25.5 (m above base): # 21.2
10
I
'
I
'
I
'
I
J- 14.2 --o- 6.6
'
I
I
'
I
'
I
:
0.1
or) o
o-~ I
E~ m8
, ,
b" 0.1
,
I
,
I
,
I
,
I
:
I
:
I
,
I
~
I
i
I
i
I
,
I
,
I
,
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb
Fig. 4-11. REE patterns of selected samples from marginal and basinal sections of the Meade Peak Member. REE concentrations normalized to WSA values.
96
R.B. Perkins and D.Z. Piper
the occurrence of sulfate-reducing bottom waters appears constrained to intermittent periods or localized areas. Such conditions could have resulted: (a) by a change in sea levels and a resulting shift in the zone of intersection between the sediment surface and the oxygen-minimum layer; (b) in restricted environments (e.g. back-shoal or lagoonal settings) due to poor bottom-water circulation; or (c) because peaks in organic-matter influx, driven by elevated primary productivity, could have exhausted oxygen and nitrate levels. An alternative explanation for the observed shifts in inter-element ratios (e.g. V/Cu) is variability in algal populations, potentially with differing nutrient demands (e.g. lower relative Cu uptake). We currently have no way of evaluating this possibility. The large offsets in marine element ratios in the upper part of the easternmost (Lakeridge) section suggest a major change in the environment of deposition. Some of the marine element ratios (e.g. Mo/Zn and Mo/Cu) seemingly reflect differing paleoenvironmental conditions. Here we examine the geochemistry and lithology of this section more closely in an attempt to elucidate the nature of these changes. The concentrations of many marine-derived metals (Cd, Cu, Mo, Ni, Se, V, Zn) in the Lakeridge core decrease abruptly by one to two orders of magnitude and organic carbon decreases by at least half at 7-8 m above the base of the Meade Peak (Fig. 4-12). This may
Lakeridge
Lakeridge
Lakeridge .
2500
.
.
.
.
.
.
.
.
.
.
.
.
.
.
-
.
sool. ..............
""TOC
....
.
.
.
.
!
.
.
.
.
,
....
2500-
i
g o
2000
2000-
._ 1500
1500-
y~
-~ 1ooo E E O
i
500 , , , i , , ,
0
20
40 %
60
80
1000
5000 i ,,.,.,,-.t-"r,.
100
0
Apatite
~,oo~' ............
1000-
5
,i
10 % TOC
i" N,'i
oooi
15
....
20
and
i,,,
25
30
0
, ,~.~ , i . . . .
0
| ....
1
2
S
S/
2500 -F
2500
2000+
2000
1500 -I-
1500
I000+
1000
3
4
600
800
TOC
9
N lOOOtr E --, O
5000
soo
500
0
1000
2000
Concentration
3000 (ppm)
4000
0"
0.01
0.1 Element
1 ratios
10
0
0
200 Total
400 La (ppm)
Fig. 4-12. Concentrations of apatite, TOC, total S, selected metals, and La (representing trivalent REE) and Cd and Mo vs. Cu ratios with depth in the Meade Peak Member, Lakeridge section. These data are from discrete samples and do not necessarily represent average compositions throughout the various sample intervals shown.
The Meade Peak Member of the Phosphoria Formation
97
be due to increased dilution of marine sediment by terrigenous detrital material. Normative calculations (Medrano and Piper, 1995) show an increase in the terrigenous fraction from an average of 26% below m24 (n = 10) to 44% from m24 to m44 ( n - 19). This increase is statistically significant at the 95% confidence level (t-score = 2.43; df = 23). The increase in terrigenous sedimentation may be explained in terms of increased accommodation space by rising sea level, an increased supply from easterly or southerly marginal areas following inundation of previously emergent barriers, or to shifting ocean or wind currents transporting terrigenous sediments. All three scenarios are consistent with the observed offset in K20/A1203 values (Fig. 4-4). The concomitant decline in organic carbon and marine trace element concentrations is then largely the result of increased dilution by terrigenous sediments. Although the concentrations of marine elements decline sharply at ~7 m, inter-element ratios are relatively constant from the base of the Meade Peak to ~ 14 m above the base (e.g. Cd/Cu, Fig. 4-12), indicating that the source of these elements remained the same for some time following the apparent change in sedimentation rates. The offsets in marine element ratios occur at or near bed m40 (~ 14 m above the base of the Meade Peak) and persist through the overlying 6 m (e.g. Cu/Zn and Mo/Zn ratios in Fig. 4-8, Mo/Cu in Fig. 4-9, V/Cu in Fig. 4-10, Cd/Cu in Fig. 4-12). Ratios of metals to Cu in this interval more closely approximate those found in modern plankton. The lack of apparent enrichment in these ratios over planktonic values would seem to preclude prolonged excursion of bottom waters into sulfate-reducing conditions. Metals to Zn ratios are shifted towards seawater ratios which, by themselves, would suggest decreasing redox conditions. There is also another decrease in organic carbon just above the same horizon in the section (Fig. 4-12), dropping from an average of 7.1% below m40 (n = 25) to an average of 0.8% above this sample (n = l l; t = 5.19 at 95% CI, df = 24). If we assume that accumulation of organic matter is driven by primary productivity (Pedersen and Calvert, 1990) rather than bottom water redox (Ryan and Cita, 1977; Rossignol-Strick et al., 1982), then such a shift may indicate a drop in primary production, which would decrease oxygen demand both in the sediment and bottom waters. Increases in maximum and average REE concentrations occur over the same interval (above 14 m; e.g. La in Fig. 4-12). Studies of REE concentrations in seawater (de Baar et al., 1988; German et al., 1991; Sholkovitz et al., 1994) showed that they reach minima in the zone of denitrification and rise dramatically in the zone of sulfate reduction, presumably due to reductive dissolution of carrier phases. Sulfate reducing conditions in bottom waters would therefore prevent REE accumulation in underlying sediment, whereas denitrifying conditions would enhance such accumulation. The decrease in organic-carbon content, concomitant with the increase in REE concentrations and the offset in marine metal/Cu ratios signify an important shift in paleoenvironmental conditions. A similar signal at the Fontanelle site (not shown) suggests a shift in paleoenvironmental conditions along the eastern margin of the Phosphoria basin. In the Lakeridge core, these changes occurred during deposition of a succession of variably calcareous and phosphatic mudstones that occur at ~14 m above the base of the Meade Peak, below the upper Meade Peak-Rex Chert transition. We suggest that changes in both terrigenous and marine components are eustatically driven, reflecting a change from transgressive to regressive conditions.
98
R.B. Perkins and D.Z. Piper
McKelvey et al. (1959; Fig. 4-3) interpreted rocks in southeast Idaho from the upper part of the Grandeur to the middle of the Rex Chert as representing one nearly complete cycle of marine transgression and regression with the central mudstone-rich units of the Meade Peak representing deepest-water conditions and cherty carbonate rocks at the base of the Rex Chert representing a return to shallow-water conditions. The remainder of the Rex Chert was thought to represent a second cycle of transgression and regression. Sheldon (1963; p. 124 and Fig. 82) interpreted the entire Meade Peak, Rex Chert, lower Shedhorn, and lower Franson units as the lower of two transgressive-regressive sequences in Permian rocks of western Wyoming. Hendrix and Byers (2000), working in the Uinta Mountains of Utah, also interpreted the Meade Peak as representing transgressive and highstand tracts, placing the surface of maximum flooding within the middle of the unit. Hiatt (1997) placed the maximum flooding surface at the boundary between the Meade Peak and the overlying Rex Chert Member. We suggest that the maximum flooding surface in the Lakeridge section is represented by the sharp offset in elemental ratios at ~14 m above the base of the Meade Peak. This horizon (bed m40) is also marked by a spike in both apatite and TOC contents, with significantly lower TOC contents persisting throughout the overlying section (Fig. 4-12). We interpret these spikes as being due to lowered terrigenous-sediment input. The maximum flooding surface - the interval that experienced the fastest rate of sea-level rise - should be the most sediment-starved interval, as water-transported terrigenous debris would have been trapped close to the shoreline. Carbonate deposition may also have been unable to keep up with the rapid sea-level rise. Bed m40 has the lowest alumina content (1.25%) of any bed from the basal unit to 17.5 m above the base, and much less than the average of the underlying (8.1%) or overlying (5.7%) 2-3 m. The content of both detrital quartz grains and authigenic silica (from sponge spicules) increases in the beds overlying m40. We interpret these increasingly siliceous beds as representing a shoaling upward cycle during highstand conditions leading to deposition of the overlying calcareous Shedhorn Sandstone and even~ tually, with lowering sea levels, carbonates and bedded evaporate deposits of the Franson Member of the Park City Group. A rise in sea level would be expected to have the most pronounced effect on the accumulation of organic carbon-rich sediments in shallow waters near the basin margin. A third-order rise in sea level of 50-100 m (Ross and Ross, 1995) may have more than doubled the water depth in such settings. A significantly greater depth would translate to oxidation of a substantially greater portion of organic matter during settling (Suess, 1980). In addition, increasing water depths would have allowed for shoreward migration of the zone of intersection of wavebase-generated turbulence along the nearly flat ramp surface, increasing mixing of surficial sediments. Such increased turbulence, mixing, and shoaling are reflected by coarser detrital quartz grains occurring in m43 and overlying beds and in the brecciated and bioclastic textures of samples somewhat higher in the section (beds m45, r50, and r54). This part of the Lakeridge section is also characterized by high Fe203, Ba, and Sc to alumina ratios, consistent with increased shoaling and reworking under higher energy conditions. Agitation of sediment could have led to further oxidation of organic material and hence further lowering the TOC content in these beds (Fig. 4-12).
The Meade Peak Member of the Phosphoria Formation
99
Increased oxidation in the upper Lakeridge is also strongly suggested by the lower metal to Cu ratios, the lower V/Cr ratios, and the high REE contents. The high metal to Zn ratios may be due to a relatively greater hydrogenous influx of metals other than Zn, augmented by lower organic input and a preferential loss of Zn or retention of other metals during early diagenesis. The discrepancy between the metal to Cu and metal to Zn ratios highlights the dangers of relying on any given elemental pair ratio to elucidate paleoenvironmental conditions.
SPATIAL VARIATIONS IN MARINE ELEMENTS General statistical parameters for measured concentrations of the marine fractions of Cd, Cr, Cu, Mo, Ni, Se, V, and Zn are determined for five of the six sections (Table 4-IV). The means of both Cd and Mo increase across the Phosphoria basin from west to east and both the means and maxima for Cd, Mo, and V are significantly lower in the Mud Spring section than in the central and eastern sections. Although TOC contents are also much lower in the Mud Spring section, Cu and Zn concentrations are similar to those in other sections. The lower V concentrations in the Mud Spring section may indicate that bottomwater redox conditions were higher than in other sections due to lower organic loading. The lower Cd and Mo concentrations in this section may be related to variations in algal populations between nearshore and distal areas. A chemofacies classification scheme for paleoenvironments was proposed by Hiatt (1997) based on the ratio of TOC and S values in modern environments where redox conditions are known, with a rather arbitrary allowance for later diagenetic TOC loss. Nearly all of the samples from the various Meade Peak sections, excluding those from the Lakeridge core, plot within the dysoxic and anoxic fields (Fig. 4-13). Most samples plot in the anoxic field. Samples from the Mud Spring and Wheat Creek sections plot within the dysoxic field as they contain little organic carbon or S. Nearly all of the samples that plot in the euxinic field are from the easternmost site, the Lakeridge core and primarily from the upper part of the Meade Peak (beds m38 and above; S/TOC values > 1.2). The high ratios are due to the decreased TOC content (Fig. 4-12) as the average S content in these beds (1.2%) is significantly lower (at the 95% CI) than the average content in the underlying beds (3.7%; t-statistic = 5.4, df = 33). The high S/TOC ratios are, therefore, not likely indicative of euxinic bottom-water conditions, but rather increased retention of S over C within the sediment column. A greater amount of reactive iron (i.e. decoupled from terrigenous siliciclastics; Figs 3 and 6) would allow for metal sulfide formation within anoxic sediments. The upper Lakeridge section is also rich in apatite (up to 60%); as much as half of the average total S content of 1.2% may be incorporated in apatite, which may contain > 2 % S O 4 (Benmore et al., 1983; McArthur, 1985). An enrichment of HREE relative to LREE in seawater occurs from east to west across the basin, as suggested by the shift in the zero-intercept regressed slopes of La versus Yb from section to section (Fig. 4-14). Enrichment of HREE in seawater has been ascribed to preferential complexation of HREE and their resulting retention in solution
TABLE 4-IV General statistics for selected marine element concentrations (mg kg-') in five sections of the Meade Peak Member, Phosphoria Formation Element
Sectiona
Mean
Standard Error
Count
Minimum
Maximum
"MS: Mud Spring (Medrano and Piper, 1995); EV: Enoch Valley (Piper, 1999); HS: Hot Springs (Piper et al., 2000); FC: Fontanelle Creek (Medrano and Piper, 1995); Lakeridge core (Medrano and Piper, 1995).
102
R.B. Perkins and D.Z. Piper
Fig. 4-13. Plot of total S vs. TOC in the Mead Peak Member of the Phosphoria Formation showing position of samples from six sections in dysoxic, anoxic, and euxinic chemofacies fields defined by Hiatt (1997).
Fig. 4-14. La vs. Yb as surrogates for light and heavy REEs in five sections of the Meade Peak Member. The values listed in the legend are zero-intercept slopes and associated errors as calculated by least squares method.
(Sholkovitz et al., 1994). Assuming little biogenic uptake from surface waters, preferential sorption of LREE would be expected where reactive Fe oxyhydroxides are abundant. The high LREE/HREE ratios in eastern sections further suggest some degree of terrigenous sedimentation from the easterly Goose Egg basin, which was rich in Fe oxides, or sediment reworking and preferential retention of REE-rich phases.
The Meade Peak Member of the Phosphoria Formation
103
REE enrichment over terrigenous levels is indicated by the WSA-normalized patterns shown in Fig. 4-11. The flatter patterns represent beds rich in terrigenous debris, with normalized values near one. Interestingly, the basal Meade Peak is generally the most REE-rich bed in most sections. This may be the result of initial concentration of carrier phases (e.g. Fe oxides) under oxic to suboxic sediments prevalent during lowstand conditions and subsequent burial by organic-rich material that would have initiated reductive dissolution and incorporation of the REE in apatite.
CONCLUSIONS Our approach has been to investigate the geochemistry of the Meade Peak Member using ratios of elements associated with either the terrigenous or marine sediment fractions and a key element representing each fraction. Despite the simplicity of this approach and the underlying assumptions, the method is useful in highlighting geochemical changes that might otherwise be masked by lithologic variability and which may be important with respect to understanding the regional Middle Permian environment. The inter-element relationships in the terrigenous component appear useful for chemostratigraphic correlation. A sharp decrease in K20/A1203 ratios occurs in all but the northeasternmost section, where an offset occurs in average and minimum values. These offsets correspond closely to the lower Guadalupian Series boundary coincident with a change from major lowstand to transgressive conditions (Behnken et al., 1986; Ross and Ross, 1995; Hiatt, 1997). The decrease may be related to transgression of the Phosphoria sea on the Wyoming shelf. Assuming that the terrigenous fraction is mostly windtransported material, such a reduction in the K/A1 values may be related to changes in paleoatmospheric circulation, to inundation of particular source areas, or to increased transport distances. Three intervals displaying high Fe203/AI203, Ba/A1203 and Sc/AI203 ratios occur in the upper beds of the easternmost sections. These intervals occur within transgressive to highstand tracts that include the upper Meade Peak and inter-tonguing beds of Rex Chert. The location of these sections near the eastern margins of the basin and the lack of excess Fe, Ba, and Sc over their terrigenous contribution in other sections suggest that Fe-, Ba-, and Sc-rich sediments from southern or eastern sources (i.e. the Goose Egg basin) were either transported into the eastern margin of the Phosphoria basin by shoaling or surface-water transport during maximum-flooding conditions. Altematively, these intervals may reflect sediment reworking and preferential retention of heavy minerals hosting these elements. The westernmost section, presumably representing the deepest portions of the Phosphoria basin, has high Ba/AI203 ratios in the uppermost beds. We suggest that these intervals represent periods of low sediment accumulation during maximum flooding and highstand conditions. Such signals could be particularly useful in correlations of shale-rich basinal sections. Inter-element relationships in the marine fraction imply that bottom waters of the Phosphoria basin were dominantly denitrifying (dysoxic to anoxic), although temporary
104
R.B. Perkins and D.Z. Piper
sulfate-reducing (anoxic to euxinic) conditions may have occurred intermittently during deposition of the most phosphate-rich sections. Ratios of Cd and Mo to Zn and Cu closely approach those in modern plankton in most of the sections, implying a major biogenic source for these elements. Lower values for these ratios occur throughout the westernmost (distal) section, possibly due to different algal populations and relative nutrient uptakes. Large shifts in marine element ratios occur in the upper part of the northeasternmost (shoreward) section. These shifts are accompanied by decreases in organic carbon and an increase in REE concentrations and are coincident with the high Fe203, Ba, and Sc to alumina ratios. We interpret these changes as indicative of increasing redox potential in bottom waters, which nonetheless remained under denitrifying conditions as evidenced by relatively high Cr concentrations. The transition to more oxic conditions occurs at a horizon in the upper-middle part of the Meade Peak Member interpreted as the maximumflooding surface. Elevated oxic conditions persisted through the overlying and increasingly siliceous interval representing highstand conditions. The decrease in organic content could have been due to increased efficiency in oxidation of organic matter via increased water depths (Suess, 1980). However, the occurrence of apatite-rich zones, including pelletal packstones, suggests a relatively high organic input to the sediment column. Therefore, we attribute the noted offsets in the eastern sections to increased oxidation of the sediments resulting from increased mixing at the sediment interface. Removal of finer sediments by winnowing may also have concentrated apatite pellets. Deepening waters would likely have resulted in shoreward migration of the zone of intersection (less dampening) of wavebase turbulence along the gently sloping ramp margin, resulting in the higher energy bottom conditions. Some observations from this study may be of use in interpreting similar systems. 1. 2.
3.
Correlative offsets in terrestrial geochemical signatures in "sediment-starved basins" may be used as chemostratigraphic horizons. Distal areas may contain relatively thick phosphatic layers due to lack of terrigenous dilution. Lower concentrations of several important trace elements, however, may reflect deposition away from areas of peak primary production. Differing populations of primary producers with differing nutrient requirements may also have affected trace-element relationships in distal sections. Both mid-shelf (middle ramp) and marginal environments may have accumulated phosphate-rich layers and high concentrations of trace elements. However, sediments in marginal areas are likely to have the most varied geochemistry because they experienced the greatest variability in terrigenous sediment influx and because even moderate eustatic changes may have dramatic effects on facies, energy levels, sediment mixing, and the amount of organic detritus reaching the sediment surface.
ACKNOWLEDGEMENTS The authors are grateful for the reviews and suggestions of Greg M611er of the University of Idaho and George Desborough of the US Geological Survey, Denver. This
The Meade Peak Member of the Phosphoria Formation
105
work was funded by the US Geological Survey M e n d e n h a l l Postdoctoral Fellowship Program.
REFERENCES Allison, P.A., Wignall, P.B. and Brett, C.E., 1995. Palaeo-oxygenation: effects and recognition. In: D. W. J. Bosence and P. A. Allison (eds.), Marine Palaeoenvironmental Analysis from Fossils. The Geological Society of London Special Publication, pp. 97-112. Baedecker, P.A., 1987. Methods for geochemical analysis. US Geological Survey, Bulletin, vol. 1770, variously paginated. Behnken, EH., Wardlaw, B.R. and Stout, L.N., 1986. Conodont biostratigraphy of the Permian Meade Peak phosphatic shale member, Phosphoria Formation, southeastern Idaho. Contrib. Geol. Univ. Wyoming, 24: 169-190. Benmore, R.A., Coleman, M.L. and McArthur, J.M., 1983. Origin of sedimentary francolite from its sulphur and carbon isotope composition. Nature, 302: 516-518. Bentor, Y.K., 1980. Phosphorites- the unsolved problem. In: Y.K. Bentor (ed.), Marine Phosphorites Geochemistry, Occurrence, and Genesis. Society of Economic Paleontologists and Mineralogists, Special Publication 29, Tulsa, OK, pp. 3-18. Berger, W.H., Fischer, K., Lai, C. and Wu, G., 1988. Ocean carbon f l u x - global maps of primary production and export production. In: C.R. Agegian (ed.), Biogeochemical Cycling and Fluxes Between the Deep Euphotic Zone and Other Oceanic Realms. NOAA Under Sea Research Program Report, pp. 131-176. Bodkin, J.B., 1977. Determination of fluorine in silicates by use of an ion-selective electrode following fusion with lithium metaborate. The Analyst, 102: 409-413. Boyle, E.A., Sclater, F.R. and Edmond, J.M., 1976. On the marine geochemistry of cadmium. Nature, 263: 42-44. Boyle, E.A., Sclater, ER and Edmond, J.M., 1977. The distribution of dissolved copper in the Pacific. Earth Planet. Sci. Lett., 37: 38-54. Breheret, J.G. and Brumsack, H.J., 2000. Barite concretions as evidence of pauses in sedimentation in the Mames Bleues Formation of the Vocontian Basin (SE France). Sediment. Geol., 130: 205-228. Briggs, P.H., 1996. Forty elements by inductively coupled plasma- atomic emission spectrometry. In: B. E Arbogast (ed.), Analytical Methods Manual for the Mineral Resource Surveys Program. US Geological Survey, Open File Report, 96-525, pp. 45-62. Briggs, P.H. and Meier, A.L., 1999. The determination of forty-two elements by inductively coupled plasma-mass spectrometry. US Geological Survey, Open-File Report, 99-165, 15 pp. Broecker, W.S. and Peng, Y.H., 1982. Tracers in the Sea. Eldigio Press, Palisades, NY, 690 pp. Bruland, K.W. and Franks, R.P., 1983. Mn, Ni, Cu, Zn, and Cd in the western North Atlantic. In: C. S. Wong, E. Boyle, K. W. Bruland, J. D. Burton and E. D. Goldberg (eds.), Trace Metals in Sea Water. Plenum Press, New York, 920 pp. Brumsack, H.J., 1986. The inorganic geochemistry of Cretaceous black shales (DSDP Leg 41) in comparison to modern upwelling sediments from the Gulf of California. In: C. P. Summerhayes and N. J. Shackleton (eds.), North Atlantic Paleoceanography. Geological Society of London Special Publication, 21, London, pp. 447-462. Burchette, T.P. and Wright, V.P., 1992. Carbonate ramp depositional systems. Sediment. Geol., 79: 3-57.
106
R.B. Perkins and D.Z. Piper
Carroll, A.R., Stephens, N.P., Hendrix, M.S. and Glenn, C.R., 1998. Eolian-derived siltstone in the Upper Permian Phosphoria Formation; implications for marine upwelling. Geology, 26:1023-1026. Clark, W.J., 1994. The Ervay cycle of the Permian Phosphoria and Park City formations, Bighorn Basin, Wyoming; sequence stratigraphy, facies, and porosity trends. PhD thesis, University of Colorado, Boulder, 606 pp. Codispoti, L.A., 1983. On nutrient variability and sediments in upwelling regions. In: E. Suess and J. Thiede (eds.), Coastal Upwelling, Its Sediment Records; Part A, Responses of the Sedimentary Regime to Present Coastal Upwelling. Plenum Press, New York, pp. 125-145. Collier, R.W., 1985. Molybdenum in the northeast Pacific Ocean. Limnol. Oceanogr., 30: 1351-1354. Collier, R.W. and Edmond, J.M., 1984. The trace element chemistry of marine biogenic particulate matter. Prog. Oceanogr., 13: 113-119. Cook, P.J., 1968. The petrology and geochemistry of the Meade Peak Member of the Phosphoria Formation. PhD Thesis, University of Colorado, 204 pp. Cremer, M.J., Klock, ER., Neil, S.T. and Riviello, J.M., 1984. Chemical methods for analysis of rocks and minerals. US Geological Survey, Open File Report, 84-0565, 148 pp. Cressman, E.R. and Swanson, R.W., 1964. Stratigraphy and petrology of the Permian rocks of southwestern Montana. US Geological Survey, Professional Paper, vol. 313-C, pp. 275-569. Curry, K.J., 1996a. Total carbon by combustion. In: B.E Arbogast (ed.), Analytical Methods Manual of the Mineral Resource Surveys Program. US Geological Survey, Open File Report, 96-525, pp. 99-104. Curry, K.J., 1996b. Total S by combustion. In: B.E Arbogast (ed.), Analytical Methods Manual for the Mineral Resource Surveys Program. US Geological Survey, Open File Report, 96-525, pp. 105-110. de Baar, H.J.W., German, C.R., Elderfield, H. and van Gaans, P., 1988. Rare earth element distribution in anoxic waters of the Cariaco Tranch. Geochim. Cosmochim. Acta, 52: 1203-1219. Emerson, S.R., Cranston, R.E. and Liss, ES., 1979. Redox species in a reducing fjord-equilibrium and kinetic observations. Deep-Sea Res., 26A: 859-878. Filippelli, G.M., 1997. Controls on phosphorus concentration and accumulation in oceanic sediments. In: G.M. Filippelli and K.C. Ruttenberg (eds.), Phosphorus Cycling and Sedimentation in Marine and Freshwater Systems. Elsevier, Amsterdam, pp. 231-240. German, C.R., Holliday, B.P. and Elderfield, H., 1991. Redox cycling of rare earth elements in the suboxic zone of the Black Sea. Geochim. Cosmochim. Acta, 55: 3553-3558. Geslin, J.K., 1998. Distal ancestral Rocky Mountains tectonism - evolution of the PennsylvanianPermian Oquirrh-Wood River Basin, southern Idaho. Geol. Soc. A. Bull., 110: 644-663. Gulbrandsen, R.A., 1979. Preliminary analytical data on the Meade Peak Member of the Phosphoria Formation at Hot Springs underground mine, Trail Canyon Trench and Condo underground mine, southeastern Idaho. US Geological Survey, Open File Report, 79-369, 37 pp. Hageman, P.E. and Welsch, E., 1996. Arsenic, antimony, and selenium by flow injection or continuous f l o w - hydride generation- atomic absorption spectrophotometry. In: B.E Arbogast (ed.), Analytical Methods Manual of the Mineral Resource Surveys Program. US Geological Survey, Open File Report, 96-525, pp. 18-24. Hendrix, M.S. and Byers, C.W., 2000. Stratigraphy and sedimentology of Permian strata, Uinta Mountains, Utah; allostratigraphic controls on the accumulation of economic phosphate. In: C.R. Glenn, L. Prevot-Lucas and J. Lucas (eds.), Marine Authigenesis; From Global to Microbial. Society for Sedimentary Geology (SEPM) Special Publication, Tulsa, OK, pp. 349-367.
The Meade Peak Member of the Phosphoria Formation
107
Hiatt, E.E., 1997. A paleoceanographic model for oceanic upwelling in a late Paleozoic epicontinental sea; a chemostratigraphic analysis of the Permian Phosphoria Formation. PhD thesis, University of Colorado, Boulder, 294 pp. Inden, R.E and Coalson, E.B., 1996. Phosphoria Formation (Permian) cycles in Bighorn Basin, Wyoming, with emphasis on the Ervay Member. In: M.W. Longman and M.D. Sonnenfeld (eds.), Paleozoic Systems of the Rocky Mountain Region. Society for Sedimentary Geology (SEPM), Denver, CO, pp. 379-404. Isbell, J.L. and Cuneo, N.R., 1996. Depositional framework of Permian coal-bearing strata, southern Victoria Land, Antartica. Palaeoceanogr., Palaeoclimatol., Palaeoecol., 125:217-238. Jackson, L.L., Brown, EW. and Neil, S.T., 1987. Major and minor elements requiring individual determination, classical whole rock analysis, and rapid rock analysis. In: P. A. Baedecker (ed.), Methods of Geochemical Analysis. US Geological Survey, Bulletin, vol. 1770, pp. G 1-G23. Knudsen, A.C., Gunter, M.E., Herring, J.R. and Grauch, R.I., 2001. Mineralogical characterization of strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation; channel and individual rock samples of measured section J and their relationship to measured sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 02-0125, 38 pp. Kolodny, Y., 1981. Phosphorites. In: C. Emiliani (ed.), The Sea- Ideas and Observations on Progress in the Study of the Seas. Wiley-Interscience, New York, pp. 981-1023. Lichte, EE., Golightly, D. and Lamothe, P.J., 1987. Inductively coupled plasma-atomic emission spectrometry. In: P.A. Baedecker (ed.), Geochemical Methods of Analysis: US Geological Survey, Bulletin, vol. 1770, pp. B I-B 10. Littke, R., Klussmann, U., Krooss, B. and Leythaeuser, D., 1991. Quantification of loss of calcite, pyrite, and organic matter due to weathering of Toarcian black shales and the effects on kerogen and bitumen characteristics. Geochim. Cosmochim. Acta, 55: 3369-3378. Martin, J.H. and Knauer, G.A., 1973. The elemental composition of plankton. Geochim. Cosmochim. Acta, 37:1639-1653. Maughan, E.K., 1966. Environment of deposition of Permian salt in the Williston and Alliance Basins. In: J.L. Rau (ed.), Second Symposium on Salt. The Northern Ohio Geological Society, Cleveland, pp. 35-47. Maughan, E.K., 1984. Geological setting and some geochemistry of petroleum source rocks in the Permian Phosphoria Formation. In: J. Woodward, EE Meissner and J.L. Clayton (eds.), Hydrocarbon Source Rocks of the Greater Rocky Mountain Region. Rocky Mountain Association of Geologists, Denver, CO, pp. 281-294. Maughan, E.K., 1994. Phosphoria Formation (Permian) and its resource significance in the Western Interior, USA. In: A.E Embry, B. Beauchamp and D.J. Glass (eds.), Pangea; Global Environments and Resources. Canadian Society of Petroleum Geologists, Calgary, pp. 479-495. McArthur, J.M., 1985. Francolite geochemistry; compositional controls during formation, diagenesis, metamorphism and weathering. Geochim. Cosmochim. Acta, 49: 23-35. McKelvey, V.E., Armstrong, EC. and Smith, L.E., 1953. Stratigraphic sections of the Phosphoria Formation in Wyoming, 1947-1948. US Geological Survey, Circular, 0210, 35 pp. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M. and Swanson, R.W., 1959. The Phosphoria, Park City and Shedhorn Formations in the western phosphate field, US Geological Survey, Professional Paper, vol. 313-A, 47 pp.
108
R.B. Perkins and D.Z. Piper
Medrano, M.D. and Piper, D.Z., 1995. Partitioning of minor elements and major-element oxides between rock components and calculation of the marine derived fraction of the minor elements in rocks of the Phosphoria Formation, Idaho and Wyoming, US Geological Survey, Open File Report, 95-0270, 79 pp. M/iller-Karger, E, Varela, R., Thunell, R., Scranton, M., Bohrer, R., Taylor, G., Capelo, J., Astor, Y., Tappa, E., Ho, T.-Y. and Walsh, J.J., 2001. Annual cycle of primary productivity in the Cariaco Basin - response to upwelling and implications for vertical export. J. Geophys. Res., 106: 4527-4542. Murata, K.J., Friedman, I. and Gulbrandsen, R.A., 1972. Geochemistry of carbonate rocks in Phosphoria and related formations of the western Phosphate Field. US Geological Survey, Professional Paper, vol. 800-D, pp. D 103-D 110. Murray, J.W., Spell, B. and Paul, B., 1983. The contrasting geochemistry of manganese and chromium in the eastern tropical Pacific Ocean. In: C.S. Wong, E. Boyle, K.W. Bruland, J.D. Burton and E.D. Goldberg (eds.), Trace Metals in Seawater, Plenum Press, New York, NY, pp. 643-669. Papp, C., Brandt, E. and Aruscavage, P., 1996. Carbonate carbon by coulemetric titration. In: B.F. Arbogast (ed.), Analytical Methods Manual for the Mineral Resource Surveys Program, US Geological Survey, Open File Report, 96-525, pp. 38-44. Pedersen, T.F. and Calvert, S.E., 1990. Anoxia vs productivity; what controls the formation of organic-carbon-rich sediments and sedimentary rocks? AAPG Bull., 74: 454-466. Peterson, J.A., 1980a. Depositional history and petroleum geology of the Permian Phosphoria, Park City, and Shedhorn formations, Wyoming and southeastern Idaho. US Geological Survey, Open File Report, 80-667, 42 pp. Peterson, J.A., 1980b. Permian paleogeography and sedimentary provinces, West Central United States. In: T. D. Fouch and E. R. Magathan (eds.), Paleozoic Paleoceanography of West-Central United States. Society of Economic Paleontologists and Mineralogists, Rocky Mountain Section, Denver, pp. 271-292. Piper, D.Z., 1999. Trace elements and major-element oxides in the Phosphoria Formation at Enoch Valley, Idaho; Permian sources and current reactivities. US Geological Survey, Open File Report, 99-0163, 66 pp. Piper, D.Z., 2001. Marine chemistry of the Permian Phosphoria Formation and basin, Southeast Idaho. Econ. Geol., 96: 599-620. Piper, D.Z. and Isaacs, C.M., 2001. The Monterey Formation: Bottom water redox conditions and photic zone primary productivity. In: C. M. Isaacs and J. Rullkotter (eds.), Monterey Formation from Rocks to Molecules. Columbia University Press, New York, pp. 31-58. Piper, D.Z. and Link, P.K., 2002. An upwelling model for the Phosphoria Sea - a Permian oceanmargin sea in the northwest United States. Am. Assoc. Pet. Geol. Bull., 86: 1217-1235. Piper, D.Z., Skorupa, J.P., Presser, T.S., Hardy, M.A., Hamilton, S.J., Huebner, M. and Gulbrandsen, R.A., 2000. The Phosphoria Formation at the Hot Springs Mine- a source of selenium and other trace elements to ground water, surface water vegetation and biota. US Geological Survey, Open File Report, 00-050, 75 pp. Poole, F.G., 1964. Paleowinds in the western United States. In: A.P.M. Narin (ed.), Problems in Paleoclimatology. John Wiley and Sons, London, pp. 394-405. Redfield, A.C., Ketchum, B.H. and Richards, EA., 1963. The influence of organisms on the composition of seawater. In: M.N. Hill (ed.), The Composition of Seawater- Comparative and Descriptive Oceanography. Wiley-Interscience, New York, pp. 26-77. Ross, C.A. and Ross, J.R.P., 1987. Late Paleozoic accreted terranes of western North America. In: C. H. Stevens (ed.), Pre-Jurassic Rocks in Western North American Suspect Terranes. Pacific
The Meade Peak Member of the Phosphoria Formation
109
Section of Society of Economic Paleontologists and Mineralologists, Los Angeles, CA, pp. 7-22. Ross, C.A. and Ross, J.R.P., 1995. Permian sequence stratigraphy. In: P. A. Scholle, T.M. Peryt and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea; Volume I, Paleogeography, Paleoclimates, and Stratigraphy. Springer-Verlag, Berlin, pp. 98-123. Rossignol-Strick, M., Nesteroff, W., Olive, P. and Vergnaud-Grazzini, C., 1982. After the deluge: Mediterranean stagnation and sapropel formation. Nature, 295:105-110. Ryan, W.B.E and Cita, M.B., 1977. Ignorance concerning episodes of ocean-wide stagnation. Mar. Geol., 23:197-215. Sarnthein, M., Winn, K., Duplessy, J.C. and Fontugne, M.R., 1988. Global variations of surface ocean productivity in low and mid latitudes- influence on CO2 reservoirs on the deep ocean and atmosphere during the last 21,000 years. Paleoceanography, 3: 361-399. Schuffert, S.D., Jahnke, R.A., Kastner, M., Leather, J., Struz, A. and Wing, M.R., 1994. Rates of formation of modern phosphorite off western Mexico. Geochim. Cosmochim. Acta, 58: 5001-5010. Sheldon, R.P., 1963. Physical stratigraphy and mineral resources of Permian rocks in western Wyoming. US Geological Survey, Professional Paper, vol. 313-B, pp. 49-273. Sheldon, R.P., Cressman, E.R., Carswell, L.D. and Smart, R.A., 1954. Stratigraphic sections of the Phosphoria Formation in Wyoming. US Geological Survey, Circular, 325, 24 pp. Sheldon, R.P., Waring, R.G., Warner, M.A. and Smart, R.A., 1953. Stratigraphic sections of the Phosphoria Formation in Wyoming, 1949-1950. US Geological Survey, Circular, 307, 45 pp. Sholkovitz, E.R., Landing, W.M. and Lewis, B.L., 1994. Ocean particle chemistry: the fractionation of rare earth elements between suspended particles and seawater. Geochim. Cosmochim. Acta, 58: 1567-1579. Smart, R.A., Waring, R.G., Cheney, T.M. and Sheldon, R.P., 1954. Stratigraphic sections of the Phosphoria Formation in Idaho, 1950-1951. US Geological Survey, Circular, 327, 22 pp. Suess, E., 1980. Particulate organic carbon flux in the oceans; surface productivity and oxygen utilization. Nature, 288: 260-263. Taggart, J.E., Lindsey, J.R., Scott, B.A., Vivit, D.V., Bartel, A.J. and Stewart, K.C., 1987. Analysis of geologic materials by wavelength-dispersive X-ray fluorescence spectrometry. In: P. A. Baedecker (ed.), Methods for Geochemical Analysis. US Geological Survey, Bulletin, vol. 1770, pp. E I-E 19. Thunell, R.C., 1998. Particle fluxes in a coastal upwelling zone - sediment trap results from Santa Barbara Basin, California. Deep-Sea Res., II, 45: 1863-1884. Tisoncik, D.D., 1984. Regional lithostratigraphy of the Phosphoria Formation in the Overthrust Belt of Wyoming, Utah and Idaho. In: J. Woodward, EE Meissner and J.L. Clayton (eds.), Hydrocarbon source rocks of the Greater Rocky Mountain region. Rocky Mountain Association of Geologists, Denver, CO, pp. 295-320. Torres, M.E., Brumsack, H.J., Bohrmann, G. and Emeis, K.C., 1996. Barite fronts in continental margin sediments; a new look at barium remobilization in the zone of sulfate reduction and formation of heavy barites in diagenetic fronts. Chem. Geol., 127: 125-139. Tsoar, H. and Pye, K., 1987. Dust transport and the question of desert loess formation. Sedimentology, 34:139-153. Tsunogai, S. and Noriki, S.A., 1987. Organic matter fluxes and the sites of oxygen consumption in deep water. Deep-Sea Res., 34: 755-767. Turekian, K.K. and Wedepohl, K.H., 1961. Distribution of the elements in some major units of the Earth's crust. Geol. Soc. A. Bull., 72: 175-191.
1 10
R.B. Perkins and D.Z. Piper
Wardlaw, B.R. and Collinson, J.W., 1986. Paleontology and deposition of the Phosphoria Formation. In: D.W Boyd and J.A. Lillegraven (eds.), Western Phosphate Deposits. University of Wyoming, Laramie, Contributions to Geology, pp. 107-142. Webster, R., 1973. Automatic soil-boundary location from transect data. J. Int. Assoc. Math. Geol., 5: 27-37. Wedepohl, K.H., 1969-1978. Handbook of Geochemistry. 4 Volumes, Springer-Verlag, Berlin. Whalen, M.T., 1996. Facies architecture of the Permian Park City Formation, Utah and Wyoming; implications for the paleogeography and oceanographic setting of western Pangea. In: M.W. Longman and M.D. Sonnenfeld (eds.), Paleozoic Systems of the Rocky Mountain Region. Society for Sedimentary Geology (SEPM), Denver, CO, pp. 355-378.
Li[e Cycle of the Phosphoria Formation." From Deposition to Post-Mining Environment Edited by James R. Hein Handbook o['Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
111
Chapter 5
REGIONAL ANALYSIS OF SPICULITE FAUNAS IN THE PERMIAN PHOSPHORIA BASIN: IMPLICATIONS FOR PALEOCEANOGRAPHY B.L. MURCHEY
ABSTRACT The sponge spiculites of the Permian Phosphoria basin, Antler high, and eastern Havallah basin were the southernmost expression of one of the largest spiculite belts in the Earth's history. This spiculite belt extended from Nevada to the Barents Sea. In Idaho and Nevada, the spicule populations of this belt are dominated by demosponge spicules and are distinctive for their abundant rhax microscleres, large monaxons, and lithistid desmas. They form an Eastern Belt of spiculites that interfingers with spicule assemblages derived from choristid demosponges and hexactinellids that lived along the eastern margin of the deeper Havallah basin. The Havallah basin assemblages are similar to those in Permian arc terranes to the west, and together the sponge populations in this domain constitute a distinct Central Belt. Radiolarians are virtually absent in the siliceous microfossil populations of the Eastern Belt, abundant in the populations of the Central Belt, and dominant in the populations of a Western Belt confined to Mesozoic accretionary complexes in the Pacific Coast States. The scattered sponge spicules in the Western Belt radiolarites were derived from hexactinellids. During the Permian, the relative abundance and apparent diversity of siliceous sponges expanded over a wide range of depths in the basins from Nevada and Idaho to the open ocean. Radiolarian preservation and apparent diversity increased in the deeper Cordilleran basins as well. In the Arctic regions, significant sponge spiculites were deposited in epicratonic basins. At the same time that siliceous sponge populations expanded along the northwestern margin of Pangea, warm-water carbonate producers disappeared. Suppression of carbonate-producing organisms along the margin was critical to the accumulation and preservation of both the demosponge spiculites in the Eastern Belt and the spicule-rich argillites of the Central Belt. Vigorous thermohaline circulation was the major control on the paleobiogeography of the late Early, Middle, and early Late Permian along northwest Pangea. It was driven by cold, nutrient- and oxygen-rich northern waters and it produced a coastal current that swept down the margin of the supercontinent. The upwelling associated with deposition of world-class phosphorites in the Phosphoria basin was a part of this larger oceanographic system.
1 12
B.L. Murchey
INTRODUCTION Siliceous sponge spiculites in the Permian Phosphoria Formation of Idaho and adjacent States are remnants of an immense belt of spiculite deposits that rimmed the northwestern quadrant of Pangea. The belt stretched from central Nevada and Idaho (Fig. 5-1) through Alberta and British Columbia (not given in the figure) all the way to the seas off Norway and Arctic Canada (Fig. 5-2) (Murchey and Jones, 1992; Beauchamp and Baud, 2002). Even if the Pangean spiculite belt may have been locally discontinuous, its aggregate length eclipsed that of any older or younger spiculite accumulations along western and northern North America. The Permian spiculites are economically important because they form oil and gas reservoirs in the northern seas (Ehrenberg et al., 2001), alternate with the world-class phosphorite deposits in the Phosphoria basin (this volume), and host disseminated gold deposits in central Nevada (Murchey et al., 1995).
Fig. 5-1. Map showing present locations of Permian basins and terranes in Idaho, Nevada, California, Oregon, and Washington. Terrane names follow Silberling et al. (1987). Numbers correspond to samples listed in Table 5-II and mentioned in the text. The shaded area encompasses localities from which rhax-bearing, demosponge-dominated spiculites (Eastern Belt) are documented. The approximate locations of the Central and Western Belt assemblages discussed herein are illustrated as well.
Regional analysis of spiculite faunas in the Permian Phosphoria basin
1 13
Fig. 5-2. Paleogeographic map of the Arctic region showing the distribution of Permian siliceous marine deposits, primarily spiculites, based on a figure in Beauchamp and Baud (2002). Numbers correspond to samples listed in Table 5-II and mentioned in the text.
Siliceous sponge spicules are the principal allochems of the Middle Permian Rex and Tosi Chert Members of the Phosphoria Formation and they are common sedimentary components of other members. For this reason, the total volume of siliceous sponges in the Phosphoria exceeds that of any other fossil group. Dissolved silica from dissolution of sponge spicules was the primary source of the silica in bedded and nodular Phosphoria chert (Sheldon, 1957). Despite the abundance of sponge spicules, the faunal compositions of the spiculites are less well-documented than the conodont and megafossil faunas of the Phosphoria (e.g. Yochelson, 1968; Wardlaw and Collinson, 1986) primarily because disaggregated spicules are not very useful for dating purposes or species identification. They have the potential, however, to provide new information regarding the environment of the Phosphoria basin and its relationship to basin systems along and across strike. This study will characterize siliceous microfossil assemblages in the southern part of the Permian spiculite belt from the Phosphoria basin in Idaho to the eastern Havallah basin and Antler high in central Nevada. In order to define and constrain the width of the spiculite belt, Permian siliceous microfossil assemblages will be compared across a broad east to west transect from Idaho and Nevada to California, Oregon, and Washington. The transect traverses paleogeographic settings from the epicontinental margin, across the deep Havallah basin, to island arc terranes and offshore oceanic basins. Siliceous sponge spicules and (or) radiolarians accumulated in many of these settings during the latter half of the Permian, an interval of especially widespread deposition of siliceous sediment in the global ocean (Murchey and Jones, 1992, 1994). In addition, a very preliminary comparison between the southern and northern siliceous spiculite belts will be made, based on the fossils from a few chert samples from the Chukchi Sea and northern Alaska.
1 14
B.L. Murchey
BACKGROUND AND PREVIOUS STUDIES The Phosphoria basin of Idaho and parts of Montana, Wyoming, Utah, and Nevada was a large marine embayment that formed above the autochthonous craton margin of western Pangea (Fig. 5-1). The largest Permian phosphorite deposits in the world accumulated in the Phosphoria basin during the latest Early and Middle Permian. Those deposits cyclically alternate with spiculite accumulations as eustatic sea levels rose and fell (Hein et al., Chapter 2). The eastern and southern margins of the basin were flat ramps across which the Phosphoria Formation intertongued with the Park City Group (Wardlaw and Collinson, 1986; Hiatt, 1997; Hein et al., Chapter 2). The southern margin of the Phosphoria basin trended due west across the Utah panhandle into the northeastern corner of Nevada where it was truncated by the Antler orogenic belt (Stevens, 1991). The western side of the basin is hidden by the Idaho batholith, which leaves key information regarding the basin topography in the realm of speculation. If the Antler high extended northward from Nevada, it would have formed a ridge, island chain, or sill along the western side of the Phosphoria basin. From the southwestern corner of the Phosphoria basin, the Antler high tracks southwest and then south to central Nevada (Stevens, 1991) (Fig. 5-1). This welt of Lower Paleozoic marine strata, also known as the Roberts Mountain allochthon, was thrust over the continental margin along the Roberts Mountain thrust fault in the Middle Paleozoic (Roberts, 1964). Stevens (1991) illustrated the Antler high as an emergent ridge during the Middle Permian, as very little marine strata of that age had been identified along the belt. However, as the result of gold exploration along the Golconda thrust fault, the Wordian Edna Mountain Formation has been recognized on the western side of the trend in a number of new localities (Murchey et al., 1995). In its type area, the Edna Mountain includes marine sandstone, siltstone, and minor conglomerate. Abundant siliceous sponge spicules and phosphatic coated grains are distinguishing characteristics of the fine-grained sandstone and brown siltstone intervals in the Edna Mountain (Murchey et al., 1995). The remapped Edna Mountain and its equivalents are used herein to help define the known southern limits of the Pangean spiculite belt (Fig. 5-1). Moore and Murchey (1998) also documented latest Early and Middle Permian radiolarian- and sponge-bearing argillite within the overlap sequence deposited on the Antler high in the Shoshone Range of central Nevada. The expansion of the number of known Middle Permian marine deposits along the Antler high suggests that it may have formed a chain of islands during Phosphoria time. Throughout the Late Paleozoic, the Antler high marked the boundary between western marine basins in which radiolarians were common rock-formers and eastern marine basins where radiolarians were only episodically present and were volumetrically insignificant. The Antler high formed the autochthonous eastern margin of the Late Paleozoic Havallah basin, which herein includes the Schoonover basin of northeastern Nevada (Fig. 5-1). The deeper reaches of the Late Devonian through Permian basin were characterized by pillow basalt, radiolarian chert, and turbiditic sandstone derived from the Antler high (Murchey, 1990; Whiteford, 1990). In central Nevada near Battle Mountain, basin and slope facies of the Havallah basin were thrust eastward over the Edna Mountain Formation no earlier than
Regional analysis of spiculite faunas in the Permian Phosphoria basin
115
the late Wordian (Roberts, 1964). The Golconda thrust fault marks the eastern edge of the allochthon (Fig. 5-1). The absence of a Triassic overlap sequence across the fault and the pervasiveness of the Mesozoic tectonic overprint on the history of Nevada have created controversy about the age of the Golconda thrust fault. Not all regional geologists believe that the present position of the allochthon's leading edge predates the Jurassic (Ketner, 1998). Nevertheless, the closure of part of the Havallah basin was well underway when the Triassic Koipato sequence of the Tobin Range was deposited unconformably on imbricated stacks of uppermost Devonian and Mississippian through Lower Permian deep-marine strata (Stewart et al., 1986). I suggest that the more westward location of deep-marine Triassic depocenters and the absence of Triassic strata within the deformed assemblage indicate that the Havallah basin had substantially closed by the end of the Permian. In the Middle Permian, therefore, the Havallah basin was more likely a relatively narrow trough than a wide and expansive marginal sea. In some paleogeographic models for the Phosphoria Formation, the Havallah basin is projected northward into western Idaho beneath the Idaho batholith and west of the Phosphoria basin (e.g. Hiatt, 1997). Permian radiolarian- and spicule-bearing argillaceous slope facies of the eastern Havallah basin have been documented both above and below the Golconda thrust fault. Murchey (1990) described them in the Havallah tectonic assemblage immediately overlying the Golconda thrust near Antler Peak. In this area, argillite and siliceous argillite interleave with packages of black spiculitic chert that represent turbidite deposits (Murchey, 1990). The Permian spiculitic chert had previously been correlated with older radiolarian chert (Roberts, 1964). In the Shoshone Range to the south, Permian radiolarian- and spicule-bearing argillaceous strata form part of the autochthonous Antler overlap sequence, as previously mentioned, and Permian black spiculitic chert occurs in allochthonous strata above the Golconda thrust fault (Moore and Murchey, 1998). The latter is the basis for extending the southern limit of the Permian sponge spiculite belt at least as far south as the Shoshone Range (Fig. 5-1). One or more Permian island arc systems were likely located to the west of the Phosphoria and Havallah basins. In southern Nevada, the structurally highest tectonic units in the Havallah tectonic assemblage tie the basin to an active Permian arc by virtue of redeposited volcanic and plutonic material (Whiteford, 1990). Permian arc remnants are preserved in a discontinuous belt of accreted terranes from California and northwestern Nevada to northeastern Oregon. The Permian arc remnants included in this study follow the terminology of Silberling et al. (1987): the Northern Sierra terrane of California, Black Rock terrane of northwestern Nevada, Eastern Klamath terrane of California, and the Grindstone terrane of northern Oregon (Fig. 5-1). With the possible exception of the Northern Sierra terrane, these arc fragments share a number of characteristics including a similar association of Early and Middle Permian fusulinids and megafossils, commonly called the McCloud fauna (Stevens et al., 1990). Although the arc remnants are the most obvious candidates for a volcanic archipelago off the western margin of the greater Havallah basin (Harwood and Murchey, 1990), there is no universal agreement that the McCloud faunal province was located near the Cordilleran margin (Stevens et al., 1990).
116
B.L. Murchey
West and north of the Permian arc terranes, Early Mesozoic accretionary complexes contain sequences of Permian chert and siliceous mudstone. Samples from the North Fork and Hayfork terranes of California, the Baker terrane of Oregon, and the Hozameen and San Juan terranes of Washington were included in the regional comparisons of this study (Fig. 5-1). These rocks are associated with basalt and their pelagic protoliths were probably deposited on oceanic crust (Murchey and Jones, 1994). This interpretation is supported by the presence in these terranes of Permian limestone blocks with fusulinids of Tethyan affinity. The blocks are exotic to North America and may have formed on fartravelled seamounts (Stevens, 1991). In summary, the Phosphoria embayment was the easternmost marine basin in a series between the interior coastline and the open ocean. During the Permian, siliceous marine sediments were deposited in all the basins (Murchey and Jones, 1992). The distribution patterns of the two main fossil groups in these strata, radiolarians and siliceous sponge spicules, can help to define provinces within the expansive tract of siliceous deposits. In a previous study, Murchey and Jones (1994) compared Permian siliceous microfossil faunas from the allochthonous and parautochthonous terranes west of the Golconda thrust
Fig. 5-3. Modified X-Y diagram showing the ratio of ruzhencevispongacid radiolarians to albaillellacids (horizontal axis) vs. percentage of sponge spicules in total microfossil population (vertical axis) for Permian fauna from l0 tectonostratigraphic terranes in the western United States (Murchey and Jones, 1994) as well as from the Phosphoria basin and Antler high (Fig. 5-1). The shaded area is based on population counts illustrated as point data in Murchey and Jones (1994). The black-filled circle corresponds to new results from spiculites in the Phosphoria basin, eastern Havallah basin, and the Antler high. The three belts of sponge spicule assemblages distinguished in this study fall in different regions of the diagram.
Regional analysis of spiculite faunas in the Permian Phosphoria basin
117
fault (Fig. 5-1). On the basis of population counts of radiolarians and siliceous sponge spicules, two major paleogeographic domains could be distinguished based on their quantitatively different faunal characteristics. The domains were defined by the ratio of two major radiolarian groups to one another and the ratio of radiolarians to sponge spicules. A domain characterized by assemblages containing both abundant radiolarians and sponge spicules encompasses the Permian volcanic arc terranes and the allochthonous part of the Havallah basin. The radiolarian populations in these assemblages were notable for their high concentrations of ruzhencevispongacid radiolarians relative to albaillellacids (Murchey and Jones, 1994) (Fig. 5-3). A western domain, characterized by abundant radiolarians but few or no sponge spicules, is defined by the assemblages in the oceanic terranes (Fig. 5-1). The radiolarian populations are notable for their very low ratios of ruzhencevispongacid radiolarians relative to albaillellacids. Whereas Murchey and Jones (1994) focused primarily on the characterization of radiolarian populations in terranes west of the Golconda fault, this study focuses on the characterization of siliceous microfossil faunas deposited east of the Golconda thrust in the Phosphoria basin and along the Antler high, as well as the sponge spicule assemblages in the more western basins. One goal is the expansion of the criteria that can be used for characterization of geographically distinct faunal assemblages.
METHODS Hundreds of rock samples from Paleozoic marine deposits of the western States were processed for siliceous microfossils. First, sponge spicules and radiolarians were etched from the rock matrix using diluted hydrofluoric acid (10% o f - - 5 0 % concentrate) (Dumitrica, 1970; Pessagno and Newport, 1972). Then, the fossils were washed off the rock surface and collected on Tyler-equivalent 250-mesh (63 Ixm openings) and 80-mesh (180 Ixm openings) screens. The fossils were examined with a binocular microscope. Of the processed samples with reasonably well-preserved siliceous microfossil faunas, 99 were determined to be Permian based on their contained fossils or on the fossil ages of surrounding strata.
Identification of sponge spicule morphotypes Ten sponge spicule morphotypes accounted for an estimated 99% of the sponge populations (Table 5-I). Their presence or absence in each sample is recorded in Table 5-II. More than 500 sponge spicules were commonly examined when sufficient quantities were extracted. In Table 5-II, X signifies that three or more specimens of a particular morphotype were observed, while R (rare) signifies that only one or two specimens were observed.
118
B.L. Murchey
TABLE 5-I List and brief description of the most common siliceous sponge spicules in Permian rocks of the western Cordillera: Idaho, Nevada, California, Oregon, and Washington. Their distribution in individual fossil assemblages is recorded in Table 5-II Spicule type
Comments
Monaxon
Class Demospongiae and Hexactinellida; broken single(?) axons and single axons with pointed ends (oxea monaxons) are the most abundant spicules in almost all samples; large monaxon spicules are present only in samples with few or no radiolarians - they are probably demosponges Class Demospongiae; small kidney-bean-shaped spicule, a microsclere; probably a demosponge selenaster Class Demospongiae, likely lithistids: polyaxon, polyactine; Tricranoclones are distinct forms that may prove to have paleogeographic significance Class Demospongiae, Hexactinellida(?); barbell-shaped monaxons Class Demospongiae, possible choristid; 4-axon, 4-actine; three actines curve away from fourth Class Demospongiae, possible choristid; 4-axon, 4-actine of approximately equal length, tetragonal symmetry Class Demospongiae, possible choristid; 4-axon, 4-actine; three actines curve downward toward fourth Class Hexactinellida; anchor-shaped spicules, some associated with root tuft Class Hexactinellida; modern forms restricted to the deep-water Amphidiscophora; morphotypes include paraclavules (ring of recurved spines at one end of straight shaft), hemidiscs (ring of recurved spines and one end of straight shaft, tiny ring of recurved spines at other end), and amphidiscs (equant rings of recurved spines at each end of shaft) Class Hexactinellida; 3-axon, 6-actine [5-, 4-actine]; more than 99% of the morphotypes observed in this study have pointed non-bifurcating terminations (oxyhexactine); variations of hexactines, such as pentactines and stauractines, are included in this category because they are difficult to distinguish from broken hexactines, and they occur with hexactines.
Rhax Desma Strongyle Protriaene Calthrops Anatriaene Anadiaene Birotule
Hexactine
Quantitative comparison of sponge spicules to radiolarians The 250-mesh screen residue (size fraction 63-180 txm) was used for relative comparisons of abundance. This size fraction commonly has the greatest abundance and diversity o f both sponge spicules and radiolarians. A ratio of sponge spicules to the total microfossil population (radiolarians plus spicules) was obtained by strewing the 250-mesh residue on a picking tray and counting 50-100 specimens lying on a line. The ratio is represented as a percentage (%S250 = (sponge/sponge + radiolarians)25o x 100%). The %S25o value for each sample is given in Table 5-II, which includes the data from Murchey and Jones (1994) for terranes lying west of the Golconda thrust fault. .
Regional analysis of spiculite faunas in the Permian Phosphoria basin
119
RESULTS An eastem belt of rhax-bearing, demosponge-dominated spiculites can be distinguished from the two faunal domains previously defined by Murchey and Jones (1994). The differences are based on quantitative as well as qualitative observations. This belt can be traced from the Phosphoria basin in Idaho to the southwest and south along the Antler high (Fig. 5-1). Redeposited spicules from the Antler high can be recognized in the structurally lowest faultbounded strata above the Golconda thrust fault. The eastern spiculite belt appears to have characteristics similar to at least some of the spiculite deposits in the Arctic.
Eastern Belt: rhax-bearing, demosponge-dominated spiculite assemblages Twenty-four samples from chert or siltstone in Idaho and Nevada are characterized by demosponge spicules (Table 5-II; Samples 1-24). Large monaxons coupled with the presence of rhaxes, strongyles, and spheroidal microscleres distinguish these spiculites from all other Permian samples from the conterminous western States. The monaxon spicules of probable demosponge (Class Demospongiae) origin are abundant in all samples. Their axial canals are commonly filled with dark material, including collophane or a related phosphate mineral. Desmas from lithistid demosponges are generally more common than protriaene, anatriaene, or calthrop spicules of probable choristid demosponges. Oxyhexactine hexactinellid spicules (Class Hexactinellida) are present in many samples, but are a small component, a few percent at most, of the total spicule population. Radiolarians are absent in all but two samples which contain less than 1% radiolarians with no albaillellacid forms (Fig. 5-2).
Rex Chert of the Phosphoria Formation in southeastern Idaho, central basin (Table 5-11; Samples 1-3) Three samples from the Rex Chert Member of the Phosphoria Formation in southeastern Idaho yielded abundant sponge spicules but no radiolarians. Two composite samples (Samples 1 and 2, Table 5-I) were collected from an outcrop of the Rex Chert Member at Trail Canyon (locality 1206 in McKelvey et al., 1953a). At this outcrop, the Rex Chert consists of 7 m of black chert with beds 2.5-30 cm thick overlain by about 9 m of pale-gray chert with beds 16-61 cm thick. The black chert beds in the lower part of the outcrop contain no shale partings. Laterally, several beds commonly fuse into a single bed. The acidetched surfaces of individual beds reveal fine sedimentary laminae. The sponge spicules in this unit (Sample 2) are small, densely packed, and fairly diverse. In decreasing order of abundance, the spicules include broken monaxons (abundant), rhaxes (common), spheroidal microscleres (common), anatriaenes (rare), desmas (rare), hexactines (rare), calthrops (rare), and monaxon strongyles (rare). The pale-gray chert beds in the upper part of the outcrop have undulating bedding surfaces. The beds are densely packed with large,
B.L. Murchey
120 T A B L E 5-II
Sponge spicule data for individual samples. X indicates the presence of spicule morphotypes
in
r e s i d u e c o l l e c t e d o n a 2 5 0 - m e s h s c r e e n (at l e a s t t h r e e o b s e r v e d ) . R i n d i c a t e s o n l y o n e o r t w o s p e c i m e n s o b s e r v e d . L a n d S i n d i c a t e l a r g e or s m a l l m o n a x o n
spicules SPONGE SPICULE MORPHOTYPES
en
~
~
SE IDAHO,
O ~
1
My 700
100
L
~
2 3 4
My699 Lico 2 13681A
100 100 99+
L L L
"-
5 6
13681B 13679 13686B 91 ~ r r
99+ 100 !00 !00
L
L L L
X X X
X X
100 100 100 100 100 100 100 100 100 100 100 100
L L L L L L L L L L L L
X X X X X X X
X X
100
L
100 100 100 100 1oo 100 100
L L L L L L L
92-P34 (292-332) 100
L
..~
Rex Chert Member of the Phosphoria Formation L~ N.EAST NEVADA Mt. lchabodarea Unnamed black chert
~ ~~ r ~..~
7
N.CENTRAL NEVADA
8
~ ~ Edna Mountain Formation ~ ~ phosphatic siltstone: ~ L~ ~ ~ m CENfRAL NEVADA ~ Havallah assemblage: ~ ~ Black spiculitic chert, 8 ~
silty--turbidites
~ ~
13-22: North Central NV Battle Mt.,Antler Peak
L; L~ ~
23_24:
O
~
m
co~
ShoshoneRange CHUKCHI SEA, NORTHWIND RIDGE ~ Chert clasts derived from a~ < Van Hauen Fm. equivalent(?)
9 10 11 12 13 14 15
911"I"046 91"II"064 91"I/'065 9 ITI'~6 82MYI6D 82MY19 82MY22 1716 82MY21 MYI64 ..~ IS MY165 "~ 19 MYI69 ~: 20 MYI73 21 MYi84 22 MY317 23 13 24 19 25 93-PI6#1 r 26 #14 #17-1 28 #17-2 29
X X
X
X
R R R
R X X
X
X X
~,R X
R X R
R
X
X
X
R
X X R R
X X X X X X X
X X
X X
X
X
X
R
X X X
X X x
X x
X
Continued
Regional analysis of spiculite faunas in the Permian Phosphoria basin
121
TABLE 5-II Continued SPONGE SPICULE MORPHOTYPES m
oo
N. CENTRAL NEVADA Havallah Assemblage:
NORTHWEST NEVADA
30 31 32 33 34 35 36 37 38 39 40 41
82MYI6X 82MY 18 82MY20 82MY24 82MY23 83MY167 83MY171 84MY325 84MY330 84MY342 82MY352 5
60 52 95 40 20 51 23 80 50 40 60 I
S S S S S S S S S S S S
BLACKROCKTERRANE .~ "~ Chert overyling carbonate. 1A ~ is base of deepening upward sequence. ~
42 43 44 45
4 3 2 IB
465 !0 <5 15
S S S
IA 51B 51A 49
42 40 50 70
S S S S
o,)
~ u.< ~
~ ~
s
r,~
Green and red siliceous argillite, commonly ,9,~ interbedded with spiculitic ~ chert containing Eastern Belt
fa-i
~
!
~ ~
'~ ~ ~
m
r~ tg ~, v
~ >
~ ~
X X X
X X X
X X X X X X
X
R
R X R R
X
RXXx
X X X
X
w ~
46 47 48 49
NORTH. CALIFORNIA, E. KLAMATH TERRANE
1~ "~"
50 51 52
GV854 Dekkasl Pitl
78 S 100 L 100 No list
X X incomplete list, no rhax ( Murchey and Jones, 1994)
Do~,a~n~i,:
~
s3
~t2
3
~Mu~o~n~Jono~. 1994
Tuffaceous chert. NORTH OREGON, GRINDSTONE NORTH. CALIFORNIA, NORTH FORK/HAYFORK TERRANES
:a~
54 Pit3 4 No list ( Murchey and Jones, 1994) 55-63 5-98%, median 25%( Murchey and Jones, 1994) Diverse spicule fauna: choristid and hexactinellids, no rhax. 64 3800 40 S 65 3801 6 S R 66 3802 30 S 67 3803 30 S R 68 3804 16 S 69 3806 12 S 70 3807 6 S 71 3808 2 S 72 3810 2 S 73 3811 14 S 74 3812 10 S 75 3814 4 S 76 3815 10 S 77 3820 40 S 78 1960 6 S 79-89 0--2 90-92 0--4 ( Murchey and Jones, 1994) 93-98 <1-9%, median 2% (Murchey and Jones, 1994)
Red, green chert and siliceous mudstone
.~
X X R
CALIFORNIA, NORTHERN SIERRA TERRANE Reeve Formation: tuffaceous red argiilite
..a
89
X
~ "~" "~ L~ "~ ~ "~ ~0 ~ ,9.o ~
o
OREGON, BAKER TERRANE N.W. WASHINGTON HOZAMEEN, SAN JUAN T. N. ALASKA, Angayucham
99
3275
6
~olist
S
X X X
R R
X
R R
R X
X X X X
R X R X R
R
R X R X X
X
122
B.L. Murchey
current-oriented spicules composed of large broken monaxons, a few whole strongyle monaxons, and desmas (composite Sample 1). A third sample from another locality in the Rex Chert (Sample 3) contains large monaxons (abundant), desmas, strongyles, calthrops, and at least 1% hexactines. Most spicules within the Rex Chert were probably subjected to current action before their final deposition. In DeLaubenfels' interpretation (oral commun., 1957 in Cressman and Swanson, 1964), the spicules of the Rex Chert were derived from sponges living at depths less than 50 m. DeLaubenfels observed from thin sections that the axial canals of the large monaxon spicules in the Rex Chert Member are differentially enlarged by dissolution and are filled with a variety of materials. Hence, the spicules may have been partially dissolved, filled, and then redeposited. The spicules in the study samples are size sorted and oriented. At some localities the Rex Chert encompasses lenses of shelly fossiliferous limestone. The limestone lenses may represent small carbonate banks (Yochelson, 1968), indicative of a moderately shallow-water origin for the surrounding spiculitic chert, or they may represent fossil-filled channels (Yochelson, 1968). The new data in this report document more spicule morphotypes in the Rex Chert than previously reported, but are otherwise generally consistent with the observations of previous workers.
Black chert, northeastern Nevada, southwestern margin of Phosphoria basin and inferred Antler high (Table 5-11,"Samples 4-7)
Four samples of black chert were collected from the Mount Ichabod area (Ketner et al., 1996). Sample 4 is from bedded chert depositionally overlain by Sample 5 chert conglomerate. The larger clasts in the conglomerate are intrabasin clasts of spiculite similar to the underlying bedded chert. Detrital white mica occurs in both the bedded chert and conglomerate; rounded quartz grains, phosphatic chert clasts, and spiculitic chert clasts also occur in the conglomerate. Residue from both samples contain abundant sponge spicules and a few rare radiolarians (Table 5-II). Both samples contain abundant, large monaxons spicules filled with dark material that may be collophane. Sample 4 contains rhax spicules and both samples contain desmas, strongyles, and some oxyhexactines. Sample 6 came from massive black chert near the base of a Mesozoic or younger thrust fault (Ketner et al., 1996) and sample 7 from another black bedded-chert unit. Both contain large monaxons, rhaxes, and strongyles. Based on the similarity of spicule faunas, presence of phosphatic material, and lithology, these spiculites represent part of the spiculite-phosphate belt that includes the Phosphoria occurrences.
Edna Mountain Formation of Nevada, overlap sequence deposited on the Antler high (Table 5-11; Samples 8-12)
This study documents the sponge spicule assemblages found in the upper part of the Edna Mountain. Recent investigations in the Battle Mountain Mining District and at Lone
Regional analysis of spiculite faunas in the Permian Phosphoria basin
123
Tree indicate that the Edna Mountain Formation is much more widespread than previously thought (Murchey et al., 1995). It crops out at Treaty Hill north of the Humboldt River and at Lone Tree Hill (Lone Tree and Stonehouse gold deposits) and the Eight South deposit south of the Humboldt River. In the Battle Mountain Mining District, rocks previously assigned to the Middle Pennsylvanian Battle Formation at the Old Marigold Mine have been reassigned to the Edna Mountain Formation, and the unit has been recognized farther south at the Fortitude gold deposit. The Edna Mountain crops out for about 1.6 km along Trout Creek in the north-central part of the North Peak quadrangle south of Battle Mountain. In this area, 200-250 m of west-dipping brown platy siltstone and fine-grained sandstone are discontinuously exposed. Based on oscillation ripple marks, the depositional environment was very shallow water. Petrographic and electron microscopic examination of the brown siltstone reveals that these rocks contain abundant sponge spicules, many of which are filled by brown collophane. The rock matrix is quite siliceous and, locally, is strongly silicified. In the general area of the Eight South deposit, rocks of the Edna Mountain Formation contain as much as 0.9 wt.% P (Murchey et al., 1995). Sponge-spicule faunas were obtained from five localities at approximately the same stratigraphic position along the strike-length of the formation. The spicule faunas are dominated by large monaxons, commonly broken. Tiny, kidney-bean shaped rhax spicules are a small but very distinctive component of these assemblages. Desmas, strongyles, and oxyhexactines are also present. The spicule faunas are indistinguishable from the previously described faunas in the Wordian Rex Chert Member of the Phosphoria Formation.
Spiculitic black chert, Havallah assemblage, Nevada, eastern basin margin facies (Table 5-11; Samples 13-24)
Well-bedded (4-80 cm) black or gray chert in the Havallah lithotectonic assemblage of the Golconda terrane contains spiculite faunas very similar to those described above (Murchey, 1990). The chert contains abundant sponge spicules (grain-supported in some samples) and quartz silt, but no radiolarians. The sponge spicules are well-sorted and commonly current-oriented. In decreasing order of relative abundance, the spicule morphotypes include large monaxons, rhaxes, strongyles, desmas, anatriaenes, oxyhexactines, anadiaenes, and birotules. Most of the samples came from north central Nevada in the Antler Peak area (Samples 13-22). However, two samples (23 and 24) with similar faunas were collected farther south in the Shoshone Range (Moore and Murchey, 1998). In the Antler Peak area, thinner black chert beds have rather flat bedding surfaces with sole marks and deep-water trace fossils (Roberts, 1964; Murchey, 1990). They exhibit partial or complete Bouma Ta__e sequences (Bouma, 1962). Beds thicker than 30 cm are gray, vitreous, and lack apparent sedimentary structures. Bedding typically thickens and grains coarsen upward. Packages of these chert beds alternate with packages of Artinskian to Kungurian and Wordian argillite and siliceous argillite but they are not independently dated. Because the bases of the argillite units in the Antler Peak area are commonly thrust
124
B.L. Murchey
faults in the Antler Peak area (Murchey, 1990), the spiculites may be Wordian, based on their similarity to assemblages in the Edna Mountain Formation and Phosphoria basin. However, the oldest part of the Phosphoria is Kungurian, and Kungurian rocks (upper Leonardian) have not been reported to occur in the Antler high, probably because they were removed during the formation of an unconformity at the base of the Edna Mountain Formation. I had concluded originally that the spiculite faunas of the Havallah basin first accumulated in an environment similar to or equivalent to the depositional setting of the Rex Chert and subsequently were redeposited downslope into the Havallah basin (Murchey, 1990). The subsequent discovery of the Edna Mountain assemblages supports that general interpretation but points to the Antler high as the closer source area.
Central Belt: mixed choristid demosponge-hexactinellid sponge assemblages associated with (ruzhencevispongacid) radiolarians Thirty-four samples from upper Artinskian and Kungurian and (or) Guadalupian red and green argillite, mudstone, or chert were collected from the Havallah assemblage of the Golconda terrane (Table 5-II; Samples 30-40) and from the previously mentioned Permian arc terranes in western Nevada, eastern California, and northern Oregon (Table 5-11; Samples 41-63). The terrane names for the Permian arc sequences follow the usage of Silberling et al. (1987) (Fig. 5-1). The arc terranes and their Permian radiolarian assemblages are described in more detail in Murchey and Jones (1994). The Artinskian to Kungurian Havallah basin samples came from thin beds (1-2 cm) of siliceous argillite interbedded with red, green, and purple argillite in Lithotectonic Unit 1 of Murchey (1990) near Antler Peak. Neither sponge spicules nor radiolarians are sorted. Therefore, the sponge spicules may not have traveled far from their living positions, while the radiolarians represent pelagic deposition. The samples from the Black Rock terrane of northwestern Nevada came from a 10 m gray and black chert unit with Middle Permian radiolarians that conformably overlies ferruginous dolostone containing Middle Permian brachiopods (Ketner and Wardlaw, 1981; Jones, 1990). The chert unit contains thick (150 cm or more) resistant beds as well as thinner, argillaceous beds that may be tuffaceous. Microfossils include radiolarians, sponge spicules, and silicified, current-oriented brachiopod(?) spines (Table 5-1; Samples 41-46) (Murchey and Jones, 1994). Artinskian to Kungurian samples from the Northern Sierra terrane were collected from the Reeve Formation, which is primarily a volcanic and volcaniclastic unit that includes lenses of red siliceous argillite with microfossil faunas (Harwood and Murchey, 1990). The samples from the Eastern Klamath terrane came from a chert sample in the Dekkas Formation and chert in the basal part of the Pit(?) Formation at outcrops on Nosoni Creek and near Lake Shasta (Murchey and Jones, 1994). Samples from the Grindstone terrane came from bedded chert or argillite (Murchey and Jones, 1994). Three generalizations describe the majority of assemblages in these samples. First, most samples contain both abundant radiolarians and sponge spicules. Second, the sponge-spicule
Regional analysis of spiculite faunas in the Permian Phosphoria basin
125
faunas are diverse and characterized by small and delicate representatives of both hexactinellids (hexactines, pentactines, birotules, anadiaenes) and demosponges (probable choristids: protriaenes, anatriaenes, calthrops). They lack the large monaxon spicules and abundant rhax and spheroidal microscleres of the Eastern Belt. Third, as previously documented by Murchey and Jones (1994), stauraxon (ruzhencevispongacid) radiolarians are more abundant than albaillellacid radiolarians. Lithistid demosponge desmas are present as a minor component of many samples, but tricranoclone desmas appear to be restricted to the shallowest areas, that is, Sample 51 from the volcanic Dekkas Formation. On the other hand, the demosponges, and even hexactinellids, were suppressed over time in the Black Rock terrane, which may have deepened upward rapidly or become less-well oxygenated. The Eastern and Central belt assemblages interfingered along the eastern margin of the Havallah basin. Both are documented above and below the Golconda thrust fault. As previously discussed, eastern belt spiculites were redeposited downslope into the Havallah basin where they were bracketed by argillaceous sediments with Central Belt assemblages. Preliminary studies of the faunas in previously mentioned argillaceous rocks in the Antler overlap sequence in the Shoshone Range (Moore and Murchey, 1998) indicate that they will be classified as Central Belt assemblages.
Western Belt." radiolarian-dominated assemblages with or without a minor component o f hexactinellid sponge spicules Thirty-five samples (Table 5-II; Samples 64-98) of red and green chert and siliceous mudstone from accretionary complexes in California, Oregon, and Washington were examined for microfossils. Murchey and Jones (1994) documented the ratio of radiolarians to sponge spicules in samples from the Baker terrane of Oregon, and the Hozameen and San Juan terranes of Washington, but did not record other spicule data, primarily because spicules were rare (Table 5-1I) and unremarkable. Samples from the North Fork and Hayfork terranes in the Klamath Mountains contain more spicules, which were reexamined for this study. The clear pattern for all of these terranes is that spicules are subordinate to radiolarians. If spicules are present at all, they rarely include morphotypes other than delicate monaxons and hexactines. Murchey and Jones (1994) documented that Permian albaillellacid radiolarians were more abundant than stauraxon forms in the oceanic terranes.
Northern basins Northwind Ridge, Chukchi Sea Cores from Northwind Ridge, a continental fragment in the borderland of the Arctic Ocean Amerasia basin, recovered a suite of Paleozoic and Triassic sedimentary clasts from
126
B.L. Murchey
a Pliocene and Miocene breccia. Grantz et al. (1998) correlated the clasts to the Sverdrup Basin based on lithologies, ages, and fossil assemblages. They concluded that the Northwind Ridge was attached to Arctic Canada and Arctic Alaska prior to the rifting that created the Amerasia basin. Cherty spiculites from two cores contain a diverse demosponge assemblage remarkably similar to the Eastern Belt assemblages of the Phosphoria and Edna Mountain but with fewer rhaxes, no strongyles, and no hexactines (Table 5-II; Samples 25-29, and Murchey in Grantz et al., 1998). Grantz et al. (1998), based on regional correlations and reconstructions, correlated the samples to spiculites in the Permian Van Hauen Formation of the Sverdrup Basin, which formed within the siliceous belt of the northern seas that was delineated by Beauchamp and Baud (2002) (Fig. 5-2).
Angayucham terrane and Northern Brooks Range
For no obvious reason, very few Permian radiolarites and no spiculites have been recognized in the terranes of the Angahucham Mountains and Brooks Range in northern Alaska, although hundreds of chert and argillite samples have been processed for microfossils. The Permian rocks are mostly confined to the lower Lower Permian (Murchey and Jones, 1992). Permian radiolarian diversity may have been low in the northern seas, or the conditions for preservation unsuitable. One sample from the Angayucham terrane contains probable Kungurian radiolarians and 6% sponge spicules comprised only of monaxons and hexactines. The Angayucham terrane is associated with pillow basalts, and not surprisingly, therefore, the sponge association is comparable to those in the Western Belt accretionary complexes of the conterminous States. More margin-proximal strata of the northern Brooks Range have also not yielded identifiable late Early or Middle Permian siliceous faunas. Older Carboniferous strata have very impoverished spicule populations (small monaxons and hexactines, local protriaenes) (Murchey, 1989) in comparison to the Chukchi Sea clasts.
DISCUSSION AND CONCLUSIONS This study has delineated an Eastern Belt of distinctive spiculite deposits that rimmed the autochthonous margin of the western conterminous States. The rhax-bearing, demosponge-dominated spiculite assemblages formed only along the margin edge, in the relatively shallow waters of the Phosphoria embayment and in shallow-marine subbasins on the trend of the Antler high, the same regions where phosphogenesis occurred during the Permian. Approximately 90% of all modern marine sponge species are demosponges and they occupy subtidal to abyssal environments (Rigby, 1983). They are the most common reef sponges (Riitzler and Macintyre, 1978). In the barrier reef near Carrie Bow Cay, Belize, the highest density and diversity of living sponges (mostly demosponges) is in the "deeper part of the inner fore-reef (spur and groove zones, 4-12 m) and on the outer fore-reef
Regional analysis of spiculitefaunas in the Permian Phosphoria basin
127
(12-60 m)" (Riitzler and Macintyre, 1978). The Permian Eastern Belt assemblages contain a variety of demosponge spicule types, but the belt is notable for the wide distribution of rhaxes and lithistid desmas. The ubiquitous rhax spicules in the Eastern Belt are probably corroded demosponge selenaster microscleres equivalent to those in the modern tropical genus Placospongia (Rfitzler and Macintyre, 1978). Placospongia inhabits the spur and groove environments on the reef front (Riitzler and Macintyre, 1978). In the Jurassic Tethys Sea, rhax-bearing sponges inhabited coral-reef, reef-talus, and sporadically, backreef environments (Wiedenmayer, 1980) and they are important components in the spicule assemblages of the Ammonitico Rosso ad Aptici of the Trento Plateau in Italy (Murchey and E Baumgartner, unpublished data). Lithistid desmas are very common in the Eastern Belt, less so in the Central Belt, and virtually absent in the Western Belt. Modern lithistids have few representatives; they are most common in warm, relatively shallow water, although their bathymetric range is subtidal to at least 1500 m (Rigby, 1983, p. 20). During the late Paleozoic, lithistid sponges lived in shelly environments as well as in basin muds (Finks, 1960, 1970). In the Pennsylvanian and Permian Delaware Basin of Texas, whole lithistid sponges are found in patch-reefs and in basin sediments that were deposited possibly as deep as 550 m (Finks, 1970). An inferred bathymetric range of 150 m or less for the Eastern Belt assemblages is determined from the model of Murchey (1989, 1990) (Fig. 5-4). In the model, the ratio of sponge spicules to radiolarians correlates inversely with increasing depth of living position. Extremely high spicule abundance (99-100%), such as that found in the spiculites of the Eastern Belt, also correlates with the presence of rhax spicules (only in the Eastern Belt), lithistid desmas, strongyles, and large monaxons. Where they have been well dated in Idaho and Nevada, the rhax-bearing demosponge spiculites are Middle Permian, Guadalupian. Where they occur as redeposited sediments
.Spicule i /
~
~
l.~,
e~
,
~
~ o
~~ ~
#
~~ y
~
~
life position :i
I
I
I
Eastern Belt
I
I
I
Estimated
. ~ . ~ / /
..-.
- - 10 0 %
Q)
nn
I
--
50m
--
150m
_ 20%
--
1000 m
-- 5%
-- 2000 m
E (1)
I
m
--
100%
I
-- 50%
I
I
I
I
i
Western Belt
Fig. 5-4. Best fits of the spicule populations in the Eastern, Central, and Western Belts (Fig. 5-1; Table 5-II) on the paleobathymetric model of Murchey (1990).
128
B.L. Murchey
in the eastern part of the Havallah basin, they are probably no younger than early Middle Permian, Wordian, the age of the youngest associated radiolarians, but they may be as old as late Artinskian(?) or Kungurian. The very preliminary examinations of sponge spiculites from the Chukchi Sea point to a similar fauna along strike far to the north, beyond the limited region of major phosphorite deposition, but in the same general physical setting along the Pangean margin. As previously stated, these spiculites were dominated by large monaxons and a diverse suite of other demosponge spicules similar to the assemblages of the Eastern Belt in Idaho and Nevada. These few samples had fewer rhaxes, no strongyles, and no hexactines. A more thorough comparison of the spiculite faunas of the Canadian Arctic and Barents Sea might reveal some consistent, measurable differences between these faunas and those of Utah and Nevada. Nevertheless, these early comparisons support the conclusion that the Eastern and Arctic spiculite belts formed as part of the same major oceanographic system. The comparisons support previous conclusions (Wardlaw and Collinson, 1986; Beauchamp, 1994) that calcareous marine faunas interfingering with the spiculites along the Pangean margin also form a distinct faunal belt from the Arctic region to the Phosphoria region. In the Phosphoria basin, the two major intervals of spiculite accumulation, represented by the Wordian Rex Chert and Capitanian Tosi Chert Members, each followed deposition of phosphatic shale (Sheldon, 1957; Wardlaw and Collinson, 1986). Sedimentary cycles in the Phosphoria have been attributed to changes in eustatic sea levels (see Chapters 1, 2, and 4). The phosphatic shales have been interpreted as representative of lowstand to transgressive conditions (Hiatt, 1997), though western deposits of the Meade Peak Member have also been interpreted as corresponding to maximum flooding and highstand conditions (Perkins and Piper, Chapter 4). The spiculites of the Phosphoria have been interpreted as coinciding with maximum highstand systems tracts (Hiatt, 1997). The flat, ramp-like eastern and southern margins of the basin (Wardlaw and Collinson, 1986; Hiatt, 1997; Hein et al., Chapter 2) permitted thick deposits of spiculites and phosphorite to accumulate in the Idaho depocenter. Along the Antler high, black spiculitic chert appears to be confined to the northern part of the trend, near the southern margin of the Phosphoria basin. Farther south, abundant Eastern Belt assemblages are preserved in the siltstone of the Wordian Edna Mountain Formation, but more pure spiculite accumulations were winnowed and redeposited into the deeper Havallah basin. Along the Antler high, eustatic sea-level changes were probably overprinted by tectonic events in the Middle Permian. The Edna Mountain Formation was deposited unconformably on rocks of the Roberts Mountain allochthon and older overlap sequence, concurrently with normal faulting at some localities (Murchey et al., 1995). Beauchamp and Baud (2002) tied the onset of spiculite deposition in the Sverdrup Basin to a maximum flood event at the beginning of the Artinskian, and the maximum extent of spiculite deposition to the Middle Permian Wordian and early Late Permian. They interpreted the spiculites of the Trappers Cove and Van Hauen Formations as highstand deposits. During this same interval, siliceous spicule-rich strata accumulated in adjacent basins as far east as Svalbard in the Barents Sea (Fig. 5-2). Ehrenberg et al. (2001) proposed a sequence stratigraphic model for Permian strata of Spitsbergen and adjacent localities in
Regional analysis of spiculite faunas in the Permian Phosphoria basin
129
the Norwegian Barents Sea wherein lowstands of sea level produced siliciclastic-limestone intervals and highstands produced spiculite deposits in a broad, gently sloping intracratonic basin. The paleobathymetric model of Beauchamp (1994) suggests that spiculite accumulations were related to the migration of siliceous sponges into shallower waters, which, when coupled with the hypothesis of dramatic cooling of seawater, may have been significant in creating a hospitable shallower-water environment for sponges and an inhospitable environment for calcareous organisms. A highstand on an epicontinental margin has also been proposed for the Eocene spiculites and spongolites (whole sponges) of southern Australia (Gammon et al., 2000). The Eocene sponge belt of Australia is quite interesting because it was almost as large as the Permian spiculite belt along northwestern Pangea. Gammon et al. (2000) interpreted the appearance of a lithistid demosponge assemblage on the drowned shelf as a fairly straightforward migration of deeper-water Tethyan-type forms into a newly compatible environment. They emphasized that the environment was probably not cold. A simple model of sponge migration from deeper to shallower water during highstands does not adequately explain the Eastern Belt assemblages of the Phosphoria-Antler region, nor seemingly of the Chukchi Sea. The sponges that produced the large monaxons, the rhax spicules, and the tricranoclone desmas did not previously occupy deeper niches in the Havallah basin, in the arc and oceanic basins farther west, or in the Alaskan basins of the Brooks Range and Angayucham Mountains. Examination of hundreds of samples from these regions has not revealed these types of spicules in older siliceous rocks of the Late Paleozoic Cordilleran basins (Murchey, 1989). In addition, the most common sponge spicules in upper Lower Permian assemblages in the deeper Havallah basin immediately to the west are very minor components of the Middle Permian Eastern Belt assemblages (Table 5-II). It is possible that the rhax spicules, the large monaxons, and tricranoclone desmas of the Eastern Belt represent exotic immigrant species and (or) new species. It is also possible that parent sponges already existing on the shelf in small numbers during carbonate phases greatly expanded their populations when carbonate production along the margin was suppressed. Siliceous sponges were very common throughout the basins and submerged arc highs of the Central Belt, and locally present though not common in the open ocean basins of the Western Belt. The delicate choristid demosponge and hexactinellid assemblages of the Central Belt represent an especially diverse suite of late Early and early Middle Permian spicules in comparison to older Permian and Carboniferous sponge assemblages in the same geographic region (Murchey, 1989, 1990). Protriaene, anatriaene, and calthrop spicules in the Central Belt probably came from choristid sponges. Similar spicules also occur in basinal limestone in the Permian Delaware Basin of western Texas (Finks, 1960). The choristids are the principal group of modern demosponges with four-rayed (tetraxon-tetractine) megascleres. They are common in reef environments (Riitzler and Macintyre, 1978) but have been reported to occur at depths of over 2000 m (Rigby, 1983). Hexactinellid sponges produced the hexactine, pentactine, birotule, and anadiaene spicules in the Central Belt assemblages. These morphotypes have been documented in whole or partial hexactinellid sponge skeletons in basinal limestone of the Delaware Basin
130
B.L. Murchey
(Finks, 1960, 1983). Hexactinellid diversity is greatest in the Texas basin facies but two specialized genera occur in shallow settings around the basin rim (Finks, 1960, 1983). Almost all modern hexactinellid sponges live at depths greater than 90 m except in the Antarctic region where siliceous sponges occur in anomalously shallow water (Dayton et al., 1974). Most hexactinellids inhabit environments between 200 and 2000 m deep, although the modem producers of birotule spicules, the deep-dwelling Amphidiscophora, inhabit depths to at least 5900 m (Reid, 1968 after Schulze, 1899; Ijima, 1927; Hartman, 1980; Finks, 1983). One can draw three conclusions from these observations of sponge distribution patterns. First, the bottom waters of the Central Belt basins were very hospitable to siliceous sponges. They were sufficiently well oxygenated to support widespread if scattered sponge populations, and salinity must have been normal. Second, the Central Belt assemblages were generally fairly deep-water deposits, between 150 and 1000 m, except on topographic highs along the western arcs (Fig. 5-4). Third, the sponge spicule assemblages qualitatively resemble spicule assemblages reported from basinal facies of the Delaware Basin in Texas. In the Delaware Basin, however, siliceous spicule accumulations were overwhelmed by calcareous debris. Therefore, suppression of carbonate production along northwestern Pangea was probably one of the more important factors leading to the production of significant deposits of siliceous spiculites or mixed spiculite-radiolarite deposits. Radiolarians are key components of the Central Belt assemblages and the principal component of the Western Belt assemblages. During Permian time, opal-secreting polycistine radiolarians were the "default" fossil-forming microplankton. Planktonic foraminifers, diatoms, silicoflagellates, and calcareous nannoplankton did not appear in the fossil record until the Mesozoic. Thus, suppression of carbonate production was not a critical factor in their preservation. Low rates of clastic sedimentation and at least moderate rates of productivity in the water column were the major factors influencing their accumulation in Permian basins. Three major radiolarian groups occupied the water column in the Central and Western Belts. Spheroidal forms were present in both regions and may have lived in more than one environment. The fiat, multi-spoked stauraxons (Ruzhencevispongacea) were more common in the Central Belt (Murchey and Jones, 1994) (Fig. 5-3). Similar modem morphotypes contain algal symbionts and, therefore, inhabit the photic zone. The heavy, elongated Albaillellacea were more common in the Western Belt (Murchey and Jones, 1994) (Fig. 5-3). They may have preferred deeper water than the stauraxons because their absence in spicule-dominated samples is clearly depth related. The widespread co-occurrence of radiolarian groups that probably inhabited different parts of the water column is additional evidence for an offshelf water column that was mostly oxygenated. The typical green and red color of the chert and argillite in the Central and Western Belt basins also supports this idea. These observations support models that call for vigorous thermohaline circulation in the oceans (Murchey and Jones, 1992; Beauchamp and Baud, 2002). Widspread ocean anoxia and breakdown of thermohaline circulation did not occur until the later part of the Lopingian (Late Permian) (Beauchamp and Baud, 2002; Kato et al., 2002). In summary, the Eastern Belt of rhax-bearing demosponge-dominated spiculites of the Permian Phosphoria basin and Antler high were the southernmost expression of one of the
Regional analysis of spiculite faunas in the Permian Phosphoria basin
131
largest spiculite belts that ever existed. The thickest deposits formed in epicratonic basins, probably during highstands. In Nevada, the Eastern Belt deposits interfingered with argillaceous radiolarite-spiculites containing spicules from the choristid-hexactinellid sponge populations that lived in the deeper, radiolarian-rich basins of the Central Belt. The Idaho batholith obscures any remnant of the Central Belt along the edge of the Phosphoria basin. The Central Belt sponges did not form massive bedded-chert sequences, but they were nevertheless important components of argillaceous hemipelagic deposits. In the pelagic radiolarite deposits of the Western Belt, scattered populations of deep-water hexactinellid sponges were minor contributors to the sediment budget. A comparison of the temporal and spatial distribution patterns of the late Early Permian and Middle Permian sponge spicule populations, reported herein, with those in older Carboniferous and early Early Permian populations of the Cordillera (Murchey, 1989) did not reveal any significant migration of the major deeper-marine sponge groups from the Havallah basin into the shallower coastal realm. The rhax-bearing demosponge populations in the Eastern Belt may have resulted primarily from the expansion, immigration, and (or) evolutionary diversification of shallower-dwelling demosponges. The deeper-dwelling hexactinellids and probable choristid demosponges formed only a minor component of the Eastern Belt assemblages. However, they formed the major component of the Central Belt assemblages, as they had done in earlier times. But even in the geographic region outlined by the Central Belt, the relative abundance and apparent diversity of late Early and early Middle Permian sponge spicules were significantly greater than those of the Carboniferous or early Early Permian, suggesting that the same could be said for the living sponge populations in the bathyal realm. Suppression of carbonate-producing organisms along the margin was critical to the accumulation and preservation of both the demosponge spiculites and the spicule-rich argillites of the Eastern and Central Belts. Beauchamp and Baud (2002) described a Middle and Late Permian world in which warm-water carbonate producers disappeared from the northwestern margin of Pangea while siliceous sponges increasingly dominated the coastal basins. In deeper basins, deposition of radiolarites was so widespread that Murchey and Jones (1992), in their description of the phenomenon, referred to it as the Permian Chert Event. Beauchamp and Baud (2002) proposed a model in which the root cause of the faunal turnover and widespread deposition of siliceous marine sediments was vigorous thermohaline circulation driven by cold, nutrient- and oxygen-rich northern waters. In their model, a major current carried these waters southward along the northwest Pangean margin. Local upwelling, such as has been proposed for the Phosphoria basin (McKelvey et al., 1953b; Sheldon, 1963; Piper and Link, 2002; see Hein et al., Chapter 2), was a component of the larger system. The model incorporates the observations and conclusions of Wardlaw and Collinson (1984), who noted the similarities between the assemblages of conodonts and brachiopods in the Phosphoria basin and those in the Arctic. The idea that cold oceanic waters suppressed the warm-water carbonate producers in the Phosphoria basin was suggested as early as 1917 by Pardee. He also proposed that decaying organic matter associated with high productivity produced CO2, which suppressed the formation of carbonates but not phosphates (see Hein et al., Chapter 2). Although there is
132
B.L. Murchey
evidence for warm water in the shallow reaches of the greater Phosphoria basin (Hiatt, 1997), the larger regional patterns of deposition along the northwest margin of the Pangea support the integrated biologic and oceanographic model of Beauchamp and Baud (2002).
ACKNOWLEDGMENTS Many colleagues collected samples that were included in this study: Keith Ketner, northeastern Nevada; Ted Theodore, Edna Mountain Formation; Thomas Moore, Shoshone Range; Arthur Grantz, Chukchi Sea; David Harwood, Northern Sierra terrane; Louis Fraticelli, Eastern Klamath terrane; Charles Blome, Grindstone and Baker terranes; Porter Irwin and Nick Mortimer, North Fork and Hayfork terranes; John Whetten, San Juan terrane; Lynn Tennyson, Hozameen terrane; and David L. Jones, Angayucham terrane. David L. Jones was extremely generous in sharing his vast knowledge of the regional geology of the Cordillera. James Hein and John Barron provided valuable comments on the first version of this manuscript.
REFERENCES Beauchamp, B., 1994. Permian climatic cooling in the Canadian Arctic. In: G.D. Klein (ed.), Pangea: Paleoclimate, Tectonics, and Sedimentation During Accretion, Zenith, and Breakup of a Supercontinent. Geol. Soc. Am. Spec. Paper 288, pp. 229-246. Bcauchamp, B. and Baud, A., 2002. Growth and demise of Permian biogenic chert along northwest Pangea: evidence for end-Permian collapse of thermohaline circulation. Palaeogeogr. Palaeoclimatol. Palaeoecol., 184: 37-63. Bouma, A.H., 1962. Sedimentology of Some Flysch Deposits: a Graphic Approach to Facies Interpretation. Elsevier, Amsterdam, 168 pp. Cressman, E.R. and Swanson, R.W., 1964. Stratigraphy and petrology of the Permian rocks of southwestern Montana. US Geol. Surv., Prof. Paper, 313C, pp. 276-568. Dayton, P.K., Robilliard, G.A., Paine, R.T. and Dayton, L.B., 1974. Biological accommodation in the benthic community at McMurdo Sound, Antarctica. Ecol. Monogr., 44: 105-128. Dumitrica, P., 1970. Cryptocephalic and cryptothoracic Nassellaria in some Mesozoic deposits of Romania. Revue Roumaine de Geolologie, Geophysique, et Geologie, Serie de Geologie, 14: 45-124. Ehrenberg, S.N., Pickard, N.A.H., Henriksen, L.B., Svana, T.A., Gutteridge, P. and Macdonald, D., 2001. A depositional and sequence stratigraphic model for cold-water, spiculitic strata based on the Kapp Starostin Formation (Permian) of Spitsbergen and equivalent deposits from the Barents Sea. AAPG Bull. 85(12): 2061-2088. Finks, R.M., 1960. Late Paleozoic sponge faunas of the Texas region: the siliceous sponges. Bull. Am. Mus. Nat. Hist., 120: 1-160. Fink, R.M., 1970. The evolution and ecologic history of sponges during Paleozoic times. Symp. Zool. Soc. Lond., 25: 3-22.
Regional analysis of spiculite faunas in the Permian Phosphoria basin
133
Finks, R.M., 1983. Fossil Hexactinellida. In: T.W. Broadhead (ed.), Sponges and Spongiomorphs (notes for a short course). University of Tennessee, Department of Geological Sciences Studies in Geology, vol. 7, pp. 101-115. Gammon, P.R., James, N.P. and Pisera, A., 2000. Eocene spiculites and spongolites in southwestern Australia; not deep, not polar, but shallow and warm. Geology, 28(9): 855-858. Grantz, A., Clark, D.L., Phillips, R.L., Srivastava, S.P. with additional biostratigraphic contributions or radiometric ages by 14 others, 1998. Phanerozoic stratigraphy of Northwind Ridge, magnetic anomalies in the Canada basin, and the geometry and timing of rifting in the Amerasia basin, Arctic Ocean. Geol. Soc. Am. Bull., 110:801-820. Hartman, W.D., 1980. Systematics of the Porifera. In: W.D. Hartman, J.W. Wendt and E Wiedenmayer (eds.), Living and Fossil Sponges (notes for a short course). Sedimenta, VIII: 24-51. Harwood, D.S. and Murchey, B.L., 1990. Biostratigraphic, tectonic, and paleogeographic ties between upper Paleozoic volcanic basinal rocks in the northern Sierra terrane, Calivornia and the Havallah sequence, Nevada. In: D.S. Harwood and M. Miller (ed.), Paleozoic and Early Mesozoic Paleogeographic Relations; Sierra Nevada, Klamath Mountains, and Related Terranes. Geol. Soc. Am. Spec. Pub., vol. 255, pp. 157-217. Hiatt, E.E., 1997. A paleoceanographic model for oceanic upwelling in a late paleozoic epicontinental sea: a chemostratigraphic analysis of the Permian Phosphoria Formation. PhD Dissertation, Univ. of Colorado. ljima, I., 1927. The Hexactinellida of the Siboga Expedition, v. 6. Leiden, Siboga-Expeditie, 383 pp. Jones, E.A., 1990. Geology and tectonic significance of terranes near Quinn River Crossing, Nevada. In: D.S. Harwood and M. Miller (eds.), Paleozoic and Early Mesozoic Paleogeographic Relations; Sierra Nevada, Klamath Mountains, and Related Terranes. Geol. Soc. Am. Spec. Pub., vol. 255, pp. 239-254. Kato, Y., Nakao, K. and lsozaki, Y., 2002. Geochemistry of Late Permian to Early Triassic pelagic cherts from southwest Japan: implications for an oceanic redox change. Chem. Geol., 182:15-34. Ketner, K.B., 1998. The nature and timing of tectonism in the western facies terrane of Nevada and California; an outline of evidence and interpretations derived from geologic maps of key areas. US Geol. Surv., Prof. Paper, Report: P 1592, 19 pp. Ketner, K.B. and Wardlaw, B.R., 1981. Permian and Triassic rocks near Quinn River Crossing, Humboldt County Nevada. Geology, 9:123-126. Ketner, K.B., Murchey, B.L., Stamm, R.G. and Wardlaw, B.R., 1996. Geologic map of the Mount Ichabod area, Elko County, Nevada. US Geological Survey, Miscellaneous Investigations Series Report, 1-2535, 1 sheet. McKelvey, V.E., Armstrong, EC., Gulbrandon, R.A. and Campbell, R.M., 1953a. Stratigraphic sections of the Phosphoria Formation in Idaho, 1947-48, Part 2. US Geol. Surv., Circular, 301. McKelvey, V.E., Swanson, R.W. and Sheldon, R.P., 1953b. The Permian phosphorite deposits of western United States. Comptes Rendus, 19th Int. Geol. Congr. sec. XI, pp. 45-64. Moore, T.E. and Murchey, B.L., 1998. Initial results of stratigraphic and structural framework studies in the Cedars Quadrangle, southern Shoshone Range. In: R.M. Tosdale (ed.), Contributions to the gold metallogeny of northern Nevada. US Geological Survey, Open File Report, 98-0338-B, pp. 119-140. Murchey, B.L., 1989. Late Paleozoic siliceous basins of the western Cordillera of North America (Nevada, California, Mexico, and Alaska): three studies using radiolarians and sponge spicules for biostratigraphic, paleobathymetric, and tectonic analyses. PhD Dissertation, Univ. of California, Santa Cruz, 188 pp.
134
B.L. Murchey
Murchey, B.L., 1990. Age and depositional setting of siliceous sediments in the upper Paleozoic Havallah sequence near Battle Mountain, Nevada: implications for the paleogeography and structural evolution of the western margin of North America. In: D.S. Harwood and M. Miller (eds.), Paleozoic and Early Mesozoic Paleogeographic Relations; Sierra Nevada, Klamath Mountains, and Related Terranes. Geol. Soc. Am. Spec. Pub., vol. 255, pp. 137-155. Murchey, B.L. and Jones, D.L., 1992. A mid-Permian chert event: widespread deposition ofbiogenic siliceous sediments in coastal, island arc and oceanic basins. Palaeogeogr. Palaeoclimatol. Palaeoecol., 96:161-174. Murchey, B.L. and Jones, D.L., 1994. The environmental and tectonic significance of two coeval Permian radiolarian-sponge associations in eastern Oregon. In: T.L. Vallier and H.C. Brooks (eds.), Geology of the Blue Mountains Region of Oregon, Idaho, and Washington: Stratigraphy, Physiography, and Mineral Resources of the Blue Mountains Region. US Geol. Surv., Prof. Paper, 1439, pp. 183-198. Murchey, B.L., Theodore, T.G. and McGibbon, D.H., 1995. Regional implications of newly discovered relations of the Permian Edna Mountain Formation, north-central Nevada. Geology and ore deposits of the American Cordillera; a symposium. Geological Society of Nevada, United States, pp. 57-58. Pardee, J.T., 1917. The Garrison and Philipsburg phosphate fields, Montana. US Geol. Surv., Bulletin, 640, Contributions to Economic Geology 1916, Part l, pp. 195-228. Pessagno Jr., E.A. and Newport, R.L., 1972. A technique for extracting Radiolaria from radiolarian cherts. Micropaleontology, 18:231-324. Piper, D.Z. and Link, P.K., 2002. An upwelling model for the Phosphoria Sea - a Permian, oceanmargin basin in the northwest United States. Am. Assoc. Pet. Geol. Bull., 86: 1217-1235. Reid, R.E.H., 1968. Bathymetric distributions of Calcarea and Hexactinellida in the present and the past. Geol. Mag., 105: 546-559. Rigby, J.K., 1983. Introduction to the Porifera. In: T.W. Broadhead (ed.), Sponges and Spongiomorphs (notes for a short course). University of Tennessee, Department of Geological Sciences Studies in Geology, vol. 7, pp. 12-39. Roberts, R.J., 1964. Stratigraphy and structure of the Antler Peak quadrangle, Humboldt and Lander counties, Nevada. US Geol. Surv., Prof. Paper, 459-A, pp. A 1-A93. Rfitzler, K. and Macintyre, I.G., 1978. Siliceous sponge spicules in coral reef sediments. Mar. Biol., 49: 147-159. Schulze, EE., 1899. Amerikanische Hexactinelliden nach dem Materiale der Albatross-Expedition. Gustav Fischer, Jena. Sheldon, R.P., 1957. Physical stratigraphy of the Phosphoria Formation in northwestern Wyoming. US Geol. Surv. Bull., 1042-E, 185 pp. Sheldon, R.E, 1963. Physical stratigraphy and mineral resources of Permian rocks in western Wyoming. US Geol. Surv., Professional Paper, 313-B, 273 pp. Silberling, N.J., Jones, D.L., Blake, M.C., and Howell, D.G., 1987. Lithotectonic terrane map of the western conterminous United States. US Geol. Surv. Miscellaneous Field Studies Map, MF- 1874-C, scale 1:2,500,000. Stevens, C.H., 1991. Permian paleogeography of the western United States. In: J.D. Cooper and C.H. Stevens (eds.), Paleozoic Paleogeography of the Western United States-II. Pacific Section SEPM, vol. 67, pp. 149-166. Stevens, C.H., Yancey, T.E. and Hanger, R.A., 1990. Significance of the provincial signature of Early Permian faunas of the eastern Klamath terrane. In: D.S. Harwood and M. Miller (eds.), Paleozoic and Early Mesozoic Paleogeographic Relations; Sierra Nevada, Klamath Mountains, and Related Terranes. Geol. Soc. Am. Spec. Pub., vol. 255, pp. 201-218.
Regional analysis of spiculite faunas in the Permian Phosphoria basin
13 5
Stewart, J.H., Murchey, B., Jones, D.L. and Wardlaw, B.R., 1986. Paleontologic evidence for complex interlayering of Mississippian to Permian deep-water rocks of the Golconda allochthon in Tobin Range, north-central Nevada. Geol. Soc. Am. Bull., 97:1122-1132. Wardlaw, B.R. and Collinson, J.W., 1986. Paleontology and deposition of the Phosphoria formation. In: D.W. Boyd and J.A. Lillegraven (eds)., Western Phosphate Deposits. Contrib. Geol. Univ. Wyoming, Laramie, pp. 107-142. Whiteford, W.B., 1990. Paleogeographic setting of the Schoonover sequence, Nevada, and implications for the late Paleozoic margin of western North America. In: D.S. Harwood and M. Miller (eds.), Paleozoic and Early Mesozoic Paleogeographic Relations; Sierra Nevada, Klamath Mountains, and Related Terranes. Geol. Soc. Am. Spec. Pub., vol. 255, pp. 115-136. Wiedenmayer, E, 1980. Siliceous Sponges-development through time. In: W.D. Hartman, J.W. Wendt and E Wiedenmayer (eds.), Living and Fossil Sponges (notes for a short course). Sedimenta, VIII: 55-85. Yochelson, E.L., 1968. Biostratigraphy of the Phosphoria, Park City, and Shedhorn Formations. US Geol. Surv., Professional Paper, 313-D, pp. 571-660.
This Page Intentionally Left Blank
Li[e Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistrv, 88 8 (M. Hale, Series Editor) Published by Elsevier B.V.
137
Chapter 6
STRAIN DISTRIBUTION AND STRUCTURAL EVOLUTION OF THE MEADE PLATE, SOUTHEASTERN IDAHO J.G. EVANS
ABSTRACT The Meade thrust plate is one of seven main plates in the imbricate Sevier fold-and-thrust belt of northern Utah, southeastern Idaho, and western Wyoming and contains four of the five presently active mines in the Western Phosphate Field. The structural evolution model of the Phosphoria Formation in the Meade plate includes: (a) compaction of the Phosphoria by Triassic and Jurassic overburden; (b) additional compaction in Early Cretaceous by the tectonic overburden of a warm Putnam-Paris plate and resulting greenschist-facies metamorphism of the footwall rocks that would later comprise the upper Meade plate; (c) transfer of principal strain in the Sevier fold-and-thrust belt to the Meade thrust during Early to Late Cretaceous when rocks of the Meade plate developed a B perpendicular to B' tectonite and layered deformation styles and the strata in the upper part of the thrust plate were shortened by as much as 38%; (d) possible enlargement of part of the upper fiat as the result of variable displacement involving incompetent strata and (or) basement structures; (e) additional, possibly passive piggyback, eastward transport of the Meade plate on younger thrusts; (f) burial of the westem Meade plate by gravel from a rejuvenated Putnam-Paris plate; (g) left-lateral and fight-lateral faulting within the Meade plate as a result of variable displacement on the Crawford, Absaroka, and possibly younger thrust faults; (h) Cretaceous faulting and (or) early Tertiary compression or extension that may have shuffled strata and disturbed Cretaceous metamorphic isograds; (i) periodic uplift and erosion that stripped much of the Putnam-Paris plate from the Meade plate and eroded much of the Meade plate; (j) middle to late Tertiary extension associated with development of the north-northwest-trending Bear Lake-Blackfoot Reservoir graben and bimodal basalt-rhyolite volcanism, possibly a distal effect of the Yellowstone hotspot, and possibly implicated in disturbance of Cretaceous isograds.
INTRODUCTION Following deposition of 9100-12,200 m of clastic-dominated Paleozoic and Mesozoic sedimentary rocks, a series of imbricate thrusts during the Early Cretaceous to Eocene
13 8
J G. Evans
Sevier orogeny transported the sedimentary section generally eastward (Armstrong and Oriel, 1965; Armstrong, 1968; Royse et al., 1975) along the Sevier fold-and-thrust belt. The belt includes seven major thrusts, listed from oldest to youngest: Putnam-Paris (Paris-Willard by some), Meade (Meade-Crawford or Meade-Willard by some), Crawford, Absaroka, Darby, Prospect, and Hogsback (Fig. 6-1). The Meade thrust plate contains the active phosphate mines in the southeastern Idaho part of the Western Phosphate Field (Fig. 6-2). The Permian Phosphoria Formation in the Meade plate, which
112~
111~
I
110~
II
I
i i i
Stuarea
I"...
i,,
i II
.~
9
42~ Banr Lake. ' ~
!".~ " ;~
i.~
' ',
I,'~',~q : ;,'~ .JO <
, o :~,
j~
," I ~ /
~" .,,,,,,, ,,.' 41~
""
I
:
,y ,,'" ) i~"' (,""
,P" " " "..O'-
,"
0 , ,Ar a
" Wyoming " .... -,-'- Utah
" " "
I
I
Fig. 6-1. Map showing location of the Sevier fold-and-thrust belt in southeastern Idaho, western Wyoming, and northern Utah, major thrust faults, and location of study area. Modified from Lamerson (1982).
Strain distribution and structural evolution of the Meade plate
139 111 ~
112 ~ i Bingham County
J
~o~
i
~,o
Bonne lille_County . . . . . . . . . . .
J___
43 ~
j Caribou County
~nry
EX R X XB
x
J
,o,x ..,
,~o~" ~
US 30
I -
Cor
o~,
Bancroft .~o~
Fx MCx
0
XS
~ '
1
MFx
-'X--
II
o'~~176 .' J
i i
,o~1,,~''~'~%-" ."/~'~
i
42"30'
[
/ o~O~
Ja 0 x i !CO
-o
.E_ E
:J ,_,
/~
)
",,
~
\
k G ~l / ~QGeorgetown " ">- - " ~k(,,.,~ ~ "~ ~" ,L~
Bear~ Lake Co~lnty
~o
ooo. \o~ _
i
0 I
10km I
Fig. 6-2. Index map of 7.5' quadrangles; quadrangles studied for phosphate resources during this and previous investigations are shown by larger letters; x, approximate location of historic open-pit mine; X, approximate location of active and recently active open-pit mines; B, Ballard; CH, Champ; CO, Conda; E, Enoch Valley; F, FMC Corp.; G, Georgetown Canyon; H, Henry; MC, Maybe Canyon; MF, Mountain Fuel; R, Rasmussen; S, Smoky Canyon. hosts the phosphate ore, is a deformed rock unit occupying a stratigraphic position between thick, relatively competent Paleozoic formations below it and thick, relatively competent Mesozoic strata above it. Pioneering studies of the Meade thrust plate and adjacent structures were made by Mansfield (1927, 1929, 1952) who mapped the geology of much of southeast Idaho and
140
J. G. E v a n s
adjacent southwest Wyoming at the scale of 1:62,500, including compilations at the scales of 1:250,000 and 1 : 125,000. Most of the more detailed geologic mapping in southeastern Idaho, done at a variety of scales over many years by many geologists, was compiled by Oriel and Platt (1980) into a geologic map of the Preston 1~ by 2 ~ quadrangle at the scale of 1:250,000. The structure of the Phosphoria Formation in the Meade plate was studied at macroscopic to microscopic scales to obtain a model of three-dimensional strain in the rocks in which the phosphate deposits are located and determine whether strain history and strain distribution correlate with mining characteristics, ore volume, or ore grade. A burial and thermal history of the Meade plate is included because it has a bearing on some of the physical conditions during thrusting and macroscopic structural development. The name Putnam-Paris thrust is used here for the pre-Meade and earliest thrust in the Sevier orogenic belt because the two thrusts that supply the name are apparently of similar age and may be the same structure. The Phosphoria Formation in the Meade plate, the focus of this study, includes three members: (a) Meade Peak Phosphatic Shale, which contains the phosphate resources, and the overlying (b) Rex Chert, and (c) Cherty Shale (McKelvey et al., 1956, 1959, 1967). The Rex Chert and Cherty Shale Members may interfinger in the central part of the plate and have been confused on some maps; the confusion in part reflects the generally poor exposures of the Phosphoria and the similar appearance of float composed of chert and cherty shale. In some places, the thicknesses of the Rex Chert and Cherty Shale vary so abruptly that faulting, including thrusting, and (or) folding within the upper Phosphoria is likely. The Triassic Dinwoody Formation, a secondary focus of this study, overlies the Phosphoria Formation and consists of thin-bedded limestone and interbedded silty and shaly limestone, and carbonate-cemented siltstone. In places where the contact with the Phosphoria is relatively well exposed, the Dinwoody is usually a pale-brown, brown, and greenish-brown weathered siltstone or shale and rarely sandstone. The contact between Phosphoria and Dinwoody usually corresponds to a topographic swale.
DEPTH OF BURIAL Claypool et al. (1978) and Herring (1995) noted that post-Permian strata were thinner on the continental shelf and thicker toward the west, farther from the paleoshore. The hinge across which the Mesozoic strata thicken is located just west of the present-day Idaho-Wyoming State line. Claypool et al. (1978) estimated that the depth of burial of Permian strata in eastern Idaho by the end of the Cretaceous was 5.5-8.3 km. The lower estimate is from the Gay mine on the Fort Hall Indian Reservation (western Chesterfield Reservoir quadrangle (Fig. 6-2) and west and north of the quadrangle). The greater estimate of thickness is from the Georgetown Canyon-Snowdrift Mountain area (Cressman, 1964) close to the Meade thrust. Edman and Surdam (1984b) estimated the depth of burial of the Phosphoria in the vicinity of Wooley Valley (Meade plate, Lower Valley quadrangle,
Strain distribution and structural evolution of the Meade plate
141
Fig. 6-2) at 4.15 km. An additional 9-12 km of tectonic overburden may have been added by emplacement of the Putnam-Paris plate (DeCelles et al., 1993).
THERMAL HISTORY
Petroleum generation and very low grade metamorphism Angevine and Turcotte (1983) developed a model of thermal evolution leading to oil generation in overthrust belts. They suggested that oil is generated in the rocks of a lower plate soon after emplacement of a thrust sheet. The thrust plate insulates the underlying rocks, causing their temperature to rise. Depending on the thickness of the thrust plate, oil may be destroyed through cracking in less than 5 Ma or last as many as 100 Ma in footwall rocks. This model applied to the Meade plate suggests that temperatures due to emplacement of the Putnam-Paris thrust graded from relatively low in the east to relatively high in the west, varying with the thickness of the Putnam-Paris plate. This relatively simple model is broadly consistent with evidence (see below) bearing on the thermal evolution of the Meade plate. Thermal models of thrusting also attribute a rise in temperature in a lower plate to formation of a reverse thermal gradient as the result of emplacement of a warm upper plate, in which relatively high temperatures are derived from the pre-emplacement geothermal gradient in the source region of the upper plate (Brewer, 1981). Several studies of hydrocarbons in the Meade Peak Member have presented a mixed picture of temperatures in the Meade Peak. Claypool et al. (1978) found the Meade Peak in the Meade plate to have been "overcooked" during burial to depths of as much as 8.5 km so that fluid hydrocarbons were destroyed by cracking. Subsequent studies including the Gay mine area on the Fort Hall Indian Reservation suggested: (a) that the samples of Claypool et al. were oxidized, weathered, and (or) biologically degraded, and therefore appeared to have been subjected to temperatures above the oil generation window (Desborough et al., 1988); (b) that conodont alteration indexes (CAI = 2-3) for Permian strata in the Gay mine area indicate that the rocks there were heated above the window for oil generation (Harris et al., 1980); and (c) that three samples from the Meade Peak Member (Rock-Eval) and one from the Wells Formation (CAI) below the Putnam-Paris thrust at the Gay mine had not been heated beyond the oil generation window (Desborough et al., 1988). Another study by Edman and Surdam (1984b) found a discrepancy in a single sample from Wooley Valley that showed temperature-related attributes characteristic of temperatures within and outside the oil generation window. This discrepancy was explained by a heat pulse from Tertiary plutons that influenced pyrolysis temperature, but not reflectance temperature. An estimate of depth of burial of the Wooley Valley segment of the Meade Peak Member of 4.15 km (Edman and Surdam, 1984b) suggests a temperature at the base of the Meade plate of nearly 250~ assuming a 32.8~ km -I geothermal gradient. In Early Cretaceous, the Meade plate was also covered by an estimated 12 km thick Putnam-Paris plate (DeCelles et al., 1993), which would have raised the temperature (assuming 32.8~ km -l)
142
J G. Evans
to about 400~ at the level of the Phosphoria and 640~ at the base of the Meade plate, given enough time. If the Phosphoria was exposed to 400~ it would most likely show more mineralogical indications of greenschist-facies metamorphism (range 300-550~ Turner and Verhoogen, 1960) than it does and possibly more abundant syntectonic mesoscopic structures that commonly accompany widespread greenschist-facies metamorphism (cleavage, foliation). The general lack of mineralogical and structural indications of incipient metamorphism in the Phosphoria may be indicative of a short duration of tectonic burial of rocks that would become the Meade plate, and later rapid activation of the Meade thrust, and uplift and erosion of the Meade plate.
Low-grade metamorphism Fractured quartz in Rex Chert in which biotite fills the cracks suggests that parts of the Meade plate experienced temperatures of low-grade metamorphism (_>300~ Turner and Verhoogen, 1960) at least briefly. However, occurrences of diagnostic metamorphic minerals are uncommon in the Phosphoria and absent in older formations of the Meade plate. This may be due to lack of the necessary chemical components required for neomineralization of diagnostic metamorphic minerals in chert, limestone, and dolostone, or the rocks may have lacked necessary fluids and permeability. Early diagenesis would have eliminated most of the permeability in the Phosphoria (Edman and Surdam, 1984a) and other parts of the section so that carbonate- or silica-rich, nearly monomineralic sedimentary rocks, isolated from pore fluids, could remain stable at and above 300~ Very finely grained silty limestone in the Dinwoody Formation of the Meade plate contains chlorite, biotite, amphibole, plagioclase, minerals of the epidote group, and garnet. The most abundant of these minerals are from the epidote group and are so abundant in places that the rock is green in outcrop. The metamorphic minerals are estimated to comprise as much as 30% of the rock in Dinwoody samples. Much of this green rock was found as far east as the Georgetown syncline in the Georgetown Canyon area. A suite of low-grade metamorphic minerals (garnet, chlorite, biotite, zoisite, and amphibole) was also found in samples of middle Paleozoic rocks from just above the Putnam-Paris thrust in southern Hatch quadrangle (Fig. 6-2). This distribution of low-grade metamorphic minerals in the upper Meade plate and the lower Putnam-Paris plate suggests that the upper strata of the Meade plate were subjected to a reverse geothermal gradient that could have coincided with emplacement of a warm Putnam-Paris plate. The thermal model ofYonkee et al. (1989) shows a P-T path for their Willard-Meade plate consistent with initial thermal gradients between 30 and 35~ km -l. Fluid pressures were less than lithostatic pressure in the hanging wall. Temperatures under 300~ in the footwall, consisting of the Crawford and Absaroka plates in this model, are also consistent with the findings of Yonkee (1983), Mitra and Yonkee (1985), and Mitra et al. (1988). The temperature of the rocks of the hanging wall (Willard-Meade plate) resulted from the pretectonic geothermal gradient in the region from which the upper plate rocks came. A reverse geothermal gradient was established after these relatively warm hanging-wall
Strain distribution and structural evolution of the Meade plate
143
rocks were thrust over the rocks that would eventually constitute the Crawford plate. According to this model, the base of the Meade plate should show at least low-grade metamorphism, but the metamorphic grade higher in the section should decrease (prehnite-pumpelleyite or zeolite facies?). The reverse appears to be true. If the Willard thrust model is applied to the Putnam-Paris thrust (Fig. 6-1), then the data from Yonkee et al. (1989) would indicate that metamorphism at the base of the Willard plate translates to a warm Putnam-Paris plate above the rocks that later became part of the Meade plate. Based on the presently perceived distribution of the epidote-bearing rocks in the Dinwoody Formation, the warm thrust plate model would require that the Putnam-Paris thrust system have extended much farther east than it does at present. It may have extended close to the present position of the Meade thrust or farther east, for a minimum amount of tectonic transport of about 40 km (compare with 48 km, Mitra and Yonkee, 1985; Allmendinger, 1992). There must be additional factors influencing the variations of paleogeothermal gradients because part of the Meade Peak Member and Dinwoody Formation near the Putnam-Paris thrust was not subjected to low greenschist-facies temperatures (Harris et al., 1980; Paul et al., 1985; Desborough et al., 1988), contrary to the warm thrust-plate model. Lacking additional information about distribution of low-grade metamorphic minerals, explanation of the presently inferred pattern of metamorphic rock and sedimentary rock capable of generating fluid hydrocarbon includes the possibility that postmetamorphic tectonism may have shuffled the strata of the upper Meade plate and disturbed Early Cretaceous metamorphic isograds. Some of this rearrangement may be due to concealed backthrusting, remobilization of the Putnam-Paris thrust in the Late Cretaceous or Early Cenozoic and emplacement of large tectonic lenses (duplexes) of unmetamorphosed Permian and Triassic rocks, Tertiary backsliding of unmetamorphosed rocks along former reverse faults or intraplate thrusts, or detachment faulting. The discussion above suggests a Cretaceous temperature profile of the Meade plate of >_300~ near the Putnam-Paris thrust(s) gradually decreasing to perhaps 250~ at the base of the Meade plate. The threshold of greenschist-facies temperature (300~ may have been reached at the stratigraphic level of the Phosphoria, possibly near the contact of the Meade Peak Member and Rex Chert, in which biotite was identified.
STRUCTURE P r e t e c t o n i c structures
The model of pretectonic changes in the Phosphoria Formation suggests that the stress affecting the rocks at that time would have been largely due to the lithostatic pressure of overlying Mesozoic sediments during compaction and following initial lithification. Boudinage, pinch-and-swell structure, and isoclinal folds can be produced in sedimentary rocks of varying lithification in the section and could have led to local anisotropic strain as the result of different yield strengths of the strata. Stylolites subparallel to bedding
144
J. G. Evans
suggest formation during Mesozoic compaction. The locally close relation of quartz and carbonate veins perpendicular to bedding and attached to stylolites suggests that the fractures that host the veins provided paths for silica- or carbonate-rich solutions that left behind laminae of insoluble residues, for example, hematite and clay minerals.
Syntectonic structures As a result of emplacement of the Putnam-Paris thrust system, most of the rocks that would eventually constitute the Meade plate were subjected to additional flattening from the overburden of the Putnam-Paris plate. Many of the same kinds of structures listed above as pretectonic may also have been produced during this phase of flattening. Allmendinger (1979) provided evidence that the rocks in the Meade plate closest to the Putnam-Paris thrust were deformed prior to emplacement of the thrust, so that the tectonic overburden in places may have been emplaced at an angle to bedding. One strategy for sorting structures into stages is to assign bedding-parallel strain features to the Sevier orogeny. Syntectonic structures can resemble structures developed during compaction (fractures, veins, stylolites, boudinage), but would theoretically be oriented as much as 90 ~ from similar pretectonic structures, assuming horizontal generally east-west compression and near-horizontal bedding. In places where bedding was rotated to steep dips on the limbs of macroscopic folds, regional northeast-southwest-trending to east-westtrending horizontal compression (see below) further flattened the rock along bedding. As a result, flattening parallel to bedding should comprise a large component of the total strain and it may have accumulated in as many as three stages: (a) during initial compaction by sedimentary overburden; (b) during emplacement of the Putnam-Paris plate; and (c) during continuing near-horizontal compression at large angles to bedding in steeply tilted strata. Some of the minor structures, like stylolites and quartz veins, that can be assigned to a younger layer-parallel compressive event according to the criteria mentioned above appear to be poorly developed in contrast to similar structures associated with compaction and later emplacement of tectonic overburden of the Putnam-Paris plate. These observations suggest that the rocks were relatively strain-hardened after early compaction events. In addition, the conditions during deformation (pressure, temperature, pore fluid volume, porosity, permeability, and a lack of appropriately oriented pre-existing planes of weakness, e.g. bedding) may have changed and been unfavorable for further development of certain kinds of minor structures by layer-parallel compression. The short duration of horizontal compressive stress would also have affected intensity of development of minor structures.
Orogenic and structural terminology Here, the name Sevier orogeny is used for thrusting episodes that took place from Early Cretaceous to Eocene (from Late Jurassic according to Armstrong, 1968; Heller et al., 1986) in the southeastern Idaho part of the fold-and-thrust belt. The Putnam-Paris thrust,
Strain distribution and structural evolution of the Meade plate
145
considered the oldest thrust in this part of the Sevier orogenic belt, was formerly included in the "Bannock overthrust" of Richards and Mansfield (1912) and Mansfield (1927), and the "Bannock thrust zone" of Armstrong and Cressman (1963) and Armstrong and Oriel (1965). Armstrong and Cressman (1963) showed that the Paris and Meade thrusts are separate imbricate thrust faults of the thrust belt. Northwest of Soda Springs, the Putnam thrust (Fig. 6-1) was recognized and named by Mansfield (1920, 1929). Kellogg (1992), Kellogg et al. (1999), and Rodgers and Janecke (1992) suggested that the Putnam thrust is connected to the Paris thrust by a thrust-transfer system, which further suggests that the Putnam thrust is partly coeval with the Paris thrust (Early Cretaceous). During recent field mapping, an attempt to map the northwestern extent of the Paris thrust as far as Chesterfield Reservoir was unsuccessful, as the critical area was covered by Cretaceous and (or) early Tertiary to Quaternary gravel and alluvium. The lower to middle Paleozoic limestone mapped in the southern part of the Hatch 7.5' quadrangle (Mansfield, 1929; Fig. 6-2) is in the Paris plate and suggests a continuous thrust trajectory between the mapped Putnam and Paris thrusts. The Meade plate, or duplex of Armstrong and Cressman (1963), has been assigned various relations with adjacent thrust plates. Here, the Crawford plate is considered to be in the footwall of the southern part of the Meade thrust. The footwall of the northern part of the Meade thrust is the Absaroka plate, as the Crawford thrust is not mapped north of 42 ~30' latitude (Dixon, 1982; Yonkee, 1983). The lower boundary of the Meade plate as used here is in part the one originally mapped by Mansfield (1927) as the Bannock overthrust, but is closer to the Meade thrust of Armstrong and Cressman (1963).
Timing o f thrusting in southeastern Idaho The oldest Mesozoic thrust in the region may be the Middle Jurassic Manning Canyon detachment of Allmendinger and Jordan (1981) and Allmendinger et al. (1984), about 70km west of the westernmost exposures of the Meade thrust plate. Movement on the Manning Canyon detachment was assigned to the Nevadan orogeny. Movement on the Putnam-Paris thrust, formerly thought to be Late Jurassic to Early Cretaceous in age (Armstrong and Cressman, 1963), was later found to be Early Cretaceous, based on fossil evidence indicating that the age of the syntectonic Ephraim Conglomerate is Aptian to Cenomanian (Heller et al., 1986). Principal movement on the Meade thrust occurred in Early to Late Cretaceous (Coniacian, about 88 Ma, Armstrong and Oriel, 1965; Albian to Turonian, about 105-90 Ma, Allmendinger, 1992). Based on provenance of clasts in the Bechler Conglomerate at Red Mountain, Idaho, about 1.5 km east of the present trace of the Meade thrust, DeCelles et al. (1993) proposed that the Meade thrust initially developed in Aptian (-118-113 Ma; Early Cretaceous). Deposition of conglomerate continued during Albian to Cenomanian time, during which the Meade thrust overrode the Bechler Conglomerate and other rocks in the footwall. Movement on the Meade thrust was probably completed by Cenomanian or Turonian time (-95-90 Ma).
146
J. G. Evans
Subsequent movements on Crawford, Absaroka, Darby, Prospect, and Hogsback thrusts range through the Late Cretaceous and Paleocene. These thrusts and their upper plates are important for understanding the structural evolution of the Meade plate because deformation in those plates may have affected the Meade plate. Geologic relations in the western part of the Meade thrust plate in the Chesterfield Reservoir quadrangle, southeast Idaho, strongly suggest that the Putnam-Paris plate to the west was uplifted after deformation of the Meade plate was complete. Evidence for this sequence of events is Cretaceous and (or) Tertiary gravel up to 400 m thick that contains conspicuous white blocks of early Paleozoic or Late Proterozoic quartzite as much as 5 m across and less conspicuous boulders of early Paleozoic limestones as much as 1 m across (Mansfield, 1927; Evans, unpub, mapping, 2000, Chesterfield Reservoir quadrangle). The uplift of resistant rocks in the Putnam-Paris plate could have accompanied movement on thrusts younger than the Meade thrust, such as the Absaroka or Darby thrusts, possibly as the Putnam-Paris plate was passively elevated above one or more thrust ramps (Schm'itt and Steidtman, 1990; DeCelles, 1994). Ramp-related uplift of the Putnam-Paris plate, however, is not obvious because geologic interpretations of seismic data (Dixon, 1982) suggest that the present extent of the Putnam plate does not overlie a younger ramp (see below). In addition, one or more of several Mesozoic or Cenozoic tectonic processes may have disrupted Cretaceous isograds and other geologic relations, so that the present relative locations of all parts of thrust plates may be different from their Cretaceous to Early Cenozoic arrangement. Also, the Putnam-Paris plate of today is a remnant of a more extensive plate, the leading edge of which may have originally been at least 40 km farther east.
Shortening of the Meade and other plates Leeman et al. (1992) suggested that the Sevier thrusting in the northwest Montananorthern Idaho area moved the upper plate of the Sevier system as much as 150 km northeast and was crustal in scale. Estimates of cumulative shortening for the Sevier foldand-thrust belt range from 43 to 54% (Rubey and Hubbert, 1959; Monley, 1971; Royse et al., 1975) and 100-150 km (Rubey and Hubbert, 1959; Royse et al., 1975; Claypool et al., 1978; Allmendinger, 1992; Craddock, 1992). Estimates have been made of more than 10-44 km of tectonic transport along the Putnam-Paris thrust (Wiltschko and Dorr, 1983; Allmendinger, 1992; Coogan, 1992; Cradddock, 1992; this chapter). Estimates of 23-48 km of tectonic transport of the Meade plate have been reported (Royse et al., 1975; Wiltschko and Dorr, 1983; Mitra and Yonkee, 1985; Allmendinger, 1992; Coogan, 1992; Craddock, 1992; DeCelles et al., 1993). According to models of Sevier thrusting, once movement along the Meade thrust ceased, the plate was passively carried an additional 60-75 km farther east on younger thrusts (Wiltschko and Dorr, 1983; Allmendinger, 1992). The Sevier orogeny was followed by Middle and Late Cenozoic Basin-and-Range regional extension, which needs to be considered in estimating Sevier-age shortening. Levy and Christie-Blick (1989) restored Basin-and-Range extension of ~250 km, most of which was west of the Meade plate. This restoration moved the leading edge of the Sevier
Strain distribution and structural evolution of the Meadeplate
147
fold-and-thrust belt 130 km west of the Idaho-Wyoming State line. This reconstruction of the entire Sevier orogenic belt is too broad in scope to assist in assessing Tertiary extension within the Meade plate in detail, but it suggests that extension in places may be about the same order of magnitude as the earlier shortening, similar to conclusions from crosssections (see below). Late Tertiary extension of the Meade plate probably concentrated in north-trending grabens that contain thick Tertiary sediment and volcanic rocks, such as Gem Valley, Blackfoot lava field, Willow Creek lava field, Grays Lake, and Bear Lake Valley. Slip on relatively minor reverse and intraplate thrusts may also have been reversed.
Direction o f tectonic transport Eastward tectonic transport of thrust plates during the Sevier orogeny is based on the assumption that movements on the thrusts are predominantly up-dip, resulting in older rocks on top of younger ones (Armstrong and Oriel, 1965; Armstrong, 1968; Royse et al., 1975) and overturning of eastern limbs of folds to the east. Other estimates are based on structural analyses. Crosby (1968, 1969, 1970) studied minor structures at 22 localities across the Idaho-Wyoming-Utah fold-and-thrust belt, and concluded that the transport directions varied across an arc of about 80 ~, from northeast to east-southeast. These transport directions are approximately perpendicular to major fold axes in the Meade plate (Fig. 6-3). Other estimates of direction of transport based on macro-, meso-, and microscopic fabric are generally about east-west (Allmendinger, 1981, 1982; Craddock, 1992).
Thickness o f the Meade thrust plate and topology o f the Meade thrust In contrast to the 2.4 km maximum thickness of the Meade plate suggested by Mansfield (1927, 1929, 1952), more recent work has inferred that the thickness is much greater, 5.8 km (Edman and Surdam, 1984b). Royse et al. (1975) showed variable thickness (2.4-5.5 km) of the Meade plate, with general thickening to the west and merging with the Putnam-Paris thrust at the regional Mississippian detachment under Gem Valley, about 5 km east of Soda Springs. Dixon (1982) interpreted proprietary seismic data from southeastern Idaho using several assumptions, some of which have not been substantiated. Nevertheless, his work is one of the most important in the region because it provides interpretations of deep structure. Dixon interpreted seismic profiles across the Meade plate to indicate a vertical thickness of the plate of as much as 3-4 km in the east and from 9.5 to 14 km in the west. Structure contours on the Meade thrust were derived from Dixon's cross-sections (Fig. 6-4). Maximum thickness perpendicular to the principal (first) ramp is about 12 km. Dixon's cross-sections 25-47 were not used because this study concentrates on the northern and broader segment of the Meade plate. Cross-sections showing the greatest depth of the thrust are numbers 1 through 9; towards the Bear Lake area (Fig. 6-1) to the south, the depth of the Meade thrust decreases to about 9.5 km.
148
J. G. Evans 111 ~ 30 r
43 ~ 15'
111 ~
I
Willow~ Creek
~)
~
Lava Field
0! " " '~-~
.... '. . . .
~
""-,
Absaroka Plate
10km I
Anticline Syncline
43 ~
\ \ \
\
Blackfoot Lava ~, Field
_
\ \ \
%, \
i 42 ~ 30'
Fig. 6-3. Central part of Meade plate showing large fold axes (after Mansfield, 1927, Plate I). Axes of most large folds trend from north-northeast in the south to northwest in the north. Figure does not show Rasmussen fault of Pratt and Oriel (1981) and Oberlindacher et al. (unpublished mapping, Lower Valley Quadrangle). The steeply dipping or vertical Rasmussen fault is about 5 km north of the Blackfoot fault, strikes east, and shows left-lateral offset of formations and structure. SA, Snowdrift anticline; GS, Georgetown syncline; LGA, Little Gray anticline.
149
Strain distribution and structural evolution of the Meade plate
112 ~
43 ~30'
Eastern L_Snake River Plain _.., . \
\\
01A t
I
~
-,0,000 -20,000
,,.,
-30,000 -40,000
111 ~
Meade plate Meade thrust Absaroka-Crawford plates Absaroka Plate
43 ~
/ Approximate location of Putnam-Paris thrust(s) 60
Putnam-Paris Plate(s) 95 ~
43 ~30'
107'
j
Crosssection
2o..___.A.~'
Crawford Plate
~' o r o00
0
10km
I
J
Fig. 6-4. Structure contours on northern part of Meade thrust; elevation data from Dixon (1982); contour interval 5000 ft (1525 m); thick lines show axes of folds in Meade thrust; north-trending faults are Cenozoic in age; scale, latitude, and longitude are approximate. Cross-section A-A' (inset) approximately parallels general strike of this part of the thrust and is drawn with no vertical exaggeration; top of cross-section is 0 ft, mean sea level.
150
J. G. E v a n s
The structure contours on the Meade thrust (Fig. 6-4) are drawn as far to the west as can reasonably be inferred from Dixon's (1982)cross-sections. A contour interval of 5000 ft (1525 m) was used and the normal faults associated with the Bear Lake fault zone show substantial offsets that were taken into account when drawing the contours in the southern part of the study area. Feet are used in Fig. 6-4 to facilitate direct reference to Dixon's cross-sections. The Meade thrust is characterized by ramp and "fiat" geometry (Fig. 6-4), which is considered typical of thrust faults (Boyer and Elliott, 1982). Below-35,000 fl (below sea level; - 1 0 , 6 7 5 m), the Meade thrust appears to flatten (trailing branch), presumably before merging with the underlying regional Mississippian d6collement from which the Putnam-Paris thrust system originated (Royse et al., 1975). The largest part of the Meade thrust is a lower ramp most of which is between -10,000 ft ( - 3 0 5 0 m) and -35,000 ft (-10,675 m). The ramp, varying from 18-20 km wide horizontally, extends from the northwest end of the plate southeast of the Snake River Plain to what appears to be a broad, central, strongly folded zone. From there, it continues southward just west of the Bear Lake fault and associated faults (north-striking normal faults in the southeastern corner of Fig. 6-4). Dip of the lower ramp varies from as much as 47 ~ southwest at the northwestern end to as little as 36 ~ west-southwest at the latitude of Dixon's (1982) cross-section 17 (42 ~ 10'-15'N latitude). About 3 km east of the leading edge of the Putnam-Paris thrust, a northwest-trending cross-section, drawn to cut the deepest part of the Meade thrust, shows the thrust as a line. Its curved path suggests shortening of about 9% parallel to the orogen, assuming the thrust originated as a planar structure. Two relatively flat segments of the Meade thrust above the first ramp are located between 42 ~30'-45'N latitude and at depths between -10,000 ft ( - 3 0 5 0 m) and near sea level, and between 43 ~0 ' - 2 0 ' N latitude and at depths between - 10,000 ft ( - 3 0 5 0 m) and - 5 0 0 0 ft ( - 1525 m). The dip of the upper "fiat" is as much as 22.5 ~ southwest in the north to 15~ west-southwest at the latitude of Dixon's cross-section 17. Although this part of the thrust is not actually "fiat" (0 ~ dip), the dip angle of the thrust changes relatively abruptly by about 50~ from the lower ramp to the second "fiat." The broad extent of this "fiat" may result from the Meade thrust following a zone of weakness in the Jurassic Twin Creek Formation (Evans and Craddock, 1985) and (or) a zone of salt in the Preuss Formation (Coogan and Yonkee, 1985). Alternatively, the shape and extent of the southern "fiat" may reflect variable displacement along the Meade thrust (see below). The variable displacement, however, may be controlled by competency of the strata encountered. The uppermost 5000 ft (1525 m)-8000 ft (2440 m) parts of the thrust above sea level in the south and the uppermost 10,000 ft (3050 m) in the north appear to steepen, suggesting that these parts of the thrust comprise the lower part of an upper ramp. Most of the characteristics of the Meade thrust are duplicated in the younger thrusts of the Sevier belt. Dixon (1982) drew structure contours on the Absaroka, Prospect, and Hogsback thrusts, and Mitra and Yonkee (1985) drew structure contours on the Crawford thrust using Dixon's data. These thrusts have lower ramps and middle "fiats," like the second "fiat" of the Meade thrust, that are defined by abrupt decreases in dip. The thrust faults steepen abruptly into upper ramps before they are truncated by the erosion surface.
Strain distribution and structural evolution of the Meade plate
151
The structure contours on these thrusts appear to show folds that trend at large angles to the strike of the thrusts like the folds in the Meade thrust (Fig. 6-3). In addition, Dixon (1982; Fig. 16) interpreted structure contours in the central part of the Darby thrust as folds with hinges subparallel to the trend of the orogen. Dixon attributed the flattening in the thrusts younger than the Meade to interaction of the thrusts with Cretaceous foredeep shales or basement blocks.
Style o f deformation o f the Meade thrust and plate Axes of apparent macroscopic folds in the Meade thrust trend from 50 to 110 ~ and are most prominent in the central part of the study area (Fig. 6-4). Trends of macroscopic to subregional folds defined by stratigraphy at the surface, however, are mostly at right angles to trends of folds in the Meade thrust (Fig. 6-3), although a few subordinate macroscopic northeast-trending folds cross the dominant set at the surface. The northeast-trending folds of the Meade thrust suggest the possibility that the rocks below the second "fiat" (below -10,000 fl; - 3 0 5 0 m) may have different structural components or components that are different in style, orientation, and (or) development from the structures in exposures. The axes of folds in the Meade thrust parallel directions of tectonic transport inferred by Crosby (1968, 1969, 1970) from mesoscopic structural data. The large north-northeast- to northwest-trending folds and the northeast-trending folds may be components of a B perpendicular to B' tectonite (Hills, 1963), suggesting that the folds formed penecontemporaneously. Northeast-trending folds may be dominant at depth, and subordinate at the surface. Above sea level, the northeast- to east-trending folds of the Meade thrust that are so obvious along the lower "fiat" and ramp are not noticeable above what can be seen of the top of the upper "fiat" (Fig. 6-4). Both observations suggest a layering of deformation styles in the plate. Concentric folds were considered the typical fold type in the Sevier fold-and-thrust belt (Mansfield, 1927, 1929, 1952; Royse et al., 1975; Dixon, 1982) despite evidence to the contrary. Others challenged the assumption of parallel flexural-slip folds in the Cordilleran thrust zones and instead described the kink-band and box fold geometry (narrow hinges and long, planar limbs; Allmendinger, 1981, 1982; Boyer, 1986; Coogan, 1992). The biggest and most easily seen box fold in the Meade plate is near the south end of the Georgetown syncline near the former site of the Georgetown Canyon mine offices and plant (in the process of dismantling and reclamation in 2001; Mansfield, 1927; Cressman, 1964). The ramp-fiat model of thrusting postulates that folds in the overriding plate form as the result of ascending a ramp, which forms a syncline at the base of the ramp and an anticline at the top as the strata are constrained to flex in one direction, unfold, and then flex in another direction (Rich, 1934; Davis et al., 1983; Suppe, 1983). While this model of strain may seem reasonable for folds near the Meade thrust, other large folds in the plate occur several kilometers above the Meade thrust and involve formations of varying competence. The relevance of ramp-related folding for these structures is not clear. Some large folds like the Rock Creek Syncline (Chesterfield Reservoir quadrangle, Mansfield, 1929)
152
J. G. Evans
are relatively close to the Putnam-Paris thrust and may have been generated in strata modified by displacements along the Putnam-Paris thrust before, during, or even after movement along the Meade thrust. Dixon's cross-sections, work by Lageson and Schmitt (1994), and suggestions from more detailed local analyses (Allmendinger, 1979, 1981, 1982) suggest that the rocks near the Meade thrust deformed differently from those higher in the plate. If so, then the rampflat model of folding may not be appropriate for the entire Meade plate, especially at depths below -10,000 ft ( - 3 0 5 0 m). Surficial structure in the Meade plate is dominated by nearly plate-long, mostly upright folds (e.g. the 75-km-long Snowdrift anticline) that trend northwest near the eastern Snake River Plain, and north to north-northeast east of Soda Springs (Fig. 6-3). Some of the synclines in the western part of the plate have western limbs overturned to the east.
Shortening and extension implied by folding and faulting in the Meade plate Numerous published cross sections drawn across the Meade thrust approximate the amount and locations of macroscopic strains that the Meade plate has undergone (Table 6-I). The age of shortening is generally not apparent in a complex orogenic zone like the Sevier orogenic belt. Some of the deformation within a thrust plate, including folding and faulting, may be associated with significant movement of an older or younger thrust. Wojtal and Mitra (1986) suggested that much of the internal shortening in a thrust plate develops concomitantly with the underlying thrust zone itself, but that internal strain may have an upper limit depending on local PTx conditions, beyond which strain tends to be concentrated in a basal strain-softened, cataclastic zone. In addition, as deformation progresses, there may develop a basal zone where the rocks are flattened and extended while rocks higher in the plate are compressed, thickened, and shortened (Allmendinger, 1979; Wojtal, 1986). As suggested above, the Meade plate may contain this layering of deformational styles. Consequently, inferences about shortening derived from surficial geology may apply only to the uppermost 10,000 to 18,000 vertical-feet (3050-5490 m) of the Meade plate. Several assumptions are used here in determining a minimum amount of shortening from cross-sections (Table 6-I): (a) the trends of the cross-sections approximate the profiles of the folds, and parallel the principal direction of shortening; (b) an insignificant amount of material moved across the plane of section; and (c) mesoscopic and microscopic strain were insignificant additions to total shortening. As mentioned above and below, assumption (b) may be untenable. There is evidence of relatively minor (joints, faults, folds, veins) extension parallel to major fold axes and, therefore, through the planes of cross-sections; the amount, however, may be less than 10%. Assumption (c) mostly points to the inability to correlate relations among macroscopic, mesoscopic, and microscopic strain components. Also, macroscopic strain may be the result of distributed smaller-scale processes and may give an estimate of total strain within the plate, irrespective of when it occurred. Measurements of the length of the upper contact of the Meade Peak Member of the Phosphoria Formation are used here as a proxy for estimating shortening because of the
Strain distribution and structural evolution of the Meade plate
15 3
TABLE 6-I Macroscopic shortening and extension within Meade plate
Cross-
Reference
section
% shortening
% shortening
% total
% extension
% cross-
Length o f
by folding
by faulting
shortening
(faulting)
section used
crosssection used (km)
1. D - D ' , plate l l
Mansfield
18
-
18
19
60
2.3 4.5
(1927) 2. F - F ' , plate 11
Same
21
-
21
-
90
3. G - G ' plate 11
Same
24
-
24
8
99
8.2
4. I - I ' plate 11
Same
16
5.7
21.7
0.3
75
16.5
5.4
18.8
5. K - K ' , plate 11
Same
16
21.4
1.1
80
6. L - L ' , plate 11
Same
7
-
7
1.8
60
9.8
7. M - M ' , plate I1
Same
12
4
16
3.2
100
19.3
plate 11
Same
7.4
-
7.4
-
50
9.4
9. O - O ' , plate 12
Same
6
0.3
6.3
1.6
90
29.4
1.3
! 7.3
0.7
75
26
-
74
7
6O
2.3
8. N - N '
10. S - S ' , plate 12
Same
!6
11. V - V ' , plate 12
Same
6
-
6
12. A-B, plate 2
Mansfield
23
-
23
2.8
--
33
1.8
(1929) 13. C - E , plate 2
Same
33
95
16.8
! 4. F- (;, plate 2
Same
33
33
1oo
9.8
15. tt-I, platc 2 16. A A' (cast)
Samc Rioux ct al.
18
18
80
6.6
15
15
2.7
57
6.5
!.3
(1975) 17. B B'
Same
20
3.3
! 3.3
18. A A'
('ressman
i3
!.8
14.8
19.5 -
11
68
8.2
100
8.3
-
86
4.7
5.5
82
9.6
-
63
5.1
and Gulbrandsen (1955) 19. B - B '
Same
19.5
20. C - C '
Same
I1
2 !. D - D '
Same
8
22. E - E '
Same
17.3
18.2
35.5
4.2
56
2.4
23. A - A '
Cressman, (1964)
36.4
1.7
38.1
-
97
8.7
8
Snowdrift Mtn. 24. B - B '
Same
31
1.2
32.2
2.9
78
8.3
25. C - C '
Same
19.5
-
19.5
-
86
4.7
26. D - D '
Same
22.5
0.6
23. I
1.1
71
7.6
27. E - E '
Same
26.2
0.9
27.1
-
68
7.2
28. F - F '
Same
26.8
-
26.8
-
60
6.3
29. A - A '
Cressman ( ! 964)
32.4
3
35.4
! 4.1
83
17.3
(Meade Peak)
Continued
154
J. G. Evans
TABLE 6-I Continued Cross-
Reference
section
% shortening
% shortening
% total
% extension
% cross-
Length of
by folding
by faulting
shortening
(faulting)
section
cross-
used
section used (km)
30. C - C ' 31. A - A '
Same Gulbrandsen
18.7 6.6
4.6 -
23.3
60
12.6
6.6
11.7
-
98
10.62
8.8
10.4
97
9.3
14.6
3.9
89
4.9
et al. (1956) 32. B - B '
Same
8.2
33. C - C '
Same
14.6
34. D - D '
Same
3.5
0.6
4.1
7.3
100
4.3
35. A - A '
Montgomery
8.5
-
8 . 5
-
100
4.7
0.6 -
and Cheney (1967) 36. B - B '
Same
9.6
9.6
-
100
10.7
37. C - C '
Same
12.8
12.8
-
100
10.7
38. D - D '
Same
16.2
16.2
-
96
9.9
39. E - E '
Same
21.3
21.3
-
97
! 0.1
principal interest in this unit. Strains are likely to be different at other stratigraphic levels owing to changes in magnitude and intensity of the strain resulting from differences in depth, temperature, proximity to bounding thrusts, history of deformation, and responses to strain of the rocks under the physical conditions at those levels. Here, the cumulative lengths of the upper contact of the Meade Peak Member are compared with the lengths of each segment of the cross-sections that show significant amounts of Meade Peak. Lengths of segments of cross-sections used initially ranged from 20 to 100% of published sections. Parts of cross-sections were ignored because of focus on the Phosphoria Formation and, therefore, only parts of cross-sections were used in which the Phosphoria could be reasonably projected above or below the erosion surface. In addition, interpretations of many structures were not well constrained in the field beyond the Phosphoria Formation with its diagnostic lithologies. Extrapolation and interpretation of fold profiles to depths below Mansfield's Meade thrust (his Bannock overthrust) is realistic because Dixon (1982) determined the thrust plate to be at least twice as thick as Mansfield's estimates. Dixon also showed that folds exposed at the surface extend down nearly to the Meade thrust in the eastern, thinner part of the plate. Some unknown factors that could be important in framing the context of the internal deformation of the Meade plate include the original eastward extent of the Meade thrust and original thickness of this missing part of the Meade plate. In addition, there is no clear way to separate deformation seen in most cross-sections listed in Table 6-I into clear stages. Allmendinger (1979, 198 I), however, showed that broad open folds in the Meade plate were truncated by the Putnam-Paris thrust. Some of the deformation in what is now the Meade plate, therefore, dates from or preceded emplacement of the Putnam-Paris thrust. Slip may have occurred on the Putnam-Paris thrust during eastward translation of
Strain distribution and structural evolution of the Meade plate
155
the Meade plate on the Meade or younger thrusts (Fig. 6-1). To complicate matters further, Mansfield showed some normal-slip arrows on faults that otherwise show shortening of the section. One interpretation of this apparent anomaly is that Cretaceous shortening was larger than Tertiary extension along these faults although not much greater. However, the reason Mansfield reached this interpretation is not clear. Detachment faults with this kind of displacement history are proposed for the Sevier orogenic belt in west-central Utah (Allmendinger et al., 1983). The effect of this extension is to reduce the apparent amount of shortening by faulting at least in the upper levels of the plate. Macroscopic extension by faulting in some cross-sections of the Meade plate is of the same order of magnitude as macroscopic shortening (Table 6-1). The most reliable estimates of shortening within and across the Meade plate are probably along the longest segments of the longest cross-sections. For purposes of this study that focuses on the northern part of the Meade plate, segments of cross-sections that are less than 50% of the original cross-sections are excluded as potentially unrepresentative of macroscopic shortening. Using this criterion, 20 cross-sections of Armstrong (1969), Allmendinger (1979), and Kellogg et al. (1989, 1999) are excluded. These short segments have a higher average shortening than the longer segments (29.7 vs. 18.8%), are considered atypical for estimating shortening in the Meade plate, and are not included in Table 6-I. However, some of these cross-sections may more accurately portray the amount of shortening in narrow domains adjacent to the floor and roof thrusts. Shortening and extension of 39 cross-sections meet the criterion of reliability. Within the Meade plate, shortening from folding varies from 3.5 to 36.4% with a median between 15 and 20% and a sub-maximum between 5 and 10%. The median and average (18.8%) are close. In general, estimates of orogen-normal shortening by folding are consistent where cross-sections from separate studies overlap or are near one another. Shortening by folding is 14-36% near the Meade thrust. In the northwestern and upper part of the Meade plate, from the Lower Valley quadrangle to Paradise Valley quadrangle (Fig. 6-2), orogennormal shortening is 18-33%. A north-northeast-trending cross-section in the Snowdrift Mountain quadrangle (Cressman, 1964) shows 6% shortening by folding parallel to the axis of the Snowdrift anticline (Snowdrift Mountain quadrangle; Figs 6-2 and 6-3). This cross-section contradicts the assumption that material was not moved parallel to the trend of the orogen. The amount of orogen-parallel shortening by folding is of the same order of magnitude as the estimated strike-parallel shortening of 9% mentioned above for the Meade thrust (Fig. 6-4, upper right). The amount of shortening related to reverse or within-plate thrust faulting is as much as 18.2%, assuming largely dip slip, but most sections portray shortening by faulting of less than 50, and many of these less than 2%. Shortening by faulting is absent in most of the sections (discussed later), but occurs in a north-trending zone that lies north of the large normal faults that strike north from the Bear Valley graben (Fig. 6-4). The southern part of this zone of faulting is truncated by the Meade thrust. Total surficial orogen-normal shortening (by folding and faulting) near the Meade thrust ranges from 17.3 to 38.1% in the Harrington Peak and Snowdrift Mountain quadrangles and between 18 and 33% north and west of Blackfoot Reservoir. Between these two areas, total
156
J. G. Evans
shortening varies from 4.1 to 35.5%; however, all the cross-sections with total shortening of 11% or less are in an east-west-trending middle zone in parts of the Johnson Creek, Dry Valley, Diamond Flat, and Stewart flat quadrangles (Figs 6-2 and 6-5). Location of the zone of shortening by faulting above the upper margin of the lower ramp in the Meade plate suggests that this faulting is related to the change in dip during emplacement when these rocks were refolded at the ramp-flat hinge as they approached the surface and responded to further stress by brittle failure rather than by more ductile processes. Ramp-related deformation may be important in the central part of the plate about 8 km horizontally from the Putnam-Paris thrust and as much as 14 km horizontally from the easternmost part of the Meade thrust (Fig. 6-5). Extension by faulting was found in 21 cross-sections. Fourteen of the cross-sections showed extension of less than 5%. Five of the cross-sections showed extension ranging from 7.3 to 19%. Twelve cross-sections that showed extension also showed shortening by faulting. Part of this association may be a result of creation of fault zones of the right orientation for extension (parallel to the lower ramp), or possibly portions of the plate that experienced early brittle compressive deformation were strain-softened by fracturing, and slip reversal had the lowest energy requirements of available deformational processes. Orogen-normal extension by normal faulting is 1.8-8% in the southwestern part of the study area, 1.1-14.1% in the Harrington Peak and Snowdrift Mountain quadrangles,
1 1 2 <~
111"
43 ~ 30'
Fig. 6-5. Strain regimes superimposed on map of structure contours on Meade thrust. Singlehachured lines mark approximate boundaries of area showing 11% or less shortening by folding; double-hachured line marks approximate boundary of area showing shortening by faulting; triple-hachured lines mark approximate boundaries of areas showing extension by faulting; western part of southern area of extension coincides with area of shortening by faulting.
Strain distribution and structural evolution of the Meade plate
157
and from 1.1 to 19% in the central part of the study area (Fig. 6-5). In addition, the ramp-flat hinge may have influenced the location of the normal faults that control the Bear Lake graben. Extension also occurred farther to the northeast than early fracturing and in a separate area to the northwest adjacent to the Putnam-Paris thrust (Portneuf 15' quadrangle of Mansfield, 1929; Fig. 6-5). The northwestern area of extension suggests that strain softening by early fracturing from the emplacement of the Putnam-Paris thrust may have occurred there, although shortening by fracturing does not show up on cross sections by Mansfield (1929). Recent mapping in the Chesterfield Reservoir quadrangle (Evans, unpub, mapping, 2000), however, indicates that faulting in that area was greatly minimized in past studies. Extension in the northwestern area (Fig. 6-5) also offers a possible explanation for apparent local juxtaposition of rocks affected by relatively low-temperatures (within oil generation window; Paul and Paul, 1986; Desborough et al., 1988) closer to the warm Putnam-Paris thrust than to overmature greenschist-facies rocks. This arrangement of low-temperature subgreenschist parts of the Meade plate may be tectonically juxtaposed in relatively minor extensional duplexes adjacent to the Putnam-Paris thrust. Unmetamorphosed rocks from higher in the Meade plate, according to this hypothesis, slid down under the Putnam-Paris thrust.
Estimates o f shortening and compression directions from other data Fabric studies by Protzman and Mitra (1990) and Craddock (1992) suggest that some of the microscopic to mesoscopic fabric in the Meade plate or its footwall, the Crawford plate, pre-dates thrusting. Craddock suggested that the carbonates recorded a pre-thrusting layer-parallel and thrust-transport-parallel strain and a syntectonic (synthrusting) non-layer-parallel strain. Calcite fabric data from the Meade plate (Craddock, 1992) indicated average maximum shortening axis and the Turner (1953) compression axis in the tectonic non-layer-parallel fabric of country rock to be close to one another and at large angles to major surficial folds in the Meade plate (Fig. 6-3); they are 9 and 16~ from the trend of the closest fold in the Meade thrust (Fig. 6-4). Shortening and compression axes in the vein fabrics in the Meade plate trend more northeasterly and plunge west-southwest and are likely to be associated with the main intraplate compressive deformation of the Meade thrust plate and may not be affected by strain connected with plate transport. Craddock (1992) determined layer-parallel compressive strains from calcite twinning lamellae in country rock from each major thrust plate and subdivided fabric data from country rock and veins. Syntectonic calcite veins showed another strain pattern. Calcite strain magnitudes ranged from 2 to 16% shortening with the largest in vein samples. The largest amount of shortening reported is theoretically at the maximum that can be obtained from twinning fabrics before strain hardening develops and causes other strain mechanisms to be favored (15%, Burkhard, 1993), and therefore, may underestimate total shortening.
158
J G. Evans
Craddock attributed rotations of the pre-thrusting fabric to dextral transpression related to plate convergence along the western margin of North America and estimated a ratio of dextral fault displacement to thrust shortening of 7.5: 1, or 1650 km of strike-slip, which may seem extreme for the Sevier orogen. However, similar magnitudes of strike-slip have been proposed for other orogens (see Craddock and van der Pluijm, 1988; Craddock, 1992, and references therein). In addition, as deformation progressed and bedding was folded, the same compressive stress responsible for earlier shortening would have a different orientation with respect to bedding and could initiate twinning at higher angles to bedding. In this case, Craddock may have measured different components of the total strain during the Meade thrusting episode of the Sevier orogeny, so that average strains inferred from his calcite fabrics would be additive, or about 11%, not far below the 18.8% average shortening from macroscopic structures. The average macroscopic shortening in this case, however, conceals the distribution of strain with greatest strain (>-30%) closest to the roof and floor thrusts. Allmendinger (1982) studied calcite e-lamellae, dolomite f-lamellae, and sub-basal I deformation lamel|ae in quartz from two folds overturned to the east in the northwestern end of the Meade plate (Northern Blackfoot Mountains). He concluded that early compression was parallel to bedding and later compression was at a large angle to bedding (same as stage 3 of flattening, see Syntectonic structures). Regional shortening and probably compressive stress were horizontal or subhorizontal throughout the deformation. Most of the strain within the thrust plates in the Northern Blackfoot Mountains is directly related to folding, not thrusting. Strain from basal traction along fault surfaces is only important within 50-200 m of faults.
Variable displacement
Dixon (1982) presented diagrams that show lateral differences in displacement along the Absaroka, Prospect, Darby, and Hogsback thrusts (Fig. 6-6). Because of the apparent alignments of displacement maxima and minima along the Absaroka and younger thrusts, the geology along certain of Dixon's cross-sections was examined for evidence of variable displacement effecting the Meade thrust. A displacement maximum on the Absaroka, Darby, and Prospect thrusts of special interest here is located along Dixon's cross-section 10 and a minimum near cross-section 20 (Fig. 6-6). Displacement differences between maxima and minima are as much as 15 miles (24 km) on the Absaroka and younger thrusts. Just to the southeast of crosssection 10, the Meade plate broadens toward the east over the large second "flat" in the southern part of the thrust (Fig. 6-4), and several intraplate east-west- and northeaststriking strike-slip faults, that are consistent with broadening of the thrust to the east, cut the plate (Fig. 6-7). Just north of the displacement minimum along cross-section 20, the eastern salient of the Meade thrust cuts sharply west, a path that seems consistent with approach of the displacement minimum. A steep east-west-striking fault or fault zone
Strain distribution and structural evolution of the Meade plate
35 30 _Darby and Pros ect thru~ t~ y 25 20 _ ~ i !~~ 15
159
m
a~
n . n
E
j.,
I
Hogsbackthrust
25 20 15 10 5 0 25
10
t-"
E o a
~
20 ~
5
s
0 30
E
8
20 121 15 10 5 0
25 20 15 10
5
10
15 20 25 30 35 Cross-sections (Dixon, 1982)
40
Fig. 6-6. Displacements on cross-sections of Dixon (1982, modified from his Fig. 17).
abuts against the southern limit of the Meade plate salient, cuts through the Crawford plate, across the Meade plate, and into the eastern part of the Paris plate. This fault is close to and subparallel to cross-section 20 and is a type of structure that might be expected to be associated with displacement discontinuities. Another manifestation of the displacement minimum may be the disappearance of the Crawford thrust close to the path of crosssection 20. The Meade plate may have been affected by a displacement maximum manifested during thrusting in broadening of the second flat and generating the eastern salient of the plate. The strike-slip faults near cross-section 10 may be the late brittle response of the Meade plate to variable displacements on the Absaroka and (or) younger plates. Displacements on the leftlateral Rasmussen and Blackfoot strike-slip faults may have helped to spread the Meade plate eastward along or close to the zone of maximum displacement. The total displacement difference between maximum and minimum in this part of the plate is about 28 km. This displacement may have accumulated during emplacement of the Meade thrust, and later in response to deformation in the Absaroka and younger thrust plates. The apparent confinement of the strike-slip faults to the Meade plate suggests that slip along the Meade thrust was also involved at that time. Dixon's (1982) estimates of maximum and minimum displacements, from almost all of his 47 cross-sections, do not show how this variation relates to geologic structures at the surface (Oriel and Platt, 1980). The dearth of macroscopic structures associated with variable displacement in the Crawford, Absaroka, and younger thrusts may point to very local accommodation of lateral variations of displacement by
160
J. G. Evans
111o15' '~~Pelican / ~ I' fault7 7 Fau/Its " ~ ,c'"-.~~'\,-'~ ~b
112~ 43~ k
~ Chesterlield ~ servoir
Blackf,ot , ~ / / ~ - . . - ~ . ~ ' = ~ " ~ reservoir ~ / ,~. Rhyolitedomes ~ \~and basaltc o n
%~,
\. '%,
/" s
e
~
plate N
:f. " I ~ ~
\
\ \
\
\
'x
Absaroka
_
-~
~_
,
\ ?',
J
Putnam-Paris plate 42 '~ao' -
~::~ ~o ~
~v'~ \ '~
e 4I Fault
c
]Crawford 7 plate ion 20
\ i
Fig. 6-7. Selected geologic structures from Preston 2~ sheet (Oriel and Platt, 1980) that are consistent with continuation of a displacement maximum along cross-section 10 and a displacement minimum along cross-section 20 projected from Absaroka, Darby, and Prospect thrusts into Meade plate.
minor thrust faults or internal deformation of incompetent strata, such as formation of cleavage like that developed in the Twin Creek Limestone in the Crawford plate (Yonkee, 1983; Mitra and Yonkee, 1985; Mitra et al., 1988). Some of the structures developed in response to variable displacement in the younger plates may not be obvious, may not penetrate to the present erosion surface, or may have weakened the rocks involved so that they are easily weathered and disaggregated, like the weathered Twin Creek Limestone. The displacement variation, especially the maximum and minimum along cross-sections 10 and 20, appear to correlate with the salients and reentrants of six of the main plates (Fig. 6-1; Dixon, 1982; Fig. 4), including the Meade plate, thereby controlling the gross outline of the fold and thrust belt. Formational thicknesses would vary to compensate for the changes of the area of the unit contacts and this process could result in material transported parallel to the orogen to compensate for thinned portions of the formations. This lateral relocation of material may be at least partly responsible for a tectonic component of variation of apparent stratigraphic thicknesses. For displacement variation to have aligned maximum and minimum deformation zones in possibly as many as six thrust sheets, as may be the case in this
Strain distribution and structural evolution of the Meade plate
161
part of the Sevier fold-and-thrust belt, basement control may be involved. One obvious type of large-scale structure suggested by the alignment of displacement maxima parallel to cross-section 10 is a deep, long-lived, crustal shear zone that parallels the trend of the eastern Snake River Plain and may have been characterized by fight-lateral slip.
CONCLUSIONS Principal events in the history of the Meade plate are summarized in Table 6-II. Only the Cenozoic events mentioned in the table have not been described extensively here, but may be inferred from the published geologic maps of the area and from Pierce and Morgan
TABLE 6-II Proposed strain history of the northern Meade plate with focus on Phosphoria Formation Time
Events
Permian to Jurassic
Compaction; soft-sediment deformation
Early to Late Cretaceous
Compaction from tectonic overburden; greenschist-facies metamorphism by upside down thermal gradient associated with possibly oblique emplacement of Putnam-Paris plate; early layer-parallel, orogen-normal compression (4-38%); activation of Meade thrust; orogen-parallel shortening of Meade plate (6-9%); Meade plate transported eastward; lateral displacement variation affects the gross outline of Meade thrust, Meade plate, and other plates possibly along incompetent strata or active basement structures; erosion of Putnam-Paris plate
Late Cretaceous
Meade plate generally immobile after displacement transfer and (or) activation of Crawford and Absaroka thrusts; Meade plate passively(?) carried eastward on younger thrusts; erosion of Putnam-Paris and Meade plates
Late Cretaceous
Rejuvenated uplift of Putnam-Paris plate(s); burial of western Meade plate by late orogenic sediments from the Putnam-Paris plate; possible emplacement of tectonic lenses of unmetamorphosed rocks of Meade plate immediately under Putnam-Paris plate
Cenozoic
Extension of Sevier fold-and-thrust belt results in westward movement of Putnam-Paris plate and parts of Meade and possibly tectonic lenses of unmetamorphosed Meade plate rocks; relatively minor local extension related to emplacement of basalt and rhyolite intrusions and the bow-wave fracturing effect of the Yellowstone hotspot
162
J. G. E v a n s
TABLE 6-III Thrust-model processes and geometry affecting the Meade thrust system Process/geometry
Evidence/interpretation
Ramp-flat geometry
Structure contours on Meade thrust
Deformation of thrust; emplacement and erosion of Meade Peak
Folds trending down dip of Meade thrust orogenic; Bechler Conglomerate
Ramp-related(?) uplift of hinterland (Putnam-Paris plate)
Late orogenic conglomerate derived from Putnam-plate buries western Meade plate following deformation of Meade plate
Ramp-related deformation
Zone of macroscopic shortening by faulting parallels and is above hinge between lower ramp and upper flat
Variable displacement along strike of Sevier orogenic belt
Crawford thrust disappears into an anticline; projection of zones of displacement maximum and minimum in Absaroka and younger thrusts affect Meade plate; large southern flat in area of maximum
Reverse geothermal gradient from and emplacement of warm Putnam-Paris plate
Greenschist-facies metamorphism in upper Meade plate and lower Putnam-Paris plate
Strain softening by fracturing
Geographic correlation of shortening by faulting with extension by faulting
Thrust and reverse faults reactivated for (west) extension
Possible offset of Cretaceous isograds and backsliding of Putnam-Paris plate relative to Meade plate
(1992). The fold-and-thrust-belt processes and the geometry observed in the Meade thrust and plate are summarized in Table 6-III. No clear evidence was found to link macroscopic structures of the Phosphoria Formation with mining characteristics. The most important controls on phosphate ore appear to be stratigraphic. However, the tectonic events did have an affect on the mobilization of elements and formation of different generations of minerals within the Phosphoria Formation (see Grauch et al., Chapter 8).
REFERENCES Allmendinger, R.W., 1979. Structural evolution of the Northern Blackfoot Mountains, southeastern Idaho. PhD dissertation, Cornell University, Ithaca, NY, 222 pp. Allmendinger, R.W., 1981. Structural geometry of Meade thrust plate in Northern Blackfoot Mountains, southeastern Idaho. Am. Assoc. Pet. Geol. Bull., 65(3): 509-525.
Strain distribution and structural evolution of the Meade plate
163
Allmendinger, R.W. and Jordan, T.E., 1981. Mesozoic evolution, hinterland of the Sevier orogenic belt. Geology, 9:308-313. Allmendinger, R.W., 1982. Analysis of microstructures in the Meade plate of the Idaho-Wyoming foreland thrust belt (U.S.A.). Tectonophysics, 85:221-251. Allmendinger, R.W., 1992. Fold and thrust tectonics of the western United States Cordillera, exclusive of the accreted terranes. In: B.C. Burchfiel, P. Lipman, and M.L. Zoback (eds.), The Cordilleran Orogen: Conterminous U.S. The Geology of North America, vol. 6-3, Geol. Soc. Amer., Boulder, CO, pp. 583-607. Allmendinger, R.W., Miller, D.M., and Jordan, T.E., 1984. Known and inferred Mesozoic deformation in the hinterland of the Sevier Belt, Northwest Utah. In: G.J. Kerns and R.L. Kerns, Jr. (eds.), Geology of Northwest Utah, Southern Idaho and Northeast Nevada. Utah Geological Association Publication, 13, pp. 21-34. Angevine, C.L. and Turcotte, D.L., 1983. Oil generation in overthrust belts. Am. Assoc. Pet. Geol. Bull., 67: 235-241. Armstrong, EC., 1969. Geologic map of the Soda Springs qua&angle, southeastern Idaho. US Geol. Survey, Misc. Geol. Inv. Map, 1-557, scale 1:48,000. Armstrong, EC. and Cressman, E.R., 1963. The Bannock thrust zone, southeast Idaho. US Geol. Survey, Prof. Paper, 374-J, 22 pp. Armstrong, EC. and Oriel, S.S., 1965. Tectonic development of the Idaho-Wyoming thrust belt. Am. Assoc. Petrol. Geol. Bulletin, 49:1847-1866. Armstrong, R.L., 1968. Sevier orogenic belt in Nevada and Utah. Geol. Soc. Am. Bull., 79: 429-458. Boyer, S.E., 1986. Styles of folding within thrust sheets: examples from the Appalachian and Rocky Mountains of the U.S.A. and Canada. J. Struct. Geol. 8(3/4): 325-339. Boyer, S.E. and Elliott, D., 1982. Thrust systems. Am. Assoc. Pet. Geol. Bull., 66(9): 1196-1230. Brewer, J., 1981. Thermal effects of faulting. Earth Planet. Sci. Lett., 56: 233-244. Burkhard, M., 1993. Calcite twins, their geometry, appearance, and significance as stress-strain markers and indicators of tectonic regime: a review. J. Struct. Geol., 15(3-5): 351-368. Claypool, G.E., Love, A.H., and McKee, E.K., 1978. Organic geochemistry, incipient metamorphism, and oil generation in black shale members of Phosphoria Formation, Western Interior United States. Am. Assoc. Pet. Geol. Bull., 62(1): 98-120. Coogan, J.C., 1992. Structural evolution of piggyback basins in the Wyoming-Idaho-Utah thrust belt. In: P.K. Link, M.A. Kuntz, and L.B. Platt (eds.), Regional Geology of Eastern Idaho and Western Wyoming. Geol. Soc. Am. Mem. 179, pp. 55-81. Coogan, J.C. and Yonkee, W.A., 1985. Salt detachments in the Jurassic Preuss redbeds within the Meade and Crawford thrust systems, Idaho and Wyoming. In: G.J. Kerns and R.L. Kerns (eds.), Orogenic Patterns and Stratigraphy of North-Central Utah and Southeastern Idaho. Utah Geol. Assoc. Pub., Salt Lake City, vol. 14, pp. 75-82. Craddock, J.P., 1992. Transpression during tectonic evolution of the Idaho-Wyoming fold-and-thrust belt. In: P.K. Link, M.A. Kuntz, and L.B. Platt (eds.), Regional Geology of Eastern Idaho and Western Wyoming. Geol. Soc. Amer. Mem., vol. 179, pp. 125-139. Craddock, J.P. and van der Pluijm, B.A., 1988. Kinematic analysis of an en echelon-continuous vein complex. J. Struct. Geol., 10: 445-452. Cressman, E.R., 1964. Geology of the Georgetown Canyon-Snowdrift Mountain area, southeastern Idaho. US Geol. Survey, Bull., 1153, 105 pp, map scales 1:48,000 and 1:24,000. Cressman, E.R. and Gulbrandsen, R.A., 1955. Geology of the Dry Valley Quadrangle, Idaho. US Geol. Survey Bull., 1015-I, pp. 257-270.
164
J. G. Evans
Crosby, G.W., 1968. Vertical movements and isostasy in Western Wyoming overthrust belt. Am. Assoc. Pet. Geol. Bull., 52(10): 2000-2015. Crosby, G.W., 1969. Radial movements in the western Wyoming salient of the Cordilleran overthrust belt. Geol. Soc. Am., 80(6): 1060-1078. Crosby, G.W., 1970. Radial movements in the western Wyoming salient of the Cordilleran overthrust belt: reply. Geol. Soc. Am., 81 ( 11): 3507-3512. Davis, D., Suppe, J. and Dahlen, EA., 1983. Mechanics of fold-and-thrust belts and accretionary wedges. J. Geophys. Res., 88(B2): 1153-1172. DeCelles, P.G., 1994. Late Cretaceous-Paleocene synorogenic sedimentation and kinematic history of the Sevier thrust belt, northeast Utah and southwest Wyoming. Geol. Soc. Am. Bull., 106(1): 32-56. DeCelles, P.G., Pile, H.T. and Coogan, J.C., 1993. Kinematic history of the Meade thrust based on provenance of the Bechler Conglomerate at Red Mountain, Idaho, Sevier thrust belt. Tectonics, 12(6): 1436-1450. Desborough, G.A., Poole, EG., Harris, A.G. and Daws, T., 1988. Thermal history and organic maturation of Paleozoic rocks, Pocatello area, southeastern Idaho. U.S. Geol. Survey, Open-File Report, 88-0289, 8 pp. Dixon, J.S., 1982. Regional structural synthesis, Wyoming salient of western overthrust belt. Am. Assoc. Pet. Geol., 49:1847-1866. Edman, J.D. and Surdam, R.C., 1984a. Diagenetic history of the Phosphoria, Tensleep, and Madison Formations, Tip Top Field, Wyoming. In: D.A. McDonald and R.C. Surdam (eds.), Clastic Diagenesis. Am. Assoc. Pet. Geol. Mem., vol. 37, pp. 317-345. Edman, J.D. and Surdam, R.C., 1984b. Influence of overthrusting on maturation of hydrocarbons in Phosphoria Formation, Wyoming-Idaho-Utah overthrust belt. Am. Assoc. Pet. Geol. Bull., 68(11): 1803-1817. Evans, J.E and Craddock, 1985. Deformation history and displacement transfer between the Crawford and Meade thrust systems, Idaho-Wyoming thrust belt. In: G.J. Kerns and R.L. Kerns (eds.), Orogenic patterns and Sratigraphy of North-Central Utah and Southeastern Idaho, Utah Geol. Assoc., Salt Lake City, vol. 14, pp. 83-95. Gulbrandsen, R.A., McLaughlin, K.E, Honkala, ES., Clabaugh, S.E., and Krauskopf, K.B., 1956. Geology of the Johnson Creek quadrangle, Caribou County, Idaho. US Geol. Survey Bull., 1042-A, 23 pp. Harris, A.G., Wardlaw, B.R., Rust, C.C. and Merrill, G.K., 1980. Maps for assessing thermal maturity (conodont color alteration index maps) in Ordovician through Triassic rocks in Nevada and Utah and adjacent parts of Idaho and California. US Geol. Survey, Map 1-1249, scale 1:2,500,000. Heller, P.L., Bowdler, H.P., Chambers, H.P., Coogan, J.C., Hagen, E.S., Shuster, M.W., and Winslow, N.S., 1986. Time of initial thrusting in the Sevier orogenic belt, Idaho-Wyoming and Utah. Geology, 14: 388-391. Herring, J.R., 1995. Permian phosphorites: a paradox of phosphogenesis. In: P.A. Scholle, T.M. Peryt, and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea, vol. 2, Sedimentary Basins and Economic Resources. Springer-Verlag, Berlin, pp. 292-312. Hills, E.S., 1963. Elements of Structural Geology. John Wiley & Sons, Inc., New York. 483 pp. Kellogg, K.S., 1992. Cretaceous thrusting and Neogene block rotation in the northern Portneuf Range region, southeastern Idaho. In: P.K. Link, M.A. Kuntz, and L.B. Platt (eds.), Regional Geology of Eastern Idaho and Western Wyoming. Geol. Soc. Amer. Mem., vol. 179, pp. 95-113. Kellogg, K.S., Oriel, S.S., Amerman, R.E., Link, P.K. and Hladky, ER., 1989. Geologic map of the Jeff Cabin Creek quadrangle, Bannock and Caribou Counties, Idaho. US Geol. Survey, Geologic Quadrangle Map, GS- 1669, map scale 1:24,000.
Strain distribution and structural evolution of the Meade plate
165
Kellogg, K.S., Rodgers, D.W., Hladky, ER., Kiessling, M.A. and Riesterer, J.W., 1999. The Putnam thrust plate, Idaho - dismemberment and tilting by Tertiary normal faults. In: S.S. Hughes and G.D. Thackray (eds.), Guidebook to the Geology of Eastern Idaho. Idaho Museum of Natural History, Pocatello, pp. 97-114. Lageson, D.R. and Schmitt, J.G., 1994. The Sevier orogenic belt of the Western United States: recent advances in understanding its structural and sedimentologic framework. In: M.V. Caputo, J.A. Peterson, and K.J. Franczyk (eds.), Mesozoic Systems of the Rocky Mountain Region, U.S.A. Denver, Colorado, Rocky Mountain Section, Soc. Econ. Pal. Min. (Soc. Sed. Geol.), pp. 27-64. Lamerson, ER., 1982. The Fossil Basin and its relationship to the Absaroka thrust system, Wyoming and Utah. In: R.B. Powers (ed.), Geologic Studies of the Cordilleran Thrust Belt. Rocky Mtn. Assoc. Geol., Denver, CO., pp. 279-340. Leeman, W.P., Oldow, J.S. and Hart, W.K., 1992. Lithosphere-scale thrusting in the western U.S. Cordillera as constrained by Sr and Nd isotopic transitions in Neogene volcanic rocks. Geology, 20(1): 63-66. Levy, M. and Christie-Blick, N., 1989. Pre-Mesozoic palinspastic reconstruction of the eastern Great Basin, western United States. Science, 245: 1454-1462. Mansfield, G.R., 1920. Geography, geology, and mineral resources of the Fort Hall Indian Reservation, Idaho. US Geol. Survey, Bull., vol. 713, 152 pp. Mansfield, G.R., 1927. Geography, geology, and mineral resources of part of southeastern Idaho. US Geol. Survey, Prof. Paper, 152, 453 pp., map scales 1:62,500, 1:125,000, and 1:250,000. Mansfield, G.R., 1929. Geography, geology and mineral resources of the Portneuf quadrangle, Idaho. US Geol. Survey, Bull., vol. 803, 110 pp., map scale 1:62,500. Mansfield, G.R., 1952. Geography, geology, and mineral resources of the Ammon and Paradise Valley quadrangles, Idaho. US Geol. Survey, Prof. Paper, vol. 238, 92 pp., map scale 1:62,500. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M., and Swanson, R.W., 1956. Summary description of Phosphoria, Park City, and Shedhorn Formations in the Western Phosphate Field. Am. Assoc. Pet. Geol. Bull., 40( 12): 2826-2863. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M., and Swanson, R.W., 1959. The Phosphoria, Park City, and Shedhorn Formations in the Western Phosphate Field. US Geol. Survey, Prof. Paper, 313-A, p. 47. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M. and Swanson, R.W., 1967. The Phosphoria, Park City, and Shedhorn Formations in Western Phosphate Field. In: L.A. Hale (ed.), Anatomy of the Western Phosphate Field, 15th Annual Field Conference, pp. 15-33. Mitra, G., Hull, J.M., Yonkee, W.A. and Protzman, G.M., 1988. Comparison of mesoscopic and microscopic deformational styles in the Idaho-Wyoming thrust belt and the Rocky Mountain forelands. In: C.J. Schmidt and W.J. Perry (eds.), Interaction of the Rocky Mountain Foreland and the Cordilleran Thrust Belt. Geol. Soc. Am. Mem., vol. 171, pp. 119-141. Mitra, G. and Yonkee, W.A., 1985. Relationship of spaced cleavage to folds and thrust in the Idaho-Utah-Wyoming thrust belt. J. Struct. Geol., 7:361-373. Monley, L.E, 1971. Petroleum potential of Idaho-Wyoming overthrust belt. In: L.H. Cram (ed.), Future Petroleum Provinces of the United States - their Geology and Potential. Am. Assoc. Pet. Geol. Mem. 15(1): 509-529. Montgomery, K.M. and Cheney, T.M., 1967. Geology of the Stewart Flat quadrangle, Caribou County, Idaho. US Geol. Survey Bull., 1217, 63 pp. Oriel, S.S. and Platt, L.B., 1980. Geologic map of the Preston 1~ by 2 ~ quadrangle, southeastern Idaho and Wyoming. US Geol. Survey, Misc. Inv. Series Map, I-1127, scale 1:250,000.
166
J. G. Evans
Paul, R.K. and Paul, R.A., 1986. Epilogue for the Permian in the western Cordillera - a retrospective view from the Triassic. Univ. Wyo., Contrib. Geol., 24(2): 243-252. Paul, R.K., Paul, R.A. and Anderson, A.L., 1985. Conodont biostratigraphy and depositional history of the Lower Triassic Dinwoody Formation in the Meade Plate, southeastern Idaho, in Orogenic patterns and stratigraphy of north-central Utah and southeastern Idaho. Utah Geol. Assoc. Pub., 14: 55-65. Pierce, K.L. and Morgan, L.A., 1992. The track of the Yellowstone hot spot: volcanism, faulting, and uplift. In: P.K. Link, M.A. Kuntz, and L.B. Platt (eds.), Regional Geology of Eastern Idaho and Western Wyoming. Geol. Soc. Am. Mem., vol. 179, pp. 1-53. Platt, L.B. and Oriel, S.S., 1981. Fault relations, Bannock Range, southeastern Idaho. US Geol. Survey, Prof. Paper, 1275, 73 pp. Protzman, G.M. and Mitra, G., 1990. Strain fabrics associated with the Meade thrust sheet: implications for cross-section balancing. J. Struct. Geol., 12: 403-417. Rich, J.L., 1934. Mechanics of low-angle overthrust faulting as illustrated by Cumberland thrust block, Virginia, Kentuck, and Tennessee. Am. Assoc. Pet. Geol. Bull., vol. 18, 1584-1596. Richards, R.W. and Mansfield, G.R., 1912. The Bannock overthrust- a major fault in southeastern Idaho and northwestern Utah. J. Geol., 20:681-707. Rioux, R.L., Hite, R.J., Dyni, J.R. and Gere, W.C., 1975. Geologic map of the Upper Valley quadrangle, Caribou County, Idaho. US Geol. Survey, Geol. Quadrangle, GQ-1194, scale 1:24,000. Rodgers, D.W. and Janecke, S.U., 1992. Tertiary paleogeologic maps of the western Idaho-Wyoming-Montana thrust belt. In: P.K. Link, M.A. Kuntz, and L.B. Platt (eds.), Regional Geology of Eastern Idaho and Western Wyoming. Geol. Soc. Amer. Mem., vol. 179, pp. 83-94. Royse, E, Jr., Warner, M.A. and Reese, D.L., 1975. Thrust belt structural geometry and related stratigraphic problems, Wyoming-ldaho-northern Utah. In: D.W. Bolyard (ed.), Deep Drilling Frontiers of the Central Rocky Mountains. Rocky Mountain Assoc. Geol., Denver, CO., pp. 41-54, 4 plates. Rubey, W.W. and Hubbert, M.K., 1959. Role of fluid pressure in mechanics of overthrust faulting; II Overthrust belt in geosynclinal area of western Wyoming in light of fluid-pressure hypothesis. Geol. Soc. Am. Bull., 70: 167-206. Schmitt, J,G. and Steidtman, J.R., 1990. Interior ramp-supported uplifts: implications for sediment provenance in foreland basins. Geol. Soc. Am., 102(4): 494-501. Suppe, J., 1983. Geometry and kinematics of fault-bend folding. Am. J. Sci., 283: 684-721. Turner, EJ., 1953. Nature and dynamic interpretation of deformation lamellae in calcite of marbles. Am. J. Sci., 251: 276-298. Turner, EJ. and Verhoogen, J., 1960. Igneous and Metamorphic Petrology. McGraw-Hill, New York, 694 pp. Wiltschko, D.V. and Dorr, J.A., Jr., 1983. Timing of deformation in overthrust belt and foreland of Idaho, Wyoming, and Utah. Am. Assoc. Pet. Geol. Bull., 67(8): 1304-1322. Wojtal, S., 1986. Deformation within foreland thrust sheets by populations of minor faults. J. Struct. Geol., 8(3/4): 341-360. Wojtal, S. and Mitra, G., 1986. Strain hardening and strain softening in fault zones from foreland thrusts. Geol. Soc. Am. Bull., 97(6): 674-687. Yonkee, W.A., 1983. Mineralogy and structural relationships of cleavage in the Twin Creek Formation within part of the Crawford thrust sheet in Wyoming and Idaho. MS thesis, Laramie, Univ. of Wyoming, 125 pp. Yonkee, W.A., Parry, W.T., Bruhn, R.L. and Cashman, P.H., 1989. Thermal models of thrust faulting: Constraints from fluid-inclusion observations, Willard thrust sheet, Idaho-Utah-Wyoming. Geol. Soc. Am. Bull., 101 (2): 304-313.
PART III.
G E O L O G I C A L AND G E O C H E M I C A L STUDIES IN S O U T H E A S T IDAHO
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
169
Chapter 7
THE EFFECTS OF WEATHERING ON THE M I N E R A L O G Y OF THE PHOSPHORIA FORMATION, SOUTHEAST IDAHO A.C. KNUDSEN and M.E. GUNTER
ABSTRACT The Permian Phosphoria Formation of the western United States is one of the largest phosphate deposits in the world. Despite the economic significance of this formation, its fine-grained nature has discouraged detailed mineralogical characterization and quantification studies. Recently, selenium and other potentially hazardous trace elements in mine waste rock have drawn increased attention and motivated extensive studies. Part of this effort has focused on a more detailed geological and mineralogical characterization of the rocks. This study uses powder X-ray diffraction (XRD) with Rietveld quantification software to quantify and characterize the mineralogy of samples collected from nine measured stratigraphic sections from the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation at four active phosphate mines, all near Soda Springs, Idaho. These measured sections include four pairs of more-weathered and less-weathered sections as well as a single, deep, least-weathered section. Mineralogical analyses of samples from these nine sections reveal how weathering has affected the mineral content, identifies the minerals that host the trace elements of environmental concern, and discusses the complex variations in the composition of the carbonate fluorapatite (CFA). Carbonate minerals decrease sharply in concentration with increased weathering, as do pyrite and sphalerite. Because these sulfide minerals have been linked to trace elements such as Se, their weathering likely contributes to the geochemical cycling of these trace elements. CFA occurs in phosphorites globally, with varying amounts of carbonate substituted for phosphate in the apatite structure. Analysis of the measured sections shows strong connections between the variable degree of carbonate substitution in the CFA structure and the host rock type. However, there is also evidence that weathering plays an important role in the composition and structure of the CFA, apparently breaking it down into nondiffracting CFA-like compounds. INTRODUCTION Mineralogical analysis of Phosphoria Formation rocks provides a better understanding of depositional and diagenetic processes and the environmental and economic concerns
170
A.C. Knudsen and M.E. Gunter
Fig. 7-1. Location map showing the four active phosphate mines in southeast Idaho. Less-weathered and more-weathered sections were collected at the Enoch Valley, Dry Valley, and Rasmussen Ridge mines. A deep least-weathered section was collected at the Enoch Valley mine from drill core (modified from Herring et al., 1999). surrounding trace-element residence. This study characterized and quantified the mineralogy of 475 samples from nine measured sections from the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation. Samples were collected from two measured sections at each of the four active phosphate mines in southeast Idaho (Fig. 7-1), measured
Phosphoria Formation: effects of weathering on mineralogy
171
sections A and B (Enoch Valley mine), C and D (Dry Valley mine), and F and E (Rasmussen Ridge mine) are more-weathered and less-weathered sections, respectively, from each mine. Measured sections G and H (Smoky Canyon mine) are both altered due to a combination of surficial weathering and subsurface alteration. Finally, measured section J (Enoch Valley mine) is a deep-core and considered the least-weathered section studied. The quantified mineralogy of these sections, along with estimates of the maximum crystallinity and carbonate substitution in carbonate fluorapatite (CFA), are reported in Knudsen et al. (2000, 2001, 2002a,b) and Knudsen (2002). The location, stratigraphy, sampling, and chemistry of these measured sections are discussed more fully in Herring and Grauch (Chapter 12), Tysdal et al. (1999, 2000a-c) and Grauch et al. (2001). Using X-ray diffraction (XRD) analysis and Rietveld refinement, we quantified and characterized the samples for the following minerals: CFA, quartz, albite, orthoclase, buddingtonite (NHaA1Si308), muscovite, illite, kaolinite, montmorillonite, dolomite, calcite, pyrite, sphalerite, and gypsum. We also analyzed the extent of carbonate substitution for phosphate in CFA using Rietveld unit-cell refinements and finally estimated the nondiffracting portion of the samples by comparing the mineralogical and chemical data for the same samples. Comparison of samples from more-weathered, less-weathered, and least-weathered strata illustrates the effects of weathering on the mineralogy of the Phosphoria Formation.
CARBONATE FLUORAPATITE CFA is the primary phosphate and ore mineral in the Phosphoria Formation and nearly all other marine phosphorites. CFA is built on the general structure of common fluorapatite with extensive substitutions. The most significant substitution in marine CFAs is that of CO2- for PO 3- and to a lesser extent, of SO 2- for PO 3-. In order to balance the charge associated with these substitutions, there are a number of other chemical modifications made relative to pure fluorapatite including: vacancies in the Ca 2+ site, substitution of a monovalent cation such as Na + for Ca 2+, and additional F- entering the structure, possibly coupled with the CO 2- to form a CO3F3- psuedotetrahedron. A more complete discussion of CFA and the substitutions within its structure can be found in Knudsen and Gunter (2002) and Pan and Fleet (2002) and references therein. Gruner and McConnell (1937) first proposed that carbonate substitutes for phosphate in the apatite structure, and Altschuler et al. (1952) showed a contraction of the apatite a-cell parameter with the presence of carbonate, confirming this theory. While other substitutions are present, the crystallographic effects of substituting the trigonal planar CO 2group for the tetrahedral PO 3- are the most pronounced and, therefore, can be indirectly measured using measurements of the CFA crystallographic axes. The X-ray work of Altschuler et al. (1952) laid the groundwork for current methods of measuring the degree of carbonate substitution for phosphate in CFA. McClellan and Lehr (1969) observed a substantial variation in the a-cell parameter of CFAs with varying chemical composition, with only minor changes in the c-cell parameter. Based on this relationship, they formulated an equation to estimate the carbonate content in CFA from the a-cell parameter. While this
172
A.C. Knudsen and M.E. Gunter
equation is useful, it requires the addition of an intemal standard to calibrate the a-cell parameter. Gulbrandsen (1970), using CFA from the Phosphoria Formation, presented a formula based on the separation of the (004) and (410) peaks. Because the (004) peak location is dependent only on the c-cell parameter, it will not vary as a function of carbonate substitution, while the (410), dependent on the a- and b-cell parameters (equivalent in the hexagonal structure of apatite), will vary. Thus, the separation of these two peaks changes as a function of CO 2- in CFA. Schuffert et al. (1990) further refined this method and it is their formula that we use here.
METHODS
Sampling and sample preparation Samples for this study came from nine measured sections representing each of the four active phosphate mines in southeastern Idaho (Fig. 7-1). US Geological Survey scientists collected samples from pairs of sections from each of the four mines, as well as a core from the Enoch Valley mine. The stratigraphy of each of these sections has been described (Tysdal et al., 1999, 2000a-c; Grauch et al., 2001). Channel samples were collected for each section, which consist of a series of samples of varying stratigraphic thickness that represent the entire section of Meade Peak rocks. Measured section J, the core from the Enoch Valley mine, was sampled for both channel samples and individual samples. Samples were disaggregated in a mechanical jaw crusher and a representative split was ground in a ceramic plate grinder to < 100mesh (<0.15 mm). Representative splits of the latter material were provided for mineral analysis. All splits were obtained with a riffle splitter to insure homogeneity among samples.
XRD analysis and RietveM refinement Powder XRD analyses were performed with a scan from 2-62 ~ using Cu Ka radiation on a Siemens D5000 diffractometer operating at 40 kV and 30mA. These scans revealed the major phases in the samples, but low peak-to-background ratios prevented accurate identification of minor phases (generally those less than 1%) depending on the crystallinity and X-ray scattering efficiency. The patterns were subsequently analyzed using the Siroquant program (Taylor, 1991), which is based on the Rietveld method to quantify the mineralogical content. First, every phase in a sample must be identified. Then the program calculates an XRD pattern based on the known crystal structure of each mineral to match the collected pattern to determine the quantities of each phase. The Rietveld method refines the calculated pattern for each phase to match the observed pattern, correcting for variable peak shape, preferred orientation, and shifts in cell
Phosphoria Formation: effects of weathering on mineralogy
173
parameters. The quantity of each phase is reported as weight percent along with an estimated experimental error.
C032- substitution in CFA Measurements of the degree of CO2- in the CFA structure were made using the empirical relationship determined by Schuffert et al. (1990): y = 10.643x 2 - 52.512x + 56.986 where y is the weight percent of CO2- present in the CFA and x is the A20(004)_(410 ). Rather than measure the separation of the (004) and (410) CFA peaks, we used Rietveld refinement to determine the unit-cell parameters. The measured cell parameters were then used to calculate the separation (A~ of the diffraction peaks, eliminating the potential for human error in determining peak location as well as circumventing the problems of peak overlap between CFA and other minerals.
Statistical analyses Principal component factor analysis, which determines unobservable relationships in the data set, was done using the SAS system. We used orthogonal factor analyses based on the Pearson correlation matrix. The correlation matrix is based on the observed variables "p." Principal component factor analysis then manipulates the correlation matrix to form a new set of unobservable "m" variables, which are uncorrelated to one another. Mathematically, the new values, being uncorrelated, can be described as mutually perpendicular vectors, the directions and lengths of which are called eigenvectors and eigenvalues, respectively. The eigenvalue of a factor represents the total significance of that factor in determining the total variance, where the sum of the eigenvalues is equal to the total variance or the number of original variables. Factors with small eigenvalues can be discarded, as they have relatively little control on the total variance of the data set, leaving fewer factors than original variables, thus simplifying the original data set. Each factor is controlled by some combination of the original variables. The variables are loaded, either positively or negatively, in each factor, with larger absolute loadings, approaching [1 or - 1[, showing a greater dependence of that factor on a given variable. Principal component factor analysis returns a "noise" variable, the total communality, along with the factors. The total communality for each variable describes the variability for each variable described by the factors, the difference between this value and 1 is the amount that is not described by the given factor analysis. For a more complete discussion of principal component and factor analysis, refer to Davis (1986) or Harman (1976).
174
A. C. Knudsen and M.E. Gunter
RESULTS Bulk mineralogy
Weathering in the Phosphoria Formation is important geologically, economically, and environmentally. In order to gain a broader picture of the Meade Peak Phosphatic Shale, and the role that weathering plays in its evolution, we grouped data for the more-weathered sections as well as the less-weathered sections. By grouping the sections, we can better evaluate the role that weathering plays on the samples, comparing the least-, less-, and more-weathered sections. For these comparisons, only the channel samples from section J are considered, and not the individual samples. The quantified mineralogy from these three groups of data are shown in side-by-side box plots (Fig. 7-2A-I). Pyrite values are shown only for samples from the Enoch Valley and Smoky Canyon mines (Fig. 7-2I) because the Dry Valley mine samples show notably higher pyrite concentrations. Consequently, Dry Valley pyrite data skew the less-weathered and more-weathered sample values higher than the least-weathered core J samples from the Enoch Valley mine. While individual strata cannot be traced between the measured sections, ore- and waste-producing intervals are recognized between the mines and can be used to group strata and compare them between mine sections. The Meade Peak Phosphatic Shale Member is divided into the upper waste (UW), upper ore (UO), middle waste (MW), lower ore (LO), and lower waste (LW). Because the stratigraphic thickness varies among samples, the values are weighted. Weighting is accomplished by rounding the sample interval to the nearest half foot (15.2 cm). This returns a weighting factor by which each sample is multiplied. Thus, a 5-foot (152 cm) channel sample is considered as 10 0.5-foot samples rather than a single 5-foot sample.
Least-weathered
Less-weathered % CFA 20 40 60 8 0 1 0 0 0
(A) 0 20 40 60 801000
More-weathered 20 40 60 8010s
RC UW
D
9
o
o
UO MW
i
~o o o
LO LW
i ~
GD
Fig. 7-2(A).
I
o
o
t
Phosphoria Formation: effects of weathering on mineralogy (B)
Least-weathered
Less-weathered More-weathered % Quartz 0 20 40 60 8 0 1 0 0 0 20 40 60 80100 0 20 40 60 80100 T
RC UW
o
o
~
o
io
omo
UO MW !
i
LW
1
t
~
nn
O
t
"l
t
GD
z
Total Feldspars (%)
(c)
0 20 40 60 8 0 1 0 0 0
20 40 60 80100 0 20 40 60 80100 '
RC
~
~ o
LO
I
'
!
I
t
oD
UW
m
i~aa
IE]-~ -
UO
~;-}---~
MW
~ - - - ~ ~176
LO
o
~o
LW
oo
t
GD
(D)
0 10 20 30 40 50
Buddingtonite (%) 0 10 20 30 40 50 0 10 20 30 40 50
t
RC UW
o
UO MW
E]3~
-s
LO
t
LW GD
I
Fig. 7 - 2 ( B - D ) .
,
175
176
A.C. Knudsen and M.E. Gunter
(E)
Least-weathered
Less-weathered More-weathered Muscovite + Illite (%) 0 10 20 30 40 50 0 10 20 30 40 50 0 10 20 30 40 50
RC
t
UW UO MW
~-~
---I
o
o
o
LO ~]~ ,W
-tZ
:
GD
Dolomite (%) 10 20 30 40 50 0
(F) 0 10 20 30 40 50
0
10 20 30 40 50
RC UW 17]
'
"
UO MW
.
: .
.
.
.
io=i
o
LO LW
i9
,~.1 ~ ='
J
GD
~]
(G)
0 102030405060
Calcite (%) 0 102030405060 J
RC
1
UW
2
UO
i
i
i
i
t
3 . r_
MW ::)4 LO
5
LW
6
GD
7
"-R
Fig. 7-2(E-G).
I
|
~
oooo
omoo
o
0 10203040506(
Phosphoria Formation." effects of weathering on mineralogy (H)
Less-weathered
Least-weathered 0
RC
1
UW
2
UO
0.5
1.5
More-weathered
Sphalerite (%) 0.5 1.5 1.5 0 1.5 0
~~
177
0.5
1.5
1.,
.-io
3
I~~
~
,i
MW ~ 4 LO
5 I~
LW
6
GD
7
~
~ -9I o
~
t
01
23456
78
01
Pyrite (%) 23456
78
01
23456
78
RC UW UO
~~
MW LO LW GD
Fig. 7-2. Schematic box plots showing the concentrations of" (A) CFA; (B) quartz; (C) feldspars (albite + orthoclase + buddingtonite); (D) buddingtonite; (E) muscovite + illite; (F) dolomite; (G) calcite; (H) sphalerite, in the "least-weathered," "less-weathered," and "more-weathered" sections. The concentrations of pyrite (I) are shown only for samples from the Enoch Valley and Smoky Canyon mines. The stratigraphic units shown on the vertical axis are the Rex Chert (RC), the Upper Waste (UW), the Upper Ore (UO), the Middle Waste (MW), the Lower Ore (LO), the Lower Waste (LW), and the Grandeur Dolostone (GD). The box includes the middle half of the data from the 25th percentile to the 75th percentile, the lower and upper quartiles are separated by the median. The mean is shown as a cross. The inner quartile range (IQR) is the distance between the 25th and 75th percentiles. The lower and upper fences are located at 1.5 x IQR below the 25th percentile and at 1.5 x IQR above the 75th percentile, respectively. The bars extend from the box to the smallest value within the lower fence and the highest value within the upper fence. Outliers are shown as small boxes.
Nondiffracting component The quantified mineralogy determined by the Rietveld method does not, without manipulation, account for the portion of the sample that does not diffract X-rays. One method to determine the amount of nondiffracting material is to spike the samples with an internal standard such as corundum, and then quantify the spiked sample. Comparison of the reported
178
A.C. Knudsen and M.E. Gunter
weight percent of the spike phase with the known amount allows for precise measurement of the nondiffracting component. This method is, however, very time consuming and, therefore, impractical for analysis of the large number of samples in this project. Instead, we developed a method to estimate the minimum nondiffracting component based on a comparison of the mineralogical and chemical data (Herring and Grauch, Chapter 12) for the same samples. The method developed is based on the principle of limiting reagents. Assuming ideal stoichiometry in the minerals, we calculated the chemistry of the crystalline portion of the samples based on the mineralogical data. We then compared this calculated chemical data with the analytical chemical data. If a sample contained no nondiffracting fraction, these values would be near equal, with slight differences due to deviations from ideal mineral stoichiometry. Likewise, differences between these data sets can be used to estimate the maximum crystallinity of a given sample. If a chemical component is overstated in the mineralogy-based calculated chemical data, it can be implied that there is a nondiffracting component. For example, if the CFA content translates to 40% P205 whereas the chemical analysis measures only 30%, then it is clear that another unaccounted for phase must be present. While there is 40% P205 in the diffracting component, knowing that there is only 30% P205 in the sample as a whole, we can infer that the sample is, at maximum, 75% crystalline, if we assume that there is no P205 in the nondiffracting component. We compared major chemical components from mineral groups which account for at least 15% of a given sample (i.e. if silicate minerals sum to less than 15%, we do not consider Si in our calculations because the experimental errors in both the XRD and chemical analytical methods produce unrealistic results). As a secondary method, we took the smallest maximum crystallinity estimates from the above method for each sample and multiplied that proportion by the calculated chemical data. This sets the chemical component, or components, which returned the maximum crystallinity value equal to the analytical data for that component. Again referring to the hypothesized sample containing 40% P205 based on the calculated chemistry, by multiplying the maximum crystallinity estimate of 75%, it is set equal to the analytical chemical data. In turn, this method then returns values for the other chemical components that are smaller than the analytical data. We then sum the differences between the analytical and the adjusted calculated values, and then add the weight percent of components that we know were not accounted for, particularly organic C. This new sum returns a second estimate for the maximum crystallinity. Because these two methods are clearly mathematically linked, they cannot be considered independent checks. Indeed, the two methods returned very similar numbers, the smaller maximum crystallinity (i.e. the larger nondiffracting component) estimate was then retained (Fig. 7-3). One notable exception to the similarity between the two methods is important though: samples with a very high organic C content produced considerably lower maximum crystallinity estimates based on the second method. This process can only be considered at best a semi-quantitative estimate because it compares two data sets, each with specific limitations, along with assumptions of ideal stoichiometry. Furthermore, it must be stressed that these estimates must be considered the maximum crystallinity. These methods are, however, based on one important and likely incorrect assumption, that at least one major chemical component exists only in the crystalline portion. However, if all of the major mineral forming chemical components are
Phosphoria Formation: effects of weathering on mineralogy
179
Least-weathered
Less-weathered More-weathered % Maximum crystallinity 50 60 70 80 90100 50 60 70 80 9010050 60 70 80 90100 RC UW
t
:
H]--~~
o
:
i
oolo
' if,if
o ~D----~
MW
LW
!
!
UO
LO
i
'
i
:
i
It11--
I"
o
i
t:
GD
Fig. 7-3. Schematic box plots showing the maximum crystallinity of samples from the leastweathered, less-weathered, and more-weathered sections, see Fig. 7-2 for a description of the box plots and the vertical axis. Maximum crystallinity is calculated by comparing quantified mineralogical and chemical data. TABLE 7-I Comparison of measured and calculated crystallinity of six samples from the least-weathered measured section J (see text: nondiffracting component) Sample
WpsJ054 WpsJ086 WpsJ128 WpsJ149 WpsJ 151 WpsJ162 WpsJ 164 WpsJ 181 WpsJ 184
Measured crystallinity (%)
Calculated maximum crystallinity (%)
95 76 86 56 68 62 60 62 59
87 98 95 72 64 87 80 82 70
present in both crystalline and noncrystalline phases, the reported maximum crystallinity will be larger than the actual crystalline proportion. This is seen in six samples from the least-weathered section J that were spiked with corundum to measure the nondiffracting portion and then compared to the maximum crystallinity estimates (Table 7-I, Fig. 7-4) for the same samples. The results show two samples in which the spiking method reports a smaller nondiffracting component, although both of these differences are close enough that they can be considered to be within error. The other samples show that our calculated method tends to significantly underestimate the portion of these samples that is
A.C. Knudsen and M.E. Gunter
180 50 o~
v
40
.~_ 30 tO t-
.~
20
E N 9
lO
w
o
,
o
,
,
,
10 20 30 40 Measured nondiffracting (%)
,
50
Fig. 7-4. Relationship of the measured crystallinity of nine samples compared to the calculated maximum crystallinity based on a comparison of mineralogical and chemical analyses. A one-to-one line is included for reference. Samples plotting to the right of the line show that the calculated method underestimates the amount of the noncrystalline proportion, and samples to the left of the line overestimate the noncrystalline proportion. nondiffracting. While our calculated values should not be used to draw conclusions about the crystallinity of a given sample, they can show trends within and among sections that pertain to stratigraphic and weathering variations.
Carbonate substitution in fluorapatite Carbonate substitution in CFA is generally between 2 and 3% by weight, in accordance with previous measurements from Phosphoria rocks (Gulbrandsen, 1970). While the carbonate ion in the apatite structure is susceptible to weathering (Lucas et al., 1979), weathering does not produce a discernable trend in our data set (Fig. 7-5). A pattern of variability can be seen in these samples that relates to the stratigraphic position of the samples where the degree of carbonate substitution tends to be higher in samples from the ore zones than the waste zones.
DISCUSSION
Weathering The most marked trend in bulk mineralogy is the disappearance of dolomite and calcite with increasing weathering (Fig. 7-2F and G). Dolomite is abundant in the least-weathered section, whereas in the less-weathered sections it decreases from the bottom, presumably
Phosphoria Formation: effects of weathering on mineralogy Least-weathered 0
1 2
3
4
5
Less-weathered % 0032- in CFA 0 1 2 3 4 5
181
More-weathered 0
1 2
3
4
5
RC
~o
UW UO
4"
MW
, ~]3-,
o
o
o
o
o I-.-~
m
o
o
LO LW
o o I-~.-I
t
t
GD
Fig. 7-5. Schematic box plots showing the degree of CO2- substitution for PO43- in CFA for samples from the least-weathered, less-weathered, and more-weathered sections, see Fig. 7-2 for a description of the box plots and the vertical axis. The degree of CO2- substitution is calculated using unit-cell parameters refined by Rietveld analysis and the equation by Schuffert et al. (1990). less exposed part of the sections, to the upper parts, presumably more exposed. In the moreweathered sections, dolomite is only a trace component. The sulfides pyrite and sphalerite follow a pattern similar to that of the carbonate minerals, with concentrations generally decreasing with increased weathering (Fig. 7-2H and I). The effect of weathering on the sulfide minerals has environmental implications because they host Se (Grauch et al., Chapter 8). As weathering oxidizes these sulfides, they break down and Se is mobilized. The susceptibility of the sulfides to weathering suggests that the breakdown of these minerals may explain much of the stratigraphic variability in Se concentrations. An important economic effect of weathering is the upgrading of P205 in potential oregrade rocks (greater than 20% CFA) as the gangue carbonates are removed (Fig. 7-6). It is notable that, while the P205 concentrations increase with weathering, the CFA concentrations do not (Fig. 7-6). As a result, the PEOs/CFA ratio (Fig. 7-6) in these high-CFA samples is higher with increased weathering. This relationship cannot be explained by higher carbonate contents in CFA, which do not appear to be related to weathering (Fig. 7-5). Instead, it is likely that P205 is present in a phase other than CFA. This trend is magnified when the CFA concentration is adjusted by multiplying it by the maximum crystallinity for the PEOs/CFAadj ratio (Fig. 7-6). The unadjusted PEOs/CFA ratio is generally near 0.4, which is the approximate value of the PEOs/apatite ratio in pure fluorapatite, and ranges from approximately 0.30 to 0.42 depending on the degree of substitutions (carbonate and sulfate for phosphate) in CFA. The values of the adjusted PaOs/CFAadj ratio are notably higher, nearer to 0.5. This upward shift suggests that P205 is being removed from CFA during weathering and redeposited as part of the X-ray amorphous component. The proximity of the unadjusted ratio to the stoichiometric PEOs/CFA ratio, suggests that nondiffracting P205 must be a significant component in much of the Meade Peak Member of the Phosphoria Formation. Assuming the P2Os/CFA ratio in the crystalline
182
A.C. Knudsen and M.E. Gunter
0
P205 10 20 30 4020
CFA P205/C FA 40 60 80 1000.1 0.3 0.5 0.7
Moreweathered
: F--~I::I i
~
~1 I .
Lessweathered
~ I*
~
~ F-I :~['
Leastweathered
~
H
"
i
~ .
.
~
~
.
.
!~
i
P205/C FAadj 0.20.40.60.81.01.2
~
i
!
~ * '
~
"~*
~ *"*
!
~
~
~176
:
Fig. 7-6. Schematic box plots for the concentrations of P205, CFA, P205/CFA, and P205/CFAadj from samples with 20% or more CFA in each of the least-weathered, less-weathered, and moreweathered sections. The CFA concentration is adjusted by multiplying the CFA concentration by the maximum crystallinity. portion of the samples is near the stoichiometric value of 0.4, in order for the unadjusted PzO5/CFA ratio (Fig. 7-6) to also be near the 0.4 value, the concentration of P205 in the nondiffracting component must be approximately the same as the P205 concentration in the crystalline portion. This suggests that in many samples, particularly in more-weathered strata, the concentration of noncrystalline PzOs-bearing phases may be significant.
C a r b o n a t e substitution in CFA
In the study of the mineralogy of phosphorites, the issue of carbonate substitution in apatite has long been contentious. Globally, this substitution ranges from 2 to 3% in the Phosphoria Formation to as high as 8-10% in Moroccan phosphorites (Table 7-II). However, the cause of this variability has not been determined. Gulbrandsen (1970), in analyzing samples from the Phosphoria Formation, noted a link between lithology and the amount of carbonate in CFA. Gulbrandsen found that samples from the western portions of the Phosphoria Formation, with more mudstone, shale, and siltstone mixed with phosphorite, have lower amounts of carbonate in the CFA, while samples from Wyoming in the eastern portion of the Phosphoria Formation, where phosphorite is more commonly found with limestone and dolostone, show higher amounts of carbonate present in the CFA structure. He suggested that this relationship is due to variability in the original depositional environment affecting the CFA chemistry. Others have pointed to differences in weathering, diagenesis, and metamorphism, which may affect the stability of the substituted carbonate ion, as well as other substitutions, as the primary cause for the chemical variability of CFAs (McArthur 1978, 1985; McClellan, 1980). Our study appears to support the work of Gulbrandsen (1970) on the Phosphoria Formation in which he analyzed 368 samples. While Gulbrandsen's data are useful, there
183
Phosphoria Formation: effects of weathering on mineralogy TABLE 7-II Carbonate content in CFA of selected phosphorites Location
Deposit
CO 2 content
Morocco Spain North and South Carolina Israel Turkey Western United States
Sidi Daoui Estremadura Pungo River Formation Mishash Formation Mazidagi Phosphoria
8.1-8.9% (McArthur, 1978) 6% (Lucas et al., 1978) 5-6% (McClellan and Lehr, 1969) 4% (Nathan, 1984) 4% (Lucas et al., 1980) 2% (Gulbrandsen, 1970)
were problems that need to be addressed. Because the samples analyzed by Gulbrandsen came from considerable distances apart, it is also possible that the strata from which the samples were collected underwent different degrees of weathering and diagenesis. This possibility can be addressed by looking at samples from the same location and comparing the CO2substitution in apatite within a single stratigraphic section rather than over a wide region. The relationships between CO2- content in CFA and lithology is shown statistically with a correlation matrix and factor analysis for four variables (Tables 7-III and 7-IV): silicates (quartz + muscovite + illite + albite + orthoclase + buddingtonite), carbonates (dolomite + calcite), CFA, and CO~- substitution in CFA; the three mineral variables proxy for mudstone and chert, dolostone and limestone, and phosphorite, respectively. Sample data are weighted to account for variability in stratigraphic thickness. This analysis only includes the section J samples (both channel and individual samples) because it is these samples that most closely resemble the original unweathered Phosphoria Formation. Factor analysis shows that 80% of the total variance for four variables can be explained by two factors, thus the two remaining factors are not reported as they are considered to be of minor importance (Table 7-IV). The first factor accounts for 49% of the variance and shows strong positive loadings for the CO2- substitution in CFA and CFA concentration. It also shows a very strong negative score for silicate fraction concentration. Finally, for the first factor, the carbonate minerals have little effect, showing a very weak negative score (Table 7-IV). In the second factor, which explains 31% of the variance, the CO2- substitution in CFA has a weaker positive score than in Factor 1. For this factor, however, the concentration of carbonates has a very strong positive score compared to moderately negative weightings for both CFA and total silicates (Table 7-IV). This factor analysis indicates that CO2- substitution in CFA is less extensive with higher concentrations of silicate minerals and greater with higher concentrations of apatite and, to a lesser degree, carbonate minerals. It is also important to note that the total communality of CO2- in CFA is 0.76, meaning that a significant portion (0.24) could not be addressed by this technique. Other explanations for the additional variability could be chemical and physical factors related to both depositional and post-depositional processes not considered here. Based on this factor analysis, it appears that host-rock types in which the CFA was formed affect the degree of CO2- substitution in apatite as suggested by Gulbrandsen
184
A.C. Knudsen and M.E. Gunter
TABLE 7-III Correlation coefficient (R) based on 128 samples and test statistic (t) for variables CO 2- (CO 2- substitution in CFA), silicates (quartz + muscovite + illite + albite + orthoclase + buddingtonite), CFA, and carbonates (dolomite + calcite), section J individual and channel samples CO 2CFA R t Carbonates R t Silicates R t
CFA
Carbonates
0.27 <0.0001 0.09 0.3126
-0.56 <0.0001
-0.34 <0.0001
-0.38 <0.0001
-0.55 <0.0001
TABLE 7-1V Unrotated orthogonal factor scores for correlation matrix in Table 7-11I; only factors with eigenvalues (EV) greater than 1 are shown; TC is total communality Factor Pattern EV CO32CFA Carbonates Silicates
Factor 1 1.98
Factor 2
0.57 0.90 -0.02 -0.91
0.27 -0.39 0.98 -0.25
1.24
TC 0.76 0.92 0.95 1
(1970). Weathering of CFA has been shown to remove the CO 2- ionic complex from the CFA structure (Lucas et al., 1979), although this does not seem to be a major factor in the Phosphoria Formation CFA where the level o f CO 2- substitution in CFA does not change substantially from the least-weathered to the more-weathered samples (Fig. 7-5). However, it appears likely that both the lithologically controlled and the alterationcontrolled hypotheses on the CO 2- in the CFA are valid. While the CO 2- concentration in the CFA structure does not appear to decline with weathering in this study, it likely destabilizes the CFA causing it to break down, lose CO 2-, and form a nondiffracting "apatite-like" phase. The instability o f CO2--substituted apatite explains the high concentrations o f nondiffracting P-bearing phases, which increase with weathering.
Phosphoria Formation: effects of weathering on mineralogy
185
This leaves the CO2- content of the crystalline CFA roughly the same in strata that have undergone different degrees of weathering.
CONCLUSION Previous work, that used only chemical identification techniques coupled with unquantiffed XRD data that do not show the nondiffracting P-bearing phases, may have misidentiffed those nondiffracting phases as CFA. It is likely that the CFA crystals in other phosphate deposits may vary in resistance to weathering and removal of the CO2- ion, allowing some to remain crystalline while losing the CO2- ion, whereas others, such as the Phosphoria Formation, simply break down when the carbonate is removed. Variability in the resistance to weathering of the non-CO 2- portion of the apatite may likely be related to the extent of other substitutions in the structure. The CFA in the Phosphoria Formation is rich with SO 2-, reportedly higher than 1% in some samples (Personal communications from G. Desborough, USGS, 2002). This and other substitutions may further destabilize the CFA structure, leading to its breakdown with the removal of CO2-. Future studies making use of chemical methods such as the electron microprobe, couple with quantitative XRD, unit cell refinements, and statistical analyses on CFA in other phosphorites will contribute to our understanding of the significance of CO2- in CFA.
ACKNOWLEDGMENTS We thank the US Geological Survey for their financial and scientific support of this project, especially Phil Moyle and Jim Herring. Special thanks to James R. Hein, Richard I. Grauch, and George Desborough for their thoughtful reviews of this chapter. We would also like to thank Tom Williams at the University of Idaho for his help with our XRD work. Finally, we would like to thank the phosphate mines in southeastern Idaho for access to the mines for sampling.
REFERENCES Altschuler, Z.S., Cusney, E.A. and Barlow, I.H., 1952. X-ray evidence of the nature of carbonate-apatite. Geol. Soc. Am. Bull., 63:1230-1231. Davis, J.C., 1986. Statistics and Data Analysis in Geology (2nd edn.). Wiley, New York, 646 pp. Grauch, R.I., Tysdal R.G., Johnson, E.A., Herring, J.R. and Desborough, G.A., 2001. Stratigraphic section and selected semiquantitative chemistry, Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-20-E. Gruner, J.W. and McConnell, D., 1937. The problem of the carbonate apatites. The structure of francolite. Z. Kristallogr., Z27:208-215.
186
A.C. Knudsen and M.E. Gunter
Gulbrandsen, R.A., 1970. Relation of carbon dioxide content of apatite of the Phosphoria Formation to the regional facies. US Geol. Survey, Professional Paper, vol. 700-, pp. 9-13. Harman, H.H., 1976. Modem Factor Analysis (3rd edn.). University of Chicago Press, Chicago, IL, 487 pp. Herring, J.R., Desborough, G.A., Wilson, S.A., Tysdal, R.G., Grauch, R.I., and Gunter, M.E., 1999. Chemical composition of weathered and unweathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation. Measured sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-147-A, 24 pp. Knudsen, A.C., 2002. A mineralogical investigation of the Permian Phosphoria Formation, southeastern Idaho: characterization, environmental concerns, and weathering. Unpublished Ph.D. dissertation, University Idaho, 205 pp. Knudsen, A.C. and Gunter, M.E., 2002. Sedimentary phosphates - an example: Phosphoria Formation, southeastern Idaho, USA. In: M.J. Kohn, J. Rakovan and J.M. Hughes (eds.), Phosphates- Geochemical, Geobiological, and Materials Importance. Reviews in Mineralogy and Geochemistry, vol. 48. Mineralogical Society of America, Washington, DC, pp. 363-390. Knudsen, A.C., Gunter, M.E. and Herring, J.R., 2000. Preliminary mineralogical characterization of weathered and less-weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation: measured sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 00-116, 74 pp. Knudsen, A.C., Gunter, M.E. and Herring, J.R., 2001. Preliminary mineralogical characterization of weathered and less-weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation: measured sections C and D, Dry Valley, Caribou County, Idaho. US Geological Survey, Open File Report, 01-072, 72 pp. Knudsen, A.C., Gunter, M.E., Herring, J.R. and Grauch, R.I., 2002a. Mineralogical characterization of strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation channel-composite and individual rock samples of measured section J and their relationship to measured sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 02-125, 46 pp. Knudsen, A.C., Gunter, M.E., Herring, J.R. and Grauch, R.I., 2002b. Mineralogical characterization of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation: measured sections E and F, Rasmussen Ridge, and Measured Sections G and H, Sage Creek area of the Webster Range, Caribou County, Idaho. US Geological Survey, Open File Report, 02-392, 44 pp. Lucas, J., Prrvrt, L., Lamboy, M., 1978. Les Phosphorites de la marge nord de l'Espagne. Chimie, Minrralogie, Grnrse. Oceanol. Acta, 1" 55-72. Lucas, J., Pr~vrt, L. and El. Mountassir, M., 1979. Les phosphorites rubrfires de Sidi Daoui. Transformation mrtrorique locale du gisement de phosphate des Ouled Abdoun (Maroc). Sci. Geol. Bull., 32:21-37 (in French). Lucas, J., Prrvrt, L., Ataman, G., Giindogdu, N., 1980. Mineralogical and geochemical studies of phosphatic formations in southeastern Turkey (Mazidagl-Mardin). In: Y.K. Bentor (ed.), Marine Phosphorites. SEPM Spec. Pub., 29, Tulsa, OK, pp. 149-152. McArthur, J.M., 1978. Systematic variations in the contents of Na, Sr, CO3, and SO4 in marine carbonate fluorapatite and their relation to weathering. Chem. Geol., 21:89-112. McArthur J.M., 1985. Francolite geochemistry- compositional controls during formation, diagenesis, metamorphism and weathering. Geochim. Cosmochim. Acta, 49: 23-35. McClellan, G.H., 1980. Mineralogy of carbonate fluorapatites. Geol. Soc. London, 137: 675-681. McClellan, G.H. and Lehr, J.R., 1969. Crystal chemical investigation of natural apatites. Am. Min., 54: 1379-1391.
Phosphoria Formation: effects of weathering on mineralogy
187
Nathan,Y., 1984. The mineralogy and geochemistry of phosphorites. In: J.O. Nriagu and EB. Moore (eds.), Phosphate minerals. Springer-Verlag, Berlin, pp. 275-290. Pan, Y. and Fleet, M., 2002. Compositions of the apatite-group minerals: substitution mechanisms and controlling factors. In: M.J. Kohn, J. Rakovan and J.M. Hughes (eds.), PhosphatesGeochemical, Geobiological, and Materials Importance. Reviews in Mineralogy and Geochemistry, vol. 48. Mineralogical Society of America, Washington, DC, pp. 363-390. Schuffert, J.D., Dastner, M., Emanuele, G. and Jahnke, R.A., 1990. Carbonate-ion substitution in francolite: a new equation. Geochim. Cosmochim. Acta, 54: 2323-2328. Taylor, J.C., 1991. Computer programs for standardless quantitative analysis of minerals using the full powder diffraction profile. Powder Diffraction, 6: 2-9. Tysdal, R.G., Johnson, E.A., Herring, J.R. and Desborough, G.A., 1999. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation, central part of Rasmussen Ridge, Caribou County, Idaho: US Geological Survey, Open File Report, 99-20-A. Tysdal, R.G., Herring, J.R., Desborough, G.A., Grauch, R.I. and Stillings, L.A., 2000a. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Dry Valley, Caribou County, Idaho: US Geological Survey, Open File Report, 99-20-B. Tysdal, R.G., Grauch, R.I., Desborough, G.A. and Herring, J.R., 2000b. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation, east-central part of Rasmussen Ridge, Caribou County, Idaho: US Geological Survey, Open File Report, 99-20-C. Tysdal, R.G., Herring, J.R., Grauch, R.I., Desborough, G.A. and Johnson, E.A., 2000c. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Sage Creek area of Webster Range, Caribou County, Idaho. US Geological Survey, Open File Report, 99-20-D.
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
189
Chapter 8
PETROGENESIS AND MINERALOGIC RESIDENCE OF SELECTED ELEMENTS IN THE MEADE PEAK PHOSPHATIC SHALE M E M B E R OF THE PERMIAN PHOSPHORIA FORMATION, SOUTHEAST IDAHO
R.I. GRAUCH, G.A. DESBOROUGH, G.P. MEEKER, A.L. FOSTER, R.G. TYSDAL, J.R. HERRING, H.A. LOWERS, B.A. BALL, R.A. ZIELINSKI and E.A. JOHNSON
ABSTRACT The Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation hosts the ore mined by the phosphate industry of southeast Idaho. It also hosts environmentally sensitive elements (ESE) such as Se, As, Hg, Ni, Cd, Zn, and Cr. Primary chemistry, elemental distribution patterns, and mineralogy within the Meade Peak were modified by element migration and possibly the introduction of elements. Fluids moved within the Meade Peak throughout its history, although the passage of fluids was highly variable in space and time, resulting in small domains of different rock chemistry and different mineralogy. Timing of major events affecting the Meade Peak and mineral habit are used to differentiate among detrital, diagenetic, epigenetic, and supergene mineral assemblages. Cross-cutting relationships among minerals are too rare to provide much paragenetic information. Carbonate fluorapatite (CFA) occurs in several forms, but dominantly as pelloids, some of which may have formed in situ during diagenesis. The other volumetrically significant form of CFA is interstitial cement that formed during diagenesis. Beginning during diagenesis and continuing intermittently, multiple generations of carbonate (dolomite and calcite) formed overgrowths and texturally complex carbonate cements. Movement and precipitation of silica followed a similar pattern. The ammonium feldspar buddingtonite, which generally rims orthoclase, also formed during diagenesis. Bacteria apparently played a significant role during diagenesis as well as during supergene processes, resulting in extreme fractionation of S isotopes and the possible bacterially mediated formation of minerals such as glauconite and sphalerite. Catagenesis, apparently culminating in oil generation, was the last significant diagenetic change. Thrusting accompanied by fluid (oil and brine) migration began during catagenesis in the Late Jurassic or Cretaceous and continued into the early Eocene. Fluorite ___carbonate _ barite _-L-bitumen veins formed as a result of brittle deformation and accompanying fluid movement. This fracturing event may have been associated with a period of extension and normal faulting (Neogene to Holocene). Passage of the Yellowstone hot spot to the north of the area during the Neogene is marked by silicic domes and basaltic
190
R.I. Grauch et al.
flows. The enrichment of Hg in fracture coatings might be the result of deposition from warm fluids associated with the emplacement of the silicic domes or a generally elevated, regional thermal gradient associated with the volcanism. Many of the fracture systems are still open and continue to provide fluid pathways that are the primary depositional sites for a wide variety of supergene minerals (such as Se, effiorescent salts) and element associations (such as Hg, Cd-S, Fe-Cr-O) in which many of the ESE are concentrated. Native Se is the most commonly identified host of Se in the studied samples. The largest concentration of Se occurs in open-fracture systems that cross-cut waste rock and ore units. The age(s) of native Se formation is not known; however, the latest period of Se mobility is the present. Direct measurement of effiorescent "salts" forming on new mine faces indicate that several ESE, including both Se and Zn, are concentrated on the faces soon after they are exposed. Zinc is present as hydrous sulfates, but the residence of Se in these "salts" is unknown.
INTRODUCTION Phosphate has been produced from the Permian Phosphoria Formation of the western United States for more than nine decades. Approximately 12% of the current US phosphate production comes from four mines in southeast Idaho (Fig. 8-1; see Jasinski et al., Chapter 3). There is concem regarding the environmental impact of the mining. Livestock have suffered from selenium (Se) toxicosis and surface water locally exceeds the US Environmental Protection Agency's guideline for Se content (Presser et al., Chapter 16). The extent of the environmental impact is addressed elsewhere in this volume (Hamilton et al., Chapter 18; Mackowiak et al., Chapter 19). The phosphate mines exploit ore from two zones of the Meade Peak Phosphatic Shale Member and discard the remaining material. The discarded waste material is thought to be the main host of the problematic Se and other environmentally sensitive elements (ESE) such as As, Cd, Ni, Zn, Cr, and Mo (Herring and Grauch, Chapter 12). The mined phosphate ore is also a potential source of the ESE, but is not typically considered an environmental problem because it is removed and processed. The stratigraphy of the Meade Peak at the four active phosphate mines in southeastern Idaho is described by Tysdal et al. (1999, 2000a-c) and Grauch et al. (2001). They measured and sampled two stratigraphic sections in active open-pits at each of the four mines. Additionally, a section of pre-mining drill core from the Enoch Valley mine was described, measured, and sampled. The sections were selected in order to compare deeper, generally less-weathered rock to shallower more-weathered, stratigraphically equivalent rocks. Rock from the core (section J) is assumed to be the least weathered. The distinction between least, less, and more weathered was based on the assumption that the depth below the pre-mining land surface approximates the extent to which the rock has been weathered. In general, this assumption is supported by the bulk chemistry and bulk mineralogy of the rock (Knudsen and Gunter, Chapter 7; Herring and Grauch, Chapter 12). Extensive literature addresses the petrography of the Meade Peak (e.g. McKelvey et al., 1959; Gulbrandsen, 1960; Mabie and Hess, 1964; Cook, 1969, 1970; and references therein)
Petrogenesis and mineralogic residence of selected elements
191
Fig. 8-1. Location map and generalized geology, modified from Blackstone and De Bruin (1987). Measured sections A, B, and J are located at the Enoch Valley mine; sections C and D, Dry Valley mine; E and F, Rasmussen Ridge mine; and G and H, Smoky Canyon mine. in which multiple stages of mineral growth are described and attributed to diagenetic processes. This detailed and informative body of research was limited by the technology available for studying the very fine-grained, Meade Peak siltstone, mudstone, phosphorite, and carbonate. The presence of opaque organic matter in many parts of the rock further
192
R.I. Grauch et al.
complicates the task of identifying fine-grained materials. Desborough (1977) was the first to systematically employ electron-microbeam techniques in the investigation of the Meade Peak. His work began the task of describing the diversity of micrometer- to submicrometer-sized mineral phases such as sulvanite (Cu3VS4)and powellite (CaMoO4) that are hosts for some of the ESE and associated metals. That work has recently been extended by several researchers (Grauch et al., 1999, 2000, 2002; Munkers, 2000; Perkins and Foster, 2002; Perkins and Foster, Chapter 10; and this chapter). In this chapter, we present new observations on the mineralogical evolution of the Meade Peak and on the mineralogical residence of some of the ESE. Emphasis is placed on Se because of its potential impact on the local environment and on Zn because of its apparent complex history of redistribution in the rocks.
GEOLOGIC SETTING AND HISTORY The Meade Peak was deposited unconformably on the Grandeur Tongue of the Permian Park City Formation and is overlain by the Rex Chert Member of the Phosphoria Formation. Unpublished geophysical data (R. Blank, USGS and oral communication from D. Campbell, USGS) and our field observations indicate that the Meade Peak basal contact is locally uneven with small (<5 m deep) solution cavities developed in the underlying dolomitic Grandeur Tongue. Major geologic events that affected the region after deposition of the Meade Peak were burial, oil generation, thrusting, extensional faulting, volcanism, and weathering (see Evans, Chapter 6). Catagenesis (thermal degradation of organic matter to form oil) and associated migration of oil from the Meade Peak into Wyoming began sometime during the Late Jurassic or Early Cretaceous and continued during the early stages of thrusting and development of the Wyoming-Idaho thrust belt (Sheldon, 1967; Momper and Williams, 1984; Fryberger and Koelmel, 1986). By the end of the Cretaceous, the Meade Peak in southeast Idaho was covered by about 8.5 km of Mesozoic sedimentary rocks (Claypool et al., 1978; Evans, Chapter 6). Thrusting continued through early Eocene (Armstrong and Oriel, 1986; Royse, 1993). Pathways established during thrusting permitted periodic fluid flow that crossed thrust plates and formations within thrust sheets (Atnipp et al., 1987). A second period of catagenesis occurred within the Wyoming-Idaho thrust belt during the last 55 Ma resulting in oil generated from Cretaceous sedimentary sources being trapped in the upper plate rocks of the Absaroka thrust (Warner, 1982). Sedimentary rock-hosted Cu-Ag-Pb-Zn occurrences and deposits are located in several of the thrust sheets (Loose, 1990). Host sedimentary rocks for the metal deposits range in age from Lower Paleozoic to Lower Cretaceous, but the age of mineralization is not well established. Loose (1990) suggested that some of the mineralization in the Absaroka thrust sheet formed during or soon after the generation of Phosphoria-sourced oil, but more likely was related to Cretaceoussourced oil and used Laramide fractures as primary fluid pathways. Extension and normal faulting began during the Neogene and continued into the Holocene, resulting in north-south trending basin and range structures and northeast
Petrogenesis and mineralogic residence of selected elements
193
striking normal faults (Armstrong and Oriel, 1986; Kellogg et al., 1999). Bimodal (mafic and silicic) igneous activity accompanied the Neogene passage of the Yellowstone hot spot to the north of the Meade thrust sheet (Pierce and Morgan, 1992). Silicic domes and basaltic flows that formed as recently as Quaternary mark its passage (Fig. 8-1; Fiesinger et al., 1982). Erosion, which marks the possible onset of chemical weathering, is the youngest major process to affect the area. Its earliest record may be the sediments shed into the Neogene basins (Kellogg et al., 1999). The four active Idaho phosphate mines are located within the Wyoming-Idaho thrust belt. All are located within the Meade thrust sheet (Fig. 8-1). We assume that our sampling sites have, in general, experienced similar physical and chemical histories during and after initial burial and early diagenesis. However, their detailed histories probably varied in response to the different structural settings of the four mines. Smoky Canyon mine is near the leading edge of the Meade thrust sheet and, like the Enoch Valley mine, sits on the west flank of a major anticline. Dry Valley mine is also located on the moderately dipping west limb of a major anticlinorium but is near a west-facing normal fault that strikes parallel to the anticlinal axis. The Rasmussen Ridge mine is situated near the crest of the nearly vertical east limb of the same anticline that contains the Enoch Valley mine. The different structural settings are supported by qualitative field observations that include the extent of brecciation, spacing and number of cleavage planes and faults, number of gash veins, and apparent thinning of beds (also see Evans, Chapter 6). The latter effect appears to be the result of several processes, including near bedding-plane-parallel faulting, tight folding, and chemical thinning (partial to complete dissolution of carbonate layers), which occurred during a complex history.
APPROACH AND METHODOLOGY The general lack of features such as veins, cross-cutting relationships, and consistent overgrowth or replacement textures that are typically used to define paragenetic sequences necessitated a different approach to unraveling the complex petrogenetic history of the Meade Peak. We used the chronological sequence of the major geologic events that affected the area as a framework for describing mineral paragenesis. For example, initial bitumen formation is assumed to have accompanied catagenesis. Minerals occurring in open-space fracture systems must have formed during or after the youngest brittle deformation. Features indicative of paragenetic position were used, where present, to confirm or refine mineral origins assigned within the geologic framework. Sample collection consisted of channel sampling of minable units defined by company staff at each mine. In some places, we subdivided those units based on obvious lithologic breaks. The resulting sample suite, therefore, represents thin (0.3-2.5 m), more or less lithologically homogenous units that could be composited to represent units thick enough to mine economically or to move as waste. Hand samples were selected for petrographic examination based on lithologic type, paragenetic textures, or mineralogic/chemical uniqueness. Of the more than 500 samples collected, 225 were made into polished
194
R.I. Grauch et al.
thin-sections for examination. Sample numbers indicate the measured section and the distance in feet above the base of the Meade Peak from which the samples were collected. For example, sample wpsJ8.8 was collected 8.8 fl (2.7 m) above the base of the Meade Peak in section J; the wps prefix indicates western phosphate stratigraphic sample (Fig. 8-2). Hand samples collected separately as part of an unpublished stratigraphic section at the Enoch Valley mine have a Z designation. Polished thin sections were examined using standard transmitted and reflected light microscopy. The opacity of thin sections (due to organic matter) and the fine-grain size of the rocks required that most minerals be identified with either scanning electron microscope (SEM) or electron microprobe techniques. Samples selected during optical examination (~50) were examined in a reconnaissance manner with a JEOL JSM-5800 LV electron microscope. Detailed examination of several samples was accomplished using a variety of SEM techniques. Entire polished thin sections were scanned at low magnifications (~ 100 x). Areas of particular interest that were identified optically were scanned at higher magnification (->2000x). Each sample was then scanned in its entirety at high magnification using enhanced contrast in backscatter electron mode to identify volumes of high or low average atomic weight or the location of individual mineral phases with high or low average atomic weight. Cathodoluminesence was used to distinguish different generations of feldspar and quartz. In addition to the examination of polished thin sections, several small samples of freshly broken rock chips or effiorescent salts collected from active mine faces were investigated. Mineral phases were chemically characterized and tentatively identified by means of semiquantitative energy dispersive X-ray analysis. A few samples that contain unusual minerals or textures were then transferred to a JEOL 8900 electron microprobe for quantitative wavelength dispersive X-ray analysis of individual minerals and detailed mapping of backscatter electrons (average atomic weight) and the distribution of specific elements based on measurements of specific X-ray wavelengths. Operating conditions (accelerating voltage, sample current, beam diameter, etc.) of both the SEM and electron microprobe were optimized for each application such as imaging, mapping, or analysis. Standards used for quantitative analyses are either natural minerals or synthetic materials distributed by C.M. Taylor Corp. Maps were obtained using JEOL-supplied software that permits the reliable, simultaneous, acquisition and production of a backscatter-electron image and element distribution maps for up to 15 elements (using both energy dispersive and wavelength dispersive X-ray systems). Maps of different size and spatial resolution were constructed based on the size of the features being mapped. The largest maps are 1000 x 1000 pixels, with each pixel = 10 i~m2. Dwell time for each pixel varied depending on the specific map, but was generally 40ms. The colors or gray shades of each map represent the intensity of the mapped X-ray signal. Colors were adjusted in order to better delineate minerals or volumes within the thin section in which a mapped element is concentrated. In order to cover most of a thin section (3 cm x 2 cm), six individual maps had to be merged. Our initial full thin-section map (Fig. 8-3) was merged in Adobe Photoshop after color adjustment of the individual maps using JEOL software. Subsequent merged maps were built within ENVI (The Environment for Visualizing Images, Research Systems, Inc.) using an unpublished
Petrogenesis and mineralogic residence of selected elements 4.
Zn(ppm)
Se (ppm)
,,,,,,,,,,,,,,,,,,,,,,,,, , , ,!!,, [ -0
Lithology
Rex Chert
195
500
V (ppm)
0
1000
0
2500
55m
170~," 50
UpzPoereOre
160, 150" 45 140" 40
120' 35
-"
110-
IIIIIIIIIIIIIIIIlllllllllllllllll
100-
MiddleWaste Zone
!
90-
25
-~.
20
"'-I
t
I
70"
60"
15
50.
40" 10
302
LowerOre Zone
1
205 10-
i1!!!iiill]!!lllll. lLlil[l[Itll!iiilii!!lllii! tlLltllll'lltl[I r ,!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
LowQr Waste L(~ne 0
Cu n.gue
"2 /
.... 7 .... 7 .... 2.... ~ .... I i 9 / --r"
/ / Z Z / : I z] ~ 1 : 7 1
Z ::_Z_;
-1
IL-
Explanation Dolostone
Mudstone
Siltstone
Carbon seam
Phosphorite
Fig. 8-2. Meade Peak Member, section J: generalized lithology, major mining intervals (zones), and Se, Zn, and V concentrations of channel samples. Limits of individual channel samples are shown by pale gray, horizontal bars in the concentration columns. Concentrations for channel samples that appear blank are too low to be represented on the figure. Data from Grauch et al. (2001) and Herring et al. (2001).
196
R.I. Grauch et al.
Fig. 8-3. (A) Transmitted light image of typical phosphorite (wpsB 133) (B) Combined backscatter electron and Se and Zn X-ray maps of phosphorite (same field of view as (A) (six sets of maps merged to form a single image, see text). The Se and Zn maps were manipulated to show single points that correspond to the most intense range of X-ray signal for each element; red pixels represent native selenium and yellow pixels sphalerite (red and yellow pixels are enlarged to facilitate visibility). Large box indicates location of C. Small box is approximate location of D. (C) Close-up of B; lower insert shows two spherical inclusions in a CFA pelloid; the sphere on the right contains CFA (gray), bitumen (dark gray), and open space (black). The sphere on the left (enlarged in the other insert) contains native Se and CFA as well as open space (black); X-ray energy spectra are shown for native Se (the C coat on the sample is responsible for the C signal). (D) Fission-track map showing different U contents of adjacent CFA pelloids; the darker the shade the greater the U content; note that the apparent fractures, which are pale gray and marked by arrows, are bitumen veins with minor U contents; see Zielinski et al. (Chapter 9) for a discussion of the fission-track technique; A, B, and C from polished thin section, upper-ore zone; D is from a separate polished thin section made from the same blank.
routine written by Ray Kokaly (USGS). The routine merges up to six maps prior to color adjustment. The advantage of this approach is that adjusting the color after merging the maps eliminates the possibility of developing different color schemes for the individual maps, which could bias interpretation o f the m e r g e d map. M e r g e d maps were then manipulated within Adobe Illustrator prior to interpretative use and publication.
Petrogenesis and mineralogic residence of selected elements
197
RESULTS AND DISCUSSION A variety of minerals were identified optically, chemically, morphologically, structurally, or by a combination of techniques (Tables 8-I and 8-II). Additionally, many element associations were recognized, but not positively accredited to a discrete mineral (Table 8-II). In some samples, the element associations are thought to represent minerals in a distinct crystallochemical class, such as silicate, carbonate, or sulfate. Other element associations seem to be mixtures of minerals that are too fine grained to be distinguished. Element associations that were determined with electron beam techniques are referred to by their identified components such as Fe-O. An element association of Fe-O could represent hematite, magnetite, or a hydrous Fe oxide such as goethite. Many minerals were positively identified and paragenetically classified (Table 8-I). Almost all minerals occur in all rock types, albeit in different proportions. Because of the difficulties of dealing with fine-grained and opaque samples, the paragenetic classification scheme adapted here is loosely defined and subject to interpretive biases. Detrital minerals are relatively easily classified by their grain shape (irregular, broken, abraded, etc.) and lack of authigenic texture. The authigenic/diagenetic mineral assemblage incorporates minerals that formed in unconsolidated sediment and/or during lithification. They are generally recognized as overgrowths, replacements, cements, and fillings of interstitial void. Glauconite and some CFA are exceptions (see following discussion). Both epigenetic and supergene minerals formed after lithification and brittle deformation of the Meade Peak. However, distinguishing between them is often difficult in these rocks; therefore, they are grouped together. They occur in veins, along grain boundaries, rarely as replacements, on fracture surfaces, and as open-space fillings. Classifying a specific mineral occurrence is often straight forward in terms of the three major categories used in Tables 8-I and 8-II. However, differentiating timing within the major categories is subject to ambiguity. Several minerals occur in multiple paragenetic positions indicating multiple generations. The following discussion loosely follows a chronologic sequence of mineral formation and element mobility.
Detrital assemblage The detrital mineral assemblage is dominated by silicates, but contains minor amounts of phosphates, carbonates, and oxides. Two K-feldspars, orthoclase and microcline, occur in this assemblage. The microcline is a minor constituent observed only as inclusions or in composite grains in which it is armored by detrital quartz. Detrital plagioclase of undetermined composition occurs in a similar manner, but the armoring phase can be either quartz or orthoclase. Albite occurs both as individual grains and as inclusions. These feldspar relationships are similar to those that led Desborough et al. (200 l) to infer that unarmored plagioclase may have been removed from the Meade Peak during diagenesis. The combination of detrital quartz, feldspars, muscovite, monazite, xenotime, tourmaline, Cl-bearing
198
R.I. Grauch et al.
TABLE 8-I Detrital and authigenic/diagenetic minerals of the Meade Peak Phosphatic Shale Member of the Phosphoria Formation in southeastern Idaho Mineral
Detrital
Authigenic/diagenetic
Minerals that are either major (greater than 3 wt. %) or minor constituents
Albite Apatite (F- and S-bearing) Buddingtonite-potassium feldspar solid solution Calcite Dolomite Muscovite Orthoclase (contains 0.5 or more wt.% Na20) Quartz
Yes Possibly No
Yes Yes Yes
Possibly Possibly Yes Yes
Yes Yes No No
Yes
Yes
Yes Yes Yes No Yes Possibly Yes Yes Yes
No No No Yes Yes Yes No Possibly No
No Yes No Yes Yes Yes
Yes Yes Yes No Yes No
Minerals that are only minor constituents
Apatite (Cl-bearing) Biotite Chlorite Glauconite lllite Kaolinite (1:1 clay) Microcline Monazite Plagioclase (undetermined composition) Pyrite Rutile (and/or anatase) Sphalerite Tourmaline Xenotime Zircon
Identification of minerals listed in this table was accomplished by a combination of X-ray diffraction, optical microscopy, quantitative X-ray dispersive spectrometry, and semiquantitative X-ray energy dispersive spectrometry. Data sources are this study; Knudsen and Gunter (Chapter 7); Johnson (unpublished data); Kristen Sanford (personal communication, 2002). apatite, and abundant zircon suggests at least one source area was rich in felsic igneous and/or metamorphic material. Based on textural criteria such as broken and rounded grains, a minor amount of the carbonate material may be detrital. Perhaps some of the Meade Peak carbonate units were reworked from up-slope carbonate depocenters. Broken CFA pelloids, mixed populations
Petrogenesis and mineralogic residence of selected elements
199
TABLE 8-II Diagenetic/epigenetic/supergene minerals of the Meade Peak Phosphatic Shale Member in southeastern Idaho Mineral
Diagenetic
Epigenetic/supergene
Barite
Zinc
No Yes Possibly No No No Yes Yes Yes Possibly No Yes Possibly No
Possibly Yes ? Yes Yes Yes Yes Possibly No No ? Yes Possibly Yes No No Yes
Association
Authigenic/diagenetic
Epigenetic/supergene
"Bitumen ''~ Fluorite Goslarite Gypsum Iodargyrite (AgI) Pyrite Vaesite-pyrite solid solution
Roscoelite Selenium
Silver Smectite
Sphalerite (spheres) Sulvanite
Uraninite
Element associations not positively identified as a mineral phase Native elements Sb Cu Fe Pb Ni Pt S Alloys Fe-Cr Fe-Ni W-Si Sulfides (-S) 2 Cd Cu Fe-As Fe-Cr Pb
Possibly Possibly Possibly Possibly Possibly Possibly Possibly
Possibly
Possibly
Continued
200
R.L Grauch et al.
TABLE 8-11 Continued Association
Authigenic/diagenetic
Epigenetic/supergene
Selenides (-Se) 2 Cu
Possibly
Halides Ag-Cl
Yes
Oxides (--0) 2 Ca-Mg Cu _+Ag Fe Fe-Ni-Zn Mn Mn-Zn-Ni Ti Ti-Fe U Zn Carbonates (-C-O) 2 Zn Sulfates (-S-O) 2 Fe Ni Ti-Fe Zn Zn-Ni Phosphates (-P-O) 2 Al Fe Fe-Al La Silicates (-Si-O) 2 Al A1-Cr Ca Ca-A1 Ca-Mg-Al Fe-A1 Fe-Cr Ni-Al K-A1 K-Fe-Ti-Mg +_Cr-A1 K-Mg _ Cr _+Fe-Al Ti
m
Yes Yes Yes
Yes n
Petrogenesis and mineralogic residence of selected elements
201
TABLE 8-I1 Continued Association
Authigenic/diagenetic
Epigenetic/supergene
Mixtures 3 Se-Cd-Ni-Ca-C-S-O S-C-V-Si-A1-Ca V-Cr-Fe-Si-A1-K-Mg-C-O-S Zn-S-O-Si-AI lBitumen is not a recognized mineral; the term is loosely used here to indicate a C phase identified by energy dispersive analysis (see text); it is included because it is a significant rock component that formed during oil generation 2Elements listed in parentheses are to be appended to the list of elements listed in the association (i.e. (-O) indicates that O is associated with all of the elements in the oxide association. 3Elements listed in order of apparent abundance in a volume that contains phases that are too small to resolve with a 1 I~m diameter electron beam. Identification of most minerals was accomplished by semiquantitative X-ray energy dispersive spectrometry and morphology; crystalline native selenium, goslarite, and smectite were confirmed by X-ray powder diffraction analysis; data sources, this study, Kristen Sanford (personal communication, 2002). Dash means data not sufficient for paragenetic classification of identified phase. Minerals tentatively identified in italics. of chemically distinct pelloids, and cross-bedded layers of pelloids (Fig. 8-4) suggest that some of the CFA pelloids are detrital and may be reworked from up-slope sources.
Authigenic/diagenetic assemblage The authigenic/diagenetic assemblage is dominated by CFA. However, as shown in Tables 8-I and 8-1I, a wide variety of authigenic/diagenetic minerals formed during Meade Peak history. This mineralogic diversity and associated textures (described below) indicate that fluids of different compositions and different ages moved through these rocks, resulting in the redistribution of different elements at different times. The textures also suggest that micro-chemical environments were extensively developed within Meade Peak rocks.
Phosphate There are at least two generations of CFA: early pelloids (Fig. 8-3) and later interstitial cement. The CFA cement may be temporally and/or genetically related to masses of CFA, felt-like in appearance (Fig. 8-5B), that seem to replace earlier generations of CFA. Differences in pelloid composition (average atomic weight) are obvious in the backscatter
202
R.I. Grauch et al.
Fig. 8-4. (A) Transmitted light image of cross-bedded (red lines) phosphatic siltstone; box indicates the location of B; the large ovoid that is partly included in the box may be a burrow filled with sediment that is different than that outside the ovoid. (B) Selinium crystals (Se), gypsum, pyrite framboids, and a variety of silicates filling a fracture around the base of a possible burrow; black part of vein is epoxy; backscatter electron image; polished thin-section wpsD 207, upper-waste zone.
Petrogenesis and mineralogic residence of selected elements
203
Fig. 8-5. (A) Backscatter electron map of dolomitic carbonate showing calcite rimmed and partly replaced by dolomite; also masses of CFA, felt-like in appearance, that apparently replaced earlier, deformed pelloids. Fe-O is a pseudomorph after pyrite framboid. (B) Felt-like mass of CFA (same sample as A). (C) CFA (perhaps deformed pelloids or interstitial fillings) of apparently the same generation showing different textures probably due to different degrees of compositionally controlled reactions with later fluids. A and B from polished thin-section wpsD 17, lower-ore zone; C from polished thin-section wpsD 207, phosphatic siltstone, upper-waste zone.
204
R.I. Grauch et al.
electron image (e.g. Fig. 8-3B). Quantitative microprobe analysis of the CFA has proven difficult because of the apparently variable C content, which we could not reliably quantify, OH- content, and generally ubiquitous micrometer- to sub-micrometer-sized inclusions. However, more than 1000 unpublished spot analyses for Ca, Na, P, S, and F suggest that there are no significant compositional differences between rims and cores of individual pelloids. There are, in some samples, minor compositional variations among pelloids within the same sample and significant differences (several wt.%), especially of Na and F contents, among pelloids in different samples. A bimodal distribution of analysis totals (sum of all analyzed components) suggests that there are differences (---2 wt.%) in OHand/or CO~- contents. Variation of U content, which is clearly reflected in differences of fission-track density shown in Fig. 8-3(D), may contribute to variations in average atomic weight shown in Fig. 8-3(B) and 8-3(C). Individual pelloids of the same generation may react differently to fluids that move through the rock. Jahnke (1984) demonstrated that even minor increases in the carbonate content of CFA cause major increases in CFA solubility. Adjacent CFA masses, apparently of the same generation, can have very different porosity (Fig. 8-5C). The greater porosity could result from preferential dissolution of inclusions (including organic matter) or from preferential dissolution of CFA with greater carbonate content. The latter explanation is supported in samples where there is no apparent variation in inclusion population.
Silicates and carbonates
Glauconite pelloids have only been observed in the upper parts of the upper-waste zone. Formation of glauconite is generally considered to be a replacement process that occurs at or just below the sediment-water interface in a reducing environment during and soon after sediment deposition, but does not necessarily identify a specific depositional environment (Chafetz and Reid, 2000). Bacteria may be involved in glauconite formation (Geptner and Ivanovskaya, 2000). Quartz overgrowths on detrital quartz grains are fairly common. Field observation of silica replacement of siltstone and mudstone suggest that at least some of the quartz overgrowths were the result of silicification that affected large parts of the upper Meade Peak. Buddingtonite (ammonium feldspar) overgrowths on detrital orthoclase are also common throughout the Meade Peak (Knudsen and Gunter, Chapter 7), but are especially concentrated in the middle-waste zone. Some buddingtonite is zoned with compositions ranging between Kf20Bd80 and Kf65Bd35 (Fig. 8-6). It is not clear whether complete solid solution occurs within this composition range. End-member buddingtonite has not been observed. Degradation of organic matter is probably responsible for the release of ammonium required for the growth of buddingtonite. Such degradation is often the result of bacterial activity during early diagenesis prior to oil formation (Ramseyer et al., 1993). Buddingtonite also occurs as a fracture filling, possibly the result of a later period of ammonium generation related to oil formation. Diagenetic albite has also been observed (Desborough et al., 2001).
Petrogenesis and mineralogic residence of selected elements
205
Fig. 8-6. Zoned buddingtonite (b) overgrowth on detrital orthoclase (Kf), with detrital albite (alb) and quartz (q); KFz3Bd77--23% end-member orthoclase and 77% end-member buddingtonite; polished thin-section wpsJ 138.2, siltstone, middle-waste zone. Roscoelite (V illite) is a common constituent in the V-rich zones of the Meade Peak (Fig. 8-2). It fills interstitial voids (Fig. 8-7) and is disseminated within thin layers (tens of micrometers thick) that may be primary depositional layers or thin alteration zones parallel to bedding. Sphalerite is included in some of the roscoelite (Fig. 8-7) possibly contributing to the apparent correlation between Zn and V, indicated by rock composition data (Fig. 8-2). However, the apparent association of Zn and V is not evident on the scale of elemental maps. On a scale of a thin section, V is widely disseminated on bedding planes and interstitially, whereas Zn occurs in discreet mineral grains that do not correspond to the greatest concentrations of V (Fig. 8-8). In several places, roscoelite is spatially associated with vaesite (NiSz)-pyritess (ss = solid solution). However, there is little correlation between V and Ni on the scale of a half centimeter (Fig. 8-8). Beginning during diagenesis and continuing intermittently, multiple generations of carbonate (dolomite and calcite) precipitated, resulting in complex carbonate cements and overgrowths of carbonate grains (Fig. 8-5A) (Cook, 1969, 1970; Hiatt, 1997). Dolomite also forms aligned, ellipsoidal, polycrystalline masses that appear to lie on cleavage
206
R.I. Grauch et al.
Fig. 8-7. Unusually large aggregate of roscoelite crystals within what may have been an interstitial void; large sphalerite crystal is engulfed by the roscoelite; polished thin section wpsJ 8.8, phosphorite, lower-ore zone. planes, suggesting fluid movement and crystal growth during a period of folding, perhaps during emplacement of the Meade thrust sheet and concomitant formation of most major folds in the area.
Sulfides
Pyrite is ubiquitous in the Meade Peak. The earliest generation formed as framboids. In rare samples, pyrite appears to have replaced shell material. Some of the early pyrite occurs as inclusions in CFA pelloids. Later generations of framboidal pyrite occur in veinlets and in bitumen veins. Framboidal and euhedral to subhedral pyrite occurs sporadically in bedding-parallel structures that contain abundant clay minerals. In several samples, all forms of pyrite are associated with vaesite-pyritess (Fig. 8-9). In this occurrence, CFA grew around a pre-existing framboid and is in turn partially rimmed by vaesite-pyritess. In rare samples, vaesite-pyritess appears to replace the walls and partially fill the interior of unidentified fossils, suggesting that at least some of this Ni-rich pyrite formed prior to extensive sedimentary compaction. We have not been able to obtain reliable compositional data for the vaesite-pyritess because of the fine grain-size. Nor have we been able to gather enough data for pyrite to determine if there are compositional differences among different generations of pyrite.
Petrogenesis and mineralogic residence of selected elements
207
Fig. 8-8. Backscatter-electron map (six sets of maps merged to form a single image, see text), rectangle indicates the area of X-ray intensity maps for Zn, V, Se, and Ni; Ni map is superimposed on the backscatter-electron map; X-ray intensity (element concentration) for Zn is shown in reds and blues with red being the greater concentration; the other elements are shown in shades of blue with greater concentration proportional to increased color saturation and brightness; polished thin-section wpsJ 8.8, phosphorite, lower-ore zone.
208
R.I. Grauch et al.
Fig. 8-9. Vaesite-pyritessovergrowths surrounding CFA which surrounds pyrite framboids; polished thin-section wpsJ 8.8, phosphorite, lower-ore zone.
Sulfur isotope analyses of pyrite from two samples yielded 834Spyvalues in the range o f - 16.4%o (+0.6%o, 3 splits, Dry Valley mine) to -56.7%o (___1.0%o, 4 splits, Rasmussen Ridge mine) (analyses by Geochron Laboratories and replicated by R. Rye, USGS). The pyrite samples, which are from siltstone in the middle-waste zone, are unusual for the Meade Peak in that they are coarse grained enough (> 50 ~m) to hand pick. The very negative value, -56.7%o, suggests extreme fractionation of S during biologically mediated sulfate reduction and pyrite formation. Such fractionation could occur in one precipitation event but more likely is the product of multiple periods of S reduction and pyrite precipitation. The latter interpretation is consistent with the hypothesis of multiple periods of fluid movement within or through the Meade Peak. Sphalerite is also widespread in the Meade Peak and probably has at least three generations: early diagenetic, later diagenetic, and supergene (discussed below). The earliest generation forms either coarse (>50 I~m) or fine-grained crystals. The coarse-grained material is generally found as inclusions in CFA and is associated with sulvanite (Cu3VS4) (Fig. 8-10). The unusual globular shape of the example in Fig. 8-10 suggests that the sphalerite replaced something. The finer-grained sphalerite also occurs as apparent inclusions in CFA and is probably disseminated in the matrix. A later diagenetic generation of sphalerite replaces CFA (Fig. 8-11). This "sphalerite disease" appears to preferentially attack only specific CFA and to leave adjacent CFA pelloids unaffected.
Petrogenesis and mineralogic residence of selected elements
209
Fig. 8-10. Large sphalerite with partial overgrowth of sulvanite, all included in CFA; top to bottom, transmitted light, reflected light, and backscatter electron image; polished thin section wpsJ 8.8, phosphorite, lower-ore zone. As with pyrite, we have insufficient analytical data to distinguish compositional differences among the multiple generations of sphalerite. Data in Table 8-III indicate differences in Cd and possibly Cu content among samples. Sphalerite with higher Cu content is from samples wpsJ 8.4 and 8.8 which also host sulvanite (Cu3VS4)and which contain elevated concentrations of V (Fig. 8-2). The apparent correlation between Zn and V in the J section rocks (Fig. 8-2) cannot be accounted for by the association of sulvanite and sphalerite.
210
R.I. Grauch et al.
Fig. 8-11. (A-C) Backscatter-electron maps showing sphalerite partly rimming and replacing CFA pelloids; bright grains in C are sphalerite, gray is CFA, and dark (black) areas are either voids or mixtures of bitumen and void; box in A is area shown in B and box in B is area shown in C; polished thin section wpsJ 8.8, phosphorite, lower-ore zone.
Sulvanite is too rare in these rocks and there is not an equivalent correlation with Cu (Herring and Grauch, Chapter 12). O f the sulfides for which we have quantitative analyses, sulvanite contains the highest concentration of Se (Table 8-11I). However, semiquantitative analyses of vaesite-pyritess suggest that it contains even more Se.
E
oo
0o
0o
oo 0o
O~
O~ 0o
fxl
~
0 r O~
~
0
~
0
0
~
0
~
Cxl Cxl 0 0 0 0
~
9
~
0 Cxl 0
0
~
0
.
~
0
.
~
~ 0
0
~
~
0
~ ,--,
'~'~ " ~ -'~ ~
~
d d d d d o o d o d o o
~
d
~
--
~
d d
d
~ d
d
9
o
~ ~
d
o ~ d ~ d o d o o M o ~
d
o M d ~ o d d o o M d ~
~
0
Petrogenesis and mineralogic residence of selected elements
m
o
0
an k~ 0
0 0 o . ,..~
~
-" "~" O0
~
s
0
Z
E 0
o
-"
O~
0
dod
6ddd
.
cn
~0 o 0 n~
E
E
0
0
t-4
o
Z~n
21 1
212
R.I. Grauch et al.
Other minerals
Uraninite occurs as inclusions in CFA and as discreet, < 1 ixm, crystals in the matrix (see Zielinski et al., Chapter 9). Matrix-hosted uraninite was observed only in samples in which sphalerite replaces CFA, suggesting that the U may have been derived from the dissolution of CFA. Xenotime and possibly monazite occur as diagenetic minerals in irregular, small (less than a few micrometers) masses along grain boundaries and as thin (<2 I~m) overgrowths. Bitumen is one of the latest phases formed during diagenesis, concomitant with catagenesis. Bitumen is used in a very loose sense to describe the organic-carbon phase that occurs in these rocks; it is generally volatilized by an electron beam, but has not been tested for solubility in organic solvents (a key characteristic ofbitunaen). It is disseminated in most rock types and forms at least two temporally distinct sets of veins. The veins contain a variety of entrained minerals, many of which are very fine-grained ( _ 1 Ixm in largest dimension), making it difficult to distinguish between X-ray signals generated from entrained material and those from elements incorporated in bitumen. Despite this confounding factor, we tentatively conclude that minor amounts of Se are incorporated in some bitumen. There is no obvious correlation between Se and age of the bitumen. The distribution of Se in bitumen appears analogous to that of U. The uniform distribution of fission tracks in the veins suggests that U is incorporated in the bitumen rather than residing in entrained grains (Fig. 8-3D; see also Zielinski et al., Chapter 9).
Epigenetic/supergene assemblage
The epigenetic/supergene assemblage is dominated by oxides, sulfates, phosphates and native metals (Table 8-II). Many minerals (reported as groups of elements) have not been positively identified, some may be hydrated phases. As with the authigenic/diagenetic assemblage, the textures and juxtaposition of normally non-coexisting minerals suggest that micro-chemical environments were extensively developed within the Meade Peak rocks.
Silicates
Buddingtonite veins that are as wide as 100 I~m occur in the middle-waste zone. Fragments ofmicrobreccia are commonly included in these otherwise monomineralic veins. The time of vein formation is unknown. They may have formed during or just before catagenesis or they may be related to a later event in which ammonium was locally generated by further degradation of organic matter. In either scenario, ammonium had to be a significant component of the fluid within the fracture system. It is not clear if a temporal relationship exists between these veins and the relatively common buddingtonite overgrowths on orthoclase.
Petrogenesis and mineralogic residence of selected elements
213
The most common veins in the Meade Peak are dominated by quartz. They are apparently most numerous high in the section near the contact with the Rex Chert.
Halides
Veins containing fluorite ___calcite ___bitumen formed during catagenesis or thereafter. The veins are usually thin (<25 I~m). The fluorite _ calcite _ barite _ quartz assemblage is associated with samples that also contain polycrystalline masses ( < 10 Ixm to ~ 1 cm) of the same minerals that appear to have replaced preexisting structures. The larger structures are ovoid in cross-section and may be deformed fossil pseudomorphs or deformed replacements of burrow-like structures such as that shown in the lower right quadrant of Fig. 8-4(A). Iodargyrite (AgI) was tentatively identified in a siltstone sample from the Rasmussen Ridge mine. It forms well-developed crystals that are volatile in the electron beam. The primary mode of occurrence in other localities is as an oxidation product of sulfide deposits. Boyle (1997) suggested that iodargyrite is an indicator mineral for arid to semiarid environments.
Other minerals
Mercury may be part of the assemblage of epigenetic elements. Samples containing large amounts of Hg are rare; most channel and hand samples contain less than 1 ppm Hg (Herring et al., 1999, 2000a-c, 2001). Anomalous samples, identified with a portable X-ray fluorescence unit, contain between ~180 and ~90ppm Hg (see Grauch et al., 2001, for description of the instrument and technique; data are unpublished). Mercury is concentrated in fracture coatings, although its mineral residence has not been determined. The surface coatings with which Hg is found include Fe-A1-P-O, A1-P-O, and iodargyrite.
Selenide (CuxSey)
Selenide minerals were observed in only one sample, wpsD 207. Several very small (< 1 Ixm) grains of Cu selenide occur adjacent to the vein shown in Fig. 8-4(A) and (B).
Native elements
Selenium occurs in a variety of settings and forms within the Meade Peak. As indicated above, it is an integral component of several different sulfides (pyrite, sphalerite, and vaesite-pyritess) that are thought to have formed during early diagenesis and may be incorporated in bitumen that formed as a result of catagenesis. However, its most common
214
R.I. Grauch et al.
mineral residence in our samples appears to be as native Se. The dominant habit of native Se is as small ( < 2 ~m) grains that occur either within the matrix or as apparent inclusions in other phases. Some problems arise with classifying the small grains as inclusions that formed contemporaneously with CFA as illustrated in Fig. 8-3(C). In this figure an unusually large mass of native Se appears to be an inclusion in a CFA pelloid. In Fig. 8-12(I),
Fig. 8-12. Selenium crystals. (A-D) Native Se crystals coating fracture surfaces; sample, wpsJ 186.8, glauconitic phosphorite, upper-waste zone; (E) native Se and sphalerite filling open space around pyrite framboids; polished thin section wpsC 153, siltstone, upper-waste zone; (F) native Se next to Fe-O pseudomorph after a pyrite framboid; freshly broken surface of wpsC 157, siltstone, upper-waste zone; (G) native Se crystal on platelets of A1-Si-O (kaolinite?); freshly broken surface of wpsC 157, siltstone, upper-waste zone; (H) native Se crystals in Fe-P-O vein that cross-cuts vein of microbreccia (dark gray) cemented by quartz and other silicates; polished thin-section wpsJ 186.8, glauconitic phosphorite, upper-waste zone; (I) native Se within cubic pit in CFA; Fe-S appears brighter than it should be because of charging on the edge of the grain; polished thin section wpsJ 186.8, glauconitic phosphorite, upper-waste zone.
Petrogenesis and mineralogic residence of selected elements
215
a small grain of native Se also appears to be an inclusion in CFA, but note that it is attached to the edge of a void. In both samples, the native Se is not supported by much matrix material; in Fig. 8-3(C), the Se-CFA-bearing sphere is similar to the adjacent one that contains bitumen and CFA but no Se. The similarity of textures between the two spheres and the open space in the Se-bearing sphere suggest that the native Se formed during a replacement process in which some of the bitumen was removed. This persistent relationship coupled with the porosity commonly observed in CFA and the previously described replacement of CFA by sphalerite (Fig. 8-11 C) suggests that most, if not all, of the small native Se grains formed in pore space after catagenesis. Native Se also fills pore spaces where it grew around pre-existing pyrite framboids (Fig. 8-12E). In some samples, large (>5 Ixm) crystals of native Se precipitated in open fractures and were subsequently encased in another mineral(s). Gypsum and a variety of silicates (predominantly clay minerals) are common later vein fillings (Fig. 8-4B). An unusual, young, fracture filling dominated by a Fe-phosphate has a similar texture with sprays of native Se crystals encased in later vein-filling minerals (Fig. 8-12H). The fracture fill appears to be part of the youngest event that created a complex set of anastomosing and, in places, cross-cutting fractures. The earliest stage of this fracture event is a microbreccia with clay minerals and quartz as the dominant matrix (Fig. 8-12H). Open-space growth of native Se crystals (Fig. 8-12A-D) has been observed at several localities, including both the upper-ore and middle-waste zones at Enoch Valley mine and the middle-waste zone at Dry Valley mine. In these samples, the crystals preserve their delicate modifications and have not been enclosed by later minerals. The open-fracture systems formed late in the history of the Meade Peak have remained open since native Se precipitation, and may be present-day conduits for ground-water flow. Native Se crystals also grew in voids (Fig. 8-12F and G). Both examples are from a freshly broken surface of a siltstone. In Fig. 8-12(F), the native Se is adjacent to a Fe-O pseudomorph after a framboid (upper right corner of the figure). The approximately 30 Ixm long native Se crystal shown in Fig. 8-12(G) apparently grew between platelets of A1-Si-O (kaolinite?). Whole rock Se isotopes from five samples range between - 2 . 5 and +2.74%0 68°/76Se (Hagiwara, 2000). The samples are from the Enoch Valley mine (wpsZ 75), Rasmussen Ridge mine (wpsE 82), Dry Valley mine (wpsC 157 and FMC 220-2), and the Gay mine, which is located -~100 km northwest of the Enoch Valley mine. The three wps samples are all from the middle-waste zone. The data are permissive of either biotic or abiotic isotope fractionation. Additional work on specific generations and forms of Se will be required to determine whether one or both of the possible fractionation processes occurred. Native Zn occurs in an oxidized portion of the lower-waste zone in section J (Fig. 8-13). The pervasive oxidation process apparently involved the removal of organic material and most primary Fe minerals (Fig. 8-13A). The process included formation of secondary oxides including hematite and Mn oxide. The latter contain thousands of ppm Zn and Ni (black spots in Fig. 8-13A). The native Zn is rimmed by Zn carbonate which in turn is partially rimmed by Zn oxide (Fig. 8-13B). In addition to replacing Zn carbonate, some of the Zn oxide pseudomorphs euhedral sphalerite tetrahedra. Sphalerite occurs within a few
216
R.I. Grauch et al.
[ O@m •
~
i
B
Zn
--
S
Zn --C --0
Native Zn - -
Zn--O
Fig. 8-13. (A) Oxidized (organic C - poor) siltstone near the base of the Meade Peak; black areas are Mn oxide that contains hundreds to thousands of ppm Ni, Zn, and Cr; box shows approximate location of B (not to scale); (B) native Zn partially rimmed by Zn carbonate that, in turn, is partially rimmed by Zn oxide; polished thin section wpsJ 0.8, lower-waste zone. micrometers of the native Zn. Timing of native Zn formation is unclear. It may be related to alteration caused by oxidized, basinal fluids or it may have formed later as part of supergene processes. In either scenario, the Zn carbonate and Zn oxide record the interaction between native Zn and evolving, oxidizing fluid(s) that was, at least initially carbonate bearing and was probably responsible for oxidizing this part of the Meade Peak. Occurrences of native Zn elsewhere in the world are all related to oxidized parts of sulfide deposits or to placers (Boyle, 1961; Clark and Sillitoe, 1970). A variety of other native metals and alloys (Table 8-II) occur as discrete, very small (<1 ~m) grains in the matrix and as apparent inclusions in CFA. Their abundance, variety,
Petrogenesis and mineralogic residence of selected elements
217
and occurrence in both polished thin sections and freshly broken samples indicate that it is unlikely that they are contaminants. The inclusions are unusual because the grains are always much smaller than the voids in which they occur. Rarely, C (bitumen) is seen within the same void. This mode of occurrence is similar to that of native metals and alloys, including Au, Ag, Bi, and Pd, precipitated in very restricted reducing environments in red-bed sequences (Hofmann, 1990, 1993).
Oxides
Hematite is only common in the oxidized portion of the lower-waste zone in section J where it forms large (-~10 txm) grains. It may occur elsewhere as minute grains, but cannot be distinguished within the complex mixtures of secondary hydrous iron oxides that commonly replace pyrite (Fig. 8-12F) and other sulfides. Manganese oxide was only positively identified in the oxidized sample (wpsJ 0.8). However, preliminary X-ray absorption fine structure spectroscopy (XANES) suggests that Mn-O occurs with Se a+ in samples from the upper-waste zone at Dry Valley mine (wpsC 157 and wpsD 207) and in the upper-ore zone at the Enoch Valley mine (wpsB 133). A variety of other oxides were observed (Table 8-II), but there is no particular discernable pattern to their distribution or mode of occurrence.
Phosphates
Secondary phosphates occur as disseminations, vein fillings, and coatings on fracture surfaces indicating that P has been and currently is locally mobile. Fe-A1-P-O and A1-P-O occur with iodargyrite and Hg. Fe-P-O also forms veins with native Se (Fig. 8-12H).
Sulfates
Gypsum is a common component in many veins, on fracture surfaces, and as part of an effiorescent salt assemblage. Barite, as mentioned above, is part of the fluorite vein and replacement assemblages. Goslarite (ZnSOa-7H20) is a common effiorescent salt that has been identified by X-ray diffraction. Several other sulfates, including other zinc sulfates, occur in polished thin sections and as effiorescent salts (Table 8-1I). The effiorescent salts form rapidly (overnight in some places) on mine faces and are ephemeral. Analysis of effiorescent salt using a portable X-ray fluorescence analyzer indicates that Se is present (in the tens of ppm range) in samples from some locations (Enoch Valley and Smoky Canyon mines), although its residence and absolute concentration are unknown. The analyzed effiorescent salts occurred on mine faces as polycrystalline masses up to 10 cm in longest dimension and several tens of millimeters thick.
218
R.I. Grauch et al.
Fig. 8-14. Spheres of Zn-S (probably sphalerite) on etched dolomite substrate from freshly exposed mine surface; possibly the result of bacterially mediated sphalerite growth (see text); sample wpsZ 45.5, carbonate-bearing phosphatic siltstone, lower-ore zone, Enoch Valley mine. Sulfides
CdS is rare in our sample suite. It is fine-grained (< 1 p~m) and appears to be an alteration product of sphalerite with which it is spatially associated. ZnS spheres (probably sphalerite) grew on a corroded dolomite substrate (Fig. 8-14). The sample (wpsZ 45.5) is from a freshly exposed mine face at the Enoch Valley mine. The size and coalescing arrangement of the spheres are similar to those described by Labrenz et al. (2000) as aggregates of nanometer-scale sphalerite particles precipitated in modern biofilms. This suggests that bacteria were or are actively mediating sphalerite precipitation in the Meade Peak.
CONCEPTUAL MODEL The following conceptual model of the history of metal movement in the Meade Peak is based on our preferred interpretation of the observations presented in this chapter. The model is outlined in Table 8-IV in which the major geologic events affecting the Meade Peak are listed in chronologic sequence along with the inferred consequences for metal mobility. References documenting tectonic events, timing of catagenesis, and formation of sediment-hosted Cu-Ag-Pb-Zn deposits are listed in the section on geologic setting.
Petrogenesis and mineralogic residence of selected elements
219
TABLE 8-IV Conceptual model of events affecting element mobility in the Meade Peak Phosphatic Shale Member of the Phosphoria Formation Time
Holocene Neogene
Regional event
-
Meade Peak event
Result
Weathering
9 Precipitation of efflorescent salts 9 Local oxidation 9 Precipitation of native metals in reducing environments? 9 Movement (redistribution) of Se, Zn, S, P, Cd, etc. 9 Movement (redistribution) of Se, K, Si, P, Fe, S, As, etc. 9 Fracture coatings 9 Introduction of Hg ? 9 Movement (redistribution/introduction?) of P, Al, Se, Ni, I, Ag, etc.
Extension and normal faulting
Fracturing
Early Eocene Early Cretaceous
Yellowstone Hot Spot passed to the north at about 10 Ma Extension and normal faulting Thrusting
Local heating
Fracturing
Vein formation
Early Cretaceous
9 Local oxidation ? 9 Precipitation of native metals in reducing environments (timing uncertain) 9 Fluorite and bitumen in beddingparallel and x-cutting veins
Late Jurassic Catagenesis Diagenesis
Mid to Late Permian
-
Deposition
9 Buddingtonite veins? 9 At least 2 generations of bitumen veins 9 Oil and some ESE left the system 9 Movement of most elements (silicification?; multiple generations of carbonate and CFA cements; framboidal pyrite formation; buddingtonite formation 9 Accumulation of sediments and possible reworking of carbonate units and CFA pelloids
Data regarding regional events from Momper and Williams (1984); Armstrong and Oriel (1986); Fryberger and Koelmel (1986); Pierce and Morgan (1992); Kellogg et al. (1999).
220
R.I. Grauch et al.
During and immediately following deposition in the Middle Permian, Se, Zn, and a variety of other chalcophile elements were incorporated into sulfides (pyrite, sphalerite, sulvanite) that formed in the sediments and in part were included in the growing CFA pelloids. Bacterial activity caused the reduction of sulfate and led to reactions causing the formation of framboidal pyrite and Se-bearing sulfides. There are no unequivocal data that suggest native Se or selenide minerals were a major component of the early diageneticmineral assemblage. As indicated by cross-bedded layers of pelloids and compositionally mixed pelloid populations, at least some of the CFA was reworked and became part of the sedimentary package. Continued diagenesis during burial resulted in multiple stages of carbonate, silicate, CFA, and sulfide growth as well as on-going maturation of organic matter. To what extent the ESE were incorporated in the original and evolving organic matter is unknown. It is unknown to what extent, if any, sources other than seawater, such as submarine seeps, contributed ESE to the organic-rich sediments. Catagenesis began as burial continued to a depth of about 8.5 km during the Late Jurassic-Early Cretaceous. Oil, containing unknown amounts of ESE such as Se, U, Ni, and V, which often occur in crude oil (Filby, 1994), left the Meade Peak system along with associated formation water that also contained unknown amounts of ESE. Secondary porosity probably developed within the Meade Peak as a result. The flow of oil and saline water was probably constrained by fracture systems associated with thrusting (Fig. 8-15; see Evans, Chapter 6) Fluids flowing near the soles of the thrust plates would have had access to most of the pre-Cretaceous sedimentary rocks. As those fluids moved through the plates, they may have leached metals from the rocks or deposited metals where reductants were encountered. At least two sets of bitumen veins formed in the Meade Peak during that period of tectonic activity, catagenesis, and fluid migration. Uranium and Se are either trace constituents of the bitumen or they are entrained as very small mineral inclusions. Fluorite +_ calcite +_ barite ___quartz and fluorite +_ calcite ___bitumen veins and replacements in the Meade Peak post-date the onset of catagenesis and may mark the waning stages of that period of activity. Alternatively, they may represent a later period of fluid movement possibly associated with Neogene extension. Cenozoic tectonic events affecting southeast Idaho added pathways for fluid flow, provided heat that would have increased circulation of fluids, and may have supplied additional metals. Post-Paleocene catagenesis of Cretaceous sedimentary rocks may have provided fluids involved in the formation of sedimentary rock-hosted Cu-Ag-Pb-Zn deposits (Fig. 8-15). Alternatively, oxidizing fluids moving within the thrust belt may have mobilized and transported the metals to reducing environments. Copper deposits hosted in sediments of presumed Triassic age in the Montpelier District (Gale, 1910) suggest that these ore-forming fluids moved within the Meade thrust sheet. The widespread presence of these metal deposits (Fig. 8-15) indicates that fluids moved throughout the thrust belt. Neogene through Holocene extension and normal faulting created the basin and range topography of southeast Idaho as well as additional fluid pathways within and across thrust plates. An increased regional thermal gradient associated with the passage of the Yellowstone hotspot may have driven movement of fluids within the thrust belt. As mentioned above, the F-bearing vein assemblages and replacements may have formed at that
Petrogenesis and mineralogic residence of selected elements
221
Fig. 8-15. Map and cross-section of southeast Idaho and southwest Wyoming showing location of metal occurrences/deposits, oil fields, and major thrust sheets. Locations of deposits and oil fields on the cross-section were projected more or less along strike but were constrained to stay within their host thrust sheet. Figure 8-15(A) map modified from Loose (1990); Blackstone and De Bruin (1987); Powers (1993); Fig. 8-15(B) cross section modified from Webel (1987).
222
R.I. Grauch et al.
time and it is possible that some of the other chemical and mineralogical changes in the Meade Peak, such as the concentration of Hg took place during that time interval. Quaternary processes affecting the Meade Peak are surficial weathering and the downward transport of ESE and other elements that are susceptible to solution and redeposition. This cycle of solution-precipitation was repeated multiple times in response to changing climatic conditions, changes in water-table level, and changes in local rock/water chemistry (such as depletion of reductant). The net effect was to deplete the more soluble elements from surficial, topographically high areas and concentrate some elements near the water table. The geometry of this supergene enrichment zone is often irregular and can have root-like downward extensions that follow joints and fractures. A similar, but more extensive supergene enrichment process is used to explain ore deposits related to porphyry Cu systems such as those at Bingham, Utah, and Miami, Arizona. Supergene enrichment explains many of our observations, especially the small-scale variations in the distribution of several of the redox sensitive ESE and the replacement textures of their host minerals. The occurrence of native Se and other native metals in very small voids, some of which contain minor amounts of bitumen, is suggestive of metal transport in an oxidizing fluid and precipitation where an active reductant was encountered. Some of the roscoelite described above may have formed in that way. Roscoelite is a common accessory mineral in native metal-bearing reduction spheroids in a variety of low-temperature environments (Hofmann, 1990, 1993). Passage of oxidizing fluids is indicated by the presence of numerous secondary oxides and probable oxyhydroxides of Fe and Mn in the Meade Peak. The porosity required for efficient fluid transport could have been self generating as initial oxidation of sulfides produced locally acidic conditions that could have dissolved CFA and carbonates. Such a process could explain solution textures such as those illustrated in Fig. 8-5. Supergene alteration of the Meade Peak would have been limited by the acid-generating capacity and fine-grain size of the rock. Those limitations would result in irregularly distributed, nonpervasive alteration, and the concentration of alteration products on fractures. The oxidized portion of the lower-waste zone of the Meade Peak as well as the solution cavities immediately below the basal Meade Peak unconformity may also be products of supergene alteration and fluid flow.
ACKNOWLEDGMENTS Thanks are due to many colleagues. Jim Hein acted as an understanding and helpful editor. Isabelle Brownfield and Michelle Anderson were of great help with SEM and electron microprobe activities. Bob Rye very kindly provided several sulfur isotope analyses. Tom Johnson and Yoshie Hagiwara, University of Illinois, collaborated by incorporating some of our samples in a reconnaissance study of Se isotopes. Early in our studies, Steve Sutley definitively identified native Se by means of X-ray diffraction. Rhonda Driscoll provided X-ray diffraction analysis of whole thin sections and efflorescent salts. Ray Kokaly developed the ENVI-based routine used for merging and manipulating electron
Petrogenesis and mineralogic residence of selected elements
223
microprobe images. Special thanks are due Kristen Sanford for sharing work from her MS thesis research at Georgia State University (anticipated completion in late 2003). George Breit, Joel Leventhal, and David Piper provided insightful and very constructive reviews of an early version of this chapter.
REFERENCES Armstrong, EC. and Oriel, S.S., 1986. Tectonic development of the Idaho-Wyoming Thrust Belt. Am. Assoc. Pet. Geol. Mem., 49: 243-279. Atnipp, V.A., Budai, J.M., Lohmann, K.C. and Dickinson, W.R., 1987. Evidence for multiple stages of fluid migration through the Wyoming overthrust belt. Geological Society of America, 1987 annual meeting and exposition, Phoenix, AZ, United States, October 26-29, 1987, Abstr Prog.Geol. Soc. Am., 19(7): 576. Blackstone, D.L. Jr. and De Bruin, R.H., 1987. Tectonic map of the Overthrust Belt, western Wyoming, northeastern Utah, and southeastern Idaho. Geological Survey of Wyoming Map Series 23. In: W. Roger Miller (ed.), Wyoming Geological Association Thirty Eighth Field Conference Guidebook. Mtn. States Litho, Casper. Boyle, D.R., 1997. Iodargyrite as an indicator of arid climatic conditions and its association with gold-bearing glacial tills of the Chibougamau-Chapais area, Quebec. Can. Mineral., 35: 23-34. Boyle, R.W., 1961. Native zinc at Keno Hill. Can. Mineral., 6: 692-694. Chafetz, H.S. and Reid, A., 2000. Syndepositional shallow water precipitation of glauconitic minerals. Sediment. Geol., 136: 29-42. Clark, A.H. and Sillitoe, R.H., 1970. Native zinc and a-Cu, Zn from Mina Dulcinea de Llampos, Copiap6, Chile. Am. Mineral., 55:1019-1021. Claypool, G.E., Love, A.H., and Maughan, E.K., 1978. Organic geochemistry, incipient metamorphism, and oil generation in black shale members of Phosphoria Formation, western interior, United States. Am. Assoc. Pet. Geol. Bull., 62(1): 98-120. Cook, P.J., 1969. The petrology and geochemistry of the Meade Peak Member of the Phosphoria Formation. PhD thesis, University of Colorado, Boulder, CO, 204 pp. Cook, P.J., 1970. Repeated diagenetic calcitization, phosphatization, and silicification in the Phosphoria Formation. Geol. Soc. Am. Bull., 81 (7): 2107-2116. Desborough, G.A., 1977. Preliminary report on certain metals of potential economic interest in thin vanadium-rich zones in the Meade Peak Member of the Phosphoria Formation in western Wyoming and eastem Idaho. US Geological Survey, Open-File Report, 77-341, 27 pp. Desborough, G.A., Tysdal, R.G., Knudsen, A.C., Grauch, R.I., Herring, J.R. and Brownfield, I., 2001. Bulk mineralogy, micromineralogy, and authigenic minerals in the Meade Peak Phosphatic Shale Member of the Phosphoria Formation, southeastern Idaho. US Geological Survey, OpenFile Report, 01-0004, 20 pp. Fiesinger, D.W., Perkins, W.D. and Puchy, B.J., 1982. Mineralogy and petrology of Tertiary-Quaternary volcanic rocks in Caribou County, Idaho. In: B. Bonnichsen and R.M. Breckenridge (eds.), Cenozoic Geology of Idaho. Bulletin- Idaho Bureau of Mines and Geology, vol. 26, pp. 465-488. Filby, R.H., 1994. Origin and nature of trace element species in crude oils, bitumens and kerogens: implications for correlation and other geochemical studies. In: J. Parnell (ed.), Geofluids: Origin, Migration and Evolution of Fluids in Sedimentary Basins. Geological Society Special Publication, vol. 78, pp. 203-219.
224
R.I. Grauch et al.
Fryberger, S.G. and Koelmel, M.H., 1986. Rangely Field: eolial system-boundary trap in the PermoPennsylvanian Weber Sandstone of northwest Colorado. In: D.S. Stone (ed.), New Interpretations of Northwest Colorado Geology. Rocky Mountain Association of Geologists, Denver, CO, pp. 129-150. Gale, H.S.,1910. Geology of the copper deposits near Montpelier, Bear Lake County, Idaho. US Geol. Surv. Bull., 430-B, p. 112-121. Geptner, A.R. and Ivanovskaya, T.A., 2000. Glauconite from Lower Cretaceous marine terrigenous rocks of England: a concept of biogeochemic origin. Lithol. Miner. Resour., 35: 434-444. Grauch, R.I., Meeker, G.P., Desborough, G.A., Herring, J.R., Tysdal, R.G. and Johnson, E.A., 1999. Selenium residence in the Phosphoria Formation. Geol. Soc. Am. Abstr. Prog., 31 (7): A-35. Grauch, R. I., Meeker, G.P., Desborough, G.A., Driscoll, R.L., Tysdal, R.G. and Herring, J.R., 2000. Selenium and Nickel Mobility in the Phosphoria Formation, southeast Idaho. Geol. Soc. Am. Abstr. Prog., 32: A-488. Grauch, R.I., Tysdal, R.G., Johnson, E.A., Herring, J.R. and Desborough, G.A., 2001. Stratigraphic section and selected semiquantitative chemistry, Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 99-20-E, 1 plate with text. Grauch, R.I., Herring, J.R., Tysdal, R.G., Desborough, G.A. and Johnson, E.A., 2002. The Permian Phosphoria Formation, environmentally sensitive elements, and phosphate mining. l lth IAGOD Quadrennial Symposium and Geocongress, Geological Survey of Namibia, Windhoek, Namibia, CD. Gulbrandsen, R.A., 1960. Petrology of the Meade Peak Phosphatic Shale Member of the Phosphoria Formation at Coal Canyon, Wyoming- a method of X-ray analysis for determining the ratio of calcite to dolomite in mineral mixtures: US Geological Survey, Bulletin, 111 l-C, D, pp. 98-105. Hagiwara, Y., 2000. Selenium isotope ratios in marine sediments and algae - a reconnaissance study. University of Illinois at Urbana-Champaign, MS thesis, 62 pp. Herring, J.R., Desborough, G.A., Wilson, S.A., Tysdal, R.G., Grauch, R.I. and Gunter, M.E., 1999. Chemical composition of weathered and unweathered strata of the Meade Phosphatic Shale Member of the Permian Phosphoria Formation. US Geological Survey, Open-File Report, 99-147-A, 24 pp. Herring, J.R., Wilson, S.A., Stillings, L.A., Knudsen, A.C., Gunter, E., Tysdal, R.G., Grauch, R.I., Desborough, G.A. and Zielinski, R. A., 2000a. Chemical composition of weathered and lessweathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation- B. Measured sections C and D, Dry Valley area, Caribou County, Idaho. US Geological Survey, Open-File Report, 99-147-B, 33 pp. Herring, J.R., Grauch, R.I., Desborough, G.A., Wilson, S.A. and Tysdal, R.G., 2000b. Chemical composition of weathered and unweathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation- C. Measured sections E and F, Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open-File Report, 99-147-C, 35 pp. Herring, J.R., Grauch, R.I., Tysdal, R.G., Wilson, S.A. and Desborough, G.A., 2000c. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation- D. Measured sections G and H, Sage Creek area of the Webster Range, Caribou County, Idaho. US Geological Survey, Open-File Report, 99-147-D, 38 pp. Herring, J.R., Grauch, R.I., Siems, D.E, Tysdal, R.G., Johnson, E.A., Zielinski, R.A., Desborough, G.A., Knudsen, A. and Gunter, M.E., 2001. Chemical composition of strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation. Channel-composited and individual rock samples of Measured Section J and their relationship to Measured Sections A and B,
Petrogenesis and mineralogic residence of selected elements
225
central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 01-195, 72 pp. Hiatt, E.E., 1997. A paleoceanographic model for oceanic upwelling in a late Paleozoic epicontinental s e a - a chemostratigraphic analysis of the Permian Phosphoria Formation. University of Colorado PhD dissertation, Boulder, CO, 294 pp. Hofmann, B.A., 1990. Reduction spheroids from northern Switzerland: mineralogy, geochemistry and genetic models. Chem. Geol., 81:55-81. Hofmann, B. A., 1993. Organic matter associated with mineralized reduction spots in red beds. In: J. Parnell, H. Kucha and P. Landais (eds.), Bitumens in Ore Deposits. Society for Geology Applied to Mineral Deposits Special Publication, vol. 9, pp. 362-378. Jahnke, R.A., 1984. The synthesis and solubility of carbonate fluorapatite. Am. J. Sci., 284: 58-78. Kellogg, K.S., Rodgers, D.W., Hladky, ER., Kiessling, M.K. and Riesterer, J.W., 1999. The Putnam Thrust Plate- dismemberment and tilting by Tertiary normal faults. In: S.S. Hughes and G.D. Thackray (eds.), Guidebook to the Geology of Eastern Idaho. Idaho Museum of Natural History, Pocatello, pp. 97-114. Labrenz, M., Druschel, G.K., Thonsen-Ebert, T., Gilbert, B., Welch, S.A., Kemmer, K.M., Logan, G.A., Summons, R.E., De Stasio, G., Bond, P.L., Lai, B., Kelly, S.D., and Banfield, J.E, 2000. Formation of sphalerite (ZnS) deposits in natural biofilms of sulfate-reducing bacteria. Science, 290:1744-1747. Loose, S.A., 1990. Ore mineralogy, paragenesis, source of metals, and timing of mineralization of the Lake Alice Cu-Ag-Pb-Zn district in the Wyoming thrust belt: Wyoming sedimentation and tectonics. Wyoming Geological Association forty-first field conference; Casper, WY, United States, August 17-23, 1990, Guidebook - Wyoming Geological Association, vol. 41, pp. 123-149. Mabie, C.P. and Hess, H.D., 1964. Petrographic study and classification of western phosphate. US Bureau of Mines, Report of Investigations, 6468, 95 pp. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M., and Swanson, R.W., 1959. The Phosphoria, Park City and Shedhorn Formations in the western phosphate field. US Geological Survey, Prof. Paper, 313-A, 47 pp. Momper, J.A. and Williams, J.A., 1984. Geochemical exploration in the Powder River basin: Petroleum geochemistry and basin evaluation. Am. Assoc. Pet. Geol. Memoir, 35:181-191. Munkers, J.P., 2000. Abiotic and biotic processes in the release and control of selenium in the western phosphate resource area. MS thesis, University of Idaho, Moscow, ID, 123 pp. Perkins, R.B. and Foster, A.L., 2002. Mineral affinities and distribution of selenium in phosphatic shales of the Meade Peak Member of the Permian Phosphoria Formation. Geol. Soc. Am. Abstr Prog., 34(6): 311. Pierce, K.L. and Morgan, L.A., 1992. The track of the Yellowstone hot spot; volcanism, faulting, and uplift. In: P.K. Link, M.A. Kuntz and L.B. Platt (eds.), Regional Geology of Eastern Idaho and Western Wyoming. Geol. Soc. Am. Memoir, vol. 179, pp. 1-53. Powers, R.B., 1993. Wyoming-Utah-Idaho thrust belt province (090): Petroleum exploration plays and resource estimates, 1989, onshore United States; Region 3, Colorado Plateau and Basin and Range. US Geological Survey, Open-File Report, 93-0248, pp. 74-92. Ramseyer, K., Diamond, L.W., and Boles, J.R., 1993. Authigenic K-NH4-feldspar in sandstones: a fingerprint of the diagenesis of organic matter, J. Sed. Pet., 63: 1092-1099. Royse, F. Jr., 1993. An overview of the geologic structure of the thrust belt in Wyoming, northern Utah, and eastern Idaho. In: A.W. Snoke, J.R. Steidtmann and S.M. Roberts (eds.), Geology of Wyoming, M e m o i r - Geological Survey of Wyoming, Laramie, WY, vol. 5, pp. 272-311.
226
R.I. Grauch et al.
Sheldon, R.E, 1967. Long-distance migration of oil in Wyoming. Mountain Geol., 4: 53-65. Tysdal, R.G., Johnson, E.A., Herring, J.R., and Desborough, G.A., 1999. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 99-20-A, 1 plate with text. Tysdal, R.G., Herring, J.R., Desborough, G.A., Grauch, R.I., and Stillings, L.A., 2000a. Stratigraphic sections of the Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Dry Valley area, Caribou County, Idaho. US Geological Survey, Open-File Report, 99-20-B, 1 plate with text. Tysdal, R.G., Grauch, R.I., Desborough, G.A., and Herring, J.R., 2000b. Stratigraphic sections of the Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, east-central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open-File Report, 99-20-C, 1 plate with text. Tysdal, R.G., Herring, J.R., Grauch, R.I., Desborough, G.A., and Johnson, E.A., 2000c. Stratigraphic sections of the Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Sage Creek area of Webster Range, Caribou County, Idaho. US Geological Survey, Open-File Report, 99-20-D, 1 plate with text. Warner, M.A., 1982. Source and time of generation of hydrocarbons in the Fossil Basin, western Wyoming thrust belt. In: R.B. Powers (ed.), Geologic Studies of the Cordilleran Thrust Belt. Rocky Mountain Association of Geologists, Denver, CO, pp. 805-815. Webel, S., 1987. Significance of backthrusting in the Rocky Mountain thrust belt. In: W. Roger Miller (ed.), The Thrust Belt Revisited. Guidebook- Wyoming Geological Association, Casper, vol. 38, pp. 37-53.
Life Cycle of the Phosphoria Formation." From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
227
Chapter 9
WEATHERING OF THE MEADE PEAK PHOSPHATIC SHALE MEMBER, PHOSPHORIA FORMATION: OBSERVATIONS BASED ON URANIUM AND ITS DECAY PRODUCTS
R.A. ZIELINSKI, J.R. BUDAHN, R.I. GRAUCH, J.B. PACES AND K.R. SIMMONS
ABSTRACT Variably weathered outcrop samples of the Meade Peak Phosphatic Shale Member of the Phosphoria Formation have 5-10% of the contained uranium (U) in a form readily extractable by 0.1 M sodium bicarbonate. Fission track radiography of outcrop samples and other less-weathered channel and core samples indicate that this mobile fraction of U is likely hosted by organic matter, secondary iron oxides and clay minerals, trace uraninite, and very fine-grained apatite cement. During weathering, this extractable U fraction is especially susceptible to redistribution, which produces small but measurable departures (1-15%) from radioactive (secular) equilibrium in the 238U decay-series. The most weathered samples show the strongest isotopic evidence for redistribution of U during the last 350 ka, but sequestration of U by alteration products limits open-system losses of U at the whole-rock scale. In less-weathered samples, isotopic evidence for minor U loss (or gain) over longer time periods (< 1 Ma) is consistent with relatively non-aggressive attack of phosphatic rock during weathering. Comparative extractability of selenium (Se) suggests that a larger fraction of Se (19%) is readily available for mobilization during the earliest stages of weathering.
INTRODUCTION Uranium (U) is a trace element of interest and concern because of its value as an energy source and its contribution to radioactive waste and environmental radioactivity. Understanding the behavior of U during ore or rock formation and its mobility during alteration and weathering are critical for guiding exploration, planning waste disposal, and predicting the movement of U and geochemically analogous elements in the environment. Earliest studies of U in phosphatic rocks were driven by the need to assess ancient phosphate deposits as low-grade U resources and sources of byproduct U (Thompson, 1954; McKelvey, 1956; Swanson, 1970). More recently, studies of U in phosphatic rocks have focused on Holocene and late Pleistocene phosphorites of the seafloor. Uranium decay-series studies of
228
R.A. Zielinsla" et al.
seafloor phosphorites evaluated the closed-system behavior of U and its daughters and provided estimates of the rate of nodule growth and the age of phosphorite formation (see review by Burnett and Veeh, 1992). Other studies attempted to define conditions of marine phosphorite formation and early diagenesis based on the microdistribution, speciation, and crystallographic residence of U (see review by Jarvis et al., 1994). Uranium decay-series measurements of older (Miocene-Pliocene) phosphatic rocks from Florida documented the extent of mobility of U and its daughters during late Pleistocene-Holocene weathering (Burner et al., 1988). This study uses measurements of U and its decay products to describe the leachability, microdistribution, and recent open-system mobility of U in variably weathered rocks of the Meade Peak Phosphatic Shale Member. Observations based on selective extractions, fission-track radiography, and U decay-series measurements build upon earlier observations of the mode of occurrence of U in the Phosphoria Formation (Thompson, 1954; McKelvey and Carswell, 1956; Sheldon, 1959; Gulbrandsen, 1966; Swanson, 1970). The focus on U provides a basis for comparisons with the behavior of other trace elements such as Se, a major subject of this volume. The results of this study are intended to illustrate the types of findings that are possible from similar but more comprehensive investigations of U behavior during weathering of phosphatic rocks. Although the focus of this study is on U mobility during weathering, the methods employed are of interest to exploration geochemists because they are universally applicable for the evaluation of any U source rock. Additional uses of U decay-series disequilibria for U-deposit exploration and description are described in Levinson et al. (1982) and Dickson and Wheller (1992).
SAMPLE COLLECTION AND DESCRIPTION Seventeen channel samples (homogenized composites) from measured sections A and B (Enoch Valley mine, Tysdal et al., 1999), and sections C and D (Dry Valley mine, Tysdal et al., 2000) were selected from a more complete archived collection of channel samples from sections of the Meade Peak Phosphatic Shale Member. The paired measured sections from each mine cover the same stratigraphic interval but are sampled at different depths below the pre-mining surface (Herring et al., 1999a,b). Sections A and C are shallower (by 30 and 46 m) than their respective pairings and are presumed to be more weathered. The samples were selected to span a large range of phosphorous concentration (1.1-16.0 wt. %) and organic-carbon content (0.2-35%; Table 9-1). Six outcrop samples representing progressively more weathered phosphorite (1, 3, 5) and closely adjacent black mudstone (2, 4, 6) were collected from the Rasmussen Ridge mine (Lee, 2001). The three pairs were collected from nearly vertical beds at three elevations (14, 71, and 80 m) above the pit floor. Access was along existing benches cut in the pit wall. The lack of continuous exposure, the contortion of beds, and the earthy texture of the shallow, most weathered pair of samples (5, 6) hindered the tracing of. individual beds. The paired samples of phosphorite and underlying mudstone were collected consistently at
Weathering of the Meade Peak Phosphatic Shale Member: Uranium study
229
TABLE 9-I Channel samples I of Meade Peak Phosphatic Shale Member selected for this study Sample2
Unit
Lithology
P
Carbonate C
Organic C
StotaI
Fetotal
(wt.%)
(wt.%)
(wt.%)
(wt.%) (wt.%)
WPSA-06 -30 -62 - 158
LO LO MW UO
Phosphorite Phosphorite Carbon seam Phosphorite
15.4 13.2 1.9 16.0
0.4 0.3 <0.1 0.4
1.2 2.3 35.0 1.4
0.52 0.37 3.95 0.23
0.95 0.65 1.16 0.30
WPSB-08 -26 -47 -131
LO LO LO UO
Phosphorite Phosphorite Phosphorite Phosphorite
14.5 13.3 10.1 12.7
0.5 0.8 1.1 0.3
2.5 2.4 5.3 12.0
0.77 0.63 0.70 1.15
0.33 0.49 0.82 0.45
WPSC- 15 -135
LO UO
Phosphorite Phosphorite
14.2 10.7
0.4 0.3
0.7 3.4
0.83 0.41
0.55 0.92
WPSD-0.5 -09 -27 -98 -126 -132 -205
LO LO LO MW MW MW UW
Phosphorite Mudstone Phosphorite Phosphorite Carbon seam Phosphorite Mudstone
14.6 5.4 12.4 6.8 9.7 12.7 1.1
1.0 0.3 0.4 0.2 0.2 0.2 0.1
0.2 1.8 1.6 18.9 5.1 1.8 4.9
0.65 0.63 0.41 4.02 4.01 2.38 3.38
0.13 1.72
0.73 1.63
2.77 1.54
2.39
IData from Herring et al. (1999a,b). 2Sample numbers indicate approximate stratigraphic level (in feet) above the base of the Meade Peak Phosphatic Shale Member. Abbreviations: LO, lower ore; UO, upper ore; MW, middle waste; UW, upper waste. the same relative depths below a traceable marker bed of limestone. The samples ranged from being relatively competent and texturally distinct near the pit floor (1, 2) to earthy and indistinct near the ridge crest (5, 6). Three core samples (WPSJ-8.4, 8.8, and 56.0) were obtained from measured section J, a drill core collected prior to development of the Enoch Valley mine (Grauch et al., 2001; Herring et al., 2001). Based on the greater below-ground depth of the J core and its collection prior to mining, it is considered to be the least weathered of the measured sections. The three core samples were included in a more detailed mineralogical investigation by Grauch et al. (Chapter 8) and were selected on the basis of lithology and trace-metal content, the latter based on analysis of nearby channel or individual samples (Herring et al., 2001). Two phosphorite samples (WPSJ-8.4 and 8.8) located near the base of the core were of particular interest because of unusually high estimated content of U ( > 4 0 0 ppm).
230
R.A. ZielinsM et al.
Three surface-water samples were collected from Angus Creek, a first-order stream of approximately 12 km length that is impacted in its upper reach by drainage from waste rock of phosphate mining (Herring and Amacher, 2001; Stillings and Amacher, Chapter 17). Water samples were collected in April 2002 during a period when surface flow was a mixture of groundwater return and early spring snowmelt. Sample AC-2 is from a seep at the base of a large waste-rock pile located at the head of the drainage. Sample AC-4 is from the outlet of a small reservoir that collects and integrates all flow from seeps, springs, runoff, and wetlands within the upper 0.6 km of Angus Creek. Sample AC-6 is from 0.6 km above the confluence of the creek with the Blackfoot River. The samples were chosen to display a possible large range of dissolved U concentration and U isotopic composition influenced by water-rock interactions in weathered waste rock from mining, and weathered local bedrock and soils. In addition to the Phosphoria Formation, major bedrock units in the Angus Creek drainage are the overlying Lower Triassic Dinwoody Formation and the underlying Grandeur Tongue of the Lower Permian Park City Formation.
ANALYTICAL METHODS G a m m a - r a y spectrometry
Activity concentrations of 238U, 226Ra, and 21~ in the 17 channel samples and the six outcrop samples were obtained by high-precision gamma-ray spectrometry. A known weight (~100g) of each dry and finely ground (<150txm) sample was packaged in a clear acrylic petri dish, which was sealed with black electrical tape to inhibit radon (Rn) loss. Samples were stored for three weeks to establish radioactive equilibrium between 226Ra and its relatively short-lived daughters in the 238U decay series (Fig. 9-1). At radioactive equilibrium, the disintegration rate (activity) of parent and daughter are equal and the more easily measured gamma-ray emissions of daughters such as 214pb and 214Bi can be used to determine the activity of their parent 226Ra. Similarly, the activity of 234Th (half-life = 24.1 days), the immediate daughter of 238U (Fig. 9-1), is used as a proxy for 238U, assuming previous establishment of equilibrium of this parent-daughter pair in the rock sample. Gamma-ray emissions were detected over a low-energy range (0-400 kiloelectron-volt (keV)) and a high-energy range (150-2000 keV) using two solid-state germanium detectors. Photopeak resolution (peak width at half peak height) on the low-energy detector varied from 0.4 keV at 46.5 keV gamma-ray energy to 1.0 keV at 352 keV. For the high-energy detector, the photopeak resolution was 1.2 keV at 352 keV gamma-ray energy and 1.8 keV at 1332 keV. Gamma-ray spectra were accumulated for periods ranging from 24 to 72 h in order to limit errors from counting statistics to less than 3% (one sigma). The activity of 238U was determined using two low-energy gamma-ray emissions of its 234Th daughter at 63 and 93 keV. 226Ra was determined using gamma rays from daughters 214pb (295,352 keV) and 214Bi (609 keV). The activity of 21~ was determined directly using its gamma-ray emission at 46.5 keV. Integrated photopeak areas in the gamma-ray spectra (as counts-per-minute (cpm)
231
Weathering of the Meade Peak Phosphatic Shale Member: Uranium study
U
238U
234U
4.47x109 years
92 Pa
2.48x105 years 1.1~}min
'234pa ~
91
6.7h
Th 90
234Th 24.1 days
~r
23~ 7.52 x 109 years
Ac
i 1~
89 Ra
226Ra
1600 years 88 Fr 87 Rn
222Rn 3.825days
86 At
I (X
85 Po
i i
218po
214pO
3.05 min
210p0
1.6x 10-4 s
84
138.4 days
f
f
Bi
19.7 min 83 Pb
f
214pb / 13 26.8 min
82 i
o~
21~
5.01days
,r
/ 13
22.6 years
o~
2~ (stable lead isotope)
Fig. 9-1. The 238Udecay series showing major decay products and their half-lives, modified from Friedlander et al. (1964). Radioisotopes discussed in this study are highlighted.
232
R.A. Zielinski et al.
minus background) were converted to disintegrations-per-minute (dpm) after applying corrections for branching ratio, energy-dependent detector efficiency, and self-absorption (Zielinski and Budahn, 1998).
Fission-track radiography Polished thin sections of core and outcrop samples were irradiated with neutrons (2.5 x 1012neutrons cm -2 s -i) in the US Geological Survey TRIGA research reactor to induce fission of the 235U isotope. During irradiation, fission fragments recoil from the surface of the thin section and pass into a sheet of mica directly overlying the thin section. The passage of fission fragments causes linear paths of damage (tracks) in the mica that are made more visible by subsequent etching of the recovered mica with 48% hydrofluoric acid. Areas of highest track density, which are readily observed by optical microscopy, correspond to areas of high U concentration in the original sample. Depending on the angle of incidence, fission-track lengths in the mica are < 1-5 p~m long. Submicrometersized uraniferous grains are recorded in the mica as point sources from which many fission tracks radiate. The length of fission tracks limits the fine-scale spatial resolution of individual point sources. More subtle gradients in homogeneously dispersed U that occur over large areas appear as gradients in fission-track density. The duration of neutron irradiation or the neutron flux can be adjusted to determine U distribution in most natural materials containing 1-1000 ppm U.
Selective extraction experiments For each of the six outcrop samples from Rasmussen Ridge, 2 g of crushed sample was mixed with 100 ml of 0.1 M NaHCO3, pH = 8.3. The mixture was stirred continuously at room temperature for 24 h and was then filtered through a cellulose acetate membrane of 0.45 I~m opening. The filtered solution was acidified to pH < 2 with ultrapure nitric acid. The sodium bicarbonate solution was chosen to selectively remove U that is adsorbed or is in readily exchangeable sites while producing minimal dissolution of mineral components. Dissolved U is stabilized as a carbonate complex. Other trace elements that are present as oxyanions in alkaline solution (Se, As, Mo, V, Cr) are also potentially mobilized by this solvent. Extracted percentages were computed using final solution compositions determined in this study and rock compositions provided by M. Lewan (USGS, unpublished data, 2001).
Mass spectrometry Three channel samples and the six outcrop samples were analyzed for 238U,234U,232Th, and 23~ by thermal ionization mass spectrometry (TIMS). A known sample weight
Weathering of the Meade Peak Phosphatic Shale Member." Uranium study
233
(8-26mg) was spiked with known amounts of a mixed isotopic tracer solution (236U, 229Th) and dissolved in a mixture of strong acids (HC1, HNO3, HC104) at subboiling temperatures. The equilibrated and spiked solution was evaporated to dryness and dissolved in warm ultrapure 7 N HNO3. Both U and Th were separated and purified using anion-exchange resin (AG 1 X 8) with HNO3 and HC1 eluants. The purified U and Th salts were analyzed using a Finnigan MAT-262 mass spectrometer equipped with a secondaryelectron multiplier operating in ion-counting mode. Uranium was loaded on the evaporation side of a double Re-filament assembly whereas Th was loaded on single Re filaments sandwiched between layers of colloidal graphite (Arden and Gale, 1974). Total procedural blanks are less than 50 pg U and less than 100pg Th. Measured atomic ratios were corrected for contributions from the blank and the added spike, and for mass discrimination. Activity ratios were calculated from the measured mass ratios and have 2-sigma precision of approximately 0.4% for 234U/238Uand between 0.6 and 1.1% for 23~ U and Th concentrations determined by isotope dilution have 2-sigma precision of better than 0.5% for U and 0.8% for Th. Extracts of three outcrop samples (1, 2, 6) and the three surface water samples from the Angus Creek drainage were analyzed for 234U/238U.A mixed spike of 233U and 236U was used. Methods for U separation and purification were the same as above. The purified U was loaded on a single Re filament doped with colloidal graphite and analyzed on an automated VG-54E mass spectrometer with an analog Daly detector. Total procedural blank for U was 5 pg and analytical precision of the measured mass ratios was better than 0.4% (2-sigma).
Other analyses Fission-track radiographs of two U-rich J core samples (WPSJ-8.4 and 8.8) were used to guide additional characterization of U hotspots by scanning electron microscope (SEM) and energy dispersive X-ray analysis (EDXRF), according to methods described in Grauch et al. (Chapter 8). Additional multi-element analyses of extracts were performed by inductively coupled plasma-mass spectrometry and hydride-generation atomic absorption (for Se), and have an estimated 1-sigma analytical precision of better than 10%.
RESULTS AND DISCUSSION
Amount o f extractable uranium and comparison to other elements Extracted percentages of U in the six variably weathered outcrop samples (1-6) range from 4.2 to 9.9% and do not correlate positively with degree of weathering (Table 9-11). The highest extracted percentage is from relatively fresh organic-rich mudstone (sample 2), and the three highest percentages are from samples 2, 3, and 4 of slight-to-moderate weathering. Similar (within a few percent) fractions of extractable U in these variably weathered
R.A. Zielinsla" et al.
234 Table 9-II
Percentages of uranium and other trace elements extracted from variably weathered outcrop samples of Meade Peak Phophatic Shale Member by 0.1 M sodium bicarbonate solution (pH = 8.3) Sample 1 2 3 4 5 6
Weathering
Mo
Se
Sb
As
U
V
Cr
Slight Slight Moderate Moderate High High
42 27 12.4 3.0 nd 2.0
19 3.5 1.0 1.5 1.3 1.2
13.5 8.8 2.1 <1 1.2 <1
7.0 3.9 1.8 2.5 1.7 1.6
5.0 9.9 7.2 6.7 4.2 5.8
3.5 2.5 1.0 2.9 <1 <1
<0.1 <0.1 <0.1 <0.1 <0.1 <0.1
Other trace elements analyzed for but extracted at < 1% included Be, Sc, Ti, Mn, Co, Ni, Cu, Zn, Ga, Li, Rb, Sr, Y, Zr, Nb, Cd, Ag, Cs, Ba, La, Ce, T1, Bi, Th. nd, not determined. samples suggests that chemically "labile" U in these rocks is not quickly depleted during incipient weathering, but is made available more gradually during prolonged weathering. Relatively slow release of U is consistent with slow dissolution of a carbonate fluorapatite host. The two most-weathered outcrop samples (5 and 6) from shallowest depth had 4.2 and 5.8% of extractable U, compared to only slightly higher percentages of extractable U (5.0, 6.7, 7.2, and 9.9) in less-weathered samples. This suggests that only several percent of original U is lost from these samples during progressive weathering. If only 5-10% of U is readily extractable and thus available for redistribution during various stages of phosphorite weathering, then only minor U-series disequilibria are predicted in these outcrop samples. Percentages of extracted Se, As, Mo, and Sb are greatest from least weathered samples 1 and 2, which also contain the highest concentrations of these elements (M. Lewan, USGS, unpublished data, 200 l). Elements such as Se that are extracted more readily than U from relatively unweathered phosphorite (sample l) probably have a greater initial association with phases other than apatite (Grauch et al., Chapter 8; Perkins and Foster, Chapter 10).
Microdistribution o f uranium in phosphorite The distribution of U in uraniferous phosphorite collected near the base of the J core section (WPSJ-8.8) is strongly influenced by the type of pellet and the degree and type of cementation (Fig. 9-2). The vast majority of phosphate pellets have a uniform distribution of U that is similar to the rather homogeneous distribution of U reported by Burnett and Veeh (1977) for the cryptocrystalline apatite matrix in marine phosphate. The pellets show no evidence of the pronounced concentration gradients described by Altschuler et al. (1958) for larger weathered and leached pebbles in Florida phosphate. Most pellets have similar U concentration as indicated by the density of fission tracks and the corresponding
Weathering of the Meade Peak Phosphatic Shale Member:Uranium study
235
Fig. 9-2. Fission-track radiograph of a thin section of highly uraniferous phosphorite (WPSJ-8.8) collected near the base of the J core, Enoch Valley mine. Long dimension of view = 4 cm. Increasing shades of gray correspond to higher densities of fission tracks and higher concentrations of U. intensity of darkening on the radiograph image. A small subset of pellets of lower U content (lighter color) is suggestive of some minor reworking and introduction of allochthonous pellets. An even smaller number of pellets have irregular, low-U overgrowths on high-U cores. Approximately half of the image appears noticeably darkened by the presence of uraniferous finer-grained apatite cement. Clear areas on the radiograph correspond to fractures or stringers of low-U carbonate. Previous petrographic descriptions of phosphorite from the Meade Peak identify several types of phosphate pellets and
236
R.A. Zielinsla" et al.
variable contributions of fine-grained phosphatic matrix and cementation (Cressman and Swanson, 1964; Cook, 1969), all of which should produce variability in U distribution. The heterogeneous distribution of U at a variety of scales confounds attempts to evaluate chemical effects of weathering based on an assumed uniform initial composition of traceable beds. Thin-section-scale radiographs of the six outcrop samples show highly variable distributions of U (Fig. 9-3). The strong association of U with apatite and the clear spatial
Fig. 9-3. Fission-track radiographs of thin sections of variably weathered outcrop samples 1-6 from the Rasmussen Ridge mine. Long dimension of views - 4 cm. Increasing shades of gray correspond to higher densities of fission tracks and therefore higher concentrations of U.
Weathering of the Meade Peak Phosphatic Shale Member: Uranium study
237
resolution of fission-track images provides detailed views of the distribution of apatite in these fine-grained rocks. Dramatic differences in U (and apatite) distribution suggest either that the stratigraphically equivalent(?) phosphorite bed samples (1, 3, 5) and mudstone samples (2, 4, 6) are highly variable in U distribution, or that individual beds were not accurately traced during sampling. In the most weathered samples, the preservation of finely laminated beds of contrasting U concentration (sample 5) suggests minimal redistribution of the apatite host. Likewise the sharply defined contacts between areas of high-U phosphorite and low-U interbeds (sample 6) suggest minimal leaching of apatite-hosted U. Radiographs observed at higher magnification revealed other hosts of U and provide evidence of the redistribution of U in alteration products and recrystallization of phosphate (Fig. 9-4). Stringers of organic matter (photo 4B) were sought because of the well-known association of organic matter and U. In all samples, U concentration in organic matter was at least an order of magnitude less than in the surrounding apatite, based on the relative density of fission tracks. A few of the larger hot spots of U in mudstone interlayers (examples in photo 4D) correspond to grains of detrital zircon. Fracture fillings of red secondary minerals containing modest concentrations of U were observed in the most-weathered outcrop samples 5 and 6 (photos 4C and D). Analysis of these fillings with EDXRF indicated major amounts ofAl, Si, and O, and lesser amounts of Fe, P, and Ca. The fillings are tentatively identified as mixtures of iron oxides and clay minerals with some very fine occluded grains of apatite. The radiograph of core sample WPSJ-56 also indicated an unusual collection of larger-than-average phosphate pellets that appear to line up along a contact with an adjacent carbonate-rich layer (photo 4A). This association suggests phosphate recrystallization (Ostwald ripening), perhaps aided by counter-diffusion of dissolved PO 4 (from phosphorite) and Ca (from carbonate) (Chai, 1974). Recrystallization that exchanged some U most probably occurred during early diagenesis or during deepest burial (8-9 km) of these fine-grained phosphate rocks during Late Cretaceous-Early Cenozoic time (Claypool et al., 1977). Returning to the radiograph of highly uraniferous-phosphorite sample WPSJ-8.8, closer inspection revealed some areas within and between pellets that contain scattered, very intense U hotspots. SEM observation and EDXRF analyses of these areas confirmed detectable (> several weight percent) concentrations of uranium oxide (uraninite?) residing in cavities in the pellets and in the matrix between pellets (Fig. 9-5). Uraninite has been described in seafloor phosphorite (Baturin, 1982) indicating a possible early diagenetic origin. However, for Permian-age rocks of complex thermal and burial history, other possibilities for post-diagenetic precipitation of uraninite and sulfide minerals exist (Grauch et al., Chapter 8). The preservation of easily oxidized uraninite in the deepest part of the J core is consistent with minimal weathering. The presence of uraninite also indicates reducing conditions of Eh < 0.1 V during formation (Langmuir, 1978, 1997), but the rare occurrence of uraninite in samples of this study suggests that such reducing conditions may be uncommon. Other studies based on the abundance and estimated accumulation rates of seawater-derived trace elements in the Phosphoria Formation suggest that early deposition and diagenesis generally occurred under suboxic, denitrifying conditions (Piper and Medrano, 1994). Rare occurrences of uranium(VI) phosphates reported in the Phosphoria
238
R.A. Zielinski et al.
Fig. 9-4. Photomicrographs (left) and corresponding fission-track radiographs (fight) showing U distribution in small areas of selected samples. Darker areas of the fission-track images correspond to higher densities of fission tracks and therefore higher concentrations of U. (A) Contact of high-U phosphorite (upper left) and carbonate in core sample WPSJ-56; long dimension = 2.5 mm. (B) Black organic stringer in channel sample WPSB-133; long dimension - 2.5 mm. (C) Fracture fill in outcrop sample 5; long dimension = 1.0mm. (D) Fracture fill in outcrop sample 6; long dimension - 1.0mm.
Weathering of the Meade Peak Phosphatic Shale Member." Uranium study
239
Fig. 9-5. (A) Magnified view of a small area of the fission-track radiograph of core sample WPSJ8.8 (Fig. 9-2); inter-pelletal hotspots of high U are indicated by dark-colored spots of high fissiontrack density. SEM image (B) of the same area in backscatter mode showing bright spots of high average atomic number. Close-up (C) showing more detail of some bright spots and their chemistry. EDXRF spectrum (D) of one bright spot indicating the presence of major U (~20wt. % UO3). Abbreviations: Ap, apatite, Kfs, potassium feldspar.
240
R.A. Zielinski et al.
Formation (McKelvey and Carswell, 1956) may be the result of oxidative weathering of a similar interval containing both apatite and uraninite (Jerden and Sinha, 2003).
Disequilibria in the
238Udecay-series determined by gamma-ray spectrometry
Activity concentrations of 238U measured in the 17 channel samples (Fig. 9-6A) and six outcrop samples (Fig. 9-6B) vary from less than 5 dpm/g to almost 135 dpm/g. For reference, the activities of 238U in dpm/g can be converted to ppm 238U by multiplying by 1.340. This multiplication factor is based on conversion factors of 2.22 dpm/picocurie and 0.336pCi/l~g of 238U. On these plots, radioactive equilibrium between 238U and daughters 226Ra and 21~ is indicated by horizontal tie lines indicating equal activities. Deviations from equal activity conditions are uncommon and small, indicating that most samples approximate closed systems with respect to disturbances within the 238U decay series. This result supports previously reported results for Phosphoria Formation rocks that were based on much less precise laboratory determinations and comparisons of chemical U and radioactivity-equivalent U (eU) (Sheldon, 1959; Swanson, 1970). The present data provide a much more rigorous test of equilibrium because U and daughter abundances are determined by a single more precise method on the same exact bulk sample. Minor 238U-series disequilibrium in phosphorite is consistent with the premise that the majority of U (and therefore U-decay products) is contained in relatively insoluble carbonate fluorapatite (Altschuler et al., 1958; Altschuler, 1973; Perkins and Foster, Chapter 10). Minor 238U-series disequilibrium indicates that differential mobilization of 23Su and its chemically distinct decay products has not been large or rapid within time scales dictated by the half-lives of the decay products. The longest-lived decay product is 234U(half life = 248 ka). Any U-series disequilibrium produced in a chemically open system is subject to self correction through either radioactive decay of excess daughter or ingrowth of deficient daughter. For example, barring further disturbance, initial partial disequilibrium between 226Ra (half life = 1.6ka) and its progeny 21~ will be corrected after approximately 4-5 half lives of the shorter-lived 21~ (half life -- 22.6 years). If disequilibrium is observed, then some open-system mobility of 226Ra, 21~ or an intermediate daughter such as 222Rn (Fig. 1) is operative within a similar time frame. Disequilibria within the upper half of the 238U-decay series are preserved over longer time periods because of the involvement of longer-lived 234U or 23~ (half life = 75 ka). The precision of the gamma-ray spectrometry data permits identification of measured activities that differ by as little as 5%. Eight channel samples and three outcrop samples show possible small departures from radioactive equilibrium (Fig. 9-6A and B). Apparent disequilibria between 226Ra and 238Ucould be influenced by selective movement of 238U, 226Ra, or the intervening long-lived daughters 234U and 23~ (Fig. 9-1). Likewise, disequilibria between 226Ra and 21~ could be caused by preferential mobility of either nuclide as well as intervening 222Rn (half life = 3.8 days). Interpretations of the most likely
B 140
140
t~
U gain
#1
WPSB-08
130
130 120 t~
120 110 110 u gain
J NPSA-30
~
100 _
100
Rn gai~__.~,~.._~ WPSB-47
L
-1"
90
WPSB-131
80 ~
80
-
J, T
a
Rn loss
~ WPSC-15
13. a
#2 70
U Ioss.__._.__.__._~
70 L wPsc~,3~ "
Y
Rn loss Rn a.~
60
"~PSA-158
~ . ~ _ vv ~ : ~
50
--
-~W~S~ w~o~
50
-~PSB-26
t J aai n
40
,-
r pv~u-u.o
U Ios~
60
WPSD-
~I
""
~NPSD-
[]
-I
20
226Ra
(=230Th?)
126
#5 30 ~ 132
,a-WPSD-205
.; '
0 238U
40 t~
.
210pb
PSA-62
,
10
#3
U gain L
A w
gain
--
-
_.-----.-------4 #6
~~
Rn gain..~_.__.__._._---4 #4
0 238U
226Ra
21opb
(=230Th?)
Fig. 9-6. Comparative activity concentrations of 238U-decay series isotopes determined by gamma-ray spectrometry in (A) 17 channel samples and (B) six outcrop samples of Meade Peak Phosphatic Shale Member. Samples labeled with larger fonts show some apparent open-system behavior of U and/or Rn. Typical analytical precision of 3% (one sigma) is indicated by error bars or is approximated by the size of symbols.
to 4~
242
R.A. Zielinsla" et al.
causes of disequilibria are based on the following assumptions: (1)
(2)
(3)
Radium is assumed to be insoluble in the oxidizing weathering regime of the Meade Peak Phosphatic Shale Member. A previously analyzed seep-water sample collected from the base of a phosphate mining waste-rock pile, and thus similar to AC-2 of this study, had near-neutral pH (7.45), modest total dissolved solids (1800ppm), and sulfate (1140 ppm) as the dominant anion (J. Herring, USGS, written communication, 2001). This chemistry is unfavorable for dissolution and transport of Ra, which is most soluble in reducing waters of high salinity and chloride content (Fisher, 1998). Mineral/solution equilibium calculations using the computer code Geochemist's Workbench (Bethke, 1998) indicated that the same water had sufficient dissolved barium (20 ppb) to achieve saturation (log[ISP/Ksp] = 0.46) with barium sulfate, a well known sequestering agent for dissolved Ra (Doerner and Hoskins, 1925). Lead is assumed to be highly insoluble in the neutral pH and sulfate-rich pore waters of Meade Peak Phosphatic Shale Member. Lead sulfate is highly insoluble and the water sample discussed above had a dissolved Pb content of <0.05 ppb. Thorium is assumed to be highly insoluble in local pore fluids of weathered shale. Extreme insolubility of Th in non-acidic, oxidizing ground water and surface water is well documented (Langmuir and Herman, 1980). Dissolved Th in the water sample from Herring mentioned above was 0.5 ppb.
Assumed insolubility of Ra and Th (and therefore their isotopes) implies that the activity of 226Ra in variably weathered bulk samples should be close to that of 23~ More involved measurements of both isotopes by radiochemical separation and alpha or mass spectrometry could test this assumption. Such an analysis was reported for 11 slightly to highly weathered phosphate rocks from Florida (Burnett et al., 1988). In only four samples (only one highly weathered) did the 226Ra/23~ activity ratio (AR) show analytically significant divergence (___4%) from equilibrium and no divergence exceeded 10%. If the above assumptions hold, then disequilibria in the measured samples are largely influenced by mobility of U and Rn. Uranium that is liberated from apatite or other possible hosts such as organic matter or iron oxides should be highly soluble as U(VI) in the local oxidizing surface water or shallow ground water. Dissolved U(VI) is further stabilized in neutral to alkaline pH solutions through the formation of strong complexes with either carbonate or phosphate (Langmuir, 1978). The mine waste-rock seep mentioned above contained 40 ppb dissolved U. For comparison, surface water outside the vicinity of U deposits can contain 0.01-100ppb U, but rarely exceeds 20ppb (Osmond and Ivanovich, 1992). For further comparison, the US Environmental Protection Agency (USEPA) has recently instituted a drinking water standard for U of 30 ppb (USEPA, 2000). Radon is a chemically inert gas that, once liberated from mineral surfaces, should move freely within soil pore water or soil gas. Finely crushed and agitated samples of variably weathered phosphate rock from Florida emanated significant fractions of total generated Rn under dry conditions (average = 12.6% Rn loss) and moist conditions (average = 32.7% Rn loss) (Burnett et al., 1988). Based on the above discussion, the sloped tie lines of Figs 9-6A, B that indicate small disequilibria are assumed to be anchored by the value for relatively immobile 226Ra and are
Weathering of the Meade Peak Phosphatic Shale Member." Uranium study
243
labeled according to the inferred relative gain or loss of 238U or 222Rn. Net open-system lOSSof 222Rn during approximately the last 100 years results in a lower activity of its 21~ decay product compared to the activity of 226Ra. Samples showing an apparent excess of 21~ decay product compared to 226Ra have experienced a similarly recent, and perhaps ongoing, flux of added 222Rn. The subset of eight channel samples with analytically significant disequilibria (Fig. 9-6A) includes representatives from all four sampled sections (A-D). Seven of the eight samples are from the lower 15 m (50ft) of the sections, which corresponds to the lower-ore zone. More analyses are needed to confirm this association, but a plausible explanation based on qualitative field observations is that rocks of this interval are more subject to brittle fracturing (R. Tysdal, USGS, oral communication, 2002). Such fracturing likely occurred as the rocks were folded and faulted during the Laramide orogeny. Fractures in these highly tilted rocks provide conduits for recent open-system movement of water, soil gas, and associated dissolved U and Rn, or gaseous Rn (Ghahremani and Banks, 1983). Of the six weathered outcrop samples, two mudstones (2, 4) and one phosphorite (3) show slight (5-15%) disequilibria (Fig. 9-6B). Interestingly, the shallow most-weathered samples (5, 6) are in apparent radioactive equilibrium. Apparent minimal mobilization of U during progressive weathering implies that the apatite host is relatively unaffected. Stability of apatite during local weathering contrasts with the obvious destruction of apatite during the highly acidic lateritic weathering of Florida phosphates (Altschuler et al., 1958; Altschuler, 1973). Apparent Rn gain, such as excess 21~ daughter, in the three samples showing disequilibria suggests that Rn from depth or from laterally beneath the exposed face is actively moving through these fractured, vertical rocks.
Disequilibria in the 238U decay-series determined by mass spectrometry Activity ratios of 234U/238U and 23~ in nine samples were determined by mass spectrometry to provide additional details on the type of disequilibrium within the upper part of the 238U decay series (Table 9-Ili). Interpretation of such information can provide additional constraints on the timing of recent U movement in the samples. Three channel samples previously analyzed by gamma-ray spectrometry were selected on the basis of contrasting apparent disequilibria between 238U and daughter 226Ra. These included WPSB26 (U gain), WPSD-0.5 (U loss), and WPSC-135 (radioactive equilibrium) (Fig. 9-6A). In addition, all six of the variably weathered outcrop samples were chosen because they represented a range of weathering intensity, U concentration, and U-Ra disequilibria (Fig. 9-6B). Results for the three channel samples indicate small amounts (1-9%) of 234U enrichment compared to 238U parent (Table 9-III and Fig. 9-7). In addition, all three samples show 23~ AR > 1.0, which indicate probable loss of 238U compared to relatively insoluble 23~ Collectively, these results indicate that the samples have been open systems with respect to both U isotopes within the last several 105 to 106 years. This conclusion is based on the fact that any excess "unsupported" 234U decays with a half life of 248 ka. Based on the respective values of 23~ sample WPSD-0.5 has experienced a net loss
244
R.A. Zielinski et al.
Table 9-III Uranium-thorium isotopic analyses of Meade Peak Phosphatic Shale Member samples and uranium isotopic analysis of sodium bicarbonate extracts and surface waters Sample
Type
Th
(ppm)
(ppm)
WPSB-26 WPSC-135 WPSD-0.5
Channel Channel Channel
65 129 119
1 2 3 4 5 6
Outcrop Outcrop Outcrop Outcrop Outcrop Outcrop
193 90.3 37.5 10.8 40.3 29.6
1 2 6
Extract Extract Extract
0.0238 0.0030 0.00026
AC-2 AC-4 AC-6
Water Water Water
1.15
,
,
,
C) r-]
,
234U/238U •
U
|
|
!
!
23~
sigma
•
23~
sigma
(calc.)
2.4 3.8 1.0
1.010 1.017 1.091
• ___0.009 •
1.006 1.013 1.086
_____0.012 _____0.012 _____0.007
0.996 0.996 0.995
7.6 9.1 9.3 10.1 7.9 9.3
1.001 0.999 0.993 1.0214 1.0631 1.0560
• • 0.004 • • • •
1.011
1.008 1.006 1.013 1.139 1.114
_____0.009 • ___+0.007 ___+0.010 • ___+0.008
1.010 1.008 1.013 0.992 1.071 1.055
-
0.973 0.993 1.117
• • •
-
1.263 1.547 2.195
• • __+0.002
I
'
Channel Outcrop
'
i
w
I
'
i
,
Equiline
' 7
/
234U/23OTh=1.0/ / /
1.1 01/
-
~
/
23 oo
/
04
D
~
1.05
D
04
Ii!-
0.95 / 0.95
/ J 7 I
I
I
I
I
1
I
I
I
I
I
1.05
I
I
I
I
1.1
i
i
I
I
1.15
23OTh/238u
Fig. 9-7. 238U-decay series disequilibria determined by thermal-ionization mass spectrometry in three channel samples and six variably weathered outcrop samples of Meade Peak Phosphatic Shale Member. The sizes of symbols approximate the two-sigma analytical uncertainty. The equiline represents radioactive equilibrium between 234Uand its immediate daughter 23~
Weathering of the Meade Peak Phosphatic Shale Member." Uranium study
245
of 238U and sample WPSC-135 a smaller net loss of 238U, in qualitative agreement with predictions based on 238U/226Ra ratios by gamma-ray spectrometry. Sample WPSB-26 shows a small net loss of 238U compared to a small apparent gain of 238U as determined by gamma-ray spectrometry (Fig. 9-6A). The disagreement may result from differences in scale; gamma-ray spectrometry measured 100 g of sample and mass spectrometry measured < 100mg. All three channel samples plot close to the equiline representing radioactive equilibrium between 234U and its immediate daughter 23~ (Fig. 9-7). This observation suggests that any open-system movement of U causing net enrichment of 234U relative to 238U has been gradual enough or old enough for 23~ daughter to maintain equilibrium with its 234U parent through ingrowth or decay. The half life of 23~ is 75 ka. In summary, the isotopic data for the three channel samples indicate that they have been subject to minor open-system movement of U (semi-continuous or episodic) within the last million years, and that movement of U is extremely slow and/or largely predates the last 350 ka (several times the 75 ka half-life of 23~ If weathering is viewed as a semi-continuous process that may precede the last million years, then the 238U-decay-series disequilibria recorded in these channel samples could characterize a type of steady-state disequilibrium condition (Latham and Schwarcz, 1987; Dequincey et al., 2002). Steady-state disequilibrium established early in the weathering process could persist to varying extents during the Quaternary. For example, the larger degree of disequilibrium in sample WPSD-0.5 relative to the other two channel samples could indicate greater openness to movements of U, but with disequilibrium still governed by very slow rates of U movement. Four of the six outcrop samples (1, 2, 3, 4) plot close to the equal-activity origin representing radioactive (secular) equilibrium and also trend closely along the equiline (Fig. 9-7). In this sense, they are similar in isotopic composition to two of the three channel samples. The remaining two samples (5, 6) are the most weathered and both plot far from the origin that represents secular equilibrium and well off the equiline (Fig. 9-7). Samples 5 and 6 show net losses of 238U compared to relatively immobile 23~ and disequilibrium excess of 23~ relative to 234U(Table 9-II1). The presence of excess, undecayed 23~ indicates that some transient, more recent influence upon U loss has acted on these nearsurface, highly weathered samples within approximately the last 350ka. Climatic variations during late Pleistocene glacial and interglacial stages could produce transient changes in the intensity of shallow weathering and mobilization of U. Despite net losses of 238U, the three channel and six outcrop samples retain a 234U/238U AR > 1.0, which is contrary to expectations for rocks that have experienced recent U removal (see below). In a study of laterite formation, Dequincey et al. (2002) proposed that samples showing both 23~ > 1.0 and 234U/238U > 1.0 experienced a net loss of U accompanied by a smaller gain of U of elevated AR. In a near-surface weathering environment, some U of high AR could be added via sorptive uptake from local surface water or shallow ground water (see below). Gamma-ray spectrometry results for 100 g aliquots of samples 5 and 6 indicated that they have very similar activities (within ___3%) of 226Ra and 238U (Fig. 9-6B), but mass spectrometry of < 100 mg aliquots showed 5.5 and 7.1% excesses of 23~ activity
246
R.A. Zielinski et al.
compared to 238U(Table 9-111). Unequal activity of 226Ra and its parent 23~ violates the previous assumption that Ra and Th are both insoluble during local weathering. Perhaps more advanced weathering of phosphate rocks can mobilize both Ra and U (Burnett et al., 1988). An alternative explanation is that in the more weathered samples recent mobilization of U occurs on a small (mg sized) scale, but on a whole-rock scale internally redistributed U is mostly retained by alteration products such as secondary iron oxides or clay minerals. Values of 234u/E38u in the three surface waters are all > 1.0 and are higher than any AR value in the nine whole-rock samples (Table 9-III). Disequilibrium excesses of 234U are common in surface water and ground water and are caused by preferential, decay-related recoil of 234Uatoms from mineral surfaces, or preferential dissolution of 234U from radiation damaged sites under conditions of natural, non-aggressive leaching (Osmond and Ivanovich, 1992). Non-aggressive dissolution of U during regional weathering is consistent with modest to low concentrations of dissolved U (3.0, 0.3 ppb) and elevated values of 234u/E38u AR (1.547, 2.195) in Angus Creek waters AC-4 and AC-6, respectively (Table 9-III). Leaching of U in the waste-rock pile generated a much higher concentration of dissolved U (24 ppb) and a lower AR (1.263) in seep water AC-2. Mining exposes many previously unweathered surfaces and should accelerate oxidation and leaching of U and other trace elements such as Se that are soluble as cations in acid solution and as oxyanions in alkaline solution (Amacher et al., 2001). Weathering should lower the 234U/238UAR values in residual solids and, in particular, in the phases most susceptible to U leaching. The 0. l M sodium bicarbonate extracts of relatively fresh samples l and 2 have AR values that are < 1.0 (Table 9-III). The extracted U may represent U that is normally left behind in the most leachable sites after incipient weathering by surface and ground water. In contrast, the extract of highly weathered surficial sample 6 has AR > 1.0 (1.117). This extract may contain U that was liberated from apatite during weathering (AR > l) but was then retained by alteration products. Alternatively, some extractable U of high AR could represent U that was recently adsorbed from surface water onto alteration products and thus mimics the U isotopic composition of local surface water.
CONCLUSIONS The vast majority of U in the Meade Peak Phosphatic Shale Member is contained in relatively insoluble carbonate fluorapatite, but a small (5-10%) fraction of U is extractable with a mild sodium bicarbonate solution. Fission-track radiography indicates that possible hosts of this labile U include organic matter, iron-bearing alteration products, trace uraninite, and very fine-grained apatite cement. During progressive weathering, this small component of U may be removed from rock samples or may reattach (sorb) onto alteration products, and some additional U may be sorbed from ground and surface water. The net effects of recent (late Pleistocene-Holocene) open-system behavior of U are recorded as small departures (1-15%) from secular equilibrium in the 238U-decay series. The nature
Weathering of the Meade Peak Phosphatic Shale Member." Uranium study
247
and extent of this disequilibrium varies as a function of degree of weathering and scale of sampling. The most-weathered samples show the largest and most recent (last several 105 years) evidence for redistribution of U on a small scale (several mg), but on a wholerock scale approximate closed systems because alteration products such as secondary iron oxides likely sequester mobilized U. Less-weathered samples contain slightly larger amounts of extractable, potentially labile U and show small disequilibria that are associated with less recent U loss. This behavior is consistent with minor, slow loss of U during the initial stages of weathering. Slow rates of U removal during weathering are also indicated by low dissolved U concentrations (<3 ppb) and elevated 234U/238UAR (> 1.5) in local surface water. An important exception is low-volume seepage water flowing from the base of a mining waste-rock pile that represents a local point source of high dissolved U (24--40 ppb). Several outcrop and channel samples show radioactive disequilibria between 226Ra and 21~ decay product that is probably the result of ongoing open-system movement of 222Rn within the last 100 years. The mobility of Rn is likely enhanced by rapid upward movement of gases along fractures in these highly tilted host rocks. When compared to U, the fraction of extractable Se in variably weathered samples is much more skewed towards high values in the least-weathered phosphorite. Greater extractability and potential availability of Se during incipient weathering of phosphorite suggests an expected greater association of Se with phases other than apatite. The behavior of mobile Se during weathering of the Meade Peak Phosphatic Shale Member is therefore most analogous to the open-system behavior of the small, non-apatite fraction of U. During weathering of Meade Peak Phosphatic Shale Member, very slow release of U is controlled by the slow rate of attack of the primary U hosts by local, mildly alkaline waters, and by the sorption of U onto alteration products. Because U and its decay products reside primarily in relatively insoluble apatite, preferential removal of other more soluble minerals should produce relative enrichment of U and its decay products in the weathered residuum, similar to the behavior of phosphate (Krieg, 1997). Slightly more rapid release of U and greater fluxes of Rn gas are predicted in more highly fractured rock strata and in recently disturbed mining areas and waste-rock piles.
ACKNOWLEDGMENTS Special thanks to M.W. Lewan, US Geological Survey, for providing the six variably weathered samples from the Rasmussen Ridge mine and for permitting access to field notes and unpublished data for their chemical composition. James R. Herring, USGS, graciously provided surface water samples from Angus Creek and chemical data for similar water samples collected previously. Other personnel of the USGS who helped with sample processing and analysis included Z. Brown (Se), P.J. Lamothe (ICP-MS), and L.A. Neymark (TIMS).
248
R.A. Zielinski et al.
REFERENCES Altschuler, Z.S., Clarke, R.S. and Young, E.J., 1958. Geochemistry of uranium in apatite and phosphorite. US Geological Survey, Professional Paper, 314-D, 90 pp. Altschuler, Z.S., 1973. The weathering of phosphate deposits - geochemical and environmental aspects. In: E. Griffith, A. Beeton, J. Spencer and D. Mitchell (eds.), Environmental Phosphorous Handbook, Wiley, New York, pp. 33-96. Amacher, M.C., Herring, J.R. and Stillings, L.L., 2001. Total recoverable selenium and other elements by HNO3 and HC104 digestion and other soil characterization data from Wooley Valley units 3 and 4 waste rock dumps and Dairy Syncline lease area soils, southeast Idaho. US Geological Survey, Open File Report, 01-69, 65 pp. Arden, J.W. and Gale, N.H., 1974. Separation of trace amounts of uranium and thorium and their determination by mass spectrometric isotope dilution. Anal. Chem., 46:687-691. Baturin, G.N., 1982. Phosphorites on the Sea Floor, Origin, Composition and Distribution. Developments in Sedimentology 33, Elsevier, Amsterdam, 343 pp. Bethke, C.M., 1998. The Geochemist's Workbench, version 3.0, A users guide to Rxn, Act2, Tact, React, and Gtplot. Hydrogeology Program, University of Illinois. 184 pp. Burnett, W.C., Chin, P., Deetae, S. and Panik, P., 1988. Release of radium and other decay-series isotopes from Florida phosphate rock. Florida Institute of Phosphate Research, Publication No. 05-016-059, 164 pp. Burnett, W.C. and Veeh, H.H., 1977. Uranium-series disequilibrium studies in phosphate nodules from the west coast of South America. Geochim. Cosmochim. Acta, 41: 755-764. Burnett, W.C. and Veeh, H.H., 1992. Uranium-series studies of marine phosphates and carbonates. In: M. lvanovich and R. Harmon (eds.), Uranium-series Disequilibrium, Applications to Earth, Marine, and Environmental Sciences, Clarendon Press, Oxford, pp. 487-512. Chai, B.H.T., 1974. Mass transfer of calcite during hydrothermal recrystallization. In: A. Hofmann, B. Giletti, H. Yoder and R. Yund (eds.), Geochemical Transport and Kinetics, Carnegie Institute of Washington, Publication 634, pp. 205-218. Claypool, G.E., Love, A.H. and Maughan, E.K., 1977. Organic geochemistry, incipient metamorphism, and oil generation in black shale members of Phosphoria Formation, western interior United States. Am. Assoc. Petrol. Geol. Bull., 62: 98-120. Cook, P.J., 1969. The petrology and geochemistry of the Meade Peak Member of the Phsophoria Formation. Unpublished PhD. thesis, University of Colorado, 124 pp. Cressman, E.R. and Swanson, R.W., 1964. Stratigraphy and petrology of the Permian Rocks of southwestern Montana. US Geological Survey, Prof. Paper, 313-C, pp. 275-569. Dequincey, O., Chabaux, F., Clauer, N., Sigmarsson, O., Liewig, N., and Leprun, J.C., 2002. Chemical mobilizations in laterites: Evidence from trace elements and 238U-234U-23~ disequilibria. Geochim. Cosmochim. Acta, 66:1197-1210. Dickson, B.L. and Wheller, G.E., 1992. Uranium-series disequilibrium in exploration geology. In: M. Ivanovich and R. Harmon (eds.), Uranium-Series Disequilibrium, Applications to Earth, Marine, and Environmental Sciences, Clarendon Press, Oxford, pp. 704-730. Doerner, H.A. and Hoskins, W.M., 1925. Co-precipitation of radium and barium sulfates. J. Am. Chem. Soc., 47: 662-675. Fisher, R.S., 1998. Geologic and geochemical controls on naturally radioactive materials (NORM) in produced water from oil, gas, and geothermal operations. Environ. Geosci., 5: 139-150.
Weathering of the Meade Peak Phosphatic Shale Member." Uranium study
249
Friedlander, G., Kennedy, J.W. and Miller, J.M., 1964. Nuclear and Radiochemistry, Wiley, New York, p. 9. Ghahremani, D.T. and Banks, P.O., 1983. Micro-fracturing and its relation to radon movement in Devonian shales of northeastern Ohio. Trans. Am. Geophys. Union (EOS), 64: 882. Grauch, R.I., Tysdal, R.G., Johnson, E.A., Herring, J.R. and Desborough, G.A., 2001. Stratigraphic section and selected semiquantitative chemistry, Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Survey, Open-File Report, 99-20-E, 1 plate with text. Gulbrandsen, R.A., 1966. Chemical composition of phosphorites of the Phosphoria Formation. Geochim. Cosmochim. Acta, 30: 769-778. Herring, J.R., Desborough, G.A., Wilson, S.A., Tysdal, R.G., Grauch, R.I. and Gunter, M.E., 1999a. Chemical composition of weathered and unweathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation A. Measured sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open-File Report, 99-147-A, 21 pp. Herring, J.R., Wilson, S.A., Stillings, L.A., Knudsen, A.C., Gunter, M.E., Tysdal, R.G., Grauch, R.I., Desborough, G.A. and Zielinski, R.A., 1999b. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation B. Measured sections C and D, Dry Valley Caribou County, Idaho. US Geological Survey, Open-File Report, 99-147-B, 32 pp. Herring, J.R., Grauch, R.I., Siems, D.E, Tysdal, R.G., Johnson, E.A., Zielinski, R.A., Desborough, G.A., Knudsen, A., and Gunter, M.E., 2001. Chemical composition of strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation. Channel-composited and individual rock samples of measured section J and their relationship to measured Sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open-File Report, 01 - 195, 72 pp. Herring, J.R. and Amacher, M.C., 2001. Chemical composition of plants growing on the Wooley Valley phosphate mine waste pile and on similar rocks in nearby Dairy Syncline, Caribou County, Idaho. US Geol. Survey, Open-File Report, 01-0025, 47 pp. Jarvis, I., Burnett, W.C., Nathan, Y., Almbaydin, ES.M., Attia, A K.M., Castro, L.N., Flicoteaux, R., Hilmy, M.E., Husain, V., Qutawnah, A.A., Serjani, A. and Zanin, Y.N., 1994. Phosphorite geochemistry: state-of-the-art and environmental concerns. Eclogae Geol. Helv., 87: 643-700. Jerden, J.L. Jr. and Sinha, A.K., 2003. Phosphate-based immobilization of uranium in an oxidizing bedrock aquifer. Appl. Geochem., 18: 823-843. Kreig, J.J., 1997. Mineralogical changes during weathering of the Meade Peak Member of the Permian Phosphoria Formation, Smokey Canyon Mine, southeast Idaho. Geol. Soc of Am. Abstracts with Programs, Annual Meeting, 29: 51. Langmuir, D. 1978. Uranium solution-mineral equilibria at low temperatures with applications to sedimentary ore deposits. Geochim. Cosmochim. Acta, 42: 547-569. Langmuir, D., 1997. Aqueous Environmental Geochemistry. Prentice Hall, Upper Saddle River, New Jersey, 600 pp. Langmuir, D. and Herman, J.S., 1980. The mobility of thorium in natural waters at low temperatures. Geochim. Cosmochim. Acta, 44: 1753-1766. Latham, A.G. and Schwarcz, H.P., 1987. On the possibility of determining rates of removal of uranium from crystalline igneous rocks using U-series disequilibria-l: a U-leach model, and its applicability to whole-rock data. Appl. Geochem., 2: 55-65.
250
R.A. Zielinski et al.
Lee, W.H., 2001. A history of phosphate mining in southeastern Idaho. US Geological Survey, OpenFile Report, 00-425, 253 pp. Available on CD-ROM or at http://geopubs.wr.usgs.gov/openfile/of00-425/. Levinson, A.A., Bland, C.J., and Lively, R.S., 1982. Exploration for U ore deposits. In: M. Ivanovich and R. Harmon (eds.), Uranium Series Disequilibrium: Applications to Environmental Problems, Clarendon Press, Oxford, pp. 351-383. McKelvey, V.E., 1956. Uranium in phosphate rock. In: L. Page, H. Stocking and H. Smith (eds.), Contributions to the Geology of Uranium and Thorium by the United States Geological Survey and Atomic Energy Commission for the United Nations International Conference on the Peaceful Uses of Atomic Energy, Geneva, Switzerland, 1955, US Geol. Survey, Prof. Paper, 300, pp. 477-481. McKelvey, V.E. and Carswell, L.D., 1956. Uranium in the Phosphoria Formation. In: L. Page, H. Stocking and H. Smith (eds.), Contributions to the geology of uranium and thorium by the United States Geological Survey and Atomic Energy Commission for the United Nations International Conference on the Peaceful Uses of Atomic Energy, Geneva, Switzerland, 1955, US Geological Survey, Prof. Paper, 300, pp. 483-494. Osmond, J.K. and lvanovich, M., 1992, Uranium-series mobilization and surface hydrology. In: M. lvanovich and R. Harmon (eds.), Uranium-Series Disequilibrium, Applications to Earth, Marine, and Environmental Sciences, Clarendon Press, Oxford, pp. 259-289. Piper, D.Z. and Medrano, M.D., 1994. Geochemistry of the Phosphoria Formation at Montpelier Canyon, Idaho: environment of deposition. US Geological Survey, Bulletin, 2023-B, 28 pp. Sheldon, R.P., 1959. Geochemistry of uranium in phosphorites and black shales of the Phosphoria Formation. US Geological Survey, Bulletin, 1084-D, pp. 83-113. Swanson, R.W., 1970. Mineral resources in Permian rocks of southwest Montana. US Geological Survey, Prof. Paper, 313-E, pp. 661-777. Thompson, M.E., 1954. Further studies of the distribution of uranium in rich phosphate beds of the Phosphoria Formation. US Geological Survey, Bulletin, 1009-D, pp. 107-123. Tysdal, R.G., Johnson, E.A., Herring, J.R. and Desborough, G.A., 1999. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, central part of Rasmussen Ridge, Caribou County, Idaho. U.S. Geological Survey, Open-File Report, 99-20-A, 1 plate. Tysdal, R.G., Desborough, G.A., Herring, J.R., Grauch, R.I. and Stillings, L.A., 2000. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Dry Valley, Caribou County, Idaho. U.S. Geological Survey, Open-File Report, 99-20-B, 1 plate. USEPA, 2000. National Primary Drinking Water Regulations; Radionuclides; Final Rule. Environmental Protection Agency 40 CFR parts 9, 141, 142. Federal Register vol. 65, no. 236, Thursday, December 7, 2000. Zielinski, R.A. and Budahn, J.R., 1998. Radionuclides in fly ash and bottom ash: improved characterization based on radiography and low energy gamma-ray spectrometry. Fuel, 77: 259-267.
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
251
Chapter 10
MINERAL AFFINITIES AND DISTRIBUTION OF SELENIUM AND OTHER TRACE ELEMENTS IN BLACK SHALE AND PHOSPHORITE OF THE PHOSPHORIA FORMATION
R.B. PERKINS and A.L. FOSTER
ABSTRACT Mineral affinities and quantitative distributions of Se and other trace elements of environmental concern in the Permian Phosphoria Formation were studied in samples from three variably weathered sections using microprobe, X-ray absorption spectroscopy (XAS), and sequential-extraction analyses. The results show a clear difference in the distribution of Cd, Cu, Ni, Se, V, and Zn between non-weathered and weathered samples. In unweathered sampies, sulfides (mainly pyrite and sphalerite) host the majority of Cd, Cu, Se, and Zn and a large proportion of the Ni and V. Most of the non-sulfide fraction of these elements in unweathered samples is associated with organic matter and oxyhydroxides. A small fraction of Se is present in elemental form. Apatite is the primary host for U. Both apatite and organic matter may host a significant fraction of Mo. Of the elements investigated, only Cr, U, and V were found to have minimal association with organic matter in unweathered rocks. Acid-insoluble phases (assumed to be silicates and oxides) host the majority of Cr and a significant amount of V. Molybdenum was the only element for which a significant fraction is easily leachable (assumed to be weakly adsorbed or in a very soluble phase). In weathered samples, acid-soluble oxyhydroxides are the primary hosts for all the aforementioned elements except Cr and U, which are associated with relatively stable phases in non-weathered samples. Organic-bound Ni, which represents a substantial part of total Ni in unweathered samples, apparently is more resistant to weathering than is organic-bound Mo, which appears to be readily lost during weathering. Despite the obvious shift in mineral residences of Se and associated trace elements upon weathering, weathered rocks may still contain high concentrations of these elements. As many of the elements of concern appear to be associated primarily with oxyhydroxides in weathered rocks, both dissolved and sorbed species released to the surface environment must be considered. Furthermore, trace elements in unweathered and minimally weathered rocks are hosted in a number of phases that have variable oxidation rates and a wide range of particle sizes. Some of these host phases occur as inclusions within phases more resistant to oxidative weathering, such as phosphate pellets. These factors indicate that the release of these elements to the environment will be a variable and long-term process.
252
R.B. Perkins and A.L. Foster
INTRODUCTION Black, organic-rich shales commonly host mineral and energy resources such as metals, oil, coal, and phosphate. However, black shales are also highly enriched in potentially toxic trace elements such as Se (Presser et al., Chapter 11; Herring and Grauch, Chapter 12). Surface mining can lead to accelerated weathering and release of these trace elements. The bioavailability of Se and other trace elements in the surface environment is controlled by their chemical form, the rate of weathering of host phases, and the rates of sorption and precipitation of the elements by and within weathering products. Determination of Se speciation and Se host phases in source rocks is thus a crucial step in modeling the mobility of this important and potentially harmful element. Previous investigations of Se in the environment have typically focused on the bioavailability of Se in soils and sediments (e.g. Chao and Sanzolone, 1989; Pickering et al., 1995; Tokunaga et al., 1996; Zhang and Moore, 1996; Martens and Suarez, 1997a, 1997b) and little work has been done to characterize Se in parent rocks. Recent studies (e.g. Piper et al., 2000) have shown that black shales and phosphorites of the Phosphoria Formation are the source of Se linked to the death of livestock in the phosphate mining district of southeast Idaho. The highest concentrations of Se within the Phosphoria Formation are found in the lowermost Meade Peak Phosphatic Shale Member, the primary source of mined phosphate. Other elements that may be of environmental concern, including Cd, Cr, Cu, Mo, Ni, U, V, and Zn, are also present in the Meade Peak Member at concentrations well above their average crustal abundance. The purpose of this study is to determine the mineralogical associations and distribution of Se and other elements of environmental concern in variably weathered phosphatic black shales of the Meade Peak Member of the Phosphoria Formation. The two distinct yet complimentary approaches used were: (a) sequential-extraction techniques to determine the distribution of trace elements among operationally defined mineral fractions; and (b) solid-phase analytical techniques to identify specific minerals hosting trace elements and to determine the range of trace-element concentrations in these phases. These latter techniques included X-ray diffraction (XRD), scanning electron microscopy (SEM), electron microprobe (EMP) analyses, and X-ray absorption spectroscopy (XAS). The findings of this study should be useful in modeling the release of Se and other trace elements resulting from natural weathering and land-use activities in regions where black shales occur.
METHODS Sample selection and preparation
The Meade Peak Member consists of interbedded, organic-carbon-rich silty mudstone, carbonates, phosphorite, and chert. This study focused on phosphorites and mudstones (Table 10-I) because bulk-rock chemical analyses indicate these units contain the highest concentrations of trace elements (Medrano and Piper, 1995; Piper, 1999; Piper et al., 2000).
253
Mineral affinities and distribution o f selenium and other trace elements
TABLE 10-I Lithology and selective chemical composition (wt.%, except Se ppm) of analyzed samples Sample 1
Lithology
*EV-P-602
Cherty mudstone Phosphatic mudstone Phosphatic mudstone
287
*EV-P-762 *EV-P- 1 8 6 2 *HS-2183
HS- 136 *HS-5214 *HS-5653 LR-WS 115 LR-M 125 *LR-M14-155 *LR-M16-185 LR-M215 *LR-M36-405
*LR-M515
Se
Inorganic C
TOC
1.1
1.0
1.7
3.2
0.2
124
9.4
0.1
6.4
3.4
0.8
73
9.4
0.1
18.1
3.2
2.0
Phosphorite Phosphorite Phosphatic mudstone Phosphorite
111 119 120
23.1 17.9 9.6
NA 1.3 1.1
NA 11.3 13.3
1.2 2.0 2.2
2.0 3.2 2.9
32.8
NA
NA
0.25
1.1
Dolostone Phosphorite Phosphatic dolostone Mudstone Phosphatic dolostone Phosphorite/ phosphatic mudstone Phosphorite
12.2 28.3 209
11.7
P205
Fe203
S
0.53 33.2 5.8
9.6 1.7 6.1
0.24 1.31 11.5
0.77 0.30 1.5
0.69 1.58 4.0
287 143
4.7 5.2
4.6 6.0
14.5 7.7
1.8 1.5
5.2 3.2
55
9.1
2.4
3.2
3.3
3.8
32
25.9
0.9
0.7
0.6
1.9
1Location denoted by prefix: EV, Enoch Valley; HS, Hot Springs; LR, Lakeridge. eBulk-rock analyses from Piper (1999a). 3This study. 4piper et al. (2000). 5Unpublished data; note: Se concentrations for HS-218 and -565 determined by ICP-MS; values are likely lower (up to --25%) than actual concentrations due to partial volatilization during digestion; note that sample LR-WS11 is from Grandeur Member of Park City Formation, immediately underlying Meade Peak Member. Samples marked with * were used in sequential-extraction study.
Fourteen samples were obtained from three stratigraphic sections (Fig. 10-1): (a) a freshly excavated trench in the lower section of the open pit at Enoch Valley Mine in SE Idaho (Piper, 1999), (b) a mine adit in the underground Hot Springs Mine in SE Idaho (Gulbrandsen, 1979), and (c) a freshly slabbed repository core obtained from General Petroleum's Lakeridge No. 43-19-G well, drilled in Sublette County, Wyoming (Sheldon, 1963; Murata et al., 1972).
254
R.B. Perla'ns and A.L. Foster
m above base of M e a d e Peak
Section Locations:
r~
~9 MONTANA "' ....... i ...... IDAHO!, WYOMING Enoch ! Valley 9I 9Lakeridge . . . . . . . . _=j Hot i' Springs L._ ,
HS-565" -.~
50-
]I
i I i i
"-. . . . i i
HS-521" .--~ ~ / ~ .f'./.~ Chert 40I
] Carbonate
~
Phosphatic Mudstones
~_J 30-
-V-P186*
~
n
Cherty Mudstone
~--HS-136
Mudstone ,R-M51 *
20EV-P76*
M36-40" EV-P60*
10-
LR-M21 HS-218" F---
.M16-18" I-M14-15" LR-WS11
_ LR-M12 Enoch valley
Hot Springs
Lakeridge
Fig. 10-1. Generalized stratigraphic sections of the Meade Peak Member of the Phosphoria Formation at three locations showing locations of samples referred to in this study. Samples marked with * were used in the sequential-extraction study. Lakeridge samples LR-M37A and LR-M41A, used in the XAS study, are not shown but are located immediately below and above sample LR-M37-40.
The samples selected represent a range of phosphate, carbonate, organic carbon, and trace-element contents (Table 10-I and Fig. 10-1). Samples obtained from the Enoch Valley Mine were qualitatively more weathered than the samples obtained from either the adit samples or the slabbed core, as evidenced by their degree of friability, oxidized mineral surfaces, and the occurrence of iron sulfates (e.g. jarosite) on fracture surfaces. Twelve samples were examined with SEM and energy-dispersive spectroscopy (EDS) and seven of those samples were analyzed by EMP. Only one Enoch Valley sample (EV-P 186) was examined by SEM. Both qualitative and quantitative analyses were made on polished sections that were cut from the interiors of hand or core samples. Stub mounts of
Mineral affinities and distribution o f selenium and other trace elements
255
freshly broken bedding and fracture surfaces were used for some SEM analyses. Gold-palladium coating was applied to SEM samples and carbon coating was typically used for EMP analyses. Ten samples were analyzed by sequential chemical-extraction methods, three each from the Enoch Valley and Hot Springs sections and four from the Lakeridge section. Samples were chosen from upper and lower parts of each section (Fig. 10-1). Two of the Enoch Valley samples, EV-P76 and EV-P186, were obtained from previously powdered samples that had been stored in glass jars. The third Enoch Valley sample (EV-P60) as well as hand samples or slabbed-core segments from the Hot Springs and Lakeridge sections were trimmed, crushed, and milled to coarse chips. These chip samples were finely ground to < 150 Ixm with an agate mortar and pestle and dried at 40-45~ to constant weight immediately prior to beginning the sequential-extraction study. A riffle splitter was used to obtain representative aliquots of samples for analyses. A method blank and several reference material samples were prepared in duplicate to evaluate the effectiveness and selectivity of various extraction steps. Each blank consisted of 4.00 ___0.01 g of a low-iron glass sand (National Institute of Standards and Technology Standard Reference Material 165a) and 0.50 ___0.02 g of acid-washed activated charcoal (Sigma). The latter was intended as a sorbent phase proxy for organic matter. Total organic carbon in the more organic-rich samples exceeded the 11 wt.% represented by the activated charcoal in the blank material. The silica sand contained measurable amounts (1-590 ppm) of A1, Cr, Fe, Ti, and Zr oxides. Each of the reference materials was made by adding various Se compounds, Se-bearing solutions, or Se-bearing minerals to the same blank material. Although the reference materials do not accurately simulate natural geologic materials, their use nonetheless provides a measure of the ability of a given reagent to extract analytes from the target phase and the overall selectivity of the sequential extraction scheme with respect to the target phase. Reference material one (REF-01) contained 0.1 ml of a 0.058 M Na2SeO3 solution. Each of the REF-01 samples was air dried prior to beginning extraction. This material was used to measure the effectiveness of removing the "soluble plus exchangeable" Se fraction. Reference material two (REF-02) contained 2 mg of red elemental Se (Pfaltz and Bauer). Reference material three (REF-03) contained 7 mg of naturally occuring Se-rich pyrite/marcasite. Although previous microprobe analyses indicated elevated levels of Se (i.e. > 200 ppm) in the pyrite/marcasite used (David John, USGS, personal communication), the material was not fully characterized in terms of its bulk Se concentration or homogeneity, so results from this reference material are only qualitative. Reference material REF-04, composed of wheat gluten added to the glass sand, was quickly judged unsuitable for liquid extractions. A fifth reference material (REF-05) containing 5.0 mg of CuSe (Alpha-Aesar, 99.5%) was used to quantitatively evaluate the stability of acidvolatile selenides during the various extraction steps. Reference materials for X-ray absorption spectroscopy (XAS) were purchased as pure phases (e.g. selenomethionine and elemental (red) Se), or obtained from natural samples (e.g. the Se-pyrite/marcasite mentioned). Concentrated reference materials (> 1 wt.% Se) were mixed thoroughly with glucose or boron nitride in a mortar and pestle to obtain an
256
R.B. Perkins and A.L. Foster
absorption length (/zx, where /x is the energy- and mass-dependent mass absorption coefficient, and x is the sample thickness) of approximately 1-1.5. The mixture was then loaded into either 1.5 mm thick aluminum or Teflon holders with Mylar windows. Dilute reference materials and all samples were ground in a mortar and pestle and mounted in 3 mm thick Teflon holders with Mylar windows.
Solid characterization o f selected samples XRD samples were prepared as bulk powder mounts and analyzed on a Philips X-ray diffractometer using Cu Ka radiation, a graphite monochromator, a step size of 0.020 ~ 20, and a scan time per step of 1 s. Samples were run from 4 to 70 ~ 20 at 40 kV and 45 mA. Peak determinations and peak matches with existing powder diffraction files were performed using the Philips X-Pert software package. Sample images were taken and qualitative chemical compositions ascertained with a LEO 982 digital field-emission SEM equipped with an energy-dispersive X-ray spectrometer (EDS) using a 15 mm working distance and an accelerating voltage of 15-20 kV. Both polished sections and stub mounts were examined with the SEM. Quantitative solid-phase chemical analyses were conducted using a JEOL 8900 Superprobe EMP with five wavelength-dispersive spectrometers (WDS) and an energydispersive spectrometer (EDS). A 15 or 20 kV accelerating voltage and 30 nA current were used in the analyses. A focused (-< 1 txm) electron beam was used to measure concentrations in small phases, particularly sulfides; a 30 txm beam was used to obtain analyses more representative of the bulk composition for matrix materials and apatite. Standardization was conducted on a regular basis using igneous apatites, sulfides, silicates, chromite, and native metals for which bulk chemical analyses were available. ZAF corrections were applied in computing final concentrations. The WDS reporting limits, using counting times of 30-40 s, varied from < 100 to 500 ppm. These limits are the average corrected concentration having a net intensity just greater than three times the standard deviation of the average background signal determined in multiple measurements of standards and samples. Precisions calculated as the percent relative standard deviation of repeated measurements made on standards were typically under 5%. Accuracy was assessed periodically throughout sample measurements by measuring concentrations of major and minor elements in known standards. Instrument recalibration was employed whenever the relative percent difference between average measured concentrations and known concentrations exceeded 5%. Typically, accuracies of better than 2% were achieved. Post-processing corrections, based on averaged deviations from known values in standards run as unknowns, were applied as needed when such deviations exceeded 0.5% for major elements to 2% (relative) for trace elements. Selenium K-edge X-ray absorption near edge spectra (XANES) of samples and reference materials were collected at the Stanford Synchrotron Radiation Laboratory (SSRL) over the energy range 12,425-12,710 eV, using a Si(220) double crystal monochromator. A 0.2-eV step was used in the XANES region in order to collect high-resolution spectra. Each XANES scan was calibrated to the maximum inflection point in the spectrum of
Mineral affinities and distribution of selenium and other trace elements
257
a native Se standard provided by SSRL (collected concurrently with each scan), which was assigned a value of 12,658.0 eV. Spectra of concentrated reference materials were collected in transmission mode (sample oriented perpendicular to X-ray beam) whereas spectra of dilute reference materials and core samples were collected in fluorescence mode (sample oriented at 45 ~ to the X-ray beam) using either an Ar- or Xe-filled Stern-Heald-Lytle detector with a 3 ixm-thick (Z-1) filter or a 13-element solid state Ge detector. XANES data processing and analysis was performed using EXAFSPAK (George, 1992) and WINXAS (Ressler, 1998) software packages following standard procedures for XAFS spectra (Sayers and Bunker, 1988). Shifts in energy of ___0.5 eV were allowed to accommodate slight differences in calibration displayed by some of the reference materials that were collected during previous experiments at SSRL.
Sequential extractions
Two sequential-extraction schemes were developed to optimize the release of target elements from their host phases. The efficiency of the extraction steps were also evaluated.
Sequential-extraction techniques
Three techniques were used to demonstrate the ability of a given extraction step to liberate the Se associated with the target mineral and to evaluate the stability of non-target minerals: (a) the use of blanks and well-characterized reference materials; (b) the use of parallel extractions that targeted elemental and organic-bound Se at different steps or with different extractants; and (c) characterization of solid residues resulting from various extraction steps. All samples were run in duplicate with each duplicate weighing 5.00 +__0.01 g. Samples were extracted in 50-ml PTFE centrifuge tubes. All labware used in the sequential extractions was cleaned with detergent, rinsed, soaked for a minimum of 4 h in a 5% HNO3 solution, triple rinsed with ultrapure (18 + MD, cm) water, and air dried. With the exception of the Se-rich pyrite in REF-03, only reagent-grade compounds were used in the extracts and reference materials. Both sequential-extraction schemes are summarized in Table 10-II. Details of the primary extraction scheme steps, labeled as to the intended mineral targets, are as follows. Scheme A, Step 1: "Soluble/Exchangeable Fraction". A 25 ml aliquot of 0.1 M K2HPO4-KHzPO4 solution (pH ~6.9) was added to each sample. The samples were then mixed thoroughly with a vortexer and placed in a reciprocating water-bath shaker at 130 oscillations per minute and 25~ for 2 h. The suspension was centrifuged at ~2000 rpm for 15 min and the supernatant decanted and filtered through a 25 mm diameter, 0.45 Ixm PVDF Acrodisc T M disposable syringe filter into a 50 ml graduated cylinder. Five to 6 ml of ultrapure water was added to the solid residue that was then carefully stirred using a Teflon-coated, stainless steel spatula (vortexing or shaking resulted in caking of the solid along the centrifuge tube walls due to the low rinsate to solid ratio) and allowed to stand
258
R.B. PerMns and A.L. Foster
TABLE 10-II Summary of sequential-extraction steps used in this study Step Scheme A 1
Extractant
Target phase(s)
Mechanism
0.1 M KH2PO4 KeHPO4 0.1 M K25208 @ 90~ 1.0 M Na2SO3, pH = 7
Water soluble, exchangeable forms Organic matter Elemental Se
Anion/ligand exchange
4 M HC1
Fe, Mn, and Al-oxides, carbonates, phosphates, acid-volatile sulfides, occluded organic matter, acid-soluble organic matter Sulfides
Concentrated (~ 16 M) HNO3 @ 90~ Scheme B 1
0.1 M KH2PO4 K2HPO4 1.0 M Na2SO3, pH = 7
Water soluble, exchangeable forms Elemental Se
4% (minimum) NaOC1
Organic matter
Oxidative dissolution Formation of selenosulfate Se(s) + SO~-(aq) --)SeSO32(aq) Acid dissolution/hydrolysis
Oxidative acid dissolution
Mass action Anion/ligand exchange Formation of selenosulfate Se(s) + SO~-(aq) ---)SeSO32(aq) Oxidative dissolution
for 15 min. The sample was then centrifuged as before and the rinsate filtered and added to the initial extract. One ml of concentrated nitric acid was added to the final extract to aid in sample preservation. Scheme A, Step 2: "Organic fraction". A 25 ml aliquot of 0.1 M K2S208 solution (pH ~ 9.5) was added to each of the solid samples. The sample and extract were mixed by vortexing and placed in a hot-water bath at 90~ for 2 h, with occasional shaking to maintain a mixed suspension. The samples were centrifuged, filtered, and rinsed with 5-6 ml of ultrapure water as described in Step 1. Scheme A, Step 3: "Elemental Se fraction" (aider Velinsky and Cutter, 1990). Twenty-five ml of 1.0 M Na2SO3 adjusted to pH 7 with HC1 was added to the solid material remaining after Step 2. Following vortexing, the samples were mixed in a reciprocating bath at 130 oscillations per minute at 25~ for 12 h. The samples were then centrifuged and filtered using 0.45 txm Acrodisc T M syringe filters into 150 ml glass beakers. An additional 5 ml of 1.0 M Na2SO3 solution was added to each of the solid residues, which were stirred
Mineral affinities and distribution of selenium and other trace elements
259
thoroughly with a Teflon-coated spatula and left to stand 15 min. The samples were again centrifuged and the secondary extract added to the initial extract. This step was repeated, replacing the Na2SO3 solution with 5-6 ml of ultrapure water. A 7 ml aliquot of concentrated nitric acid was then added to each of the filtered solutions. These solutions were capped and left to stand for one hour to oxidize Se. The beakers were then placed on a hot plate and the solutions slowly evaporated to < 4 ml over several hours to remove the nitric acid. A 3-4 ml aliquot of ultrapure water was added to each of the resulting crystallinerich suspensions and these solutions again evaporated to near dryness. Upon cooling, 10 ml of 4 M HC1 were added to each of the resulting thick suspensions and stirred until all visible solids were dissolved. Slight warming (< 50~ was usually required to facilitate the dissolution process. These solutions were then transferred to 50 ml graduated cylinders that had been rinsed several times with ultrapure water. Then ultrapure water was added to the graduated cylinders to achieve a final, nominal 30 ml volume. Scheme A, Step 4: "Oxyhydroxide, Carbonate, and Phosphate Fraction". A 25 ml aliquot of 4 M HC1 was added to the remaining material in each sample. The samples were first vortexed and then mixed in a reciprocating shaker water bath at 130 oscillations per minute and 25~ for 4 h. They were then centrifuged, filtered, and rinsed as per Steps 1 and 2. Scheme A, Step 5: "Crystalline Sulfide Fraction". Concentrated nitric acid (3.5 ml, nominal 16 M HNO3) was added to the solid material remaining after Step 4. The samples were capped and sonicated, then placed in a hot-water bath at 90~ for 30 min. Following cooling, 20 ml of ultrapure water was added to each sample and mixed by vortexing. The samples were heated to 90~ for 90 min, cooled, and then centrifuged, filtered, and rinsed as per Steps 1, 2, and 4. An alternative scheme was used on selected samples to verify the organic and elemental Se fractions. No extraction beyond the organic fraction was attempted. The secondary extraction scheme steps are as follows. Scheme B, Step 1: "Soluble/adsorbed fraction". Scheme B, Step 2: "Elemental Se fraction".
Identical to Scheme A, Step 1.
Identical to Scheme A, Step 3.
Scheme B, Step 3: "Organic fraction". A 20 ml aliquot of NaOC1 (Aldrich, minimum 4%) was added to each sample. The samples were capped and sonicated, then placed in a reciprocating water bath at 130 oscillations per minute and 25~ for approximately 16 h. An additional 4 ml of NaOC1 was added and the sample mixed for an additional 4 h or until no further reaction (bubbling) was noted. The samples were then centrifuged, filtered, and rinsed as described for Scheme A, Steps 1, 2, 4, and 5.
Analyses of extracts
Given the potential for changes in the oxidation state of Se and other redox-sensitive elements brought about by both the extraction and subsequent sample handling and
260
R.B. Perkins and A.L. Foster
storage, sample extracts were analyzed for total elemental concentrations only by inductively coupled-plasma mass spectrometry (ICP-MS) as per Lamothe et al. (1999). The ICP-MS unit was calibrated using reagent blanks spiked with commercial multielement standards to achieve matrix matching with samples. Detection limits for the various analytes were based on previously determined values for the settings used, the extract dilutions, and the ~6:1 extract/solid weight ratios used in the extractions.
lnitial and residual solids characterization
All solid samples were analyzed for element concentrations using inductively coupledplasma atomic emission spectroscopy (ICP-AES; Briggs, 1996), except for HS-218 and HS-565, which were analyzed by ICP-MS (Briggs and Meir, 1996). The ground samples were first digested using a mixture of hydrochloric, nitric, perchloric, and hydrofluoric acids. Selenium in samples other than HS-218 and HS-565 was determined by hydride generation and atomic-absorption spectrometry (HG-AAS) following multi-step digestion with heated acid mixtures (Hageman and Welsch, 1996). Total carbon and sulfur were determined as CO2 and SO2 using infrared detectors on combusted samples (Curry, 1996a,b). Carbonate carbon was determined by coulometric titration of monoethanolamine-sorbed CO2 evolved from the sample with hot perchloric acid (Papp et al., 1996). Organic carbon was calculated as the difference between total and carbonate carbon. Total concentrations for all samples except HS-218 and HS-565 were determined as part of previous investigations (unpublished data; Piper, 1999; Piper et al., 2000). The two Hot Springs samples were analyzed for this study. An additional, triplicate-sample aliquot was prepared for selected samples in order to obtain residues for solid characterization. While these samples were subjected to the same sequential-extraction processes as the other samples, their extracts were not analyzed. Both starting samples and these residual solids were analyzed using XRD and SEM with EDS. Residual solids from several samples were also digested in acid and analyzed by ICP-MS as per Briggs and Meier (1996).
RESULTS Solid characterization Scanning electron microscopy
Scanning electron microscopy shows that phosphorite-rock samples are composed of pellets and ooids of carbonate fluorapatite (apatite) varying in size from ~40 I~m to > 2 mm in diameter (Figs. 10-2A-F). Some samples are composed only of pelletal apatite grains and matrix. Ooid layers often show significant variations in backscattered electron (BSE) intensity. The lower BSE-intensity layers often contain sufficient organic-matter content to enable detection of C and S by EDS analyses.
Mineral affinities and distribution o f selenium and other trace elements
261
Fig. 10-2. (A) Reflected light photomicrograph of large apatite pellet surrounded by smaller ooids (HS-565); (B) BSE image of apatite pellets (pale gray) in mixed phosphate/silicate (medium gray) and carbonate (dark gray) matrix (LR-M12); minute white grains are barite; (C) BSE image of large apatite pellets (HS565); dark material rimming pellets is organic-rich; white area in lower right corner is sulfide-filled fossil mold; (D) BSE image of apatite pellet inclusions (LR-MI2); (E) highcontrast BSE image showing complex composite ooid structure (HS-565); (F) BSE image of phosphate ooid formed around quartz grain (LR-M12). Major and minor minerals include apatite, calcite, dolomite, microcrystalline quartz, and detrital quartz and feldspar. Other, presumably detrital phases, including monazite and zircons, are far less common. Apatite or micritic carbonate form the matrix in most samples (Fig. 10-2B), although microcrystalline quartz is common in the upper Lakeridge samples. Illite, smectite, and kaolinite are present in variable amounts in some samples. Organic matter tends to occur in small fractures, as rims around phosphate grains (Fig. 10-2C), or as occluded material in apatite grains (Fig. 10-2F). The use of EDS analyses to identify phases with percent levels of elements was facilitated by limiting analyses to phases that produced a strong BSE signal, indicating a high
262
R.B. Perkins and A.L. Foster
Mineral affinities and distribution of selenium and other trace elements
263
Fig.10-3. BSE images showing examples of sulfides in Phosphoria samples. (A) Ni-bearing pyrite framboids in dark, organic-rich matrix between apatite pellets (HS-565); (B) aggregate mass of coalesced pyrite framboids (LR-M37). (C) masses of pyrite in matrix between apatite pellets (LR-M37); (D) close-up of euhedral pyrite crystals in quartz matrix; note spicules (LR-M37); (E) replacement of apatite pellet by pyrite; note the lack of alteration in adjacent grains (LR-M12); (F) replacement of K-feldspar by a Cu-V sulfide; also fine grains of Cd-bearing sphalerite interspersed throughout matrix (LR-M17); (G) apparently sheared mass of Cd-bearing sphalerite in mixed detrital/organic-matter matrix with finer pyrite crystallites to either end (LR-M15); (H) sphalerite and sub-i~m-sized pyrite in a fossil mold (HS-136); (I) pyrite grain in vug showing cuneiform structure (LR-M37); (J) rosette formed by radiating Fe-S (marcasite?) crystals (LR-M43); (K) Fe-Cu-V sulfide blebs in K-aluminosilicates (LR-M17); (L) BSE image comparison of pyrite framboids (LR-M37, upper left; HS-565, upper right) with Fe-oxyhydroxide-replaced framboids (EV-P186; lower two images); note the remnant pyrite (white specks) in lower left image.
mean atomic number relative to the major phases such as apatite, calcite, and silicates. The most common phases with a high BSE response in the Lakeridge and Hot Springs samples were sulfides (Figs. 10-3A-L), although heavy-metal-bearing silicates, oxides, and (or) sulfates were also present in every sample. Sphalerite appears more abundant in the Hot Springs samples while pyrite is ubiquitous. Iron sulfides occur in several different forms. (a) Framboidal pyrite (Fig. 10-3A) is typically associated with pellet-rims, organic-rich matrix material, and occasionally forms massive aggregates (Fig. 10-3B). However, both discrete framboids (~1-10 I~m in diameter) and aggregates occur within apatite grains. (b) Euhedral to subhedral pyrite grains, generally < 100 p~m in diameter, are found in the matrix and replace apatite pellets (Figs. 10-3C-E). (c) Cryptocrystalline (sub-micrometer) Fe sulfides are typically found in concentrated masses in organic-rich matrix or in microenvironments, such as infilled vugs or fossil molds (Figs. 10-3G-I). (d) Acicular, often radiating, iron sulfides (marcasite?) are found only in the upper Lakeridge section, typically in microcrystalline quartz matrix (Fig. 10-3J). (e) Micrometer-sized blebs of Fe-sulfides, typically containing other transition metals, are associated with layered K- and Na-aluminosilicates in the lower Lakeridge section (Fig. 10-3K). Most sphalerite is subhedral to euhedral and occurs within apatite grains, vug and fossil-mold infillings, and matrix materials (e.g. Fig. 10-3H). The only sulfides in Enoch Valley sample EV-P186 are small remnants of pyrite in what appear to be relict framboid structures that had been nearly completely replaced by Fe oxyhydroxides (Fig. 10-3L).
264
R.B. Perkins and A.L. Foster
Quantitative analyses using the electron microprobe
The composition of apatite pellets and ooids, pyrite, sphalerite, and less common Cu/ V-rich sulfides (sulvanite) were determined by EMP analyses (Tables 10-III and 10-IV). Low oxide totals (averaging 95-96%) in apatite pellets and ooids are almost certainly due to carbonate and Sr substitutions for phosphate and Ca and to organic inclusions. Exceptionally low totals (45-85%) were common in analyses of framboidal pyrite and microcrystalline sulfide masses, presumably due to high absorption resulting from uneven surfaces (angled facets or plucked micro-crystals). Most of these analyses were deemed unsuitable for inclusion in the tables. However, the results of analyses of framboidal pyrite in Hot Springs sample HS-565 is included in Table 10-IV as an example (average total = 82.3%). Element totals o f - 9 5 % typically resulted from analyses of sphalerite. Cadmium substitution for Zn is evident in all the sphalerites analyzed. While Cd Ka peak interference is possible from Cu, the Cu content in sphalerite is typically low and Cd background counts are not particularly elevated in sphalerite analyses. Thus, the low totals are likely the result of missing analytes such as Ag, As, Co, Ge, Hg, Mn, and Sn, although standardization problems cannot be entirely ruled out. Low totals for sulvanite are attributed to lack of As analyses, as discussed below. The EMP results indicate little significant enrichment of trace elements (above bulk concentrations) in apatite pellets and ooids. Although measurable Mo in some samples may be due to micro-inclusions, the persistent and relatively consistent Mo concentrations in other samples (e.g. LR-MI4 and-37) suggest possible incorporation of Mo within the apatite structure. The occasional high Zn content, particularly in sample HS-565, is assumed to be due to micro-inclusions of organic matter or sphalerite. Sulfide phases are variably enriched in several trace elements. Sphalerite contains as much as 4.8% Cd, averaging 2-3%. Though these levels are high, they are within the range (< 5%) of previously reported Cd concentrations. The concentration of other trace elements in sphalerite are not as high as in pyrite, although Ni may reach 0.1-0.5~ and both Cu and Se may be enriched over bulk-rock concentrations (e.g. LR-M21). Pyrite generally has a 10- to 100-fold enrichment of Se over bulk-rock concentrations, with average concentrations exceeding 2000 ppm in four of five samples and nearing 1% in two of the five samples. Selenium contents of up to 2.0% were measured in individual pyrite grains in LR-M12, the highest Se concentrations detected. Interestingly, this is nearly the same maximum value reported by Desborough (1.9% Se, 1977), who analyzed 24 pyrite grains in Meade Peak samples from Coal Canyon, Wyoming. Coleman and Delevaux (1957) noted that pyrite in U-ore deposits in sedimentary formations of the Colorado Plateau contain up to 3% Se. Although average Cr concentrations exceeded the calculated reporting limit in all but one sample (LR-M37), Cr was not consistently detected in any sample. Chromium may represent micro-inclusions of a Cr-bearing phase such as Cr(OH)3 or Cr203. Average Cu and Ni concentrations in pyrite exceed 0.1% in some samples. Very high concentrations of both (1.3 and 1.8%, respectively) were obtained for individual pyrite grains from sample LR-M40. Measurable V in pyrite was consistently found in only two of five samples analyzed. Small (_< 5 Ixm), framboidal
T A B L E 10-III
t~
A v e r a g e c o m p o s i t i o n b y m i c r o p r o b e ( w t . % oxides a n d F; p p m others) o f apatite pellets a n d ooids in selected s a m p l e s
Sample I
P205
CaO
F
FeO
MgO
Na20
SiO2
SO3
Cd
Cr
HS-565: N = 29 Average oxide total = 96.1%; residual attributed to CO3 and Sr substitutions + organic inclusions Average: 38.3 52.9 3.2 0.05 0.07 0.24 0.08 1.3 < 130 < 190 Maximum: 42.1 57.8 3.8 1.0 0.13 0.41 0.76 3.0 510 260 % Detects3:
100
100
100
LR-M12: N = 22 Average oxide total = 96.1% Average: 36.1 52.2 4.3 Maximum: 38.6 53.6 5.1 % Detects:
100
100
100
LR-M14: N = 5 Average oxide total = 96.5% Average: 36.0 52.5 4.1 Maximum: 37.9 55.0 5.0 % Detects:
100
100
100
LR-M37: N = 14 Average oxide total = 96.1% Average: 35.3 51.9 4.1 Maximum: 35.9 52.6 4.4 % Detects:
100
100
100
LR-M40: N = 24 Average oxide total = 96.2% Average: 36.9 53.0 3.5 Maximum: 38.5 53.8 4.2 % Detects:
100
100
100
45
0.10 0.80 82
0.35 1.3 80
0.12 0.45 93
0.05 0.32 88
100
97
NA -
NA -
-
-
0.05 0.09
0.55 0.74
52
0.06 0.47 77
<0.02 <0.02
100
100
0
NA -
NA -
0.87 4.8
-
-
NA -
NA -
-
-
100
0.19 1.3 79
97
3.1 4.6 100
2.9 5.0 100
3.5 4.6 100
3.4 4.1 100
Cu
Mo
NA -
Ni
Se
V
Zn
<500 540 47
< 100 < 100 17
NA -
950 7300 76
48
76
-
(170) 600 72
< 130 190 64
< 190 550 68
<450 610 50
(170) 410 86
<500 500 55
<100 160 68
<1 160 55
< 460 < 460 0
< 130 < 130
< 190 < 190
NA
280 650
NA -
< 100 < 100
NA
0
40
100
-
20
-
630 1950 60
390 100
<500 <500 36
<100 <100 100 200 43 71
< 460 < 460 0
<450 870
230 490
<500 (380)
< 100 < 100 110 120
50
96
< 460 < 460 0
< 130 140
(160) 360
<450 1670
43
86
43
< 130 200 54
(160) 390 92
-
220
33
50
ISee footnote, Table 10-I for explanation of prefixes. 2Values in parenthesis are near but below calculated reporting limits (average concentrations having > 3 • background net intensity). 3% Detects: percentage of analyses with detectable amounts of analyte, regardless of whether below calculated reporting limit. Note: A1203 also measured in every sample; average content = 0.05%, except in LR-M37 (0.20%). Average K20 content in HS-565 and LR-M14 < 0.01.
46
t~ t~
to
TABLE 1O-IV
tO O~
Average compositions (wt.% S, Fe; ppm others) of sulfides in selected samples as measured with electron microprobe
Sample 1
S
Fe
Zn
Cd
Cr
Cu
Mo
Ni
Se
LR-WS11" Pyrite (0.01-0.05 mm, subhedral) N = 29 Average element total = 98.9% Average: 52.7 45.3 < 460 < 130 360 Maximum: 54.2 47.9 < 460 230 1,960 100 100 52 48 83 % Detects:
< 450 2,130 69
< 190 < 190 0
< 500 1,930 45
3,500 16,300 100
< 100 310 45
LR-M12" Pyrite ( < Average: Maximum: % Detects:
0.01 mm, subhedral) N = 19 Average element total = 99.3% 52.4 44.3 < 460 < 130 850 53.4 46.2 < 460 220 5,860 100 100 53 47 84
2,790 6,740 100
< 190 < 190 0
1,660 6,840 100
7,840 19,500 100
130 360 63
LR-M21" Pyrite (0.01-0.1 mm, subhedral) N = 7 Average element total = 97.3% Average: 52.4 43.1 < 460 < 130 260 Maximum: 53.4 45.3 660 200 490 100 100 29 71 86 % Detects:
1,380 2,000 100
NA
1,830 3,800 100
9,660 14,900 100
280 430 100
6,560 6,870 100
< 500 < 500 33
< 100 220 73
< 100 120 33
NA
1,970 4,980 100
2,010 18,100 100
120 430 67
LR-M37: Pyrite (0.04-0.10 mm, subhedral to euhedral) N = 15 Average element total = 98.3% Average: 53.4 43.3 < 460 < 130 < 190 500 Maximum: 54.1 47.3 < 460 240 350 1,130 100 100 40 47 33 100 % Detects: LR-M40: Pyrite (0.01-0.05 mm, subhedral) N = 27 Average element total = 98.2% Average: 51.6 43.9 < 460 < 130 810 Maximum: 53.4 46.8 < 460 140 6,190 % Detects: 100 100 35 22 93
2,880 13,900 93
%
b~
HS-565: Framboidal pyrite (~5 I~m dia) in organic-rich matrix N = 5 Average element total = 82.3% 2 < 190 Average: 41.0 38.7 940 330 1060 2,340 < 190 Maximum: 42.9 43.6 2,500 830 3,220 5,880 0 100 100 80 80 100 100 % Detects:
2,330 6,050 100
1,510 4,030 100
174 340 100
< 100 < 100 40
LR-M21" Sphalerite (typically 0.01-0.05 mm, euhedral; also masses of microcrystals) N = 20 Average element total = 95.2% 2 740 Average: 32.8 0.067 58.6% 31,600 < 190 (440) NA 1,010 2,210 Maximum: 33.3 0.351 60.1% 47,700 220 1,630 4,580 100 100 95 100 100 70 70 80 % Detects:
< 100 380 70
HS-565" Sphalerite Average: Maximum: % Detects:
(0.01-0.10 mm, subhedral to euhedral) N = 5 Average element total = 94.4% 2 32.6 0.044 59.9% 18,000 < 190 < 450 32.9 0.072 60.5% 25,500 < 190 < 450 100 80 100 100 40 80
< 190 < 190 0
16,800 19,900 100
< 500 (400) 80
r~
LR-M21: Sulvanite (euhedral grains (> 0.02 mm) with sphalerite and large (~0.1 mm) masses of microcrystals) N = 11 Average element total = 92.6% 2 Average: 34.2 Maximum: 34.9 % Detects: 100
0.026 0.149 64
720 1,260 100
< 130 150 45
6,700 7,520 100
48.8% 49.3% 100
NA -
650 1,250 91
4,460 5,350 100
82,300 90,600 100
o~
e5
~See footnotes, Table 10-III. 2See text for explanations of low totals.
tO
268
R.B. Perkins and A.L. Foster
pyrite in sample HS-565 appears to be the only Fe sulfide with a V concentration (0.15%) significantly greater than that measured in the bulk-rock (0.04%). Cu-V sulfides, typically in association with sphalerite, were observed in several samples from the lower Lakeridge section. Analyses of subhedral Cu-V sulfide crystals (~10 Ixm across) in LR-M21 indicate average S, Cu, and V contents of 34.4, 48.8, and 8.2%, respectively. The S average closely matches that of stoichiometrically ideal sulvanite (Cu3VS4; 34.7% S) while the Cu and V averages are lower than that of ideal sulvanite (51.6% Cu and 13.8% V). However, sulvanite forms a series with arsenosulvanite (Cu3(As,V)S4).Although As was not measured, such a phase containing 8.2% V (Xv = 0.61) would contain 7.7% As, very close to the average residual of 7.4%. The difference between the measured and ideal S content (< 0.6%) is within the likely analytical error. The deficiency in Cu (< 1.5%) could be due to substitution of other transition metals, including Ag, which was not measured. The chemistry of this phase thus closely matches that of an As-bearing sulvanite, that is enriched in both Cr and Se (average of 6700 and 4500 ppm, respectively) and, to a lesser degree, in Ni and Zn (< 1000 ppm).
X-ray absorption spectroscopy
X-ray absorption near edge spectra (XANES) are useful in determining bulk oxidation state because the position of the absorption edge (i.e. primary inflection point) of Se K-edge XANES spectra shifts as a function of Se valence (Pickering et al., 1995). In this study, XANES spectra of reduced (sulfide associated or native) Se model materials had measured absorption-edge energies between 12,656.5 and 12,658.5 eV, spectra of Se(IV) model materials had edge energies between 12,662.0-12,663.0 eV, and Se(VI) model compounds had edge energies between 12,666.0 and 12,667.0 eV (Fig. 10-4). The measured absorption-edge position of Lakeridge sample spectra averaged 12,658.8 ___0.08 eV, consistent with Se in a reduced form. In contrast, the edge positions of Enoch Valley spectra averaged 12,662 ___0.10 eV, consistent with those of Se(IV) reference materials. Selenium K-edge XANES spectra are also sensitive to chemical speciation. Such spectra of model materials have previously been used as "fingerprints" to identify and quantify bulk Se species in soils, fungi, and aqueous solutions by linear, least-squares fitting (Pickering et al., 1995). In this study, least-squares fits indicate that the XANES spectral lineshape of upper Lakeridge samples (LR) M37a, M36-40, and M41a are well matched by the XANES spectrum of the same seleniferous pyrite/marcasite material used in REF03 (fits not shown, but compare spectra in Fig. 10-4A). Least-squares fits to XANES spectra of Lower Lakeridge samples (LR) M14 and M16-18 indicate that seleniferous pyrite and Cu-sulfides or Cu-selenides are the primary hosts for Se, although it should be noted that the fits to these samples are poorer than the fits to the upper Lakeridge samples. XANES spectra of Enoch Valley samples EV-76 and EV-195 are best fitted by the XANES spectrum of Se(IV) adsorbed on synthetic vernadite (MnO2), whereas the spectrum of EV-185 is best fitted by the spectrum of Se(IV) adsorbed on synthetic manganite (MnOOH; fits not shown, but compare spectra in Fig. 10-4B). This result is somewhat
Se(0)-red, amorphous Se(0) black, amorphous
"--'-~..~
Se(IV)-MnO2 (vernadite)
Seleniferous pyrite/marcasite
Se(IV)-MnO OH (manganite) ~....
,,_._...._ Se(IV)-Fe oxyhydroxide
~
Se(IV)-kaolinite
LR-M41A LR-M37-40 LR-M37A
1
"""-"-
r~
EV-P195
Selenocysteine EV-P186 LR-M14 c~ -.-------- EV-P76
LR-M16-18 Cu(I)Se
/ i , t l l l l l l a , , l l , l l l l l l l l
12.65 12.66 12.67 12.68 12.69 12.70 Enerav (KeV~
~.,.~,~.~._Aqueous Se(Vl) l
I l
II
11
II
II
ll
II
II
,
I
II
II
I.
12.65 12.66 12.67 12.68 12.69 12.70 Enerav (KEY)
Fig. 10-4. Comparison of X-ray absorption spectra from selected samples and relevant reference materials; note the higher (0.3-0.4 KeV) K-edge energies for Se(IV) reference materials and Enoch Valley samples in right-hand plot.
270
R.B. Perla'ns and A.L. Foster
problematic as both bulk chemical and extraction results indicate these samples are not particularly enriched in Mn. Furthermore, these samples contain ample Fe oxyhydroxides, which have been shown to scavenge Se(IV) more effectively than MnO2 (Balistrieri and Chao, 1990). For this reason, we believe that the results of the least-squares fingerprinting analysis may be in error in this case. One possible reason for the error could be that the model XANES spectrum used to represent Se(IV) sorbed to amorphous Fe oxyhydroxides is not representative. Evidence for this argument comes from the XANES spectra of Se(IV) sorbed to the Mn oxyhydroxides in Fig. 10-4B; the Mn oxyhydroxides are structurally and compositionally distinct and Se(IV) adopts different sorption geometries on the two substrates, producing distinctly different XANES spectra. An analogous situation might hold for the Se(IV)-Fe oxyhydroxide system, but further work is needed to determine if this is true. Regardless of this result, XANES analysis and extended X-ray absorption fine structure (EXAFS) analysis (not discussed here) clearly indicate that Se(IV) in the Enoch Valley samples is associated with metal oxyhydroxides.
S e q u e n t i a l extractions Reference materials
Although recoveries from reference samples provide some measure of the effectiveness of the technique (Tables 10-Va and 10-Vb), they may not accurately reflect recoveries from natural samples in which other secondary reactions (sorption and precipitation) may occur. For sequential-extraction Scheme A, the total recovered Se for each material was within acceptable weighing and analytical measurement errors of the total expected Se (< 20% difference). However, full recovery from the specific target extract was not achieved in three of the four reference materials. Extraction of the NazSeO3-spiked material (REF-01) resulted in measurable Se concentrations in all five extracts although 71% was recovered in the initial target extraction ("P-buffer") and 94% was recovered in the first two extractions (P-buffer; K2S2Os). The "bleed-over" may have been due, in part, to inadequate rinsing of the first extract. The complete lack of recovery of Se in the target extraction step for the Scheme B reference material suggests either analytical error or that precipitation or sorption of selenite may occur despite acidification and refrigeration of extracts. This process may be strongly time dependent as the time between extraction and analyses of Scheme B reference samples (~ 7 weeks) was considerably longer than for the Scheme A reference samples (< 2 weeks). The percentage of elemental Se recovered from REF-02 during the target (Na2SO3) extraction was 74 and 94%, respectively, for Schemes A and B. The recovery of 10.3% of the total Se in REF-02 during step 2 of Scheme A suggests greater oxidation of elemental Se by the K2S208 extract than was measured by Martens and Suarez (< 1.6%; 1997b). Extraction of elemental Se as a second step, as in Scheme B, is therefore suggested for future studies. In both Schemes A and B, significant amounts of the measured Se (15.4 and 7.6%, respectively) were recovered subsequent to the target step. This may indicate remobilization or could result from other forms of Se in the reagent; the manufacturer
TABLE 10-V
t~
Selenium concentrations (mg kg-1 of solid and % of total) in various fractions from quality assurance reference materials using (a) Scheme "A" sequential partial-extraction method and (b) Scheme "B" sequential partial-extraction method
Reference 2
Scheme A Fractions I
Total Selenium
1
2
3
4
5
Fraction sum
Calculated
%Difference 3
(a) BLK-01-A
< 0.41
< 0.40
< 0.40
< 0.40
< 0.41
< 2
0
0
REF-01-A Na2SeO3 REF-02-A E1 Se REF-03-A
63 _ 3 (71.0%) < 0.41 < 0.41
20 ___2 (23.0%) 37.1 ___0.9 (10.3%) < 0.41
0.8 -4- 0.05 (0.9%) 268 +__2 (74.3%) < 0.41
0.6 (0.7%) 4.9 +_ 0.4 (1.4%) < 0.41
3.8 +__0.1 (4.4%) 51 ___4 (14.0%) 2.7 +__0.2
88 +_ 5
Se-pyrite REF-05-A CuSe
1.4 + 1.3 (0.23%)
28 + 12 (5%)
110 + 30 (19%)
_
0.7 + 0.1 (0.12%)
361 +_ 7 2.7 -+- 0.2
(100%)
440 + 30 (76%)
580 + 80
Scheme B Fractions ~
(b) BLK-01-B
REF-01-B Na2SeO3 REF-02-B
670 14 (w/in measurement error) Total Selenium
1 (same as #1 in Scheme A)
2 (same as #3 in Scheme A)
3
Fraction sum
< 0.40
< 0.40
5.3 _+ 0.0
5.3 _+ 0.4
< 0.41
15.3 +__0.09 (17.2%) 340 4- 18
11.7 +__0.05 (13.4%) 27 ___3
27 -Z-_0.1 (30.5%) 367 +__21
< 0.40
100 12 (w/in measurement error) 440 18 (w/in weighing error) Unknown -
Calculated
t~ t~ t~ t~
% Difference 3
(no measurable Se in scheme A) 100 73 420
13 Ix.)
Continued
..
tO tO
TABLE 10-V Continued Reference 2
E1 Se REF-05-B fuSe
Scheme B Fractions 4
Total Selenium
1 (same as #1 in Scheme A)
2 (same as #3 in Scheme A)
3
Fraction sum
< 0.41
(94%) 29 __+4 (5.0%)
(7.6%) 297 ___11 (51%)
(102%) 326 ___11 (56%)
Calculated
% Difference 3
(w/in weighing error) 670 51
1Fraction 1, "Soluble + Exchangeable"; 2, "Organic"; 3, "Elemental"; 4, "Oxide + Carbonates + Apatite + Acid-volatile sulfides"; 5, "Crystalline sulfides". Errors are standard deviations based on analyses of duplicate samples; no errors listed if no measurable difference in duplicates. 2BLK-01" Blank consisting of 4.00 g iron-poor silica sand + 0.50 g activated charcoal; REF-01" Blank + 0.1 ml of 0.057 M NaSeO3 solution; REF-02: Blank + ~2 mg elemental (red) Se (Pfaltz & Bauer); REF-03" Blank + ~7 mg Se-rich marcasite (unknown Se concentration or purity); REF-05: Blank + ~5 mg Cu (II) selenide (Alfa-Aesar, 99.5%). 3percent difference = Calculated (expected) Total - Fraction Sum/Calculated Total. 4Fraction 1 -"Soluble + Exchangeable", same method as #1 in scheme A; 2 -"Elemental", same method as #3 in scheme A; 3 -"Organic", alternate method; errors are standard deviations based on analyses of duplicate samples; no errors listed if no measurable difference in duplicates; % of totals based on totals as measured in "full" extraction scheme A; organic (3rd) fraction concentrations in reference materials corrected for Se detected in blank.
b~ r~
Mineral affinities and distribution of selenium and other trace elements
273
(Pfaltz and Bauer) did not certify purity. Amorphous red Se prepared by reduction of NaSeO3 by ascorbic acid (only one of many methods for the preparation of this material) can contain small amounts of Se(IV), as indicated by a positive 1-2 eV shift in absorption edge position of this material relative to monoclinic or hexagonal forms of native Se (Fig 4A). Se(IV) present in red, amorphous Se(0) may exist as an adsorbed species or it may be occluded in the Se(0) matrix. Our previous work indicates that repeated washings with purified water do not remove it. All measurable Se in REF-03 was obtained from the final extract (~16 M HNO3, 90~ as expected for Se in pyrite. Recovery efficiency for REF-03 could not be calculated as the actual Se concentration and overall sample homogeneity of the pyrite/marcasite is unknown. Given the low total recovery (2.7 ppm), significant amounts of Se may have been solubilized in previous steps but still not detected. However, a prior study by Shannon and White (1991) found that 92% of the Fe added as FeS2 was recovered in their sulfide-targeted extraction, suggesting that the measurable Se in our extracts reflects the majority of the Se present in the reference material. The measured Se recoveries from REF-05 (CuSe) demonstrate potential difficulties in targeting monoselenides or monosulfides. Significant amounts of Se were extracted in both schemes during steps designed to target only elemental or organic-bound Se. Extraction with NaOC1 in Scheme B resulted in the liberation of nearly half (46%) the total recovered Se. Shannon and White (1991) also found that monosulfides (up to 25% of spiked FeS) could be extracted prematurely using MgCI2 and NaOAc extractants. In general, analyses of the reference materials reveal that the sequential-extraction methods used result in less than 100% recovery of target phases and may dissolve some portion of non-target phases. Nonetheless, Scheme A appears to result in reasonable total Se recoveries (> 80%) and specificity (71-100% of recoverable Se liberated in target extraction) and the results of the sample analyses should be useful for determining the overall significance of specific phases in hosting trace elements and in providing estimates of trace-element distributions. Samples
As with the reference materials, there are some discrepancies between Schemes A and B with respect to the amount of Se liberated in the initial ("P-buffer") extraction (Tables 10-Via and b). However, in neither scheme does the percentage of total recoverable Se extracted during the initial step exceed 5% and only in HS-565-B and LR-M 16-18-B does the percentage of total recoverable Se liberated in the first step exceed 2%. Precipitation or sorption of Se from the extract is unlikely to account for the low totals obtained from the P-buffer extract as many of the Scheme A samples were submitted and analyzed along with the reference materials for which 71% of the spiked selenite was recovered. Elemental Se appears to be a minor phase by either scheme, accounting for less than 5% of the total recoverable Se in all but one of the Scheme A samples (EV-P 186, 8%). Two Scheme B samples, EV-P186 and LR-M36-40, have average elemental Se contents of 11 and 16%, respectively. However, LR-M36-40 also has the greatest duplicate variability so that the percent of total Se occurring as elemental Se is uncertain. No elemental Se was
tO --..I 4~
TABLE 10-VI(a) Selenium concentrations (mg kg-1 of solid and % of total) in various fractions of rock samples
Scheme A Fractions I Sample 1
EV-P60 EV-P76 EV-P186 HS-521-77 HS-565-77 HS-218-74 LR-M14-15 LR-M16-18 LR-M36-40 LR-M51
Total Selenium
1
2
3
4
5
0.50 (0.2%) 0.4 _+ 0.04 (0.5%) 0.19 _+ 0.26 (0 - 1.0%) 2.7 _+ 0.09 (2%) < 0.40
1.0 (0.4%) 6.1 _+ 0.05 (7%) 3.9 _+ 0.4
0.00 (0.0%) < 0.40 3.2 _+ 0.2
30 _+ 5 (11.5%) 20 _+ 14 (22%) 14.0 _+ 0.7
229 _+ 4 (88%) 61 _+ 1 (71%) 17.6 _+ 0.6
(10%)
(8%)
(36%)
(46%)
11.3 (8.2%) 1.50
4.56 (3.3%) < 0.40
34.5 _+ 0.2 (25%) 5.3
86.1 _+ 0.1 (63%) 4.0 _+ 0.08
-
(14%)
-
(49%)
(37%)
1.98 (2%) 4.1 -+ 0.04 (2%) 0.6 (0.2%) 0.5 (0.8%) 0.25 _+ 0.35
11.3 -+ 0.4 (11%o) 12.3 -+ 0.1 (6%) 21.7 _+ 0.8 (7%) 2.76 (4%) 5 _+ 1
2.6 _+ 0.3 (2.6%) < 0.40 13.6 _+ 0.2 (4%) 2.0 _+ 1.9 (3%) 2.8 -+ 0.04
18 -+ 1 (18%) 42.9 _+ 0.7 (21%) 62 _+ 6 (20 _+ 2%) 5.60 (9%) 5.3 _+ 0.07
67.5 _+ 0.4 (67%) 142 _+ 1 (71%) 218 _+ 8 (69%) 52.6 _+ 0.4 (83%) 35 -+ 5
(10%)
(4%)
(11%)
(75%)
(0.0-1.2~
Fraction Sum
261 ___9
Analyzed in Bulk Rock
% Difference
287
9.1
87 _+ 14
124
30
39___2
73
47
137 ___4
120
-14
10.8 ___0.1
11.7"
7.7
101 _+2
111"
9.0
202 _+ 2
209
3.3
316 _+ 10
287
-10
63 _+ 2
55
- 15
47 _+ 6
32.1
-46
t~ ~t
TABLE 10-VI(b) t~
Comparison of selenium concentrations (mg kg-1 of solid) in various rock fractions determined by two different sequential extraction methods on select rock samples t.,~. t,,~.
Fractions 3 t~
EV-P186 Scheme A
(1) Soluble/ Sorbed
Scheme A (2) Organic
0.19 (0.50%) < 0.40
3.9 (10%)
Scheme B HS-521-77 Scheme A
2.7 (2%) < 0.40
Scheme B LR-M14-15 Scheme A Scheme B
<0.40
Scheme A (4)
Scheme A (5)
14
17
8.1 (21%)
3.2 (8%) 4.2 (11~
35
29 (21%)
2.3 (2%) 8.4 (6%) < 0.40
5.3
1.5 (14%)
Subtotal: 32 -
< 0.40
< 0.40
12 (6%)
16 (8%)
RPD Org 5
% Total After 1 - 3 6
39
70%
82
T o t a l - X(1-3): 27
69
86 Subtotal: 121
137
88%
T o t a l - ~(1-3): 100
88 73
4.0 10
128%
T o t a l - ~(1-3): 3.5 43
93 35
142 Subtotal: 185
42 (21~
Total 4
-
Subtotal: 9.3 6.8 (63~
0.53 (5%) 4.1 (2%) 0.53 (0.3%)
(A-3 or B-2) Elemental
11 (8%)
Scheme B HS-565-77 Scheme A
Scheme B (3) Organic
T o t a l - ~(1-3): 143
202
109%
92 71
Continued tO
t~
TABLE 10-VI(b) Continued Fractions 3
Sample 3 LR-M16-18 Scheme A
(1)Soluble/ Sorbed 0.56 (0.2%) 10
Scheme A (2) Organic 22 (7%)
Scheme B
0.50 (1%) 0.50 (0.9%)
-
44 (15%)
(4%)
Scheme B LR-M36-40 Scheme A
Scheme B (3) Organic
2.8 (4.4%)
-
18 (17%)
(A-3 or B-2) Elemental 14 (4%) 14 (5%)
2.0 (3%) 10 (16%)
Scheme A (4)
Scheme A (5)
62
Total 4
RPD Org s
% Total After 1-36
316
67%
89
218 Subtotal: 280
Total- ~(1-3)" 248 5.6
78
53 Subtotal" 58
-
63
146%
Total- ~(1-3): 44
70
1See footnote 2, Table 10-Va. 2percent difference = Analyzed Total as determined by triple acid digestion and HGAAS - Fraction Sum/Analyzed Total. Note: Se concentrations denoted by * were determined by ICP-MS and may be lower than actual (by up to ~25%) due to partial volatilization during digestion. 3See footnotes, Table 10-Va 4Total as measured with Scheme A 5Relative percent difference in "organic" fraction by separate methods: Scheme B Value- Scheme A Value/Average 6percent of total (as measured using Scheme A) remaining after steps 1-3 ("soluble" + "organic" + "elemental").
92
Mineral affinities and distribution of selenium and other trace elements
277
detected in samples EV-P60 or EV-P76 by Scheme A or in sample HS-565 by either method (Tables 10-VIa and b). The amount of "organically bound" Se determined by Scheme B (step 3) is higher in every sample than the amount determined by Scheme A (step 2) with a relative percent difference between 40 and 150% (Table 10-VIb). Although the organic-Se fraction appears significant in all but one sample (EV-P60; 0.4% total Se), extraction with K2S208 results in no more than 14% of the total recoverable Se. Using Se-rich plant residues, Martens and Suarez (1997b) found the K2S208 extract could liberate as little as 40-60% of the organic Se in a sample, depending on the types of organic matter involved. Conversely, extraction with NaOC1 may liberate significantly more non-target phases, as demonstrated with the CuSe reference material (REF-05). However, even assuming that the NaOC1 extraction more accurately represents the total organic-Se fraction, this fraction accounts for less than 25% of the total recoverable Se. There is a strong possibility that neither of the extraction methods were able to liberate 100% of the organically bound Se present in these samples. Despite the samples being ground to < 150 Ixm, some organic matter likely remained occluded within apatite phases, escaping oxidation. A portion of this occluded matter would be liberated concurrently with dissolution of the apatite during the fourth step. Acidsoluble organics may not be digested until extraction step 4 and highly humic-rich material may not be digested until extraction step 5. The fourth extraction in Scheme A is the least specific in that it digests oxyhydroxides, apatite, carbonates, and acid-volatile organics and sulfides. However, the EMP analyses and previous studies of trace-element fractionation help to constrain the source of trace elements liberated in this step. Manning and Burau (1995) found an average of only 3.2% of total recoverable Se was liberated from the carbonate (NaC2H302 at pH 5) fraction of Kesterson Reservoir sediments. Factor analyses of bulk-rock chemical data from Meade Peak samples (unpublished data; Piper, 1999b) suggest that none of the elements investigated in this study are associated with carbonates although Cr, Cu, Ni, and U may be associated with apatites. The EMP analyses (Table 10-III) indicate little, if any, enrichment of Cr, Cu, or Ni above bulk-rock levels, although the total amount of these elements hosted by apatite in highly phosphatic samples (e.g. HS-218, HS-565, LR-M51) may still be significant. XRD analyses show the dissolution of both apatite and carbonate minerals and the preservation of pyrite and sphalerite in this step (Table 10-VII). Oxyhydroxides and apatite can, therefore, be considered the primary sources of trace elements liberated in this step, with additional contributions by acid-volatile sulfides and occluded acid-soluble organic matter as discussed above. An estimate of the effectiveness of the HCI treatment in dissolving carbonate and apatite is provided by the Ca and Sr concentrations measured in extracts (Table 10-VIII). Calcium may be present in various silicates that would not be extracted with the treatments used; however, most of the Ca is anticipated to be present in either apatite or carbonates and should be liberated during the fourth step with little Ca present in the last ("crystalline sulfide") extract. Likewise, bulk-rock chemistry suggests Sr is associated with P205 (correlation coefficient of 0.86; Piper, 1999), presumably due to substitution for Ca in apatite. The percentage of total recoverable Ca extracted in the final step exceeded 10% in three
278
R.B. Perla'ns and A.L. Foster
TABLE 10-VII X-ray diffraction mineralogy of initial samples and solid residues after selected sequential partial-extractions steps Components Sample/fraction I
Major
Moderate
Minor/Trace
Quartz Quartz
CFA, dolomite, albite Albite
Goethite
CFA
Quartz
Quartz
Muscovite
Albite, buddingtonite(?), muscovite K-feldspar
CFA CFA Quartz
Quartz Calcite, Quartz Albite, Ti-oxide
Buddingtonite(?), muscovite Albite, graphite K-feldspar
CFA, quartz Quartz
Illite, bixbyite, pyrite Albite, K-feldspar
Quartz
Dolomite Muscovite, pyrite/sphalerite Muscovite
CFA, quartz Quartz
Dolomite Albite
Pyrite K-feldspar, muscovite, Mg/Fe-silicate(?), celsian(?)
CFA CFA
Calcite Calcite
Quartz, pyrite, Cu-sulfide(?) Barite/celestine Cu-sulfide/selenide (?) gypsum (assumed product)
Dolomite Quartz
CFA, quartz, calcite,
Albite, goethite(?) Albite, illite
Dolomite, quartz Dolomite, quartz
CFA CFA
A: Step 4
Quartz
A: Step 5
Quartz
Albite, muscovite, pyrite Albite, muscovite
Calcite, muscovite, pyrite Albite, calcite, K-feldspar, muscovite, pyrite K-feldspar
EV-P60 Initial A: Step 5 EV-P76 Initial A: Step 5 EV-P186 Initial B: Step 3 A: Step 5 HS-218 Initial A: Step 4 A: Step 5 HS-521 Initial A: Step 5 HS-565 Initial A: Step 2
LR-M14-15 Intial A: Step 5 LR-M16-18 Initial B: Step 3
Albite, K-feldspar
K-feldspar Continued
279
Mineral affinities and distribution o f selenium and other trace elements
TABLE 10-VII Continued Components Sample/fraction I LR-M36-40 Initial A: Step 3 A: Step 4 A: Step 5
Major
Minor/Trace
Moderate
Albite, Cu-sulfide(?), illite, muscovite, pyrite Illite, muscovite, pyrite
CFA, dolomite, Quartz CFA, dolomite, Quartz Quartz Quartz
Illite, muscovite, pyrite Illite, K-feldspar, muscovite, Ti-oxide(?)
~See footnote 1, Table 10-Va; note: CFA, carbonate fluorapatite.
TABLE 10-VIII Percent of total concentration of three matrix indicators extracted in each fraction or remaining in solid residual
Element (total concentration) EV-P60 Ca (3.8%)* Fe (2.2%) Sr (139 ppm)* EV-P76 Ca (11.0%) Fe (2.4%) Sr (680 ppm)* EV-P186 Ca (8.7%) Fe (2.2%) Sr (522 ppm)* HS-218 Ca (22.2%) Fe (0.84%) Sr (606 ppm)*
% of Bulk-rock total Soluble + Sorbed (%)
Apatite/Carb Organic (%) +Oxide/AVS (%) Sulfide (%) Residual (%)
1.0 0 0.6
7.2 0.0 6.6
88.9 25.3 89.3
2.9 31.4 3.5
0 43.3 0
0.2 0 0
3.1 0 5.4
78.2 43.3 88.2
5.1 22.3 12.9
13.5 34.4 0
0.1 0 0
3.9 0.4 4.3
86.1 63.9 90.1
4.0 4.6 5.8
5.9 31.1 0
0 0 0.1
1.3 0.5 2.0
65.1 15.8 69.4
27.4 73.2 28.5
6.1 10.5 0 Continued
280
R.B. Perkins and A.L. Foster
TABLE 10-VIII Continued Element (total concentration) HS-521 Ca (9.0%) Fe (1.5%) Sr (371 ppm) HS-565 Ca (33.4%) Fe (0.17%) Sr (965 ppm) LR 14-15 Ca (16.3%) Fe (1.0%) Sr (367 pm) LR 16-18 Ca (13.8%)* Fe (1.2%) Sr (417 ppm)* LR 36-40 Ca (12.4%) Fe (2.3%) Sr (445 ppm) LR 51 Ca (25.1%) Fe (0.44%) Sr (1925 ppm)*
% of Bulk-rock total Soluble + Sorbed (%)
Apatite/Carb Organic (%) +Oxide/AVS (%) Sulfide (%) Residual (%)
0.3 0.0 0.6
2.6 0.1 4.5
91.6 9.4 85.0
2.6 47.6 2.7
2.9 39.3 7.2
0 0 0
0.9 0 1.5
73.4 20.9 77.5
23.8 44.2 21
1.9 34.9 0
0.1 0 0.4
1.5 0.4 3.1
80.4 11.8 85.3
9.7 42.2 5.3
8.2 25.2 6.0
0.2 0 0
2.0 0.5 3.9
95.5 26.1 92.1
2.4 50.3 4.1
0 23.1 0
0.1 0 0.3
1.9 0.2 2.5
91.1 5.6 92.4
6.9 51.2 4.6
0 59.7 0.3
0 0 0.1
1.1 0.5 4.5
78.0 21.8 77.9
21.1 59.6 17.6
0 18.1 0
Note: Totals determined by ICP-AES or - M S after 4-acid digestion except for elements denoted by * where fraction sums > measured totals.
samples, HS-218, HS-565, and LR-M51, the only samples with Ca contents > 20 wt.%. The percentage of total recoverable Sr in the fifth extract exceeded 10% in four samples (HS-218, HS-565, LR-M51, EV-P76), the samples with the highest bulk-rock Sr contents, although the amount of Sr extracted from EV-P76 during the last step was much less than the others. The high levels of both Sr and Ca in HS-218, HS-565, and LR-M51 suggest incomplete dissolution of both carbonates and apatite. It is possible that 20-28% of the other elements liberated in the final extract in these three samples could be from the apatite or carbonate fractions. However, neither apatite nor carbonate phases were identified by XRD analyses in HS-218 residues remaining after the fourth step (Table 10-VII) and previously discussed considerations, including EMP analyses, suggest the trace-element
Mineral affinities and distribution of selenium and other trace elements
281
content contributed by these phases would be minor except for U. Stronger HC1 (e.g. 6 M) or a secondary treatment is suggested for future studies. The percentage of total recoverable Se liberated in step 4 ranges from 9 to 50% and averages 18%. Given the very low loadings for Se on factors associated with apatite (< 0.01, unpublished data; Piper, 2001), we conclude that the Se in this fraction is primarily released by dissolution of Fe or Mn oxyhydroxides, although some may be from acid-volatile sulfides and selenides. Analyses of Fe contents in the extracts (Table 10-VIII) indicate that while most of the Fe is liberated in the final extract or remains in the solid residue, up to 64% (in EV-P 186) is liberated during step 4, presumably by dissolution of iron oxyhydroxides. XRD analyses (Table 10-VII) indicate the absence of previously measurable goethite in solid residues from sample EV-P60 following step 4. The largest fraction of Se is obtained from the final extraction. The fraction accounts for 37-88% (average = 67%) of the total recoverable Se using Scheme A. XRD data show that pyrite, sphalerite, and unidentified Cu sulfides are the most important phases dissolved during this step (HS-218, LR-M 16-18, LR-M36-40; Table 7). Final residual solids consist primarily of quartz and other silicates with minor amounts of insoluble oxides (primarily Ca- and Ti-oxides) and barite. The extraction techniques used in this study were designed for selective liberation of Se. Thus, interpretation of the resulting data for transition metals requires special consideration. The step of greatest uncertainty is in the extraction designed to liberate elemental Se via reaction with sulfite. However, the concentration of other elements extracted during this step averages < 2.3% of their recoverable totals except for Mo (average of 8.2%) and Cu (6.9%). Given that the detected concentrations of most of the elements is likely a carryover from the previous (organic) extraction step and the lack of firm reaction mechanisms for Cu and Mo, the concentrations measured in this step are added to the organic fraction for all of the elements in the discussion that follows. Cadmium, Cr, U, and Zn appear to have distinctly different mineral affinities from one another and from Se (Tables 10-IX-XII). Excluding the Enoch Valley samples, Cd (Table 10-IX) is similar to Se in that most appears in the final extract (average 58.5%). However, much more of the Cd in the Enoch Valley samples is liberated in step 4 (60.7%) and relatively little ( < 6%) in step 5. A significant portion of the total Cr remained in the final solid residue. This is especially true for samples with relatively high concentrations of Cr (> 1000 ppm; Table 10-X). Uranium is distinctly absent from the organic fraction and heavily concentrated in the oxyhydroxide-apatite-carbonate fraction (Table 10-XI). This is consistent with the measurements of Zielinski et al. (Chapter 9) that show U is associated mainly with apatite in Meade Peak rocks, and with bulk-rock chemical studies that show that U is largely correlated with apatite (Piper, 1999; Piper et al., 2000). The only samples from which > 50% of the total recoverable U was obtained in the last (step 5) fraction are HS-565 and LR-M51, the only two samples with P205 contents > 25% (Table 10-I) and two of the three samples for which Ca and Sr data suggest incomplete dissolution of apatite and carbonates during step 4. Concentrations of Zn show greater variability in the duplicate analyses than do those of other elements; however, total recoveries closely match total bulk-rock Zn measurements
bo oo bo
TABLE 10-IX Cadmium concentrations (mg kg-1 of solid and % of total) in various fractions of rock samples Sample I
EV-P60 EV-P76 EV-P186 HS-521-77 HS-565-77 HS-218-74 LR-M14-15 LR-M16-18 LR-M36-40 LR-M51
Scheme A fractions I 1
2
3
1.2 _+ 0.04 (1%) <0.01
27.5 _+ 0.3 (21%) 11.7 (26%) 41 _+ 5 (37%) 51.1 _+ 0.3 (20%) 0.04 _+ 0.03 (0.3%) 51.8 _+ 0.6 (33%) 29.8 _+ 0.08 (23%) 72 _+ 3 (21%) 21.5 _+ 0.5 (18%) 9.0 _+ 0.4 (22%)
15 _+ 3 (11%) 1.7 _+ 0.2 (4%) 3.9 _+ 0.04 (4%) 1.80 (0.3%) 0.01 (0.1%o) 0.84 (0.5%) 2.6 _+ 0.2 (2%) 6.7 _+ 0.3 (2%) 1.3 _+ 0.04 (1%) 0.12 (0.3%)
0.06 _+ 0.09 (0-0.11%) 1.18 (0.5%) <0.01 0.25 (0.2%) 0.3 _+ 0.04 (0.3%) 0.60 (0.2%) <0.12 <0.01
ISee footnotes, Tables 10-Va and I 0-VIa.
Total cadmium 4
84 _+ 3 (63%) 28.8 _+ 0.6 (64%) 60 _+4 (55%) 41.1 _+ 0.04 (16%) 6.6 _+ 0.05 (42%) 23.1 _+ 0.4 (15%) 28.7 _+ 0.1 (21%) 71 _+ 2 (21%0) 20.0 _+ 0.2 (16%) 6.8 _+ 1.3 (16%)
5
6.4 _+ 0.1 (5%) 2.7 (6%) 5 _+ 1 (5%) 167 _+ 2 (64%) 9.2 _+ 0.08 (58%) 83.4 _+ 0.8 (52%) 71.04 (54%) 195 _+ 6 (57%) 78.7 _+ 0.9 (65%) 28 _+ 14 (60%)
Fraction sum
134 _+6
Analyzed in bulk rock
% Difference I
150
11
46
2.2
110_+7
120
8.3
262 _+ 2
276
5.0
45 _+ 0.8
15.8 _+ 0.1
16.8
5.9
159_+2
178
11
133 _+0.3
149
11
345 _+7
423
18
121 _+ 1
147
18 r~
44_+16
41
-7.3
TABLE 10-X
t~
Chromium concentrations (mg kg-l of solid and % of total) in various fractions of rock samples Sample ~
3
4
5
Fraction sum
Analyzed in bulk rock
13.2 • 0.9 (4~ 5.12 (0.5%) 2 . 2 • 1.3 (0.4%) < 1.2
1.20 (0.3%) 4.5 • 0.4 (0.5%) 1.20 (0.2%) < 1.2
490
27
940 • 20
3600
74
<0.06
158 • 6 (44~ 414 • 4 (44%) 310 • 20 (61%) 246 • 1 (72%) 120 • 1 (77%) 493 • 8 (88) 214 • 3 (62%) 223 • 5 (60%) 140.1 • 0.9 (60%) 600 • 95 (88%)
360 • 10
1.80 (1%) 3.3 • 0.4
184 • 9 (52%) 520 • 20 (55%) 194 • (38%) 96.7 (28%) 33.9 • 0.9 (22%) 67 • 2 (12%) 131 • 3 (38%) 150 • 9 (40%) 94 • 2 (40%) 81 • 2 (12%)
1
EV-P60
< 1.2
EV-P76
0.9 • 0.5 (0.1%) <0.03
EV-P186 HS-521-77
< 1.2
HS-565-77
<0.06
HS-218-74
<0.06
LR-M14-15
<1.2
LR-M16-18
<0.06
LR-M36-40
< 1.2
LR-M51
Total Chromium
Scheme A fractions I
<0.06
2
1.8 (0.5%) 0.10 • 0.14 (0 - 0.02%) < 1.2 3.6 (0.5%)
1See footnotes, Tables 10-Va and 1O-VIa.
0.12 (0.02%) <1.2 0.40 < 1.2 <0.06
% Difference I
t~ t~
510 • 20
8400
94
343 • 1
1910
82
156 • 2
182
14
563 • 9
695
19 t%
347 • 4
1670
79
373 • 10
1750
79
234 • 2
1190
80
685 • 96
697
1.7
t~
OO
tO O0
TABLE 10-XI Uranium concentrations (mg kg-! of solid and % of total) in various fractions of rock samples
Sample ~
Scheme A fractions ~ 1
EV-P60
<0.12 -
EV-P76 EVoP 186 HS-521-77 HS-565-77 HS-218-74 LR-M14-'15
6.2 (13%) 6.2 (10%) <0.12 _ 3.2 _+ 4.5 (0-8.2%) 0.09 _+ 0.04 (0.10%) <0.12 -
LR-M16-18 LR-M36-40
6.2 (18%) <0.12 _
LR-M51
<0.12 -
2
<0.12 0.13 (0.26%) 0.7 _+ 0.3 (1%) <0.12 _ <0.12 1.5 _+ 0.04 (2%) <0.12 <0.12 <0.12 _ 2.5 _+ 0.3 (4%)
1See footnotes, Tables 10-Va and 10-VIa.
3
0.18 (1%) 0.6 _+ 0.04 (1%) 3.5 4_ 0.3 (5%) 0.36 (0.5%) <0.12 0.6 _+ 0.04 (0.6%) 0.5 (2%) 0.42 (1%) 0.24 _+ 0.08 (2%) 0.39 _+ 0.05 (0.6%)
Total Uranium 4
12.6 (90%) 34.8 _+ 0.6 (71%) 41 _+ 2 (64%) 36.0 _+ 0.6 (94%) 13.9 _+ 0.1 (19%) 60 _+ 6 (62%) 14.1 _+ 0.3 (49%) 18.8 (55%) 13.9 _+ 0.01 (85%) 10.4 _+ 0.2 (16%)
5
1.3 _+ 0.1 (9~ 7.6 _+ 0.1 (16%) 13.3 _+ 0.4 (21%) 2.2 _+ 0.04 (6%) 57.5 _+ 0.1 (77%) 34.1 _+ 0.3 (36%) 14.0 _+ 0.5 (49%) 9.1 _+ 0.1 (26%) 2.1 _+ 0.04 (13~ 53 _+ 4 (80%)
Fraction sum
Analyzed in bulk rock
% Difference I
14.1 __ 0.1
11
-28
49.3 -+ 0.6
48
-2.7
65_+2
69
5.7
39.4
2.5
75_+5
83
9.6
96_+6
102
5.9
38.4 _+ 0.6
28.5 _+ 0.8
<100
34.5 _+ 0.1
<100
16.2 _+ 0.1
<100
66_+4
<100
TABLE 10-XII Zinc concentrations (mg kg-! of solid and % of total) in various fractions of rock samples q..~ ~~
Sample I
Scheme A fractions I 1
EV-P60 EV-P76 EV-P186
7.2 • 0.04 (0.3%) <1.8 < 1.8 -
HS-521-77 HS-565-77
19.3 • 0.2 (0.5%) < 1.8 -
HS-218-74 LR-M14-15 LR-M16-18
8.6 • (0.3%) 8• 2 (0.4%) <1.8 -
LR-M36-40 LR-M51
0.00 (0.0%) <1.8 -
2
411 • 8 (17%) 298 • 8 (25%) 100 • 25 (40%) 978 (27%) < 1.8 1360• 10 (45%) 651 • 4 (34%) 1040 • 40 (25 %) 663 • 21 (25%) 470 • 45 (29%)
l See footnotes, Tables 10-Va and 1O-VIa.
3
94 • 14 (4%) 11.5 • 0.1 (1%o) < 1.8 35.1 < 1.8 14 • (0.5%) 46 • 2 (2%) 102 • 11 (2%) 33 • 6.8 (1%) 3.1 • 0.4 (0.2 %)
Total Zinc 4
1190 • 60 (48%) 659 • 15 (55~ 121 • 4 (48%) 413 • 4 (12%) 180 • 4 (45%) 327 • 17 (11%) 292 • 1.7 (15%) 680 • 45 (16%) 345 • 2.3 (12%) 119 • 25 (7%)
5
765 • 32 (31%) 241 • 3 (20%) 31 • 2 (12%) 2140 • 34 (60%) 221 • 1 (55%) 1310 • 10 (43%) 909 (48%) 2400 • 100 (57%) 1752 • 22 (63%) 1000 • 460 (63%)
Fraction sum
Analyzed in bulk rock
% Difference 1
2470 • 70
2700
8.5
1210 • 20
1300
6.9 t~ t~
253 • 25
280
9.6
3570 • 34
3350
-6.6
401 • 4
434
7.6
3020 • 40
3000
-0.7
1910 • 5.8
1920
0.7
4200 • 110
4580
8.3
2790 • 48
2960
5.8
1600 • 500
1710
6.4
"X
t~ t~
to oo
286
R.B. Perkins and A.L. Foster
Fig. 10-5. Concentrations of Cu, Mo, Ni, and V in operationally defined mineral fractions (Exch = readily exchangeable; Org = organically bound; Ox = oxides + apatite + carbonates; Sulf = sulfides) in selected samples from largely unweathered (Hot Springs and Lakeridge) and weathered (Enoch Valley) sections. Percentage of measured bulk-rock element concentration recovered from all sequentialextraction steps is given in legends, y-axis scales are the same for each element in both columns. (<10% relative difference; Table 10-XII). A significant portion of total Zn (-----25% for all but one sample) is present in the organic fraction. However, the greatest portion of Zn appears to be in the crystalline sulfide fraction in the Hot Springs and Lakeridge samples, consistent with the occurrence of sphalerite. Similar to Cd, the Zn content is higher in the oxyhydroxide-apatite-carbonate fraction than in the crystalline-sulfide fraction in Enoch Valley samples. The majority of Cu is extracted in the final (crystalline sulfide) step in the Hot Springs and Lakeridge samples (Fig. 10-5). However, most of the Cu in the Enoch Valley samples
Mineral affinities and distribution o f selenium and other trace elements
287
is liberated during the fourth step, similar to Cd and Zn. The percentage of Cu liberated in the organic fraction averaged 15%. Significant portions of Mo in Hot Springs and Lakeridge samples were present in both the soluble/readily exchangeable fraction (average 9.2%) and in the organic fraction (average 28%) (Fig. 10-5). Again, the Enoch Valley samples were different in that little Mo was present in these two fractions (average 0 and 4.3%, respectively) and that the samples had lower overall Mo contents. The average Mo content liberated during step 4 from the Hot Springs and Lakeridge samples exceeded that liberated in step 5 by a factor of 1.4. Ni shows a persistent and strong affinity with organic matter in all samples (11-55% of total recoverable Ni, averaging 34%; Fig. 10-5). Most of the Ni in the Hot Springs and Lakeridge samples (> 40% in each) appears to be associated with crystalline sulfides. However, an average of 45% of the total recoverable Ni was liberated from the Enoch Valley samples during extraction step 4. The vast majority of recoverable V in samples from the Hot Springs and Lakeridge sections (average 70%) was liberated in the final extract (Fig. 10-5). The Enoch Valley samples are again distinct in that most of the recoverable V (average 66%) was released by the step 4 extraction. Total V recoveries, as determined by comparison with bulk-rock sample analyses, are less than 70% for six samples and less than 50% for four samples (EV-P 186, HS-521, LR-M16-18, LR-M36-40). This indicates that most of the V in these samples is present in silicates or in other insoluble phases. The fraction sums for the other four samples average 89% of total V. As two of these samples have the highest V contents (> 1000 ppm), very high V enrichment apparently does not result from these insoluble phases. The concentration of elements measured in residual solids could not be directly compared to the sequential-extraction results because sample weight loss was not monitored through the extraction procedure. Nonetheless, analyses of residual solids from five samples by ICP-MS indicated between 2000 and 5000 ppm Cr. This clearly represents the majority of the total Cr, consistent with the observed difference between the bulk-rock Cr and the sum total of Cr measured in all of the extracts. Likewise, V concentrations of between 200 and 2000 ppm were measured in the solid residues, accounting for the previously noted differences between bulk and extract V concentrations. Less than 1 ppm to a few tens of ppm of Cd, Cu, Mo, Ni, Se, and U were detected while 100-200 ppm of Ni and Zn were detected in three of five samples. These concentrations would represent < 1% (for Cd) to perhaps 10-12% (for Ni) of the total measured element contents if corrected for mass loss. While these concentrations could conceivably represent association with insoluble residual phases, they could just as well be the result of analytical errors, inadequate rinsing, or re-adsorption during the last extraction step.
DISCUSSION The results of the XAS analyses indicate that the dominant form of Se in the Enoch Valley samples is Se(IV). These results conflict with those of the sequential-extraction study, which show that the greatest portion of Se is liberated in the final extraction step,
288
R.B. Perkins and A.L. Foster
which is supposed to target sulfides. A possible explanation is that a high degree of crystallinity in Se-bearing oxyhydroxides may have prevented their full dissolution in extraction step four. Chao and Zhou (1983) found that less than 50% of some well-crystallized Fe oxides were dissolved after boiling in 4 M HC1 for 30 min. Another possibility is that Se(IV) initially liberated from Enoch Valley samples during step 4 was re-adsorbed on the surfaces of other phases from which it was then released during step 5; expectedly, this was not found for transition metals, as indicated by their high concentrations in the fourth extract. Sulfide mass-balance calculations were used as an overall check on the sequentialextraction results. Available S was estimated from bulk-rock sulfur contents corrected for S in apatite and organics. The corrections used assume average apatite S and P205 contents of 1.2 and 36.3% (0.5 and 38.2% for Hot Springs samples), respectively, based on EMP analyses (Table 10-III), a moderately high S content of 5% in organic matter, and an organic-carbon content of 72% in organic matter based on a generalized stoichiometry of C4H20. Molar Fe:S calculated from the Fe content measured in the final extract and the estimated available S are 6.1 and 1.4, respectively, in EV-P60 and EV-P76. These values are much higher than the ratios calculated for the other samples (~ 0.30), and for ideal pyrite (0.50), indicating that these two Enoch Valley samples have excess reactive Fe. Futhermore, the maximum estimated pyrite content in EV-P60, based on the available S, is simply too low to account for the amount of Se present in the final extract unless we assume an unrealistically high (-19%) Se content. These calculations are consistent with the XAS results and indicate incomplete dissolution of crystalline Fe oxyhydroxides in step 4. The mass-balance calculations are also consistent with the results of the solid-phase analyses of the Hot Springs and Lakeridge samples. The estimated net available S content is sufficient to account for a suitable mass of sulfides in all of these samples using the EMP-measured average pyrite-Se content of 5000 ppm (Table 10-IV). The Fe contents measured in the final extract are more limiting with respect to the amount of pyrite that might be present and require pyrite-Se concentrations of 10,700-13,100 ppm in samples LR-M14-15 and LR-MI6-18 to achieve the Se contents measured in the final extracts (Table 10-XIII). Nonetheless, these values are within the ranges of Se contents measured by EMP. Furthermore, the average and maximum Se contents of pyrite in lower Lakeridge samples LR-M 12 and LR-M21, which bracket LR-M 14-15 and M 16-18 in the section, were the highest measured (up to 19,500 ppm; Table 10-IV). Similar calculations can be made to verify or constrain the sequential-extraction results for other elements. For example, all of the Cd released in step 5 (> 50% of the total in Hot Springs and Lakeridge samples) can be attributed to sphalerite, based on the measured Zn and estimated S concentrations and assuming a Cd content of 0.5-4.7%, values that are within the range measured by EMP (Table 10-IV). A majority of Cu (average of 72.4%) is associated with the sulfide fraction in Hot Springs and Lakeridge samples. Copper contents varied widely in pyrite, from not detectable to 1.4%, and were consistently low in sphalerite, averaging < 450 ppm. Pyrite Cu contents of 0.5-12.5% Cu would be required to account for the Cu concentrations
289
Mineral affinities and distribution o f selenium and other trace elements
TABLE 10-XIII Calculated maximum wt.% pyrite and equivalent Se content in pyrite that would be required for measured Se content in sulfide fraction Sample
EV-P60 EV-P76 EV-P 186 HS-218 HS-521 HS-565 LR-M14-15 LR-M 16-18 LR-M36-40 LR-M51
Wt.% Fe Available for sulfides 1
Wt.% S Available for sulfides 2
Molar ratio of available Fe:S
Wt.% equivalent Pyrite 3
Measured Se in sulfide fraction (ppm)
Calculated equivalent Se in Pyrite (ppm)
0.99 0.77 0.15 0.62 1.03 0.07 0.62 0.86 1.69 0.38
0.07 0.23 0.45 1.16 2.25 0.17 3.05 4.05 3.28 1.00
8.12 1.92 0.19 0.31 0.26 0.23 0.12 0.12 0.30 0.22
0.12 0.42 0.32 1.33 2.21 0.15 1.33 1.85 3.63 0.82
230 62 17 68 86 4 143 242 52 39
189,000 14,600 5,200 5,100 3,900 2,400 10,700 13,100 1,500 4,800
~Determined from wt.% of total Fe in final extraction. 2Determined from wt.% of total S, corrected for 1.2 wt.% S content in apatite (0.5% for Hot Springs samples) and 5.0 wt.% S content in organic matter (except EV-P60 and EV-P76, for which only a 3.0% S content in organic matter is used). Available S = Bulk-rock S - 0.033 (or 0.013 for HS-samples) x Bulk-rock P205 - 0.05 x Bulkrock total organic carbon x 1.37; where 1.37 is assumed conversion of organic carbon to organic matter based on generalized stoichiometry of C4H20. 3Wt.% pyrite based on lowest value calculated from either available Fe or S, assuming ideal stoichiometry.
determined from the final extract. Therefore, a substantial fraction of Cu must be present in other phases, such as sulvanite. The measured Cu concentrations sharply constrain the amount of sulvanite that might be present to <- 0.03%, indicating that this phase is insignificant as a host for trace elements other than Cu or V. Molar ratios of V:Cu, based on the amounts present in the final extracts, exceed those of stoichiometric sulvanite in all sampies, indicating that any excess Cu may be attributed to sulvanite but that another phase is required to account for the excess V. Excess V in the final fraction cannot be attributed to commonly observed sulfides given the low V concentrations measured in pyrite and sphalerite, although unidentified, possibly cryptocrystalline sulfides cannot be ruled out. A significant amount of V remained in the final solid residue in some samples. A V-rich Ti-oxide phase, which appeared as aggregates ( < 50 txm in length) composed of micrometer crystallites, was
290
R.B. Perkins and A.L. Foster
observed in solid residues retained from the EV-P60 sequential extraction. Partial dissolution of more resistant phases is not the source of V in the final extracts as two of the highest V contents were measured in the final extract in samples for which the total V recovery was 85 and 98% of that measured in bulk-rock analyses. A significant part of recoverable V (10--40%) appears associated with the oxyhydroxide-apatite-carbonate fraction. This V may be in the form of V205 which could have precipitated from seawater under suboxic conditions (Piper, 1999), although a V-bearing phosphate, phosphovanadylite, has been identified in Phosphoria rocks (Medrano et al., 1998). The percentage of recoverable Ni in the organic fraction (average of 33.7%) is higher than the percentages of other elements in this fraction, consistent with high Ni contents in kerogen, bitumen, and crude oil, which typically exceed those of other metals except V (Giordano, 2000). The average Ni content in organic matter is calculated as 1400 ppm, based on the Ni concentrations liberated in step 2 and average total organic carbon (TOC) contents, and assuming that TOC constitutes ~70% of the organic matter. This value is well under the maximum Ni content reported for kerogen (2160 ppm) in the New Albany Shale (Filby, 1994). The Ni contents liberated in extraction step 5, however, require between 0.9 and 5.0% Ni in pyrite (average 2.9%). While NiS2 forms a partial solid solution with FeS2 (up to 10 mol.% NiS2; Deer et al., 1962) the calculated values for most samples far exceed those measured in pyrite, sphalerite, or sulvanite crystals by EMP analyses (maximum of 2.0% in framboidal pyrite). We suggest that the Ni liberated in this step is from a mixture of framboidal pyrite, cryptocrystalline sulfides, and possibly humic-rich organic matter. Total Mo contents are < 100 ppm in all but a few samples. The percentage of total recoverable Mo that is soluble or exchangeable exceeds 5% except in the Enoch Valley samples and LR-MI6-18. The lack of soluble-exchangeable or significant organic-bound Mo in any of the Enoch Valley samples suggests these fractions are readily lost during weathering. Non-soluble, non-exchangeable Mo is rather evenly partitioned between organic, crystalline-sulfides, and oxyhydroxides-apatite-carbonate fractions, though slightly more occurs in the latter fraction in most samples (average of 36.4%, excluding Enoch Valley samples). This may reflect the presence of Ca or Mg molybdates; for example Desborough (1977) identified calcium molybdate (powellite?) in one Coal Canyon sample. However, persistent detection of Mo in apatite pellets in some samples (Table 10-III) suggests incorporation of MoO42 in the apatite structure. Although it might be expected that Mo(VI) would be reduced in organic-rich sediments, some previous studies suggest that Mo reduction may not take place at "higher pH" (Wood, 2000). EMP analyses also indicate that large, euhedral pyrites in LR-M37 are enriched in Mo with average concentrations of 6560 ppm. Interestingly, these same pyrites are depleted in Cu, Ni, and Se compared to pyrites in other samples (Table 10-IV). The majority of Cr was retained in the final solid residue (Table 10-X) except in the four samples with < 1000 ppm total Cr. Thus, extremely high enrichments of Cr are due to the presence of relatively insoluble, Cr-rich phases. SEM examination of solid residues remaining after the sequential extraction study showed masses of relatively equidimensional, colloidal-sized material in which only Cr and O could be detected. This phase is likely either Cr(OH)3 or Cr203 that may have precipitated directly from seawater during
Mineral affinities and distribution o f selenium and other trace elements
291
deposition of Meade Peak sediments (Piper, 1999). Another phase consisted of elongated to blocky, sub-micrometer to micrometer grains that may be a F e - C r - N i silicate. Desborough (1977) concluded that the principal host of Cr in unweathered samples from the Maby Canyon mine was mica, which would also be consistent with the poor extraction recoveries.
CONCLUSIONS The results of this study indicate that the largest fraction of Se in unweathered or minimally weathered samples (i.e. Hot Springs and Lakeridge samples) is associated with sulfides. Most of the non-sulfide Se is associated with organic matter and oxyhydroxides and a small amount is present in elemental form, as reported by Grauch et al. (1999, Chapter 8). Se(IV) associated with oxyhydroxides is the dominant form of Se in more weathered (i.e. Enoch Valley) samples, implying oxidation of primary sulfide and organic Se host phases. The high TOC content (up to 18 wt.%) and the scarcity of sulfides in weathered samples indicate that sulfide minerals are preferentially lost relative to organic matter during weathering. The mineral affinities for Se and other trace elements in weathered and unweathered rocks are summarized in Table 10-XIV. Selenium concentrations > 2000 ppm were measured in sphalerite, > 5000 ppm in Cu-V sulfides (sulvanite), and up to 2% in subhedral pyrite present within the matrix and as inclusions in phosphate pellets. Framboidal pyrite and masses of cryptocrystalline sulfides are ubiquitous in Meade Peak rocks, especially within organic-rich matrix material, and may contain even higher concentrations of Se as well as other elements of environmental concem. Cadmium and Zn are hosted primarily by sphalerite and organic matter in unweathered samples while strong sorption on oxyhydroxides appears to dominate in weathered sampies. Copper is hosted primarily by pyrite and sulvanite in unweathered samples. The fraction of Cu hosted by organic matter is more variable than for Cd and Zn, but may exceed 35%. Sorption on oxyhydroxides or incorporation in amorphous, acid-volatile sulfides appears to account for the majority of Cu in weathered samples. A significant amount (11-55%) of the Ni in Meade Peak rocks, weathered or not, is present in organic matter. However, the largest fraction of the Ni in unweathered rocks is in sulfides, likely in framboidal pyrite or cryptocrystalline masses, which appear much more enriched in Ni than larger, more well-developed pyrite. Again, there is a shift from sulfides to oxyhyroxides in more weathered samples. Molybdenum, although typically less abundant than Ni, is also strongly associated with organic matter which may account for > 40% of the Mo in some samples. A smaller but significant fraction of Mo is also relatively soluble and is easily lost during weathering, as is the organic-bound fraction. Molybdenum occurs in relatively high concentrations in some sulfides (> 5000 ppm). A greater percentage of Mo appears to be present either as acid-soluble molybdates or in apatite, although measured concentrations of Mo in apatite are relatively low (< 650 ppm). Sorption of Mo on oxyhydroxides cannot be ruled out.
292
R.B. Perkins and A.L. Foster
TABLE 10-XIV Summary of trace element mineral affinities in unweathered and weathered rocks Element
Unweathered and minimally weathered rocks (Lakeridge, Hot Springs)
Weathered rocks (Enoch Valley)
Se
SULFIDES Organic matter Selenides? Oxyhydroxides Elemental Se SULFIDES Organic matter
OXIDES Organic matter Elemental Se
Cd, Cu, Zn Ni
Mo
U
Cr
V
ORGANIC MATTER SULFIDES Oxides OXIDES &/or APATITE Organic matter Sulfides Soluble or loosely sorbed APATITE Occluded organic matter? Soluble or loosely sorbed REFRACTORY PHASES (Cr Oxide, hydroxide; Fe-, Cr silicates) Acid-soluble oxides SULFIDES REFRACTORY PHASES (Ti-, V oxide?) Oxides &/or apatite
OXIDES Recalcitrant organic matter Cu: occluded sulfides? OXIDES ORGANIC MATTER Occluded sulfides? OXIDES &/or APATITE Occluded sulfides or organic matter?
APATITE Occluded organic matter? Soluble or loosely sorbed REFRACTORY PHASES (Cr oxide, hydroxide; Fe, Cr silicates) Acid-soluble oxides OXIDES &/or APATITE REFRACTORY PHASES Occluded sulfides?
Note: Predominant host phase(s) denoted by all upper case letters.
Uranium is chiefly present in apatite (see Zielinski et al., Chapter 9) as also evidenced here by bulk-rock chemical associations between U and phosphate and an average recovery of 73% of the total U in extraction step 4 for samples with nearly complete apatite and carbonate dissolution. Little V is present in organic matter, despite its common enrichment in bitumen, kerogen, and crude oil (Giordano, 2000). Instead, the majority of V in unweathered samples is present in sulfides (sulvanite and unidentified phases) and acidsoluble vanadates or oxides. Insoluble (residual) oxides, and either soluble oxyhydroxides or V-bearing phosphates (Medrano et al., 1998) account for > 60% of V in weathered samples. High Cr concentrations ( > 1000 ppm) in Meade Peak rocks are the result of
Mineral affinities and distribution o f selenium and other trace elements
293
insoluble (residual) phases. Although Cr20 3 or Cr(OH)3 are likely the most important of these, Fe-Cr silicates have also been observed. The results of this study show a dramatic shift from reduced forms of Se in unweathered rocks to Se(IV) in weathered rocks. However, the high Se bulk-rock concentrations of some of the weathered rocks (e.g. 287 ppm in sample EV-P60) attest to the fact that the oxidized Se is not necessarily readily lost and may be sorbed to Fe oxyhydroxides formed in response to weathering. The concentrations of other elements that occur predominantly as sorbed phases in weathered rocks can also be relatively high in comparison to their crustal abundance (e.g. 376 ppm Cu in EV-P 186). These results imply that: (a) transport of elements from Meade Peak rocks to the environment as sorbed species on carrier phases is important and needs to be considered along with dissolved species; and (b) release of these elements from weathered rocks and particles will be strongly dependent on the pH, Eh, and exchangeable ion contents of surface and groundwater. Furthermore, Se and associated trace elements in Meade Peak rocks are hosted in a number of phases that have variable oxidation rates and a wide range of particle sizes. Some of these host phases occur as inclusions within phases more resistant to oxidative weathering, such as phosphate pellets. These factors indicate that the release of these elements to the environment in southeast Idaho will be a variable and long-term process.
ACKNOWLEDGEMENTS The authors are grateful for the reviews and suggestions of Andrea Koschinsky of the International University Bremen, Dean Martens of the USDA-ARS Southwest Watershed Research Center, and Richard Sanzolone, USGS, Denver. We would also like to thank Paul Lamothe, USGS, Denver, for providing timely ICP-MS analyses of extracts and Brandie Mclntyre and Michelle Lopez for their help with XRD analyses. This work was partially funded by the USGS Mendenhall Postdoctoral Fellowship Program.
REFERENCES Balistrieri, L.S. and Chao, T.T., 1990. Adsorption of selenium by amorphous iron oxyhydroxide and manganese dioxide. Geochim. Cosmochim. Acta, 54: 739-751. Briggs, P.H., 1996. Forty elements by inductively coupled plasma-atomic emission spectrometry. In: B.F. Arbogast (ed.), Analytical Methods Manual for Mineral Resources Surveys Program. US Geological Survey, Open File Report, 96-525, pp. 45-62. Briggs, P.H. and Meier, A.L., 1996. The determination of forty-two elements in geological materials by inductively coupled plasma - mass spectrometry. In: B.E Arbogast (ed.), Analytical Methods Manual for Mineral Resources Surveys Program. US Geological Survey, Open File Report, 96-525, pp. 76-88. Chao, T.T. and Zhou, L., 1983. Extraction techniques for selective dissolution of amorphous iron oxides from soils and sediments. Soil Sci. Soc. Am. J., 47: 225-232.
294
R.B. Perkins and A.L. Foster
Chao, T.T. and Sanzolone, R.E, 1989. Fractionation of soil selenium by sequential partial dissolution. Soil Sci. Soc. Am. J., 53: 385-392. Coleman, R.G. and Delevaux, M., 1957. Occurrence of selenium in sulfides from some sedimentary rocks of the western United States. Econ. Geol., 52: 499-527. Curry, K.J., 1996a. Total carbon by combustion. In: B.E Arbogast (ed.), Analytical Methods Manual for Mineral Resources Surveys Program. US Geological Survey, Open File Report, 96-525, pp. 99-104. Curry, K.J., 1996b. Total sulfur by combustion. In: B.E Arbogast (ed.), Analytical Methods Manual for Mineral Resources Surveys Program. US Geological Survey, Open File Report, 96-525, pp. 105-110. Deer, W.A., Howie, R.A. and Zussman, J., 1962. Rock-Forming Minerals, Non-Silicates. John Wiley and Sons, Inc., New York, NY, 371 pp. Desborough, G.A., 1977. Preliminary report on certain metals of potential economic interest in thin vanadium-rich zones in the Meade Peak Member of the Phosphoria Formation in western Wyoming and eastern Idaho. US Geological Survey, Open File Report, 77-341, 27 pp. Filby, R.H., 1994. Origin and nature of trace element species in crude oils, bitumens, and kerogen. In: J. Parnell (ed.), Geofluids: Origin, Migration, and Evolution of Fluids in Sedimentary Basins. Geological Society of London, pp. 203-219. George, G., 1992. EXAFSPAK: A Suite of Computer Programs for Analysis of X-ray Absorption Spectra. Stanford Synchrotron Radiation Laboratory, available from authors. Giordano, T.H., 2000. Organic matter as a transport agent in ore-forming systems. In: T.H. Giordano, R.M. Kettler and S.A. Wood (eds.), Ore Genesis and Exploration: The Roles of Organic Matter: Reviews in Economic Geology. Society of Economic Geologists, Boulder, CO, pp. 133-156. Grauch, R.I., Meeker, G.P., Desborough, G.A., Tysdal, R.G., Herring, J.R., Moyle, P.R. and Anonymous, 1999. Selenium residence in the Phosphoria Formation, Geological Society of America Annual Meeting, Boulder, 35 pp. Gulbrandsen, R.A., 1979. Preliminary analytical data on the Meade Peak Member of the Phosphoria Formation at Hot Springs underground mine, Trail Canyon Trench and Condo underground mine, southeastern Idaho. US Geological Survey, Open File Report, 79-369, 37 pp. Hageman, P. and Welsch, E., 1996. Arsenic, antimony, and selenium by flow-injection or continuous flow - hydride generation-atomic absorption spectrophotometry. In: Arbogast, B.E (ed.), Analytical Methods Manual for Mineral Resources Surveys Program. US Geological Survey, Open File Report, 96-525, pp. 18-24. Lamothe, P.L., Meier, A.L. and Wilson, S., 1999. The determination of forty-four elements in aqueous samples by inductively coupled plasma-mass spectrometry. US Geological Survey, Open File Report, 99-151, 14 pp. Manning, B.A. and Burau, R.G., 1995. Selenium immobilization in evaporation pond sediments by in situ precipitation of ferric oxyhydroxide. Environ. Sci. Technol., 29: 2639-2646. Martens, D.A. and Suarez, D.L., 1997a. Selenium speciation of marine shales, alluvial soils, and evaporation basin soils of California. J. Environ. Qual., 26: 424-432. Martens, D.A. and Suarez, D.L., 1997b. Selenium speciation of soil/sediment determined with sequential extractions and hydride generation atomic absorption spectrophotometry. Environ. Sci. Technol., 31: 133-139. Medrano, M.D. and Piper, D.Z., 1995. Partitioning of minor elements and major element oxides between rock components and calculation of the marine derived fraction of the minor elements in rocks of the Phosphoria Formation, Idaho and Wyoming. US Geological Survey, Open File Report, 95-279, 79 pp.
Mineral affinities and distribution o f selenium and other trace elements
295
Medrano, M.D., Evans, H.T., Jr., Wenk, H.-R. and Piper, D.Z., 1998. Phosphovanadylite; a new vanadium phosphate mineral with a zeolite-type structure. Am. Mineralog., 83: 889-895. Murata, K. J., Friedman, I. and Gulbrandsen, R. A., 1972. Geochemistry of carbonate rocks in Phosphoria and related formations of the western Phosphate Field. US Geological Survey, Professional Paper, 800-D, pp. D 103-D 110. Papp, C., Brandt, E. and Aruscavage, P., 1996. Carbonate carbon by coulometric titration. In: B.E Arbogast (ed.), Analytical Methods Manual for Mineral Resources Surveys Program. US Geological Survey, Open File Report, 96-525, pp. 38--44. Pickering, I.J., Brown, G.E., Jr., and Tokunaga, T.K., 1995. Quantitative speciation of selenium in soils using X-ray absorption spectroscopy. Environ. Sci. Technol., 29: 2456-2458. Piper, D.Z., 1999. Trace elements and major-element oxides in the Phosphoria Formation at Enoch Valley, Idaho; Permian sources and current reactivities. US Geological Survey, Open File Report, 99-0163, 66 pp. Piper, D.Z., Skorupa, J.P., Presser, T.S., Hardy, M.A., Hamilton, S.J., Huebner, M. and Gulbrandsen, R.A., 2000. The Phosphoria Formation at the Hot Springs M i n e - a source of selenium and other trace elements to ground water, surface water vegetation and biota. US Geological Survey, Open File Report, 00-050, 75 pp. Ressler, T., 1998. WINXAS: A program for X-ray absorption spectroscopy data analysis under MS-Windows. J. of Synchr. Radiat., 5:118-122. Sayers, D.E. and Bunker, B.A., 1988. Data Analysis. In: D.C. Koningsberger and R. Prins (eds.), X-ray Absorption: Principles, Applications, Techniques of EXAFS, SEXAFS, and XANES. John Wiley and Sons, New York, NY, pp. 211-225. Shannon, R.D. and White, J.R., 1991. The selectivity of a sequential extraction procedure for determination of iron oxyhydroxides and iron sulfides in lake sediments. Biogeochemistry, 14: 193-208. Sheldon, R.P., 1963. Physical stratigraphy and mineral resources of Permian rocks in western Wyoming. US Geological Survey, Professional Paper, 313-B, 273 pp. Tokunaga, T.K., Pickering, l.J. and Brown, G.E., Jr., 1996. Selenium transformations in ponded sediments. Soil Sci. Soc. Am. J., 60: 781-790. Velinsky, D.J. and Cutter, G.A., 1990. Determination of elemental selenium and pyrite-selenium in sediments. Anal. Chim. Acta, 235: 419-425. Wood, S.A., 2000. Organic matter: supergene enrichment and dispersion. In: T.H. Giordano, R.M. Kettler and S.A. Wood (eds.), Ore Genesis and Exploration: The Roles of Organic Matter: Reviews in Economic Geology. Society of Economic Geologists, Boulder, CO, pp. 157-192. Zhang, Y. and Moore, J.N., 1996. Selenium fractionation and speciation in a wetland system. Environ. Sci. Technol., 30:2613-2619.
This Page Intentionally Left Blank
PART IV.
G E O E N V I R O N M E N T A L STUDIES
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
299
Chapter 11
THE P H O S P H O R I A FORMATION: A M O D E L FOR FORECASTING GLOBAL SELENIUM SOURCES TO THE ENVIRONMENT
T.S. PRESSER, D.Z. PIPER, K.J. BIRD, J.P. SKORUPA, S.J. HAMILTON, S.J. DETWILER and M.A. HUEBNER
ABSTRACT Mining of the Permian Phosphoria Formation- a marine, oil-generating, phosphatic s h a l e - provided the selenium (Se) source implicated in the recent deaths of livestock in southeast Idaho. Field studies and the geohydrologic balance of Se in southeast Idaho confirm risk to animals from exposure to Se through leaching of mined waste shale into streams, discharge of regional drainage, and impoundment of drainage in wetland areas. Forage grown to stabilize waste rock contoured into hills or used as cross-valley fill provides an additional mechanism of Se exposure for the environment (Mackowiak et al., Chapter 19). The average Se concentration of the Meade Peak Member of the Phosphoria Formation is an order of magnitude higher than those of other exploited marine shales that have been linked to incidences of Se toxicosis via oil refining and irrigation in the western United States. The Phosphoria Formation accumulated in an environment that preserved organic matter and contributed to the formation of economic-grade phosphate and oil deposits. The addition of this phosphate-mining case study enables a comprehensive approach to the identification of marine sedimentary Se sources and a more complete range of ecotoxic field studies on which to establish the conditions and anthropogenic connections that determine uptake, release, and recycling of Se in food webs. A constructed conceptual model of Se pollution indicates that ancient organic-rich depositional marine basins, unrestricted by age, are linked to the contemporary global distribution of Se source rocks. A global plot shows (a) the areal association of major basins hosting phosphate deposits and petroleum source rocks and (b) the importance of paleo-latitudinal setting in influencing the composition of the deposits. Given the geographic patterns, Se emerges as a contaminant within specific regions of the globe that may limit phosphate mining, oil refining, and drainage of agricultural lands because of potential ecological risks to vulnerable food webs. Selenium also may serve as a geochemical exploration tool that signals an ancient productive biological environment.
300
T.S. Presser et al.
INTRODUCTION The initial accumulation and elevated concentrations of phosphate and other trace nutrients in organic-rich marine sedimentary rocks, specifically black shales, petroleum source rocks, and phosphorites, depend on the role of these elements in the primary productivity of the oceans (Sheldon, 1981; Cutter and Bruland, 1984; Piper, 1994). For Se, these fundamental processes extend further to the ability of bacteria, algae, fungi, and plants to synthesize Se-containing amino acids de novo (Stadtman, 1974). Once Se-containing biological compounds are established in food webs, they are consumed by progressively more complex species (Presser et al., 1994). Dietary levels exceeding those necessary for nutrition affect the reproductive system, immune system, and growth, including embryo deformities in higher-level predators (US Department of the Interior, 1998). These processes of primary productivity in the Permian Phosphoria sea led to selected trace-element deposition, largely as organic detritus, on the sea floor under denitrifying conditions (Piper, 2001). Elevated concentrations of phosphate, Se, and other trace nutrients (e.g. Cd, Ni, Mo) in the rocks resulted from the loss of roughly 90% of the host organic matter, mostly during early diagenesis via bacterial respiration, retention of nutrients by the sediment, and an extremely low-accumulation rate of otherwise diluting phases. Concentrations of phosphate of up to 40% (as P205) were achieved. The Phosphoria Formation currently is still highly organic-rich (up to 15% total organic carbon) based on amounts (> 1.5-2% residual organic carbon) necessary to qualify for petroleum exploration, as well as phosphorite exploitation (McKelvey et al., 1959, Claypool et al., 1978). In this chapter, we present data from field studies of Se pollution in the western United States, including those for southeast Idaho, to illustrate the biogeochemical pathways of Se in the environment. Concentrations of Se in geologic sources, receiving waters, food webs, and vulnerable species show the range and severity of ecological impacts from such anthropogenic activities as irrigation of agricultural land, refining of petroleum, and mining of phosphate when applied to Se source rocks. The addition of phosphorites as a category of Se-containing rocks to that of other carbon-rich source rocks enable a forecast of global Se sources. Although black shales and their recoverable organic fractions are widely recognized as sources of trace elements, the implications of worldwide reservoirs, sitespecific fluxes, and persistent biologic cycling of Se are not. However, enough is known in terms of guidelines for health and risk to establish a first-order understanding of effects, should Se be discharged into vulnerable environments.
METHODS AND SOURCES OF DATA Selenium guidelines
Compilations of Se health and risk criteria are available for: (a) nutrition, adequate/chronic toxicity (Puls, 1988; US Department of Health and Human Services,
The Phosphoria Formation -forecasting modelfor global selenium sources
301
1989, 1996); (b) protection of aquatic life (US Environmental Protection Agency, 1987; US Department of Health and Human Services, 1996); (c) protection of animal life (Puls, 1988); (d) human health advisory, consumption of fish (Fan et al., 1988); and (e) human health advisory, drinking water (US Department of Health and Human Services, 1996). Guidelines for risk to aquatic life (i.e. ecological thresholds) take into account food webs in that they were derived for water, sediment, diet, fish tissue, and bird eggs (Stanley et al., 1996; US Department of Interior, 1998). The criterion for a hazardous Se waste is designed to protect ground water from the leaching of toxic substances based on an extraction procedure of solid wastes (US Department of Health and Human Services, 1996). Values for solids are given in dry weight, except as noted.
Western United States (Colorado River watersheds, San Joaquin Valley, and San Francisco Bay-Delta Estuary, California) Seawater Se concentrations for the North Pacific are reported in Bruland (1983) and in Cutter and Bruland (1984). Selenium source-rock data are taken from compilations in Presser (1994) and Piper and lsaccs (1995). Extensive data sets (e.g. 2055 bird eggs) for drainage, receiving waters, food webs, and predator species of concern from the western United States are non-random representing contaminated sites only, except for coot and duck eggs (Presser and Ohlendorf, 1987; Saiki and Lowe, 1987; Ohlendorf and Hothem, 1994; Presser, 1994; Hamilton, 1998; Skorupa, 1998; US Department of Interior, 1998; Luoma and Presser, 2000). Where avian eggs were collected regionally, including uncontaminated reference sites, the minimum-to-maximum range was used. Where only contaminated sites were sampled, the regional minimum is unconfirmed and only the maximum -I was used (i.e. ?-maximum). It is expected, however, that regional minima are < 1 Ixg g Se for coot and duck populations (Skorupa, 1998).
Idaho Selenium source-rock data are taken from compilations in McKelvey et al. (1986) and Piper et al. (2000). The methodology for measuring Se concentrations in bird eggs from Idaho that are reported here is a fluorescence-based micro-digestion (Fan et al., 1997). Collection methodologies and extent and duration of sampling for other environmental media in southeast Idaho have evolved since 1997 under cooperative agreements between mining companies and federal agencies, with the most impacted area monitoring in accordance with a federal Comprehensive Environmental Response, Compensation, and Liability Act (TRC, 1999; Montgomery Watson, 1998, 1999, 2000, 200 la,b). The number of samples in each category for Idaho varies (rocks, n = 378; wasterock seeps and drains, n = 35; receiving waters, n = 231; benthic invertebrates, n = 67; avian eggs, n = 74 including 27 from American coot; forage fish, n = 53; gamefish, n = 61). Samples of sheep liver are limited due to conditions of carcasses (n = 7) (Piper et al., 2000). Selenium concentrations in water samples collected in association with bird
302
T.S. Presser et al.
egg samples that are shown here graphically were analyzed by fluorescence-based microdigestion (Fan et al., 1997).
Global distribution of phosphate deposits and petroleum basins The distribution of world phosphate deposits was compiled from Notholt et al. (1989) and the distribution of world productive petroleum (oil and gas) basins was adapted from Klemme and Ulmishek (1991). Klemme and Ulmishek (1991) divided petroleum source rocks into those with dominantly types I and II kerogen (oil prone) and those containing dominantly type III kerogen and coal (gas prone). Thirty-one of the 47 petroleum basins considered in the analysis are of type II kerogen (marine oil shales). However, 13 are of type III kerogen and/or coal (continental deposits) and three are of type I kerogen (mainly lacustine deposits). In the future, it may be productive to expand our coverage to include coal deposits and power production that uses seleniferous coal because local extinctions of fish populations have occurred from contamination from fly ash that was elevated in soluble Se (Lemly, 1996; Skorupa, 1998). The temporal distribution of major phosphate deposits was based on data from Cook and Shergold (1986) and Notholt et al. (1989). Identified phosphate resources (in million tonnes) were compiled by country, distributed by age, and normalized to a total tonnage for the United States, Latin America and Mexico, Asia and the Pacific, Africa, the Middle East, Europe, and Central Eurasia. Time distribution of effective petroleum source rocks was adapted from Klemme and Ulmishek (1991). They normalized the data as percentages of the original global petroleum reserves generated by these rocks. Grouping by age of percentages of petroleum source rocks is different in two instances from those shown by Klemme and Ulmishek (1991) to accommodate depiction with phosphate resources: the Oligocene and Miocene originally grouped together as 12.5% are represented as 6.25% in both the Oligocene and Miocene; the PennsylvanianEarly Permian originally grouped together as 8.0% are represented as 4% in both the Pennsylvanian and Early Permian. These generalizations affect individual percentages for phosphate and petroleum, but not the overall conclusions shown by the time distributions.
CONCEPTUAL MODEL The marine pathway through the environment accounts for the largest natural flux of Se (Haygarth, 1994). The marine system is also the major sink in the global cycle of Se. However, the current release of Se to the environment shows that the mobilization through anthropogenic activities is larger than the total flux of natural marine, terrestrial, and atmospheric sources on an annual basis (Haygarth, 1994). Haygarth (1994) concluded that this escalated rate of release over natural transfer in the environment indicates human interference is a major factor in the distribution, fate, and transport of Se.
The Phosphoria Formation -forecasting modelfor global selenium sources
303
We introduce here a conceptual model of Se pollution, annotated with environmental concentrations of Se, which illustrates specific biogeochemical pathways connected to irrigation, petroleum refining, and mining in the western United States. Overall the model shows the Se cycle from sources through food chain pathways to vulnerable predators. Processes important for Se cycling and defining the toxicity of Se in the environment, Se enrichment, Se mobilization, and establishment of Se-containing biological compounds, form the basis for the model. The model shows the diversity of environments that are affected including wetlands, rivers, estuaries, and forage lands. Several important ecosystems in the western United States are documented including those of the Colorado River basins, the San Francisco Bay-Delta Estuary, the San Joaquin Valley, California, and the Blackfoot River watershed, Idaho. The model encompasses: 9 oceanic depositional environments; 9 organic carbon-rich marine sediments that are source rocks for petroleum and phosphate ores; 9 specific Se source rocks; 9 anthropogenic activities that facilitate transfer to the environment; 9 source waters or pollutant streams that reflect concentrated Se leachates or effluents; 9 affected receiving water bodies; 9 food webs that have bioaccumulated Se to toxic dietary levels; and 9 predator species (birds, fish, and livestock) whose tissue Se concentrations exceed toxic risk thresholds. Selenium effects, in order of increasing sensitivity as toxic endpoints, are: adult mortality, juvenile mortality, teratogenesis, mass wasting in adults, embryo mortality, reduced juvenile growth, and immuno-suppression (Skorupa, 1998). Regulated public health effects presently are related to consumption of fish and wild bird tissue in California, Colorado, Utah, and Texas (US Department of Health and Human Services, 1996). In general, issues of public safety also arise concerning subsistence lifestyles and consumption of wild fish and game and, in the long-term, deterioration of groundwater aquifers.
Selenium biochemistry and guidelines Selenium is an essential micronutrient in bacteria and animals (Stadtman, 1974). Beneficial effects in humans stem mainly from the role of Se as an antioxidant. However, Se is the most toxic of all biologically essential elements in mammals (Venugopal and Luckey, 1978). Toxic effects occur via biochemical pathways unable to distinguish Se from S, thus substituting Se-containing amino acids in structural and functional proteins during critical stages of development and growth (Stadtman, 1974). Hence, dramatic effects such as congenital anomalies (monstrosities) occur in embryos of aquatic birds (Presser and Ohlendorf, 1987; Skorupa, 1998). Nutritional guidelines and national guidelines for risk have been developed (Fig. 11-1). The guidelines for food webs show the narrow difference between concentrations
Seawater,
Depositional environment
(dissolved 0.075-0.19 tag L-' particulate 3.0-6.2 tag g-' North Pacific)
L~
i
4~
I
Marine deposits
Organic-carbon-enriched sediment
I
I Se source formations (western United States) (pgg-1)
Pierre and Niobrara equivalents Kreyenhagen, Moreno, Tumey
Monterey, Kreyenhagen, Moreno
Phosphoria (Meade Peak member)
(1-103, average 4)
(<1-56. average 12)
(1-1200, average 182)
1 I Irrigati~ I
Anthropogenic activities
I
I
Source waters (lagL-1) Receiving waters (#gL -1)
(ggg-1)
Waste-rock seeps/drains
(<1-7300)
(50. effluent limit)
(2-4300)
I
I
Colorado River Watershed,
San Joaquin River Water shed, Kesterson and Tulare Ponds, California (9-t t.3oo)
Invertebrates (lO-71)]1 Plants (indicator) (1-14.920) I c o o t (egg 0.8-120);
San Francisco BayDelta Estuary, California
Aquatic insects
Clams
(16-330)
(4-20)
minnow (whole-body 12-31). razorback sucker cattle; horse
shorebirds egg (2-164); d u c k (egg 0.6-37): catfish, blackfish, mosquitofish, and carp (whole-body 22-370) Coot (egg 0.6-74);
Nutrition (adequate/chronictoxicity) 0.03-0.4 / 1.9lagg-1ww Human 0.03-0.4 / 5-40lagg-1 Horse 0.03-0.4 / 5-25 lagg-~ Sheep Human health advisory Consumption of fish (flesh)
Drinking water
< 1 8 g d -1 at 2lagg -1 ww
50lag L -~
Idaho (<1-400) i
Benthic invertebrates (1-150) Aquatic macrophytes (0.4-65) Plants (forage) (<1-952)
I
I
pike-
Upper Blackfoot River Watershed,
(0.1-O.3) I
I
(flesh 12-54); b o n y t a i l ;
Guidelines
I
Refinery wastewater
I
Species of concern
I
Subsurface drainage
(9-170)
Food webs (ggg-1)
I
] Mining]
Refining
I
trout forage fish (whole-body 1-35); horse; elk, sheep (liver 6-19, wet weight) Coot (egg ?-80);
scaup (liver 8-114) scoter (liver 13-386); sturgeon (flesh 2-50); splittail; salmon; flounder; Dungeness crab
coot (egg ?-57);
Protection of aquatic life Water quality < 5 lag L -1
(flesh 0.4-33);
Risk to aquatic life; ecological thresholds None
Protection of animal life (liver) Sheep (wet weight) 15-30 mg g-1 Hazardous waste water extract 1000lag L -1 Solids (wet weight) 100 lag g-~
Marginal
Substantive
Freshwater (lag L -1) Sediment (lagg-~) Diet (lag g-~)
<2 <2 <3
2-5 2-4 3-7
>5 >4 >7
Fish (lagg-~)(whole-body) Avian eggs (lag g-1)
<4 <6
4-6 6-10
>6 > 10
(8 lag g-~ dw at 75% moisture)
Fig. 11-1. Conceptual model of Se pollution based on Se concentrations in the environment from examples in the western United States. Guidelines for protection from Se toxicity, ecological risk thresholds, and hazardous wastes are listed for comparison. Food webs focus on vulnerable species. Selenium concentration ranges (in dry weight, dw, except as noted; ww = wet weight) are given in parentheses, together with averages for the sourcerock shales. Sources of data are in the Methods and Sources of Data Section.
(% (% ,,,,,.,
The Phosphoria Formation -forecasting model for global selenium sources
305
considered safe and those considered harmful. Nutritional guidelines may not be directly comparable across classes or within species because Se toxicity is dependent on the sensitivity of the animal, the chemical form of the Se, and the dose and duration of exposure. For example, the threshold for chronic toxicity in humans (wet weight, whole diet) is based on a longer exposure time than that for horses and sheep (Puls, 1988). Despite these complexities, five western states at 11 sites have human-health advisories for fish consumption because of Se (US Department of Health and Human Services, 1996). Children (less than age 15) and pregnant women are advised not to consume fish or game from a posted area, whereas adult males and non-pregnant women should consume no more than one meal each 2-week period of approximately 120 g of fish or game flesh (i.e. muscle) (Fan et al., 1988). Ecological risk threshold ranges are indicative of the endpoints used to measure adverse biological effects (US Department of Interior, 1998). Thresholds for marginal and substantive risk are well established for water, sediment, diet, fish tissue, and bird eggs (Fig. 11-1). Because the difference between essential and toxic levels for Se is narrow, concern of marginal risk levels, which are between levels considered safe (no effect) and those considered harmful (substantive risk or the toxicity threshold), are intended to provide protection for the environment. The equality of the criterion for the protection of aquatic life and the ecological threshold at which substantive risk occurs (i.e. 5 I~g L-~ Se) demonstrates a need to establish a set of criteria that fully encompasses both aquatic and semi-aquatic food web components (US Environmental Protection Agency, 1998).
Selenium ocean chemistry
The dissolved concentration of Se currently in the North Pacific Ocean ranges from 0.075 ~g L-1 Se in surface water to 0.19 I~g L-i Se in deep water (Cutter and Bruland, 1984). The depth-profile of Se parallels those of phosphate and nitrate (Bruland, 1983), essential and limiting nutrients to primary productivity (Codispoti, 1989). Organic selenide makes up 80% of the total dissolved Se in surface water (Cutter and Bruland, 1984). The reduced-state organic selenide maximum, supposedly consisting of seleno-amino acids, coincides with the maxima of primary productivity, suggesting entrance into food-chain organisms. The downward flux of particulate Se, found primarily in the - 2 oxidation state, decreases with depth. Sediment trap material for the North Pacific (100-970 m) ranged in Se concentration from 3.0 to 6.2 I~g g-! Se (Cutter and Bruland, 1984).
Field case-studies and environmental selenium concentrations
The sources and biogeochemistry of Se combine to make contamination by this element an ecological issue of concern (Fig. 11-1). Environmental Se concentrations from ecotoxic field case-studies in the western United States provide comparison for dispersal of Se investigated in Idaho (Trelease and Beath, 1949; Presser, et al., 1994; Hamilton, 1998; Skorupa,
306
T.S. Presser et al.
1998). Illustrated examples from California and the watersheds of the Colorado River also include a range of processing activities that call attention to anthropogenic connections (disposal of oil refining effluents and subsurface drainage into wetlands and estuaries), in addition to surface processes (weathering, erosion, and runoff) that can ultimately mediate contamination. Selenium source rocks in the western United States encompass a wide range of sedimentary deposits, from marine shales mildly enriched in organic carbon to oil shales strongly enriched in organic matter, biogenic silica, phosphate, and trace elements (Fig. 11-1) (Presser, 1994; Piper and Isaccs, 1995; Piper et al., 2000). The resource extraction histories of shales demonstrate how sedimentary rocks exposed to weathering and erosion at basin margins may be buried deeply enough in the central part of a basin to produce oil. The marine shales of the Colorado River watersheds and those that provided source sediment for the alluvial fans of the San Joaquin Valley of California provide enriched, but disseminated Se sources (Fig. 11-1). Selenium transport is by both mass wasting and delivery of Se in dissolved load from surface runoff and groundwater throughflow. These watershed transport mechanisms provide soluble selenate and a secondary solid-source Se usually of large mass, but of a comparatively low-level concentration, which acts as a continuously renewed source. The Moreno and Kreyenhagen Formations of the Coast Ranges of California also provide oil for nearby refineries surrounding the San Francisco Bay-Delta Estuary. Here, a more concentrated source of Se is processed as reflected in effluent loads discharged to the estuary (Luoma and Presser, 2000). Restrictions for aquatic discharges recently have been enacted under regulatory permits necessitating a partitioning of Se into a solid waste for disposal on land as part of the refining process. A model developed in the 1940s (Trelease and Beath, 1949) and refined in the 1980s for the western United States (Presser et al., 1994; Luoma and Presser, 2000) shows that Se is mainly oxidized to soluble selenate. Consequently, in areas of semi-arid to arid climates where evaporation greatly exceeds precipitation and where drainage is impeded (i.e. in areas of net negative annual water budgets), Se accumulates as salts in soils or in aquifers. The affected lands support agriculture only through massive irrigation, which leaches salts and Se, and management of subsurface drainage flows (Fig. 11-1). Installation of subsurface drains increases the speed, volume, and control of the drainage of shallow groundwater that impedes agricultural production. Collection of drainage from irrigated soils in drainage canals enables efficient discharge into surface waters. In the San Joaquin Valley and the Colorado River basins (Fig. 11-1), Se concentrations in agricultural drains exceeded the criterion for a water-extracted hazardous Se waste (1000 Ixg L-~ Se). Selenium released to aquatic systems can result in Se being bioaccumulated to toxic levels in plants, fish, bird eggs, and livestock (Presser and Piper, 1998; Skorupa, 1998). The general term bioaccumulation can be applied to all of the biological levels of Se transfer through the food web. Linked biological and geochemical reactions affect how readily Se enters food webs, initiates food-web transfer, and cycles through particulate matter, sediments, consumer organisms, and predators (Luoma and Presser, 2000). Because Se concentrations can be magnified at each step of food-web transfer, upper trophic level species are most vulnerable to adverse effects from Se contamination.
The Phosphoria Formation -forecasting modelfor global selenium sources
307
Beginning in the 1930s, seleniferous open-range forage plants associated with the Pierre and Niobrara Formations (or their equivalents) were found to poison livestock mainly in Wyoming, Nebraska, and South Dakota (Fig. 11-1) (Trelease and Beath, 1949). Lands were withdrawn from use by livestock and leaching through subsurface drains was proposed as a means to remediate saline and seleniferous soils. The most well-known case of Se poisoning in an aquatic ecosystem was at Kesterson National Wildlife Refuge in the San Joaquin Valley, California (Fig. 11-1) (Presser and Ohlendorf, 1987; Presser, 1994). Widespread fish mortality and deformities in ducks, shorebirds, grebes, and coots occurred in wetlands fed by agricultural irrigation drainage. The deformities most frequently observed in birds were defects of eyes, feet or legs, beak, brain, and abdomen (Ohlendorf and Hothem, 1994). Further south in the San Joaquin Valley, a higher level of tetratogenicity (56.7%) occurred in shorebirds inhabiting ponds where accelerated evaporation was taking place as part of a management program (Skorupa, 1998). Levels of Se in the San Joaquin Valley wetlands, streams, and rivers that support beneficial uses for fish and birds exceeded levels for protection of aquatic life (> 5 txg L-s Se) (Fig. 11-1). Although food chains are specific to each water body, food-web biota exceeded ecological thresholds for dietary Se toxicity (> 7 ixgg -i Se) (Fig. 11-1) (Saiki and Lowe, 1987; US Department of Interior, 1998). Levels of Se in tissues of fish, mostly non-native, exceeded ecological thresholds for substantive risk (> 6 txg g-s Se), as did Se concentrations in bird eggs (> 10 txgg -I Se) (Skorupa, 1998). Agricultural drainage canals are posted with human-health advisories against consumption of fish because of Se contamination (Fan et al., 1988). Selenium contamination in the upper and lower Colorado River basins (Fig. 11-1) has contributed to the decline of native fish (i.e. pikeminnow, Ptychocheilus lucius; bonytail, Gila elegans; and razorback sucker, Xyrauchen texanus) and now represents a high hazard for effects in fish from Se exposure (Hamilton, 1998; Hamilton et al., 2002). A recovery program has been enacted in an effort to stabilize and enhance populations of endangered fish, especially razorback suckers, in the upper Colorado River. Eggs from American coot (Fulica americana) collected in wetlands affected by irrigation drainage in the Green River watershed showed an average Se concentration of 50 txg g-l Se (Skorupa, 1998). The level of teratogenicity was approximately 10% in eggs that survived to full term and the level of egg inviability (failed-to-hatch eggs) was > 85%. The San Francisco Bay-Delta Estuary (Fig. 11-1) is characterized by enhanced biogeochemical transformations to bioavailable particulate Se (Luoma and Presser, 2000). Efficient Se uptake to toxic levels occurs in clams and then diving ducks and bottomfeeding fish, even though waterborne concentrations in the estuary are more than fivefold below the freshwater criterion for the protection of aquatic life and the ecological threshold at which substantive risk occurs (Fig. 11-1) (Luoma and Presser, 2000). Selenium in food webs is sufficient to be a threat to listed or endangered species such as the Sacramento splittail (Pogonichthys macrolepidotus) and a concern to human health if those species are consumed. Resources that show declining populations such as Dungeness crab (Cancer magister) and chinook salmon (Oncorhynchus tshawytsch) also may be at risk from Se, in addition to other causative factors.
308
T.S. Presser et al.
IDAHO CASE STUDY
Phosphate production and shale exposures The Meade Peak Member of the Phosphoria Formation extends throughout southeast Idaho, and adjacent areas of Wyoming, Montana, and Utah (McKelvey et al., 1959). Outcrops occur over a vast part of that area as a result of folding, faulting, uplift, and subsequent erosion of younger deposits. The area supports phosphate mining, livestock grazing, fishing, and hunting. Over the last half of the twentieth century, mining in Idaho provided approximately 4.5% of world demand for phosphate, used mainly in fertilizer (US Department of Interior, 2000; Jasinski et al., Chapter 3; Jasinski, Chapter 22). About 49% of the total production has occurred since 1985. This tonnage represents approximately 15% of the estimated one billion tons accessible to surface mining within the Phosphoria Formation (US Department of Interior and US Department of Agriculture, 1977). Out of 19 mining sites in Idaho, four are presently active (Dry Valley Mine, Enoch Valley Mine, Rasmussen Ridge Mine, and Smoky Canyon Mine) (Causey and Moyle, 2001) and two are categorized as existing mine operations (Maybe Canyon Mine and Lanes Creek Mine). Expansion at five existing mining operations and development of five new sites is expected during the next 15 years (US Department of Interior et al., 2000). The Phosphoria Formation also is estimated to have generated about 3 • 10 ~~metric tons of oil (Claypool et al., 1978). The Meade Peak Member contains up to 1200 txg g-I Se, a value exceeding a solid Se hazardous waste (100 txg g-i Se, wet weight; or 111 txg g-i Se, dry weight at 10% moisture), if this criterion was applied to mining waste (Fig. 11-1) (McKelvey et al., 1986; US Department of Health and Human Services, 1996). Average concentrations range from 48 to 560 txg g-i Se in westernmost Wyoming (Lakeridge core, subsurface depth of 4200 m) and in southeast Idaho (Hot Springs Mine, Enoch Valley Mine; Vanadiferous Zone of Bloomington Canyon) (McKelvey et al., 1986; Piper et al., 2000). Selenium is dispersed throughout the deposit, but achieves its highest concentration in a waste-shale zone between two major phosphate-ore zones of the Meade Peak Member. The waste-shale beds are phosphate lean, but enriched in organic carbon compared to the ore zones (Claypool et al., 1978; Herring and Grauch, Chapter 12). The lower-ore zone is about 12 m thick, the waste zone 27 m thick, and the upper-ore zone 5 m thick, each of which approximately maintains its thickness over 21,500 km 2 (McKelvey et al., 1959).
Geochemical mechanism of dispersal and selenium discharges Mining removes phosphate-rich beds and exposes organic carbon-rich waste rock to subaerial weathering. Waste rock is generated at a rate of 2.5-5 times that of mined ore (US Department of Interior and US Department of Agriculture, 1977). Individual dumps contain 6-70 million tons of waste-rock that is either contoured into hills, used as crossvalley fill, or used as back-fill in mine pits. In terms of Se chemistry, when Se hosted by
The Phosphoria Formation -forecasting model for global selenium sources
309
organic matter in source rocks is exposed to the oxic atmosphere and surface and ground water, Se is oxidized from relatively insoluble selenide (Se 2-) and elemental Se ~ to soluble oxyanions, selenite (SeO 2-) and selenate (SeO4z-) (Presser, 1994; Piper et al., 2000). Organic Se (operationally defined as organic selenide) also can exist in the dissolved phase. The cross-valley fills at the Smoky Canyon mine (50 million tons) and South Maybe Canyon mine (30 million tons) are stabilized with under-drains (TRC, 1999; US Department of Interior and US Department of Agriculture, 2002). Discharges from these drains are source waters for Pole Creek and Maybe Creek. Concentrations of Se in these two drains, as well as in a dump seep at the inactive Conda Mine, were equal to or exceeded the criterion for a water-extracted hazardous waste of 1000 Ixg L-l Se (Fig. 11-1) (US Department of Health and Human Services, 1996; TRC, 1999; US Department of Interior and US Department of Agriculture, 2002). In the upper Blackfoot River watershed, where the majority of mines are located, Se concentrations in streams draining both active and inactive mines contained up to 400 txg L-! Se (Fig. 11-1) (Montgomery Watson, 2001 a), exceeding the protective guidelines of 2-5 Ixg L -! Se for freshwater (US Department of Health and Human Services, 1996; US Department of Interior, 1998). These concentrations compare in order of magnitude to those discharged from subsurface drains to reservoirs and rivers in field case studies from the western United States (Presser, 1994; Presser et al., 1994; Skorupa, 1998). Temporal stream sampling in Idaho showed that waste-rock dump seeps and surface streams exhibit annual cycles in Se concentration that peak during the spring period of maximum flow (Fig. 11-2) (Montgomery Watson, 1999; Presser et al., Chapter 16).
loo
Load S
._1 :::L v
~
,~~Concoeianrt
1
//
#
E G'/F
cO
Fall, Summer
Spring
a-Conda Mine Drain b-State Land Creek c-Sage Creek, N. Fork d-Dry Valley Creek e-Georgetown Creek f, g-Sheep Creek
El Nino i
O.Ol
1
i
lOO
I
Water flux (m 3 s -1) Fig. 11-2. Relation between surface flow and Se concentrations in streams from southeast Idaho. Solid curves connect data collected from drains and streams in the Blackfoot River watershed in May (upper case) and again in September (lower case) of 1998 (Montgomery Watson, 1999). Broken lines represent generalized trends of flow, concentration, and load data from the San Joaquin Valley in California (Luoma and Presser, 2000).
310
T.S. Presser et al.
Concentration data are also available without accompanying flow data for Idaho (Montgomery Watson, 1998, 1999, 2000, 2001a,b). For example, in 1998, the stream Se concentration maximum for all samples collected in May was 260 Ixg L-1 Se, whereas it was 32 I~g L-i Se in September. In May 2000, the Se concentration in East Mill Creek was 400 i~gL -l Se and was 19 I~gL -l Se in September 1999. The observation that increased ground- and surface-water flows in the mining area result in increased Se concentrations and, by inference, increased loads is similar to that seen in the San Joaquin Valley, California (Luoma and Presser, 2000). Only during periods of exceptional runoff in California (e.g. an E1 Nifio year) did the high flow achieve a diluting effect of the large internal reservoir of Se that influences water quality (Fig. 11-2). Although the Idaho data sets for flow and concentration were not adequate to extrapolate to Se loads on an annual basis, the similarity in mobility trends in the two regions suggests massive Se storage in Idaho that is now subject to transport, as has occurred in California.
Biological reactions a n d selenium concentrations in biota
Mass balance is important in assessing ecosystem-level Se contamination (Presser and Piper, 1998). Biological reactions dominate Se partitioning once Se enters aquatic environments, leading to Se transfer through food webs (bioaccumulation). Pathway bioaccumlation models consider food as the main route of transfer and link environmental Se concentrations to biological effects in upper trophic level animals (Luoma and Presser, 2000; Piper et al., 2000). Hence, Se toxicity depends not only on exposure via aquatic loading, but also on the processes that influence Se bioavailability in food webs and Se trophic transfer to vulnerable higher-level predators.
Plants, invertebrates, and fish
In Idaho, Se concentrations in biota showed a similar seasonal trend to that of drainwater and surface streams (Montgomery Watson, 1998, 1999, 2000, 200 l a,b). Submerged macrophytes reached a maximum of 56 I~g g-~ Se in spring and 44 I~g g-! Se in fall. Benthic macroinvertebrates showed a maximum of 150 Ixg g-I Se in spring and 63 I~g g-! Se in fall (Fig. 11-1). Both these food-webs bioaccumulate Se to an extent that they can provide a diet above 7 Ixg g-i Se, which is defined as causing substantive risk for higher trophic level species (US Department of Interior, 1998). Based on tissue, whole-body Se concentrations in forage fish (suckers, sculpins, minnows, and salmonids < 15 cm) exceeded the substantive risk threshold of 6 Ixg g-~ Se for growth and survival during both seasons (maxima: 35 ixgg -I Se in spring and 11 i~gg -! Se in fall) (Fig. 11-1) (US Department of Interior, 1998). These values also exceeded the substantive risk threshold for diet, if these fish are eaten by larger fish. Concentrations of Se in gamefish (> 15 cm) showed a maximum skin-on fillet concentration of 33 ixgg -1 Se in spring and 17 ixgg -l Se in fall (Fig. 11-1). Depending on the conversion factor used (Piper et al., 2000), whole-body
The Phosphoria Formation -forecasting model for global selenium sources
311
Se concentrations in gamefish would reach 55-77 txgg -1 Se in spring. These included Yellowstone cutthroat trout (Oncorhynchus clarki bouvieri), brook trout (Salvelinus fontinalis), and rainbow trout (Oncorhynchus mykiss). The proportion of trout above the 2 txg g-I Se wet-weight guideline for human consumption of fish flesh (Fig. 11-1) (US Department of Health and Human Services, 1996; US Department of Interior, 1998) increased from < 1% in fall to 30% in spring.
Birds and mammals Eggs of a common water bird, American coot (Fulica americana), collected at three impoundments down-gradient of mine-waste dumps (i.e. high-risk sites based on Se concentrations in ponds and wetlands) showed a trend of increasing Se concentration with increasing Se in pond waters (Fig. 11-3). Selenium-contaminated impoundments appear to present greater risks to wildlife than Se-contaminated streams and rivers (Seiler, 1995; Skorupa, 1998). Samples were collected in spring when ephemeral vernal wetlands provide habitat and breeding birds are present. Coot eggs reached 80 Ixg g-l Se, above the 10 Ixg g-l Se embryo viability threshold and the 65 txg g-l Se concentration above which 100% teratogenesis in coot embryos has been observed (Figs. 11-1 and 11-3) (Skorupa, 1998; US Department of Interior, 1998). Reproductive impairment was found at one impoundment (Fig. 11-4) in spite of the fact that egg collection was limited (i.e. a population-level
100
100 eggs
80
80 7 .
.
.
.
.
.
.
.
L
O (f) O'J =L
60
6o N
O
40
4~
20
20
v
O O
o
t
A
A
A
A B1 Location
B2
0
C
Fig. 11-3. Selenium in coot eggs and associated aquatic habitats sampled in spring 1999 in southeast Idaho. Sites A and C are in the upper Blackfoot River watershed and site B is in the Salt River watershed, all influenced by the Meade Peak Member of the Phosphoria Formation. The upper broken line gives the Se concentration above which 100% teratogenesis is observed for coots; the lower broken line gives the avian embryo viability threshold (i.e. threshold of substantive risk) (Fig. 11-1) (Skorupa, 1998; US DOI, 1998).
312
T.S. Presser et al.
Fig. 11-4. Deformed American coot (Fulica americana) embryo from a nest in the vicinity of a southeast Idaho phosphate mine tailings reservoir. The deformity exhibited here, "curly toe," is similar to that induced by Se in chickens (Detwiler, 2002) (B, deformed; C, normal). The scale bar represents 10 mm in each image. This coot egg was artificially incubated and analyzed using a fluorescence-based micro-digestion (Fan et al., 1997).
exposure assessment would necessitate the collection of more eggs). The egg tissue contained 12 ~g g-1 Se, a value just above the threshold (l 0 p~gg-1 Se) for substantive risk to embryo viability (Skorupa, 1998; US Department of Interior, 1998). In all, of the 27 coot eggs collected, nine embryos were assessable for presence or absence of overt deformities. One deformity in nine embryos is a factor of 75 above the background rate for overt deformities. This deformity is considered "mild" and, as such, is considered a suggestive, not definitive, endpoint for Se toxicity. When considered with the sets of data for tissue, water, and diet from Idaho (Fig. I l-l), it represents additional evidence of risk to resident birds and those migrating through southeast Idaho. Because acute dietary exposure led to the death of eight horses and approximately 250-300 sheep since 1996 at six sites in the mining district, hunter-killed elk are being evaluated for public-health risks and permits for grazing have been suspended for some mine-disturbed areas (Idaho Department of Environmental Quality, 2002). The elk survey shows a direct correlation between elevated concentrations of Se in liver versus the distance of the harvested elk from the nearest phosphate mine. Samples in 2001 at two mining areas (Mackowiak et al., Chapter 19) showed mean Se concentrations in forage plants (legume and grass) grown on waste-rock dumps exceed thresholds of dietary toxicity for horses (5-40 txg g-1 Se) and sheep (5-25 Ixg g-1 Se) (Fig. 11-1). Location within a dump site, as well as species of plant, were factors in determining Se concentrations in vegetation (Piper et al., 2000; Mackowiak et al., Chapter 19). For example, in 2001, alfalfa bioaccumlated Se to a greater extent (mean 150 txg g-1 Se, maximum 952 txg g-1 Se) than grasses (mean 27 Ixg g- 1 Se, maximum 160 Ixg g- 1 Se) (Mackowiak et al., Chapter 19).
The Phosphoria Formation -forecasting model for global selenium sources
313
Maximum Se concentrations in grass and mean and m a x i m u m Se concentrations in legume would qualify the plant material itself, regardless of dietary considerations, as hazardous based on the criterion for a hazardous Se solid waste (100 Ixg g-1 Se, wet weight; or 143 Ixg g - 1 Se, dry weight at 30% moisture) (Fig. 11-1) (US Department of Health and Human Services, 1996).
GLOBAL OCCURRENCE OF PHOSPHORITES AND PETROLEUM Prediction o f selenium sources Selenium emerges as an ecological issue of more widespread concern than illustrated by our case studies (Fig. 11-1) when possible candidates for Se sources are extrapolated from major basins hosting phosphate and petroleum source rocks. From the combined global distribution of phosphate deposits and petroleum-generating basins, it is possible to produce a world-wide map that shows the distribution of organic-carbon enriched sedimentary basins (Fig. 11-5) (Notholt et al., 1989; Klemme and Ulmishek, 1991). This map presents a base on which to predict environments that may be affected by Se loading. It is notable that 68% of the global petroleum reserves and more than 70% of phosphate resources were deposited at low latitudes in the Tethyan oceanic realm
Fig. 11-5. The distribution of phosphate deposits (Notholt et al., 1989) is overlain onto that of productive petroleum (oil and gas) basins (Klemme and Ulmishek, 1991) to generate a global plot of organic carbon-rich sedimentary basins. The Tethyan basins, which are emphasized, far outweigh the productivity of other defined realms encompassing much greater areas (i.e. Boreal or northern group of basins, the Pacific accreted terrains, and Gondwana or southern group of basins) (Klemme and Ulmishek, 1991). The Tethys was a progression of seaways subjected to many tectonic openings and closings that separated North America, Europe, and Asia to the north from South America, Africa, India, Australia, and Antarctica to the south.
314
T.S. Presser et al.
(Notholt et al., 1989; Klemme and Ulmishek, 1991). Yet, the Tethyan realm constitutes less than one-fifth of the global land area and continental shelves (Fig. 11-5) (Klemme and Ulmishek, 1991). The Tethyan realm was an east-west corridor for oceanic circulation nearly parallel to the equator. The Tethyan basins were characterized by a warm, moist climate that sustained abundant organic richness in broad, shallow continental shelves or in epicontinental seas during transgressive oceanic events throughout the Mesozoic and into the Cenozoic. Originally, source-rock age was hypothesized as controlling Se sources, with Cretaceous sedimentary rocks such as the Pierre and Niobrara Formations identified as important sources of Se (Trelease and Beath, 1949). However, formations selected for our case studies (Fig. 11-1) range in age from the Permian Phosphoria Formation to the Miocene Monterey Formation, with significant Cenozoic sources in the California Coast Ranges. A compilation by age of major phosphate resources and effective petroleum source rocks (normalized as a percentage of total global resources or of original petroleum reserves, respectively), show that some geologic ages are more important than others in determining productivity, but no apparent predictable periodicity can be discerned (Fig. 11-6) (Cook and Shergold, 1986; Notholt et al., 1989; Klemme and Ulmishek, 1991). Some 50% of phosphate resources were deposited in the Eocene and Miocene (Fig. 11-6) (Cook and
Fig. 11-6. Temporal distribution of major phosphate resources (Cook and Shergold, 1986; Notholt et al., 1989; Trappe, 1998) and of effective source rocks (Klemme and Ulmishek, 1991) normalized as percentages of total global resources and the original global petroleum reserves, respectively. Generalizations imposed on the data affect individual percentages for phosphate and petroleum, but not the overall conclusions shown by the time distributions (see Methods and Sources of Data Section).
The Phosphoria Formation -forecasting model for global selenium sources
315
Shergold, 1986; Notholt et al., 1989; Trappe, 1998). Ninety percent of the global discovered oil resources come from six stratigraphic intervals, with the middle Cretaceous and upper Jurassic accounting for 54% (Fig. 11-6) (Klemme and Ulmishek, 1991).
Commodities and exploration
Current anthropogenic activity, when combined with our forecasts based on distribution of organic carbon-rich rocks (Fig. 11-5), helps locate areas that may warrant investigations of Se dynamics during exploitation. The US has remained the world's largest producer of phosphate rock throughout most of the last century and into the twenty-first century (US Department of Interior and US Department of Agriculture, 1977; US Department of Interior, 2000). North Africa and the Middle East together produce a comparable amount. Major oil production is from the Middle East (6870 million barrels per year) with Latin America, Central Eurasia, Asia and the Pacific, the United States, and Europe each contributing in the range of 2500 million barrels per year (Oil and Gas Journal Energy Database, 1996). Areas of the Alaskan North Slope, North Africa, and Kazakhstan represent areas where both commodities are available or where industries possibly will expand (Bird and Magoon, 1987; US Geological Survey Energy Assessment Team, 2000). These enriched resource areas (Fig. 11-5) also now encompass most of the great deserts of the world, which may add severity to Se effects as evidenced by areas considered susceptible to Se contamination in the western United States (Presser et al., 1994; Seiler, 1995).
CONCLUSIONS Further refinements in models for Se deposition, source-rock weathering, and efficiency of bioaccumulation need to be made to complete our understanding of site-specific factors that may exacerbate or ameliorate Se retention and bioavailability. For example, the Miocene through Holocene phosphate deposits in North Carolina have a range of 4-9 p,gg-1 Se with a correspondingly low organic carbon content (Riggs et al., 1985), which possibly reflect deposition under oxic conditions rather than denitrifying conditions identified for the Phosphoria basin. This Se concentration range is far lower than the average for Idaho, but is comparable to that of Se source rocks in the California Coast Ranges, where serious and enduring Se contamination problems are occurring from disposal of both agricultural and oil refinery waste effluents (Presser, 1994; Luoma and Presser, 2000). However, the problems in California arose more from irrigation and refinery practices than from source rocks extraordinarily enriched in Se. Hence, the magnitude of commodity production as a measure of anthropogenic activity represents a relevant factor. Our analysis here reveals that a broad range of Se source rocks exists that encompasses essentially all organic carbon-rich marine sedimentary rocks. Consequently, we predict that the development of protective criteria and remediation technologies for controlling Se pollution in many geographic regions will become increasingly critical to natural resource exploitation activities, as well as to fish and wildlife conservation efforts.
316
T.S. Presser et aL
ACKNOWLEDGMENTS Field collection of avian eggs, embryo assessments, and chemical analyses were supported by the US Fish and Wildlife Service, Boise, Idaho, and the University of California at Davis, Departments of Land Air and Water Resources and of Animal Sciences. The authors are especially grateful to R. Brassfield and J. Jones for help with field logistics in Idaho and T. Fan, R. Higashi, and M. Fry for providing laboratory space and other technical support.
REFERENCES Bird, K.J. and Magoon, L.B., 1987. Petroleum geology of the northern part of the Artic National Wildlife Refuge, northeastern Alaska. US Geol. Surv., Bull., 1778, 329 pp. Bruland, K.W, 1983. Trace elements in sea-water. In: J.P. Riley and G. Skirrow (eds.), Chemical Oceanography, Academic Press, London, Vol. 8, pp. 158-220. Causey, J.D. and Moyle, ER., 2001. Digital database of mining-related features at selected historic and active phosphate mines, Bannock, Bear Lake, Bingham, and Caribou Counties, Idaho. US Geol. Surv., Open-File Report, 01-142, 46 pp. Claypool, G.E., Love, A.H., and Maughan, E.K., 1978. Organic geochemistry, incipient metamorphism, and oil generation in black shale members of Phosphoria Formation, western interior US. Am. Assoc. Pet. Geol. Bull., 62: 98-120. Codispoti, L.A., 1989. Phophorus vs. nitrogen limitation of new and export production. In: W.H. Berger et al. (eds.), Productivity of the Oceans, Past and Present. John Wiley, New York, pp. 377-394. Cook, P.J. and Shergold, J.H., 1986. Phosphate Deposits of the World, Vol. 1, Proterozoic and Cambrian Phosphorites, Cambridge University Press, Cambridge, 386 pp. Cutter, G.A. and Bruland, K.W., 1984. The marine geochemistry of selenium: a re-evaluation. Limnol. Oceanogr., 29:1179-1192. Detwiler, S.J., 2002. Toxicokinetics of selenium in the avian egg: comparisons between species differing in embryonic tolerance. PhD Thesis, Univ. California, Davis, 192 pp. Fan, A.M., Book, S.A., Neutra, R.R., and Epstein, D.M., 1988. Selenium and human health implications in Califomia's San Joaquin Valley. J. Toxicol. Environ. Health, 23: 539-559. Fan, T.W.M., Lane, A.N., and Higashi, R.M., 1997. Selenium biotransformation by a euryhaline microalga isolated from a saline evaporation pond. Environ. Sci. Technol., 31: 569-576. Hamilton, S.J., 1998. Selenium effects on endangered fish in the Colorado River Basin. In: W.T. Frankenberger Jr. and R.A. Engberg (eds.), Environmental Chemistry of Selenium. Marcel Dekker, New York, pp. 297-317. Hamilton, S.J., Holley, K.M. and Buhl, K.J., 2002. Hazard assessment of selenium to endangered razorback suckers (Xyrauchen texanus). Sci Total Environ., 291:111-121. Haygarth, P.M., 1994. Global importance and global cycling of selenium. In: W.T. Frankenberger Jr. and S. Benson (eds.), Selenium in the Environment. Marcel Dekker, New York, pp. 1-27. Idaho Department of Environmental Quality, 2002. Area Wide Human Health and Ecological Risk Assessment and related memorandum by R. Clegg. Tetra Tech EM Inc., Boise Idaho, 156 pp. Klemme, H.D. and Ulmishek, G.E, 1991. Effective petroleum source rocks of the world: stratigraphic distribution and controlling depositional factors. Am. Assoc. Pet. Geol. Bull., 75:1809-1851.
The Phosphoria Formation -forecasting model for global selenium sources
317
Lemly, A.D., 1996. Assessing the toxic threat of selenium to fish and aquatic birds. Environ. Monit. Assess., 43: 19-35. Luoma, S.N. and Presser, T.S., 2000. Forecasting selenium discharges to the San Francisco Bay-Delta Estuary: ecological effects of a proposed San Luis Drain extension. US Geol. Surv., Open-File Report, 00-416, 358 pp. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M. and Swanson, R.W., 1959. The Phosphoria, Park City, and Shedhorn Formations in the western phosphate field. US Geol. Surv., Prof. Paper, 313-A, 47 pp. McKelvey, V.E., Strobell, J.D. and Slaughter, A.L., 1986. The Vanadiferous Zone of the Phosphoria Formation in western Wyoming and southeastern Idaho. US Geol. Surv., Prof. Paper, 1465, 27 pp. Montgomery Watson, 1998. Fall 1997 Interim Surface Water Survey Report, Southeast Idaho Phosphate Resource Area, Selenium Project (Montgomery Watson, Steamboat Springs, Colorado), Chapters 1-5; Appendices A and B. Montgomery Watson, 1999. Final 1998 Regional Investigation Report, Southeast Idaho Phosphate Resource Area, Selenium Project (Montgomery Watson, Steamboat Springs, Colorado), Chapters 1-7; Appendices A-H. Montgomery Watson, 2000. 1999 Interim Investigation Data Report, Southeast Idaho Phosphate Resource Area, Selenium Project (Montgomery Watson, Steamboat Springs, Colorado), Chapters 1-5; Appendices A-L. Montgomery Watson, 2001a. Draft 1999-2000 Regional investigation data report for surface water, sediment and aquatic biota sampling activities, May-June 2000, Southeast Idaho Phosphate resource area, Selenium Project (Montgomery Watson, Steamboat Springs, Colorado), Chapters 1-5; Appendices A-E Montgomery Watson, 200lb. Draft 1999-2000 Regional investigation data report for surface water, sediment and aquatic biota sampling activities, September 1999, Southeast Idaho Phosphate resource area, Selenium Project (Montgomery Watson, Steamboat Springs, Colorado), Chapters 1-5; Appendices A-F. Notholt, A.J.G., Sheldon, R.P. and Davidson, D.E, 1989. Phosphate Deposits of the World, Vol. 2, Phosphate Rock Resources. Cambridge University, New York. 566 pp. Ohlendorf, H.M. and Hothem, R.L., 1994. Agricultural drainwater effects on wildlife in central California. In: D.J. Hoffman, B.A. Rattner, G.A. Burton Jr. and J. Cairns (eds.), Handbook of Ecotoxicology. Lewis, Boca Raton, pp. 577-595. Oil and Gas Journal Energy Database, 1996. Energy Statistics Sourcebook PennWell, Tulsa, 258 pp. Piper, D.Z., 1994. Seawater as the source of minor elements in black shales, phosphorites and other sedimentary rocks. Chem. Geol., 114:95-114. Piper, D.Z., 2001. Marine chemistry of the Permian Phosphoria Formation and Basin, southeast Idaho. Econ. Geol., 96: 599-620. Piper, D.Z. and Isaacs, C.M., 1995. Geochemistry of minor elements in the Monterey Formation, California: seawater chemistry of deposition. US Geol. Surv., Prof. Paper, 1566, 41 pp. Piper, D.Z., Skorupa, J.P., Presser, T.S., Hardy, M.A., Hamilton, S.J., Huebner, M. and Gulbrandsen, R.A., 2000. The Phosphoria Formation at the Hot Springs Mine in southeast Idaho: a source of selenium and other trace elements to surface water, ground water, vegetation, and biota. US Geol. Surv., Open-File Report, 00-050, 73 pp. Presser, T.S., 1994. The Kesterson effect. Environ. Manage., 18: 437-454. Presser, T.S. and Ohlendorf, H.M., 1987. Biogeochemical cycling of selenium in the San Joaquin Valley, California, USA. Environ. Manage., 11:805-821.
318
T.S. Presser et al.
Presser, T.S. and Piper, D.Z., 1998. Mass balance approach to selenium cycling through the San Joaquin Valley: From source to river. In: W. Frankenberger and R.A. Engberg (eds.), Environmental Chemistry of Selenium. Marcel Dekker Inc., New York, pp. 153-182. Presser, T.S., Sylvester, M.A. and Low, W.H., 1994. Bioaccumulation of selenium from natural geologic sources in western states and its potential consequences. Environ. Manage., 18: 423-436. Puls, R., 1988. Mineral Levels in Animal Health: Diagnostic Data, (2nd edn.). Sherpa International, Clearbrook, British Colombia, Canada, 356 pp. Riggs, S.R., Snyder, S.W.P., Hine, A.C., Snyder, S.W., Ellington, M.D. and Mallette, P.M., 1985. Geologic framework of phosphate resources in Onslow Bay, North Carolina continental shelf. Econ. Geol., 80:716-738. Saiki, M.K. and Lowe, T.P., 1987. Selenium in aquatic organisms from subsurface agricultural drainage water, San Joaquin Valley, California. Arch. Environ. Contam. Toxicol., 16: 657-670. Seiler, R.L., 1995. Prediction of areas where irrigation drainage may induce selenium contamination of water. J. Environ. Qual., 24: 973-979. Sheldon, R.P., 1981. Ancient marine phosphorites. Annu. Rev. Earth Planet. Sci., 9:251-284. Skorupa, J.P., 1998. Selenium poisoning of fish and wildlife in nature: lessons from twelve realworld examples. In: W.T. Frankenberger Jr. and R.A. Engberg (eds.), Environmental Chemistry of Selenium. Marcel Dekker, New York, pp. 315-354. Stadtman, T.S., 1974. Selenium biochemistry. Science, 183: 915-922. Stanley, T.R., Smith, G.J., Hoffman, D.J., Heinz, G.H. and Rosscoe, R., 1996. Effects of boron and selenium on mallard reproduction and duckling growth. Environ. Toxicol. Chem., 15:1124-1132. Trappe, J., 1998. Phanerozoic phosphorite depositional systems: A dynamic model for a sedimentary resource system. Lecture Notes in Earth Sciences, 76, Springer, New York, 316 pp. TRC Environmental Corporation, 1999. Maybe Canyon (south) Site Investigation, Caribou National Forest, Caribou County, Idaho. Englewood, Colorado, 121 pp., Appendices A-W. Trelease, S.E and Beath, O.A., 1949. Selenium: Its Geological Occurrence and Its Biological Effects in Relation to Botany, Chemistry, Agriculture, Nutrition, and Medicine. Trelease and Beath, New York, 292 pp. US Department of Health and Human Services, 1989. Toxicological profile for selenium (Agency for Toxic Substances and Disease Registry, Public Health Service, US Department of Health and Human Services, Atlanta, Georgia), 293 pp.; Appendices A-C. US Department of Health and Human Services, 1996. Toxicological profile for selenium (Agency for Toxic Substances and Disease Registry, Public Health Service, US Department of Health and Human Services, Atlanta, Georgia), 185 pp. US Department of the Interior (US Fish and Wildlife Service, Bureau of Reclamation, Geological Survey, Bureau of Indian Affairs), R.A. Engberg (ed.), 1998. Guidelines for interpretation of the biological effects of selected constituents in biota, water, and sediment (National Irrigation Water Quality Program, US Department of Interior, Bureau of Reclamation, Denver, Colorado), pp. 139-184. US Department of the Interior, US Geological Survey, 2000. Minerals Yearbook, Metals and Minerals, Vol. I, US Government Printing Office, Washington, DC, 57-1 to 7-12. US Department of the Interior and US Department of Agriculture, 1977. Final environmental impact statement: development of phosphate resources in southeastern Idaho, Vol. I, US Government Printing Office, Washington, DC, 429 pp. US Department of the Interior and US Department of Agriculture, 2002. Final supplement environmental impact statement for Smoky Canyon Mine, panels B and C (US Bureau of Land Management, Pocatello, Idaho), Chapters 1-7, Appendices A1-4C.
The Phosphoria Formation -forecasting model for global selenium sources
319
US Department of Interior (Bureau of Land Management), US Department of Agriculture (Forest Service), US Army Corps of Engineers, 2000. Final environmental impact statement for Dry Valley Mine, South Extension Project (US Bureau of Land Management, Pocatello, Idaho), Chapters 1-7, Appendices A-K. US Environmental Protection Agency, 1987. Ambient Water Quality Criteria for Selenium - 1987 (EPA-440/6-87-008; PB88-142237). Office of Water Regulation and Standards, Criteria, and Standards Division, Washington, DC, 121 pp. US Environmental Protection Agency, 1998. Report on the peer consultation workshop on selenium aquatic toxicity and bioaccumulation. US Environmental Protection Agency, Washington, DC, 59 pp.; Appendices A-E US Geological Survey World Energy Assessment Team, 2000. World petroleum assessment 2000. US Geol. Surv. Digital Data Series 60 available at http://geology.cr.usgs.gov/energy/WorldEnergy/ DDS-60/ Venugopal, B. and Luckey, T.D., 1978. Metal Toxicity in Mammals, Vol. 2. Plenum, New York. 409 pp.
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
321
Chapter 12
LITHOGEOCHEMISTRY OF THE MEADE PEAK PHOSPHATIC SHALE MEMBER OF THE PHOSPHORIA FORMATION, SOUTHEAST IDAHO
J.R. HERRING and R.I. GRAUCH
ABSTRACT This study focuses on the geochemical composition of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation in southeast Idaho. We present the composition and distribution of elements in nine measured sections that were sampled by a continuous series of channel samples through the Meade Peak. Two sections were sampled at each of four operating phosphate mines, one close to the pre-mining ground surface and one from a deeper level. This permited comparison of near-surface, weathered sections with deeper, less-weathered sections. The Meade Peak ranges from 36 to 58 m thickness in the measured sections. The member has lower and upper phosphorite ore zones of about 12.1 and 5.1 m thickness, respectively. A waste-rock unit about 24.6 m thick lies between the two ore zones. During mining, the middle waste-rock unit and other waste-rock units below and above the ore zones are removed and placed in waste-rock dumps. It is the removed rock that has led to concerns over the release of trace elements into the environment. The Meade Peak is a phosphatic black shale that is notably enriched in several trace elements compared to most other black shales and even to many other phosphatic black shales. Compared to the average world-shale composition, the Meade Peak waste-rock is exceptionally enriched in Ag, Cd, Cr, Se, U, and Zn. Concentrations of Hg and TI, which average 0.5 and 2 ppm, respectively, are not strongly enriched over concentrations in other shale. Five principal components in the Meade Peak are carbonates, detrital minerals, quartz, phosphate, and organic carbon. The proportions of these components vary as a function of depositional processes and subsequent alteration. Weathering of near-surface strata completely removes carbonate and greatly lowers organic-carbon content while slightly elevating concentrations of phosphate and detrital minerals. Where the Meade Peak is highly altered from interaction with oxidizing groundwater, the concentrations of many elements are greatly reduced. The trace elements that are most easily removed by weathering are Hg, Ni, Se, and to a lesser extent Cr, Cu, Mo, Sb, and Zn. A few elements, notably Ag, Ba, U, V, and Zr, are slightly enriched in the highly altered rocks. Uranium is enriched less than its host carbonate fluorapatite in the altered zones which may result from partial oxidation and removal of U from phosphorite by
322
JR. Herring and R.I. Grauch
bicarbonate-rich groundwater. Alteration has removed enormous quantities of many contaminant trace elements from the surficial Meade Peak. At a minimum, several hundred kilograms of Se and several other trace elements for each meter slice of rock along the strike of the Meade Peak have been released by weathering.
INTRODUCTION
Purpose and background This study summarizes lithogeochemical analysis of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation in southeast Idaho at four active phosphate mines (Fig. 12-1). The Meade Peak, which extends over parts of five States, has been mined for phosphate for nearly a century and remains an important resource. However, recent concern has arisen over several contaminant elements that exist in the waste-rock associated with mining and their potential release into the environment. In particular, the presence of high concentrations of Se is of great concern to the extraction of phosphate and reclamation of disturbed lands. Selenium occurs throughout the Meade Peak in varying concentrations and with no specific association with phosphatic units. In addition, many other trace elements with potential deleterious effects on the environment exist in these rocks. Consequently, we examined the data for these elements in whole sections through the Meade Peak and in various lithologic subunits. Over the past 5 years, awareness of geoenvironmental issues associated with contaminant trace elements focused much interest on understanding the distribution and mobility of these elements. Among elements in the Meade Peak, Se is of key interest, because of recent observations of toxic effects on livestock by Se derived from the waste-rock of the Meade Peak (see Hamilton et al., Chapter 18). However, there are many other elements with geoenvironmental significance, especially Ag, As, Ba, Cd, Co, Cr, Cu, Hg, Mn, Mo, Ni, Pb, Sb, Se, Th, TI, U, V, Zn, and Zr. This study improves on published studies concerning element concentrations in the Meade Peak in several ways. First, several complete sections through the Meade Peak are analyzed as a series of contiguous channel samples that sampled zones of uniform lithology. This allowed calculation of mean element concentrations as a function of lithology. Using thickness data, it also permitted calculation of overall weighted mean concentrations of elements in the various zones and for the entire member. Although many early studies of the Meade Peak included analyses of similar channel samples across the entire member, these studies focused on phosphorite resources and did not calculate mean element concentrations. Commonly, these early studies did not report concentrations of most of the trace elements of current geoenvironmental interest or, when trace-element concentrations were included, they were based on discrete samples, not channel samples. Finally, sampiing at the working phosphate mines allowed us to obtain complete sections at greater depths than many older studies, which often sampled in trenches at ground surface. This allowed us to compare deeper, less-altered rocks with shallower rocks.
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
323
Fig. 12-1. Location map of the four active phosphate mines in southeast Idaho where sections were measured and samples collected.
The Phosphoria Formation in the vicinity of the measured sections consists of three members, which in ascending order are the Meade Peak Phosphatic Shale, the Rex Chert, and the Cherty Shale (McKelvey et al., 1959; Montgomery and Cheney, 1967; Brittenham, 1976; Oberlindacher, 1990). The Meade Peak overlies the Grandeur Tongue of the Permian Park City Formation, a unit composed of dolomite and lesser amounts of other minerals.
324
JR. Herring and R.I. Grauch
The Cherty Shale Member in southeast Idaho is overlain by the Triassic Dinwoody Formation, a silty mudstone. The measured sections used for this report focus on the Meade Peak (Fig. 12-2).
M e a s u r e d sections
We conducted geochemical sampling at each of four active phosphate mines in southeast Idaho. Continuous channel sampling was completed through two well-described sections for the entire Meade Peak at each mine. The intention was to sample the Meade Peak where it was well exposed and to compare the composition at a locality where it was highly altered with a nearby section at the same mine where the unit was less altered. The comparison pairs (Fig. 12-1) are labeled A and B (Enoch Valley mine), C and D (Dry Valley mine), E and F (Rasmussen Ridge mine), and G and H (Smoky Canyon mine). Drill Core Section J is also from the Enoch Valley mine site. Stratigraphic sections are described by Tysdal et al. (1999, 2000a-c) and Grauch et al. (2001). Channel samples were collected from the measured sections such that they exactly correspond to the described intervals. The analytical data are archived in a series of reports by Herring et al. (1999, 2000a-c, 2001). These two sets of reports are best used together to obtain both descriptive and analytical information about the sections. Informal zone names for the sequence from the base upward consist of lower-wasterock, lower ore, middle-waste-rock, upper ore, and upper-waste-rock. Contacts of units within ore and waste-rock zones were picked by mine personnel. For more detailed sampling, unit contacts within the middle- and upper-waste-rock zones generally were picked by USGS personnel and corresponded to intervals of uniform lithology. The stratigraphic sections were measured along surfaces exposed by mining equipment. Bed and unit thicknesses in the measured sections and core are true thicknesses. No detailed descriptions were made of the strata in the sections. Stratigraphic units of the middle waste, for example, are shown mainly as mudstone (Fig. 12-2), although interbeds of other rock types also exist in the middle waste. The measured sections vary in overall thickness, thickness of lithologic units, depth below present surface, and extent of alteration (Table 12-I). Even at the same mine, the thickness of the Meade Peak can vary by up to 15%, chiefly because of tectonic thinning or thickening, especially in the middle-waste-rock (e.g., Section E). Moreover, not all of some sections, especially Section F, were well exposed and some strata of the middlewaste-rock might be tectonically repeated (see Evans, Chapter 6), which could have been overlooked in the field. Possible anomalous thickening was noted in Sections A, B, G, and H and thinning in Sections D and J (Table 12-I). As often as possible, the sections were sampled along horizontal transects with constant depths below the pre-mining ground surface. This was true for sections A, B, C, D, E, and E Strata of Sections A, B, and J dip approximately 55 ~ westward, on the west limb of a major anticline. In Section B, mudstone within the phosphorite sequence of the lower strata contains a thickened, poorly exposed zone that is interpreted to host a low-angle thrust fault.
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
325
Fig. 12-2. Graphic log of Measured Section J. Thickness measured from base of fish-scale bed. **Denotes zones sampled for intermixed dolostone and organic material (see text).
TABLE 12-I Meade Peak measured sections A-J: relative degree of alterations and thicknesses of ore and waste intervals (in feet) Section
A
B
C
D
E
F
G
Shallow. Relative Mostly depth of intermediate section alteration with below zones well ground altered; little surface. carbonate Relative alteration of present; overall organic carbon section mostly low
Intermediate depth. Mostly intermediate alteration with zones well altered; little carbonate present
Very shallow to shallow, Intermediate to well altered
Intermediate to deep. Less altered, with occasional altered zones
Intermediate depth. Mostly intermediate alteration with zones well altered; little carbonate present
Very shallow. Mostly intermediate alteration with zones well altered; little carbonate present; organic carbon low
Intermediate depth, Intermediate alteration; extensive along faults and fractures
H Intermediate depth, Intermediate alteration; extensive along faults and fractures; lower ore repeated by
J Deepest. Least altered; exhibits local alteration directly below fish-scale bed, likely along unconformity
fault Upper waste 6+
25
49.5
15 +
28
25 +
1+
29.5
23.5
Upper ore >18% P205
14
10.5
26
1.5 +
16.5
16
23.5
27.5
16.5
Middle waste
97.5-
74.5
73
108.5 +
55
88
58.5-
71
101 +
Lower ore >18% P205
43.5
50-
30
31.5
32.5
48
33.5-
50-
39+
Lower waste 4
5
12
10.5
3
3
1+
0+
8
Meade Peak Member
165
190.5
167
135
180
117.5
178
188
165
Note: Intervals marked with + have been tectonically thinned by faulting or folding or interval contacts were not discernable and likely are thicker than reported; those w i t h - are tectonically thickened and likely are thinner than reported.
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
327
The fault repeats nearly the entire lower-ore zone, although the fish-scale bed is not repeated. Section A was about 12 m and Section B about 36 m below the pre-mining surface. Sections C and D are of similar thickness. Section C has its basal part at the pre-mining ground surface and its upper strata are deeper below this surface, whereas Section D is uniformly about 50 m below the pre-mining ground surface. Sections E and F are located on the near-vertical east limb of the same anticline as Sections A and B. Section E is located about 500 m south of Section F and about 50 m below the pre-mining land surface. Lower strata of Section F were measured at about 8 m below the pre-mining surface and upper strata at about 12 m below. Rocks of Section F are intensely weathered. Those of the much deeper Section E also are extensively altered, probably because fluid intrusion was facilitated by abundant fractures. The lower third of Section G was measured along a horizontal surface and the upper two-thirds was measured along a steeply inclined pit wall flanking a cut bench about 8 m higher than the lower half of the section. Uppermost strata of the Meade Peak were not exposed at the Section G site. Section H was measured along a steeply inclined pit-wall, gradually stepping upward across three 10 m high benches from the base to the top of the section. Sections G and H have zones of extensive alteration that result from fluid interaction with the rocks along fractures and faults. Section J is a continuous vertical core that was drilled at a site that subsequently was developed into the Enoch Valley mine. The core samples the entire thickness of the Meade Peak. The Meade Peak extends from 55 to 151 m below the ground surface at this site, which is the greatest below-ground depth of Meade Peak that was sampled in this study. The Section J core was drilled before the start of mining and the rocks were not exposed to the atmosphere or surficial weathering, or fractured as a result of blasting and other mining activities as were the rocks from other described sections. Consequently, samples from Section J are the deepest and least altered of those studied here. We present unpublished element concentrations for samples from Section Z, also from the Enoch Valley mine. The section was measured to provide a framework for closely spaced channel samples collected to test compositional variations of the waste-rock over small distances and along strike. The depth below the ground surface is slightly deeper than typical mid-depth for that mine. The lithologic character of the section indicates that it is generally less altered compared to most of the other sections. Only a few intervals were sampled in this section, mostly in the middle-waste-rock, but these intervals were sampled in detail using adjacent channel sample spacing that ranged from 4 to 10 cm. This is the closest-spaced channel sampling included in our study.
METHODS
Sampling Samples for geochemical and petrological analysis were scraped or chiseled in a consistent manner along a channel across each entire interval of uniform lithology. For Section J core,
328
JR. Herring and R.I. Grauch
a sawed split of consistent shape was taken through the entire interval. In both types of sampling, the procedure provided a single sample that represented the entire interval. The choice of sampling intervals is intended to characterize strata of more or less uniform lithology and of a broad thickness that can be handled by typical mine equipment should our results indicate that separate handling of such zones would be advantageous. Within these broad intervals, we sampled thinner intervals, as thin as 20 cm or less, where rock types were different or distinct from the thick interval as a whole. Typically, about 40 samples were taken through adjoining intervals of the Meade Peak for each measured section. About 0.5-1 kg of rock was collected for each channel sample. The bulk samples were jaw crushed and then ground to < 100 mesh (0.15 mm). Splits were obtained using mechanical splitters and shipped to the analytical contractor in a randomized sequence for analysis, set aside for further USGS analysis, and archived. A set of 85 individual (non-channel) samples was taken from Section J for various special interests, such as anomalous concentrations of elements, or over short intervals that had an abrupt change in lithology. The locations of individual samples were based on visual examination or analysis using a hand-held X-ray fluorescence (XRF) unit (see Grauch et al., 2001, for explanation of this equipment). Mostly, samples were taken over stratigraphic intervals of only a few centimeters. Three suites of these individual samples were chosen to study organic-carbon-rich zones within dolostone of the middle-waste rock. Typically, these zones occur over stratigraphic intervals of about 1-2 m. Each zone base begins with the appearance of tiny stringers of thin organic-carbon seams or carbonrich areas within dolostone. Stratigraphically upward, the carbon seams thicken and become more numerous, culminating in a carbon seam that is several centimeters thick. Directly overlying the carbon seam, the dolostone again becomes visibly pure, with little carbon-rich material. XRF analysis of these zones indicated that trace-element concentrations increased as carbon seams become more abundant.
Analyses
Descriptions of analytical techniques used are given by Baedecker (1987) and Arbogast (1996). A split of each sample was analyzed for 40 major, minor, and trace elements (referred to as ICP-40) using concentrated hydrochloric, hydrofluoric, nitric, and perchloric acid digestion in conjunction with inductively coupled plasma-atomic emission spectrometry (ICP-AES) analysis (Briggs, 1996). Another split was fused in lithium metaborate and then analyzed by ICP-AES after acid dissolution. This technique (referred to as ICP-16) provided analysis of 16 elements including all major elements and a few minor and trace elements. This is the only analytical technique used that measured Si concentrations. Titanium and Cr were analyzed using both ICP techniques. However, the fusion technique is superior to acid digestion because of its ability to better digest resistant minerals that might contain those elements; so only the ICP-16 data for Ti and Cr are used here. Concentrations for Bi and Sn determined by ICP were eliminated due to a bias
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
329
in analysis. Manganese concentrations determined by ICP-40 were used rather than from ICP-16 because of lower detection limits. Strontium concentrations were used arbitrarily from the ICP-40 data set, but those data correlate well with those from ICP-16 with an r 2 of > 0.99. Selenium was determined using hydride generation followed by atomic absorption (AA) spectroscopy. The hydride-AA technique was also used to determine As, Sb, and most T1 concentrations. Mercury was determined using a cold-vapor AA. Total S and total C were measured using combustion in oxygen followed by infrared measurement of the evolved CO2 and SO2. Carbonate carbon was measured as evolved CO2 after acidification of the sample, and organic carbon was calculated as the difference between total and carbonate carbon. Ferrous Fe was measured using titration on 11 of the individual samples from Section J. The technique used a three-acid digestion, eliminating the oxidizing nitric acid from the above acid-digestion technique, and followed digestion with titration for ferrous Fe using potassium dichromate with a sodium diphenylamine sulphonate indicator. Energy-dispersive X-ray analysis (EDXRF) methodology is described by Siems (2000). Splits of about 5 g of the samples ground to < 100 mesh (0.15 mm) were pressed into pellets in mylar cups and covered with prolene film. Comparison of EDXRF analysis with other instrumental methods for Section J samples was reported by Herring et al. (2001). In general, the agreement is excellent with correlation coefficients exceeding 0.9 for 17 of the 19 elements. EDXRF measured U to a detection limit of I ppm and the quality of those data were tested on a subset of 70 samples using delayed neutron (DN) techniques, with the two data sets showing a correlation coefficient of >0.99. However, there is a systematic bias in that EDXRF data are consistently about 15% higher than DN data. Nonetheless, this comparison shows that the EDXRF technique, used on all samples for this project, provides an acceptable and consistent measure of U content.
RESULTS
Compositional averages o f elements in the Meade Peak When no more than half of the element concentrations are censored because they are below the lower detection limit (LDL), the censored values must be replaced to perform statistical analyses. Our data set is relatively robust in terms of censored values, with only 3% on average being replaced. Most elements, with the exception of Ag, Co, and Cr, had all or nearly all unqualified data (Table 12-11). The two elements with the largest number of replaced values are Co and Cr, each with about 30% of their concentrations being below the LDL. The choice of the replacement values commonly used range from 0.5 to 1 times the LDL. Here we use 0.7 of the LDL (Miesch, 1976). For so few replacement values as occur in this study, the choice of replacement value is negligible for calculation of the mean and the distribution values (Cohen, 1959).
TABLE 12-II Lithogeochemical summary of major oxides (%) and minor elements (ppm) in the Meade Peak rocks, southeast Idaho Channel samples
% above lower detection limit of 374 samples Minimum Maximum Grand mean Geometric mean Central range (-68% of values) Expected range (-95% of values) Meade Peak average (this study) Phosphoria Formation TF I Meade Peak Phosphorite 2 World Shale Average 3 North American shale composite 4 Average shale selected for comparison Enrichment (>l), depletion (<1) of grand mean vs. average shale Meade Peak Zones, wtd. means Upper waste shale Upper ore Middle waste shale Lower ore Lower waste shale Degree of alteration Highly, Individual samples average Highly, ch. sample wtd. average Highly + intermed., individual samples average Highly + intermed., ch. sample wtd. average Intermed., individual samples average Intermed., ch. sample wtd. average Least, individual samples average Least, ch. sample wtd. average
Total C
Carbonate C
Organic C
Total S
AI203
CaO
Fe203
K20
MgO
Na20
P205
SiO 2
TiO 2
100
99.5
100
99.5
100
100
100
100
100
100
100
100
100
0.22 35.0 5.7
0.0021 11.8 1.4
0. l 35.0 4.4
0.035 4.9 I, 1
0.5 14.7 6. l
0.6 53.3 22.1
0.2 6.1 2.1
0.1 3.5 1.5
0.0 18.0 1.7
0.1 2.7 0.6
0.1 39.0 13.4
4.1 83.0 35.2
0.0 0.9 0.4
2.6 0.9-7.5 0.3-20.4 4.8 . 1.5 0.67
0.6 0.2-2 0.1-6.3 1.4
0.67
0.24
4.7 2.3-9.7 1.2-19.2 5.6 16.7 1.8 14.7-15.5 16.9 15.5 0.4
18.0 7.7-42.3 3.4-96.1 23.7 2.5 43.8 3.09 3.58 3.2 7
1.7 0.8-3.4 0.4-6.6 2.0 6.2 1.1 6.0 6.4 6.2 0.3
1.2 0.6-2.4 0.3-4.7 1.4 4.2 0.5 3.19 3.94 3.5 0.4
0.7 0.2-2.5 0.06-8.9 2.3 1.2 0.4 2.49 2.85 2.6 0.6
0.5 0.3-1 0.1-1.9 0.6 0.68 0.8-1.3 1.13 1.1 0.6
8.6 2.6-28.3 0.8-88.9 13.5 0.2 29.8 0.07--0.16 0.14 0.15 89
27.3 13.6-54.7 7-106.5 32.6 58.3 12.3 15.6-58.4 64.8 60 0.6
0.3 0.28-0.6 0.15-1.3 0.3 0.5 0.15 0.86 0.80 0.83 0.4
4.0 1.5-10.4 0.6-25.9 6.6 . 2.63 .
0.4 0.1-2.3 0.01-12.5 1.8 . . 0.7 0.6 0.6 .
.
0.9 0.24
.
3.4 3.7 7.0 5.3 4.0
0.5 0.6 1.4 1.9 2.4
2.9 3. l 5.6 3.4 ! .6
1.4 0.6 1.4 0.8 0.5
8.0 4. l 7.4 3.7 7.2
6.0 32.8 16.8 33.5 12.8
2.5 1.4 2.6 1.2 3.4
1.7 0.9 1.8 1.1 2.0
1.1 0.6 1.7 2.1 4.0
0.57 0.51 0.76 0.48 0.31
3.4 23.1 9.5 20.6 5.0
61.6 27.6 39.8 19.9 47.1
0.47 0.27 0.44 0.21 0.48
2.6 2.8 3.3
0.3 0.3 0.7
2.3 2.5 2.5
0.4 0.4 0.5
6.9 6.9 6.7
21.7 21.4 22.3
2.4 2.4 2.3
1.7 1.7 1.6
0.6 0.5 1.1
0.5 0.5 0.5
15.8 15.8 15.1
38.1 37.4 37.2
0.43 0.43 0.41
3.4
0.6
2.8
0.5
6.6
22.4
2.3
1.6
0.9
0.5
15.6
36.1
0.41
5.1 5.5 7.4 7.4
2.0 1.8 2.1 1.9
3.2 3.7 5.3 5.5
0.9 0.9 1.5 ! .5
6.0 5.8 5.3 5.7
23.8 25.8 24.0 21.8
2.1 1.9 1.9 2.0
1.5 1.5 1.4 1.5
2.7 2.3 2.4 2.2
0.7 0.6 0.7 0.7
13.3 15.1 13.0 11.8
34.5 32.3 31.2 34.6
0.36 0.34 0.31 0.33
Oxide Sum
Ag
As
Ba
Cd
Co
% above lower detection limit of 374 samples minimum maximum Grand mean
85.8
97.1
100
100
96.0
83
0.7 92 lO
0.8 124 23
16 965 183
Geometric mean
82.3
8 3-17 2-36 10.2 0.5 3-10
19 9-38 5-75 25 32 40
0.07(0.07-0.7) _
13, 13-21
154 87-271 51-465 175 429 100 546 636 600 0.3
Channel samples
Central range (-68% of values) Expected range (--95% of values) Meade Peak average (this study) Phosphoria Formation TF l Meade Peak Phosphorite 2 World Shale Average 3 North American shale composite 4 Average shale selected for comparison Enrichment (>1), depletion (<1) of grand mean vs. average shale
_ _ 82
90 90 101 93 0.9
O.l 104
28.4 15 2
Cr
Cu
Hg
Mn
Mo
66.6
100
100
lO0
98.7
97.6
1.4 1250 57
1.4 48 4
83 10510
8
0.02 4.22 0.42
2.8 14460 126
1.4 437 26
30 8-113 2-400 70 1 <50
3 2-6 1-12 6 20 <10
69 20-240 6-798 248 130 30
17 7-47 3-121 35 2 3O
19, 1.0-2.6
72 37-140 20-264 93 35 100 58 _
0.33 0.18-0.63 0.094-1.2 0.41 _ _
0.2, 0.3-2.0 _
778 366-1665 175-3455 1044 83 1000 83 124.5 100 ll
0.4 _
2.6, (4-5O) _
58 2
0.4 1
850, 150-600 600 600
655 918 1470 882 525
59 79 104 100 67
0.24 0.31 0.49 0.40 0.25
133 124 120 87 718
0.3 189
25.7 10 0.4
1145
979 95
0.2
5 5
Meade Peak zones, wtd. means
226 154 211 124 236
Upper waste shale Upper ore Middle waste shale Lower ore Lower waste shale
22 77 30 105 112
Degree o f alteration
Highly, individual samples average Highly, oh. sample wtd. average Highly + intermed., individual samples average Highly + intermed., ch. sample wtd. average Intermed., individual samples average Intermed., ch. sample wtd. average Least, individual samples average Least, ch. sample wtd. average
13 13 12
23 23 23
212 215 201
1198 1396 1103
107 113 99
0.50 0.50 0.46
197 122 183
12
22
202
1299
I06
0.47
123
8 lO 9 9
21 22 23 24
171 162 162 169
823 997 976 1035
0.33 0.38 0.37 0.38
143 128 188 127
Con~nued
Table 12-II Continued Channel samples % above lower detection limit of 374 samples Minimum Maximum Grand mean Geometric mean Central range (-68% of values) Expected range (~95% of values) Meade Peak average (this study) Phosphoria Formation TF I Meade Peak Phosphorite 2 World Shale Average 3 North American shale composite 4 Average shale selected for comparison Enrichment ( > 1), depletion (< 1) of grand mean vs. average shale
Ni
Pb
Sb
Se
Sr
Th
T!
U
V
Zn
Zr
100
95.7
98.9
100
100
66.0
97.9
99.7
100
100
99.7
12 1720 220
2.8 93 13
0.42 71 6
1 ! 040 65
49 5340 584
1.4 12 4
0.07 54.8 2
1.4 274.8 59
27 8280 553
62 14530 1214
7 480 176
169 74-387 33-856 288 59 ! 00
12 7-21 4-35 15 10 < 10
4 2-9 1-19 6 7
26 6-103 2-391 71 13
451 212-959 103-1981 576 70 1000
3 2-7 1-14 6 13 -
2 0.6-5 0.2-12 3 -
37 13-103 5-279 63 -
344 130-907 51-2300 590 180
955 428-2134 198-4615 1653 60
140 71-273 38-520 167 -
71 58 60 4
25 25 0.5
1.5
0.6, 1
300 (24-287)
1.5 4
0.8 81
12 12 12 0.4
1.4
142 200 3
185 113 256 190 508
11 17 11 16 28
2 4 6 7 11
60 44 68 70 39
146 644 623 672 216
6 3 5 3 6
179 158 194
14 13 14
8 7 7
28 38 30
611 686 594
5 5 5
. 1.4 2
-
3.7, 2.5 .
.
300
300
30
130, (110--260)
95, 100-200
160, 167-210
.
.
3 20
200 3
150 8
200 0.9
2 3 1 4 12
26 108 32 100 33
208 645 296 1038 1231
763 906 1086 1552 3349
234 148 202 110 301
3 3 3
68 68 66
708 614 685
1053 952 I 188
210 210 200
Meade Peak zones, wtd. means Upper waste shale Upper ore Middle waste shale Lower ore Lower waste shale Degree of alteration Highly, individual samples average Highly, ch. sample wtd. average Highly + intermed., individual samples average Highly + intermed., ch. sample wtd. average
173
13
7
40
658
5
3
69
635
1073
195
Intermed., individual samples average Intermed., ch. sample wtd. average Least, individual samples average Least, ch. sample wtd. average
240 220 264 254
14 14 14 13
6 6 5 5
34 45 78 82
544 572 560 530
4 3 4 4
3 3 3 2
62 74 53 51
616 703 489 494
1587 1450 1400 1315
167 148 153 161
Notes: Grand Mean for Meade Peak weighted average of all channel-sampled intervals; ch. = channel; wtd. = weighted; intermed. = intermediate. IFrom Medrano and Piper (1995) for terrigenous fraction. 2Modal concentration from Gulbrandsen (1966). 3World shale average preferentially from Wedepohl (1969-1978); when no value was given, from Turekian and Wedepohl (1961); the latter values italicized. 4North American shale composite (Gromet et al., 1984).
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
333
The Meade Peak has similar major element concentrations to the world-shale composite (WSC) and North American shale composite (NASC; Gromet et al., 1984) (Table 12-11); however, there are some striking differences. The presence of phosphorite in the Meade Peak causes exceptionally high concentrations of CaO and P205. Furthermore, if the A1203 concentration is used as a surrogate for the aluminosilicate fraction, it is clear that the Meade Peak is more than a simple phosphatic dilution of WSC or NASC. WSC has an average P205 content of 0.15%, whereas the Meade Peak averages 13.4% P205 (Table 12-11). Assuming an average Meade Peak phosphorite to be 39.7% P205 (see section on Lithologic Characterization), the average Meade Peak shale has a phosphorite content of 34%. Simple dilution of average shale by this amount of phosphorite would reduce the other components to about two-thirds of their original abundance in the average shale. The A1203 concentration in WSC would be reduced from 15.5 to 10.3%. However, the overall A1203 concentration in the Meade Peak is 6.1%. Clearly, there is less aluminosilicate detritus in the Meade Peak than can be explained by phosphatic dilution of shale. This additional dilution is explained by organic carbon and carbonate enrichments in the Meade Peak compared to the WSC (Table 12-I1).
Geometric means a n d deviations
The initial approach used here was to compare concentration data based on assumed log normal distributions of various elements. Element concentration distributions often include a few highly elevated concentrations that greatly skew arithmetic means toward higher values. These log-transformed concentrations remove the weighting bias caused by those few higher values. Consequently, the geometric means of the concentrations are less than their arithmetic means (Table 12-II). For elements with censored values, the geometric means and deviations were estimated by the technique of Cohen (1959) for singly truncated distributions. Elements that show lognormal concentration distributions are especially amenable to distribution and central tendency analysis using logarithms of concentrations. These directly provide the geometric mean (GM) concentration. Estimates of the concentration distributions about the GM are provided by the geometric deviation (GD) (Table 12-II). The estimated central range of element concentrations, in which about 68% of the population is estimated to occur, is given by GM/GD to GM x GD. The expected range, estimated to contain about 95% of the population of concentrations, is given by GM/GD T M to GM X GD 1"96.The GD is a multiplier or divisor to the GM and, as such, has a compelling advantage over the arithmetic standard deviation - the central and expected ranges encompass only positive numbers, whereas arithmetic (standard) deviations often produce negative numbers for the lower part of the distribution range. Distributions of un-scaled and logarithm-scaled Se concentrations as an example of log normal distribution are shown in Fig. 12-3. Distributions for other elements are shown in the CD, Appendix 3. The summaries of element concentrations in the Meade Peak are listed in Table 12-11 and depicted in Fig. 12-4.
334
J.R. Herring and R.1. Grauch
Fig. 12-3. Histograms of Se concentration: upper, arithmetic-scale; lower, log-scale.
Lithologic characterization
Samples collected from the sections were identified in the field as mudstone, siltstone, phosphorite, dolostone, limestone, carbon seam (carbonaceous), or chert. The field descriptor was compared with the major element composition for each sample and adjusted if necessary. The resulting abundances for all channel samples are 4% carbonaceous rocks, 7% dolostone, 0.4% limestone, 47% mudstone, 6% siltstone, and 35% phosphorite. If, instead, the percentages are calculated based on the weighted average presence of major elements (Table 12-II), which has the advantage that each channel sample is not forced into only a single lithologic category, the following average mass percentages are obtained using derivation coefficients explained later in the section: 6% carbonaceous rocks (equal to organic carbon times 1.4), 11% carbonate (equal to carbonate carbon times 7.7), 26% phosphorite (equal to P205 x 1.93), and 57% combined mudstone and siltstone (by difference from other major components). From this comparison, it is apparent that
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
335
Fig. 12-4. Comparison of mean concentration of major oxides and elements (%) and trace elements (ppm) for the Meade Peak in southeast Idaho. Top shows high and low concentrations, weighted average arithmetic mean (left tick mark), and geometric mean concentrations (right tick mark). Middle shows geometric deviations with central range (left bar) and expected range (fight bar); geometric means are located at bar center. Bottom shows geometric mean concentrations (squares) compared to world shale average (circles) and North American shale composite (triangles). Data from Table 12-II.
336
JR. Herring and R.I. Grauch
carbonaceous rocks and carbonates are underestimated in the field and phosphorite is somewhat overestimated. Our chemically defined terms describe three end-member components (detrital, phosphorite, and carbonate) of Meade Peak rocks (Cressman and Swanson, 1964; Gulbrandsen, 1966; Piper, 2001). The chemical constituents we use to define the detrital component, while originally deposited as detritus from outside the basin, have undergone subsequent diagenesis and remobilization that results in a mineral composition that no longer is strictly detrital. The detrital (siliciclastic) component is defined as the oxide sum of AI, K, Na, Si, and Ti. The carbonate component is calculated assuming stoichiometric dolomite, and the phosphorite component is represented by P205. Two bands occur on a Temary plot, one is along the phosphorite-detrital join and the second more diffuse band is along the carbonate-detrital join (Fig. 12-5). Note the near absence of large amounts of carbonates with either the phosphate or detrital components. Carbonate is virtually absent in highly altered rocks. However, these altered rocks form a complete range of mixing concentrations between detrital and phosphorite components. Phosphorite shows a complete range of mixing concentrations with the detrital component that seems independent of the extent of alteration. The carbonate as a whole can be silty or phosphatic with concentrations of either minor component in excess of 30%. However, silty carbonate is more common than phosphatic carbonate. These observations are similar to those of Cressman and Swanson (1964) for dominant mineral components in the Retort Phosphatic Shale Member of the
Fig. 12-5. Temary plot of relative proportions of phosphorite, detrital, and carbonate components.
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
337
Phosphoria Formation in Montana. Rocks rich in organic carbon occur along two bands similar to those of carbonate (Fig. 12-5). Aluminum correlates with K and shows a strong correlation with Ti and a few trace elements, notably Ba and Zr. These correlations occur across the range of rock alteration. The sum of major oxides (Table 12-II) does not include contributions from oxides of carbon, sulfur, or nitrogen. For example, it excludes nitrogen contained in buddingtonite (an ammonium-bearing feldspar in solid solution with orthoclase) that occurs in the rocks of the Meade Peak (Gulbrandsen, 1974; Knudsen and Gunter, Chapter 7; Grauch et al., Chapter 8). If the various forms of C and total S are added to the Grand Mean of all Meade Peak oxide sum of 83% (Table 12-II), then the total becomes 97%, which with the average of 3% F in the Meade Peak phosphorite (Gulbrandsen, 1966) equals 100%. The Grand Mean of 13.4% P205 indicates that the Meade Peak weighted average of CFA is 33.8%. Further, the weighted average CaO in the Grand Mean of the Meade Peak is 22.1%, of which 18.6% or slightly over 80% is in the CFA lattice. Another way to calculate CaO that is not in CFA is to plot P205 vs. CaO (Fig. 12-6). The lower slope boundary of the relationship is CaO:P205 -- 1.34. This number times the Grand Mean of P205 yields 18%. Note that most values showing excess CaO above the phosphate boundary line are for samples that range from least altered to intermediate altered. Using a relationship of 1.4 to transform organic-carbon concentrations into carbonaceous matter concentrations (Isaacs, 1980; Piper, 2001), 51% of the samples contain less than 4% carbonaceous matter and 79% of the samples contain less than 10%. These relationships of carbonaceous matter are comparable to those in the Retort Phosphatic Shale Member (Cressman and Swanson, 1964). The frequency distribution of organic carbon in the Meade Peak is log normal and similar to that of many trace elements. Higher relative
Fig. 12-6. Scatter plot of CaO vs. P205 showing lower boundary with a slope of 1.34.
338
JR. Herring and R.I. Grauch
proportions of both carbonates and phosphorites show little carbonaceous matter. Beds of carbonaceous material are often easier to find in the highly altered units where carbonate is nearly absent because carbonaceous seams resemble shiny bituminous coal. For example, a 20 cm thick carbonaceous bed occurs in highly altered Section A and contains the highest organic-carbon concentration (35%) analyzed. Cressman and Swanson (1964) noted a correlation of 0.6, significant at the 1% level, between carbonaceous matter and A1203 and attributed this to the carbonaceous matter being concentrated in finer-grained sediments of the Retort. Here, there is no similar correlation between concentrations of organic carbon and A1203, but if element masses are used rather than concentrations, a similar correlation is obtained.
Trace elements Gulbrandsen (1966) considered that the following trace elements are concentrated in the phosphorites of the Phosphoria Formation because their modal concentrations are elevated by a factor of two or more over crustal abundances: Ag, Zn, V, Cr, Mo, As, Sb, Se, Sr, U, and rare earth elements (REE). We note a general agreement between our geometric means and his modal concentrations, but we show higher geometric mean concentrations of Ag and Se and a substantially higher maximum concentration of Cr. In addition, Cd and Ni (Table 12-I1) are enriched over WSC by factors of nearly 200 and 10, respectively. Gulbrandsen further noted that trace elements hosted by CFA include Sr, U, Y, La, and REE. He mentioned that the association of U with CFA is the principal one, but that some U must be associated with organic matter or other components. He also concluded that V is another element mainly associated with CFA, but like U also has correlations with other components.
Individual rock samples Individual rock samples from the J core have element concentrations commonly equal to or greater than those of the channel-sampled interval average that includes them (Herring et al., 2001). Typically, these enrichments are 2-10-fold, occasionally greater. Only infrequently are concentrations in individual rock samples less than those in the channel-sampled interval. In part this reflects the compositional heterogeneity through the interval and in part an intentional bias for which many of the individual samples were selected. The logarithmic scaling of trace-element concentrations permits tracking of fine detail at lower concentrations but also de-emphasizes some large and significant changes in concentration. These large changes are especially notable between the dolostone of the Grandeur Tongue and the basal meter of the Meade Peak, including the fish-scale bed. Concentrations for As, Se, and Zn vary over three orders of magnitude over this small interval, while those for Hg, T1, Cr, Ba, Cd, Cu, Mo, V, and U vary over two orders of magnitude. Only Mn, with
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
339
its affinity for carbonate minerals, shows a decrease in concentrations in the lowermost Meade Peak compared to the Grandeur. In addition, there is a progressive increase in concentrations of Mo, V, Ni, and Zn at the top of the middle-waste rock and continuing into the lower part of the upper-ore unit. The C seam in Section J is repeated over four sequences within the middle-waste rock. In one of the sequences, which begins at 24.7 m, the carbonate constitutes 85% of the sample, which drops over the next meter to around 2-3% and then at 25.9 m increases back to a high of 92%. Organic carbon has the opposite behavior. At the base of the sequence organic carbon is at a relatively low concentration of about 3%; then at 20 cm above the base, it increases to 9%, then increases to 15% over the next 60 cm. At 26 m it drops back to 1%. Alumium, S, Fe, P, and Si mimic the organic-carbon concentration, whereas Ca and Mg track dolostone. Nearly all trace elements, organic carbon, phosphorite, and detrital components have the same trend in this section. The variable dolostone content, 2-92%, functions as a governing diluent. With the exception of Cd, all other trace elements have a highly significant average correlation coefficient with organic matter of r 2= 0.81. The same trace elements (excluding Mn) have average correlation coefficients of r 2= 0.74 and 0.63 with the detrital and phosphorite components, respectively. The dolostone is relatively free of these trace elements and all else becomes seemingly covariant. Consequently, it is not readily possible to sort out those trace elements that are affiliated only with the organic matter in this sequence. The four organic-carbon layers in the fine-grained dolostone may reflect cycles of increasing productivity in the Phosphoria sea. Using the average sedimentation rate of Piper (2001) for the Meade Peak, these productivity events may have lasted between 0.1 and 0.2 Ma. However, it is not known whether the two components, carbonaceous material and carbonate, accumulated at that average rate. The events represent enhanced upwelling, which became more persistent through time until a major change in the oceanic system diminished the upwelling. Alternatively, the cycle could also be viewed as the upward decreasing deposition of fine-grained carbonate mud against a constant sedimentation of organic material. In Sections A through H, these organic carbon-dolostone sequences have not been found, presumably because much of the fine-grained dolostone has been leached from the sections. Thus, it is not possible to correlate these repeated events in Section J to other sections or to know if they are common and correlative across southeast Idaho. However, even in Sections A through H, remnant organic-carbon layers are recognizable. The 20 cm thick carbonaceous interval in the middle-waste rock of weathered Section A may reflect one of those events. Section J includes Fe 2+ analyses for 11 samples (Herring et al., 2001). The percentage of Fe 2+ to total Fe ranges from nearly 0 to 100%. On average, Fe 2+ is about 20% of total Fe. If all of the Fe 2+ is present as pyrite, then it would require only about 10% of the total S in the sample. XRD analysis (Knudsen and Gunter, Chapter 7) indicates an average pyrite content in these 11 samples of about 4%, about a factor of 10 greater than can be explained by the average Fe 2+ content. The variable presence of this Fe 2+ component indicates that mineralogy of the iron minerals cannot be calculated from bulk chemistry when all iron is assumed to be ferric.
340
JR. Herringand R.I. Grauch
Close-spaced channel samples, Section Z Two intervals of middle-waste-rock from Section Z are described here. Interval 59 was 1 m in length and began 27 m above the base of the fish-scale bed. It was subdivided into 10 equal subintervals and each of those subintervals was sampled as a channel sample. Interval 75 was 1.2 m in length and began from 40 m above the base of the fish-scale bed. It was subdivided into 12 equal-length subintervals with each subinterval sampled like interval 59. In addition, a single channel sample was taken through both intervals and the bulk concentrations of Se in each sample was determined by Monsanto using XRF analysis. Finally, interval 59 was sampled again as a set of four equal subintervals after tracing the interval laterally along strike approximately 5 m and a second time after another 5 m. At these two locations, the unit was visibly altered from the original dark gray to black shale into a thinner ocher mudstone. The alteration was likely the result of water infiltration along a fracture surface. In the detailed sampling of intervals 59 and 75, Se and U concentrations exhibit varied and generally weak negative correlations (Fig. 12-7). For interval 59, the average subinterval concentration of Se is 1319 ppm, which is considerably higher than the 721 ppm measured for the interval single-channel sample. Differences in concentrations for different intervals are likely a function of lateral inhomogeneity between the rocks of the two sets of channel samples. Interval 75 has an average subinterval Se concentration of 961 ppm, likewise considerably higher than the single-channel sample for the whole interval, 262 ppm. The lowermost two subintervals have Se concentrations in excess of 4500 ppm. In both lateral altered intervals, the average Se concentration decreases nearly fivefold. Uranium in subintervals of intervals 59 and 75 ranges from high concentrations of about 50 ppm to lows below the LDL of 2 ppm. The alteration of interval 59 has little consequence on U concentrations. Selenium and U show similar concentration trends in the upper parts of both intervals but not in the lower. Highly variable concentrations both stratigraphically and along strike suggest that considerable concentration variance for the section as a whole can be expected. This inhomogeneity may result from the localized presence of phases highly enriched in Se, such as those noted by Grauch et al. (Chapter 8). Further, weathering significantly lowers the concentration of Se. It is unknown whether the fine-scale bulk chemical heterogeneity that exists over intervals on the order of 10 cm might also exist at even finer scales; mineralogical heterogeneity does exist at very fine scales (millimeters to micrometers; Grauch et al., Chapter 8).
Compositional changes due to weathering, Enoch Valley channel samples Comparison of element concentrations in Sections A, B, and J characterizes alteration of Meade Peak rocks. Element concentrations in Section J, the least altered of the three sections, are distinctly different from those in highly altered Section A and altered Section B. Selenium concentrations in Sections J and B are consistently greater than those in Section A and reflect the greater extent of weathering in Section A, the closest of the three measured
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
341
120m
t,-
100-
>
--o-
Interval 59
Interval 59, first altered rock transect .--e-- Interval 59, second altered rock transect
>
80-
.0
0
Interval 75
--o-
0 OrJ
:
60-
t~ L_
40E oI = 20O-
,
,
,
10
|
,
|
I||
100
i l|,,l
!
1000
10000
Se (ppm) 120
Interval 75
m~oo
"-
"6
80>
0
::===-o
--o-
Interval 59
--o--
Interval 59, first altered rock transect Interval 59, second altered rock transect
=
6O
0t}
40
E c 20-
o~
|
0
I
20
=
40
60
U (ppm)
Fig. 12-7. Variations of Se (top) and U (bottom) concentrations with stratigraphic position for the detailed sampling in intervals 59 and 75 of Section Z (concentrations at 2 ppm are < 2 ppm).
sections to the pre-mined ground surface. Zone average concentrations of elements for these three sections (Fig. 12-8) reveal the importance of their removal through alteration and the variability of concentrations from zone to zone. Elements enriched at the base of Section J include T1, Mo, and Zn (Fig. 12-8). The weighted average of 22 ppm for T1 is the highest of any section interval. Also, carbonate, organic C, and total S are greatly enriched in Section J compared to the other two sections. Some elements show the opposite behavior and have higher concentrations in Section A. For example, A1, Fe, Cr, and Zr generally exhibit considerable enrichment in Section A, and Hg, Sb, Ti, Ba, and silica show lesser enrichments. Phosphate and elements associated with it are also enriched in Section A,
342
JR. Herring and R.I. Grauch
Chert p
I
Up. Waste
I
I
I
Up. Ore !1
r Mid. Waste
L. Ore t t
L. Waste I I
\
\
/
/
Idll
Grandeur 0 5 10 15 Carbonate C (%)
/
/
/
5 10 0 10 20 30 0 50 100 0 100 200 Org. C (%) Ag (ppm) As (ppm) Cd (ppm)
0
/
100 200 Me (ppm)
9 SectionB
Chert []
\
SectionJ 9
SectionA
Up. Waste ~ =
II
Up. Ore
tI
ib
,\'
Vlid. Waste
I~
L. Ore
L. Waste
/
l
Grandeur / 0
500 1000 Ni (ppm)
0
200 400 Se (ppm)
0
/ 100 200 U (ppm)
0
,/ 1000 2000 0 2500 5000 V (ppm) Zn (ppm)
Fig. 12-8. Comparison of weighted average concentrations of trace elements and carbon for Meade Peak zones based on Section A, B, and J channel samples.
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
343
although U, curiously, is not. This enrichment in the weathered Section A reflects the removal of carbonate and organic C and consequent relative enrichment of phosphorite and detrital components. Other trace elements also exhibit significant concentration changes with alteration. For example, the enrichment of V at the top of the middle-waste-rock zone and just into the upper-ore zone, the "Vanadiferous Zone" discussed by McKelvey et al. (1986), is clearly evident in Section J. Cadmium, Mo, Ni, and Zn also are enriched in this zone. A second, apparently thicker, vanadiferous zone straddles the top of the lower-waste zone and extends into the lower-ore zone. To better compare element concentrations in zones of different thicknesses, concentrations were recalculated as average concentrations, weighted by interval length of each zone of Sections A, B, and J. These weighted average concentrations for the lower-waste rock, lower ore, middle-waste-rock, upper ore, and upper-waste-rock zones are shown on Fig. 12-8. This approach clearly shows a lower vanadiferous zone and a suppressed signature of the upper, previously defined, Vanadiferous Zone.
WEATHERING AND OTHER ALTERATION Alteration of Meade Peak rocks from water-rock interaction principally by surficial weathering profoundly changed their composition. The major effects were partial oxidation of organic matter, partial to complete leaching of carbonate, and a consequent enrichment of the phosphorite and detrital components. Minor component changes involved partial oxidation of pyrite and minor solution of some phosphorite. With oxidation of pyrite, limonite and sulfuric acid were produced. The acid reacted with phosphorite producing gypsum and phosphorus in solution. In turn, this may have resulted in intergranular phosphatic cement. In addition, F in the CFA would have been released and likely produced the fluorite found in these rocks (Grauch et al., Chapter 8). Alteration also changes rock color from black to dark or pale brown, which is predominantly due to the presence of carbonaceous matter. Variably colored coatings form on joint surfaces and the rocks change from hard to soft. Initially, the extent of alteration of the Meade Peak seems to be a function of depth below ground surface. However, alteration varies over regional to local scales and in places is extreme throughout much of a single mine, including the deeper strata mined. Besides changing composition, solution alteration may significantly reduce the thickness of the Meade Peak principally through loss of carbonate and organic carbon. For example, at several locations on the mine wall at Section G, indurated, silty carbonate layers about 1 m thick were reduced to a layer of claystone about 20 cm thick where fractures allowed water to dissolve the carbonate. Pure carbonate can be completely lost by solution alteration, whereas silty carbonate reduces to the thickness of the consolidated silt after removal of the carbonate. Loss of the relatively less-dense organic matter greatly reduces the thickness of the residual shale in an increasing amount proportional to the original organic-carbon content. For purpose of visualization, assume that the mass of organic matter is 1.4 times the
344
JR. Herring and R.I. Grauch
mass of organic carbon (Isaacs, 1980) and that it has a density of 1.9 g c m - 3 (the measured value for the carbon seam of Section A), and that the density of typical Meade Peak shale without carbonaceous matter is 2.6 g cm -3 (Gulbrandsen and Krier, 1980). The oxidation and removal of 1 or 30 wt.% organic C from a vertical section would reduce the thickness of a column of rock containing those amounts of carbonaceous material to 98 or 50% of the original thickness, respectively. In this J core section, the Meade Peak contains 6.7% organic C, equivalent to 12% of the section thickness, and 2.6% carbonate C, equivalent to 19% of the section thickness (assuming stoichiometric dolomite). In Section J, complete loss of organic C and carbonate, assuming the carbonate converts to 20% of its original thickness as silty clay, would result in a thickness reduction of 6.8 and 10.9 m, respectively. Therefore, complete solution alteration of these two lithic components of Section J would reduce the thickness of the Meade Peak from 57.3 to 39.5 m, 69% of original thickness. This hypothetically reduced thickness is near the lower range of thickness for the Meade Peak in southeast Idaho, 43-77 m (McKelvey et al., 1959). In our measured sections we note that the least-altered Section J is thicker than all others, except Section C (Table 12-I). This certainly suggests that thickness reduction in highly altered sections is likely. The alteration of the Meade Peak is predominantly an interaction with water, even though the rocks have relatively low porosity and permeability. The porosity in the unweathered mudstone is around 4% (Gulbrandsen and Krier, 1980) and the rocks are relatively impermeable compared to, for example, the overlying weathered Dinwoody siltstone. However, it is possible for water to move along bedding-plane surfaces and along the abundant fracture and fault surfaces. Movement of water along bedding planes, fractures, or faults can be observed directly in the pit walls of the phosphate mines. Even in the least-weathered Section J, there are localized effects of substantial solution alteration. For example, at more than 150 m below the ground surface in this section, the uppermost few tens of centimeters of Grandeur dolostone directly below the fish-scale bed are altered to brown claystone from the penetration of water along the unconformity surface at the top of the Grandeur. In addition, the Meade Peak in southeast Idaho experienced a regional alteration imprint through hydrocarbon generation (oil window) (Claypool et al., 1978), and possibly hydrothermal fluids. Based on field criteria for degree of alteration, highly altered samples have carbonate C o f -< 1% and organic C of <3%. These two characteristics must be taken together, as there can be localized exceptions in organic-carbon content in the carbon-rich seams, even in the altered sections. A characteristic of highly altered sections in general would be that the weighted average P205 concentration for the entire section should exceed the Meade Peak weighted average of 13.4%. For example, the weighted average concentration in altered Section A is 17.6%. Using all these criteria, the set of 373 channel samples was divided into a subset of 106 highly altered, 36 intermediate-altered, and 231 least-altered samples. The classification of intermediate alteration is the most difficult to categorize; it was determined based on samples of intermediate composition of carbonate and organic carbon that did not fall clearly into either of the other two categories. Weighted average concentrations of carbonate carbon for the most- vs. least-altered samples are 0.3% vs. 1.9%; weighted average concentrations of organic C for the most- vs. least-altered samples
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
345
are 2.5% vs. 5.4%. These average compositional differences are significant at >99.99% confidence level.
Trace-element associations as a function o f alteration Pearson correlation coefficients (Tables 12-III-12-V) and factor analysis were used to examine element relations. Two sample sets were used: (a) all analyzed samples of the Meade Peak and (b) the same set, but separated into degrees of alteration- highly altered, intermediate-altered, and least-altered. Carbonate C shows a strong positive correlation only with Mg, indicating the dolomite phase, but also has weaker positive correlations with Ca and Mn. Organic C has strong positive correlations with total S and Se and weaker positive correlations with Cr, Cu, Hg, Mo, and Ni, suggesting at least partial association with organic phases for some proportion of these elements. The detrital component is interpreted to consist of A1, Fe, K, Si, Ti, Ba, Th, As, and Zr. Arsenic correlates strongly only with Fe and has weaker correlations with all major elements in the detrital component for even highly weathered samples. P205 has strong positive correlations with Sr and U and weaker positive correlations with Cd, Cr, Cu, Hg, and V. Silver strongly correlates with Cr, Cu, Hg, and Sb and has weaker correlations with Ba, Sr, U, and V. Cadmium has significant correlations with TI, U, and V. Copper has significant correlations with Ag, Cr, and Hg, and weaker correlations with Sb and U. The host phase of Hg is unknown but possibly is an organic phase as it is for these other biologically active elements. Molybdenum has no strong correlations, but weakly correlates with Ni, which, in turn, strongly correlates with Zn. All three elements may be associated with organic matter; their correlations with organic C are weak to moderate and range from 0.3 to 0.5 for the set of all samples. Selenium only correlates strongly with organic C and total S, and therefore, likely occurs in part in organic compounds and sulfides. Its association with sulfur also could be due to coexistance with sulfur in organic compounds. Thallium has strong correlations with Cd and V and a negative correlation with organic C. Vanadium has a strong correlation with Cd and TI and a weaker correlation with P, suggesting that it is partly associated with CFA. For the least altered set of rocks (Table 12-IV), most correlations are similar to the set of all rocks. Organic C increases its number of significant correlations, adding Cr, Cu, and Hg to the previous set. Silver has a strong correlation with V. Copper strengthens its correlations with Ni and Sb, Hg with Sb, and Mo with Ni. Many of the significant trace-element correlations noted for the least altered Meade Peak rocks do not hold for the most-altered rocks. Alteration removes many elements from the rocks, particularly the organic-affiliated trace elements. Only Cr retains a significant correlation with organic carbon. Alteration has only a minor effect on the detrital component. Arsenic must be partially removed from that phase, but retains a significant correlation with an Fe-rich phase. Carbonate C in the altered rocks has a strong correlation only with Ca, which may indicate the presence of secondary calcite. Carbonate C also retains a
TABLE 12-III Correlation coefficients for all channel samples from the Meade Peak
Org. C Total S AI Ca Fe K Mg Na P Si Ti Ag As Ba Cd Co Cr Cu Hg Mn Mo Ni Pb Sb Se Sr Th T1 U V Zn Zr
Carb. C
Org. C
Total S
A!
Ca
Fe
K
Mg
Na
P
Si
Ti
Ag
As
Ba
Cd
Co
Cr
Cu
0.01 0.15 -0.40 0.53 -0.38 -0.30 0.75 -0.04 -0.13 -0.46 -0.45 -0.25 -0.05 -0.59 -0.13 0.02 -0.33 -0.31 -0.44 0.34 -0.04 -0.09 -0.24 -0.22 0.00 0.12 -0.56 -0.34 -0.22 -0.22 -0.07 -0.52
0.83 0.20 0.02 0.20 0.25 0.00 0.13 0.18 0.05 0.09 0.34 0.26 0.13 -0.07 0.01 0.62 0.53 0.52 -0.24 0.51 0.51 0.19 0.23 0.75 0.24 0.02 -0.08 0.10 0.02 0.28 0.03
0.07 0.06 0.10 0.13 0.07 0.24 0.12 -0.03 -0.04 0.07 0.22 0.02 -0.21 0.20 0.37 0.29 0.31 -0.08 0.47 0.39 0.21 0.04 0.67 0.20 -0.06 -0.14 0.05 -0.14 0.20 -0.02
-0.71 0.92 0.96 0.02 0.46 -0.40 0.89 0.96 0.07 0.53 0.73 -0.27 0.39 0.34 0.23 0.42 0.10 0.26 0.46 0.06 0.37 0.30 -0.41 0.78 0.04 -0.43 -0.13 0.14 0.80
-0.66 -0.63 0.05 -0.35 0.66 -0.79 -0.69 0.28 -0.33 -0.56 0.33 -0.38 0.04 0.13 -0.05 -0.17 -0.13 -0.29 0.02 -0.07 -0.16 0.73 -0.69 -0.05 0.55 0.19 -0.02 -0.67
0.87 -0.03 0.47 -0.34 0.83 0.89 0.08 0.67 0.73 -0.29 0.40 0.35 0.21 0.43 0.16 0.35 0.48 0.11 0.39 0.31 -0.32 0.77 0.04 -0.38 -0.11 0.16 0.78
0.08 0.47 -0.36 0.82 0.91 0.13 0.56 0.66 -0.22 0.38 0.35 0.29 0.44 0.08 0.32 0.50 0.10 0.45 0.35 -0.36 0.71 0.06 -0.39 -0.06 0.20 0.76
0.01 -0.54 -0.06 -0.06 -0.35 0.08 -0.37 -0.29 0.12 -0.33 -0.34 -0.41 0.40 0.02 0.03 -0.32 -0.18 0.09 -0.35 -0.21 -0.33 -0.55 -0.30 -0.11 -0.19
-0.23 0.45 0.52 -0. !6 0.44 0.35 -0.31 0.39 -0.07 -0.10 0.08 0.20 0.18 0.22 0.02 0.04 0.26 -0.18 0.42 -0.08 -0.36 -0.24 0.04 0.54
-0.45 -0.37 0.63 -0.16 -0.03 0.52 -0.32 0.44 0.51 0.42 -0.42 0.08 -0.04 0.33 0.21 -0.02 0.78 -0.28 0.32 0.89 0.44 0.22 -0.25
0.88 -0.09 0.40 0.74 -0.32 0.39 0.15 0.04 0.21 0.12 0.16 0.33 0.02 0.17 0.18 -0.52 0.76 0.03 -0.46 -0.22 0.02 0.77
0.07 0.50 0.73 -0.23 0.35 0.29 0.19 0.38 0.07 0.18 0.35 0.05 0.34 0.21 -0.41 0.78 0.07 -0.39 -0.09 0.07 0.86
0.21 0.22 0.63 -0.23 0.68 0.78 0.69 -0.39 0.32 0.31 0.35 0.66 0.16 0.48 0.03 0.44 0.61 0.63 0.42 0.09
0.44 -0.10 0.41 0.27 0.21 0.33 0.31 0.57 0.57 0.11 0.52 0.38 -0.08 0.40 0.12 -0.22 0.08 0.37 0.49
-0.02 0.30 0.42 0.31 0.47 0.04 0.26 0.41 0.28 0.36 0.17 -0.15 0.70 0.31 -0.04 0.06 0.20 0.72
-0.21 0.19 0.42 0.20 -0.23 0.14 0.11 0.41 0.40 -0.14 0.26 -0.16 0.66 0.67 0.80 0.49 -0.08
-0.15 -0.17 -0.03 0.57 0.23 0.46 0.12 0.02 0.05 -0.29 0.33 0.05 -0.34 -0.19 0.33 0.37
0.89 0.84 -0.38 0.39 0.42 0.29 0.55 0.41 0.41 0.21 0.12 0.41 0.29 0.27 0.19
0.85 -0.39 0.41 0.43 0.41 0.65 0.32 0.44 0.15 0.31 0.53 0.48 0.38 0.14
Hg Mn Mo Ni Pb Sb Se Sr Th TI U V Zn Zr
-0.36 0.39 0.43 0.3 i 0.62 0.38 0.38 0.35 0.21 0.37 0.32 0.3 ! 0.33
Mn
0.10 0.20 -0.09 -0.12 -0.09 -0.32 0.07 -0.07 -0.43 -0.22 0.10 0.08
Mo
Ni
0.62 0.27 0.49 0.52 0.05 0.14 0.23 0.09 0.24 0.54 0.24
0.30 0.51 0.45 -0.08 0.32 0.25 -0.04 0.17 0.79 0.32
Pb
Sb
0.35 0.14 0.09 0.18 0.55 0.49 0.50 0.44 0.19
0.25 O. 18 0.26 0.43 0.23 0.59 0.53 0.29
Se
Sr
Th
0.05 0.17 - 0.04 -0.08 0.01 0.23 0.19
-0.38 0.01 0.62 0.18 0.06 -0.40
0.16 -0.26 -0.07 0.11 0.75
Notes: Based on log concentrations. Values o f t - > 0.65 or <- - 0 . 6 5 in bold. These values o f r differ significantly from 0 by >99.5% confidence.
T1
0.49 0.67 0.51 0.25
U
V
0.64 0.28 -0.22
0.51 0.07
Zn
0.16
TABLE 12-IV Correlation coefficients for least altered samples from the Meade Peak Carb. C Organic C Total S AI Ca Fe K Mg Na P Si Ti Ag As Ba Cd Co Cr Cu Hg Mn Mo Ni Pb Sb Se Sr Th TI U V Zn Zr
-0.21 - 0. i 7 -0.33 0.49 -0.33 -0.24 0.79 -0.23 -0.24 -0.43 -0.38 -0.20 -0.04 -0.56 -0.11 -0.13 -0.32 -0.28 -0.42 0.46 -0.14 -0.21 -0.40 -0.07 -0.13 0.08 -0.52 -0.34 -0.32 -0.19 - 0.20 -0.50
Org. C
0.78 0.39 - 0. i 0 0.41 0.42 -0.18 0.12 0.26 0.22 0.26 0.52 0.39 0.36 0.03 0.04 0.77 0.73 0.73 -0.26 0.60 0.56 0.25 0.54 0.74 0.31 0.19 0.04 0. ! 8 0.16 0.35 0.22
Total S
0.36 - 0.18 0.41 0.39 -0.12 0.27 0.14 0.22 0.23 0.21 0.42 0.33 -0.18 0.33 0.52 0.47 0.56 -0.05 0.55 0.44 0.26 0.33 0.66 0.22 0.21 -0.01 0.06 -0.08 0.23 0.26
AI
- 0.70 0.92 0.96 0.00 0.64 -0.34 0.90 0.95 0.09 0.53 0.72 -0.29 0.49 0.39 0.26 0.45 0.10 0.34 0.49 0.06 0.37 0.38 -0.40 0.77 -0.06 -0.40 -0.18 0. ! 2 0.80
Ca
-0.65 -0.62 0.05 -0.46 0.59 -0.79 -0.68 0.26 -0.23 -0.53 0.34 -0.44 0.01 0.13 -0.07 -0.11 -0.14 -0.23 -0.03 0.00 -0.13 0.76 -0.68 0.02 0.49 0.26 0.05 -0.69
Fe
0.86 -0.06 0.64 -0.26 0.84 0.88 0.10 0.66 0.73 -0.32 0.49 0.42 0.25 0.46 0.15 0.43 0.50 0.11 0.38 0.41 -0.29 0.74 -0.05 -0.34 -0.17 0.13 0.78
K
0.07 0.65 -0.30 0.84 0.91 0.15 0.55 0.66 -0.23 0.48 0.39 0.31 0.46 0.09 0.38 0.51 0.07 0.46 0.40 -0.34 0.70 -0.03 -0.38 -0.11 0.16 0.75
Mg
-0.13 -0.57 -0.08 -0.08 -0.34 -0.01 -0.43 -0.27 -0.02 -0.37 -0.37 -0.45 0.48 -0.10 -0.15 -0.50 -0.12 -0.07 -0.36 -0.24 -0.38 -0.60 -0.31 - 0.26 -0.22
Na
-0.14 0.60 0.71 0.02 0.52 0.56 -0.25 0.45 0.12 0.07 0.29 0.14 0.20 0.21 0.11 0.25 0.20 -0.18 0.60 0.02 -0.28 -0.16 0.01 0.71
P
-0.37 -0.32 0.64 0.02 0.10 0.51 -0.24 0.46 0.54 0.46 -0.45 0.19 0.16 0.43 0.28 0.15 0.81 -0.20 0.41 0.89 0.51 0.39 -0.18
Si
0.90 -0.06 0.38 0.72 -0.30 0.46 0.22 0.07 0.27 0.09 0.22 0.36 0.04 0.17 0.23 -0.52 0.75 -0.04 -0.40 -0.23 0.03 0.81
Ti
0.07 0.50 0.71 -0.25 0.46 0.34 0.20 0.40 0.06 0.25 0.37 0.04 0.33 0.28 -0.41 0.77 -0.02 -0.37 -0.13 0.05 0.85
Ag
0.36 0.26 0.62 -0.12 0.70 0.81 0.70 -0.38 0.46 0.52 0.40 0.72 0.37 0.51 0.01 0.45 0.58 0.65 0.60 0.10
As
0.49 -0.01 0.40 0.41 0.34 0.45 0.23 0.59 0.54 0.13 0.61 0.40 0.05 0.33 0.14 -0.09 0.10 0.36 0.49
Ba
0.00 0.44 0.49 0.37 0.54 0.05 0.37 0.58 0.39 0.40 0.31 -0.11 0.69 0.29 0.05 0.07 0.34 0.74
Cd
-0.15 0.17 0.41 0.18 -0.29 0.24 0.26 0.45 0.38 -0.01 0.27 -0.17 0.68 0.66 0.83 0.63 -0.08
Co
Cr
0.00 -0.02 0.14 0.43 0.22 0.38 0.19 0.09 0.02 -0.29 0.44 0.02 -0.28 -0.18 0.23 0.50
0.89 0.85 -0.33 0.49 0.61 0.29 0.64 0.54 0.45 0.24 0.12 0.38 0.29 0.40 0.23
Cu
0.85 -0.37 0.52 0.65 0.41 0.72 0.47 0.50 0.15 0.33 0.52 0.50 0.55 0.15
Hg Mn Mo Ni Pb Sb Se Sr Th TI U V Zn Zr
-0.35 0.50 0.60 0.36 0.67 0.55 0.42 0.36 0.23 0.37 0.31 0.43 0.33
Mn
Mo
Ni
Pb
0.00 0.07 -0.14 -0.13 -0.18 - 0.29 0.02 -0.17 -0.49 - 0.35 - 0.08 0.01
0.67 0.31 0.62 0.55 0.15 0.17 0.33 0.18 0.28 0.59 0.32
0.37 0.65 0.40 0.08 0.36 0.32 0.15 0.26 0.79 0.37
0.33 0.20 0.19 0.21 0.59 0.56 0.50 0.5 I 0.24
Sb
Se
Sr
Th
TI
-0.37 O. 10 0.63 0.27 0.20 -0.41
0.08 -0.20 -0.10 0.13 0.75
0.59 0.71 0.60 0.22
U
V
Zn
0.49 0.26
O. 19
0.19 0.40 0.24 0.52 0.60 0.28
0.26 0.08 O. 10 0.16 0.24 0.27
Notes: Based on log concentrations. Values of r_> 0.65 of < - 0 . 6 5 in bold. Values of r _>0.65 or < - 0.65 significantly differ from 0 with >99.5% confidence.
0.68 0.46 -0.19
0.60 0.05
0.19
TABLE 12-V Correlation coefficients for most altered samples from the Meade Peak
Org. C Total S AI Ca Fe K Mg Na P Si Ti Ag As Ba Cd Co Cr Cu Hg Mn Mo Ni Pb Sb Se Sr Th TI U V Zn Zr
Carb. C
Org. C
Total S
AI
-0.07 0.18 -0.54 0.76 -0.45 -0.45 0.28 -0.08 0.57 -0.57 -0.51 0.22 -0.22 -0.52 0.41 0.01 -0.17 -0.11 -0.18 0.08 -0.11 -0.14 0.18 -0.09 -0.36 0.42 -0.51 0.06 0.46 0.13 0.20 -0.45
0.81 0.15 0. i 0 0.07 0.17 0.20 -0.13 0.12 -0. i I 0.10 0.46 0.04 -0.02 -0.01 -0.33 0.69 0.54 0.56 -0.44 0.08 0.27 0.06 0.11 0.60 0.14 -0.08 -0.19 0.07 0.02 -0.03 -0.06
-0.21 0.40 -0.22 -0.15 0.08 -0.19 0.40 -0.45 -0.25 0.47 -0.17 -0.32 0.17 -0.45 0.58 0.44 0.42 -0.43 0.05 0.03 0.11 0.05 0.39 0.31 -0.35 -0.28 0.35 0.10 -0.06 -0.33
- 0.70 0.92 0.96 0.35 0.36 -0.65 0.81 0.97 -0.01 0.64 0.72 -0.44 0.34 0.22 0.20 0.41 0.13 0.28 0.60 0.02 0.35 0.52 -0.39 0.80 0.16 -0.63 -0.13 0.24 0.82
Ca
-0.64 -0.60 -0. i ! -0.31 0.95 -0.79 -0.69 0.49 -0.47 -0.67 0.54 -0.34 0.16 0.16 0.05 -0.27 -0.22 -0.43 0.14 -0.1 i -0.35 0.69 -0.70 -0.11 0.81 0.18 -0.09 -0.61
Fe
K
Mg
Na
P
Si
0.90 0.34 0.36 -0.61 0.74 0.91 -0.01 0.79 0.68 -0.41 0.34 0.19 0.15 0.39 0.23 0.38 0.63 0.12 0.42 0.46 -0.37 0.82 0.18 -0.61 -0.08 0.30 0.80
0.37 0.34 -0.58 0.76 0.95 0.10 0.64 0.64 -0.31 0.26 0.25 0.27 0.44 0.07 0.34 0.60 0.16 0.46 0.54 -0.37 0.75 0.21 -0.53 -0.01 0.30 0.80
-0.01 -0.30 0.20 0.33 0.15 0.27 0.25 0.02 0.27 0.15 0.15 0.12 0.20 0.23 0.44 0.28 0.12 0.17 -0.36 0.22 0.29 -0.19 0.09 0.34 0.23
-0.31 0.38 0.46 -0.33 0.34 0.22 -0.37 0.17 -0.30 -0.25 -0.06 0.23 0.09 0.12 -0.15 -0.04 0.18 -0.16 0.40 -0.12 -0.44 -0.24 0.04 0.51
-0.75 -0.64 0.51 -0.49 -0.63 0.48 -0.40 0.24 0.22 0.13 -0.34 -0.25 -0.50 0.06 -0.10 -0.31 0.76 -0.65 -0.16 0.85 0.15 -0.20 -0.57
0.82 -0.33 0.50 0.78 -0.58 0.37 -0.09 -0.05 0.01 0.23 0.17 0.39 -0.10 0.10 0.35 -0.52 0.79 0.11 -0.77 -0.33 0.02 0.68
Ti
-0.05 0.64 0.72 -0.45 0.32 0.15 0.14 0.35 0.15 0.27 0.52 0.05 0.29 0.48 -0.42 0.81 0.16 -0.63 -0.15 0.19 0.89
Ag
0.02 -0.22 0.49 -0.26 0.63 0.64 0.59 -0.35 0.16 0.11 0.31 0.47 0.16 0.42 -0.21 0.19 0.53 0.44 0.15 -0.11
As
0.48 -0.29 0.38 0.08 0.07 0.27 0.41 0.60 0.65 0.16 0.55 0.37 -0.29 0.63 0.21 -0.46 0.12 0.41 0.63
Ba
-0.49 0.30 0. I1 0.00 0.10 0.15 0.16 0.30 -0.03 0.06 0.27 -0.36 0.67 0.11 -0.58 -0.21 -0.04 0.64
Cd
-0.17 0.10 0.27 0.01 -0.09 0.02 -0.05 0.40 0.32 -0.23 0.21 -0.50 0.46 0.72 0.72 0.39 -0.34
Co
-0.40 -0.35 -0.19 0.75 0.17 0.52 -0.04 0.08 -0.17 -0.26 0.32 0.35 -0.40 -0.09 0.50 0.33
Cr
0.87 0.80 -0.46 0.22 0.09 0.27 0.32 0.46 0.31 0.04 -0.10 0.30 0.21 -0.10 0.03
Cu
0.80 -0.41 0.28 0.10 0.44 0.46 0.35 0.25 0.07 0.05 0.36 0.36 0.02 0.04
Hg
-0.33 0.31 0.27 0.23 0.48 0.43 0.30 0.21 -0.07 0.13 0.20 0.11 0.24
Mo Ni Pb Sb Se Sr Th TI U V Zn Zr
Mn
Mo
Ni
Pb
Sb
Se
Sr
Th
TI
U
V
Zn
0.32 0.41 0.01 0.04 -0.12 -0.34 0.23 0.33 -0.32 0.05 0.47 0.24
0.46 0.21 0.54 0.32 -0.27 0.26 0.14 -0.11 0.38 0.38 0.34
0.18 0.55 0.39 -0.37 0.43 0.36 -0.42 0.17 0.76 0.42
0.53 0.06 -0.12 0.09 0.43 0.31 0.59 0.33 0.09
0.28 -0.02 0.24 0.40 0.09 0.67 0.60 0.29
-0.21 0.34 -0.07 -0.33 0.05 0.09 0.38
-0.44 -0.30 0.58 -0.04 -0.17 -0.41
0.16 -0.61 -0.21 0.12 0.71
0.05 0.48 0.49 0.20
0.48 -0.06 -0.50
0.47 -0.02
0.22
Note: Based on log concentrations. Values of r _> 0.65 or <- - 0 . 6 5 in bold. These values of r differ significantly from 0 by >99.5% confidence.
352
JR. Herring and R.I. Grauch
weak association with P, which in part could be carbonate that occurs in CFA. P20 5 retains correlations with Sr and U and Cd retains its strong correlations with U and V. Cobalt, which had no strong correlation in the least-altered samples, has a strong correlation with Mn perhaps as grain coatings. Chromium, Cu, and Hg still correlate with organic C, but not as strongly as in the data for the least-altered samples. Nickel and Zn strongly correlate, but neither correlate with organic C. Factor analysis provides a linear grouping of co-associating trace elements. Piper (2001) used an orthogonal transformation solution to group rocks sampled from the Enoch Valley mine into five factors: CFA; detrital; a trace element factor (hydrogenous) with largest loadings of Cd, V, and Zn; carbonate; and organic carbon. For the same trace elements as those considered here, he reported the following associations: CFA (Cr, Cu, Sr); detrital (As, Ba, Ni, Pb,); carbonate-none; and organic carbon (S, Ag, Cr, Cu, Mo, Se). For our analysis we computed solutions with different numbers of factors using an oblique transformation solution (Table 12-VI). The difference between the two solution types is generally small, with the oblique solution tending to generate slightly lower loadings onto factors. However, the oblique solution has an advantage in that orthogonal factor scores show zero intercorrelations and assume that all factors are independent, whereas oblique scores allow intercorrelations among factors. While these coefficients are not particularly high, they nonetheless indicate intercorrelations among factors, particularly between the carbonate and organic-C factors and between the trace-element and CFA factors (Table 12-VI). Similar to the results noted by Piper (2001), a five-factor result identifies components that, based on their association with major elements, are: (factor 1) detrital, with AI, Fe, K, Si, Ti, Ba, Th, and Zr; (factor 2) a partial loading for CFA with Cd, Pb, Sb, V, and Zn; (factor 3) organic carbon with Mo, Ni, and Se; (factor 4) a smaller loading for organic carbon, a larger loading for CFA with Ag, Cr, Cu, Hg; and (factor 5) carbonate (Table 12-VII). Factor 1 is a detrital component that corresponds to the terrigenous factor of Piper, but differs from Piper's factor in having considerably reduced loadings of Na, Ni, and especially Pb, but, in turn, shows major loadings of Th and Zr. Factor 2, with large loadings of Cd, Pb, Sb, T1 V, and Zn, corresponds approximately with the hydrogenous factor of Piper (2001). It is possible to extract nearly all of the above information with only four factors. Usually, it is an advantage in factor analysis to use the fewest number of factors that TABLE 12-VI Correlations of five factors from oblique solution factor analysis For an oblique solution Factor 2 Factor 3 Factor 4 Factor 5
Factor 1
Factor 2
Factor 3
0.10 0.15 -0.02 -0.01
0.09 0.28 -0.18
0.16
0.23
Factor 4
-0.29
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
3 53
TABLE 12-VII Factor analysis of all samples using a five-factor oblique solution
Detrital, factor 1 Carbonate C Organic C Total S A1 Ca Fe K Mg Na P Si Ti Ag As Ba Cd Co Cr Cu Hg Mn Mo Ni Pb Sb Se Sr Th TI U V Zn Zr
-0.56 -0.03 -0.16 0.95 -0.83 0.90 0.89 -0.10 0.45 -0.47 0.94 0.97 -0.01 0.48 0.77 -0.28 0.33 0.21 0.11 0.32 0.09 0.12 0.30 0.00 0.31 0.12 -0.52 0.86 0.07 -0.46 -0.13 0.00 0.87
Note: Values >0.50 are in bold.
Partial CFA plus trace elements, factor 2
Organic carbon, factor 3
Partial CFA plus trace elements, factor 4
Carbonate, factor 5
-0.04 -0.12 -0.17 -0.07 0.10 -0.02 0.00 -0.11 -0.13 0.27 -0.13 -0.07 0.48 0.30 0.13 0.79 0.24 0.06 0.29 0.13 0.23 0.42 0.49 0.58 0.59 -0.10 0.02 0.00 0.82 0.45 0.81 0.82 0.11
0.15 0.86 0.99 0.06 0.04 0.14 0.11 -0.01 0.37 0.17 -0.02 -0.05 0.02 0.33 0.09 -0.25 0.37 0.34 0.25 0.30 0.07 0.59 0.55 0.29 0.01 0.75 0.25 -0.02 -0.12 0.05 -0.19 0.35 0.00
-0.17 0.28 -0.07 0.13 0.28 0.09 0.17 -0.16 -0.33 0.41 -0.06 0.13 0.68 -0.04 0.05 0.18 -0.69 0.75 0.71 0.67 -0.67 -0.03 -0.12 -0.17 0.47 0.12 0.50 -0.03 -0.13 0.32 0.23 -0.20 -0.08
0.61 -0.08 -0.24 0.15 0.09 0.11 0.24 0.80 -0.17 -0.42 -0.03 0.09 0.10 0.31 -0.28 0.00 -0.09 -0.02 0.01 -0.09 0.29 0.14 0.16 -0.43 0.37 0.00 -0.17 -0.18 -0.22 -0.44 0.04 0.06 -0.14
354
JR. Herring and R.I. Grauch
explain the most variance. With the set of all channel sample data, a four-factor model explains 73% of the variance as opposed to the five-factor model, which only increases the explained variance to 78%. The only loss of information is the elimination of the carbonate factor, which only has a loading for Mg. The four-factor model produces a betterdefined single high loading for P along with associated Cr, Cu, Hg, Sr, and U. For the least-altered samples (Table 12-VIII) of the Meade Peak, a three-factor model explains 69% of the variance. A four-factor model only improves this to 76%, but it does offer a slight improvement over the three-factor model in defining the loading of Mg and Mn onto the carbonate phase. Alternatively, the three-factor model better shows the loadings of a group of trace elements onto a factor that includes P205, whereas many of the same trace elements in the four-factor model are loaded onto a factor that has no major component. For the three-factor model, factor 1 is a siliciclastic component, comprised of A1, Fe, Ka, Si, Ti, Ba, Co, Th, and Zr. Factor 2 is the CFA factor with P, Cd, Pb, T1, U, V, Zn, Ag, and Cu. Organic C and total S are the major components loaded onto factor 3, with Ag, As, Cr, Cu, Hg, Mo, Ni, Sb, Se, and Sr. For the most altered samples (Table 12-VIII), factor 1 is the detrital factor with A1, Fe, K, Si, Ti, As, Ba, Se, Th, and Zr. Factor 2 comprises organic C and total S, Ag, Cr, Cu, Hg, and Se. Note the appearance of Se on factors 1 and 2, indicating that Se, which had primarily been with organic C factor in the least-weathered rocks, now has partial affiliation with the detrital component. Factor 3 is a trace-element component with no particular association with a major-element phase: Cd, Mo, Ni, Pb, Sb, TI, V, and Zn. In summary, principal component analysis of Meade Peak rocks is relatively unsuccessful for identifying a set of factors unique to a grouping of trace elements with its highly variable degrees of alteration. Principal component analysis works better for identifying these components when restricted to end-member compositions of the least and most-altered rocks.
I n d i v i d u a l trace e l e m e n t s
Three trace elements have been selected for special consideration. Selenium is emphasized because of its potential damaging effect on the environment and U and V have been studied in the past because of their economic potential in these rocks. All three elements have been mentioned in previous studies as being associated with organic matter and the phosphate phase (Sheldon, 1959; McKelvey and Carswell, 1967; McKelvey et al., 1986; Zielinski et al., Chapter 9). Consequently, the relationships for Se, U, and V with phosphate and organic C was examined. For Se, there is a distinct relationship with organic C, which has increasing scatter at lower relative concentrations of phosphate. There is no discernable trend with alteration, although a large scatter exists between Se and organic C for highly altered samples at lowest concentrations of phosphate. Uranium is predominantly associated with phosphate and alteration seems to have had no great effect on that association. At low concentrations of organic C, there is a relationship between V and organic C that ranges over nearly the entire range of phosphate. At higher concentrations of organic C,
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
355
Table 12-VIII Factor analysis for highly altered and least altered data sets using a three-factor oblique solution Highly altered (106 samples)
Carbonate C Organic C Total S A1 Ca Fe K Mg Na P Si Ti Ag As Ba Cd Co Cr Cu Hg Mn Mo Ni Pb Sb Se Sr Th T1 U V Zn Zr
Least altered (231 samples)
Factor 1
Factor 2
Factor 3
Factor 1
-0.76 0.19 -0.19 0.93 -0.88 0.86 0.84 0.15 0.48 -0.79 0.93 0.93 -0.32 0.56 0.82 -0.82 0.18 0.15 0.04 0.27 0.02 0.18 0.38 -0.25 0.01 0.59 -0.46 0.86 -0.19 -0.86 -0.50 -0.16 0.78
-0.22 0.82 0.72 0.22 0.17 0.15 0.26 0.00 -0.17 0.29 -0.08 0.17 0.62 -0.01 0.04 -0.04 -0.62 0.94 0.82 0.83 -0.72 0.09 -0.03 0.11 0.22 0.58 0.38 0.02 -0.34 0.24 0.04 -0.29 0.02
0.34 -0.14 -0.10 0.08 0.11 0.20 0.20 0.42 -0.15 -0.01 -0.14 0.06 0.40 0.42 -0.12 0.68 0.44 0.00 0.21 0.12 0.49 0.52 0.61 0.67 0.77 -0.06 -0.17 0.03 0.78 0.26 0.83 0.90 0.15
-0.67 -0.06 0.03 0.83 -0.92 0.77 0.74 -0.30 0.70 -0.39 0.94 0.89 -0.16 0.30 0.79 -0.18 0.58 0.00 -0.09 0.14 0.09 0.09 0.26 0.24 0.05 0.01 -0.68 0.90 0.21 -0.32 -0.11 0.08 0.94
Note: Values >0.5, < - 0 . 5 in bold.
Factor 2
Factor 3
-0.81 -0.19 -0.30 -0.13 -0.05 -0.15 -0.18 -0.88 0.00 0.60 0.02 -0.01 0.43 -0.18 0.39 0.78 -0.04 0.09 0.27 0.21 -0.52 0.00 0.13 0.75 0.13 -0.21 0.09 0.24 0.89 0.82 0.77 0.52 0.30
0.35 0.96 0.82 0.31 0.25 0.38 0.41 0.24 0.09 0.24 0.00 0.14 0.59 0.64 0.13 -0.09 0.03 0.83 0.77 0.73 0.02 0.72 0.63 -0.04 0.74 0.81 0.56 -0.08 -0.22 0.02 0.02 0.33 -0.06
356
JR. Herring and R.I. Grauch
the association shifts to V and phosphate. As with U, there seems to be no association with the degree of alteration of the samples.
Selenium
In Section J, there are enrichments of Se up to 7000 ppm in a phosphorite layer that directly underlies the Rex Chert. We postulate that this results from Se oxidation and subsequent transport from overlying parts of the steeply dipping Meade Peak. The Se was carried by water into zones where conditions of sufficient organic matter and lack of oxygen reduced the Se to elemental form (see Grauch et al., Chapter 8). Minor Se also occurs in pyrite, organic matter, pyrite-vaesite solid solution, sulvanite, and sphalerite (Grauch et al., Chapter 8; Perkins and Foster, Chapter 10).
Uranium
Uranium is enriched in the Meade Peak rocks: 16% of the samples exceed 103 ppm (upper limit of central range) and 2.5% of samples exceed 279 ppm (upper limit of expected range) based on the geometric mean and deviation of the concentrations, respectively (Table 12-11). WSC and NASC have U concentrations of about 3 ppm and a Th/U ratio of about 4, whereas the Meade Peak has 58 ppm U on average and Th/U of 0.07 based on analysis of the channel samples. Uranium, which is mainly associated with phosphate minerals, does not show the same enrichment as phosphate in highly altered rocks. For example, altered rocks of Section A have a weighted average P2Os/U of 0.27, while the weighted average for all Meade Peak samples is 0.23 and that for least-weathered Section J is 0.25. However, this relationship is not uniform. The highly altered Section F weighted average of the ratio is 0.20. Nonetheless, others (e.g. Sheldon, 1959) determined that the weathered Meade Peak has a slightly increased P2Os/U ratio relative to less-weathered rocks. This suggests that additional processes may affect and remove U in the altered rocks relative to the host CFA. One likely mechanism involves leaching of carbonate. Oxidizing waters would become enriched in bicarbonate ions from carbonate dissolution, and the strong tendency to form a bicarbonate-U complex would compete for and in part remove some U in the rocks. Most if not all U in the phosphate occurs as U 4§ substituting for Ca in the CFA lattice (Altschuler et al., 1958; Sheldon, 1959). The exposure of the U to oxidizing groundwater laden with Ca would also tend to remove U from the CFA lattice by oxidizing U to the more soluble +6 state in the presence of abundant dissolved Ca that could replace it. A scatter plot indicates that there might be two superimposed linear relationships between U and phosphate (Fig. 12-9a). One occurs at lower concentrations of phosphate (<20%) and the second extends to all values of phosphate including those approaching pure CFA(about 40%). If the relationship is plotted using log U concentrations (Fig. 12-9b), there
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
357
Fig. 12-9. Scatter plots of U vs. P205 for all samples of the Meade Peak with linear (a) and semilog (b) scales.
is greater clarity of a possible bivariate relationship. Although the best fit to these data is a power curve, it is also possible to model with two linear associations. One occurs at low concentrations of phosphate and allows for the uptake of relatively large amounts of U over small changes in phosphate concentration. The second relationship begins when phosphate concentrations exceed 5%.
358
JR. Herring and R.I. Grauch
Vanadium
Vanadium occurs in concentrations up to 8280 ppm in channel samples. Samples from the vanadiferous zones have all degrees of alteration. This suggests that V is insensitive to alteration and occurs either in phases resistant to dissolution in oxidizing groundwater or, that if some V is mobilized, it does not travel far before being reincorporated into a solid phase. Grauch et al. (Chapter 8) note that V occurs in at least two distinct minerals, sulvanite, which only occurs as inclusions in CFA, and roscoelite, which is not associated with CFA.
Other geoenvironmentally significant trace elements Silver
Silver occurs in Meade Peak rocks at concentrations up to 92 ppm in channel samples. The Grand Mean concentration of all Meade Peak channel samples is 8 ppm. In the close-spaced samples of Section Z, Ag concentrations show a factor of four variation in both intervals. The altered samples of Interval 59 show little change in concentration from the less-altered samples. Silver is associated in part with organic matter in highly altered and least-altered samples. Although its correlation coefficient with organic carbon is only about 0.5 for both alteration types, Ag has a high loading in the organic-matter factors of both alteration types. Grauch et al. (Chapter 8) note the presence of both native Ag and iodargyrite (AgI).
Arsenic
Arsenic has its only significant correlation with Fe for all channel samples. This relationship does not change when comparing most-altered with least-altered samples. The average As concentration for alteration end-member samples is the same. This indicates that As is not removed by alteration. Arsenic is associated with Fe in the least-altered rocks, in part with pyrite as confirmed by Grauch et al. (Chapter 8). For As to retain its significant correlation with Fe and not change concentration in the highly altered rocks suggests that it transfers from the reduced to oxidized Fe minerals.
Cadmium
Strict rules on allowable Cd concentration in phosphoric acid have recently been adopted in California and Washington as reported at the May 2002 annual meeting of the Intemational Fertilizer Association. The rules are causing difficulty in supply compliance for some westem US phosphate suppliers. Cadmium concentrations are up to 1000 ppm and the ore zones have higher weighted average Cd concentrations than the waste-rock zones, with the lower ore usually having higher concentrations than the upper ore. High degrees of alteration tend to smear
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
359
out the differences between the ore and middle-waste rock zones, perhaps through partial dissolution and transport of Cd. This observation is of use in planning the processing of the phosphate ore using the wet-acid process. In this process, Cd in the phosphate rock goes into the phosphoric acid product. Stockpiling some of the ore from the altered zones would allow blending with deeper ore to obtain a lower average concentration of Cd.
Thallium
Both the weighted average and geometric mean concentrations for T1 are 2 ppm. However, some high concentrations, from 40 to 50 ppm, occur in the lower-waste-rock of Section J, with the overlying lower-phosphate zone having some concentrations between 10 and 20 ppm. Weighted average concentrations for these two zones are 14 and 4 ppm, respectively, clearly higher than the Meade Peak mean concentration. Thallium concentrations decrease with alteration. Because of its acute toxicity (Christie, 1961), TI occurrence should continue to be studied.
ALTERATION MODEL It is possible to quantify the change in trace-element concentrations given the overall change from least to most altered rocks. While carbonate is more sensitive to alteration than organic matter, combining the two improves the sensitivity of carbonate as an indicator of alteration. For example, there is a cluster of approximately 200 channel samples that are ~<0.5% carbonate C but that have a considerable range of organic C, from near 0 to 15%. The same relationship exists between carbonate and total S. Differences in average concentrations for the elements in most- vs. least-altered rock samples are summarized in Table 12-IX. The means listed as highly significant have < 1% chance that their highly altered vs. their least-altered mean concentrations are identical. Those elements listed as having a significant difference have between 1 and 5% chance that their mean concentrations are identical. Elements with mean concentration differences that have a > 5% chance of being identical are listed as no change. Weighted arithmetic mean concentrations are derived by multiplying each concentration by its channel-sample length then dividing the sum of those products by total length. Weighted geometric mean concentrations are derived by multiplying each log of the concentration by its channel-sample length and then dividing the sum of those products by total length and taking the antilog of the result. Trace elements that show a highly significant concentration decrease with increasing alteration are Co, Mo, Ni, Se, and Zn. The detrital major components show significant to highly significant increases in concentration with increasing alteration as do Ba, Th, Zr, Ag, Cd, Cr, Cu, Hg, Sb, T, U, and V. Of those elements, Sb, Th, and T1 typically have low concentrations and the significance in mean difference may be an artifact of these small numbers. Those elements showing no significant change between their highly altered and less-altered mean concentrations are As, Mn, Pb, and Sr.
360
J R . H e r r i n g and R.I. Grauch
TABLE 12-IX Changes of element concentrations with increasing alteration B a s e d on arithmetic means
Decrease with alteration Significant (-< 1% chance of identical means) Increase with alteration Significant (<- 1% chance of identical means) Less significant (1-5% chance of identical means) No change (--- 5% chance of identical means)
Total C, Carb. C, Org. C, Total S, Mg, Na, Mo, Ni,Se, Zn A1, Fe, K, Ti, Si, Ag, Ba, Hg, Sb, Zr P, Sr, Th, TI, U, V Ca, As, Cd, Co, Cr, Cu, Mn, Pb
B a s e d on geometric means
Decrease with alteration Significant (-< 1% chance of identical means) Less significant (1-5% chance of identical means) Increase with alteration Significant (<_ 1% chance of identical means) Less significant (1-5% chance of identical means) No change (-> 5% chance of identical means)
Total C, Carb. C, Org. C, Total S, Mg, Na, Co, Mo, Ni, Se Ca, Zn AI, Fe, K, Ti, Si, Ag, Ba, Cd, Cr, Cu, Hg, Sb, Th, TI, U, V, Zr P As, Mn, Pb, Sr
Those elements that show the most significant concentration decreases with increased alteration can be examined in combination to see if they define an alteration index. For example, carbonate C, organic C, total S, and Se can be considered against the ratio of K20/Na20. In shale, this ratio increases with increasing alteration (Garrels and Mackenzie, 1971); for example, albite progressively alters to clay minerals. Unfortunately, none of these quantities provide a smoothly decreasing index of alteration, although the highest values for K20/Na20 do correspond to values of these trace elements in the most-altered samples as defined by our criteria. We developed a model of trace-element mass removed as the result of alteration (Table 12-X). It is intended to provide a general indication of the masses of trace elements removed from a section of the Meade Peak during extensive alteration. The most significant assumptions are: (a) the rocks are uniformly and highly altered throughout the alteration zone; (b) this zone extends to 30 m below the ground surface; and (c) the starting composition for the Meade Peak prior to extensive alteration is that of the least-altered Meade Peak, Section J. The depth of alteration can be highly variable and the choice of the 30 m depth of uniform and extensive alteration is a compromise value to provide a general indication of the masses of trace elements that are removed. We take the original thickness of the Meade Peak at 53 m (Table 12-I), which is in good agreement with Gulbrandsen and Krier's (1980) estimate of 54 m.
T A B L E 12-X Section J geometric mean and weighted geometric mean concentrations ( C - T i O %: A g - Z r ppml: M e a d e Peak alteration loss or gain
Section ~I data Sect. J Geo. Mean Sect. J Wtd. Geo. Mean Lwr. Ore Geo. Mean Lwr. Ore Wtd. Geo. Mean Mid. Waste Geo. Mean Mid. Waste Wtd. Geo. Mean
C
Carb. C
Org. C
S
AIOx
CaOx
FeOx
KOx
MgOx
NaOx
POx
SiOx
TiOx
Ag
As
Ba
Cd
Co
Cr
7.93
0.94
5.29
i.69
4.0
20.1
1.46
i.06
1.15
0.46
8.1
22.2
0.17
7
22
131
25
4
8.43
1.12
5.52
1.82
4.5
18.3
1.62
1,20
1.34
0.50
6.7
24.1
0.19
6
23
134
21
4
693 692
8.75
1.99
5.15
1.58
2.6
28.9
1.06
0.90
1.78
0.51
13.3
14.2
0.10
10
23
87
55
3
483
8.34
i.78
5.07
1.54
2.2
31.0
0.93
0,76
1.18
0.50
14.5
12.1
0.09
10
22
80
60
3
418
10.97
1.48
7.05
2.20
5.6
19.9
1.90
i.43
1.75
0.52
6.2
27.9
0.23
7
26
148
12
5
1007
10.69
1.47
6.96
2.22
5.8
18.5
!.94
1.49
1.81
0.55
5.7
28.8
0.24
7
27
151
12
5
966
Weighted log mean concentration All M e a d e Peak highly altered All M e a d e Peak least altered
2.04
0. ! 2
1.77
0.32
6.0
16.2
2.07
1.46
0.33
0.39
11.9
32.6
0.37
11.5
18.5
200.1
48.9
2.2
1181
6.09
0.63
4.05
1.07
4.6
17.0
1.64
1.21
0.85
0.58
7.3
27.8
0.26
6.8
20.3
148.1
22.9
3.2
811
Gains with alteration ~;
.
-
-
-
35
113
Alteration o f J. reduced by ratio o f all highly to least altered
2.8
0.2
2.4
A m o u n t lost. concentration (J - M e a d e Peak alteration ratio)
5.6
0.9
3.1
.
.
30
-
26
20
0.5
17.4
-
-
0.5
.
.
.
.
.
21.2
-
-
2.9
-
1.3
0.9
-
-
0.8
.
.
.
.
.
2.1
-
-
1.4
-
6
.
64
17
40
68
-
46
Meade Peak loss per m strike, 30 m alteration -_one: major elements tons, trace elements kg Vertical Dip. vol. = 1 6 5 0 m 3 45 ~ dip. vol. = 2333 m 3
241 341
39 56
133 188
55 78
38 54
35
.
.
.
.
.
9
-
-
50
.
.
.
.
.
12
-
-
8
15 ~ dip, vol. = 6 3 7 5 m -~
930
152
515
212
147
135
.
.
.
.
.
34
-
-
23
3.9E -,- 08
3.6E + 08
.
.
.
.
.
-
-
A p p r o x i m a t e M e a d e Peak loss from 1200km-' area east o f Soda
2.5E -*- 09
4,0E + 08
1.4E + 09 5.6E + 08
-
9.1E + 04
6.1E + 04
Springs. Idaho (see text), tons
Continued
T A B L E 12-X Continued Cu
Hg
Mn
Mo
Ni
Pb
Sb
Se
Sr
Th
TI
U
V
Zn
Zr
Section J data Sect. J G e o . M e a n
67
0.29
95
31
257
14
4
60
480
2.5
1.5
33
299
1416
115
Sect. J Wtd. G e o . M e a n
66
0.30
lO0
29
259
13
4
62
452
2.8
1.3
27
265
1273
125
Lwr. O r e G e o . M e a n
71
0.25
73
45
203
18
6
89
632
1.8
3.9
56
614
1589
86
Lwr. O r e Wtd. G e o . M e a n
66
0.24
60
43
181
19
5
79
717
1.7
3.7
63
634
1476
8l
Mid. W a s t e G e o . M e a n
79
0.37
IIl
24
314
9
4
67
520
2.8
0.6
18
176
I119
122
Mid. W a s t e Wtd. G e o . M e a n
76
0.36
118
24
304
9
4
66
487
3.2
0.6
17
173
1067
129
Weighted log mean concentration All M e a d e P e a k s h i g h l y altered
I 01.2
0.46
48.4
14.8
130.3
I 1.6
6.1
16.9
536.3
4.6
2.1
52.4
447.6
793.9
186.0
All M e a d e P e a k s least a l t e r e d
7 ! .2
0.32
72.2
20.9
19 ! .9
11.1
3.9
44.9
427.4
3.2
1.4
31.4
294.7
962.8
131.8
A m o u n t g a i n e d with alteration, %
42
42
-
-
-
4
55
-
25
52
-
41
-
-
67.4
20.4
175.9
-
-
23.5
.
.
.
.
.
1050
-
-
-
33.1
8.4
83.0
-
-
38.9
.
.
.
.
.
44
49
67
A l t e r a t i o n o f J, r e d u c e d b y ratio o f all h i g h l y to least altered A m o u n t lost, c o n c e n t r a t i o n (J M e a d e Peak alteration ratiol
223.3
-
-
Meade Peak Loss per m strike, 30 m alteration -one: major elements tons, trace elements kg Vertical Dip, vol. = 1 6 5 0 m 3
-
-
142
36
356
-
-
167
.
.
.
.
.
958
45 ~ dip, vol. = 2333 m 3
-
-
201
51
504
-
-
236
.
.
.
.
.
1354
-
15 ~ dip, vol. = 6375 m 3
-
-
549
140
1376
-
-
645
.
.
.
.
.
3701
-
-
-
-
-
A p p r o x i m a t e M e a d e Peak loss f r o m 1200 k m 2 area east o f S o d a Springs, I d a h o (see text), tons
1.5E -,- 06
3.7E -,- 05
3.7E - 06
1.7E -,- 06
.
.
.
.
.
.
9.8E + 06
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
363
Topographic slopes other than horizontal are ignored. Three scenarios for alteration of the Meade Peak correspond to dips of strata of vertical, 45, and 15~ (Table 12-X). Volumes calculated for dips other than these can be computed by dividing the volume for the vertically dipping section by the sine of the dip in degrees. A density of 2.6 g cm-3 and porosity of 4% (Gulbrandsen and Krier, 1980) are used for Section J. Trace-element loss through alteration in the Meade Peak is calculated by multiplying the starting composition of Section J by the geometric mean concentrations of all Meade Peak highly altered samples divided by the geometric mean of the least-altered samples. The mass losses in Table 12-X are given per meter slice of Meade Peak along strike for the three dip scenarios. The purpose of this model is to demonstrate that, as a minimum, several hundred kilograms of trace elements like Se per meter slice of strike of the Meade Peak can be released into the environment by alteration of surficial rocks.
CONCLUSIONS This study improves on many historical studies of Meade Peak element concentrations. Compared to previous investigations, this study provides a more complete picture of contaminant-element concentrations and associations in the Meade Peak in the region of phosphate mining in southeast Idaho. In addition, it shows that the important process of alteration and consequent compositional change of the rocks can be modeled and quantified. Weathering resulted in the loss of an enormous mass of contaminant-trace elements and their presumed release into the environment. We document significant stratigraphic variations in chemistry, some of which are related to changes in lithology. Considerable compositional variation occurs laterally along strike as the result of weathering and alteration along fractures. Analysis of geochemical compositional data shows that the middle-waste-rock of the Meade Peak, whether highly altered or less-altered, contains elevated concentrations of several trace elements compared to average shale. These elevated concentrations are sufficient to suggest possible potential geoenvironmental problems in conjunction with disposal of the waste-rock. Based on elevated concentrations of trace elements, the less-altered waste-rock offers significantly higher potential to release large quantities of trace elements than altered waste-rock.
ACKNOWLEDGMENTS We thank the mining companies for allowing access and sampling. We appreciate help in sample preparation by M. Fallin and K. Long. P. Lamothe provided helpful insights into the quality of instrumental techniques and quality of analytical data. J. Connor and G. Desborough provided useful discussion about principal component analysis and geochemical relationships. J. Hein, M. Amacher, and G. Desborough provided helpful reviews. Brandie Mclntyre provided invaluable manuscript assistance.
364
J.R. Herring and R.I. Grauch
REFERENCES Altschuler, Z.S., Clarke, R.S., Jr. and Young, E.J., 1958. Geochemistry of uranium in apatite and phosphorite. US Geol. Surv., Prof. Paper, 314-D, pp. 45-90. Arbogast, B.E (ed.), 1996. Analytical methods manual for the Mineral Resource Surveys Program, US Geol. Survey. US Geol. Surv., Open File Report, 96-525, 248 pp. Baedecker, P.A. (ed.), 1987. Methods for geochemical analysis. US Geol. Survey, Bulletin, 1770, variously paginated. Briggs, P.H., 1996. Forty elements by inductively coupled plasma-atomic emission spectrometry. In: B.E Arbogast (ed.), Analytical Methods Manual for the Mineral Resource Surveys Program, US Geol. Survey. US Geol. Surv., Open File Report, 96-525, pp. 77-94. Brittenham, M.D., 1976. Permian Phosphoria carbonate banks, Idaho-Wyoming thrust belt. In: J.G. Hill (ed.), Symposium on Geology of the Cordilleran Hingeline. Rocky Mountain Association of Geologists - 1976 symposium, Denver, pp. 173-191. Christie, A., 1961. The Pale Horse. Collins, London, 148 pp. Claypool, G.E., Love, A.H. and Maughan, E.K., 1978. Organic geochemistry, incipient metamorphism, and oil generation in black shale members of Phosphoria Formation, Western Interior United States. Am. Assoc. Pet. Geol. Bull., 62: 98-120. Cohen, A.C., Jr., 1959. Simplified estimators for the normal distribution when samples are singly censored or truncated. Technometrics, 1:217-237. Cressman, E.R. and Swanson, R.W., 1964. Stratigraphy and petrology of the Permian rocks of southwestern Montana. US Geol. Surv., Prof. Paper, 313-C, pp. 275-569. Garrels, R.M. and Mackenzie, ET., 1971. Evolution of Sedimentary Rocks. Norton, New York, 397 pp. Grauch, R.I., Tysdal, R.G., Johnson, E.A., Herring, J.R., and Desborough, G.A., 2001. Stratigraphic section and selected semiquantitative chemistry, Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-20-E, 1 plate with text. Gromet, EL., Dymek, R.E, Haskin, L.A., and Korotev, R.L., 1984. The "North American shale composite"- its compilation, major and trace element characteristics. Geochim. Cosmochim. Acta, 48: 2469-2482. Gulbrandsen, R.A., 1966. Chemical composition of phosphorites of the Phosphoria Formation. Geochim. Cosmochim Acta, 30: 769-778. Gulbrandsen, R.A., 1974. Buddingtonite, ammonium feldspar, in the Phosphoria Formation, southeastern Idaho. US Geol. Surv., J. Res., 2: 693-697. Gulbrandsen, R.A. and Krier, D.J., 1980. Large and rich phosphorus resources in the Phosphoria Formation in the Soda Springs area southeastern Idaho. US Geol. Surv., Bull., 1496, 25 pp. Herring, J.R., Desborough, G.A., Wilson, S.A., Tysdal, R.G., Grauch, R.I., and Gunter, M.E., 1999. Chemical composition of weathered and unweathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation- A. Measured sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-147-A, 24 pp. Herring, J.R., Wilson, S.A., Stillings, L.A., Knudsen, A.C., Gunter, M.E., Tysdal, R.G., Grauch, R.I., Desborough, G.A., and Zielinski, R.A., 2000a. Chemical composition of weathered and less weathered strata of the Meade Phosphatic Shale Member of the Permian Phosphoria Formation - B. Measured sections C and D, Dry Valley, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-147-B, 34 pp.
Lithogeochemistry of the Meade Peak Phosphatic Shale Member
365
Herring, J.R., Grauch, R.I., Desborough, G.A., Wilson, S.A., and Tysdal, R.G., 2000b. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation- C. Measured sections E and F, Rasmussen Ridge, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-147-C, 35 pp. Herring, J.R., Grauch, R.I., Tysdal, R.G., Wilson, S.A., and Desborough, G.A., 2000c. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation - D. Measured sections G and H, Sage Creek area of the Webster Range, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-147-D, 38 pp. Herring, J.R., Grauch, R.I., Siems, D.E, Tysdal, R.G., Johnson, E.A., Zielinski, R.A., Desborough, G.A., Knudsen, A. and Gunter, M.E., 2001. Chemical composition of strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation. Channel-composited and individual rock samples of Measured Section J and their relationship to Measured Sections A and B, Central Part of Rasmussen Ridge, Caribou County, Idaho, US Geol. Surv., Open File Report, 01-195, 72 pp. Isaacs, C.M., 1980. Diagenesis in the Monterey Formation examined laterally along the coast near Santa Barbara, California. US Geol. Surv., Open File Report, 80-060, 329 pp. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M., and Swanson, R.W., 1959. The Phosphoria, Park City, and Shedhorn Formations in the Western Phosphate Field. US Geol. Surv., Prof. Paper, 313-A, 47 pp. McKelvey, V.E. and Carswell, L.D., 1967. Uranium in the Phosphoria Formation. In: L.A. Hale (ed.), Anatomy of the Western Phosphate Field. Salt Lake City lntermountain Association of Field Geologists, 15th Annual Field Conference Guidebook, pp. 119-123. McKelvey, V.E., Strobell, Jr., J.D., and Slaughter, A.L., 1986. The Vanadiferous Zone of the Phosphoria Formation in western Wyoming and southeastern Idaho. US Geol. Surv., Prof. Paper, 1465, 27 pp. Medrano, M.D. and Piper, D.Z., 1995. Partition of minor elements and major-element oxides between rock components and calculation of the marine-derived fraction of the minor elements in rocks of the Phosphoria Formation, Idaho and Wyoming. US Geol. Surv., Open File Report, 95-270, 79 pp. Miesch, A.T., 1976. Sampling designs for geochemical surveys - Syllabus for a short course. US Geol. Surv., Open-File Report, 76-772, 127 pp. Montgomery, K.M. and Cheney, T.M., 1967. Geology of the Stewart Flat quadrangle, Caribou County, Idaho. US Geol. Surv., Bull., 1217, 63 pp. Piper, D.Z., 2001. Marine chemistry of the Permian Phosphoria Formation and Basin, southeast Idaho. Econ. Geol., 96: 599-620. Oberlindacher, H.P., 1990. Geologic map and phosphate resources of the northeastern part of the Lower Valley quadrangle, Caribou County, Idaho. U.S. Geol. Surv., Misc. Field Studies Map, MF-2133, scale 1: 12,000. Sheldon, R.P., 1959. Geochemistry of uranium in phosphorites and black shales of the Phosphoria Formation. US Geol. Surv., Bull., 1084-D, pp. 81-113. Siems, D.E, 2000. The Determination of 30 elements in geological materials by energy-dispersive x-ray fluorescence spectrometry. US Geol. Surv., Open File Report, 00-475, 13 pp. Turekian, K.K and Wedepohl, K.H., 1961. Distribution of the elements in some major units of the earth's crust. Geol. Soc. Am. Bull., 72: 175-191. Tysdal, R.G., Johnson, E.A., Herring, J.R. and Desborough, G.A., 1999. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of the Permian Phosphoria
366
J.R. Herring and R.I. Grauch
Formation, central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-20-A, 1 plate with text. Tysdal, R.G., Herring, J.R., Desborough, G.A., Grauch, R.I. and Stillings, L.A., 2000a. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Dry Valley, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-20-B,1 plate with text. Tysdal, R.G., Grauch, R.I., Desborough, G.A. and Herring, J.R., 2000b. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation, east-central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-20-C, 1 plate with text. Tysdal, R.G., Herring, J.R., Grauch, R.I., Desborough, G.A. and Johnson, E.A., 2000c. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Sage Creek area of Webster Range, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-20-D, 1 plate with text. Wedepohl, K.H. (ed.), 1969-1978. Handbook of Geochemistry. Springer-Verlag, Berlin.
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
367
Chapter 13
ROCK LEACHATE GEOCHEMISTRY OF THE MEADE PEAK PHOSPHATIC SHALE M E M B E R OF THE P H O S P H O R I A FORMATION, SOUTHEAST IDAHO
J.R. HERRING
ABSTRACT The interaction between water and the Meade Peak Phosphatic Shale Member of the Phosphoria Formation has been studied in a series of laboratory leachate experiments. These experiments directly measure reactivity of rock powders with water and indicate the potential for dissolution and release of various potentially contaminant trace elements into the environment. The leachate protocol used deionized water with pH of about 5.5 as the leaching medium with a water/rock ratio of 20 by mass. The leachate experiments were conducted passively, with only a gentle initial shake to suspend the rock particles. The experiments mostly lasted 24 h, although splits of a few samples were allowed to react for shorter and longer times to study the effects of leachate time. Most leachates were reacted under the oxidizing conditions of atmospheric exposure, but a few were reacted in an argon atmosphere. The leachate rock samples were ground to <100 mesh (0.15 mm). This particle size is fine grained compared to natural or industrial processes involving these rocks. However, the leachate procedure used splits of the same powders that were used for chemical and mineralogical analyses, which provides direct comparison with data from those analyses. Near-surface strata of the Meade Peak have been highly altered by oxidative weathering, which greatly controls rock composition. Extensive alteration results in complete loss of carbonate and considerable reduction in organic-carbon content and many associated trace elements. In the leachate experiments, the highly altered rocks produced lower concentrations of Mo, Ni, Se, and Zn than the less-altered rocks; the soluble fraction of these elements as a percentage of total-element mass in the bulk rock was greatly reduced as well. In contrast, the less-altered rocks had leachate concentrations and leachate proportions that were much greater relative to of the original rock mass than did the highly weathered rocks. About 10% of the original mass of Mo, Ni, and Zn in the least-altered rocks leached into solution in 24 h, while Cd, Cu, and Se leached between 1 and 10% of the original element mass. Arsenic, Ba, Cr, U, and V exhibited less-reactive behavior with typically only about 0.1% of the original mass of each of these elements released into solution in 24 h.
368
JR. Herring
The rock samples consist of variable proportions of phosphatic, detrital, and organic carbon-rich phases. Leachate concentrations of several trace elements, notably Co, Ni, Cu, Zn, As, Se, Cd, and Sb, are associated with the organic-carbon content of the rock. In turn, a strong correlation exists between these element concentrations and conductivity and sulfate. This demonstrates that rocks enriched in organic matter will leach abundant concentrations of these trace elements. The detrital component of the rocks produced elevated leachate concentrations of AI, Cr, Fe, Ba, Th, and U. Molybdenum does not have a strong correlation with the major elements although it has a weak correlation with organic carbon. The phosphate-rich rocks do not have strong element correlations, although, the concentrations of soluble P and V correlate. The results show that rocks of the Meade Peak are highly reactive with water and release significant quantities of trace elements from finely ground-rock samples in periods as short as 1 h. Furthermore, the Meade Peak rocks that are least altered and organic matter-rich release the greatest amount of potentially contaminant trace elements into solution.
INTRODUCTION
Background The Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation is an important source of phosphate rock. However, recent concern has arisen over several contaminant (potentially toxic) elements, especially Se, that exist in the waste rock associated with mining and their potential release into the environment (Herring and Grauch, Chapter 12). Selenium occurs throughout the Meade Peak in varying concentrations with no particular association with the phosphatic units. Other trace elements with potentially deleterious effects on the environment exist in these rocks as well. This chapter focuses on the following trace elements: As, Ba, Cd, Cr, Cu, Mo, Ni, Sb, Se, Th, TI, U, V, and Zn.
Relevance of leachate experiments Experimental leachate data directly measure reactivity of the rocks of the Meade Peak with water in the laboratory and simulate their potential for reaction with meteoric water, dissolution, and release of trace elements into the environment. Experiments such as these are not necessarily intended to measure the rates of release of the various trace element as much as to serve as a relative screening device that indicates which rock types are most likely to release trace elements. As such, it is less important that the experimental protocol exactly follow some previously published procedure, but it is critically important that the technique be consistent throughout the experiments. Considerable attention has been paid to the bulk chemistry of rocks of the Meade Peak (see references in Herring and Grauch, Chapter 12). However, little attention has been paid to the fraction of trace elements that might be leached from that rock by water. Rocks of
Rock Leachate geochemistry of the Meade Peak Phosphatic Shale Member
369
the Meade Peak are known to be reactive with water, especially when fleshly exposed, as demonstrated by the common occurrence of white precipitates on rock faces that were newly uncovered and wetted by rain (see Grauch et al., Chapter 8). Concentrations of trace elements in Meade Peak rocks do not necessarily bear on their release because even enriched trace elements in some rocks can occur in mineralogical phases that are comparatively resistant to dissolution. For example, much of the abundant Ti and Zr in the rocks occurs in refractory rutile and zircon, respectively. Consequently, Ti and Zr are detected only infrequently in the leachate fluid and in concentrations that are several orders of magnitude less than for some trace elements whose rock concentrations are much smaller than those of Ti and Zr. The leachate samples used in this study cover the entire range of bulk chemical and mineralogical composition of the Meade Peak for typical mining depths in southeast Idaho, as well as the range of their alteration. Meade Peak rocks have a wide range of chemical and mineralogical compositions and can be enriched in detrital minerals, CFA, carbonate minerals, and organic matter (see summary of rock chemistry in Herring and Grauch, Chapter 12, and mineralogy in Knudsen and Gunter, Chapter 7). In addition, Meade Peak rocks have undergone a wide range of alteration, principally because of weathering of near surface rocks which has changed the suite of possible contaminant elements. Generally, extensive alteration leaches much of the original trace-element content of the Meade Peak and changes the chemical and mineralogical composition of the host phases of those elements. The leachate samples used here are splits taken from previously analyzed rock samples of the Meade Peak. Consequently, the leachate chemistry can be directly compared with the bulk chemical composition and mineralogy.
Rock samples
Herring and Grauch (Chapter 12) provide location and background information on the rock samples used in the leachate experiments, including description of the stratigraphic sections of the Phosphoria Formation from which the rock samples were taken. In the field, intervals of strata with more or less uniform lithology were described and sampled. Typically, about 40 channel samples were taken through adjoining intervals of the Meade Peak for each measured section. About 0.5-1 kg of rock was collected for each channel-sample interval. The bulk samples were jaw crushed then ground to < 100 mesh (0.15 mm). Splits for additional samples were obtained using mechanical splitters. The analytical data from these analyses have been archived in a series of reports (Herring et al., 1999, 2000a-c, 2001 ). The sections and the phosphate mines from which the rock samples for the leachate studies were taken are: A and B (Enoch Valley mine), C (Dry Valley mine), and F (Rasmussen Ridge mine). Section J is a continuous vertical core that was drilled at a site that subsequently became the Enoch Valley mine. Section J is the deepest and least altered by weathering of any sections sampled. Finally, the study includes leachate concentration
370
JR. Herring
data from rock samples that were taken from Section Z. This section also was taken from the Enoch Valley mine site. Its depth below ground surface is slightly deeper than typical mid-depth for that mine. The lithologic character of the section indicates that it is generally less altered by weathering compared to Sections A, C, and B, but more altered than Section J. Only a few intervals were sampled in Section Z, mostly in the middle-waste shale of the Meade Peak, but within these intervals, the rocks were sampled in detail using close channel sample spacing that ranged from 4 to 10 cm. The intervals that were selected for sampling had relatively high concentrations of Se, although a few intervals were selected with relatively low concentrations of Se for comparison. Consequently, the rock and leachate concentration data for Section Z are not representative for the entire section of Meade Peak at that locality. The six sampled sections vary in degree of alteration. Sections, A, C, and F were closest to the premining ground surface and were the most altered. Section F was sampled only a few meters below the pre-mining ground surface and is the most altered section of the study. Only one sample of phosphatic mudstone from the middle waste rock of Section F was included in the leachate analysis. Sections B and Z are mid-depth rocks for typical current mining practices and are considered to be less altered. Section J is the deepest section sampled in the study and is considered to be the least altered. In addition to channel samples, a carbon-seam sample, J95.4K, was analyzed from Section J (Herring et al., 2001).
METHODS
Sample preparation The technique used in this study is a modification of the synthetic precipitation leaching procedure (SPLP) termed Method 1312 by the US Environmental Protection Agency (1994). Previous researchers successfully used this technique or modifications of it to determine the relative tendency of various contaminant trace elements to leach from rocks into the surrounding environment (Desborough et al., 1999; Hageman and Briggs, 2000; Hageman et al., 2000a,b). The leachate procedure used here varies from Method 1312 and is more similar to that of Hageman and Briggs (2000). Here, even smaller particle sizes of the rock samples were used, there was no mild acidification of the leachate water, and the samples were not continuously agitated during the leaching. The analytical protocol for the rock-water leachate experiments uses 2.5 g aliquots of each ground (<100 mesh; <0.15 mm) rock sample mixed with 18 MI) deionized water in a water/rock ratio of 20 by mass. The pH of the deionized water was unregulated and typically was around 5.5 after equilibration with the atmosphere but prior to mixing with the rock. As the water had no effective buffering capacity, it quickly attained a pH that was determined by its reaction with the minerals of the sample. Leaching was passive, with no continuous agitation or shaking other than an initial gentle shake using multiple inversions
Rock Leachate geochemistry of the Meade Peak Phosphatic Shale Member
3 71
of the capped centrifuge tube to ensure wetting of all ground rock followed by a second similar shake to repeat the resuspension of solids after 1 h. Rock samples were allowed to react with water at room temperature, about 22-25~ mostly for 24 h, but shorter and longer reaction times also were used for a few samples to study changes in leachate concentration with time. Leachate samples were maintained at room temperature and generally oxidizing conditions by allowing the headspace volume of the centrifuge tube to equilibrate with the atmosphere. Either the cap was left loose after the second agitation or it was removed and quickly replaced to allow equilibrium between the headspace gas and the atmosphere. Samples that were allowed to react longer than one day were gently shaken once daily to re-suspend the material and maintain contact between the rock and water; also, the caps of these samples were loosened to maintain continued headspace gas equilibrium with the atmosphere. After the reaction time, samples were centrifuged to separate most of the solids from the solution, then the solution was decanted into a syringe and filtered through a 0.45 p~m pore size, cellulose nitrate, surfactant-free, depth filter. Both syringe and filter were triple rinsed with deionized water and shaken dry prior to filtration. Immediately after filtration, the conductivity, total dissolved solids (TDS), and pH of the filtrates were determined. Filtrate samples for metals and trace element analysis were acidified with Ultrex HNO3 to a pH between 1 and 2, and filtrates for anion were left untreated and refrigerated until analysis. A subset of 12 24-h leachate samples was selected for varying leachate conditions. The subset was picked to include some samples that had extreme differences in conductivity in the leachate of the 24-h experiments; in 24-h leachable Se concentrations; and in the organic-carbon content of the original bulk-rock samples. The 24-h leachate concentrations of Se ranged from 1 to 485 p~g L-~. Concentrations of elements in the leachate solutions are a function of surface area of the grains that were leached (e.g. Hageman and Briggs, 2000; Smith et al., 2000). In the protocol used for the experiments in this study, there was no specific control over particle size other than to grind the entire rock sample to < 100 mesh (0.15 mm). Consequently, the final particle-size distribution is unknown. However, all samples were treated and handled the same. The key reason for using this material is that the leachate rock sample was a split of the ground well-characterized rock. Thus, leachate results can be compared with the chemical and mineralogical composition of the solid sample.
A nalyses
Acidified leachate samples were analyzed by inductively coupled plasma-mass spectrometry (ICP-MS) for trace elements and by ion chromatography (IC) for anions. The ICP-MS techniques, including quality control, accuracy, and precision, are detailed by Lamothe et al. (1999) and, those for IC, by d'Angelo and Ficklin (1996). Statistical treatment of the data to remove qualified concentration data are detailed by Herring and Grauch (Chapter 12).
372
JR. Herring
RESULTS
Many rocks of the Meade Peak, especially when little altered, contain abundant carbonate minerals that buffer them against acid generation as they weather. This causes most water-rock interaction solutions, including the leachate experiments here, to have generally abundant bicarbonate ions and circumneutral pH. Although bicarbonate was not measured directly, the approximate bicarbonate concentration was estimated by calculating the charge difference between the major cation concentrations and those of the anions (Herring and Grauch, Chapter 12). These estimated bicarbonate concentrations range from 4 to 741 mg L -~ . Typically, they are roughly equal to the sulfate concentration for samples with low conductivity up to approximately 500 txS cm-~; for samples with conductivities 180~ 160~
~ Se measured
40 Se log-scaled
3530-
u~ 120 E if) o
~
25-
100 20-
80
.Q
E -,
60
Z
40 20
1600
-1
log concentration
Concentration (ppb) of 16 equal intervals 25O
1
U log-scaled
U measured 200
~ 15o.j N
50-'
E= Z
0 0
120
-2.75
log concentration
2.5
Concentration (ppb) of 20 equal intervals
Fig. 13-1. Representative histograms of concentrations o f trace elements in leachate solutions.
Rock Leachate geochemistry of the Meade Peak Phosphatic Shale Member
373
of about 500 txS cm-l and higher, the sulfate usually exceeds the bicarbonate by a factor of 2-10. However, even the least-altered rocks have some intervals without carbonate minerals in which the initial leachate solutions have a pH as low as 4. As a first assumption, leachate concentration data are expected to mimic elemental concentration distributions in the Meade Peak. In other words, the distributions of the leachate concentrations for the trace elements should be log-normal if they follow typical traceelement concentration distributions (Herring and Grauch, Chapter 12). The initial approach used here is to examine concentration distributions, for example that of Se (Fig. 13-1), and visually compare the frequency distributions of the concentrations vs. the logarithm of those concentrations. The concentration frequency data for all elements demonstrate that the logarithm-scaled concentrations have a more normal frequency distribution than those of the measured data. This indicates that element concentration distributions often include a few highly elevated values that greatly skew their arithmetic means toward higher values. Because the concentration distributions for the elements are more closely log-normal (Fig. 13-1), the geometric mean (GM) is reported with the arithmetic mean in Table 13-I. Herring and Grauch (Chapter 12) provide a complete discussion of GMs and deviations along with their calculation. Measured and log-scaled leachate concentration distributions for various elements are shown in the CD Appendix D. Those elements that have their highest leachate concentrations in the least-altered rock are SO42-, Mg, Ca, Co, Ni, Zn, Se, Sr, Mo, Cd, and W. Those with highest leachate concentrations in the altered rock are AI, Si, P, Ti, V, Cr, Y, La, Ce, and Nd.
DISCUSSION For all elements, the GM is lower than the arithmetic mean and in some samples is as much as a factor of 10 lower (Table 13-I). Most of these marked differences occur in leastaltered Section J samples and the others occur in less-altered Sections B and Z, which suggests that less-altered rock produces the greatest leachate concentrations for most elements. The trace elements that exhibit the most marked differences between their arithmetic and GM concentrations in the leachates are Co, Ni, Cu, Zn, and Cd. A considerable difference in element concentrations exists as a function of the alteration of the rock samples (Table 13-I). Weathering leached much of the trace-element load from the rocks. The five sections in Table 13-I are listed in terms of decreasing alteration. TDS concentrations in the altered rocks are much lower, as well. GM conductivity in the least-altered Section J samples is at least a factor of 10 greater than for rock leachates from the two most-altered Sections, C and A, and the two less-altered Sections, B and Z, have intermediate TDS concentrations.
Correlation analysis Pearson correlation coefficients were calculated for the dataset using the logarithms of the leach element concentrations and measured values of pH and conductivity (Table 13-1I).
Companson of anthmetlc and G M element concentraltans I" 24-h leachate solut~ons conduct~\~t~e\. pS~.m . an~on,. mg L pH
Src rjun F-h~ghl,ulrrrr,d stnglr \an!plr Sectton C-altered Ar~thmettcmean Ceometrlc mean Geametrlc De\aat!on Central Range. upper Central Range. louer Expected Range. upper Expecled Range. lower Sectton A-alrerrd Ar~thrnet~c mean Ceometrtc mean Geometrlc Deblatlon Central Range. upper Central Range. lower Expected Range. upper Expected Range. lower Srcrmn B-le,, alter<,d Anthmet~cmean Geometr~cmean Geornetr~cDe\ latlon Central Range. upper Central Range. lower Expected Range. upper Expected Range. lower Secoon Z - l m alrvred Ar~thmet~c mean Geornetrlc mean Geometr~cDei tanon Central Range. upper Central Range. lower Expected Range. upper Expected Range. louer Secoon J-learr ulrrrrd Arlthmetlc mean Geometrlc mean Geometr~cDe\latlon Central Range, upper Central Range. lower Expected Range. upper Expected Range. lower Detection ratio
Conductl\ ItV
F
CI
NO;
SO,
LI
\a
'. othcrr. &gL
'
W
4
P
Section
Section F-highly altered," single sample Section C-altered Arithmetic mean Geometric mean Geometric Deviation Central Range. upper Central Range, lower Expected Range, upper Expected Range, lower
Cu 12
8 4 0.5 4.8 1.3 8.9 0.7
Ga
Ge
i0
Zn
0.03
0.02
49 14 i.0 29.3 28.0 29.9 27.5
0.2 0.1 0.57 0.18 0.06 0.30 0.03
0.1 0.03 0.63 0.10 0.04 O. 15 0.02
As 2
4 4 0.31 17.7 1.70 54.4 0.55
Se 4
5 0.3 1.09 2.79 3.30 2.58 3.57
Y
Mo
Cd
Sb
Cs
W
TI
2.4
Rb
6
Sr
O. 15
1.6
0
1.5
0.03
0.3
Ba
0.09
La
0.08
Ce
0.05
Nd
0.20
0.30
0.05
Th
0.21
U
2 2 0.4 8.7 1.1 23.3 0.4
14 11 0.6 28.9 9.9 48.4 5.9
"~.0 1.0 0.86 0.37 0.27 0.42 0.23
2.6 2.0 1 12 7 15 6
2.0 1.0 0.7 !.9 0.8 2.8 0.5
1.0 0.7 0.5 2.8 0.6 6.1 0.3
0.1 0.05 0.53 0.12 0.03 0.22 0.02
7.6 2.5 0.60 3.21 1.17 5.22 0.72
1.2 0.57 0.81 0.26 0.17 0.31 0.14
0.6 0.28 0.74 0.15 0.08 0.20 0.06
0.9 0.39 0.80 0.17 0.11 0.22 0.09
0.1 0.09 0.66 0.16 0.07 0.23 0.05
0.3 0.18 0.49 0.66 0.15 1.31 0.08
0.07 0.03 0.57 0.06 0.02 0.10 0.01
0.7 0.3 0.64 0.29 0.12 0.45 0.08
(~ r ~"
~_,
Section A-altered Arithmetic mean Geometric mean Geometric Deviation Central Range, upper Central Range, lower Expected Range, upper Expected Range, lower
tt~~
4 3 0.4 1.6 0.2 4.1 0.1
32 15 0.6 18.5 7.8 28.0 5.2
O. 1 0. ! 0.44 1.56 0.30 3.47 0.13
0.2 0.1 0.55 0.18 0.05 0.31 0.03
10 9 0.33 20.7 2.31 59.3 0.81
15 5 0.63 60.4 24.3 93.5 15.7
5 5 0.2 23.1 1.2 96.0 0.3
10 8 0.4 185.5 31.1 437.4 13.2
0.6 0.3 0.62 0.04 0.02 0.07 0.01
12.6 8.2 1 178 98 238 73
1.2 0.8 0.6 1.1 0.4 1.9 0.2
3.2 2.7 0.3 10.0 1.0 30.7 0.3
O. 1 O. I 0.21 0.50 0.02 2.22 0.01
1.1 0.8 0.39 7.62 1.18 18.63 0.48
0.5 0.22 0.57 0.04 0.01 0.07 0.01
0.3 0. I 1 0.39 0.04 0.01 0.11 0.003
0.3 0.14 0.47 0.04 0.01 0.08 0.004
1.0 0.09 0.68 0.48 0.22 0.69 0.15
0.9 0.58 0.77 0.41 0.24 0.53 0.19
0.06 0.04 0.49 0.03 0.01 0.05 0.003
0.3 0.1 0.38 1.91 0.28 4.78 0.11
2 1 0.5 2.9 0.7 5.9 0.3
253 16 1.0 27.2 28.0 26.9 28.4
0.3 0.2 0.62 0.32 0.12 0.50 0.08
0.4 0.2 0.59 0.19 0.07 0.32 0.04
12 I0 0.27 18.5 1.38 64.0 0.40
55 25 1.18 8.65 12.0 7.38 14.1
7 5 0.3 16.0 1.2 54.8 0.4
37 21 0.6 61.7 20.7 104.2 12.3
0.2 0.1 0.78 0.12 0.07 0.15 0.06
81.9 33.8 ! 37 37 37 37
2.4 1. i 0.6 2.0 0.8 3.2 0.5
3.0 2.0 0.3 6.2 0.7 18.1 0.2
0.1 0.1 0.33 0.27 0.03 0.78 0.01
1.6 1.0 0.49 4.06 0.97 8.07 0.49
0.2 0.08 0.71 0.10 0.05 0.14 0.04
0.1 0.04 0.57 0.07 0.02 0.13 0.01
0.1 0.05 0.64 0.08 0.03 0.12 0.02
0.2 0.04 0.68 0.38 0.17 0.56 0.12
1.4 0.84 0.57 0.82 0.27 1.41 0.16
0.02 0.02 0.50 0.07 0.02 0.14 0.01
0.7 0.3 0.53 0.56 0.16 1.02 0.09
3 1 0.5 2.4 0.5 4.9 0.3
2606 72 1.4 49.4 103.7 34.6 148.0
0.3 O. 1 0.72 O. 17 0.09 0.23 0.07
0.8 0.5 0.41 I. 16 0.20 2.72 0.08
6 5 0.30 15.8 1.43 50.4 0.45
143 89 0.41 214.5 36.72 500.6 15.74
8 6 0.3 18.7 2.0 54.8 0.7
123 61 0.7 88.4 41.6 127.0 28.9
0.6 0.04 0.91 0.05 0.04 0.05 0.04
304.0 ! i 5.2 1 153 87 201 66
49.9 1.1 1.4 0.8 1.6 0.6 2.2
9.3 6.6 0.4 17.0 2.6 42.0 1.0
0.1 O. 1 0.33 0.26 0.03 0.75 0.01
3.5 2.3 0.48 4.86 1.10 9.92 0.54
0.3 0.03 0.80 0.04 0.02 0.05 0.02
0.2 0.03 0.75 0.04 0.02 0.05 0.02
0.2 0.02 0.72 0.03 0.02 0.05 0.01
1.5 0.38 0.94 0.40 0.35 0.43 0.33
0.8 0.22 0.81 0.27 0.18 0.33 0.14
0.02 0.01 0.42 0.02 0.003 0.04 0.001
0.3 0.2 0.63 0.27 0.11 0.42 0.07
160 3 1.0 11.1 12.1 10.6 12.6 84%
14142 475 1.7 165.4 451.9 102.1 732.0 95%
1.2 0.5 0.70 0.22 0.11 0.31 0.08 93%
0.4 0.2 0.72 0.19 0.10 0.26 0.07 8800
I0 7 0.38 21.4 3.05 54.5 1.20 100%
194 114 1.23 11.9 17.8 9.78 21.7 93%
6 6 0.4 10.5 1.6 26.1 0.6 1000,o
315 230 0.9 45.2 37.5 49.4 34.3 100%
23.9 O. 1 1.11 0.70 0.87 0.63 0.97 74%
404.3 141.6 1 16 9 20 7 100~
173.9 7.4 1.3 4.9 8.7 3.8 11.5 9300
3.1 2.3 0.4 4.2 0.6 10.7 0.2 100%
0.1 O. 1 0.44 0.18 0.04 0.41 0.02 97%
10.1 7.6 0.68 4.34 1.98 6.32 1.36 99%
9.5 0.07 1.06 0.46 0.51 0.44 0.54 70%
0.7 0.30 0.54 0.13 0.04 0.22 0.02 86%
1.0 0.36 0.51 0.91 0.24 1.73 0.13 87%
1.91 0.04 0.76 0.07 0.04 0.09 0.03 77%
4.3 0.5 0.88 0.32 0.24 0.36 0.22 100%
"* r
Section B-less altered Arithmetic mean Geometric mean Geometric Deviation Central Range. upper Central Range. lower Expected Range, upper Expected Range, lower
~"
~' ,~.
Section Z-less altered Arithmetic mean Geometric mean Geometric Deviation Central Range, upper Central Range, lower Expected Range, upper Expected Range, lower
-;~ r "~
Section J-least altered Arithmetic mean Geometric mean Geometric Deviation Central Range, upper Central Range, lower Expected Range, upper Expected Range, lower Detection ratio
11.4 13.0 0.05 0.06 1.09 1.08 0.23 0.28 0.27 0.32 0.21 0.26 0.29 0.35 67% 67%
tad -.-,.I
tao TABLE 13-II Correlation coefficients between leachate elements based on log-scaled concentrations and whole-rock mineralogy: values i>0.65 and <~- 0.65 are bold
Conductivity F CI NO3 SO4 Li Na Mg Ai Si P K Ca Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Rb Sr Y Mo Cd Sb Cs Ba La Ce Nd W TI Th
pH
Cond.
F
CI
NO3
SO4
Li
Na
Mg
AI
- 0.11 -0.54 0.13 0.09 -0.20 -0.31 -0.07 0.17 -0.53 -0.13 -0.39 -0.08 -0.06 -0.23 0.11 -0.31 -0.47 -0.62 -0.54 -0.58 -0.67 -0.65 0.35 -0.24 -0.06 -0.10 -0.17 -0.05 -0.67 0.36 -0.63 0.24 -0.16 -0.23 -0.63 -0.62 -0.60 0.45 -0.42 -0.42
0.42 0.59 0.54 0.97 0.46 0.09 0.88 -0.03 -0.30 -0.41 0.37 0.96 -0.38 -0.46 -0.16 0.77 0.22 0.77 0.73 0.12 0.62 0.40 0.43 0.05 0.75 0.59 0.91 -0.13 0.63 0.42 0.32 0.48 0.63 -0.13 -0.10 -0.13 0.31 0.09 0.08
0.15 0.17 0.39 0.54 0.37 0.21 0.47 0.26 0.35 0.47 0.43 0.24 0.10 0.40 0.52 0.47 0.51 0.59 0.61 0.61 0.12 0.39 0.40 0.26 0.47 0.43 0.52 0.06 0.66 0.10 0.33 0.32 0.53 0.43 0.46 -0.14 0.50 0.48
0.49 0.58 0.25 0.25 0.63 0.08 -0.19 -0.43 0.23 0.56 -0.39 -0.28 -0.05 0.33 0.12 0.32 0.26 -0.07 0.20 0.36 0.20 0.02 0.45 0.34 0.53 -0.19 0.41 O. 19 0.30 0.30 0.30 -0.16 -0.14 -0.18 0.23 0.05 0.04
0.53 0.40 0.32 0.57 -0.09 -0.03 -0.46 0.46 0.49 -0.28 -0.10 -0.14 0.38 -0.14 0.30 0.26 -0.17 0.20 0.27 0.26 0.12 0.51 0.54 0.45 -0.36 0.52 0.09 0.45 0.45 0.24 -0.34 -0.34 -0.35 0.29 0.19 -0.00
0.53 0.19 0.81 0.11 -0.35 -0.44 0.31 0.93 -0.47 -0.53 -0.16 0.77 0.22 0.79 0.73 0.13 0.65 0.41 0.37 0.07 0.75 0.51 0.91 -0.16 0.67 0.59 0.31 0.41 0.64 -0.14 -0.13 -0.17 0.25 O. 11 0.19
0.43 0.39 0.41 0.30 0.12 0.51 0.48 0.07 0.04 0.33 0.48 0.42 0.44 0.53 0.37 0.53 0.33 0.33 0.22 0.31 0.56 0.46 0.34 0.20 0.52 0.06 0.67 0.51 0.34 0.30 0.33 0.08 0.42 0.47
0.16 0.17 0.28 0.11 0.42 0.09 0.08 0.11 0.09 0.08 0.09 0.08 0.10 0.11 0.13 -0.08 0.16 0.08 0.16 0.26 0.05 0.12 -0.04 0.21 0.02 0.23 -0.03 0.12 0.08 0.09 -0.13 0.32 0.16
-0.07 -0.22 -0.49 0.34 0.88 -0.42 -0.41 -0.16 0.56 0.09 0.52 0.47 -0.05 0.37 0.49 0.29 0.03 0.67 0.50 0.77 -0.24 0.59 0.25 0.33 0.47 0.48 -0.23 -0.19 -0.22 0.33 -0.02 -0.00
0.27 0.48 -0.03 -0.02 0.10 0.12 0.72 0.15 0.71 0.12 0.16 0.65 0.29 0.17 0.01 0.22 -0.12 0.01 -0.03 0.76 -0.45 0.53 -0.35 0.25 0.26 0.79 0.76 0.76 -0.47 0.39 0.62
Si
0.66 0.26 -0.17 0.56 0.34 0.39 -0.07 0.21 -0.20 - 0.04 0.28 0.05 -0.04 -0.15 0.22 -0.41 0.03 -0.21 0.41 -0.41 0.07 -0.26 0.21 - 0.05 0.40 0.34 0.40 -0.22 0.32 0.33
P
K
Ca
Ti
V
Cr
Mn
Fe
Co
Ni
-0.41 -0.35 0.33 0.65 0.54 -0.14 0.26 -0.19 -0.07 0.50 0.11 -0.08 -0.17 0.37 -0.40 -0.46 -0.32 0.47 -0.50 0.42 -0.39 -0.21 -0.02 0.50 0.41 0.45 -0.51 0.37 0.46
0.36 0.03 -0.00 -0.07 0.39 0.06 0.37 0.38 -0.03 0.32 0.05 0.45 0.08 0.35 0.84 0.30 -0.00 0.32 0.25 0.35 0.67 0.20 -0.02 -0.05 -0.04 0.14 0.47 -0.04
-0.37 -0.43 -0.10 0.77 0.23 0.74 0.71 0.13 0.61 0.45 0.39 0.04 0.67 0.56 0.95 -0.11 0.59 0.41 0.29 0.52 0.68 -0.11 -0.10 -0.12 0.31 0.07 0.10
0.37 0.36 -0.13 0.08 -0.18 -0.02 0.36 -0.00 -0.18 -0.07 0.50 -0.32 -0.06 -0.43 0.37 -0.43 -0.09 -0.24 -0.10 -0.36 0.36 0.32 0.36 -0.21 0.15 0.24
0.36 -0.44 -0.15 -0.44 -0.33 0.11 -0.27 0.02 0.13 0.57 -0.21 -0.12 -0.44 0.09 -0.12 0.01 0.11 -0.06 -0.38 0.12 0.05 0.09 -0.10 0.37 0.14
0.00 0.63 -0.05 0.04 0.64 0.18 0.32 0.02 0.28 -0.25 0.01 -0.09 0.74 -0.43 0.31 -0.21 0.29 0.21 0.76 0.74 0.76 -0.29 0.22 0.62
0.48 0.92 0.88 0.36 0.83 0.13 0.45 -0.03 0.55 0.56 0.73 0.20 0.33 0.64 0.12 0.51 0.67 0.19 0.22 0.18 0.06 0.27 0.28
0.48 0.49 0.65 0.60 0.08 0.17 -0.01 0.03 0.23 0.19 0.81 -0.36 0.59 -0.27 0.42 0.43 0.82 0.85 0.83 -0.34 0.34 0.67
0.95 0.39 0.90 0.01 0.55 -0.02 0.63 0.59 0.71 0.21 0.32 0.68 0.19 0.51 0.62 0.19 0.22 0.17 0.02 0.28 0.22
0.50 0.95 0.01 0.55 0.11 0.59 0.57 0.68 0.28 0.28 0.73 0.15 0.51 0.59 0.26 0.26 0.23 -0.03 0.35 0.29
2. 0~
U CFA Quartz Muscovite Illite Albite Orthoclase Buddingtonite Dolomite Calcite Kaolinite Montmorillonite Pyrite Sphalerite Gypsum Organic C
0.03 -0.16 -0.35 -0.31 -0.19 -0.05 -0.06 0.05 0.41 0.64 -0.06 0.12 0.14 0.23 -0.16 -0.13
0.30 -0.06 -0.16 -0.02 -0.13 0.13 -0.03 0.04 0.42 0.41 -0.21 0.01 0.49 0.12 0.18 0.65
0.21 0.47 0.08 0.15 -0.14 -0.14 -0.07 -0.17 -0.41 -0.49 -0.02 -0.05 0.06 -0.25 0.13 0.26
0.20 -0.14 -0.19 -0.23 -0.29 0.13 0.09 0.12 0.23 0.28 0.04 -0.14 0.30 0.03 -0.04 0.31
-0.01 -0.02 -0.18 0.00 -0.26 0.05 0.10 0.18 0.31 0.10 -0.10 0.08 0.25 -0.01 0.10 0.51
0.23 0.01 -0.20 -0.04 -0.17 0.12 -0.04 0.05 0.47 0.32 -0.18 0.02 0.50 0.18 0.19 0.66
0.38 0.26 -0.01 0.13 0.02 -0.07 -0.05 -0.09 -0.02 -0.18 -0.13 0.19 0.29 0.05 0.19 0.52
-0.05 0.18 -0.06 0.00 -0.19 -0.03 0.09 0.05 -0.06 -0.22 0.14 0.35 0.10 0.07 -0.03 0.14
0.33 -0.21 -0.22 -0.11 -0.17 0.17 0.02 0.17 0.54 0.57 -0.19 0.06 0.49 0.11 0.11 0.59
Cu
Zn
Ga
Ge
As
Se
Rb
Sr
Y
Zn Ga Ge As Se Rb Sr Y Mo Cd Sb Cs Ba La Ce Nd W TI Th U CFA Quartz Muscovite Illite Albite
0.58 0.05 0.20 0.34 0.02 0.05 0.13 0.77 -0.35 0.64 -0.22 0.17 0.28 0.78 0.74 0.74 -0.36 0.36 0.58 0.32 0.12 0.33 0.22 0.19 -0.07
-0.04 0.48 0.10 0.46 0.48 0.58 0.43 0.11 0.81 0.05 0.5 i 0.60 0.41 0.39 0.37 -0.15 0.41 0.41 0.18 0.13 0.17 0.21 0.15 0.01
0. I 0 0.23 0.20 0.20 0.47 -0.07 0.43 -0.01 0.25 0.34 0.43 -0.01 0.01 0.02 0.44 -0.16 0.19 0.59 -0.01 -0.27 -0.05 -0.07 0.04
0.30 0.68 0.59 0.33 0.02 0.45 0.46 0.55 0.45 0.18 0.02 0.01 -0.01 0.07 0.29 0.04 0.06 0.33 -0.10 0.14 -0.10 -0.08
0.40 0.02 0.20 0.05 0.21 -0.04 0.02 -0.06 -0.28 -0.21 0.01 -0.10 -0.07 -0.17 0.01 -0.01 Mo
0.21 0.25 0.28 0.20 0.20 -0.22 0.01 -0.09 -0.67 -0.54 0.15 -0.03 -0.46 -0.34 -0.11 -0.38
0.04 0.52 0.25 0.14 0.18 -0.46 -0.16 -0.45 -0.66 - 0 .7 6 0.09 -0.15 -0.65 -0.19 -0.03 -0.40
0.16 0.22 0.00 0.14 -0.19 -0.02 0.12 0.13 -0.12 -0.09 -0.16 0.09 0.45 -0.01 0.06 0.37
0.36 0.01 -0.18 -0.01 -0.11 0.08 -0.10 0.01 0.43 0.41 -0.24 0.00 0.47 0.13 0.18 0.64
-0.06 0.16 0.40 0.31 0.02 0.00 0.20 0.12 -0.61 -0.60 0.22 -0.16 0.01 -0.35 0.03 -0.18
-0.01 0.35 -0.07 -0.06 -0.07 -0.35 -0.00 -0.27 -0.43 -0.32 -0.07 -0.10 -0.38 -0.18 -0.12 -0.21
0.51 0.13 0.12 0.08 0.22 -0.20 -0.05 -0.08 -0.36 -0.38 0.08 -0.02 -0.04 -0.30 0.03 -0.08
0.21 -0.06 0.19 0.14 0.06 0.19 -0.01 0.04 0.05 -0.10 -0.17 -0.07 0.33 0.04 0.18 0.44
0.51 -0.10 0.34 0.20 0.31 0.07 0.05 -0.05 -0.20 -0.30 0.06 -0.08 0.07 -0.22 -0.01 0.02
0.15 0.00 0.15 0.18 0.04 0.13 -0.03 -0.02 0.04 -0.04 -0.17 -0.09 0.29 0.05 0.17 0.53
Cd
Sb
Cs
Ba
La
Ce
Nd
W
TI
Th
0.16 0.11 0.17 0.23 0.07 0.08 - 0.04 -0.04 -0.08 -0.18 -0.14 -0.13 0.21 0.04 0.20 0.58
t~ r
t~ ~" ',~ ~ r
r 0.27 0.09 -0.02 0.09 0.18 0.25 0.25 0.06 -0.14 0.12 0.06 0.07 -0.06 0.25 0.23 0.09 0.24 -0.02 0.12 -0.20 -0.09
0.55 0.59 -0.27 0.65 0.37 0.50 0.37 0.30 -0.26 -0.23 -0.28 0.17 0.11 -0.07 0.03 -0.03 -0.18 0.06 -0.21 0.17
0.50 0.05 0.46 0.31 0.47 0.81 0.37 0.04 0.03 0.03 0.18 0.38 0.07 0.25 0.14 -0.05 0.10 -0.18 0.07
-0.10 0.59 0.39 0.30 0.45 0.77 -0.10 -0.10 -0.11 0.37 0.03 0.11 0.33 0.09 -0.24 -0.03 -0.10 -0.00
~.~ ~'~ -0.62 0.52 -0.39 0.21 0.22 0.99 0.96 0.97 -0.50 0.37 0.65 0.38 0.05 0.43 0.25 0.35 -0.05
-0.02 0.67 0.29 0.29 -0.60 -0.58 -0.60 0.61 -0.13 -0.26 0.08 0.07 -0.33 -0.06 -0.31 0.09
-0.03 0.36 0.46 0.52 0.48 0.46 -0.30 0.59 0.52 0.22 0.23 0.19 0.24 0.12 -0.08
0.28 0.02 -0.36 -0.36 -0.38 0.42 -0.10 -0.27 -0.01 0.24 -0.31 0.03 -0.36 -0.08
0.53 0.21 0.21 0.22 0.17 0.36 0.26 0.46 0.12 -0.09 0.02 0.03 -0.10
0.21 0.22 0.22 0.18 0.08 0.27 0.47 0.03 -0.04 -0.02 0.12 0.01
~" _~ 0.97 0.98 -0.49 0.36 0.67 0.42 0.06 0.42 0.24 0.31 -0.05
0.99 -0.46 0.31 0.67 0.44 -0.10 0.49 0.27 0.35 0.06
-0.45 0.33 0.68 0.46 -0.03 0.45 0.25 0.35 -0.02
-0.27 -0.20 0.13 0.14 -0.27 0.04 -0.06 -0.14
0.35 0.11 0.29 0.06 -0.03 -0.09 -0.18
0.40 0.09 0.21 0.28 0.26 -0.12 Continued taO "..I "..I
--..I OO
TABLE 13-II Continued
Orthoclase Buddingtonite Dolomite Calcite Kaolinite Montmorillonite Pyrite Sphalerite Gypsum Organic C
CFA Quartz Muscovite Illite Albite Orthoclase Buddingtonite Dolomite Calcite Kaolinite Montmorillonite Pyrite Sphalerite Gypsum Organic C
Cu
Zn
Ga
Ge
As
Se
Rb
Sr
Y
Cd
Sb
Cs
Ba
La
Ce
Nd
W
TI
Th
0.03 -0.09 -0.49 -0.52 0.10 -0.06 -0.18 -0.15 0.11 0.09
-0.07 -0.13 -0.11 -0.26 -0.14 -0.12 0.13 -0.00 0.13 0.45
-0.06 0.08 0.38 0.42 -0.23 0.02 0.50 0.21 0.13 0.45
-0.14 -0.02 -0.27 -0.13 -0.14 -0.05 0.28 0.09 0.24 0.60
0.01 -0.03 -0.34 -0.10 -0.17 -0.27 -0.03 -0.18 0.04 0.35
-0.01 0.17 0.40 0.33 -0.22 -0.01 0.52 0.17 0.17 0.80
0.09 0.17 0.07 0.02 -0.20 0.02 0.56 0.00 0.15 0.56
-0.15 -0.10 0.43 0.36 -0.24 0.09 0.44 0.19 0.18 0.59
0.06 -0.09 -0.50 -0.53 0.18 -0.10 -0.11 -0.28 -0.02 -0.19
-0.03 0.11 0.50 0.45 -0.28 0.12 0.57 0.39 0.23 0.66
Mo
-0.04 -0.20 -0.26 -0.33 -0.10 -0.13 0.07 0.02 0.16 0.40
0.04 0.31 0.18 0.22 -0.12 -0.05 0.34 0.27 0.27 0.55
-0.05 0.07 0.27 0.07 -0.17 -0.03 0.45 -0.04 0.05 0.40
-0.14 -0.14 0.37 0.18 -0.17 0.11 0.41 0.09 0.10 0.31
0.06 -0.06 -0.48 -0.52 0.20 -0.13 -0.09 -0.27 0.00 -0.17
0.13 -0.00 -0.42 -0.47 0.22 -0.12 -0.01 -0.27 -0.01 -0.18
0.08 -0.05 -0.43 -0.50 0.20 -0.11 -0.03 -0.28 -0.01 -0.19
-0.11 -0.04 0.42 0.15 -0.22 0.12 0.43 0.30 0.19 0.28
0.15 -0.40 -0.34 -0.38 -0.11 -0.05 -0.13 -0.06 -0.15 -0.00
-0.07 -0.19 -0.23 -0.36 -0.07 -0.05 0.15 -0.10 0.05 0.05
U
CFA
qtz
musc
illite
albite
orth
budd
dolo
calcite
kaolin
mont
pyrite
sphal
gypsum
-0.12 0.02 0.08 0.19 -0.04 -0.01 0.03 0.18 0.38 0.02 0.07 0.38 -0.03 0.17 0.17
-0.29 0.07 -0.18 -0.53 -0.39 -0.40 -0.25 -0.45 -0.07 0.00 -0.11 0.10 0.17 0.15
0.43 0.26 0.48 0.43 0.22 -0.45 -0.35 0.23 -0.18 -0.03 -0.13 - 0.03 -0.17
0.21 0.11 0.02 0.08 -0.27 -0.23 0.15 0.05 0.22 0.03 0.25 0.25
-0.03 -0.12 -0.11 -0.12 -0.19 0.10 0.16 -0.08 -0.16 - 0.08 -0.18
0.44 0.44 0.01 0.19 0.02 0.02 0.34 0.16 - 0.04 0.11
0.31 0.02 -0.07 0.12 -0.01 0.16 0.15 0.03 -0.05
0.15 0.37 0.17 -0.02 0.33 0.14 0.28 0.22
0.50 -0.19 0.11 0.31 0.18 - 0.04 0.28
-0.30 0.13 0.26 0.19 0.08 0.20
0.15 -0.14 -0.15 0.08 -0.16
0.07 0.17 0.18 0.01
0.15 0.23 0.56
0.20 0.30
0.35
Rock Leachate geochemistry of the Meade Peak Phosphatic Shale Member
379
The pH of the leachate solution does not have a strong positive correlation with any element. It has strong negative correlations with Cu and Zn, which suggests that these elements are released into solution by oxidation of pyrite and other sulfide minerals with consequent lowering of pH. Conductivity has a weak negative correlation with pH, indicating in part that lowered pH from oxidation of sulfide releases increasing masses of elements into the leachate. Clearly, other processes besides sulfide oxidation provide a source of SO42-, as the correlation between conductivity and SO42- is very high. For the other elements, conductivity has its strongest correlation coefficients with Mg, Ca, Mn, Co, Ni, Se, and Sr. Magnesium, Ca (in part), Mn, and Sr associate with carbonate minerals (Herring and Grauch, Chapter 12), and their mutual association with conductivity noted here suggests that these elements could be leached from that carbonate, perhaps by acid generated from oxidation of sulfide minerals. Of the anions measured, only SO 2- has several strong correlations, and these are with the same elements that are strongly correlated with conductivity. Aluminium has strong correlations with Cr, Fe, Cu, and some lanthanide elements. Vanadium strongly correlates with P in the leachate, but this association is not evident in the bulk-rock chemistry. Manganese is strongly correlated with Co, Ni, Zn, Sr, and Ba. Cobalt strongly correlates with Ni, Zn, and Cd. Nickel, Cu, and Zn also share a strong correlation with Cd. Arsenic has no strong correlations. Leachate Se strongly correlates with conductivity, SO42-, Mg, Ca, and Ge in the leachate solution and with organic-carbon concentration in the solid phase. Molybdenum strongly associates with SO42-, Se, Sb, and organic-carbon. Several of the lanthanide elements, as expected, show varying strong correlations with each other.
Correlations with major components Sodium is a major element in the leachate solutions and shows no strong correlations with trace elements. Magnesium correlates strongly with Se and Sr, and K correlates strongly with Rb and Cs. Calcium has several strong correlations: Mn, Co, Ni, Se, Sr, and Ba. Its relationship with Mn and Sr has already been suggested to result from association with carbonate minerals. The correlation with Ba may also result from trace occurrence of this element, a congener of Ca, in the carbonate. In the solid-phase samples, Ba correlates strongly with a detrital-mineral fraction that should be relatively insoluble compared to other fractions in the Meade Peak. The Ba concentrations in the leachate solutions are mostly < 10 p~g L-!, and this small amount was likely leached from carbonate minerals.
Correlation with mineralogy Correlation coefficients are also shown leachate concentration data and the unscaled solid phase (Knudsen and Gunter, Chapter among major elements in the leachate and
(Table 13-II) between the logarithm-scaled concentrations of the various minerals in the 7). Some strong positive correlations exist various minerals, but not between leachate
380
J.R. Herring
trace elements and minerals. Molybdenum and Se in the leachate solution show strong correlations with organic-carbon content.
Factor analysis Leachate concentrations of the elements were examined by factor analysis. Organic carbon in the solid samples correlates strongly with several potentially contaminant trace elements, so was included in the factor analysis. For factor analysis, the logarithms of the concentration data, except for pH and conductivity, were used. High values of the loadings do not mean large concentrations of that element in the leachate; the high values denote only common association. Factor loadings are derived using a five-factor principal components analysis of element concentration data using an oblique, rotated matrix solution. Scores or loadings determine the amount of a new hypothetical variable, the factor, that is formed by linear combinations with each of the previous variables, the element concentrations. Factor 1 includes the concentration of the organic carbon in the solid sample and the leachate concentrations of Co, Ni, Cu, Zn, As, Se, Cd, and Sb (Table 13-III). This indicates that leachate concentrations of these elements are proportional to the presence of the organic matter. The strong correlation with conductivity and sulfate demonstrate that the samples enriched in organic matter contain abundant dissolvable material that includes sulfate and several trace elements. This does not identify the origin of the sulfate in the solid sample, which could be either sulfide or soluble sulfate minerals. Factor 2 is similar to a detrital factor that was noted for Meade Peak rock chemistry (Herring and Grauch, Chapter 12). It contains loadings for AI, Cr, Fe, Ba, Th, and U. Factor 3 contains P, V, As, and TI, but the dissolved P concentration does not strongly correlate with CFA content of the rock samples. Factor 4 contains strong loadings for silica, Na, and K, which suggests that it may represent leaching of a clay mineral or feldspar. No trace elements are associated with Factor 4. Molybdenum, W, and Th have high Ioadings in Factor 5 and organiccarbon in the solid sample has a lower loading. The factors that represent the leachate concentrations of the trace elements have some remarkable differences from the factors that describe relationships among trace elements in solid-phase concentrations (Herring and Grauch, Chapter 12). For example, the presence of U in Factor 2 of the leachates is notable, as the element correlates strongly with P in the solid phase. Further, U has no statistically significant loading onto the factor with high loading of leachate P. These differences indicate that the processes that lead to dissolution in the leachate samples are more complex than simply a process that leaches elements into solution in proportion to their abundance in the solid phase. For example, Factor 1 has high loadings of sulfate and Mg, which could be explained by the oxidation of sulfide minerals, which would lower the pH but be buffered by the dissolution of Mg carbonate until such buffering capacity is fully used. Note that pH of the leachate has a weak negative correlation with sulfate (Table 13-II). This would be consistent with lowering of pH by continued sulfide mineral oxidation along with acid dissolution of the
Rock Leachate geochemistry of the Meade Peak Phosphatic Shale Member
381
Table 13-III Factor analysis of log-scaled leachate concentration data and organic-carbon content of rock samples. Loadings ->0.5 and -<-0.5 are bold
pH Cond. SO4 Na Mg A1 Si P K Ca Ti V Cr Mn Fe Co Ni Cu Zn As Se Sr Mo Cd Sb Ba W TI Th
Factor 1
Factor 2
Factor 3
Factor 4
Factor 5
-0.33 0.95 0.93 0.02 0.67 0.47 0.47 0.12 0.54 0.93 0.88 -0.17 0.26 0.63 -0.22 0.75 0.83 0.88 0.85 0.79 1.05 0.93 0.17 0.89 0.87 0.28 -0.75 -0.18 -0.33
-0.07 0.22 0.19 -0.08 -0.09 0.65 -0.45 -0.22 -0.27 0.06 0.26 -0.14 0.76 0.29 1.02 0.11 0.00 0.29 0.00 -0.33 -0.34 -0.10 -0.63 0.00 -0.81 0.67 0.12 0.42 0.73
-0.36 -0.24 -0.27 -0.10 -0.98 -0.01 0.26 0.97 -0.19 -0.04 -0.20 0.88 0.13 -0.04 0.06 0.01 0.10 -0.02 0.08 0.51 0.02 0.17 0.21 0.08 -0.02 0.18 0.16 0.69 0.37 -0.09 -0.19
0.26 0.00 0.08 0.91 0.09 -0.11 0.63 -0.04 0.58 0.07 0.07 -0.09 -0.25 0.37 0.10 0.28 0.16 -0.34 0.11 -0.22 -0.16 -0.02 -0.08 0.01 -0.26 -0.01 0.18 0.27 0.15 -0.28 -0.33
-0.02 0.00 0.01 -0.01 0.17 -0.13 -0.21 -0.34 0.23 -0.04 0.08 0.18 0.01 -0. l0 -0.02 -0.12 -0.03 0.00 -0.14 -0.07 0.04 -0.03 0.62 0.12 0.29 0.07 0.79 0.25 0.54 0.39 0.42
U
0.15
0.89
Organic
1.02
-0.42
dolostone. The strong negative loading of Mg in the factor that includes P also may be linked to this process. CFA will dissolve in sulfuric acid solution, producing a weaker phosphoric acid, and this dissolution consumes some of the initial leachate acidity. Note that leachate pH also has a weak negative correlation with P. The linkage of these observations suggests a pH drop by oxidation of the dissolving sulfide minerals that is neutralized by dissolution of carbonate minerals until they are exhausted. When all carbonate
382
J.R. Herring
minerals have dissolved, phosphate minerals will react with the remaining acidity from the sulfide oxidation and release P into the leachate.
Correlation with bulk chemistry The 24-h leachate element concentrations show several relationships with the concentration of the same element in the rock samples from highly altered Sections A and C, less-altered Sections B and Z, and least-altered Section J (Fig. 13-2). Concentrations in the solids are calculated as weighted averages by including the length of the channel sample interval in the calculation of the average concentration. Sections A, C, B, and J include nearly the complete Meade Peak. As such, these weighted averages approach the composition of the entire Meade Peak section as a whole for each section. Even highly altered rocks remain reactive to water. For example, Sections A, C, and F lost about 1% of their As, Mo, Sb, Se, T1, and V in 24-h leachate solutions. The elements for which the bulk concentrations in rocks are higher in the less- and least-altered rocks are Mo, Ni, Se, and Zn (Fig. 13-2), as well as sulfate. These elements, along with Fe, Cd, and Cu, also have a greater percentage of the rock that is soluble in the leachate solution for less-altered samples. This trend is identified as a series of shorter bars as the sections progress to the right from highly altered to least altered (Fig. 13-2). Arsenic, Ba, Sb, TI, U, and V show little change of mean concentration in the solids as a function of
Fig. 13-2. Comparison between the amount of an element in the 24-h leachate (bar bottom) and its bulk content in the original solid sample (bar top); the bulk original content is a weighted mean concentration of the total element in the rock as a mass fraction. Shorter bar lengths mean more of the total mass of the element was leached. Each factor of 10 of concentration range ratio represents a factor of 10 of percentage solubility. Sections are ordered from most-altered (left) to least-altered (right).
Rock Leachate geochemistry of the Meade Peak Phosphatic Shale Member
383
alteration and their percentage solubilities between solid and leachate samples do not change with alteration.
L e a c h a t e condition variations
The important parameters of the leaching experiment are rock grinding, grain size, oxygenation, agitation, solution composition and temperature, solid/leachate ratio, and leaching time. These variables may change over the course of the leaching experiment. For example, grains may disaggregate by hydration or other reactions during early stages of the leach, which then produces consequent increase in grain-surface area. Also, solution composition can change during the course of the leach from reaction with the rock and atmosphere. Different solid/leachate ratios, which can affect leachate results (Desborough et al., 1999), were not investigated here. The parameters that were varied in this study were particle size, leachate time, anoxic leachate conditions, freeze-thaw effects, and multiple leaching.
Particle size
Some limited information on the effect of particle size on leachate concentrations is provided by analyzing leachate splits of two samples that were ground to an even finer size prior to leaching. Samples B059 and B134, phosphatic mudstone from the middle-waste rock and upper-ore zones of Section B, were chosen for their differences in bulk Se and organic-carbon concentrations in the solid sample and because of different behavior in the 24-h leachate experiments. Sample B059 had higher conductivity and Se concentration in the 24-h leachate by factors of 4 and 20, respectively, than corresponding values for sample B134. Splits of the samples that had been ground to <100 mesh were then ground to <250 mesh (0.061 mm). For spherical particles at the upper end of the size cutoff, this would increase the surface area by a factor of 15. Data for the 24-h leachates of both samples ground to < 100 mesh and those more finely ground are listed in Table 13-1V. The more finely ground sample produced greater concentrations of leachate elements for both samples. Sample B059 had increased leachate concentrations (about 20-50%) of all the geoenvironmentally significant trace elements except Cr. For B134, the percentage increases for most elements in the leachate solutions from the more finely ground sample ranged from about 50 to 100%. The concentration increases indicate that finer particle sizes produce greater leachate concentrations of the various elements in rocks of the Meade Peak.
Leachate time
Additional experiments were conducted with shorter and longer leach times than the standard 24-h leachate experiments. Varying leachate time experiments were conducted for some more-, less-, and least-altered samples (Fig. 13-3).
384
T
-I
d
d
Z
U=
.,!
o
.=_
m
m,,~
~'~
V
~-~ I~.
t.~
eq
--
V
r'~
u,-~ ~
r-,I o',
,,,~- t~l - - - - V
~
t~
e~
t-,l
e,~ e~
V
"
~
u,~ - -
m
~,"~ t"q
V
oo
oo
V
~-~ ~
oo
0',
V
~
~
V V
~t-~ o~
~o
~"-
~'V
~
I"-- - -
~
t-,i
~
oo
~
~
--
~
~
o
~
e,~ t ~
~D ~
e,'~,~l-
~
~
~
00
~-, ,_...,
,,~
oe
V
e~
t-~
o o c 5 c 5 c 5 ~ 5 - ~
t-,l
o o o o c 5 ~ 5 o
eq
t-,l
r
t-q
V
V ' - V
%
oo
~
u,-~ o
e4
V
'~1" - -
V
V
V
~'~ - -
--
--
~",
i.--. ~
t"q
V
t-,~ t~l
t~
~
e~
c 5 c 5 o c 5 o c 5 c 5
t-~
.,~-
"~"
V
t'q
V V V
V
V
V
V
,4D ~,- o o
V
--
V
V
.
t-q
.
~
,I-,
V
V
--
~
V
V
--
oo,4D
V
V
--
t-q
o~
~
V
t~l
~
o~
.
e*h
o~
.
I---
.
J.R. Herring
e,i~ o
V
,,~
V
,,~
--
"
i--- o o
t',,I
.
~
oo
V
--
.
I'--
I~
V
V
--
~
~o
I~-
o c 5 c 5 o o c 5
V
"
I~
I-~
~'~
~ol~-
~r~ LI-~ u,~ ~r'~
o0
~-~ ~
Continued
ta.a Oo TABLE 13-IV Continued Sample
pH
Conduc-
HCO31
Na
Mg
AI
Si
P
SO4
K
Ca
Ti
V
Cr
<1 <1
Mn
Fe
Co
300 479
<50 <50
20.90 26.60
11 37 6 11 4 6
641 726 596 824 682 841
2,170 498 1,050 290 596 4,320
65.10 74.80 73.90 83.10 68.00 85.60
5 7
. 860 1,030
317 515
85.80 93.20
Th
U
tivity JI68-12D J168-43D
6 6.5
281 342
8 -
1.2 1.3
2.1 2.7
39 235
6.8 11.8
0.40 0.20
101 136
1.9 2.1
42 49
1.6 2.1
27 16
JI86-1D JI86-1D F,,q" J186-2D JI86-2D F/T JI86-6D J 186C-6D Ar
5.1
4.7 4.83
717 715 682 772 757 751
151 197 178 237 127 136
0.9 0.7 1.1 1.1 0.8 1.4
2.9 2.3 3.5 3.5 2.4 4.2
3,640 4,640 3,330 5,030 3,540 4,450
2.0 2.1 2.1 <0.2 3.7 5.3
3.40 3.70 2.00 2.80 2.40 4.70
311 273 267 327 260 339
2.8 2.5 2.2 2.7 2.6 3.3
118 123 114 138 102 133
3.3 4.6 3.2 4.9 4.0 2.7
87 94 96 95 124 85
4.39 4.92 4.35
225 796 963
. 173 162
. 1.3 1.5
. 3.6 4.1
. 4,970 7,460
. 5.6 6.0
. 178 202
J 186-7D resup J186-12D JI86-43D
4.7
.
.
6.8 11.9
Sample
Ni
Cu
Zn
As
Se
Sr
AO62-1H A062-2H A062-1D A062-2D A062-6D A062-12D A062-43D
17 25 15 16 18 22 28
1 2 <0.5 <0.5 <0.5 <0.5 1
55 54 13 12 21 17 22
2 2 2 2 2 3 4
22 25 6 8 11 17 38
10 14 5 6 6 5 5
1.3 1.5 1.4 1.5 2.0 2.4 3.6
1 1 0 0 0 0 0
277 319 304 315 329 375 392 780 9 10 11 9 10 34 13
1 1 1 1 1 1 2 6 4 4 5 3 5 31 5
770 1,000 838 1,010 870 1,100 1,120 2,350 13 15 20 15 17 263 20
4 12 8 4 5 13 6 8 6 6 7 5 7 11 10
200 467 223 241 224 272 283 343 4 3 4 3 4 6 5
29 23 33 23 30 31 33 52 7 6 11 7 8 64 8
1.0 0.9 1.1 4.3 1.7 4.0 1.9 2.1 7.1 5.7 8.5 5.0 11.7 11.9 22.8
12 12 14 12 13 10 15 36 1 1 1 1 1 14 1
A080-1D A080-ID F.,q" A080-2D A080-2D F/q" A080-6D A080C-6D Ar A080-12D A080-43D AI31-1D A131-1D F..rl" AI31-2D A131-2D Frl" AI31-6D A131-12D AI31-43D
Mo
. 3.30 3.50
Cd
. 345 421 Sb 0.4 0.6 0.5 0.6 0.7 0.7 0.9
. 3.2 3.6 Ba
. 142 163 La
TI
Pb
.
37.7 26.8 0.2 0.2 0.9 0.3 0.8
1.64 2.35 0.02 0.04 0.20 0.04 0.21
0.06 0.10 0.09 0.10 0.10 0.20 0.10
0.20 0.30 <0.05 <0.05 0.05 0.07 0.08
0.33 0.35 0.10 0.08 0.09 0.09 0.13
0.44 0.66 0.01 0.02 0.06 0.02 0.05
0.8 2.0 0.7 2.1 1.1 2.1 0.7 1.5 1.5 2.7 1.6 2.1 1.6 5.4 1.8 4.0 1.1 0.3 0.7 0.7 1.4 1.3 0.7 0.3 1.7 0.4 1.9 104.0 2.7 0.6
0.02 0.01 0.02 0.01 0.02 0.02 0.04 0.22 0.02 0.05 0.13 0.02 0.03 3.56 0.05
0.50 0.40 0.50 0.30 0.40 0.40 0.40 0.66 0.60 0.40 0.70 0.55 0.72 3.30 0.83
<0.05 <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 ,0.05 <0.05 0.08 <0.05 <0.05 0.50 <0.05
0.02 0.01 0.01 0.01 0.02 <0.005 0.01 0.03 0.16 0.03 0.13 0.22 0.12 0.54 0.12
0.02 0.01 0.02 0.02 0.01 0.02 0.03 0.05 0.01 0.01 0.02 0.01 0.01 1.00 0.01
.
0~
B025-1H B025-1D B025-2D B025-6D B025-12D B025-43D
2 2 2 2 7 3
1 1 <0.5 1 4 1
8 5 5 9 252 14
4 6 6 7 9 8
2 1 2 1 2 2
17 21 22 25 60 43
3.7 5.0 4.7 5.7 4.9 8.8
1 1 1 1 54 2
0.3 0.4 0.4 0.5 0.6 0.7
0.4 0.5 0.6 1.2 26.8 3.5
<0.01 <0.01 <0.01 <0.01 12.40 <0.01
1.30 1.10 1.30 1.40 2.20 1.70
<0.05 <0.05 <0.05 0.08 0.50 <0.05
0.01 0.04 0.02 0.03 0.09 0.04
1.73 2.71 2.83 3.32 6.75 4.62
B047-1D B047-2D B047-6D B047-12D B047-43D
13 13 14 53 15
1 1 1 8 1
43 63 54 1,140 73
13 13 12 17 11
12 13 12 17 16
30 33 39 191 56
16.9 17.0 21.2 23.3 33.6
2 2 3 79 4
!.0 !. 1 1.2 1.3 1.2
0.9 1.2 1.6 60.5 3.0
<0.01 0.02 <0.01 21.8 0.12
2.00 2.30 2.30 3.70 2.30
<0.05 0.06 0.07 1.80 <0.05
0.14 0.08 0.12 0.47 0.14
0.17 0.20 0.40 28.20 1.52
B059-1D B059F-ID B059-2D B059-6D B059-12D B059-43D
8 11 9 9 26 10
1 2 1 1 9 1
5 11 9 11 216 14
11 15 14 12 15 16
144 187 159 144 160 182
41 56 48 53 95 69
27.1 38.7 31.7 35.8 41.0 61.6
1 1 1 1 14 1
1.9 2.9 2.2 2.4 2.5 3.1
1.6 2.5 2.2 2.5 29.8 5.7
<0.01 <0.01 <0.01 <0.01 9.90 0.01
0.30 0.40 0.30 0.30 0.50 0.30
<0.05 <0.05 <0.05 <0.05 0.94 0.09
0.08 0.05 0.05 0.06 0.10 0.08
1.61 1.93 2.05 1.84 4.62 2.79
B 134-1D B134F-ID B134-2D B134-6D B134C-6D Ar B 134-7D resup
7 63 7 9 10
1 49 1 2 3
24 548 23 38 13
4 15 5 6 13
7 37 8 9 35
5 144 5 7 15
4.8 13.0 5.8 7.4 20.0
1 35 1 I I
0.4 1.8 0.5 0.6 1.7
0.6 43.9 0.5 1.0 1.4
0.03 45.8 0.01 0.15 0.13
1.20 3.10 1.30 1.10 1.40
<0.05 4.80 <0.05 0.10 0.10
0.09 2.13 0.06 0.10 0.02
0.03 10.80 0.03 0.06 0.04
BI34-12D BI34-43D
40 9
16 2
446 41
8 9
9 10
40 7
8.0 14.2
27 1
0.7 0.9
122.0 1.0
4.74 0.08
2.90 1.10
1.50 0.05
0.15 0.10
2.33 0.03
4 5 4 5 4
12 12 10 12 13
20 10 3 11 9
i 2 2 2 2
3 4 4 4 5
4 6 7 10 8
1.2 1.6 1.5 2.2 1. !
0 0 0 1 1
0.9 1.5 1.6 2.0 2.0
0.2 0.3 0.2 0.8 0.3
0.08 0.09 0.04 0.62 0.28
0.20 0.30 0.20 0.20 0.20
0.10 <0.05 <0.05 0.20 <0.05
0.08 0.05 0.10 0.22 0.02
0.26 0.21 0.20 0.44 0.28
27 6
133 13
178 6
4 3
8 5
113 10
2.3 3.3
31 1
2.3 2.6
136.0 0.7
15.4 0.52
0.77 0.20
1.80 0.20
0.87 0.25
18.80 0.36
J084-1H J084-2H J084-1D J084-1D F/q" J084-2D J084-2D F q J084-6D J084C-6D Ar J084-7D resup
530 712 223 603 268 572 212 214 .
1 2 1 3 1 2 1 1
900 2,040 982 3,450 1,360 3,180 1,150 1,260
1 2 2 11 11 11 2 13
319 503 377 624 499 858 513 574
259 406 231 406 377 422 480 463 .
234.0 471.0 355.0 627.0 519.0 740.0 884.0 889.0 .
5 10 4 9 6 7 7 7
4.7 9.6 5.9 6.9 9.5 6.6 12.8 11.3
18.1 14.6 4.4 8.2 7.1 7.3 10.0 9.6
0.03 0.03 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 .
<0.05 <0.05 <0.05 0.06 0.06 0.09 0.05 0.09 .
0.05 0.09 <0.05 <0.05 <0.05 <0.05 <0.05 <0.05
0.22 0.20 0.07 0.11 0.08 0.36 0.12 0.02
1.78 3.75 2.18 4.76 4.67 5.67 7.33 6.02
J084-12D J084-43D
224 175
458 583
889.0 1,340.0
0.12 <0.01
0.06 0.06
<0.05 <0.05
0.05 0.17
7.97 10.00
F 121 - 1H FI21-1D FI21-2D F 121-6D F 121C-6D Ar F 12 I-7D resup FI21-12D FI21-43D
.
. 2 2
. 1,350 1,300
. 2 14
. 603 734
.
. 7 9
. 11.4 13.9
. 10.8 12.4
.
Continued
ta~ OO OO
TABLE 13-IV Continued Sample JI25-1H JI25-2H J125-1D JI25-2D Jl25-6D J 125C-6Dar JI25-12D J 125-43 D JI68-1H J168-1D J 168-2D J 168-6D JI68-12D J 168-43D J186-1D J 186- I D F/l" JI86-2D J186-2D F/T J186-6D J186C-6Dar J 186-7Dresup JI86-12D J 186-43D
Ni
Cu
Zn
As
Se
Sr
Mo
Cd
Ba
La
3,870 4,470 3,970 3,390 3,500 4,370 3,220 2,300
427 369 249 46 3 1 3 2
23,700 27,500 24,900 19,900 18,300 24,700 14.500 6,640
<1 < 1 <1 2 <1 2 <1 4
41 59 67 73 91 76 128 176
321 387 380 339 339 323 298 290
4.2 2.0 2.0 2.4 8.9 22.4 31.3 106.0
57 68 58 46 39 33 29 16
0.5 0.7 0.7 0.6 1.0 2.1 1.5 3.6
46.6 29.6 11.2 4.5 2.3 1.6 1.9 0.6
37.6 39.3 29.7 13.1 0.85 1.44 0.10 0.02
1.10 1.30 1.10 0.90 0.80 0.40 0.67 0.63
0.10 0.06 <0.05 <0.05 <0.05 <0.05 <0.05 0.07
0.18 0.23 0.18 0.13 0.13 0.09 0.05 0.15
2.78 1.88 0.97 0.18 0.06 0.05 0.05 0.11
2,140 1.640 1,290 1,090 1,180 2,500
4 4 4 4 4 2
14 16 18 20 25 32
172 180 163 173 175 216
15.5 34.5 38.3 61.8 76.6 92.0
3 3 3 3 4 10
0.4 0.8 1.0 1.5 1.7 2.4
5.6 3.8 3.7 3.9 4.9 6.6
0.02 <0.01 <0.01 <0.01 <0.01 0.02
0.75 0.64 0.70 0.70 0.69 0.90
<0.05 <0.05 <0.05 0.06 0.10 0.07
0.03 0.08 0.05 0.06 0.03 0.07
0.04 0.02 0.02 0.02 0.02 0.03
7,010 9,590 6,930 10,300 7,970 9,810
8 25 18 14 11 35
485 888 527 686 586 696
547 551 583 586 641 645
26.5 19.2 24.8 17.6 38.1 45.9
214 210 213 242 287 263
1.3 0.9 1.4 0.8 1.8 2.0 . 1.8 1.9
17.1 13.0 13.2 12.7 17.9 15.2 . 17.0 23.0
1.87 1.65 1.91 1.79 1.59 1.46
3.10 2.20 3.10 2.10 3.50 3.20
<0.05 <0.05 <0.05 <0.05 <0.05 <0.05
0.28 0.51 0.22 1.41 0.27 0.10
0.84 0.57 0.72 0.52 0.85 0.58
<0.05 0.10
0.12 0.29
0.75 1.39
727 754 670 696 719 936 3,430 4,220 3,400 4,700 3,670 4,320 . 4,100 4,790
<0.5 <0.5 <0.5 ! 1 2 113 123 92 135 115 45 .
. 123 233
. 9,680 13,100
. 15 42
879 1,260
.
. 649 883
. 44.6 46.3
Sb
. 323 459
.
Ti
. 1.79 4.43
Pb
Th
U
. 3.50 4.50
1Estimated, see Methods Section. SufFix codes refer to sample numbers: E fine-grain (<250 mesh): H. hour: D. day (6D = six days): Ar, argon atmosphere; F/T/freeze-haw; resup, multiple leaches.
.M
o~
t~
~..,~
Fig. 13-3. Time-series leachate concentrations of various elements and mineralogical analyses for highly altered (A), less-altered (B), and least-altered (C) rock samples. Element legend is the same for all parts. Leachate concentration of Fe, when less than the LDL of 50 I~g L-1, has been replaced with the LDL concentration so that the significant changes with time can be followed.
ta~
OO
Fig.
13-3B.
Continued ~,,,.,,~
Rock Leachate geochemistry of the Meade Peak Phosphatic Shale Member
~5
~ ,,,,~
391
392
JR. Herring
Meade Peak rocks react quickly with water. Even highly altered rock samples can release 0.01-100 ~g L -! of many trace elements in times as short as 1 h. Most elements show concentration increases with increasing leachate time, although concentrations of a few elements decreased with extended time, for example, with Fe in sample A062, continued exposure to oxidizing conditions allowed the dissolved Fe to precipitate. This phenomenon has been noted in other leachate studies (Desborough et al., 1999) and can be a source of increasing acidity due to hydrolysis and precipitation of the Fe. In turn, the increased acidity can contribute to further leaching of the solids and release of additional trace elements. Leachate samples that contained sulfide minerals showed little change of the concentration of Fe with time if there was sufficient carbonate mineral present to neutralize the acidity, as in sample J084. Highly altered samples with residual sulfide minerals but little or no carbonate, such as A062, exhibit a decrease in Fe concentration in the leachate with time as the abundant CFA neutralizes excess acidity from sulfide oxidation. The highly altered samples had stable pH and conductivity values over the time of the extended leachate experiments. This is consistent with their having previously lost much of their leachable elements through alteration.
Highly altered rocLs"
Sample A062 showed low conductivity through the duration of the experiment. The initial concentration of sulfate (5 mg L-I) was due to gypsum dissolution, which changed the solution conductivity insignificantly. The initially high Fe content was due to oxidation of the highly reactive minor amount of framboidal pyrite that depressed pH slightly but still resulted in precipitation of the Fe as oxide or oxyhydroxide. This initially depressed pH dissolved minor calcite to raise pH to about 7, 12 days after immersion. However, no dolomite and only a minor amount of calcite were available to counter acidity produced by the pyrite and, as the pyrite continued to slowly react, the carbonate was consumed and the pH again lowered to 6 by day 43. Among the dissolved elements, only Fe and AI had order of magnitude concentration changes, first markedly decreasing due to oxidation and precipitation then slowly increasing as pH dropped. Sample A080 showed pH values between 5 and 6 until day 12, which had only minor effects on dissolved concentrations of elements, sulfate, and conductivity. Between day 12 and day 43, some hydrolysis or other reaction (such as oxidation of undetected pyrite or sphalerite) depressed pH to below 5, thereby increasing the solubility of Zn, Al, Ni, Fe, and sulfate. Sample F 121 had similar results as A080. Sample A131 showed conductivity values that began and remained low during the experiment. The consistently low concentration of sulfate suggests that bicarbonate or orthophosphate were balancing anions for the dissolved metals. The evidence for possible orthophosphate is the concentration of dissolved P (Table 13-I), which ranged between 1 and 7 mg L-i. Concentrations of Fe and A1 also had order of magnitude decreases then increases as in sample A062.
Rock Leachate geochemistry of the Meade Peak Phosphatic Shale Member
393
Less-altered rocks
Relatively low conductivity values for sample B025 suggest that it behaved similarly to the least-altered samples (Fig. 13-3B). Note the marked increase of Zn at day 12 at a time that corresponded to the highest concentration of dissolved P, 15 mg L-i. This concentration of Zn may represent the near complete dissolution of the minor amount of sphalerite in the sample. The large values of dissolved P indicate the importance of CFA as an acidneutralizing species and that it provides significant quantities of a complex-forming anion. Consistently high values of pH prevented A1 or Fe concentrations from becoming large. Sample B047 behaved like sample B015. Note the consistent presence of large concentrations of V and the marked increase in the concentration of Zn at day 12. Sample B059 had a very high acid-neutralizing capacity due to the large amount of calcite, and the absence of detectable Fe sulfide. The sulfate concentration and associated conductivity values are higher than those for the least-altered rocks, and they increase only slightly during the 43-day exposure. The elevated Zn concentration at day 12 could be due to association with a carbonate complex. Selenium concentrations remained nearly constant.
Least-altered rocks
Although sample J084 has almost 5% pyrite, the significant amount of calcite and dolomite prohibited pH values from dropping below about 6.5. The minor changes in sulfate concentrations and associated conductivity reflect the lack of oxidation of pyrite, which is indicated by the high and stable pH (Fig. 8-3C). Sample J125 showed a general correlation among conductivity, sulfate, and pH. The sulfate concentrations and conductivity showed small increases over time, and the sulfate maintained a nearly constant ratio to conductivity, which implies that pyrite continually dissolved. However, the lack of calcite initially allowed the pyrite to oxidize and generate significant acidity so that the pH dropped to about 4.5, which caused much Fe, A1, and Zn to be leached until dolomite started slowly dissolving. As dolomite dissolved, the pH increased to slightly above 6 at day 12 which resulted in precipitation of virtually all Al and most Fe. Some Zn and Ni remained in solution as the pH rose to near 7 at day 43, probably due to their solubility related to carbonate in solution. The high initial value of dissolved Zn may have been associated with reaction of organic matter. Selenium and Mo steadily increased over time. Sample J168 consists of quartz and CFA, with no carbonate minerals. Reactivity was dominated by gypsum and pyrite. Sulfate concentrations, conductivity, and pH were apparently controlled by initial pyrite oxidation causing pH values of around 5.7. It is speculated that some CFA or some undetected minor amount of calcite or dolomite dissolved, thereby raising the pH to about 6.5 and causing precipitation of Fe and AI. The increase of A1 from 40 to > 100 ~zg L-1 between day 12 and day 43 is unexplained. As in the previous sample and for the least-altered rocks in general, Se and Mo gradually increased with time.
394
J.R. Herring
Sample J186 is a CFA- and quartz-rich rock with very few minerals detected by XRD that could be reactants in water except for gypsum - which does not alter pH when it dissolves. Besides CFA, there are no other acid-neutralizing minerals. Dissolution of the relatively abundant CFA, which is marked by concentrations of dissolved P that range from 2 to about 4 mg L-~, is insufficient to neutralize the acidity in this sample. The initial pH of about 5.2 resulted in relatively high Fe, AI, Zn, and Ni concentrations in the range of 2-7 mg L -~. However, it appears that oxidation and associated precipitation of Fe as Fe-oxide and later Fe-oxyhydroxide resulted in a lowering of final pH to about 4.3. This lowered pH allowed A1, Zn, and Ni concentrations to increase modestly. Modest changes in sulfate and conductivity are consistent with these interpretations. Selenium gradually increased to a final concentration of just over 1 mg L -~. This amount of Se represents about 5% of the Se that was originally present in the rock.
Anoxic conditions
It is not known whether an anoxic environment exists in waste piles of Meade Peak rocks. However, reduced chemical species have been found in water draining from waste rock disposal piles that suggest the presence of anaerobic conditions inside the piles (Stillings and Amacher, Chapter 17). Anoxic conditions were examined by using an Ar atmosphere to enclose a subset of samples during leaching. Prior to mixing with the ground rock, the deionized leachate water was degassed with bubbling Ar. Mixing of the water and rock was done in a glove bag inflated with Ar and the samples were maintained in the bag with a positive pressure of flowing Ar during the 6-day duration of the leachate experiment. One of the highly altered samples, A080, increased the leachate concentrations of Fe in the absence of oxidizing conditions (Table 13-1V), but the other highly altered sample, F 121, did not. There was little difference in conductivity or pH between the 6-day leachate under oxidizing or reducing conditions for these two samples. In sample A080, Fe, Zn, and As were elevated in leachate concentration in the reducing compared to the oxidizing environment. Selenium was slightly elevated, by 21%, in the reducing leachate, while Cd and Th decreased in concentration. The sample from highly altered Section F showed little difference in concentrations of most elements between reducing and oxidizing environment leachates. Presumably, the intense alteration to which Section F had been exposed had removed most of the soluble fractions of many of the trace elements. In the comparatively less-altered sample B134, conductivity, sulfate, Si, and P increased in the reducing leachate. However, most of the trace elements in the reducing leachate, with the exception of V, either showed no increase or had a slight decrease in concentration. In the least altered samples, Section J samples J084, J125, and J186, there was little change in pH or conductivity between the reducing and oxidizing environment. Sulfate showed slight to nearly two-fold increases in the reducing leachate solutions. The dissolution of all major elements was enhanced in the reducing leachate. Sample J084 showed little change in the concentration of Fe, reflecting its relatively low abundance in
Rock Leachate geochemistry of the Meade Peak Phosphatic Shale Member
395
the bulk rock, but the other two samples showed dramatic increases in Fe in the anoxic leachates. In the time-series study, sample J125 had initial Fe concentrations in the leachate around 40,000 Ixg L -l at 1 and 2 h reaction time. This decreased to around 1500 txg L-l by 6 days and to below the LDL, 50 txg L-1 by 12 days. However, the 6-day leachate sample kept in Ar maintained concentrations near 40,000 txg L-1. Sample J 186 had a concentration of Fe in the 6-day leachate sample kept in Ar that was nearly double that of the high concentration for the oxidizing leachate samples. Clearly, the lack of oxidizing conditions allowed the Fe and trace elements that are associated with it (Co, Ni, and Zn) to remain in solution for longer times. Selenium, As, Mo, Cd, and Sb showed little change in concentration between reducing and oxidizing leachates. Thorium and U concentrations decreased in the anoxic leachate.
Freeze-thaw effects
The area of phosphate mining in southeast Idaho has cold winters with freezing conditions. Consequently, the effect of freeze-thaw conditions during leaching were studied by quickly freezing the leachate and ground-rock samples in a bath of dry ice and acetone. These rock samples were splits of four of the 1 and 2 day leachate samples and were processed exactly the same up until freezing. The samples were kept in the bath until frozen, about 20 rain, then allowed to defrost at room temperature. The freeze cycle was repeated twice each day. In general, this process produced little effect on pH or conductivity of the highly altered samples A080 and A I31 (Table 13-1V). For sample A080, Zn is the only trace element that showed an increase after the freeze-thaw process in both the 1- and 2-day leachate samples. Curiously, sample A I31 has a higher Zn content in the original bulk rock by nearly a factor of four but produces a leachate Zn concentration of only about 1-2% of that of sample A080. One explanation for this is the relative difference in organic-carbon content of the original bulk rock. Samples A080 and A I31 have 8 and 1.7% organic carbon, respectively. Sample A131 has more Zn, but it must be in mineral phases that are resistant to dissolution. The bulk-rock Zn content in A080 is less than that of A 131, but much of it must occur in the high organic-carbon fraction and it is this phase that was more easily leached and mechanically disrupted by the freeze-thaw process to accelerate dissolution. In the two less-altered samples, J084 and J186, the freeze-thaw process produced slightly enhanced dissolution as indicated by increased conductivity. Sulfate generally increased in the freeze-thaw samples compared to the same leachate samples that were not frozen. Chromium and Mn concentrations increased in the freeze-thaw leachates of both samples, but Fe remained below its LDL in one sample and significantly decreased in the other. Molybdenum, Ni, Se, and Zn increased in the freeze-thaw leachates of both samples. Molybdenum increased in the freeze-thaw leachate of sample J084, but decreased in the other sample. This was the sample of the pair with much greater content of organic carbon in the bulk rock. Molybdenum is partially associated with organic matter in rocks of the Meade Peak.
396
J.R. Herring
Multiple leaching The effect of resuspension of a previously leached rock sample was studied in samples F 121, J084, and J 186 (Table 13-IV). The sediment from the 1-day leachate was resupended at a water/rock ratio of 20 with deionized water and allowed to react for an additional 6 days. As expected, easily leached elements were greatly reduced because of the initial leaching. Sulfate and conductivity in the two resuspended samples of day 6 from Section J were reduced to about 25% of that in the initial leachate of day 6. Conductivity in the sample from Section F was little reduced in the resuspended sample, but all leachate samples from this highly altered section had very low conductivities, typically less than 25 IxS cm -l. Nonetheless, multiple leachates of the same rock sample, especially if minerals such as pyrite are present, can continue to release trace elements for many cycles of leaching (Desborough et al., 1999).
CONCLUSIONS These leachate experiments show that Meade Peak rocks are highly reactive with water and that they can release significant quantities of several trace elements in periods as short as 1 h. Furthermore, the initial release of potential contaminant trace elements into solution is dominated by those rocks of the Meade Peak that are least altered and rich in organic matter.
ACKNOWLEDGMENTS I thank the phosphate mining companies for providing access for rock sampling. I appreciate help in sample preparation by M. Fallin, K. Long, and C. Santos. P.L. Hageman, G.A. Desborough, J.R. Hein, and L.B. Kirk provided comments on the manuscript.
REFERENCES d'Angelo, W.M. and Ficklin, W.H., 1996. Fluoride, chloride, nitrate, and sulfate in aqueous solution by chemically suppressed ion chromatography. In: B.F. Arbogast (ed.), Analytical methods manual for the Mineral Resource Surveys Program, US Geological Survey. US Geological Survey, Open-File Report, 96-525, 248 pp. Desborough, G.A., Leinz, R., Smith, K., Hageman, P.L., Briggs, P.H., Fey, D. and Nash, T., 1999. Acid generation and metal mobility of some metal-mining related wastes in Colorado. US Geological Survey, Open File Report, 99-332, 18 pp. Hageman, P.L., Briggs, P.H., Desborough, G.A., Lamothe, P.J. and Theodorakos, P.J., 2000a. Update and revisions for Open-File Report 98-624, Synthetic precipitation leaching procedure (SPLP) leachate chemistry data for solid mine-waste composite samples from the Silverton and Leadville Districts in Colorado. US Geological Survey, Open File Report, 00-150, 16 pp.
Rock Leachate geochemistry of the Meade Peak Phosphatic Shale Member
397
Hageman, EL., Desborough, G.A., Lamothe, EJ. and Theodorakos, P.J. 2000b. Synthetic precipitation leaching procedure (SPLP) leachate chemistry data for solid mine-waste composite samples from southwestern New Mexico, and Leadville, Colorado. US Geological Survey, Open File Report, 00-033, 18 pp. Hageman, EL. and Briggs, P.H., 2000. A simple field leach test for rapid screening and qualitative characterization of mine waste dump material on abandoned mine lands, in ICARD 2000: Proceedings from the Fifth International Conference on Acid Rock Drainage, Denver, Colorado, May 21-24, 2000. Society for Mining, Metallurgy, and Exploration, Inc., pp. 1463-1475. Herring, J.R., Desborough, G.A., Wilson, S.A., Tysdal, R.G., Grauch, R.I. and Gunter, M.E., 1999. Chemical composition of weathered and unweathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation - A. Measured sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 99-147-A, 24 pp. Herring, J.R., Wilson, S.A., Stillings, L.A., Knudsen, A.C., Gunter, M.E., Tysdal, R.G., Grauch, R.I., Desborough, G.A. and Zielinski, R.A., 2000a. Chemical composition of weathered and less weathered strata of the Meade Phosphatic Shale Member of the Permian Phosphoria FormationB. Measured sections C and D, Dry Valley, Caribou County, Idaho. US Geological Survey, Open File Report, 99-147-B, 34 pp. Herring, J.R., Grauch, R.I., Desborough, G.A., Wilson, S.A. and Tysdal, R.G., 2000b. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation - C. Measured sections E and E Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 99-147-C, 35 pp. Herring, J.R., Grauch, R.I., Tysdal, R.G., Wilson, S.A. and Desborough, G.A., 2000c. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation- D. Measured sections G and H, Sage Creek area of the Webster Range, Caribou County, Idaho. US Geological Survey, Open File Report, 99-147-D, 38 pp. Herring, J.R., Grauch, R.I., Siems, D.E, Tysdal, R.G., Johnson, E.A., Zielinski, R.A., Desborough, G.A., Knudsen, A. and Gunter, M.E., 2001. Chemical composition of strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation. Channel-composited and individual rock samples of Measured Section J and their relationship to Measured Sections A and B, Central Part of Rasmussen Ridge, Caribou County, Idaho, US Geological Survey, Open File Report, 01-195, 72 pp. Lamothe, P.J., Meier, A.L. and Wilson, S., 1999. The determination of forty four elements in aqueous samples by inductively coupled plasma-mass spectrometry. US Geological Survey, Open File Report, 99-151, 14 pp. Smith, K.S., Walton-Day, K. and Ranville, J.E, 2000. Evaluating the Effects of Fluvial Tailings Deposits on Water Quality in the Upper Arkansas River Basin, Colorado: Observational Scale Considerations, in ICARD 2000: Proceedings from the Fifth International Conference on Acid Rock Drainage, Denver, CO, May 21-24, 2000. Society for Mining, Metallurgy, and Exploration, Inc., pp. 1415-1424. US Environmental Protection Agency, 1994. Test method for evaluating solid waste, physical/chemical methods (SW-846), 3rd edn., update 2B. Environmental Protection Agency, National Center for Environmental Publications, Cincinnati, OH 45268, order #EPASW-846.3.2B. Accessible at URL: http://www.epa.gov/epaoswer/hazwaste/test/sw846.htm.
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
399
Chapter 14
REX CHERT M E M B E R OF THE PERMIAN P H O S P H O R I A FORMATION: COMPOSITION, WITH EMPHASIS ON E L E M E N T S OF ENVIRONMENTAL CONCERN
J.R. HEIN, B.R. MclNTYRE, R.B. PERKINS, D.Z. PIPER and J.G. EVANS
ABSTRACT We present bulk chemical and mineralogical compositions, as well as petrographic and outcrop descriptions, of rocks collected from three measured outcrop sections of the Rex Chert Member of the Phosphoria Formation in southeast Idaho. The three measured sections were chosen from l0 outcrops of Rex Chert that were described in the field. The Rex Chert overlies the Meade Peak Phosphatic Shale Member of the Phosphoria Formation, the source of phosphate ore in the region. Rex Chert removed as overburden constitutes part of the material transferred to waste-rock piles during phosphate mining. It is also used to surface roads in the mining district. It has been proposed that the chert be used to cap and isolate waste piles, thereby inhibiting the leaching of potentially toxic elements into the environment. The rock samples studied here are from individual chert beds representative of each stratigraphic section sampled. The Cherty Shale Member of the Phosphoria Formation that overlies the Rex Chert in measured section 1 and the upper Meade Peak and the transition zone to the Rex Chert in section 7 were also described and sampled. The cherts are predominantly spiculite composed of granular and mosaic quartz, and sponge spicules, with various but minor amounts of other fossils and detrital grains. The Cherty Shale Member and transition rocks between the Meade Peak and Rex Chert are siliceous siltstones and argillaceous cherts with ghosts of sponge spicules and somewhat more detrital grains than the chert. The dominant mineral is quartz. Carbonate beds are rare in each section and are composed predominantly of calcite and dolomite in addition to quartz. Feldspar, mica, clay minerals, calcite, dolomite, and carbonate fluorapatite are minor to trace minerals in the chert. The concentration of SiO2 in the chert averages 94.6 wt.%. Organic-carbon content is generally very low, but can be as much as 1.8% in Cherty Shale Member samples and as much as 3.3% in samples from the transition between the Meade Peak and Rex Chert. Likewise, phosphate (P205) is generally low in the chert, but can be as much as 3.1% in individual chert beds. Selenium concentrations in Rex Chert and Cherty Shale Member samples vary from <0.2 to 138 ppm, with a mean concentration of 7.0 ppm. This mean Se content is heavily dependent on two values of 101 and 138 ppm for siliceous siltstone from
400
J.R. Hein et al.
the lower part of the Rex Chert, which contains rocks that are transitional in character between the Meade Peak and Rex Chert Members. Without those two samples, the mean Se concentration is < 1.0 ppm. Other elements of environmental interest, As, Cr, V, Zn, Hg, and Cd, generally occur in concentrations near or below that in average continental shale. Stratigraphic changes, equivalent to temporal changes in the depositional basin, in chemical composition of rocks are notable either as uniform changes through the sections or as distinct differences in the mean composition of rocks that comprise the upper and lower halves of the sections. Q-mode factors are interpreted to represent the following rock and mineral components: chert-silica component consisting of Si (_+Ba); phosphorite-carbonate fluorapatite component composed of P, Ca, As, Y, V, Cr, Sr, and La (+_ Fe, Zn, Cu, Ni, Li, Se, Nd, Hg); shale component composed of A1, Na, Zr, K, Ba, Li, and organic C (_+ Ti, Mg, Se, Ni, Fe, Sr, V, Mn, Zn); carbonate component (dolomite, calcite, silicified carbonates) composed of carbonate C, Mg, Ca, and Si (_+Mn); and, tentatively, organic matter-hosted elements (and/or sulfide-sulfate phases) composed of Cu (_+ organic C, Zn, Mn, Si, Ni, Hg, Li). Selenium shows a dominant association with organic matter and to lesser degrees associations with other shale components and carbonate fluorapatite. Consideration of larger numbers of factors in Q-mode analysis indicates that native Se (a factor containing Se ( + Ba)) may also comprise a minor component of the Se complement. Comparison of our data with those from newly exposed outcrops in active phosphate mines indicates that weathering of typical Rex Chert outcrops likely plays an important role in removing environmentally sensitive elements.
INTRODUCTION The Rex Chert Member conformably overlies the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation. It comprises part of the overburden that is removed to reach the phosphate ore at mines in SE Idaho and therefore comprises a part of the waste-rock dumps. In addition, the Rex Chert is used to surface roads in the mining district. It has been proposed by land-use managers that the chert be used to cap and isolate waste-rock dumps to prevent the release of selenium (Se) and other potentially toxic elements to the environment. Critical to this issue is that previous spot analyses of the Rex Chert indicated that Se might occur in high concentrations in some chert beds. Further, seleniferous cherts occur in other areas; for example, the Lower Permian carbonaceous cherts of Hubei Province in China average about 0.16% Se (Yao and Gao, 2002). Consequently, we sampled several outcrop sections to determine the composition of the Rex Chert. We did not sample all beds in each outcrop, but each bed analyzed was sampled through its entire thickness. One outcrop section sampled also includes the overlying Cherty Shale Member of the Phosphoria Formation and another section includes the upper
Composition of Rex Chert Member
401
part of the Meade Peak Phosphatic Shale Member and the transition zone between the Meade Peak and Rex Chert Members.
PREVIOUS STUDIES The Rex Chert Member was named after Rex Peak in the Crawford Mountains in Utah (Richards and Mansfield, 1912; see discussion by Hein et al., Chapter 2). Mansfield (1927) provided the first broad discussion of the Rex Chert and characterized it as having regular bedding, fine-grained texture, uniform composition, paucity of fossils (except abundant sponge spicules in one bed), and a great thickness and areal extent. He concluded that the Rex Chert was deposited as a gelatinous mass of inorganic silica in shallow water and that the bedding resulted from minor climatic oscillations that disrupted the deposition of this vast layer of silica gel. The silica was ultimately derived from river input. Keller (1941) provided excellent petrographic descriptions of the Rex Chert and showed that the chert contains various amounts of carbonate (calcite, dolomite), carbonate fluorapatite (CFA), detrital minerals (quartz, mica), carbonaceous matter, and in places sponge spicules. The proportion of each phase depended on the paleogeographic setting in which the siliceous sediment accumulated. He concluded that the bedded chert formed from precipitation of silica gel from seawater and that chert nodules in carbonates formed by diagenetic replacement. Sheldon (1957) suggested that cherts composed of sponge spicules are the result of winnowing out of fine-grained clastic material on the seafloor. He found that pyrite and glauconite occur in some chert beds and also concluded that much of the silica in the cherts migrated during diagenesis. He found that sponge spicules commonly dissolved and the silica was re-deposited in the chert; cherts where spicules were not present showed textural evidence for diagenetic recrystallization. McKelvey et al. (1959) correctly surmised that the source of silica for the Rex Chert was biosilica that dissolved during diagenesis and only the larger, more robust spicules remained as fossils. They further recognized that chert, along with black shale and phosphorite, are upwelling indicators and that nutrient-rich upwelled waters were important in supplying silica for the silica-secreting organisms. Cressman and Swanson (1964) showed that the Rex Chert likely formed by a relatively early diagenetic process involving the transformation of biosilica (opal) to quartz, possibly through an intermediate cristobalite or tridymite stage. This mineral transformation in which biogenic opal (opal-A) transforms to opal-CT, which then transforms to quartz with increasing temperature and time is now well established (e.g. Murata et al., 1977). The Rex Chert is Wordian in age (Guadalupian; Wardlaw and Collinson, 1984; see Hein, Chapter 1) and was part of a global episode of expanded accumulation of siliceous sediments that lasted for about 10 Ma (Murchey and Jones, 1992; Murchey, Chapter 5). The phosphorite and chert of the Phosphoria Formation were deposited in a zone of upwelling (McKelvey et al., 1959), which varied in intensity both temporally and spatially. We suggest that the phosphorites were deposited under a more intense upwelling regime, which produced suboxic bottom-water conditions (Perkins and Piper, Chapter 4), whereas
J.R. Hein et al.
402
the sponges that were the source of silica for the chert thrived under conditions of moderate upwelling that allowed the bottom waters to remain oxic.
METHODS
Field s a m p l i n g This region of southeast Idaho has supported extensive phosphate mining over the past several decades and currently has four active mines. Ten sections of Rex Chert located in the vicinity of Dry Valley, northeast of Soda Springs in SE Idaho (Fig. 14-1 and Table 14-I), were described in the field during June 2001. Three of the sections (1, 5, 7) were measured and sampled in detail (Fig. 14-2) for chemical, mineralogical, and petrographic analyses. The samples within the measured sections represented the entire thickness of individual beds, where possible. Where beds were too thick to collect in their entirety, portions of beds were collected. This approach provided an opportunity to determine the changes in composition of a single rock type through the history of deposition of the stratigraphic sections. The beds sampled were intended to be representative of each section. In addition, unusual rock types were sampled, for example a large dolomitic body in the Cherty Shale Member. About 0.1-1 kg of rock was collected from each sampled bed.
TABLE 14-I GPS coordinates and elevations; both accurate within <15 m (49 ft) Section number 1 2 3 4 51 52 7 8 9 10 11
Latitude (N)
Longitude (W)
Elevation (m)
42 ~ 42.16' 42 ~ 41.78' 42 ~ 42.82' 42 ~ 42.17' 42 ~ 39.38' 42 ~ 39.39' 42 ~ 43.91' 42 ~ 42.15 '3 42 ~ 44.85' 42 ~ 37.55' 42 ~ 36.89'
111 ~ 29.07' 111 ~ 24.61' 111 ~ 21.93' 111 ~ 22.04' 111 ~ 19.76' 111 ~ 19.42' 111 ~ 17.37' 111 ~ 17.07 '3 111 ~ 17.79' 111 ~ 21.13' 111 ~ 20.25'
2122 1967 1974 1961 2042 2042 2248 25603 2100 2314 2134
i Southwest end. 2 Northwest end. 3 Taken from topographic map.
c~
Fig. 14-1. Index map of Southeast Idaho and location of sections of Rex Chert (1-5, 7-11); sections 1, 5, and 7 were measured and sampled.
404
J R . Hein et al.
Fig. 14-2. Measured and sampled sections of Rex Chert: units H I-H3 in section 1 belong to the overlying Cherty Shale Member; units A-D in section 7 belong to the underlying Meade Peak Member; location of samples indicated to right of columns.
Composition of Rex Chert Member
405
Rock sample preparation A representative slab was cut through the entire thickness of each sample, crushed in a mechanical-jaw crusher, and then powdered in a roller mill. A riffle splitter was used to ensure homogenous subsamples. An aliquot of the powdered material was analyzed by X-ray diffraction for mineral content. A second aliquot of 15-50 g was sent to a contract laboratory for chemical analyses. The remaining powders for all samples are archived at the USGS. A second slab of each chert-bed sample was cut into one or more thin sections for petrographic analysis.
Geochemical analyses Samples were analyzed for 40 major, minor, and trace elements (ICP-40) using acid digestion in conjunction with inductively coupled plasma-atomic emission spectrometry (ICP-AES; Jackson et al., 1987). Also, Sr and Ba concentrations were determined by both techniques and the two data sets are closely comparable (R2= 0.999 for Sr and 0.997 for Ba). The ICP-40 technique measures concentrations of the following elements above the indicated detection limits: Au, 8 ppm; Bi 50 ppm; Sn, 50 ppm; Ta, 40 ppm; and U, 100 ppm; however, no sample had a concentration above these quantification limits and those elements are not reported. Another split of each sample was fused with lithium metaborate and then analyzed by ICP-AES after acid dissolution of the fusion mixture. This technique, ICP-16, provides analysis of all major elements, including Si, and a few minor and trace elements. Accuracy of Si determinations is probably about 2-4% based on the total-oxide sums. Titanium and Cr were determined using both ICP techniques, but only data from the ICP-16 technique are reported because the fusion technique more completely digests resistant minerals that might contain those elements. Selenium, As, Sb, and T1 concentrations were determined using hydride generation followed by atomic absorption (AA) spectrometry. The hydride analytical technique is considered to be more sensitive than the acid digestion ICP-AES technique and is the source of the data reported here. Mercury was determined by cold-vapor AA spectrometry. Total S and total C were measured using combustion in oxygen followed by infrared measurement of the evolved gas. For other forms of carbon, carbonate carbon was measured as evolved CO2 after acidification and organic carbon was calculated as the difference between total and carbonate carbon. The compilations by Arbogast (1996) and Baedecker (1987) include additional discussions about the analytical methods used here.
Statistical analyses The concentration of each element is reported as received from the analysts. However, qualified data (detection limit values) were modified for use in statistical analyses. An
406
JR. Hein et al.
element was not used in statistical analyses if more than 30% of the data points for that particular element were qualified. If there were fewer than 30% qualified values for an element, then the qualified values were multiplied by 0.5 and data for that element were used in the statistical analyses. Statistical analyses were performed on the combined data from all three sections (except for the Meade Peak in section 7 and the two carbonate beds in section 1) and separately for the individual data sets from sections 1 and 7. The Pearson product-moment correlation coefficient was used to calculate correlation coefficient matrices. A 99% confidence level was used to calculate the zero-point of correlation. For Q-mode factor analysis, each variable percentage was scaled to the percent of the maximum value before the values were row-normalized and cosine-theta coefficients calculated. Factors were derived from orthogonal rotations of principal-component eigenvectors using the Varimax method (Klovan and Imbrie, 1971). All communalities are -> 0.90.
Mineralogical analysis Mineral compositions were determined by X-ray diffraction using a Philips diffractometer with a graphite monochromator and Cu ket radiation. Samples were run from 4-70 ~ 2~) at 40 kV, 45 mA, and 10 counts per second. Semiquantitative mineral contents were determined and are grouped in Table 14-11 under the classifications of major (>25%), moderate (5-25%), and minor (<5%).
RESULTS
Lithostratigraphy The Phosphoria Formation in the Soda Springs area consists of three members which, in ascending order, are the Meade Peak Phosphatic Shale, the Rex Chert, and the Cherty Shale (McKelvey et al., 1959; Montgomery and Cheney, 1967; Brittenham, 1976; Oberlindacher, 1990). The Meade Peak unconformably overlies the Grandeur Tongue of the Permian Park City Formation, and the Cherty Shale Member is overlain by the Triassic Dinwoody Formation. The measured sections of this chapter focus on the Rex Chert; the Cherty Shale Member was measured in one section. The contacts between the Meade Peak and the Rex Chert and between the Rex Chert and the Cherty Shale are gradational. The transitional rocks generally contain carbonates or carbonate-rich beds. Section l was measured through a road cut that exposed the Rex Chert and an adjacent quarry wall that exposed the Cherty Shale Member. Section 5 was measured in a small quarry where the Rex Chert was being mined for surfacing roads. The underlying Meade Peak was not exposed and the overlying Cherty Shale was not present. Section 7 was measured along the exposed face of the now-abandoned South Maybe Canyon Mine and includes the upper Meade Peak rocks and transition rocks that comprise the lowermost part of the Rex Chert. Measurements record true thickness of strata.
TABLE 14-1I X-ray diffraction mineralogy of Rex Chert and associated samples r~ ~o q..,.
Major mineral
Moderate
Minor/trace
Cherty-calcareous shale bed Laminated siliceous shale Siliceous dolostone Argillaceous chert bed Argillaceous chert bed Shaly interbed Cherty shale bed Phosphatic-ferruginous chert bed Chert bed Base of massive thick chert bed Chert bed Chert bed Chert bed Chert bed Chert bed Siliceous limestone bed
Quartz Quartz Quartz Quartz Quartz Quartz Quartz Quartz
Feldspar Feldspar Dolomite Feldspar
Dolomite, gypsum, smectite(?) Illite, smectite, CFA Calcite, feldspar, smectite(?) Goethite Dolomite, feldspar, smectite, gypsum(?) Feldspar, smectite, illite, CFA Feldspar, illite, CFA, calcite, dolomite Clay minerals
Chert bed Leached zone in chert Chert bed Chert bed
Quartz Quartz Quartz Quartz
Type and interval
Section 1" 1 J - 1 N = C h e r ~ Shale Member
601-27-1N 601-27-1M 601-27-1 L2 601-27-1L 1 601-27-1 K2 601-27-1K 601-27-1J 601-27-11 601-27-1H 601-27-1G 601-27-1F 601-27-1E 601-27-1D 601-27-1C 601-27-1B 601-27-1A
CFA
Quartz Quartz
Illite, CFA, smectite(?) CFA, heulandite(?)
Quartz Quartz Quartz Quartz Quartz Calcite
Smectite(?) CFA, smectite(?), chlorite or kaolinite(?) Clay mineral Dolomite, calcite, smectite(?) Dolomite, calcite, smectite(?) Dolomite, smectite(?)
Quartz
Section 5
601-26-1A 601-26-1B 601-26-1C 601-26-1D
Kaolinite or chlorite, goethite, bixbyite Illite, chlorite or kaolinite Illite, chlorite or kaolinite Continued
4~
TABLE 14-II oo
Type and interval
Major mineral
Moderate
Minor/trace
Section 7 : 1 A - 1 C = Meade Peak M e m b e r
601-28-1R 1 601-28-1Q 601-28-1P 601-28-1N 601-28-1M 601-28-1L 601-28-1K 601-28-1 J3 601-28-1 J2 601-28-1J 1 601-28-11 601-28-1H 601-28-1G 1 601-28-1 G2 601-28-1 G3 601-28-1F 601-28-1E 601-28-1D 601-28-1C 601-28-1B 601-28-1A
Breccia Chert bed Chert bed 1N-1K = parts of a 80 cm thick white chert bed Chert Chert Chert Chert Chert Chert Chert bed Chert bed Chert Chert Chert Phosphatic cherty shale bed Siliceous siltstone Siliceous siltstone, marker bed Phosphatic-calcareous shale Carbonate nodule in black shale Black phosphatic shale
Quartz Quartz Quartz Quartz
Dolomite CFA(?) Calcite, dolomite, clay minerals
Quartz Quartz Quartz Quartz Quartz Quartz Quartz Quartz Quartz Quartz Quartz Quartz Quartz Quartz
Clay minerals Smectite(?) Smectite(?) CFA(?)
Quartz
CFA Plagioclase Feldspar
Smectite(?) CFA(?) CFA(?), smectite(?) CFA(?), gypsum CFA, clay minerals CFA, smectite Illite, smectite(?) Illite Illite, smectite(?), jarosite(?) Illite, natrojarosite
Calcite
CFA, calcite, feldspar Dolomite, CFA
Quartz
Feldspar, CFA
Calcite, dolomite, illite, pyrite
Feldspar, K-feldspar plus plagioclase; clay minerals, undifferentiated mixed-layer clay minerals.
Quartz
Composition of Rex Chert Member
409
In section 1, the Meade Peak Member is covered and the lowermost beds exposed include limestone alternating with black chert (Fig. 14-2, unit A). Most of the lower part of the Rex Chert is thin-bedded black and gray chert, which gives way up section to thick-bedded massive gray chert (units B-F). Load casts occur at the base of the thick chert. The uppermost massive chert beds are composed of composite beds displaying pinch-and-swell structures (unit F). Above the uppermost massive thick-chert bed are several thin, iron-rich, weathered chert beds (unit G). Thin sections reveal that most of the Rex Chert in this section is spiculite. The transition to the overlying Cherty Shale Member is covered in the measured section, but in the outcrop on the opposite (south) side of the road, that position may be occupied by a chert-pebble conglomerate that is not seen on the north side of the road. The lower third of the Cherty Shale Member consists of thin-bedded (---6 cm) siliceous shale, with some beds displaying load casts (unit HI). Beds generally thicken up section. The middle third of the section consists of thin-bedded (6-15 cm) argillaceous chert with thinner (< 1-3 mm) interbedded siliceous shale (unit H2). The upper third of the section consists of moderately thick-bedded (10-30 cm) argillaceous chert with siliceous-shale partings commonly being shear surfaces (unit H3). Near the top of unit H3, a 0.5 m thick body of pale-brown siliceous dolostone appears to be the hinge of a fold. The remainder of the bed(?) is not exposed laterally from that body. This was probably originally a dolostone bed or laterally extensive lens. A bed (40 cm thick) of laminated and sheared black siliceous shale occurs near the top of the section, which is capped by a nodular siliceousshale bed. The upper part of the section is highly sheared and is likely bounded by a fault. It is also likely that the thickness of the Rex Chert Member and Cherty Shale Member has been altered by folding and faulting. Bedding (So) in much of the Rex Chert has been obscured by development of generally low-relief stylolites (Sl) oriented subparallel to bedding. Locally, bedding has been transposed ($2) so that it parallels axial planes of isoclinal folds that may have formed during compaction, or possibly during emplacement of the Paris thrust plate (see Evans, Chapter 6). $2 remained an active structural element as shown by quartz veins that formed perpendicular to $2 and are offset along Sz. Section 5 was measured at its northwest end in a quarry where the thickness of the Rex Chert is a minimum because the lower part of the section is covered and the upper part has been removed by quarrying. The Rex Chert consists of dark-gray, medium- to thick-bedded chert that is divided into two sections by an intervening friable brown layer of ferruginous chert (unit B) with an earthy texture. This 30-40 cm thick brown zone may be alteration along a fault or leaching of a chert bed. The chert in this section is spiculite, as determined from thin sections. The chert shows bedding-parallel pressure-solution cleavage as indicated by truncated microfossils and accumulations of relatively insoluble components such as sericite and hematite. In section 7, the upper part of the Meade Peak Member (units A-D, Fig. 14-2) and transitional rock in the Rex Chert (units E-F) are well exposed. Alternating beds of black shale, phosphorite, and carbonate characterize the Meade Peak exposure. Carbonates (calcite and dolomite) occur as continuous beds, horizons of disconnected concretions, and as isolated concretions in some shale beds. Generally, the carbonates thicken up section from nodules to thin beds to thicker beds, except near the base of the exposed section where a distinctive
410
JR. Hein et al.
0.6 m thick carbonate "marker bed" occurs. These marker beds crop out over a distance of about 3 miles (4.8 km) in the North, Middle, and South Maybe Canyon mines in the Dry Ridge Mountains. Shale and phosphorite laminae wrap around the carbonate concretions indicating an early diagenetic origin (pre-compaction). The Meade Peak is overlain by a 0.8 m thick marker bed of siliceous siltstone (Fig. 14-2, unit E). That bed is overlain by a section that is partly covered, but consists predominantly of thin-bedded siliceous siltstone (unit F). That transition zone of siliceous siltstone is overlain by thin- to medium-bedded (5-30 cm) black chert with brown upper and lower margins on each bed (unit G). Load casts occur at the base of some beds. A few thin siliceous-siltstone beds also occur in unit G. Those cherts are overlain by 1 m of thin-bedded white chert that is overlain by a 1 m thick white chert marker bed (unit H). The white cherts are overlain by 10 m of thin- to mediumbedded (to 30 cm) black and gray chert (unit I) that is very similar to cherts in unit G. The section is capped by an 8 m thick, discontinuous chert megabreccia with a calcite-bearing quartz cement. This breccia body may be a slump deposit or fault-zone deposit.
Petrography
The dominant characteristic of the chert beds is the presence of sponge spicules, which vary from relatively well preserved to faint ghosts. Most beds can be classified as spiculites. These spiculites are laminated and commonly show a preferred orientation of elongate grains parallel to bedding. One sample (601-28-1H) shows preferred orientation in some laminae but not in others, indicating that the alignment of grains was caused by bottom currents rather than by compaction or tectonics. A sparse to common occurrence of rhombs characterize most spiculite beds, which are likely quartz-replaced dolomite. Glauconite, mica, and feldspar are present in some beds. Various combinations of bivalves, fish debris, radiolarians(?), and calcareous algae(?), are seen in some beds. Spiculite beds in section 7 show sedimentary structures that include cross bedding and cut-and-fill scouting. The upper part of section 7 does not consist of spiculite beds, but rather replaced carbonates. These chert beds are generally white to grayish, centimeters to a meter thick, and in thin section consist of abundant rhombs partly to completely replaced by quartz. Some replaced rhombs show relict carbonate twinning. Laminae are compacted around some large rhombs, which indicates that they formed prior to compaction. The textures suggest that carbonate and silica fossils were deposited on the seafloor, dolomite rhombs formed in unconsolidated sediment during early diagenesis, compaction took place with increasing burial, and finally carbonate grains were replaced and cement was precipitated during silica diagenesis. The Cherty Shale Member from section 1 consists of siliceous-siltstone beds and one siliceous-dolostone bed. Beds are laminated, contain ghosts of spicules, and some beds contain various combinations of sparse rhombs, bivalves, fish debris, feldspar, fibrous clay minerals, and calcite. Carbonate minerals are more common than they are in the underlying chert. Grains are well sorted. Uncommon sedimentary structures include burrows and reverse grading. The siliceous siltstone from the Meade Peak-Rex Chert transition zone (lower Rex Chert) in section 7 is somewhat different from the siliceous siltstone in section 1. The section 7
Composition of Rex Chert Member
41 1
siltstone is more compacted and shows a preferred fabric created by aggregate extinction of clay minerals, or by parallel orientation of thin wavy iron-rich or organic matter-rich lenses. Grains are moderately well sorted and range from angular to subrounded, but are predominantly subangular. Similar to section 1, siltstone beds are laminated, contain ghosts of spicules, and some beds contain various combinations of bivalves, fish debris, feldspar, and fibrous clay minerals, as well as mica and chlorite; however, no rhombs or calcite were seen.
Mineralogy The mineral content of the Rex Chert and Cherty Shale Member is dominated by quartz (Table 14-II). Several beds also have major amounts of carbonate minerals. For example, the lowermost bed in section 1 (601-27-1A) contains major amounts of calcite as well as quartz; a siliceous dolostone (601-27-1L2) in section 1 contains major amounts of dolomite as well as quartz. Feldspar (combined K-feldspar and plagioclase), dolomite, and carbonate fluorapatite (CFA) occur in moderate amounts in a few beds (Table 14-II). Clay minerals occur in minor amounts. In the upper part of the Meade Peak Member in section 7, calcite is a major phase in some samples and CFA is moderately abundant in all samples analyzed.
Chemical composition The mean concentrations of elements in the Rex Chert and the Cherty Shale Members are overwhelmingly dominated by silica, which averages 94.6% for the three sections studied and ranges from 92.2% for section 1 to 96.9% for section 5 (Table 14-III). The sums of the major oxides are reasonably close to 100%, but are a little high for most of the stratigraphically highest 13 samples from section 7, as much as 103% (Table 14-III). These high values provide a measure of the analytical accuracy of silica determinations for these very silica-rich rocks. Samples with low sums of the major oxides result from the exclusion in those totals of organic C and S compounds, which are significant components of the Meade Peak rocks in section 7. Organic C contents are generally very low in the chert, but are as much as 1.8% in samples from the Cherty Shale Member in section 1 and as much as 3.4% in samples from the transitional rocks (lower Rex Chert) in section 7 (Table 14-III). Similarly, phosphate (P205) is generally low in the chert, but can be high (as much as 3.1%) in individual beds; these beds do not consistently occur at any particular stratigraphic level in the sections. Selenium concentrations for samples of the Rex Chert and Cherty Shale Members vary from the detection limit (<0.2) to 138 ppm, with a mean concentration of 7 ppm (Table 14-IV), or < 1.0 ppm if two outliers are removed. The mean Se concentration for section 5 rocks (0.65 ppm) is equivalent to that of average shale, 0.6 ppm (Krauskopf, 1979), whereas that of section 1 (1.32 ppm) is 2.2 times the concentration in average shale and section 7 (12 ppm) is 20 times the average-shale concentration (Tables 14-V-14-VII). The reason for these differences is that section 7 includes the lowermost Rex Chert, which contains rocks of transitional character, and section 1 includes the Cherty Shale Member. All chert beds
TABLE 14-111 Chemical composition of rocks from three sections i !, 5, 7: Figs 14-1 and 14-2)of Rex Chert and adjacent rocks: ~ m p l e s are listed in stratigraphic order: major oxides, C, and S in wt.%, others in ppm
Lab no.
Sample no.
Sample description
Lithology
t'~
SiO, ICP-16
..\lzO~ ICP-16
Fe:Os ICP-16
TiO_, ICP-16
CaO ICP-16
K,O ICP-16
MgO ICP-16
Na.zO ICP-16
P:O.~ ICP-16
Y_CO_, Acid.
Total
Section 1. Rex Chert 1,4 11: Cherty Shale Member IJ IN 601-27-1N Brown. calcareous C- 197024 60 i -27- I M Black C- i 97023 601-27-1L2 Pale brown to gray = fresh C-197022 601-27- I L 1 Brown, earthy, porous C-197021 601-27-1 K2 Brown, calcareous C-197020 601-27- ! K Brown C-197019 601-27-1J Brown. calcareous C-197018 601-27- ! 1 Brown. Fe stained C-197017 60 ! -27-1H Gray C-197016 601-27-1G Gray. spicular C-197015 601-27-1F Gray C-197014 601-27- ! E Gray C-197013 601-27-1D Black C-197012 601-27-1C Black, calcareous C-197011 601-27-1B Black, calcareous C-197010 601-27-1A Gray C-197009
Cherty-calcareous shale bed Laminated siliceous shale Siliceous dolostone Argillaceous chert bed Argillaceous chert bed Shal~ interbed Chert~ shale bed Phosphatic-ferruginous chert bed Chert bed Base of massive thick chert bed Chert bed Chert bed Chert bed Chert bed Chert bed Siliceous limestone bed
87.5 78.7 48.8 92.6 '92.0 '90.9 86.2 83.2 98.2 95.6 9`9.9 96.3 98.2 '97.5 93.5 44.3
3.27 8.41 1.13 2.74 2.40 3.48 3.82 2.55 0.79 1.19 0.26 0.62 0.60 0.43 0.59 0.19
0.77 2.66 0.71 1.92 0.64 0.'96 1.16 4.68 0.76 1.43 0.23 0.44 0.33 0.19 0.17 0.06
0.217 0.550 0.067 0. ! 67 0.167 0.234 0.234 0. I 17 0.017 0.033 <0.020 <0.020 0.017 0.017 0.033 <0.020
2.28 0.98 16.2 0.43 ! .41 0.84 2.01 3.69 0.50 0.97 0.39 1.27 0.21 1.12 2.31 30.9
0.61 1.95 0.16 0.30 0.43 0.64 0.78 0.64 0.13 0.22 0.02 0.08 0.10 0.06 0.07 0.04
1.09 0.80 9.91 0.18 0.73 0.32 0.95 0.45 0.07 0.12 <0.02 0.05 0.05 0.18 0.93 0.48
0.63 0.59 0.22 0.61 0.31 0.42 0.49 0.08 0.04 0.03 0.01 0.03 0.04 0.04 0.05 0.04
0.48 0.69 0.16 0.32 0.23 0.44 0.44 3.05 0.37 0.85 0.28 0.96 0.09 0.16 0.28 0.05
1.69 0.03 23.4 0.02 I. 15 0.23 1.58 0.08 0.02 0.02 0.01 0.02 0.03 0.79 2.15 25.1
98.6 95.4 100.8 99.3 99.5 98.5 97.7 98.5 100.9 100.5 101.1 99.7 99.7 100.5 100.1 101.1
Section 5 C-197005 C-197006 C-197007 C-197008
Chert bed Leached zone in chert Chert bed Chert bed
98.6 93.9 97.5 97.5
0.55 1.49 1.04 I. 10
0.31 2.65 0.47 0.49
0.017 0.083 0.033 0.050
0.22 0.43 0.29 0.25
0.07 0.17 0.16 0.18
0.05 0.12 0.10 0.08
0.04 0.04 0.05 0.07
0.14 0.30 0.18 0.14
0.02 0.01 0.03 0.01
100.0 99.2 99.9 99.9
9'9.5 101.0 96.7 102.1) 101.8 101.4 101.6 100.3 101.6 101.2 96.7 99.5 96.0 90.7 95.4 89.0 74.9 78.5 48.3 1.07 46.0
0.38 0.38 I).25 0.28 0.32 I).23 0.36 11.28 0.32 0.42 I).77 I).43 1.38 0.93 1.64 3.48 10.7 I 1.1 8.35 0.15 7.29
0.07 0.06 0.04 0.17 0.04 0.13 0.10 0.14 0.04 0.09 0.21 0.17 0.37 0.26 0.57 ! .24 2.09 0.96 2.75 0.06 2.37
<0.020 <0.020 <0.020 <0.020 <0.020 <0.020 <0.020 <0.020 <0.020 <0.020 0.033 0.017 0.050 0.050 0.083 0.183 0.784 0.817 0.484 <0.020 0.484
1.05 0.46 2.92 0.28 0.18 0.27 0.35 0.28 0.20 0.34 0.50 0.34 1.01 0.67 1.48 2.46 0.17 0.34 17.1 48.7 I I.I
0.05 0.04 0.01 0.04 0.01 <0.01 0.02 0.02 0.02 0.02 0.10 0.05 0.22 0.14 0.27 0.69 2.58 2.57 1.88 0.01 1.94
0.55 0.02 0.70 0.02 0.02 0.02 0.02 0.02 <0.02 0.02 0.03 0,03 0.08 0.05 0.12 0.30 0.65 0.61 0.70 3.30 1.44
0.05 0.05 0.04 0.04 0.03 0.03 0.03 0.04 0.03 0.03 0.08 0.07 0.12 0.11 0.12 0.15 1.20 1.32 0.66 0.1 ! 1.21
0.14 0.28 0.09 0.18 0.09 0.16 0.23 0.18 0.11 0.21 0.32 0.23 0.69 0.39 1.01 1.72 0.09 0.05 5.50 2.75 4.42
1. I 0 0.02 2.78 0.01 0.01 0.01 0.02 <0.01 <0.01 0.01 0.02 0.02 0.05 0.02 0.07 0.11 <0.01 <0.01 8.18 39.3 5.49
i 02.9 102.3 103.5 103.1 102.5 102.2 102.7 101.3 102.3 102.3 98.8 100.8 100.0 93.3 100.8 99.3 93.1 96.3 93.9 95.4 81.7
601-26- I A 601-26-1B 601-26- I C 601-26- ! D
Dark gray, fractured Earthy, bro~,n Dark gray, fractured Dark gray, fractured
Section 7: Meade Peak 1,4 I C: Rex Chert I D I R 601-28- I R 1 Pale brown-gray chert clast C-197045 601-28-1Q Dark gray, fractured C-197044 601-28-1P Gray, mottled C-197043 601-28-1N White. upper 15cm margin C-197042 601-28-1M White, middle 7 c m C-197041 601-28-1L White, 7 - 1 5 c m above base C-197040 601-28-1K White, lower margin 18cmJ C-197039 601-28-1J1 Pale brown, upper bed margin C-197036 601-28-1J2 White-gray, main bed. dense C-197037 601-28-1J3 White-gray, lower bed margin C-197038 601-28- I 1 Black-bro~'n C-197035 601-28I H Pale brown, silty C-197034 601-28-1G1 Pale brown, upper bed margin C-197031 601-28-1G2 Brown, main bed C-197032 601-28-1G3 Pale brown, lower bed margin C-197033 601-28-1F Gray-brown, fossili ferous C-197030 601-28- ! E Black, thin-bedded C-197029 601-28-1D Gray-brown, mid. thick bed C-197028 601-28-1C Black C-197027 601-28- I B Black nodule C-197026 601-28-1A Black, carbonaceous C-197025
Breccia Chert bed Chert bed I N - I K = parts o f a 80cm thick \Vhite chert bed Chert Chert Chert Chert Chert Chert bed Chert bed Chert Chert Chert Phosphatic chert,,' shale bed Siliceous siltstone Siliceous siltstone, marker bed Phosphatic-calcareous shale Carbonate nodule in black shale Black phosphatic shale
~ - ~ 2~ ~ _
'c
9
9
_:
,
- " ' - - - S T " '
"-:
.
~
~
.
.
.
v
_.
.
v
.
.
v
_.
V
.
.
v
.
.
v
.
.
V V V
_ .
v
.
~,
.
.
.
,~
.
.
.
,~.
.
v
.
~_
~,/
.
.
.
v
.
.
,,
V
.
.
~
~
V V V V V V V V V V
,,
.
.
.
,..
.
.
????Ij~'??,??t,'t,'t,'t,'t,'t,'~'~,'????~g~'t~'?~'i;~;
.
,~
.
.
'v
,~,
v
.
v
.
.
,,
v
.
~
,~
.
~
~
~
~
- ~ _ z , . , ~ ~
,,
"7. '~. ~.
,,
,,
v
~!
,,
,,
~.
'v' V
.
V
.
.
,1
"7.
~
~
,,
~
~
~
V
~
o,
V
r
v"'
,-~, ~
--
~ t,'t,'~'~j . . . . t , ~ ? ? , , - ,
"~
? "' ? ~ ,':~' ? ? ? ? ? 1,~ ? t~' ,? ? ? ? ? t,' ? ? ? ? i~' ,"~' :,,' . . . . ? ~' ? ? " ~ - r-: ~' . . . . "-, ,'-.,
V
Composition o f Rex Chert M e m b e r
.r.
z,
.d E
c~
$3
413
TABLE 14-111 Conr~nuvd Sample no
LI Mn Mo Nb 1 ~ p - 4 0 1 ~ p - 4 0 1 ~ p - 4 0 1~p.40
601-27-IN 601-27-IM 601-27-lL2 601-27-ILI 601-27-IK? 601-27-IK 601-27-11 601-27-11 601-27-IH 601-17-16 601-27-IC 601-27-It 601-27-ID 601-27-IC 601-27-18 601-27-IA 601-26-I A 601-26-18 601-26-LC 601-26-ID 601-28-IRI 601-28-10
14 27 16 29
601-28-lP 601-28-IN 601-28-IM 601-28-IL 601-28-IK 601-28-111 601-28-I12 601-28-113 601-28-11 601-28-IH 601-28-161 601 -28-lG2 601-28-163 601-28-IF 601-28-lE 601-28-ID 601-28-IC 601-28-IB 601-28-IA
J
17 I3 I? 43 8 8
< 6 Y 4 4 4
Nd 1~p.40
NI Pb l ~ p l O ICP-40
Sh H?dnde
Sc ICP-40
<J
25
37
76 298 333
<2
<4 6
37 10 13
97 14 45
68 60 91 31 13
1
<4 <4 <4 4
14 22 197 ?I
29 40 4. 21 10
<4
<2
<4
<.(
30 I?
13 4
<4 <4
(2 <2
<4 <4
23
<J
<9
<4 9 <4 <4 4 <4
<2
21 20 8 11
16 16
38 42 37 5"
<2
-
7
<.I
22
9 7
07 06
9
4 <4 5
(2 1 2
7
6 li
06 (1')
-" -
4 4 .7
<2
Sr ICP-I6
Th ICP-40
28 09
85 66 I24
92 70 126
<6 7 <6
01 0.5 -31
32 86 I4
36 49 13
?3 52 63 215
26 65 56 65 231
<6 <6 <6 <6 <6
03 01 0.1 0.1 01
27 20 31 37 I88
20 16 31 32 321
09 l? 04
27 34 I6
30 40 17
<6 <6 <6
<01 03
24 38 4
30 54 16
19 UX <02 03
44 I4 23 34 I88 22 24
47 I4 23 36 183 23 28
<6 <6 <6 <6 <6 <6 <6
<0 1 <0.1 <01 <0 1 <01 <0.1 0.3
21 19 5 14
46 8 6 13
5 11 35
3 10 20
15 18
1: 13
5 3
I
0
9 I 8 19 4 11
M
ICP-40
Zr
Zn
ICP-40
ICP-I6
2 3
110 154 28
95 164 31
2
112 49 98 99 52 43 135 30
2
78 133 57
70 72 101 91 42
81 32 29 244 45 50
16
38 24 73 I21 28 109 94 153
12
62 79
<4
<9
148 29
<2
<4 <J
<9
1710 36
<4 <4
15 I1
98 II
<.I
07
05
24
<4 <4 <4 <4
9
13 h
<4 <4
U7 OX
<2
09
51 19
27 54 ?I
<6 <6 <6
<01 <0.1
5 23 X 4
<4 <4
?? IOh 21 160 92 57
<4
<2 <2 2
02
17 I5 II 8
17 15 15
<6 <6 <6 <6
<01 <0 1 <01 <0.1
3 <2
5 5 4 7 3
3 13
191 221
<9
10 I1
<4 <4
'2
<02
10 I?
<6 <6 <6
<0 1
3 5 3
6 8 6
15 17 32
<6 <6 <6
<01 <01 <01
<2 4 13
4 8 15
' 5. -
I4 I2 I5 29
I1 13 15
07 13 19 1 3
I6 43 36 58
17 46 35 60
<6 <6 <6 <6
<01 07 04 08
5 47 27 63
9 29 19
91 90
14
44
1
253 123 437
19 13 24
3X 138 101
95 80 7X
97 83 82
<6 I2 9
09 13 10
130 I?? 63
76 10 9
4 2 2
569 96 115
52 394 397
275 696 437
282 699 436
<6 <6 <6
I1 <01 I1
110 280 196
206 21 144
10
61 99 5
658 201 1460
233 23 243
10 1)
10 5 4
4 J
7
299 25" 227
4.1
<4 <4
<2
<J
<9 <9 <9
4
x
20 I5 21
5
<.I
10 <.I
<4
<2
c? <2 <2
<4 <4
<9 <9
13 13
<4 5X
07 114
<2
<4 <4
13 <9
31 29
6 <-I
10
<2 <2
6 6
107
12 8 16 18
55 51 58 34
<4 <4
21 37
45 86
<4 5
?I 18 ?I 42 31
32 24 496 69 I46
12 I4 I8 65
<4 5
18 20
33 25
14 13
5 4
93 I6 59
251 27 361
I4
AX
<06
Yb ICP-40
Sr lCPl0
i
XI
TI V Hydr~de ICP-40
Y
Se H?dndc
2
11
I? h
08 I5 IX 09
3 4 X 7
lX 09 62
"72 <2 4
8
415
Composition of Rex Chert Member
TABLE 14-IV Statistics of chemical data for 36 Rex Chert and Cherty Shale samples collected from measured sections 1,5, and 7; for comparison mean values for 9 Rasmussen Ridge mine and 13 Enoch Valley mine composite channel Rex Chert samples Element
N
Si02 (wt %)
36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36
A1203
Fez03 TiOz CaO K2O MgO Na20 p2os
Cco2 Total
ct Cc Carg SI As (ppm) Ba Ba Ce Cr Cu Hg La Li Mn Nd Ni Sb Se Sr V Y Zn Zr
--
Mean
Median
94.6 1.92 0.75 0.1 1 0.91 0.37 0.27 0.20 0.43 0.34 99.9 0.45 0.09 0.35 0.05 4.37 117 116 12 152 50 0.04 17 11 133 18 26 0.94 7.60 40.2 31.2 27.8 115 51
96.7 0.78 0.35 0.033 0.48 0.1 15 0.09 0.05 0.255 0.02 100 0.18 0.01 0.13 0.025 2.6 50.5 49.5 5 73 32 0.03 9.5 8 56 10 20 0.6 0.85 25.5 18.5 13 92.5 15.5
SD
Minimum
7.07 2.73 0.98 0.20 0.88 0.65 0.33 0.32 0.56 0.69 2.46 0.67 0.19 0.63 0.08 4.59 147.5 144 17.4 234 58.8 0.04 29.2 8.62 283 32.2 24.0 1.85 27.9 38.5 41.4 53.1 1 10 92.1
74.9 0.23 0.04 < 0.01 0.17 0.005 0.01 0.01 0.05 0.005 93.1 0.01 < 0.0015 0.007 < 0.025 0.3 17 16 < 2.5 5 6 < 0.01 3 1 13 < 4.5 4 < 0.3 <0.1 8
-
R.I. Grauch and J.R.Herring, unpublished data.
Maximum
Rasmussen mean2
Enoch mean2
TABLE 14-V Statistics of chemical data for 14 Rex Chert and Cherty Shale samples collected from measured section 1 Element
N
SiO, (wt %)
14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
A1203 Fez03
Ti02 CaO K2O MgO Na20 P20s CCO2 Total CI CC Corg
SI As ( P P ~ ) Ba Ba Ce Cr Cu Hg
La Li Mn Nd Ni Sb Se Sr V Y Zn Zr
Mean
Median
SD
Minimum
Maximum
in the upper part of section 7 have Se concentrations of c0.2-0.8 ppm. The mean concentration of 12 ppm is heavily dependent on two sample values of 101 and 138 ppm. Without those two samples, the mean Se concentration in section 7 rocks would be 0.8 ppm, close to the average-shale content.
Composition of Rex Chert Member
TABLE 14-VI Statistics of chemical data for four Rex Chert samples collected from measured section 5 Element
N
SiO, (wt %)
4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
A1203
Fe203 TiOz CaO K2O MgO NazO Pz05 CCO* Total
c, cc Corg
st As ( P P ~ ) Ba Ba Ce Cr Cu Hg La Li Mn Nd Ni Sb Se Sr
v Y Zn Zr
Mean
Median
SD
Minimum
Maximum
Other elements of environmental interest include As, Cr, V, Zn, Hg, and Cd. Arsenic concentrations vary by a factor of 56, from 0.3 to 16.8 ppm, with a mean of 4.2 ppm, which is slightly less than the concentration in average shale of 6.6 ppm (Govett, 1983). Chromium concentrations vary by a factor of 220, from 5 to 1100 ppm, with a mean of
418
JR. Hein et al.
TABLE 14-VII Statistics of chemical data for 18 Rex Chert samples collected from measured section 7 Element Si02 (wt. %) A1203 Fez03 TiO, CaO Kz0 MgO Na20 p205 CcOz Total
ct Cc
Corg
s, As (PPm) Ba Ba Ce Cr Cu Hi2 La Li Mn Nd Ni Sb Se Sr V Y Zn Zr
N
Mean
Median
SD
Minimum
Maximum
18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18
143 ppm, which is higher than the 100 ppm concentration in average shale. Vanadium concentrations vary by a factor of 188, from 1 to 188 ppm, with a mean of 30 ppm, which is much lower than the 130 ppm concentration in average shale. Zinc concentrations vary by a factor of 24, from 24 to 569 ppm, with a mean of 110 ppm, which is somewhat greater than
Composition of Rex Chert Member
419
the concentration in average shale of 80 ppm. Mercury concentrations vary by a factor of 16, from < 0.01 to 0.16 ppm, with a mean of 0.04 ppm, which is an order-of-magnitude less than the concentration in average shale of 0.4 ppm. Cadmium concentrations are uniformly below the limit of quantification (< 2 ppm) except for rocks from the lowermost part of section 7; we are unable to ascertain whether the average may be higher than the 0.3 ppm of average shale.
Stratigraphic changes in chemical composition Stratigraphic changes (equivalent to temporal changes in the depositional basin) in the chemical composition of rocks are notable either as uniform changes through the sections or as distinct differences in the mean composition of rocks that comprise the upper and lower halves of the sections. In this regard, the concentrations of most elements increase up section at the expense of silica in section 1, whereas they decrease up section in sections 5 and 7. The up-section increases in section 1 are partly due to inclusion of the Cherty Shale Member samples. However, there are still up-section increases in roughly half the elements even if the Cherty Shale samples are excluded. Silica has the opposite trend of the other elements. For section 1 (including the Cherty Shale Member), the following elements increase up section: AI, Fe, Ti, K, Na, As, Ba, Ce, Cr, Hg, La, Li, Ni, Sc, Sr, V, and Zr; in contrast, silica decreases up section and Ca, Mg, C, and Mn decrease to near mid-section then increase farther up section. For section 5, the following elements decrease up section: A1, Fe, Ti, K, Na, C, Ce, Ni, Sr, V, Y, and Zn, whereas Si increases up section. For section 7, the following elements decrease up section: A1, Fe, Ti, K, Na, organic C, Ba, Ce, Cr, La, Li, Ni, Se, Sr, T1, V, Y, Zn, and Zr; whereas Si and Mn increase up section.
Phase associations of elements The phase associations of elements were determined by comparing results from element correlations (Tables 14-VIII-14-X), rotated factor loadings from Q-mode factor analyses (Figure 14-3), and mineralogy as determined by XRD (Table 14-II). Three data sets were analyzed including data from sections 1, 5, and 7 combined, section 1 data, and section 7 data (excluding Meade Peak and Rex Chert carbonate beds). We consider four to five Q-mode factors that account for 95% or more of the variance in each of the three data sets. The factors are interpreted to represent the following rock and mineral components. Factor 1 is a chert-silica component consisting solely of Si, except for the combined data set where Ba shows a minor but statistically significant factor loading. Factor 2 is a phosphorite-carbonate fluorapatite component composed of P, Ca, As, Y, V, Cr, Sr, and La (_ Fe, Zn, Cu, Ni, Li, Se, Nd, and Hg depending on the data set). Factor 3 is a shale component composed of A1, Na, Zr, K, Ba, Li, and organic C (• Ti, Mg, Se, Ni, Fe, Sr, V, Mn, and Zn depending on the data set). Factor 4 is a
4~
TABLE 14-Vlli Correlation coefficient matrix for 36 samples from sections 1.5. and 7 combined (listed in TABLE 14-111): the poinl of zero correlation for n = 36 at the 99% confidence level is 10.421 Element AI203 Fe_,O3 TiO~ CaO K,O MgO Na,O _ P205
CO, CT Cc C,,~ ST As Ba Ce Cr Cu Hg La Li Mn Nd Ni Sb Se Sr V Y Zn Zr
SiOz
Nd Ni Sb Se Sr V Y Zn Zr
AI:O~
Fe:O;
-0.905 -0.686 0.504 - 0.882 0.995 0.451 -0.369 0.048 0.398 -0.897 0.995 0.494 -0.682 0.545 0.314 -0.824 0.937 0.357 -0.347 0.083 0.68 -0.11 -0.061 -0.146 -0.784 0.758 0.348 -0.108 -0.064 -0.149 -0.798 0.821 0.413 -0.673 0.746 0.232 -0.574 0.462 0.731 -0.546 0.557 0.204 -0.806 0.691 0.838 -0.72 0.529 0.851 0.036 0.003 -0.03 -0.861 0.779 0.823 -0.501 0.27 0.789 -0.715 0.577 0.854 0.068 -0.083 0.277 -0.451 0.21 0.807 -0.552 0.468 0.56 -0.028 0.024 0.063 -0.642 0.805 0.225 -0.718 0.493 0.77 -0.762 0.604 0.825 -0.384 0.116 0.766 -0.194 0.159 0.165 -0.825 0.963 0.385 Ni
0.168 0.091 -0.004 0.022 0.054 0.887 0.323 0.793 0.496 0.99 0.164 0.058 0.652 0.109 0.303
Sb
riO,
CaO
0.007 0.992 0.047 0.542 0.617 0.951 -0.012 0.013 0.679 -0.044 0.643 0.764 0.211 -0.047 0.641 0.823 0.033 0.775 -0.029 0.407 0.307 0.564 0.077 0.659 0.444 0.464 0.558 -0.008 -0.018 0.748 0.202 0.219 0.615 0.526 0.408 -0.077 -0.141 0.157 0.613 0.423 0.244 0.025 - 0 . 0 2 1 0.837 -0.168 0.446 0.665 0.549 0.512 0.057 0.642 0.116 0.247 0.98 -0.053
MgO
Na,O
P-O,
CO:
0.534 0.915 0.566 0.091 0.098 -0.068 0.739 0.768 0.734 -0.071 0.736 0.835 0.558 0.757 0.497 0.456 0.178 0.539 0.543 0.708 0.448 0.539 0.389 0.001 -0.165 0.787 0.373 0.292 0.212 0.555 0.361 -0.118 -0.1 i I 0.23 0.212 (I.42 0.325 0.037 - 0 . 0 6 1 0.827 0.284 0.501 0.465 0.614 0.335 0.134 0.154 0.129 -0.021 0.963 0.492
-0.086 0.038 0.735 0.035 0.768 0.805 0.251 0.607 0.577 0.324 -0.039 0.618 0.125 0.491 -0.098 0,071 0,325 0.016 0,814 0.377 0,421 -0.027 0.036 0.955
-0.117 -0.051 -0.119 -0.018 -0.14 0.604 -0.073 0.65 0.812 0.131 0.444 0.909 0.659 -0.107 0.896 0.295 0.086 -0.136 0.822 0.794 0.938 0.391 -0.058
0.334 ! 0.056 0.089 -0.229 0.158 -0.074 -0.091 -0.198 -0.205 -0.1 -0.134 -0.07 -0.(176 -(I.023 -0.109 -0.115 0.04 -0.149 -0.082 -0.14 -0.047
K:O
Se
Sr
V
Y
0.064 0.08 0.266 0.098 0.405 0.095 -0.069 0.047 -0.01 0.022 0.915
0.897 0.866 0.214 0.395
0.775 0.436 0.488
0.12 0.005
Zn
0.05
(;
0.332 0.96 0.822 0.343 0.509 0.488 0.416 -0.079 0.616 0.122 0.35 -0.14 0.083 0.326 0.009 0.663 0.362 0.428 0.(107 -0.(138 0.742
(t
0.054 0.086 -0.232 0.153 -0.078 -0.094 -(I.201 -0.207 -0.103 -0.137 -0.072 -0.(}79 -(I.026 -(I.108 -0.117 0.036 -(I, 152 -0.085 -(I.142 -0.05
(-,~
ST
As
Ba
0.844 0.433 0.493 0.54 0.468 -0.024 0.714 0.16 0.412 -0.127 0.112 0.353 0.041 0.736 0.373 0.498 0.033 0.002 0.8
0.263 0.653 0.462 0.222 -0.077 0.565 0.082 0.288 -0.107 0.018 0.116 0.04 0.868 0.329 0.375 -0.068 -0.056 0.832
0.126 0.566 0.742 0.233 0.732 0.558 0.609 0.288 0.499 0.729 0.032 0.26 0.59 0.816 0.498 0.624 0.333
0.355 0.216 -0.13 0.397 0.072 0.338 -0.11 0.066 0.217 -0.039 0.397 0.4 0.243 -0.002 -0.064 0.552
Ce
0.866 -0.025 0.816 0.861 0.846 -0.104 0.836 0.304 0.145 0.498 0.903 0.882 0.77 0.061 0.621
Cr
Cu
0.066 0.764 0.864 0.851 -0.127 0.84 0.514 0.078 0.19 0.881 0.936 0.821 0.311 0.356
0.078 0.044 0.08 0.034 -0.014 0.198 -0.008 -0.024 0.022 0.14 0.02 0.601 -0.028
Hg
La
Li
0.599 0.75 0.772 0.158 -0.102 -0.025 0.56 0.984 0.755 0.586 0.2 0.477 0,118 0.114 0.056 0.576 0.107 0.263 0.689 0.916 0.812 0.811 0.854 0.828 0.499 0.978 0.717 0,209 0.153 0.218 0,694 0.173 0.455
Mn
-0.074 0.501 4.56E-04 -0.095 -0.167 -0.075 -0.085 0.183 -0.059
TABLE 14-IX Correlation coefficient matrix for 14 samples from section I (listed in TABLE 14-111): the point of zero correlation for n = 14 at the 99~ confidence level is 10.6511 ,~.~~ Element
SiO 2
AI203
Fe203
TiO.
CaO
Ale0 3 Fe203 TiO-, CaO K,O MgO Na-,O P_,O5 CO, CT Cc Co~ S-r As Ba Ce Cr Cu Hg La Li Mn Nd Ni Sb Se Sr V Y Zn Zr
-0.884 -0.717 -0.858 -0.573 -0.895 -0.697 -0.666 -0.476 -0.181 --0.754 -0.175 - 0.805 -0.344 -0.738 -0.432 -0.737 -0.812 0.007 -0.802 -0.512 -0.722 -0.188 -0.517 -0.797 -0.604 -0.513 -0.677 -0.698 -0.483 -0.393 -0.843
0.514 0.993 0.136 0.984 0.548 0.784 0.134 -0.004 0.774 -0.008 0.906 0.293 0.644 0.436 0.458 0.61 0.055 0.7 0.167 0.56 0.274 0.178 0.968 0.259 0.329 0.318 0.421 0.132 0.546 0.959
0.436 0.523 0.549 0.132 0.234 0.851 -0.322 0.101 -0.326 0.254 -0.1 0.907 0.002 0.94 0.944 -0.051 0.855 0.869 0.928 0.176 0.883 0.411 0.812 0.325 0.823 0.955 0.867 0.148 0.384
0.105 0.968 0.58 0.814 0.05 0.053 0.8 0.048 0.912 0.338 0.566 0.482 0.393 0.541 0.052 0.638 0.094 0.511 0.291 0.101 0.967 0.196 0.31 0.264 0.343 0.054 0.526 0.976
0. ! 9 0.608 0.055 0.704 0.502 0.322 0.498 0.167 0.313 0.383 0.247 0.695 0.589 -0.215 0.415 0.717 0.446 -0.163 0.708 -0.008 0.743 0.527 0.843 0.663 0.714 -0.176 0.145
Nd
Ni
Sb
Se
Sr
V
Y
Zn
Ni Sb Se Sr V Y Zn Zr
0.034 0.928 0.391 0.94 0.958 0.998 -0.175 0.071
0. ! 23 0.262 0.165 0.292 -0.006 0.635 0.933
0.502 0.929 0.911 0.919 -0.065 0.186
0.592 0.378 0.382 0.106 0.438
0.92 0.929 -0.137 0.274
0.947 0.047 0.287
-0.177 0.025
0.501
~
K20
MgO
Na20
0.535 0.672 0.592 0.212 0.012 - 0 . 1 4 8 -0.027 0.806 0.233 0.776 0.863 0.669 -0.032 0.803 0.229 0.9 i 8 0.67 0.683 0.251 0.684 0.562 0.707 0.151 0.22 0.395 0.654 0.605 0.516 0.26 0.18 0.681 0.273 0.221 0.09 -0.107 -0.059 0.763 0.243 0.295 0.244 0.074 - 0 . 0 9 7 0.55 0.204 0.429 0.121 0.086 0.648 0.254 0.063 -0.091 0.93 0.471 0.807 0.326 0.231 0.059 0.302 0.449 0.462 0.378 0.355 0.12 0.493 0.181 0.082 0.209 0.039 -0.131 0.51 0.214 0.484 0.908 0.627 0.89
P_.O~
CO.
('~
Cc
Corg
ST
As
Ba
Ce
Cr
Cu
Hg
La
Li
Mn
o~
-0.257 -0.114 -0.261 - 0.023 -0.135 0.732 -0.103 0.894 0.827 -0.117 0.699 0.967 0.702 -0.167 0.976 0.009 0.868 0.407 0.904 0.927 0.984 -0.136 0.023
0.551 1 0.224 0.564 -0.377 0.429 -0.151 -0.216 -0.164 -0.309 -0.204 -0.252 -0.01 -0.226 -0.05 -0.051 0.209 0.035 -0.229 -0.225 -0.07 0.14
0.547 0.937 0.53 0.236 0.547 0.186 0.307 0.122 0.357 -0.069 0.14 0.031 -0.074 0.739 0.141 0.377 0.188 0.129 -0.1(18 0.455 0.8(11
0.22 0.558 0.385 - 0 . 3 7 9 0.434 0.422 0.461 - 0 . 1 5 7 0.282 -0.222 0.451 - 0 . 1 6 6 0.212 -0.313 0.547 - 0 . 2 0 9 0.008 - 0 . 2 5 8 0.271 - 0 . 0 0 8 0.039 -0.231 0.011 -0.052 0.883 - 0 . 0 5 8 0.188 0.201 0.356 0.028 0.207 - 0 . 2 3 4 0.249 -0.229 -0.029 - 0 . 0 6 7 0.559 0.134 0.878
-0.081 0.947 0.04 -0.026 -0.105 0.056 -0.091 0.059 0.066 -0.086 0.245 0.036 0.611 0.206 -0.062 -0.113 0.091 0.442
0.076 0.85 0.927 -0.036 0.912 0.741 0.795 -0.023 0.756 0.536 0.656 0.249 0.706 0.883 0.731 0.207 0.469
0.128 0.089 -0.07 0.198 -0.043 0.188 0.067 -0.037 0.382 0.042 0.58 0.251 0.025 -0.073 0.108 0.569
0.961 -0.026 0.817 0.946 0.87 -0.026 0.949 0.312 0.932 0.451 0.95 0.982 0.93 -0.02 0.363
0.058 0.919 0.866 0.854 -0.038 0.871 0.486 0.854 0.355 0.862 0.969 0.848 0.143 0.47
0.163 -0.061 -0.001 -0.1 -0.066 0.149 0.136 -0.087 -0.091 0.03 -0.065 0.502 -0.004
0.7 0.809 0.022 0.709 0.641 0.659 0.311 0.715 0.858 0.686 0.337 0.547
0.766 -0.137 0.997 0.027 0.931 0.369 0.939 0.952 0.994 -0.213 0.061
0.415 0.768 0.502 0.731 0.355 0.766 0.85 0.744 0.135 0.491
-0.142 0.402 -0.111 0.025 -0.118 -0.064 -0.15 0.298 0.355
.la, t,o
4~
TABLE 14-X
I'O Correlation coefficient matrix for 18 samples from section 7 (listed in Table 14-111): the point of zero correlation for n = 18 at the 99~ confidence level is i0.582 Element AI203 Fe203 TiO 2 CaO K20 MgO Na~O P205 CO, CT Cc Co~ ST As Ba Ce Cr Cu Hg La Li Mn Nd Ni Sb Se Sr V Y Zn Zr
Ni Sb Se Sr V Y Zn Zr
SiO2 - 0.942 -0.897 - 0.93 -0.104 -0.936 -0.675 - 0.909 -0.144 0.057 --0.786 0.058 -0.801 --0.841 -0.573 -0.939 -0.944 -0.702 0.005 -0.959 -0.721 -0.806 0.531 -0.641 -0.503 -0.019 -0.891 -0.865 -0.817 -0.284 -0.247 -0.903
AI203
Fe203
0.864 0.998 0.841 -0.084 0.098 0.999 0.861 0.645 0.549 0.987 0.791 -0.007 0.325 -0.156 -0.188 0.754 0.811 -0.158 -0.19 0.793 0.858 0.872 0.85 0.451 0.786 0.997 0.842 0.993 0.891 0.591 0.877 -0.013 0.109 0.912 0.919 0.632 0.817 0.775 0.853 -0.443 -0.469 0.587 0.757 0.344 0.618 0.013 0.059 0.961 0.831 0.786 0.883 0.731 0.934 0.128 0.438 0.126 0.418 0.988 0.812
TiO2
CaO
-0.12 0.999 0.648 0.995 -0.073 -0.145 0.763 -0.146 0.799 0.883 0.401 0.997 0.988 0.541 -0.034 0.903 0.58 0.741 -0.417 0.534 0.286 0.014 0.972 0.744 0.687 0.063 0.066 0.995
-0.103 0.469 - 0.17 i 0.578 0.714 --0.002 0.713 -0.172 -0.218 0.379 -0.125 -0.052 0.388 0.114 -0.057 0.399 0. i 9 -0.171 0.46 0.611 -0.091 -0.213 0.325 0.314 0.575 0.556 -0.174
K,O
MgO
0.65 ! 0.99 0.64 -0.041 - 0 . 0 7 5 -0.148 0.616 0.772 0.682 -0.15 0.614 0.809 0.536 0.888 0.584 0.436 0.305 0.997 0.626 0.992 0.64 0.574 0.382 -0.027 -0.184 0.912 0.548 0.605 0.359 0.758 0.447 -0.43 -0.286 0.559 0.377 0.314 0.254 0.017 - 0 . 0 6 0.971 0.634 0.764 0.508 0.715 0.462 0.094 0.025 0.093 0.013 0.993 0.646
Na,O
P:O5
-0.155 -0.137 0.753 -0.139 0.788 0.879 0.325 0.991 0.972 0.459 -0.043 0.888 0.508 0.697 -0.405 0.457 0.201 0.023 0.974 0.688 0.619 -0.02 -0.013 0.997
-0.15 --0.169 -0.151 -0.132 -0.207 0.738 -0.046 0.048 0.72 0.408 0.097 0.764 0.479 -0.334 0.785 0.875 0.011 -0.21 0.598 0.625 0.986 0.949 -0.155
Nd
Ni
Sb
Se
Sr
V
Y
Zn
0.904 2.37E-04 0.394 0.926 0.92 0.853 0.844 0.463
-0.002 0.16 0.812 0.833 0.925 0.906 0.205
0.024 0.005 0.016 0.014 0.002 0.005
0.635 0.619 -0.079 -0.076 0.983
0.963 0.703 0.667 0.684
0.719 0.685 0.632
0.964 -0.021
-0.012
CO,
0.126 1 -0.113 -0.117 -0.161 -0.175 -0.16 -0.159 -0.214 -0.199 -0.204 - 0.216 0.104 -0.144 -0.033 -0.103 -0.127 -0.161 -0.181 -0.151 -0.145 -0.137
CT
0.126 0.971 0.952 0.479 0.745 0.759 0.52 -0.115 0.847 0.377 0.562 -0.312 0.293 0.186 0.031 0.866 0.525 0.601 -0.062 -0.08 0.778
C(
-0.113 -0.118 -0.163 -0.177 -0.162 -0.161 -0.218 -0.199 -0.206 - 0.217 0.103 -0.147 -0.035 -0.099 -0.120 -0.163 -0.183 -0.152 -0.148 -0.139
Corg
0.982 0.519 0.788 0.799 0.559 -0.063 0.896 0.427 0.615 -0.337 0.329 0.195 0.055 0.898 0.565 0.646 -0.025 -0.044 0.812
ST
As
0.435 0.872 0.413 0.869 0.497 0.51 0.935 -0.095 0.314 0.908 0.596 0.415 0.86 0.625 0.724 -0.321 - 0 . 4 8 2 0.329 0.835 0.152 0.805 0.032 - 0 . 0 4 2 0.962 0.363 0.574 0.816 0.621 0.901 -0.088 0.776 -0.096 0.737 0.899 0.345
Ba
0.987 0.555 -0.031 0.91 0.6 0.753 -0.414 0.549 0.315 0.007 0.963 0.763 0.701 0.091 0.089 0.989
Ce
0.637 0.012 0.923 0.672 0.798 -0.452 0.629 0.394 0.1 0.949 0.811 0.768 0.181 0.181 0.974
Cr
Cu
Hg
0.196 0.686 0.078 0.945 0.331 0.676 0.809 0.3 0.805 - 0 . 4 7 6 -0.061 - 0 . 5 2 6 0.917 0.324 0.569 0.883 0.299 0.424 0.015 - 0 . 0 6 3 0.101 0.477 - 0 . 0 8 6 0.909 0.907 0.221 0.808 0.976 0.225 0.803 0.794 0.404 0.23 0.746 0.593 0.205 0.481 - 0 . 0 5 6 0.887
La
0.875 -0.521 0.978 0.907 0.019 0.458 0.963 0.959 0.847 0.824 0.511
Li
-0.42 0.815 0.689 0.024 0.668 0.9 0.888 0.578 0.58 0.706
Mn
-0.469 -0.284 -0.078 -0.357 -0.585 -0.528 -0.382 -0.251 -0.379
~'~"
c~
~ ~
o
O' ~
~
O.
,'-t
t ~
~
(1;)
r~
oe
N" ~..
~
~-~ ,~.
,<
","
og
~'~
~ ' ~
~
~
~'t"j
0.
,,,.
O"
r~
r~
,~
~
~,
o
.o
.o
~
~.
~
Varimax rotated factor scores o o o o
o
Varimax rotated factor scores o o o
.o
Varimax rotated factor scores
=
o
o
/
...
--
o
~ o
~-~.
o
~
o
o
o.i ~
Varimax rotated factor scores o o o o o ~)
I
- ~
o
,,0
Varimax rotated factor scores .o on
-!
""
._.
Varimax rotated factor scores
,
o0
o
~
~
~n
kJ
c~
~
oi
~
o
o
o ~
0,
~
"-
o
~.
o k0
o
~
o ~
o
~
~o
o
~0
~
r.
o~
o
Vanmax rotated factor scores
Varimax rotated factor scores oo o o o o o
0_.
Varimax rotated factor scores o oro o o o ~
~
o
ro
(~
o
--
r0
co
o~
~
b
~
I
o
~ z
--
-
~
k0
~
o
-n
~n
"-
o
-. cn
o
k~
o
~ o~
o
~
o
Vadmax rotated factor scores on
o
o.
~-=~
~ 9~o~ ~
--
Varimax rotated factor scores (~ o o o
o
~/admax rotated factor scores o o o o
-~
~
~
o
o
i
o
m
o
o
o
_.
P~
o-"
o
cn
Varimax rotated factor scores
~,o
~~
Varimax rotated factor scores o o o
o
Varimax rotated factor scores
-,
.,4
0
~_.
o
{/)
._L
o
o _~.
(n (D
,.L
0
I,o L~
424
J.R. Hein et al.
carbonate component (dolomite, calcite, silicified carbonates) composed of carbonate C, Mg, Ca, and Si (_Mn). Factor 5 is tentatively interpret as representing organic matter (and/or sulfide-sulfate phases) hosting elements that include Cu (+__organic C, Zn, Mn Si, Ni, Hg, and Li depending on the data set). Copper correlates only with Zn, but Zn also correlates with Ni, As, and some elements associated with the shale component. Silica is dominantly in the chert fraction (factor 1) and does not appear in the aluminosilicate fraction (factor 3) even though it is clearly part of the feldspars and clay minerals that comprise that fraction. Likewise, organic C is dominantly in factor 3 and does not appear in factor 5 (except for section 1) even though it is possible that the elements in factor 5 are hosted by organic matter. These characteristics are an artifact of analyzing a data set that is overwhelmingly dominated by one variable, silica, and the distribution of some elements in more than one phase. Selenium shows a dominant association with the shale component, but correlations and Q-mode factors also indicate that organic matter (within the shale component) and carbonate fluorapatite may host a portion of the Se. Consideration of larger numbers of factors in Q-mode analysis indicates that native Se (a factor containing Se (_ Ba)) may also constitute a minor component of the Se complement.
DISCUSSION AND CONCLUSIONS: ENVIRONMENTALLY SENSITIVE ELEMENTS Analyses of outcrop samples from the Rex Chert and adjoining members of the Phosphoria Formation in southeast Idaho provide clear evidence that the chert contains low concentrations of environmentally sensitive elements, but that adjacent and transitional rocks may contain high concentrations of those elements. The chert beds have low Se concentrations, <1 ppm, whereas the Cherty Shale Member rocks have somewhat higher Se concentrations, with a mean of 1.8 ppm. Some beds in the siliceous siltstone of the lowermost Rex Chert, which comprises the zone of transition with the Meade Peak, have high Se concentrations, up to 138 ppm. These transitional beds overall comprise a small part of the Rex Chert and would be the only significant source of Se that would be released to the environment during weathering of the Rex Chert Member. The low Se contents determined for the chert-bed samples here are not characteristic of Se concentrations found for composite channel samples taken through the Rex Chert at the Rasmussen Ridge and Enoch Valley Mines (R.I. Grauch and J.R. Herring, unpublished data, 2002). Those channel samples comprise a continuous composite record of Rex Chert composition through the entire section. Those weighted (for stratigraphic thickness) mean Se concentrations are 63 ppm for Rasmussen Ridge samples and 18 ppm for Enoch Valley samples (Table 14-IV). These differences can be explained in several ways: (a) weathering of the outcrops that we sampled; (b) the higher organic-matter contents of the composite samples, which are ninefold higher for Rasmussen Ridge samples and fourfold higher for Enoch Valley samples; (c) higher (two to threefold) CFA (P205) in the composite samples, and (d) the inclusion of shale and siliceous shale beds that may be interbedded with cherts in the analyses of composite samples. Explanation 2 is one of
Composition of Rex Chert Member
425
the likely manifestations of explanation 1. Greater amounts of terrigenous material (explanation 4) in the composite samples is supported by their lower silica and higher A1203, K20, etc., contents (Table 14-IV). In the outcrop sections studied here, shale interbedded with the chert consists only of thin partings, except in the Cherty Shale Member, and those partings would not likely contribute significantly to the mean Se concentration for each outcrop section. It is not known why there are more (or thicker) shale interbeds, or more argillaceous cherts in the composite sections than those studied in outcrops. Chert is very resistant to weathering and little Se should be leached by weathering of the outcrops. However, the most significant factor may be the much higher contents of organic matter and the moderately higher contents of CFA in the composite samples. Organic matter is commonly a host phase for metals, as indicated in our statistical analyses. Organic matter can be readily oxidized and removed during weathering, along with its associated complement of metals. Other elements of environmental interest include As, Cr, V, Zn, Hg, and Cd. Of these elements, only mean Cr and Zn values are higher, 30 and 27% respectively, than their respective values in average shale. Cadmium could not be evaluated because most concentrations are below the limit of quantification of 2 ppm. As with Se, the concentration of these elements in our outcrop samples are lower than they are in the composite samples from the mines. The likely reason is the same as it is for Se, the higher shale component, CFA, and especially the higher organic-matter content of the mine composite samples are the hosts for these elements. This difference in the contents of Se between the two groups of samples suggests to us that the organic fraction is indeed the dominant host of Se. This observation is consistent with that found for Permian organic C-rich, seleniferous chert in China (Yao and Gao, 2002) and for the Miocene siliceous Monterey Formation of California (Piper and Isaacs, 1995). The much lower concentrations of Se in samples of Rex Chert presented here suggest that Se and the organic matter may have been lost through weathering that extended over many years. Thus, freshly exposed outcrop samples of Rex Chert may contain higher concentrations of Se and other elements of environmental concern than samples from outcrops that have existed for extended periods of time. These geochemical relationships are especially pronounced for the lower part of the Rex Chert where rocks are of transitional character with the Meade Peak Member and in the Cherty Shale Member rocks that overlie the Rex Chert. In the outcrop samples analyzed here, those stratigraphic intervals have higher terrigenous fraction and organic matter, Se, and trace metal concentrations than the more highly siliceous Rex Chert.
ACKNOWLEDGMENTS We thank Richard I. Grauch, Margaret A. Keller, and Florence L. Wong for helpful reviews of this chapter. We thank R.I. Grauch and J.R. Herring for kindly providing unpublished data on composite channel samples from the Enoch Valley and Rasmussen Ridge mines. Access to those mine sites is gratefully acknowledged.
426
J R . Hein et al.
REFERENCES Arbogast, B.E (ed.), 1996. Analytical methods manual for the Mineral Resource Surveys Program, US Geological Survey. US Geological Survey, Open-File Report, 96-525, 248 pp. Baedecker, P.A. (ed.), 1987. Methods for geochemical analysis. US Geological Survey, Bulletin, 1770, variously paginated. Brittenham, M.D., 1976. Permian Phosphoria carbonate banks, Idaho-Wyoming thrust belt. In: J.G. Hill (ed.), Symposium on geology of the Cordilleran hingeline. Rocky Mountain Association of Geologists - 1976 symposium, Denver, CO, pp. 173-191. Cressman, E.R. and Swanson, R.W., 1964. Stratigraphy and petrology of the Permian rocks of southwestern Montana: US Geological Survey, Professional Paper, 313-C, pp. 275-569. Govett, G.J.S., 1983. Handbook of Exploration Geochemistry, vol. 3, Rock Geochemistry in Mineral Exploration. Elsevier, Amsterdam, 461 pp. Jackson, L.L., Brown, EW. and Neil, S.T., 1987. Major and minor elements requiring individual determinations, classical whole rock analysis, and rapid rock analysis. In: P.A. Baedecker (ed.), Methods for Geochemical Analysis. US Geological Survey, Bulletin, 1770, pp. G 1-G23. Keller, W.D., 1941. Petrography and origin of the Rex Chert. Geol. Soc. Am. Bull., 52: 1279-1298. Klovan J.E. and Imbrie J., 1971. An algorithm and FORTRAN-IV program for large-scale Q-mode factor analysis and calculation of factor scores. Math. Geol., 3:61-77. Krauskopf, K.B., 1979. Introduction to Geochemistry. McGraw-Hill, New York, 617 pp. Mansfield, G.R., 1927. Geography, geology, and mineral resources of part of southeastern Idaho. US Geological Survey, Professional Paper, 152, 453 pp. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M., and Swanson, R.W., 1959. The Phosphoria, Park City and Shedhorn Formations in western phosphate field. US Geological Survey, Professional Paper, 313-A, 47 pp. Montgomery, K.M. and Cheney, T.M., 1967. Geology of the Stewart Flat quadrangle, Caribou County, Idaho. US Geological Survey, Bulletin, 1217, 63 pp. Murata, K.J., Friedman, I., and Gleason, J.D., 1977. Oxygen isotope relations between diagenetic silica minerals in Monterey Shale, Temblor Range, California. Am. J. Sci., 277: 259-272. Murchey, B.L. and Jones, D.L., 1992. A mid-Permian chert event: widespread deposition ofbiogenic siliceous sediments in coastal, island arc and oceanic basins. Palaeogeogr. Palaeoclimatol. Palaeoecol., 96:161-174. Oberlindacher, H.P., 1990. Geologic map and phosphate resources of the northeastern part of the Lower Valley quadrangle, Caribou County, Idaho. US Geological Survey, Miscellaneous Field Studies Map, MF-2133, scale 1:12,000. Piper, D.Z. and Isaacs, C.M., 1995. Geochemistry of minor elements in the Monterey Formation, California - seawater chemistry of deposition. US Geological Survey, Professional Paper, 1566, 41 pp. Richards, R.W. and Mansfield, G.R., 1912. The Bannock overthrust. J. Geol., 20: 683-689. Sheldon, R.P., 1957. Physical stratigraphy of the Phosphoria Formation in northwestern Wyoming. US Geological Survey, Bulletin, 1042-E, 185 pp. Wardlaw, B.R. and Collinson, J.W., 1984. Conodont paleoecology of the Permian Phosphoria Formation and related rocks of Wyoming and adjacent areas. In: D.L. Clark (ed.), Conodont Biofacies and Provincialism. Geological Society of America Special Paper 196, pp. 263-281. Yao, L. and Gao, Z., 2002. Abnormal enrichment of selenium in Yutangba carbonaceous cherts, southeast Enshi, Hubei Province, China. Abstracts, Goldschmidt Conference 2002, Geochim. Cosmochim. Acta, 66: A859.
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
427
Chapter 15
GASEOUS SELENIUM AND OTHER ELEMENTS IN NEAR-SURFACE ATMOSPHERIC SAMPLES, SOUTHEAST IDAHO P.J. LAMOTHE and J.R. HERRING
ABSTRACT Concentrations of selenium (Se) and other selected elements were determined in air samples taken near the Unit 4 waste-rock pile of the Wooley Valley phosphate mine in southeastern Idaho. The waste-rock pile is approximately 3 million m 3 (4 million yard 3) in volume. Twenty-four-hour air samples were taken in June 2000 at two locations downstream from the waste-rock pile along Angus Creek in Little Long Valley and 48-h air samples were taken in May 2001 at two locations near the toe of the waste-rock pile. The purpose of the study was to determine if Se and other environmentally significant volatile elements that are present in the waste-rock pile are also present in the nearby atmosphere. Three of the sampling sites were in wetland areas near the toe of the waste-rock pile through which drainage travels. The wetland is a candidate source for the release of volatile trace elements including Se from the soil and/or plants. The results indicate that small but measurable elevations of Se concentration (30-90 n g m -3) occur in the near-ground atmosphere at the wetland compared to a nearby background location. No other trace elements had elevated concentrations in the atmosphere at the wetland site.
INTRODUCTION In the Little Long Valley study area in southeast Idaho, the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation occurs as a folded, often steeply dipping unit with elongated surface exposures or near-surface occurrences with north to northwest strike. Depending on the dip of the strata, a typical phosphate mine is several hundred meters deep and several kilometers long. Over the 15-20-year life of a typical mine, 20-40 million metric tonnes of ore are extracted and an amount of waste rock (shale, dolostone, and chert) of two to five times that amount requires disposal. The Meade Peak is approximately 50-55 m (164-180 ft) thick. Its formation has two phosphate ore zones, lower and upper, of approximately 10 m (32 ft) and 5 m (16 ft) thicknesses, respectively. The two ore zones are separated by a middle-waste unit of shale that is approximately 18-20 m (59-66 ft) thick. The stratigraphy and description of measured
428
P.J Lamothe and JR. Herring
sections at each of the four working mines in southeast Idaho are summarized in reports by Tysdal et al. (1999, 2000a,b,c) and Grauch et al. (2001). When the Wooley Valley mine was in operation, the two ore zones were removed and the middle-waste shale and other rocks were removed and placed in, among other waste piles, a cross-valley fill waste pile designated as the Unit 4 waste-rock pile. Elevated concentrations of Se and other environmentally sensitive trace elements (e.g. As, Cd, T1, U) within the middle-waste shale of the Phosphoria (see Herring and Grauch, Chapter 12) raised concerns about the introduction of these trace elements into the ecosystem as a result of mining and shale disposal. A table in Piper et al. (2000) summarized concentrations of Se that occur in various components of the ecosystem in SE Idaho. The purpose of the present study is to determine if a detectable amount of Se in the vicinity of the Unit 4 waste-rock pile of the Wooley Valley mine is being released to the atmosphere in the form of gas-phase Se species.
LOCATION Air samples were collected in June 2000 and May 2001 at two sites within Little Long Valley (Fig. 15-1). The study area lies approximately 32 km (20 miles) northeast of Soda Springs, Idaho, in a region of southeast Idaho that has had extensive phosphate mining over the past several decades and currently has four active phosphate mines. The sampling sites were chosen at locations in the proximity of the headwaters of Angus Creek because these headwaters drain a wetland adjacent to the Unit 4 waste-rock pile of the Wooley Valley phosphate mine. The waste-rock pile is approximately 3 million m 3 (4 million yard 3) in volume (W. Johnson, Bureau of Land Management, 1999, oral communication; wastepile tonnage is estimated at 5-7 metric megatons and the assumed density is about 1.8 metric ton m-3). Furthermore, extensive geochemical sampling and analyses have been performed on solid, aqueous, and plant samples from this locale (Herring et al., 1999, 2000a,b,c; Stillings et al., 2000). Notable enrichments of Se and other trace elements occur in those samples.
STUDY DESIGN It has been well established (Herring et al. 1999, 2000a,b,c) that Se is highly enriched in rocks of the Meade Peak in southeast Idaho. Furthermore, Se is present in the shale as components that are labile under weathering and/or other alteration processes. For example, samples of ground shale when placed in water will release significant quantities of Se and other trace elements in times as short as 1 h (Herring, Chapter 13). The primary purpose of this study was to gather information regarding the extent to which gas-phase Se compounds might enter the atmosphere at a reclaimed phosphate mine waste-rock pile that contains middle-waste shale from the Meade Peak. Note that the study was not designed to be an extensive air monitoring effort covering a large geographic area over an extended
o~ ~~
t~ o~
t~
t~
t..~
Fig. 15-1. Location map of study area showing air sampling sites. The area labeled "Waste dump" is the Wooley Valley Unit 4 waste-rock pile (modified from Stillings et al., Chapter 17). tO
430
PJ. Lamothe and JR. Herring
period of time, nor was it intended to identify specific volatile compounds of Se. Rather, the work focused on obtaining a snapshot of the magnitude of airborne total Se species through several diurnal cycles at sites where geological, biological, and hydrological conditions likely favor the formation of vapor-phase compounds of Se and other volatile trace elements. The wetland site studied is located in a marshy area containing grass, forbs, and willow at the toe of the Unit 4 waste-rock pile. A series of seeps in this area drain from the base of the waste-rock pile and flow ephemerally throughout the wetland. The wetland site was chosen because of its abundant vegetation and presence of anoxic mud (Stillings et al., 2000; Stillings and Amacher, Chapter 17) - parameters that may favor the formation of volatile Se compounds due to the interaction of soil, water, and plants (Hansen et al., 1998). A background site is located 1.5 km (0.9 mile) northwest of the mine waste-rock pile and is downstream along Angus Creek from the wetland site (Fig. 15-1). The background site was chosen because it is in the same drainage as the wetlands site, but at a distance sufficiently far and generally upwind of the wetland site such that presence of any airborne Se species from the wetland should be minimized and the Se concentration at this site should approximate background levels in this geographical area.
SAMPLE COLLECTION AND ANALYSIS Airborne volatile-Se compounds were collected using a conventional air sampling train (Fig. 15-2). The pre-filter was a 47 mm diameter, 0.45 Ixm pore size TELFON| membrane filter. Its purpose was to exclude any solid-phase Se species from entering the sampling train. An identical filter was placed in the sample train directly behind the second bubbling tower to prevent droplets of the reagent liquids from entering the pump. Ambient air was drawn continuously through the sampling train at a flow rate of 0.12 m3h -I (2 Lmin -~) over a period of 24--48 h at each sampling site. A constant flow rate was maintained using a flow-limiting orifice in the sampling train just ahead of the pump. This orifice limits flow to the above number as long as a pressure differential of at least 50 kPa (approximately 0.5 atm.) is maintained across the orifice. Samples were taken on consecutive days from June 20 to 22, 2000, and from May 7 to 11,2001. Weather conditions during sample collections were fair, partly cloudy, with a high temperature of 18-20 ~ during the day and a low of 3-7 ~ during the night. Winds were generally calm (0-3 miles h -l) at night and for most of the day, but gusts of 10-15 miles h-i were encountered during the late afternoon hours preceding sunset. The reagents used in each of the gas scrubbing flasks are the same as those used by previous workers (Zieve and Peterson, 1984; Weres et al., 1989) to collect all volatile Se compounds without regard to specific species. The first gas scrubbing flask in the sampling train contained 100 mL of 6 M Ultrex nitric acid plus 25 mL of 30% Ultrex H202, and the second flask contained 100 mL of 0.05 M NaOH plus 25 mL of 30% H202. The inlet of the sampling train was 0.3 m (1 ft) above the ground for all June 2000 samples.
431
Selenium and other elements in air samples in southeast Idaho Post-filter 0.45 lam pore size
Pre-filter 0.45 lLtm pore size Air inlet
~
Vacuum pump ~
~.~
"~
Flow control orifice
6 M Nitric acid/ 30% H202
0.05 M NaOH/
30% H202
Fig. 15-2. Air sampling train used to collect Se species. Samples were collected at a flow rate ofO.12 m3h -I
For the May 2001 measurements, air samples were collected at a seep at the toe of the waste-rock pile and at a second wetland site 60 m (200 ft) west of the seep. The seep at the toe of the waste-rock pile has a surface area of approximately 10 m 2, and it is the first expression of fluid flowing from beneath the waste pile coming into contact with the atmosphere. We chose to collect air samples at the seep because three separate measurements made in June 1999 revealed that the water in the seep contained from 544 to 725 ~g L-~ Se. Furthermore, this seep lies at the head of a large wetland area encompassing approximately 5000 m 2. Two identical air-sampling trains were employed at each site to simultaneously collect air samples at two vertical heights, 0.3 m (1 ft) and 2.13 m (7 ft) above ground, in order to measure differences in airborne Se concentrations as a function of vertical distance as well as horizontal distance from the theoretical source. Field blanks of each solution were bottled on site and were returned, along with the samples, to the laboratory for chemical analysis. As the complex chemical and biochemical interactions within anoxic mud and those of plant uptake and transpiration can produce a host of different volatile Se-containing compounds, such as elemental or methylated, our experimental design used a combination of strongly oxidizing reagents to permit total recovery of all vaporous Se compounds likely to be present (Hansen et al., 1998). All sample solutions and their corresponding blanks were analyzed for Se and other trace elements using an inductively coupled plasma mass-spectrometry (ICP-MS) technique described by Lamothe et al. (1999). However, only Se yielded results above the lower detection limits of the method.
432
P.J Lamothe and JR. Herring
RESULTS AND DISCUSSION For samples collected in June 2000, the ground-level concentration of Se in the air at the wetland near the waste-rock pile was elevated sevenfold compared to the background site further down valley (Table 15-1). The ground-level sample collected in May 2001 at the seep site (sample 01S1L) showed a 13-fold increase in airborne Se concentrations relative to the value obtained just 60 m (200 fi) away at an elevation of 2.13 m (7 fi) above ground (sample 01W2H). Furthermore, the airborne Se values measured at both sites in the vicinity of the toe of the Unit 4 waste-rock pile decreased with elevation and decreased with horizontal distance from the toe of the pile. These data indicate that small but measurable amounts of Se are entering the atmosphere due to biological and environmental conditions that exist in the wetland areas near the Wooley Valley Unit 4 waste-rock pile. Note that the airborne Se value of 7 n g m -3 measured in May 2001 at a height of 2.13 m and horizontal distance of 60 m from the toe (sample 01W2H) of the waste-rock pile was identical to the background value measured in June 2000 at a distance of 1500 m from the waste pile (sample 00B 1L). This fact demonstrates that the Se emitted by the wetland site at the toe of the pile is rapidly diluted to background concentrations by ambient winds. Unfortunately, few published data are available for Se concentrations in ambient air because earlier studies measured Se volatilization rates from wetlands using open-bottom
TABLE 15-I Concentration of Se in air samples taken at locations in the vicinity of the Wooley Valley Unit 4 waste-rock pile in Little Long Valley, Idaho Sample ID
Date
Distance from toe of waste pile (m)
Height above ground (m)
01S 1L 01S1H 01W 1L 01W2H 00W1L 00B1L
5/9/2001 5/9/2001 5/7/2001 5/7/2001 6/20/2000 6/21/2000
0.0 0.0 60 60 60 1500
0.3 2.13 0.3 2.13 0.3 0.3
4.64 0.52 1.32 0.31 1.12 0.16
23.2 NIST-16402 11.0 NIST-1643d2
True True
QC Summary NIST- 16402 Found NIST- 1643d2 Found
Se in Finalvol. solution of solution (txgL i) (L)
0.110 0.125 0.132 0.138 0.134 0.129 22.0 11.4
Vol. of air Se in air sampled (ng m- 3) (m3)
5.7 5.7 5.7 5.7 2.88 2.88
90 11 31 7 52 7
% Recovery % Recovery
105% 96%
aNIST-1640 and NIST-1643d are aqueous Standard Reference Materials prepared by the National Institute of Standards and Technology that have certified values for selenium concentrations in solution.
Selenium and other elements in air samples in southeast Idaho
433
collection chambers to cover a fixed area of the selected site. For example, Zawislanski and Zavarin (1996) measured Se volatilization rates over a variety of vegetated sites at the Kesterson Reservoir in 1993 and found rates ranging from 50 to 175 p~gm-2 day-1. However, Zawislanski and Zavarin's study site was dominated by cattails and grass species such as Polypogon and Bromus, whereas at our sites, the flora consists of a temperate, mid-latitude, higher altitude (2 km) mix of grasses, forbs, and small trees (Mackowiak et al., Chapter 19). Similarly, Weres et al. (1989) used enclosures over plants at the Kesterson Reservoir and obtained Se volatilization rates ranging from 24 to 190 i~gm -2 day -l. Zhang and Moore (1997) obtained rates of Se volatilization of 1-50 p~gm -2 day-l from microcosms consisting of Se-containing water (3 p~g L - l Se), sediment (1-7 p~gkg-l Se), and plant materials collected from a wetland system in Montana. Their results showed that plants and sediments were the major producers of volatile Se and that volatilization rates increased with increasing temperature, airflow rate, and decomposition of wetland plants. Considering that the highest airborne Se concentration we found (sample 01S 1L) was at the edge of a seep that has a surface area of approximately 10 m 2, and allowing for the diluting effect of ambient winds because we did not place enclosures over the site, the airborne Se concentrations measured are certainly consistent with Se volatilization rates determined by other investigators. To illustrate, if we use a Se volatilization rate of 50 p~gm-2 day -~, which is within the range of values measured by the studies cited above, then we would expect that 500 ~g Se was being emitted each day from the 10 m 2 seep site where sample 01S 1L was collected. The total amount of Se we collected in our air sampling train at site 01SIL over a 48-h period was 0.5 p~g, which equals 0.25 p~g day-l Se. This indicates that we collected approximately 0.05% of the total daily amount of Se being emitted by the s e e p - a fact that is reasonable considering that we used a point sampling approach and not an area sampling approach. However, it must be emphasized that our study was designed to measure average airborne Se concentrations in a system open to the ambient air whereas earlier studies were designed to measure Se volatilization rates from closed systems. Therefore, it is impossible to make direct comparisons of our results to these earlier studies because detailed site-specific atmospheric data were not available for our study area. It is interesting to note that the highest Se volatilization rate reported in the literature for Se-laden wetland sites is 190 p~g Se m -2 day-~ (Weres et al., 1989). Assuming that the total surface area of all wetlands in the vicinity of the Wooley Valley Unit 4 waste-rock pile is approximately 5000 m 2, then the total amount of Se released to the atmosphere from this wetland area is likely to be less than 350 g year-I.
CONCLUSIONS Clearly, volatile-Se species exist in the atmosphere directly above the wetlands at the study site at concentrations ranging from 30 to 90 ng m -3. Other environmentally sensitive trace elements that were analyzed for, but found to be below the limits of detection were: Ag, As, Cd, Co, Cr, Cu, Fe, Mn, Mo, Ni, Pb, Sb, Sn, T1, V, and Zn.
434
P j Lamothe and JR. Herring
We conclude that (a) the wetland at the toe of the Wooley Valley Unit 4 waste-rock pile is acting as a source of volatile Se species, and (b) that the volatilization rate of Se from the wetland is low and that ambient winds quickly dilute the volatile-Se emission down to a background concentration of 7 ng Se m -3.
ACKNOWLEDGEMENTS The samples were collected on the site of the Wooley Valley phosphate mine (inactive), operated by Rhodia Inc., near the Rasmussen Ridge phosphate mine, operated by Agrium US Inc. We thank Rhodia Inc. and Agrium US Inc. for providing access.
REFERENCES Grauch, R.I., Tysdal, R.G., Johnson, E.A., Herring, J.R. and Desborough, G.A., 2001. Stratigraphic section and selected semiquantitative chemistry, Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-20-E, 1 plate with text. Hansen, D., Duda, P.J., Zayed, A. and Terry, N., 1998. Selenium removal by constructed wetlands: role of biological volatilization. Environ. Sci. Technol., 32: 591-597. Herring, J.R., Desborough, G.A., Wilson, S.A., Tysdal, R.G., Grauch, R.I. and Gunter, M.E., 1999. Chemical composition of weathered and unweathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation - A. Measured sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-147-A, 24 pp. Herring, J.R., Wilson, S.A., Stillings, L.A., Knudsen, A.C., Gunter, M.E., Tysdal, R.G., Grauch, R.I., Desborough, G.A. and Zielinski, R.A., 2000a. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation - B. Measured sections C and D, Dry Valley, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-147-B, 34 pp. Herring, J.R., Grauch, R.I., Desborough, G.A., Wilson, S.A. and Tysdal, R.G., 2000b. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation - C. Measured sections E and F, Rasmussen Ridge, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-147-C, 35 pp. Herring, J.R., Grauch, R.I., Desborough, G.A., Wilson, S.A. and Tysdal, R.G., 2000c. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation- D. Measured Sections G and H, Sage Creek area of the Webster Range, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-147-D, 38 pp. Lamothe, P.J., Meier, A.L. and Wilson, S., 1999. The determination of forty-four elements in aqueous samples by inductively coupled plasma- mass spectrometry. US Geol. Survey, Open File Report, 99-151, 14 pp. Piper, D.Z., Skorupa, J.P., Presser, T.S., Hardy, M.A., Hamilton, S.J., Huebner, M. and Gulbrandsen, R.A., 2000. The Phosphoria Formation at the Hot Springs Mine in Southeast Idaho: a source of selenium and other trace elements to surface water, ground water, vegetation, and biota. US Geol. Survey, Open File Report, 00-050, 73 pp.
Selenium and other elements in air samples in southeast Idaho
435
Stillings, L.L., Amacher, M.C. and Herring, J.R., 2000. Selenium transport through a wetland, Caribou National Forest, southeast Idaho (abstract). Geological Society of America meeting, Abstracts with Programs, vol. 32, p. A 191. Tysdal, R.G., Johnson, E.A., Herring, J.R. and Desborough, G.A., 1999. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation, central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-20-A. Tysdal, R.G., Herring, J.R., Desborough, G.A., Grauch, R.I. and Stillings, L.A., 2000a. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Dry Valley, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-20-B. Tysdal, R.G., Grauch, R.I., Desborough, G.A. and Herring, J.R., 2000b. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation, east-central part of Rasmussen Ridge, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-20-C. Tysdal, R.G., Herring, J.R., Grauch, R.I., Desborough, G.A. and Johnson, E.A., 2000c. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Sage Creek area of Webster Range, Caribou County, Idaho. US Geol. Survey, Open File Report, 99-20-D. Weres, O., Jaouni, A. and Tsao, L., 1989. The distribution, speciation and geochemical cycling of selenium in a sedimentary environment, Kesterson Reservoir, CA, U.S.A. Applied Geochem., 4: 543-563. Zawislanski, P.T. and Zavarin, M., 1996. Nature and rates of selenium transformations; a laboratory study of Kesterson Reservoir soils. Am. J. Soil Sci., 60:791-800. Zhang, Y. and Moore, J.M., 1997. Environmental conditions controlling selenium volatilization from a wetland system. Environ. Sci. Technol., 31:511-517. Zieve, R. and Peterson, P.J., 1984. Volatilization of selenium from plants and soils. Sci. Total Environ., 32: 197-202.
This Page Intentionally Left Blank
Li[e Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
437
Chapter 16
SELENIUM LOADING THROUGH THE BLACKFOOT RIVER WATERSHED: LINKING SOURCES TO ECOSYSTEMS
T.S. PRESSER, M. HARDY, M.A. HUEBNER and P.J. LAMOTHE
ABSTRACT The upper Blackfoot River watershed in southeast Idaho receives drainage from 11 of 16 phosphate mines that have extracted ore from the Phosphoria Formation, three of which are presently active. Toxic effects from selenium (Se), including death of livestock and deformity in aquatic birds, were documented locally in areas where phosphatic shales are exposed (Piper et al., 2000; Presser et al., Chapter 11). Current drainage conditions are leading to Se bioaccumulation at concentrations that pose a risk to fish in the Blackfoot River and its tributaries (Hamilton et al., Chapter 18). A gaging station on the Blackfoot River was re-activated in April 2001 to assess hydrologic conditions and concentration, load, and speciation for Se discharges on a watershed scale. The gaging-station data are considered to represent regional drainage conditions in the upper Blackfoot River watershed because of its location near the outlet of the watershed and directly upstream of the Blackfoot Reservoir. Watershed discharges for 2001 and 2002 were below minimum hydrologic conditions for the gage as documented by the historical record. Drought emergencies were declared in the area in both 2001 and 2002. Unmonitored diversions for irrigation that routinely take place during the snowmelt season also affected conditions downstream. Annual cycles in Se concentration, load, and selenate (Se6+) reached maxima in the spring during the period of maximum flow at the gaging station. Thirty-seven to 44% of annual flow occurred during the three-month high-flow season (April through June) in 2001 and 56% of annual flow occurred during that time period in 2002. Extrapolation from historical hydrographs for average and wet years and a limited data set of regional Se concentrations for 2001 and 2002 indicated potential for a 3.6- to 7.4-fold increase in Se loading because of increased seasonal flows in the Blackfoot River watershed. Supplementation data indicate that: (a) the difference between total Se and dissolved Se, as a measure of the contribution of particulate Se, was < 10% except at the peak of concentration when total Se was 18% more than dissolved Se; (b) selenite (Se 4+) represented less than 10% of the dissolved species during all months of 2001; and (c) dissolved Se was approximately a 50:50 mixture of selenate and organic selenide (operationally defined Se 2-) during summer 2001 (June through August).
438
T.S. Presser et al.
Ecological risk based on regional Se drainage occurred during both the high- and low-flow seasons. Seventy to 83% of the Se load occurred during the high-flow season. During early May of both years, dissolved-Se concentrations exceeded the criterion for the protection of aquatic life and the ecological threshold of 5 Ixg L-1 Se at which substantive risk occurs. During the majority of the three-month high-flow season, dissolvedSe concentrations exceeded the 2 IxgL -l Se concern level for aquatic biota. The Se concentration in suspended material during high flow in 2002 was within the range of marginal risk to aquatic life (2-4 Ixg g - ' Se, dry weight). Selenate was the major species during peak flows, with both selenate and organic selenide being major species during relatively low-flow periods in summer. A change in speciation to reduced Se may indicate elevated biotic productivity during summer months and could result in enhanced Se uptake in food webs. In addition to the magnitude of regional Se release in the Blackfoot River watershed, Se concentrations in individual source drains and waste-rock seeps, and those predicted by experimental column leaching of proposed mining overburden materials, also indicate that drainage options that currently meet existing demands for phosphate mining cause ecological risk thresholds to be exceeded. At times, the drinking-water Se standard (50 Ixg L - ' Se) and the criterion for hazardous Se waste (1000 IxgL-' Se) (US Department of the Interior, 1998; US Environmental Protection Agency, 1987) are also exceeded. For water-years 2001 and 2002, seasonal increased input of water in the mining area resulted in increased Se transport, suggesting a mechanism of contamination that involves a significant Se reservoir. Hence, recognition and monitoring of Se loading to the environment on a mass balance basis (i.e. inputs, fluxes and storage within environmental media, and outputs) are essential to evaluating how to control Se concentrations within environmentally protective ranges (Presser and Piper, 1998). In areas where release of Se to aquatic systems is anticipated as a product of future expansion of phosphate mining, continuous monitoring of flow and development of seasonal Se loading patterns would help to model watersheds in terms of sources, flow periods, and environmental-Se concentrations that most influence bioavailability. These data, in turn, could be linked to Sebioaccumulation models specific to food webs and vulnerable species of the impacted areas to accurately project ecological effects. Gaging at this site on the Blackfoot River is planned to continue in order to establish a long-term (> 10 year) record of hydrologic conditions.
INTRODUCTION Accurate forecasting of the environmental fate of Se is crucial because Se released from irrigation drainage, power production, refining, and mining has caused: (a) fish mortality and deformities in waterfowl; (b) posting of human health advisories for consumption of contaminated food; and (c) termination of grazing (Trelease and Beath, 1949; Presser, 1994; Skorupa, 1998; Hamilton, 1998; Lemly, 2002). Selenium is enriched in organic-rich marine shales that are source rocks for phosphorites and petroleum and is
Selenium loading through the Blackfoot River watershed
439
secondarily enriched in soils and runoff derived from weathering of shales in many semiarid regions of the western United States (Presser et al., 1994; Presser et al., Chapter 11). Long-term assessment of Se discharges associated with the regional geology of Seimpacted areas quantifies reservoirs of Se and allows modeling of exposure and risk to fish and wildlife. In addition to traditional variables such as loads and water column concentrations, assessment of biological effects in ecosystems depends on understanding Se speciation, partitioning of Se in sediment and water, Se bioaccumulation in food webs, and trophic transfer of Se to higher predators (Luoma and Presser, 2000). The upper Blackfoot River watershed, northeast of Soda Springs, has the highest density of phosphate leasing areas and areas for expansion of phosphate mining in southeast Idaho (US Department of Interior et al., 2000). The upper Blackfoot River watershed receives drainage from 11 existing mines (Ballard, Champ, Conda, Dry Valley, Enoch Valley, Henry, Lanes Creek, Maybe Canyon (North and South), Mountain Fuel, Rasmussen Ridge, and Wooley Valley) and their associated mining disturbances (e.g. open pit mines, overburden dumps, stockpiles, processing facilities). Dry Valley, Enoch Valley, and Rasmussen Ridge are the only active mines in the watershed. Areas of possible expansion in the Blackfoot River watershed by 2015 include: Rasmussen Ridge, Lanes Creek, Enoch Valley, Maybe Canyon, and Dry Valley (US Department of Interior et al., 2000). New mine sites, mine sites under environmental review, or under phosphate lease include the areas of Caldwell Canyon, south Rasmussen Ridge, Dairy Syncline, and Trail Creek. The total surface area of mining disturbance is estimated at 32.4 km 2 or 3.6% of approximately 910 km 2 in the watershed. In terms of volume, individual dumps contain 6-70 million tons (Mt) of waste rock that is contoured into hills or used to fill valleys (e.g. cross-valley fill) and pits. Streams and drains issuing from mined waste rock and forage grown to stabilize waste rock provide pathways for dispersal of Se in the environment, which lead to toxic effects in livestock, aquatic birds, and fish (Piper et al., 2000; Presser et al., Chapter 11; Hamilton, Chapter 18; Mackowiak et al., Chapter 19). In addition, a risk assessment performed in 2002 showed that Se affected by far the greatest number of biological receptors throughout three evaluated watersheds (Idaho Department of Environmental Quality, 2002). Those watersheds were the Blackfoot River/Little Blackfoot River, the Salt River (drainage from Smokey Canyon Mine), and the Georgetown River (drainage from Georgetown Mine). Hazard was identified with potential ingestion by humans of fish from the Blackfoot River watershed. Twenty-seven percent of the stream segments in the Blackfoot River watershed were estimated as impacted by Se when assessed for human-health risk (Idaho Department of Environmental Quality, 2002). Ecological risk to both avian and mammalian terrestrial receptors was highest in the Blackfoot River/Little Blackfoot River watershed (Idaho Department of Environmental Quality, 2002). The extremely high hazard quotients (>2000; quotients > 1 are indicative of risk) for some terrestrial receptors indicated that there is a high probability of significant risk occurring in local areas. Aquatic receptors were not as extensively evaluated for risk (i.e. the potential risk to aquatic receptors could not be ruled out), but the average Se concentration for impacted area surface water was estimated at 67 txg L-l Se (Idaho Department of Environmental Quality, 2002).
440
T.S. Presser et al.
This average exceeded the criterion for protection of aquatic life (5 i~gL -1 Se; US Environmental Protection Agency, 1987) by a factor of 13.4. A maximum concentration of Se in impacted surface water for the risk assessment was estimated at 1140 txg L - l Se, a level above the hazardous water-extracted Se waste criterion (1000 IxgL -1 Se, US Department of Health and Human Services, 1996). A maximum Se concentration reported for sediment (188 Ixg g-1 Se, dry weight or dw) exceeds the hazardous Se waste criterion of 100 Ixg g- 1 Se, wet weight, for a solid waste if it is assumed the sediment contained 40% moisture and this criterion applied to mining waste (US Department of Health and Human Services, 1996). The risk assessment also used an average Se concentration of 19.7 Ixg g-1 Se (dw) for impacted sediment (Idaho Department of Environmental Quality, 2002), which is approximately fivefold above the 4 txg g-1 Se (dw) ecological substantive risk threshold concentration (US Department of Interior, 1998). In terms of effects on the environment, a recent compilation of sampling results in the vicinity of phosphate mine sites in southeast Idaho showed, in general, "elevated levels of Se in virtually every media and species of wildlife tested." (Idaho Department of Environmental Quality, 2002). Areas of concern specifically noted were as follows. Elk: "Elk surveys conducted by Idaho Fish & Game and Idaho Mining Association in 1999
show a direct correlation between elevated concentrations of Se in elk liver versus the distance of harvested elk from the nearest phosphate mine. Approximately 70% of the elk harvested within 2 miles of historic reclaimed mining areas exhibit elevated Se accumulations in their organs. These results also demonstrate upper range subpopulation elk-liver accumulations (~38 mg Se kg-l dw) approaching referenced toxic threshold liver concentration ranges (20--60 mg Se kg -l dw) for other large mammals in a limited number of the individual elk tested." Birds: "... Over 10% of the 117 bird eggs collected in the mining areas exceeded 10 mg kg-1
(dw) egg-shell Se concentration even though the collection effort appears to have missed a significant portion of the higher impacted riparian zones. The mean egg Se (MES) effects threshold (EC 10) for Se is reported to occur between 6 and 16 mg kg-1, however, effects are universally accepted to be both site- and species-specific .... " Fish: "... Whole body fish samples collected by Idaho Department of Environmental
Quality in impacted stream segments also ranged up to 33 mg kg -1 (dw) Se as compared to typical reported background levels of 1--4 mg kg-1., Note that the substantive risk threshold for fish occurs at >6 mgkg-I Se, whole-body, dw (US Department of Interior, 1998). Small Mammals: "Small mammal whole body sample concentrations in impacted areas
were up to 7 mg kg-1 (wet weight or ww) with reported background levels typically from 1-4 mg kg- 1 ( d w ) . . . " Vegetation: ".... The former practice of increasing forage productivity on reclaimed sites has inadvertently resulted in providing an enhanced pathway for wildlife exposure .... All reclaimed waste rock piles are exhibiting vegetation concentrations well in excess of the typical 5 mgkg -1 grazing recommendations... The Agency (Idaho Department of Environmental Quality) considers continued livestock grazing losses of the magnitude observed in the past to be unacceptable .... "
Selenium loading through the Blackfoot River watershed
441
In this chapter, we present an analysis of field data for the Blackfoot River gaging station above the Blackfoot Reservoir to evaluate Se loading from the upper Blackfoot River watershed. This approach to assessing a regional source of Se in southeast Idaho addresses: (a) the geologic inventory of Se available as a source of influx; and (b) natural drainage as a source of effiux. How Se loads are delivered in terms of dissolved Se concentration, speciation, and Se concentrations in particulate materials are also assessed to help evaluate Se bioavailability for foodwebs. The watershed parameters are compared to: (a) past and current hydrographs for the Blackfoot River; (b) Se guidelines and ecological thresholds that represent risk to aquatic life; and (c) data compiled from environmental assessments. These provide perspective on the release potential of Se from phosphate mining waste rock and its ability to cause adverse environmental impacts. Loading scenarios, which are based on seasonal and climate-year type (wet, dry, or average) discharges, are developed in an effort to help define a comprehensive approach to control environmental-Se concentrations within environmentally protective ranges.
SITE LOCATION AND DESCRIPTION In April 2001, the US Geological Survey (USGS) reactivated a gaging station on the Blackfoot River above the Blackfoot Reservoir (USGS gaging station #13063000, Fig. 16-1) to provide continuous-flow measurements. Operation of the gage is to continue to obtain a 10-year record of hydrologic conditions. Reports of Water Resources Data for Idaho, which now includes station #1306300, are published annually as part of the USGS Water-Data Report Series (e.g. O'Dell et al., 2001). Real-time data for the gaging station are available at http://idaho.usgs.gov. The station records drainage from an area of approximately 910km 2 in Caribou County. The gaging-station data are considered to represent regional-drainage conditions in the upper Blackfoot River watershed because of its location near the outlet of the upper watershed and directly upstream of the Blackfoot Reservoir. The drainage area is designated as the upper Blackfoot River watershed, hydrologic unit 17040207 (Fig. 16-1) (O'Dell et al., 2001). The upper watershed is shown differentiated from the lower Blackfoot River watershed and the Blackfoot Reservoir (Fig. 16-1). Both the Blackfoot River and Little Blackfoot River drain to the reservoir (Fig. 16-1). The Blackfoot River is the largest tributary to the Blackfoot Reservoir. The reservoir has a storage capacity of 509 million cubic meters (Mm3). Storage in the reservoir began in the spring of 1910 and drainage from mining activities in the watershed began in 1920. The Blackfoot River is a tributary of the Snake River and the watershed ultimately drains to the Pacific Ocean through the Columbia River. The gaging station is located downstream of ten mines in the upper Blackfoot River watershed (Ballard, Champ, Conda, Dry Valley, Enoch Valley, Lanes Creek, Maybe Canyon (North and South), Mountain Fuel, Rasmussen Ridge, and Wooley Valley) (Figs 16-1 and 16-2). The eleventh mine in the upper watershed (Henry Mine) drains downstream of the gaging station, as do parts of the Enoch Valley and Wooley Valley mines (Fig. 16-2). Other
442
T.S. P r e s s e r et al.
Fig. 16-1. The Blackfoot River watershed (upper and lower) and location of US Geological Survey Blackfoot River gaging station (#13063000). Generalized locations of I I phosphate mines are shown: Ballard (1), Champ (2), Conda (3), Dry Valley (4), Enoch Valley (5), Henry (6), Lanes Creek (7), Maybe Canyon North (8.1), Maybe Canyon South (8.2), Mountain Fuel (9), Rasmussen Ridge (10), and Wooley Valley (1 l). The gaging station is located downstream of l0 mines.
v=
Upper Slug C r e e k ~ Lower Slug Creek
Mountain Fuel Mine "=v Goodheart Creek - ~ 1
Champ Mine
~o
East Mill Creek
North Maybe Mine
'.=
South Maybe Mine
,.= French Drain
Dry Valley Mine
v
~ Spnng Creek ~
Maybe Creek
d :1
Dry Valley Creek v
25% Rasmussen Ridge M i n e ~ >75% Wooley Valley Mine ~ ~ ! 75% Enoch Valley Mine
Blackfoot River (above reservoir)
Angus Creek
~,~ ~
French Drain
,,-- State Land Creek
75% Rasmussen Ridge Mine---I~
Sheep Creek
.J
50% Conda Mine
Lanes Creek Mine Trail Creek
Lanes Creek
m,=
"l
;I
Ballard Mine
il
Henry Mine 25% Enoch Valley Mine <25% Wooley Valley Mine 25% Gay Mine
~
Lincoln Creek
75% Gay Mine
~
Ross Fork
Smoky Canyon Mine Smoky Canyon Mine
,.J Sage Cree~ ~1 v I Pole Creek
Georgetown Mine
"='
~1 Georgetown Creek Montpelier Mine
vI
Diamond Gulch Mine
~
Dry Canyon Creek
50% Conda Mine
~
Formation Creek
Little Blackfoot River
Snake River
=9 Blackfoot River
v
(mid reservoir) h.
Blackfoot River (below reservoir)
=9 Portneuf River
,d ~1 Sage Creek
"1
b. w-
Salt River
h=,
;I BearRiver :1
=9 Great Salt Lake
y
r I
Fig. 16-2. A hierarchy of streams and the phosphate mines that affect drainage quality in southeast Idaho. The mining disturbance drained to each creek is generalized to 25% increments. The Blackfoot River, Little Blackfoot River, Salt River, and Portneuf River watersheds drain to the Snake River, while the Bear River drains to the Great Salt Lake.
4~ L~
444
T.S. Presser et aL
watersheds impacted by phosphate mining in southeast Idaho are: Bear River, Portneuf River, and Salt River (Fig. 16-2). The Conda Mine drains both to the Blackfoot River and Bear River watersheds (Fig. 16-2). The Bear River ultimately drains to the Great Salt Lake, Utah, while the Portneuf River and Salt River drain to the Snake River (Fig. 16-2). Previous periods of record for the gaging station were April 1914 to September 1925 (from 5.3 to 6.4 km downstream of the current site) and August 1967 to September 1982 (at the current site). No winter records are available for the period 1914-1925 except for 1915. The gaging station is located at latitude 42.816 ~ and longitude 111.506 ~ at 1908 m above sea level. The gage is on the fight bank 21 m upstream from the railroad bridge; immediately upstream from the Monsanto Chemical Company Haul Road; 8 km upstream from Blackfoot Reservoir flow line; 9.6 km south of Henry; and 17.7 km north of Soda Springs. Diversions for irrigation take place upstream of the gaging station. A 1966 survey showed irrigation of approximately 18.2 km 2 of land (O'Dell et al., 2001). The volume of diversions has not been estimated.
METHODS Flow
River-stage measurements were made using a submersible pressure transducer. Data were logged every 15 min using a Sutron data-collection platform. Manual measurements of discharge (Rantz et al., 1982) were related to stage measurements to obtain a continuous record of discharge (Kennedy, 1983; Buchanan and Somers, 1984). Personnel visited the site approximately every six weeks, and more frequently during snowmelt, to service gage equipment and make manual measurements of stream discharge. The logged data are transmitted to US Geological Survey computers every 4 h via the Geo-stationary Operations Environmental Satellite (GOES) for posting as real-time data (http://idaho.usgs.gov).
S a m p l e collection Water
More frequent sampling occurred during snowmelt when the majority of discharge occurred. A combination of samples collected manually and with an automated sampler were used to document the water year, depending on access to the gage and snow conditions. Samples were collected by manual methods when crews were on site and by an automatic sampler at times when crews could not be on site often enough to document rapidly changing hydrologic conditions. Pairs of samples collected manually and with an automated sampler also served to document that the placement of the sampler intake was adequate for representing integrated samplings across the stream channel and that
Selenium loading through the Blackfoot River watershed
445
contamination was not introduced by the sampling method. Field-equipment and field blanks were collected using USGS certified inorganic-free water (USGS Water Quality Field Supply Unit, Ocala, Florida) to determine the potential for contamination of samples by equipment and processing of manual and automatic samples. Duplicate manual samples were submitted to the laboratories to help define analytical precision. Manual samples were collected using an equal-width-increment method to integrate sampling across the stream width and processed using a modified clean-sampling procedure, which includes further homogenization using a churn (Wilde et al., 1999). Analyses designated as "manual dissolved" were filtered onsite through 0.45 Ixm capsule filters that were certified free of trace-element contamination (USGS Water Quality Field Supply Unit, Ocala, Florida). Dissolved Se-species samples were preserved with high purity hydrochloric acid to a pH less than 2. Dissolved cation and trace-metal samples were preserved with high purity nitric acid to a pH less than 2. Anion samples were not acidified. At the time of sample collection, in-stream measurements of temperature, pH, dissolved oxygen, and specific conductance were made at the center of flow using calibrated electrometric meters. From October 2000 through September 2001, alkalinity measurements were made immediately after sample collection using a fixed-endpoint titration (Radtke et al., 1998). After September 2001, alkalinity measurements were made at the USGS National Water Quality Laboratory (Fishman and Friedman, 1989). Several samples collected manually during low flow were lost during shipment to laboratories. An automatic sampler (ISCO Model 3700) was installed only during the April-October period when freezing conditions did not exist. The automatic sampler and the streamdischarge measurement equipment are co-located in the same housing. This sampler collected point samples through an intake located in flowing water near the north side of the stream. All components of the auto-sampling system were cleaned with detergent and acid solutions followed by flushing with deionized water prior to each installation. The sampler was programmed to collect frequent samples during snowmelt to help define water-quality changes in response to flow variations in the stream. Samples were retrieved from the sampler during routine crew visits, capped, and stored in darkness until the end of runoff. Selected samples based on timing and duration of hydrograph peaks were then processed and submitted for laboratory analyses. Total Se samples were preserved with high purity hydrochloric acid to a pH less than 2. Analyses designated as "automatic dissolved" were filtered at the time of selection through 0.45 txm capsule filters certified to be free of traceelement contamination (USGS Water Quality Field Supply Unit, Ocala, Florida). Preservation protocols used for dissolved species are the same as those described for manual sampling.
Sediment
Suspended-sediment samples were collected on April 10 and 17, 2002 at the gaging station for a preliminary assessment of sediment loading. Steam samples were pumped with a peristaltic pump through the automatic sampler intake into 19 L plastic carboys.
446
T.S. Presser et al.
This method was used instead of the standard equal-width increment or equal-discharge increment integrated sampling methods (Horowitz et al., 1989) because of the relatively low amount of suspended material expected to be transported by the river at the gaging site during drought conditions and the relatively large quantity of suspended material needed for analysis. Pumping through the automatic sampler intake for a long period of time was necessary to obtain enough material for processing and analysis. Intensive sampling throughout the snowmelt period and depth-integrated cross-sectional samples would provide a more accurate assessment of suspended-sediment concentrations than that given here.
Analysis
Samples were analyzed for dissolved Se, total Se, dissolved Se speciation (selenate, selenite, and organic Se), quantity of sediment, and suspended-material Se. Other analyses included dissolved major cations and anions and dissolved trace elements. Only the results of the Se analyses are reported here. Selenium in water samples was determined using inductively coupled plasma-mass spectrometry (ICP-MS) (Lamothe et al., 1999) and hydride generation atomic-adsorption spectrophotometry (AA-hydride) (Cutter, 1978, 1983). Determination of dissolved Se speciation was by differential digestion and AA-hydride (Cutter, 1978, 1983). Suspended-sediment samples were recovered using a high-speed centrifuge and decanted (Horowitz et al., 1989). Samples were air-dried for weighing. Selenium in sediment was analyzed by AAhydride. (XRAL Laboratories, http://www.sgs.ca/xral/method/methods.html).
RESULTS Quality assurance and quality control
Records for the Blackfoot River gaging station are rated as "good" (i.e. approximately 95% of daily discharges are within 5% of the true measurement) (Meyers, 1996; O'Dell et al., 2001). The stage-discharge relation or station rating relating the water-surface elevation (gage height or stage) in the fiver to the volume of water flowing in the channel per unit time is periodically updated (http://idaho.usgs.gov). Numerous quality-control measures were conducted to ensure Se data quality. The close agreement of dissolved-Se analyses for pairs of samples collected manually and with an automated sampler documented that (a) the placement of the sampler intake was adequate for representing integrated samplings across the stream channel and (b) contamination was not introduced by the sampling method (Table 16-I). Duplicate samples within manual- and auto-sample categories showed that adequate analytical precision was achieved during sampling conditions (Table 16-I). Field-equipment blanks, automaticsampler blanks, and atmosphere blanks were below detection limits for Se analysis
Selenium loading through the Blackfoot River watershed
447
T A B L E 16-I Quality assurance and quality control for dissolved and s u s p e n d e d - s e d i m e n t
selenium analyses.
S a m p l e s c o l l e c t e d d u r i n g 2001 a n d 2 0 0 2 at the B l a c k f o o t R i v e r g a g i n g station, e x c e p t as n o t e d . E n t r i e s i n d i c a t e d w a t e r - d i s s o l v e d Se (Ixg L - l )
Manual
Manual
Auto
Auto
Methodology
Independent
ICP-MS
duplicate
ICP-MS
duplicate
duplicate
analysis (ICP
ICP-MS
AA-hydride
ICP-MS
- h y d r i d e or AA-hydride
W a t e r y e a r 2001 4/18/01
-
-
1.4
1.5
-
-
4/19/01
1.5
-
1.3
-
1.1
-
5/01/01
7.4
-
7.3
-
5.0
5/11/01
4.6
-
4.6
-
4.0
-
5/17/01
3.9
4.0
3.3
3.3
3.2/3.2
-
5/15-18/01
.
.
.
.
.
2.61
5/21/01
3.6
-
3.4
-
2.2
-
5/31/01
2.5
-
2.8
-
1.8
-
6/07/01
2.6
-
1.8
6/12-16/01
.
-
2.4
.
.
.
.
2.3 l
6/19/01
2.5
-
2.2
-
1.4
-
8/23/01
1.3
1.3
1.5
-
1.1/0.7
-
9/14/01
1.0
-
0.7
9/18-21/01
.
-
-
.
.
.
.
1.61
Water year 2002 -
-
-
0.6
2/21/02
12/21/01
1.8
1.7
-
0.4/0.4
4/10/02
2.0
-
2.0
4/17/02
3.1
-
3.4
5/8/02
D
3.3
0.6/0.6 0.7/0.9/1.3/0.7
7.4
-
7.2
0.7/1.5
5/10/02
-
-
7.6
2.0
5/11/02
-
-
-
3.4
-
3.3
6/5/02
7.0 2 0.7/0.7
S u s p e n d e d - s e d i m e n t Se (Ixg g - l , d r y w e i g h t ) Lab # 1 AA-hydride
L a b #2 A A - h y d r i d e
4/10/02
3.2
2.6
4/17/02
3.2
2.2
l l n d e p e n d e n t v e r i f i c a t i o n : w w w . d e q . s t a t e . i d . u s / n e w s / m a y 2 3 _ 0 2 . h t m , s a m p l i n g site o n B l a c k f o o t River, a p p r o x i m a t e l y 3.2 k m u p s t r e a m o f g a g i n g s t a t i o n (i.e. B l a c k f o o t R i v e r u p s t r e a m o f D r y V a l l e y C r e e k inflow). 2 I n d e p e n d e n t v e r i f i c a t i o n : M o n t g o m e r y W a t s o n , 2 0 0 2 at B l a c k f o o t R i v e r g a g i n g station.
448
T.S. Presser et aL
methodologies. Standard reference materials used in the analysis of dissolved Se determined by ICP-MS were within 10% of expected (true) values. Samples analyzed for Se analysis by ICP-MS for Se were also analyzed by AA-hydride (Table 16-1). During 2001, trends seen in the data produced by the two methods were similar. However, the maximum value determined by AA-hydride was 28% lower than that determined by ICP-MS. Lower values deviated less. Independent sampling as part of regional assessments by mining companies conducted at a site near the gaging station showed Se concentrations comparable to both methodologies used here during periods of concentration in the river below 3 l~g L- 1 Se (Idaho Department of Environmental Quality, 2001; www.deq.state.id.us/news/may23_02.htm). During 2002, Se concentrations determined by AA-hydride were significantly lower than those determined by ICP-MS, although a peak during snowmelt was observed. Independent sampling as part of regional assessments by mining companies conducted at the Blackfoot River gaging site (Montgomery Watson, 2002) showed a Se concentration similar to the maximum concentration at the hydrograph peak determined by ICP-MS (7.0 Ixg L-1 Se on May 11, 2002 vs 7.6 I~g L -1 Se on May 10, 2002). The data sets for dissolved Se for 2001 and 2002 determined by ICP-MS were used in the data assessment and interpretation that follow. Data considered supplemental to the assessment of dissolved Se loading are speciation of Se (AA-hydride in 2001) and Se loading via particulate material (total Se in water in 2001 and suspended sediment Se in 2002). These collections and analyses were done to provide an estimate of parameters essential to understanding Se exposure through the watershed. These data sets are considered limited in number, scope, and quality checks. However, Se concentrations for suspended sediment were analyzed by two laboratories, both using AA-hydride during 2002 (Table 16-I). All values for solids are given in dry weight (dw), except as noted.
Regional water-discharge and selenium concentration, speciation, and loading Hydrologic conditions Drought conditions (i.e. low snow pack) in southeastern Idaho (www.idwr. state.id.us/about/issues/drought.htm) affected discharges based on comparison to the historical record for the gaging station (Table 16-II). Historical-record-years used for comparison are: minimum year, 1977 (US Geological Survey, 1977); average year, 1980 (US Geological Survey, 1980); maximum based on peak height, 1974 (US Geological Survey, 1974); and maximum based on total discharge, 1982 (Harper et al., 1982). Diversion of flows for irrigation upstream of the gaging site also occured from thaw to July 1 (O'Dell et al., 2001). Results for 15 months of data collection are shown based on water years (WYs), which begin October 1 (Figs 16-3-16-5). Data from two snowmelt seasons (2001 and 2002) (Figs 16-3 and 16-4) are the focus of analysis because that is when the majority of runoff occurred. Analysis of data for flow and Se load was based on division of the WY into a three-month high-flow season (April through
Selenium loading through the Blackfoot River watershed
449 -4
2000000 2001 flow Historical flow _ .... 2001 load
1600000
3.5
2.5
1200000
r
2
E
800000
~
,'"
S
o
",,
,
"-'--,~
E
- 1.5
~,
O
400000
0.5 i
1
|
i
i
i
|
i
i
|
i
1
Date
Fig. 16-3. Flow and Se load at the Blackfoot River gaging station from April 17 through June 30, 2001, defined as the three-month high-flow season. For comparison is the historical mean daily discharge based on 28 years of record (O'Dell et al., 2001; http://idaho.usgs.gov). Calculations of Se loads were made using mean daily flow and a linear extrapolation of Se concentrations between sampling points for the three-month high-flow season (see Fig. 16-6). See Table 16-II and Results section for details of calculations of Se loads.
2000000
4 i
1600000
M
~ 1200000 E o
.n
~
,A.,. ,,: IIt.'~'-,.
800000
2002 flow --.- Historical flow
3.5
...
3
00 ,oa0
2.5
.,i'--
t~
2
! "
"-.,_."
';
~ E 1.5 ~
"~.
0
400000
o
,,,,r-"
,,,
d,<>, '
0.5 !
i
i
!
|
i
Date
Fig. 16-4. Flow and Se load at the Blackfoot River gaging station from April 1 through June 30, 2002, defined as the three-month high-flow season. Given for comparison is the historical mean daily discharge based on 28 years of record (O'Dell et al., 2001; http://idaho.usgs.gov). Calculations of Se loads were made using mean daily flow and a linear extrapolation of Se concentrations between sampling points for the three-month high-flow season (see Fig. 16-6).
TABLE 16-1I Flow in million cubic meters (Mm 3) and Se loads [(kg), (kg d a y - l ) , and (kg yr-1)] for designated three-month high-flow snowmelt period and nine-month low-flow period at the Blackfoot River gaging station. Historical-record-years used for comparison are: minimum (1977), average (1980), maximum based on peak height (1974), and maximum based on total discharge (1982). Calculations of Se loads were made using mean daily flow and a linear extrapolation of Se concentrations (IxgL-l Se) between sampling points observed during 2001 and 2002 (see Fig. 16-6). Patterns of Se concentration observed in 2001 and 2002 were assigned to historical flows for 1977, 1980, 1974, and 1982 to show ranges of potential Se loading. See Results Section for data limitations Water year (starts October 1)
Mm 3 (% annual flow)
Mm 3 d -1 (range)
kgSe I (% annual Se load)
kg d -l Se (range)
Se loads (water years 2001 and 2002) Three month period (April through June) 20012 2001 20024
Mm 3
Mm 3 d -1 (range)
kg Se I (at 1 ~g L-1)
kgSed -1 (range)
Mm 3 kgy -l y-1 Se
Nine month period (July through March)
Annual 75 80 140 107 127
13.7 (37%) 18.1 (44%) 29.8 (56%)
0.059-0.48 0.059-0.48 0.095-1.13
51 (68%) 56 (70%) 116 (83%)
0.14-2.05
23.43
0.023-0.047
0.023-0.41
0.023-0.474
24 24 24
0.032-0.41
37.0 41.4 53.2
0.32-3.55
23.43
Se load forecast 23.0 (37%) 19777 min 23.0 (37%)
0.068-1.10 0.068-1.10
685 (56%) 886 (56%)
0.14-1.67 0.23-2.95
38.8 38.8
0.057-0.122 0.057-0.122
39 39
0.045-0.14 0.045-0.14
61.9 61.9
t~ t~
t~
19808 avg 19749max (peak height) 19821~ max (peak discharge)
101.9 (67%) 101.9 (67%) 108.9 (64%)
0.15-2.91 0.15-2.91 0.27-4.40
4055 (56%) 4516 (56%) 4215 (56%)
0.23-19.55 0.32-13.18 0.41-21.8
49.3 49.3 62.2
0.0764).592 0.076-4).592 0.083-0.490
50 50 62
0.091-0.59 0.091-0.59 O.091-0.50
151.2 151.2 170.9
455 501 483
108.9 (64%) 130.9 (66%)
0.27-4.40 0.16-3.74
4796 (56%) 5255 (56%)
0.55-16.4 0.23-24.6
62.2 69.0
0.083-0.490 0.078-0.881
62 69
0.091-0.50 0.091-0.86
170.9 199.8
541 594
130.9 (66%)
0.16-3.74
6056 (56%)
0.32-22.7
69.0
0.078-0.881
69
0.091-0.86
199.8
674
O~
E"
I Se (Ixg L -1) • volume (m 3) • (106) = load (kg Se). 2 April-June only; O'Dell et al., 2001. 3 July-December 2001 and January-March 2002. 4 http://idaho.usgs.gov. 5 Patterns of Se concentrations were applied from 2001. 6 Patterns of Se concentrations were applied from 2002. 7 Flow: US Geological Survey, 1977. 8 Flow: US Geological Survey, 1980. 9 Flow: US Geological Survey, 1974. lo Flow: Harper et al., 1982.
4u.
452
T.S. Presser et aL
Selenium loading through the Blackfoot River watershed
453
June) and a nine-month low-flow season (July through March) (Table 16-11). Several data limitations and assumptions occur in the data analysis. In 2001, data were available only from April 17, when gaging was reactivated. An additional flow total for WY 2001 is given, which includes estimated conditions in early April 2001 (O'Dell et al., 2001) (Table 16-11). Data are not included from a period when the gage registered ice (12/11/01-2/18/02). Flow data for the nine-month period of July 2001 through March 2002 were used for analysis of both WYs because only 15 months of data were available for analysis (Table 16-II). Patterns of flow were similar in 2001 and 2002, with maximum flow occurring on April 19 in 2001 and April 16 in 2002. Annual total discharges (37.0--41.4 Mm3yr -1 in WY 2001 and 53.2 Mm3yr -1 in WY 2002) were below minimum hydrologic conditions defined by the historical record for the gage (61.9 Mm 3 yr -1 in 1977) (Table 16-11). The three month high-flow season for 2001 and 2002 is compared to historical record hydrographs for the pre-2001 minimum year, the average year, and the two maxima years for the gaging station (Table 16-I1 and Fig. 16-5). Thirty-seven to 44% of annual flow occurred during the three-month high-flow season in 2001 and 56% of annual flow occurred during that time period in 2002 (Table 16-II). In general, daily mean discharges from April 2001 through June 2002 were well below the historical mean daily discharge based on 28 years of data (Figs 16-3 and 16-4) (O'Dell et al., 2001; http://idaho.usgs.gov).
Selenium concentrations
Selenium concentrations as measured at the gaging station (Fig. 16-6) reached maxima of 7.4 txg L-1 Se in May 2001 and 7.6 IxgL-1 Se in 2002 in May, with a gradual decrease to an overall minimum of 1.0 ixgL -l Se in September 2001. Levels of marginal (>2 ixgL -1 Se) and substantive risk (>5 ixgL -1 Se) to aquatic life also are shown on Fig. 16-6 (US Department of Interior, 1998). In detail, the peak of the hydrograph occurred first, followed by a peak in Se concentration 12 days later in 2001 and 24 days later in 2002 (Fig. 16-6). Selenium concentration maxima for 2001 and 2002 both occurred within the first 10 days of May.
Fig. 16-5. Comparison of flow for 2001 and 2002 to historical record years (minimum year, 1977; average year, 1980; maximum based on peak height, 1974; and maximum based on total discharge, 1982) at the Blackfoot River gaging station. The period shown is defined as the three-month highflow season. The range of Se loads for the snowmelt periods of 2001 and 2002 is also shown. Loads for 1977, 1980, 1974, and 1982 were forecast by applying the Se concentration patterns of 2001 and 2002 to historical hydrographs. The assumption of dry-year concentration patterns may not be applicable for high flow years. However, the forecasts are valuable in that they illustrate the timing and relative scale of potential seasonal loading in a system subjected to a snowmelt climatic regime, but do not necessarily indicate the exact magnitude for wet years.
454
T.S. Presser et al.
Selenium speciation - supplemental data
In 2001, selenite (Se 4+) represented less than 10% of the dissolved species during all months. Selenate (Se 6+) increased to a maximum of 74% with increased flow during May 2001 (Fig. 16-7). During June through August 2001, dissolved Se was approximately
Fig. 16-6. Stage (m) and Se concentration (~g L -1 Se) as measured at the Blackfoot River gaging station from April 17 (the day the gaging station was reactivated) through June 30, 2002 ( 15 months).
Fig. 16-7. Speciation (% of total dissolved Se) and flow (m3 day -t) at the Blackfoot River gaging station from April through September 2001. Dissolved selenate (Se6+), selenite (sea+), and organic Se (operationally defined, Se2-) are shown.
Selenium loading through the Blackfoot River watershed
455
a 50:50 mixture of selenate and organic selenide. In 2001, selenate peaked 16 days after the dissolved concentration maximum. During relatively low-flow in September 2001 when Se concentration had decreased, the organic selenide (Se 2-, operationally defined) fraction of the dissolved species reached 90%. Both speciation of waterborne Se and partitioning of Se between water and organic material substantially affect the potential for bioaccumulation in diet and tissue. Reduced organic forms of Se and, by inference, elevated biotic productivity, as occurred in low-flow summer months, may indicate increased Se uptake in food webs (US Department of Interior, 1998; Luoma and Presser, 2000). Additional data for Se speciation are needed to verify seasonal trends and adequately assess variability.
Suspended sediment- supplemental data
Flow conditions during 2001 and 2002 were likely to minimize the amount of sediment and sediment-associated constituents that were transported in the Blackfoot River. Suspended-sediment concentration was approximately 100 mg L-i during two sampling periods in April 2002. The difference between total Se and dissolved Se, as a measure of the contribution of particulate Se, was < 10% except at the peak of concentration when total Se was 18% greater than dissolved Se. Concentrations in the suspended sediment were 3.4 Ixg g-i Se on April 10, 2002 and 3.2 Ixg g-i Se on April 17, 2002. Concentrations for a suite of major and minor elements in the suspended-sediment samples (i.e. an example of regional eroded material) are compared to average concentrations for the Meade Peak Member at the Enoch Valley Mine (i.e. an example of source rock) (Table 16-III). Without additional monitoring of suspended material in the mining area, it is difficult to directly relate suspended-sediment concentrations at the gaging station to sources. However, the source-rock Se concentration (48 Ixg g-I Se) exceeds the substantive risk threshold to aquatic life from sediment (4 Ixg g-~ Se) and the suspended sediment Se concentration remains within the range of concern for risk to aquatic life from sediment (2--4 ixgg -l Se) (US Department of Interior, 1998). Additional data for Se loading via particulate material are needed to verify amounts and assess the adequacy of methodologies under snowmelt conditions.
Dissolved selenium load calculations
Calculation of Se loads was done through use of the generalized equation: concentration • volume - load Specifically, calculation of daily Se loads (Table 16-II; Figs 16-3-16-5) during the threemonth high-flow season were made using daily mean flow and a linear extrapolation of Se concentrations between sampling points observed during 2001 and 2002 (Fig. 16-6). For the nine month low-flow season, an assigned value of 1 ~g L-i Se (i.e. the minimum
456
T.S. Presser et al.
TABLE 16-III Concentrations of major and minor elements in suspended sediment collected at Blackfoot River gaging station Se (~xgg-I) 4/10/02 4/17/02 MPM 4
4/10/02 4/17/02 MPM 4
3.4 3.2 48
OC Z C (%) (%)
S (~
p2 (%)
As (txgg -~)
4.0 4.3 2.8
0.08 0.08 0.38
0.15 0.16 5.2
3.7 4.4 32
0.9 0.6 2.03
3 2 30
65 76 520
5.2 5.3 3.9
Sb Mo (~gg-~) (~gg-~)
V (ixgg -~)
Hg (txgg -~)
Cd (ia,gg -~)
Pb (i,zgg -~)
Cr (txgg -j)
Cu (txgg -~)
Zn (la,gg -~)
Ni (la,gg -~)
Sr (la,gg -~)
0.045 0.056 -
<2 <2 60
17 20 12
83 95 1194
17 23 96
133 161 1137
27 33 213
154 139 553
OC, organic carbon. 2 Converted concentrations: P205 -- 0.34%, 0.37%, 11.8%, from top to bottom, respectively. 3 Sb from Hot Springs Mine (Piper et al., 2000). 4Average concentration for the Meade Peak Member (MPM) at the Enoch Valley Mine for comparison (from Piper, 1999; Piper et al., 2000);- not analyzed.
concentration) was used because available concentration data for this period were equal to or nearly equal to the minimum concentration (Fig. 16-6). This concentration was applied to the total volume of drainage for the nine-month low-flow period. Sixty-eight to 83% of the Se load occurred during the high-flow season (Table 16-11). Selenium loads for both years ranged from 0.14 to 3.6 kg day-~ during the three-month high-flow season and from 0.02 to 0.41 kg day-i during the nine-month low-flow season (Table 16-II; Figs 16-3 and 16-4). The range of maximum Se loading was 2.0-3.6 kg day-i. Annual loads for 2001 and 2002 were 80 and 140 kg Se, respectively. Both years defined new minimum flow levels, and by inference, minimum Se loads.
Selenium load forecasts j b r average and wet years - supplemental data
Future monitoring during average and wet years in the Blackfoot River watershed will give additional information on regional Se loading under a range of hydrologic conditions. However, the limited data set of Se concentrations for dry and drought conditions of 2001 and 2002 can be applied to historical hydrographs for average and wet years to give an indication of the magnitude of potential Se loading due to increased flow in the Blackfoot River watershed. These forecasts help provide a range of possible Se loads for the watershed to
Selenium loading through the Blackfoot River watershed
457
assess whether likely combinations of concentrations, load, hydrology, climate, Se reactivity, and Se bioavailability pose significant ecological risks in the Blackfoot River watershed. Applying Se concentration patterns of 2001 and 2002 (both years ranged from 1 to -7.5 txg L -l Se) to historical dry-, average-, and wet-year hydrographs for the gaging station (Table 16-11 and Fig. 16-5) showed annual potential Se loadings of: (a) 107-127kg for a minimum year; (b) 455-501 kg for an average year; (c) 483-541 kg for a maximum year based on peak height; and (d) 594-674 kg for a maximum year based on total discharge. The extrapolation for a historical dry-year load shows a 29% increase over the 2001 load of 80 kg Se and a 10% decrease over the 2002 load of 140 kg Se. For average and wet years, these extrapolations indicate potentials for a 3.6-7.4-fold increase in Se loading because of increased seasonal flows in the Blackfoot River watershed, if Se concentrations remain the same as in 2001 and 2002. In general, the forecasts for wetter years are valuable in that they illustrate the timing and relative scale of potential seasonal loading in a system subjected to a snowmelt climatic regime, but not necessarily the exact magnitude for wet years. The use of 2001 and 2002 concentration patterns to forecast average- and wet-year Se loads may not be valid for higher flow years, given that with more flow, Se concentrations increase. Additionally, only two years of snowmelt were observed. Both years in which Se concentrations were measured were atypical in that drought emergencies had been declared in southeast Idaho. The range of Se concentrations and timing of Se concentration maxima for both years were similar however, leading to an approximate 10% difference within average and wet year forecast categories (Table 16-11). In terms of potential daily Se loading, the ranges are, dry: 0.04-3.0kg day -I" average, 0.09-19.5kg day-I; wet based on peak discharge, 0.09-21.8 kg day-~; and wet based on net discharge, 0.09-24.6 kg day -~ (Table 16-11; Fig. 16-5). Thus the forecasted maximum daily Se loads during the snowmelt period of a normal to wet year are approximately 20-25 kg day-~. These daily loads are in contrast to maximum daily measured Se loads of 2-3.6 kg day-l during the snowmelt period for 2001 and 2002, representing the high-flow period of critical or dry years.
REGIONAL SELENIUM RESERVOIR
Geohydrologic balance Seasonal high flow in the mining area resulted in increased Se concentrations and loads as measured at the gaging station (Figs 16-3, 16-4, and 16-6). This approximately three-month temporal trend appears to exceed that expected from a short-duration spring flush of evaporative salts that occurs atter dry periods (King, 1995). Increase in both concentrations and loads with increasing flows suggests a regional reservoir of Se that is a function of the geohydrologic balance of Se between the primary geologic inventory of Se (as the source of influx) and drainage (as the source of efflux). The similarity in this mobility trend in an area of mine-waste drainage to trends occurring in Se-impacted areas in California, where subsurface drains have been installed (Luoma and Presser, 2000; Presser et al., Chapter 11), suggests significant Se storage in the mining area of southeast Idaho that is now subject to transport.
458
T.S. Presser et al.
Selenium sources and source drainage
Regional Se drainage, as measured at the gaging station, is linked to release of Se through mining activities associated with the Phosphoria Formation. Streams, seeps, and drains with elevated Se concentrations issue from phosphate mining waste-rock that has been contoured into hills, used to fill valleys, or used as backfill for mine pits. Studies related to Se drainage have shown the following: 9 Waste rock is generated at a rate of 2.5-5 times that of mined ore and individual dumps contain 6-70 Mt of waste rock (US Department of Interior and US Department of Agriculture, 1977). 9 Prior to the knowledge of Se release and impacts, waste rock was used as a construction material for under-drains for valley fill (US Department of Interior and US Department of Agriculture, 1977) and was considered suitable as a growth medium for reclamation (i.e. vegetation of slopes for stability and grazing) (Maxim Technologies Inc., 2002). 9 The Meade Peak Member contains up to 1200 txg g-! Se (dw) (Bloomington Canyon, McKelvey et al., 1986), a value exceeding a solid Se hazardous waste (100 mg g-i Se, ww; or 1 l l m g g -I Se, dw at 10% moisture), if this criterion was applied to mining waste (McKelvey et al., 1986; US Department of Health and Human Services, 1996). The highest Se concentrations occur in the waste shales (Perkins and Piper, Chapter 4; Perkins and Foster, Chapter 10). 9 Average concentrations of Se in the Meade Peak Member at different mine locations are: Enoch Valley, 51 ixgg -l Se (range 1-410txgg -I Se); Hot Springs, 70txgg -! Se (range 6-240 Ixg g-! Se), and Lakeridge, 48 Ixg g-l Se (0.5-380 Ixg g-! Se) (Piper, 1999; Piper et al., 2000). More recent data from a channel section of the uppermost part of the Meade Peak Member show a maximum of 425 txg g-~ Se (Grauch et al., Chapter 8). 9 Selenium concentrations in tributary streams draining both active and inactive mines in the Blackfoot River watershed contained up to 400 txg L -I Se (Piper et al., 2000; Montgomery Watson, 2001a,b), which exceeds the guideline for protection of aquatic life by 50-fold (5 txg L-J Se, US Environmental Protection Agency, 1987). 9 Selenium concentrations in source drains and waste-rock seeps can exceed the criterion for a water-extracted hazardous waste of 1000 txgL -~ Se (US Department of Health and Human Services, 1996; TRC Environmental Corporation, 1999; Piper et al., 2000; US Department of Interior, 2002). Besides these field data, recent data from controlled leaching of mine waste in laboratory column tests also relate to the overall magnitude of the source reservoir of Se available for release on a regional basis (Table 16-IV) (US Department of Interior and US Department of Agriculture, 2002). This testing was conducted as part of environmental documentation for expansion of mining in southeast Idaho (US Department of Interior and US Department of Agriculture, 2002). Mining of the Meade Peak Member will be the major source of phosphate and waste rock in the mining expansion areas. However, this set
TABLE 16-IV Selenium concentrations in mine waste leachate as exemplified by column tests of proposed overburden for mining expansion and in seeps and springs (adapted from US Department of Interior and US Department of Agriculture, 2002) Weighted
run-of-mine
Hanging-wall mudst,
(average) Column tests 1 (l~gL-1 Se) Overburden2 (%) 1st pore volume 2nd pore volume 3rd pore volume 1st, 2nd, 3rd (average) 5th pore volume 7th pore volume 9th pore volume 10th pore volume 1-10 (average) Overburden seeps and springs 3 Regional, range Regional, average/median Panel D 4 seeps, grab Panel E4 seeps, grab Panel D seeps, avg. (1999 through June 2001) Panel E seeps, avg. (1999 through June 2001)
Hanging-wall phosphatic shale
Center-waste shale, altered
Center-waste shale, reduced
Run-of-mine, Chert Footwall surface mudst. (Panel A) t~
100 149-181 58-64 38-47 81 47 47 41 44 55
24.4 232 47 30 103 26 25 22 20 57
7.2 273 101 61 145 56 95 52 57 99
30.3 (altered + reduced) 205-404 117-134 131-141 66-126 60-124 49-53 132-223 77-104 52-129 52-59 45-117 58-68 47-111 52-68 51-120 59-78 84-164 65-84
0 951 459 230 547 195 220 39 141 334
33.3 6 3 2 4 2 < 2 2 2 3
4.8 109 62 75 82 62 52 40 42 63
(Ixg L-l Se) 5-1,617 252/114 280 240 716 310
Continued
~~ t~
4~
TABLE 16-IV Continued Predicted seepage based on column tests ~ < 100 years > 100 years
(I.tg L-I Se) 81 (average of pore volumes 1-3) 55 (average of pore volumes 1-10)
IAll leaching solutions were filtered (Maxim, 2002). 2A weighted average run-of-mine reflects percentages in proposed overburden (Maxim, 2002). 3Seepage quality throughout southeast Idaho Phosphate District based on 23-sample data set (US Department of Interior and US Department of Agriculture, 2002) for comparison. 4Local overburden panels in area of mining expansion (US Department of Interior and US Department of Agriculture, 2002).
Selenium loading through the Blackfoot River watershed
461
of column test results (Table 16-IV) serves as only one example that may be specific to proposed local overburden (i.e. waste-rock). Further column testing of overburden from other mine sites is needed to define limits of Se concentrations in seepage from mine waste and the predictive utility of column tests (e.g. Herring, Chapter 13). A compilation of data from 23 seeps was provided as part of the environmental documentation to evaluate predictions from column testing concerning groundwater quality in areas of expansion (US Department of Interior and US Department of Agriculture, 2002; Maxim Technologies Inc., 2002) (Table 16-IV). Two local seeps specific to existing overburden in the area of mining expansion averaged 310 and 716 lug L-~ Se, with the range of Se concentrations in source drainage from 5 to 1617 pug L-l Se (Table 16-IV). This data set was considered representative of seepage quality through the phosphate-mining area of southeast Idaho, but also was thought limited when applied to systems as complex as those involved in mine expansion (US Department of Interior and US Department of Agriculture, 2002). The controlled leach data analysis was based on a run-of-mine (i.e. ore as it is mined in its natural, unprocessed state) weighted average reflecting percentages in proposed overburden (US Department of Interior and US Department of Agriculture, 2002; Maxim Technologies Inc., 2002). In order to assess a weighted average of run-of-mine, the overburden was estimated to contain 24.4% hanging-wall mudstone; 7.2% hanging-wall phosphatic shale; 30.3% center-waste shale (altered and reduced); 33.3% chert; and 4.8% footwall mudstone (Table 16-IV). The fine-grained characteristics of leached material led to what was considered a high-reaction rate. In addition, Se concentrations obtained from leaching of strongly weathered surface overburden were considered probable maxima for this series of tests (Maxim Technologies Inc., 2002). In the column tests, the results of the first pore volume were considered not in equilibrium with reduced phases (Table 16-IV). The average of the first three pore volumes calculated as a run-of-mine weighted average was considered as the near-term leaching condition (<100 years, 81 p,g L -I Se) (Table 16-IV). Use of all pore-volume data (1 through 10), calculated as a run-of-mine weighted average, was considered representative of long-term leaching (> 100 years, 55 Ixg L -! Se) that would eventually produce a source depletion (Table 16-IV). Specifically, field-weathered run-of-mine surface overburden showed the highest rate of Se release (951 and 141 Ixg L-~ Se, first pore volume and tenth pore volume Se concentration, respectively) followed by the altered (i.e. weathered) center-waste shale and the hanging-wall phosphatic shale (273 and 57 i~gL -I Se). The unaltered (i.e. reduced) shale showed less release of Se than the altered shale. Chert showed the lowest release (6 and 2 lug L-I Se).
CONCLUSIONS Data presented here suggest a significant Se reservoir that is now subject to mobility in the mining area of southeast Idaho. Ecological impacts in the Blackfoot River watershed are linked to the magnitude of regional Se release. Both source and watershed data help assess
462
T.S. Presser et al.
whether likely combinations of hydrology, climate, Se loading, and Se bioavailability pose significant risks to fish and wildlife. Mine-drainage options that currently meet existing demands for phosphate mining cause ecological risk thresholds to be exceeded. Annual cycles in Se concentration, load, and the Se-species selenate reached maxima in the spring, during the period of maximum flow at a gaging station on the Blackfoot River (Figs 16-3-16-7). During peak periods of flow during May, dissolved-Se concentrations exceeded the criterion for protection aquatic life and the ecological threshold at which substantive risk occurs (5 Ixg L-l Se) (US Environmental Protection Agency, 1987). During the majority of the three-month high-flow period, dissolved Se concentrations exceeded the 2 Ixg L -I concern level for aquatic biota (US Department of Interior, 1998). Suspended-material Se concentration during high flow in 2002 was within the range of marginal risk to aquatic life (Table 16-Ili). Both selenate and organic selenide (operationally defined Se e- ) were major species during relatively low-flow periods in summer. A change in speciation to reduced Se may indicate elevated biotic productivity during summer months and could result in enhanced Se uptake in food webs (US Department of Interior, 1998; Luoma and Presser, 2000). Determination of a Se budget throughout the watershed is crucial because sources of Se are changing as a result of climate and mining management and expansion. Extrapolation from historical hydrographs for average and wet years and a limited data set of regional Se concentrations for 2001 and 2002 indicated potential for a 3.6- to 7.4-fold increase in Se loading because of increased seasonal flows in the Blackfoot River watershed. A site just below the Blackfoot River gaging station showed a high hazard based on Se concentrations in water, sediment, macroinvertebrates, and fish in June 2000 (Hamilton et al., Chapter 18). Sites documented above the gaging station nearer source drainages showed moderate to high hazards (Hamilton et al., Chapter 18). For the Blackfoot River site, the geometric mean Se concentration in invertebrates was 5.4 Ixg g-z Se, dw (composite sample of several taxa) and in fish was 7.8 Ixgg-I Se, whole-body, dw (range 5.8-14.1 ixgg -i Se) (Hamilton et al., Chapter 18). Regional Se drainage, as measured at the gaging station, is linked to release of Se through mining activities associated with the Meade Peak Member. Selenium concentrations in source rocks can exceed the criterion for a solid waste (100 Ixg g-l Se, ww) and Se concentrations in individual source drains and waste-rock seeps associated with phosphate mining can exceed the criterion for a water-extracted hazardous waste (1000 ixgL -l Se) (US Department of Health and Human Services, 1996). An example given in environmental documentation of expected Se release in source seepage from mining expansion in SE Idaho, based on column leaching of proposed overburden (Table 16-IV) (Maxim Technologies Inc., 2002; US Department of Interior and US Department of Agriculture, 2002), predicted Se concentrations source seepage of 81 Ixg L-! Se in the near term (within 100 years) with decreasing concentrations to 55 ixgL -l Se over the longer term (> 100 years). These predicted Se concentrations are site-specific, but give some perspective for comparison of Se concentrations derived from leaching of source rocks to Se concentrations in regional drainage. Both of the predicted concentrations are above the drinking-water
Selenium loading through the Blackfoot River watershed
463
standard for Se (50 I~g L -l Se, US Department of Health and Human Services, 1996) and 11- to 16-fold above the Se criterion for protection of aquatic life (5 Ixg L -1 Se, US Environmental Protection Agency, 1987). Column leaching of specific shale zones of the Meade Peak Member showed higher concentrations (up to 404 txg L-l Se), as did leaching of weathered overburden (up to 951 txg L-l Se) (Table 16-IV). Determination of a Se mass balance for the watershed and Se cycling through the components of the watershed's ecosystems are also crucial because of Se bioaccumulation (Presser and Piper, 1998). A comprehensive linked approach would include all considerations that cause systems to respond differently to Se contamination. Both speciation and partitioning of Se between water and particulate material substantially affect the potential for bioaccumulation in diet and tissue of predators. Even though Se concentrations decrease during low-flow months, risk may increase due to increased biotic productivity and longer residence times during summer. Risk during the high-flow season may be associated with increased Se loading and concentrations in streams. As management and remediation plans for Se are proposed, the kind of analysis done in this chapter would help in modeling the Blackfoot River and tributary watersheds to accurately project ecological effects based on sources, flow periods, and environmental Se concentrations that most influence Se reactivity and bioavailability in food webs. This kind of analysis provides inputs into Se-pathway bioaccumulation models based on site-specific food webs to define Se exposures to sensitive higher-level predators. Regulatory approaches that emphasize source load quantification and reduction could work towards an overall load target determined by vulnerable food webs. A Se mass balance approach through the Blackfoot River watershed would provide a comprehensive understanding of the physical, chemical, and biological processes that underlie Se exposure and offer opportunities to identify resolutions of phosphate-mining management and expansion questions that may arise.
ACKNOWLEDGMENTS This work was supported in part by the US Bureau of Land Management (BLM) and the Idaho Department of Environmental Quality. The authors are grateful to Jay Bateman and Donald Cole of the Idaho District field office at Idaho Falls for collection of field data and samples. We especially thank Peter Oberlindacher and Bill Stout of the BLM for helping to re-initiate gaging of the Blackfoot River.
REFERENCES Buchanan, T.J. and Somers, W.P., 1984. Discharge measurement at gaging stations. US Geological Survey, Techniques of Water Resources Investigations, Book 3, Chapter A8, 65 pp. Cutter, G.A. 1978. Species determination of selenium in natural waters. Anal. Chim. Acta, 78: 59-66.
464
T.S. Presser et al.
Cutter, G.A., 1983. Elimination of nitrite interference in the determination of selenium by hydride generation. Anal. Chim. Acta, 149: 391-394. Fishman, M.J., and Friedman, L.C. (eds.), 1989. Methods for determination of inorganic substances in water and fluvial sediments. US Geological Survey, Techniques of Water Resources Investigations, Book 5, Chapter A1,545 pp. Hamilton, S.J., 1998. Selenium effects on endangered fish in the Colorado River Basin. In: W.T. Frankenberger Jr. and R.A. Engberg (eds.), Environmental Chemistry of Selenium. Marcel Dekker, New York, NY, pp. 297-317. Harper, R.W., Sisco, H.G., O'Dell, I. and Cordes, S.C., 1982. Water resources data, Idaho, water year 1982. US Geological Survey, Water-Data Report, ID-82-1, p. 106. Horowitz, A.J., Elrick, K.A. and Hopper, R.C., 1989. A comparison of instrumental dewatering methods for the separation and concentration of suspended sediment for subsequent trace element analysis. Hydrol. Processes, 2:162-184. Idaho Department of Environmental Quality, 2002. Area Wide Human Health and Ecological Risk Assessment and related memorandum by R. Clegg. Tetra Tech EM Inc., Boise Idaho, 156 pp. Idaho Department of Environmental Quality. http://www.deq.state.id.us/news/may23_02.htm, view Appendix H, draft data summary. Idaho Department of Water Resources. www.idwr.state.id.us/about/issues/drought.htm Kennedy, E.J., 1983. Computation of continuous records of streamflow. US Geological Survey, Techniques of Water Resources Investigations, Book 3, Chapter A I3, 53 pp. King, T.V., 1995. Environmental considerations of active and abandoned mine lands: lessons from Summitville, Colorado. US Geological Survey, Bulletin, 2220, 38 pp. Lamothe, P.L., Meier, A.L. and Wilson, S., 1999. The determination of forty-four elements in aqueous samples by inductively coupled plasma-mass spectrometry. US Geological Survey, Open-File Report, 99-15 I, 14 pp. Lemly, A.D., 2002. Selenium assessment in aquatic ecosystems: a guide for hazard evaluation and water quality criteria. Springer, New York, NY, pp. 5-17. Luoma, S.N. and Presser, T.S., 2000. Forecasting selenium discharges to the San Francisco BayDelta Estuary: ecological effects of a proposed San Luis Drain extension. US Geological Survey, Open-File Report, 00-416, 358 pp. Maxim Technologies Inc., 2002. Simplot Smoky Canyon Expansion Environmental Impact Statement Column Test Report: prepared for J.R. Simplot Company, Pocatello Idaho. Maxim Technologies Incorporated, Bozeman, Montana, pp. 1-37. McKelvey, V.E., Strobell, J.D. and Slaughter, A.L., 1986. The vanadiferous zone of the Phosphoria Formation in western Wyoming and southeastern Idaho. US Geological Survey, Professional Paper, 1465, 27 pp. Meyer, R., 1996. Surface-water quality assurance plan for the California District of the US Geological Survey. US Geol. Survey, Open-File Report, 96-618, 68 pp. Montgomery Watson, 2001a. Draft 1999-2000 Regional Investigation Data Report for Surface Water, Sediment and Aquatic Biota Sampling Activities, May-June 2000, Southeast Idaho Phosphate Resource Area, Selenium Project: (Montgomery Watson, Steamboat Springs, Colorado), Chapters 1-5; Appendices A-F. Montgomery Watson, 200lb. Draft 1999-2000 Regional Investigation Data Report for Surface Water, Sediment and Aquatic Biota Sampling Activities, September 1999, Southeast Idaho Phosphate Resource Area, Selenium Project (Montgomery Watson, Steamboat Springs, Colorado), Chapters 1-5; Appendices A-E
Selenium loading through the Blackfoot River watershed
465
Montgomery Watson, 2002. Draft Interim Spring 2002 Surface Water and Sediment Investigation Report, Southeast Idaho Phosphate Resource Area, Monsanto Mine-Specific Selenium Program (Montgomery Watson, Steamboat Springs, Colorado), water-quality data spreadsheet for Blackfoot River sampling sites, 1 p. and map. O'Dell, I., Lehmann, A.K., Campbell, A.M., Beattie, S.E. and Brennan, T.S., 2001. Water resources data, Idaho, water year 2001. US Geological Survey, Water-Data Report, ID-01-1, p. 106. Piper, D.Z., 1999. Trace elements and major-element oxides in the Phosphoria Formation at Enoch Valley, Idaho: Permian sources and current reactivities. US Geological Survey, Open-File Report, 99-163, 66 pp. Piper, D.Z., Skorupa, J.P., Presser, T.S., Hardy, M.A., Hamilton, S.J., Huebner, M. and Gulbrandsen, R.A., 2000. The Phosphoria Formation at the Hot Springs Mine in southeast Idaho: a source of selenium and other trace elements to surface water, ground water, vegetation, and biota. US Geological Survey, Open-File Report, 00-050, 73 pp. Presser, T.S., 1994. The Kesterson effect. Environ. Manage., 18: 437-454. Presser, T.S. and Piper, D.Z., 1998. Mass balance approach to selenium cycling through the San Joaquin Valley: From sources to river to bay. In: W Frankenberger and R.A. Engberg (eds.), Environmental Chemistry of Selenium. Marcel Dekker Inc., New York, NY, pp. 153-182. Presser, T.S., Sylvester, M.A. and Low, W.H., 1994. Bioaccumulation of selenium from natural geologic sources in western states and its potential consequences. Environ. Manage.,l 8: 423-436. Radtke, D.B., Wilde, ED., Davis, J.V. and Popowski, T.J., 1998, Alkalinity and acid neutralizing capacity. In: ED. Wilde and D.B. Radtke (eds.), National field manual for the collection of water-quality data. US Geol. Survey, Techniques of Water Resources Investigations, Book 9, Chapter A6.6, 33p. Rantz, S.E. et al., 1982, Measurement and computation of streamflow. US Geological Survey, WaterSupply Paper, 2175, vol. 1, 2, 631 pp. Skorupa, J.P., 1998. Selenium poisoning of fish and wildlife in nature: lessons from twelve realworld examples. In: W.T. Frankenberger Jr. and R.A. Engberg (eds.), Environmental Chemistry of Selenium, Marcel Dekker, New York, NY, pp. 315-354. Trelease, S.E and Beath, O.A., 1949. Selenium: its geological occurrence and its biological effects in relation to botany, chemistry, agriculture, nutrition, and medicine. Trelease and Beath, New York, NY, 292 pp. TRC Environmental Corporation, 1999. Maybe Canyon (south) Site Investigation, Caribou National Forest, Caribou County, Idaho: Englewood, Colorado, 121 pp.; Appendices A-W. US Department of Health and Human Services, 1996. Toxicological profile for selenium (Agency for Toxic Substances and Disease Registry, Public Health Service, US Department of Health and Human Services, Atlanta, Georgia), 185 pp. US Department of the Interior (U.S. Fish and Wildlife Service, Bureau of Reclamation, Geological Survey, Bureau of Indian Affairs), R.A. Engberg (ed.), 1998. Guidelines for interpretation of the biological effects of selected constituents in biota, water, and sediment. (National Irrigation Water Quality Program, US Department of Interior, Bureau of Reclamation, Denver, CO), pp. 139-184. US Department of the Interior (US Geological Survey and Bureau of Land Management) and US Department of Agriculture (Forest Service), 1977. Final environmental impact statement: development of phosphate resources in southeastern Idaho vol. I (US Government Printing Office, Washington, DC), 429 pp. US Department of the Interior (Bureau of Land Management) and US Department of Agriculture (Forest Service), 2002. Final supplement environmental impact statement for Smoky Canyon
466
T.S. P r e s s e r et al.
Mine, panels B and C (US Bureau of Land Management, Pocatello, Idaho), Chapters 1-7, Appendices A 1-4C. US Department of the Interior (Bureau of Land Management), US Department of Agriculture (Forest Service) and US Army Corps of Engineers, 2000. Final environmental impact statement for Dry Valley Mine, south extension project (US Bureau of Land Management, Pocatello, Idaho), Chapters 1-7, Appendices A-K. US Environmental Protection Agency, 1987. Ambient water quality criteria for selenium - 1987 (EPA-440/6-87-008; PB88-142237, Office of Water Regulation and Standards, Criteria and Standards Division, Washington, DC), 121 pp. US Geological Survey, 1974. 1974 water resources data for Idaho. Part 2. Water Quality Records. US Geological Survey, Water Resources Division, p. 146. US Geological Survey, 1977. Water resources data for Idaho, water year 1977. US Geological Survey, Water-Data Report, ID-77-1, pp. 181-182. US Geological Survey, 1980. Water resources data, Idaho, water year 1980. US Geological Survey, Water-Data Report, ID-80-1, pp. 117-118. Wilde, ED., Radtke, D.B., Gibs, J. and lwatsubo, R.T. (eds.), 1999. Collection of water samples and processing of water samples in national field manual for the collection of water-quality data. US Geological Survey, Techniques of Water Resources Investigations, Book 9, Chapters A4, A5 [variously paged].
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
467
Chapter 17
SELENIUM ATTENUATION IN A W E T L A N D FORMED FROM MINE DRAINAGE IN THE PHOSPHORIA FORMATION, SOUTHEAST IDAHO
L.L. STILLINGS and M.C. AMACHER
ABSTRACT The black shale-hosted phosphorite of the Phosphoria Formation has been mined for phosphate during most of the twentieth century. Selenium (Se) occurs in high concentrations throughout much of the formation, leading to concerns that mining activities may enhance its release to the environment. Seepage water from waste-rock dumps has been discovered to contain up to 1800 tug L-~ dissolved Se. Our study focuses on the removal of Se from surface waters in a wetland located at the base of a waste-rock dump. Samples of surface water and sediment were collected from a line of seeps emerging at the base of the waste-rock pile and throughout the length of the wetland. Sediment samples (collected to a depth of 15 cm) were analyzed with selective extractions for exchangeable, carbonate, Mn oxide, Fe oxide, and organic matter + sulfide + residual fractions, with each fraction being analyzed for total bulk chemistry. Water samples were also analyzed for major, minor, and trace elements. Seepage waters are easily distinguished from background waters. Seepage waters display a Ca-SO 4 chemistry compared to the Ca-HCO 3 chemistry of background water. Major ions and Se occur at much higher concentrations in seepage waters, and the range of solute concentrations is much greater in seepage waters than in background waters. Selenium concentrations in seepage water are highly variable and depend on location and discharge volume of the seep. Although we did not quantify discharge, the high discharge at the main seep site in June 1999 contained a greater Se concentration (520 Ixg L -l) than the lower discharge in September 1999 (11-23 Ixg L-i). Se concentrations were also higher during the months following a deeper snow pack (520 pug L-i in June, 1999, vs. 38 pug L-! in June, 2000). Se is quickly attenuated from surface water as it flows from the seeps through the wetland, and concentrations drop from a range of 11-520 Ixg L-~ at the main seep site to <5 txg L within 50 m of the seeps. Wetland sediments within this 50 m distance show the highest concentration of total Se, up to 693 mgkg-~, and in all but one sample most of the Se is found within the non-crystalline Fe-oxide fraction of the sediment. The total bulk concentration of Se in the sediments (mg kg-l) can be estimated with a linear expression: [Setot]sed = 8.09[ncFe203]se d + 0.874
468
L.L. Stillings and M.C. Amacher
where [ncFe203]se d is the concentration of non-crystalline Fe oxides in the sediment (as percent Fe203). This expression fits the data with an r 2 of 0.968 (n = 7). Wetland waters do not appear to result from simple linear mixing between seepage and background waters. Instead, we hypothesize that Se adsorption and/or coprecipitation with non-crystalline Fe oxides is the process responsible for sequestering Se from wetland waters to the sediments.
INTRODUCTION Selenium in the environment is a paradoxical topic. On one hand, it is known that trace levels of selenium have beneficial effects on metabolism and development of domestic livestock, yet higher concentrations are known to cause toxicosis and teratogenesis in both livestock and wildlife. Cases of human Se toxicity are rare but have occurred from consumption of Se-rich produce and well water (Engberg et al., 1998). During the 1980s, researchers focused attention on anthropogenic Se release to the environment from activities related to fossil fuel operations, metal smelting, and irrigation of seleniferous soils. In the San Joaquin Valley, California, Se-rich agricultural drainage caused a massive number of deaths and deformities of wildfowl in the Kesterson Reservoir (Ohlendorf and Santolo, 1994). Concentrations of dissolved Se in the San Luis agricultural drain, which discharged into the Kesterson Reservoir, averaged approximately 300 txg L(Weres et al., 1989; Engberg et al., 1998). By comparison, the current US Environmental Protection Agency water-quality standard for Se is a limit of 5 p.g L-~ total recoverable Se for chronic exposure (the Criterion for Continuous Concentration, CCC; US Environmental Protection Agency, 2002). Similar problems with wildfowl reproductive impairment were encountered in a constructed wetland designed to remove pollutants from the waste effluent from the Chevron Oil Refinery in Richmond, California, in the mid1990s. Hatching success and post-hatch survival rates were reduced due to high concentrations of Se in the effluent water discharged to the wetland (Ohlendorf and Gala, 2000; Lemly, 2002). The biogeochemistry of Se, especially in wetland environments, has remained a pressing topic of research due to continued discharge of high Se agricultural wastewaters to California surface waters, and the use of constructed wetlands and ponds for treating industrial wastewaters. Many studies have concluded that Se can be removed from wetland surface waters through biological volatilization (Masscheleyn et al., 1991; Hansen et al., 1998; Lin et al., 1999; Pilon-Smits et al., 1999), uptake by wetland plants (Pilon-Smits et al., 1999), and sequestration into wetland sediments. Selenium in reduced sediments has been found commonly in the organic fraction and as elemental Se (Weres et al., 1989; Pickering et al., 1995; Tokunaga et al., 1996; Martens and Suarez, 1997). Anthropogenic release of Se to the environment is not confined to agricultural and fossil fuel operations. Mining activities that disturb rock with high Se concentrations may also release Se to the environment. In 1996, livestock grazing near mine waste dumps in southeast Idaho developed symptoms of Se poisoning and died. The dumps contain waste rock resulting from mining phosphate in the Phosphoria Formation, a marine black shale known
Selenium attenuation in a wetland
469
to contain locally high concentrations of Se. Seepage waters from waste-rock dumps in the region have been shown to carry as much as 1800 ~g L-i Se (Amacher, unpublished data). The purpose of this study was to investigate the fate of Se discharged in seepage from mine waste rock from mining in the Phosphoria Formation, southeast Idaho. A natural wetland formed at the seep site and this paper focuses on the interaction between Se in the wetland surface water and sediment. Other investigators have discussed Se volatilization from this wetland (Lamothe and Herring, Chapter 15), and Se uptake by the wetland plants (Mackowiak et al., Chapter 19).
METHODS Site Our study was conducted in the Caribou National Forest, southeast Idaho. Since 1906 (Lemly, 2002), the Phosphoria Formation in that region has been mined for phosphate for use in the agriculture and food industries. The study site is a wetland that formed from seepage at the base of a mine waste-rock pile created in the 1970s, denoted as the Wooley Valley Unit #4 dump (Fig. 17-1). The wetland is approximately 450 m long and 50 m wide. At the head of the wetland near the main seepage site (S 1), surface water is approximately 2-5 cm deep and flows through grasses and shrubs. In the middle of the wetland, about 80 m downstream from the seepage, water flow becomes channelized and flows through a series of eight beaver ponds until it exits the area at approximately 450 m from the seepage. Drainage from the wetland flows into the head of Angus Creek, a tributary of the Blackfoot River. Samples of surface waters and sediments were collected at the same locations from the Wooley Valley wetland during the period between October 1998 and June 2000. Sample locations are marked in the wetland schematic (Fig. 17-1) and described in Table 17-I.
Collection and analytical methods Surface waters
Surface waters were collected from the seeps that emerge at the base of the waste dump, from the wetland, and from the beaver ponds. Background samples for the site were collected from a stream flowing from the east of the area, in a drainage that did not contain mine waste (Fig. 17-1). Samples were collected as grab samples with a plastic syringe, then dispensed into acid-washed-and-rinsed polyethylene bottles. Aliquots were collected for analysis of major cations and trace elements by ICP-AES and ICP-MS, anions by ion chromatography (IC), and alkalinity by Gran titration. Samples collected for ICP and IC analyses were filtered through a 0.45 p.m mixed cellulose ester filter, and concentrated ultra pure HNO3 was added to the ICP samples to create a 1% solution. Samples were kept
t~
Fig. 17-1. Schematic map of the field site at Wooley Valley; seepage from the dump flows north through the beaver ponds into the settling pond for eventual discharge into Angus Creek; sample locations marked by letters (keyed to Tables 17-I and 17-II); latitude-longitude datum is NAD27.
471
Selenium attenuation in a wetland
TABLE 17-I Surface waters collected at the Wooley Valley wetland: site identification, sample dates, and selenium concentration Fig. 1 location
Site ID
Description
Sample date
Se (l~g L - I )
a a a o o d d c c b b b m s s r q q p p n n n e e e e e e f g g g g h
BK BK BK DT Dta J3 J3 J4 J4 J5 JSa J5b MF P1 Pl P2 P4 P4 P6 P6 P8 P8 P8 S1 S1 S la S la S 1b S lb $2 $3 S3a S3b S3m $4
Background site Background site Background site Dead tree area Dead tree area, upstream East side seep 3 East side seep 3 East side seep 4 East side seep 4 East side seep 5 East side seep 5, upstream East side seep 5, downstream Middle ferrihydrite deposition area Pond l Pond 1 Pond 2 Pond 4 Pond 4 Pond 6 Pond 6 Pond 8 Pond 8 Pond 8 Seep 1 Seep l Seep l, above Fe oxide precipitates Seep l, upstream Seep l, below Fe oxide precipitates Seep l, downstream Seep 2 Seep 3 Seep 3, upstream Seep 3, downstream Seep 3, middle Seep 4
June 1999 Sept 1999 June 2000 June 2000 June 2000 Sept 1999 June 2000 Sept 1999 June 2000 Sept 1999 June 2000 June 2000 June 2000 June 1999 Sept 1999 June 1999 June 1999 Sept 1999 June 1999 Sept 1999 June 1999 Sept 1999 June 2000 June 1999 June 2000 Oct 1998 Sept 1999 Oct 1998 Sept 1999 Sept 1999 June 2000 Sept 1999 Sept 1999 Sept 1999 Sept 1999
8.4 7.6 6.9 0.9 2.0 120 210 53 20 240 210 160 2 7.7 2 5.7 4.0 2.0 4.9 2.0 9.3 (0.2 4.0 520 38 44 11 16 23 18 5.0 14 2.0 4.0 7.1 Continued
472
L.L. Stillings and M.C. Amacher
TABLE 17-1 Continued Fig. 1 location
Site ID
Description
Sample date
h i j k k 1
$4 $5 OBP 1 OBP2 OBP2 UF
Seep 4 Seep 5 Old beaver pond 1 Old beaver pond 2 Old beaver pond 2 Upper ferrihydrite deposition area
June Sept Sept Sept June June
2000 1999 1999 1999 2000 2000
Se (txgL -1) 3.0 <0.2 <0.2 <0.2 2.0 2.0
Soil or sediment sample Exchangeable ions: 1 M NH4CI Carbonates: pH 5, 1 M NH 4 acetate Mn oxides: pH 2, 0.1 M NH2OH.HCI Non-crystalline Fe oxides: 0.2 M NH4 oxalate + 0.2 M oxalic acid in dark Crystalline Fe oxides: 0.2 M NH4 oxalate + 0.2 M oxalic acid + 0.1 M ascorbic acid in boiling water bath Organic matter + sulfides + residual: concentrated HNO 3 + concentrated HCI + 30% H202
Fig. 17-2. Sequence of sequential extractions used to identify elements associated with specific solid-phase fractions of the sediment sampled at Wooley Valley.
cool until analysis. ICP and IC analyses (Arbogast, 1996) were performed at the US Geological Survey in Lakewood, Colorado. Gran titrations (Stumm and Morgan, 1996) were performed either on site, or at the USGS laboratory at the University of Nevada, Reno. Titrations were finished within two weeks of sample collection.
Sediments
Sediment cores were collected to a depth of 15 cm with a BMH-53 piston-type core sampler, or for coarse-sediment samples, with a stainless steel scoop. Each entire sample was air dried and sieved through a 2 mm stainless steel sieve to remove coarse fragments. Selective extractions were performed to determine the Se concentrations associated with the exchangeable, carbonate, Mn oxide, non-crystalline Fe oxide, crystalline Fe oxide, and organic matter + sulfide + residual fractions (Fig. 17-2; Amacher and Brown, 2000). Each
Selenium attenuation in a wetland
473
extract was analyzed for total Se by ICP-AES. Total recoverable Se in each sediment sample was determined by digestion in HNO3 + HC104 followed by ICP-AES analysis (Amacher and Brown, 2000).
RESULTS Water samples
Water collected from the seep sites is clearly discriminated from background water due to elevated concentrations of Ca, Mg, Na, SO4, alkalinity, and Se (Fig. 17-3). Background water is characterized as Ca-HCO3-type water, whereas seepage from the waste-rock dump has Ca-SO4-type chemistry (Fig. 17-4). Selenium concentrations in samples collected from the seeps are highly variable (Fig. 17-3). Concentrations vary by seep location, ranging from <0.2-240 ixgL -1 for samples collected in September 1999 (Table 17-I), and appear to be influenced by discharge volume. The highest Se concentration of 520 ixgL -1 was observed at the main seep (S 1) in June 1999, when the seep discharge was high from snowmelt following a winter of 65 inches (165 cm) of snow, as recorded at the airport at Soda Springs, Idaho. During baseflow conditions in September 1999, seepage at S1 showed concentrations of 11-23 Ixg L -1. In contrast, in June 2000, Se concentrations were 38 ixgL -l following a year with 34 inches (86 cm) of snow (Western Regional Climate Center, 2002). Selenium concentrations in surface water decreased with distance from the seeps (Fig. 17-5). Concentrations in seepage waters at S 1 ranged from 11 to 520 Ixg L-1 throughout the study period, yet they decreased from 5 to 10 Ixg L-1 within 50 m downstream of the seep.
Sediment samples
Total Se in the bulk sediment decreased with distance from the seeps, ranging from a high of 693 m g k g -1 in sediments at S1, to 9 mgkg -1 at P2, site r in Fig. 17-1 (Fig. 17-6). Data from the selective extractions indicate that Se is found predominately in the organic matter + sulfide + residual, non-crystalline Fe-oxide, and crystalline Fe-oxide fractions (Fig. 17-6). At S1, most Se in the sediment is associated with the organic matter + sulfide + residual fraction. However, in the remaining samples Se is predominately found in the non-crystalline Fe-oxide fraction, and to a lesser extent, in the crystalline Fe-oxide fraction. The discrepancy between total recoverable Se and the sum of Se concentrations in the various sediment fractions is because of analytical uncertainties in the very low Se concentrations in the lower portion of the wetland, and because the nitric + perchloric acid digestion (for total recoverable Se) recovers more Se than the sequential-extraction methods at high Se concentrations in the wetland sediments (upper portion of the wetland).
474
L.L. Stillings and M. C. Amacher
Fig. 17-3. Element concentration box plots for three sample populations from Wooley Valley: seepage water, wetland surface water, and background water; lower boundary of a box is 25th percentile, the median, a line within the box, and top boundary of a box is 75th percentile; vertical lines above and below a box are 90th and 10th percentiles; and points above and below those vertical lines are outliers. DISCUSSION There are two obvious points of comparison between seepage and background waters. First, the concentrations of the major ions and Se are much higher in seepage waters than in background water; and second, the seepage waters exhibit a much larger range of dissolved ion concentrations than the background water (Fig. 17-3). Dissolved ions in the
Selenium attenuation in a wetland
475
Fig. 17-4. Piper diagram of major ion chemistry of three sample populations from Wooley Valley: seep water, wetland surface water, and background water.
wetland surface waters generally show slightly lower concentrations than seepage waters and the range of values is not quite as large (Fig. 17-3). These observations are consistent with the dissolution of sulfide and carbonate minerals in the waste rock dump. General expressions describing these reactions are: FeS2 (pyrite) +y7 O2+H20 __.) Fe2++ 2SO 2- +2H +, CaMg(CO3)2 (dolomite) + 2H + ~ Ca 2+ + Mg 2+ + 2HCO3.
(17-1) (17-2)
Equation (17-1) is written with 02 as the oxidant, although Fe 3+ is also a common oxidant for pyrite (Stumm and Morgan, 1996). If reactions (17-1) and (17-2) occur simultaneously within the waste dump, then the net result would be elevated concentrations of Fe 2+, SO 2-, Ca 2+, Mg 2+ and alkalinity (HCO3) in the seepage water. On the Piper diagram (Fig. 17-4), water chemistry would be expected to evolve from the HCO3-dominated chemistry of background waters to the SO2--dominated chemistry of seepage waters as
476
L.L. Stillings and M. C. Amacher
Fig. 17-5. Concentration of dissolved Se as a function of distance from seepage sites; Se in seep waters is represented by the range of values at 0 m distance.
Fig. 17-6. Selenium concentrations in sediment fractions as a function of location in the wetland (sample abbreviations in Table 17-II; nc, non-crystalline; c, crystalline).
477
Selenium attenuation in a wetland
the water flows through the waste-rock dump. The highly variable chemistry of the seepage waters is likely due to myriad flow paths available for water as it percolates through the waste-rock dump. In a heterogeneous waste-rock dump, background water might react with many different mineral assemblages, with various contact times, depending upon its flow path. Note that the acidity (H +) produced by reaction (17-1) is neutralized by reaction (17-2), hence with an abundance of carbonate minerals the pH of the seepage waters is not expected to differ from that of background waters. All the waters at Wooley Valley (background, seepage, wetland surface) have a circum-neutral pH, ranging from 6.8 to 8.3. Correlation between decreased Se concentrations in surface waters and high concentrations of Se in wetland sediments, within 50 m of the seep, suggests that Se sequestration in wetland sediments is an important mechanism for Se attenuation in surface waters. The non-crystalline Fe-oxide fraction is the major reservoir for Se in all sediment samples except the sample collected at S 1, the main seepage site, where most of the Se was found in the organic matter + sulfide + residual fraction. The sediment chemistry at the main seep, S 1, was different from the others, with higher concentrations of reduced sulfur and total organic carbon (TOC), and lower concentrations of non-crystalline Fe oxides than sediments at the other sites (Table 17-II). Because the non-crystalline Fe-oxide fraction is the main reservoir of Se at most of the sample sites, total concentrations of Se in the sediments can be expressed as a linear function of the concentration of non-crystalline Fe oxides: [Setot]sed - 8.09[ncFe203]se d + 0.874
(17-3)
where [Setot]sed is the total extractable concentration of Se in the sediments (mg kg-l) and [ncFe203]sed is the concentration of non-crystalline Fe oxides in the sediment (as % Fe203). TABLE 17-II Partial chemical analysis of the sediment samples collected from Wooley Valley wetland in June 1999 Fig. 1 location e m n p q r s t
Sample
S1 FDA P8 P6 P4 P2 P1 GS
Total extractable Se (mg kg-i) 693 252 211 38.3 28.6 8.90 17.4 42.9
Non-crystalline Fe oxides
Total organic carbon (TOC)
(% Fe203)
(%)
5.01 32.8 23.0 5.65 4.59 3.32 1.95 2.04
6.74 4.00 4.19 3.58 3.48 2.90 3.16 3.78
Total reduced S (ppm) 3051 291 243 395 514 765 648 862
478
L.L. Stillings and M. C. A m a c h e r
Fig. 17-7. Correlation of total Se to non-crystalline Fe oxides; total Se in bulk sediment can be estimated from percentage of non-crystalline Fe oxides; One outlier from seep site had little Fe oxide compared to amount of TOC and reduced S.
The linear model fits the data with an r 2 of 0.968 (for n = 7; Fig. 17-7). Use of a linear model implies that the retentive capacity of the Fe oxides for Se is constant at 809 mg Se per kg of Fe (as Fe203). This constant could result from saturation of available adsorption sites, or it could be the maximum amount of Se that can coprecipitate within the structure of the non-crystalline Fe oxide. The association of Se with the Fe-oxide fraction of wetland sediments is an uncommon observation. While many studies have quantified the adsorption of dissolved Se species to the Fe-oxide surface (Balistrieri and Chou, 1987, 1990; references within Dzombak and Morel, 1990), studies of Se in wetland sediments have typically cited the association of Se with organic material. For example, in bulk sediments at the Kesterson Reservoir, Weres et al. (1989) observed a linear correlation of total Se with organic-C content, and suggested that Se was initially deposited with the organic fraction of the sediment. Martens and Suarez (1997) also observed a linear relationship between total Se in the sediments of Kesterson Reservoir and San Luis Drain, and organic-C contents. The close association of Se with Fe oxides at the Wooley Valley wetland is clearly a result of the precipitation of abundant Fe oxides at the site, although organic matter also sequesters Se in these sediments (Fig. 17-6 and the outlier point, the filled triangle, in Fig. 17-7). This outlier point in Fig. 17-7 represents total Se in the sediments at the S 1 seepage site, which was high compared to other sites that contained about the same amount of Fe oxide (~5%) in the sediment (Table 17-11). The main differences in sediment chemistry between S 1 and the others sites were the TOC content (6.74% vs. 3-4%; Table 17-II), and the high concentration of reduced sulfur (3051 m g k g -1 vs. 250-860 m g k g - 1 ; Table 17-1I) at S1.
S e l e n i u m attenuation in a w e t l a n d
479
Interestingly, neither the empirical relationship observed by Weres et al. (1989) nor the relationship observed by Martens and Suarez (1997) provide a good estimate of total Se at S 1. It is likely that empirical relationships such as Eq. (17-3) are site specific and dependent on solution and adsorbent composition. Martens and Suarez (1997) emphasized that empirical relationships, regardless of their form, are to be expected only when all materials have similar inputs of Se, and that organic-C content cannot be a general predictor of Se content in sediments. So far, the discussion has focused on two reservoirs for Se in the wetland system, surface water, and sediments. Yet, other reservoirs exist. Lack of quantitative water-flow data prohibits the construction of a mass budget for Se partitioning among the reservoirs, but data exist for Se volatilization (Lamothe and Herring, Chapter 15) and uptake in plant tissue (Mackowiak et al., Chapter 19). In Chapter 19, Mackowiak et al. demonstrate that plants growing in the wetland at the base of the Unit #4 waste dump have the same mean Se concentration (~15 mg kg -1 Se) as plants growing in disturbed rock on the surface of the dump, and that these concentrations are higher than mean Se concentrations in plant tissue collected from the reference watershed (~0.5 mg kg-1 Se; Fig. 19-5, Chapter 19). By looking more closely at the distribution of Se in plant tissue within the wetland, they saw that the mean Se concentration in plants collected at the main seep, S 1, was ~80 mg kg-1, and that Se concentrations in plant tissue decreased with increasing distance along the flowpath, falling to approximately 5 mg kg-1 at 150 m from the seep. In fact, the graph of Se concentration in plant tissue versus distance (Fig. 19-6, Chapter 19) has an exponential shape similar to Fig. 17-5, showing that the highest concentration of Se in both the plant and Fe-oxide reservoirs is found within 50-70 m of the seep. It appears that the processes of plant uptake and sorption to non-crystalline Fe oxides sequester Se at the head of the wetland, within 50-70 m of where the main seep, S 1, emerges from the disturbed rock of the waste dump. There are still many questions to be answered regarding Se mobility in the wetland. We do not yet know whether Se is released from plant tissue during senescence and decay, or whether it is released from non-crystalline Fe oxides during redox and aging transformations. Also, if Se is released from these reservoirs, will it be retained by another reservoir, perhaps within the sediment or biota, or will it be transported downstream as a dissolved species in ground and/or surface waters? These questions might be answered by continuous monitoring of surface-water chemistry and discharge, and with calculations of a Se mass budget for the wetland. We have argued that adsorption/coprecipitation with non-crystalline Fe oxides is a major mechanism for Se attenuation from the surface waters of the wetland, yet dilution with upwelling groundwater might also result in the observed decrease in dissolved Se concentrations. This possibility can be tested with a linear mixing curve between two end points, using concentrations of dissolved Se and a conservative ion (Na+) in the seep and background samples (Fig. 17-8). If surface waters were a mixture of water from these two sources and if Se was conservative, then the data would plot on a straight line between the two end members. Because the seepage water chemistry was highly variable, the end member data are plotted as sample medians with error bars representing the 25th and
480
L.L. Stillings and M.C. Amacher
Fig. 17-8. Mixing diagram of dissolved Se and Na + for background, seepage, and surface waters (end members are background and seepage waters, which are plotted as sample medians with error bars of the 25th and 75th percentiles). 75th percentiles. The data points, representing surface water samples, are clustered close to, and slightly below, the range of Se-Na concentrations for seep waters. None of the surface waters contain Se-Na concentrations near the mixing line. Because surface waters have Na concentrations close to those of the seep waters, it does not appear that background water is discharging to the wetland and causing significant dilution. A process other than dilution must be responsible for decreased Se concentrations in surface water, as compared to seep water. We argue that adsorption/coprecipitation with non-crystalline Fe oxides and plant uptake (Mackowiak et al., Chapter 19) are responsible for Se attenuation from the surface water.
CONCLUSIONS Seepage emerging at the base of the mine waste-rock dump at Wooley Valley exhibits concentrations of dissolved Se that vary over time between < 0.2 and 520 ~g L-1, depending on discharge volumes and location of seeps. Yet regardless of the initial concentration, Se concentrations in the wetland surface water drop to approximately 5-10 I~g L-1 within about 50 m downstream from the seeps. Total Se concentrations in the bulk sediments are also highest close to the seep, decreasing from 693 mgkg-1 at the seep to 39 mgkg -1 at 150 m downstream. Comparisons between the water and sediment data suggest that Se was removed from surface waters by adsorption/coprecipitation with the non-crystalline Fe-oxide fraction of wetland sediments. At the seep site itself, where the Fe-oxide concentration of the sediments is relatively small, the main reservoir for Se is the organic matter + sulfide + residual
S e l e n i u m attenuation in a w e t l a n d
481
fraction of the sediments. Total Se concentrations in the remaining bulk-sediment samples can be estimated as a function of the concentration of Fe oxide in the sediment. For the conditions present in this wetland system, the Fe oxide appears to contain 809 mg Se Kg-1 of Fe (as Fe203). Dilution by surface waters is not considered a mechanism for reducing dissolved Se concentrations because the data do not follow a mixing curve between the seep and background water end members.
ACKNOWLEDGEMENTS We are grateful for the many people who assisted with this work: Deb Kutterer and Tracy Christopherson helped with sediment-sample collection, preparation, and analysis; Paul Briggs, AI Meier, and Pete Theodorakis of the USGS analytical laboratories provided chemical analysis of the water samples; and the Utah State University Analytical Labs performed ICP analysis of the sediment extracts. We also appreciate the support and assistance provided by Terry Harwood of the US Forest Service Washington Office and Jeff Jones of the Caribou National Forest. Jim Hein, Ben Perkins, and Theresa Presser provided thoughtful, thorough, and timely reviews of our manuscript.
REFERENCES Amacher, M.C. and Brown, R.W., 2000. Mine waste characterization. In: D. Barcel6 (ed.), Sample Handling and Trace Analysis of Pollutants: Techniques, Applications, and Quality Assurance. Elsevier, Amsterdam, pp. 585-622. Arbogast, B.E (ed.), 1996. Analytical methods manual for the Mineral Resource Surveys Program. US Geol. Survey, Open File Report, 96-525, 148 pp. Balistrieri, L.S. and Chao, T.T., 1987. Selenium adsorption by goethite. Soil Sci. Soc. Am. J., 51: 1145-1151. Balistrieri, L.S. and Chao, T.T., 1990. Adsorption of selenium by amorphous iron oxyhydroxides and manganese dioxide. Geochim. Cosmochim. Acta, 54:739-751. Dzombak, D.A. and Morel, EM.M., 1990. Surface Complexation Modeling. John Wiley & Sons, New York, NY, 393 pp. Engberg, R.A., Westcot, D.W., Delamore, M. and Holz, D.D., 1998. Federal and state perspectives on regulation and remediation of irrigation-induced selenium problems. In: W.T. Frankenberger, Jr. and R.A. Engberg (eds.), Environmental Chemistry of Selenium. Marcel Dekker, New York, NY, pp. 1-25. Hansen, D., Duda P.J., Zayed A. and Terry, N., 1998. Selenium removal by constructed wetlands: Role of biological volatilization. Environ. Sci. Technol., 32:591-597. Lemly, A.D., 2002. Selenium Assessment in Aquatic Ecosystems. Springer, New York, 162 pp. Lin, Z.Q., Hansen, D., Zayed, A. and Terry, N., 1999. Biological selenium volatilization: method of measurement under field conditions. J. Environ. Qual., 28:309-315. Martens, D.A. and Suarez, D.L., 1997. Selenium speciation of marine shales, alluvial soils, and evaporation basin soils of California. J. Environ. Qual., 26: 424--432.
482
L.L. Stillings and M.C. Amacher
Masscheleyn, P.H., Delaune, R.D. and Patrick, W.H., Jr., 1991. Arsenic and selenium chemistry as affected by sediment redox potential and pH. J. Environ. Qual., 20: 522-527. Ohlendorf, H.M. and Gala, W.R., 2000. Selenium and Chevron Richmond refinery's water enhancement wetland: a response to A.D. Lemly, 1999. Hum. Ecol. Risk Assess., 6: 903-905. Ohlendorf, H.M. and Santolo, G.M., 1994. Kesterson reservoir- past, present, and future: an ecological risk assessment. In: W.T. Frankenberger, Jr. and S. Benson (eds.), Selenium in the Environment. Marcel Dekker, New York, NY, pp. 69-117. Pickering, I.J., Brown, G.E., Jr. and Tokunaga, T.K., 1995. Quantitative speciation of selenium in soils using X-ray absorption spectroscopy. Environ. Sci. Technol., 29: 2456-2459. Pilon-Smits, E.A.H., de Souza, M.P., Hong, G., Amini, A., Bravo R.C., Payabyab S.T. and Terry, N., 1999. Selenium volatilization and accumulation by twenty aquatic plant species. J. Environ. Qual., 28:1011-1018. Stumm, W. and Morgan, J.J., 1996. Aquatic Chemistry (3rd edn.). John Wiley & Sons, New York, NY, 1022 pp. Tokunaga, T.K., Pickering, I.J. and Brown, G.E., Jr., 1996. Selenium transformations in ponded sediments. Soil Sci. Soc. Am. J., 60:781-790. US Environmental Protection Agency, 2002. National Recommended Water Quality Criteria: 2002. EPA-822-R-02-047, 33 pp. Weres, O., Jaouni, A.R. and Tsao, L., 1989. The distribution, speciation and geochemical cycling of selenium in a sedimentary environment, Kesterson Reservoir, California, USA. Appl. Geochem., 4: 543-563. Western Regional Climate Center, 2002. A webpage of the Desert Research Institute, http://www, wrcc. dri. edu/summary/climsmid, html.
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
483
Chapter 18
SELENIUM AND OTHER TRACE ELEMENTS IN WATER, SEDIMENT, AQUATIC PLANTS, AQUATIC INVERTEBRATES, AND FISH FROM STREAMS IN SE IDAHO N E A R PHOSPHATE MINING S.J. HAMILTON, K.J. BUHL and P.J. LAMOTHE
ABSTRACT Nine stream sites in the Blackfoot River watershed in SE Idaho were sampled in June 2000 for water, surficial sediment, aquatic plants, aquatic invertebrates, and fish. Selenium (Se) and other elements were measured in these aquatic ecosystem components and a hazard assessment was performed on the data. Water quality characteristics were relatively uniform among the nine sites examined. Of the aquatic components assessed, water was the least informative, especially its analysis for Se contamination because measured concentrations were substantially below the national water quality criterion of 5 txg L-~. In contrast, Se and several other elements were elevated in sediment, aquatic plants, and aquatic invertebrates from several sites, indicating accumulation in sediments and cycling through plants and invertebrates. Only Se in fish was elevated to concentrations of potential concern. A hazard assessment of Se in the aquatic environment suggests low hazard at Trail Creek and Sheep Creek, moderate hazard at upper Slug Creek and lower Slug Creek, and high hazard at Angus Creek near the mouth, upper East Mill Creek, lower East Mill Creek, Dry Valley Creek, and lower Blackfoot River.
INTRODUCTION Phosphorus is present in economically mineable quantities in organic-rich black shales of the Permian Phosphoria Formation, which constitutes the Western Phosphate Field. There are four active open pit mines in the SE Idaho Phosphate District that produce phosphate from the Meade Peak Phosphatic Shale Member, and 10 inactive mines (Montgomery Watson, 1999; Jasinski et al., Chapter 3). Mining is by open-pit surface mining and waste materials are generally disposed of on the surface in tailing piles, ponds, landfills, and dumps. Many of the waste-rock piles have drainage systems designed to move surface water and groundwater away from waste piles where it reaches other surface
484
S.J. Hamilton, K.J. Buhl and P.J. Lamothe
waters eventually draining into tributaries and finally the Blackfoot River and Blackfoot Reservoir. Thus, water movement distributes potentially toxic trace elements to aquatic and terrestrial ecosystems. The Blackfoot River watershed has several active and inactive phosphate mines that potentially adversely affect aquatic resources in several tributaries of the Blackfoot River (see Fig. 16-2, Chapter 16). As early as 1970-1976, concerns were expressed about contamination of the Blackfoot River and its tributaries by elements released from phosphate mining (Platts and Martin, 1978). Recent concerns have been the subject of several reports (Montgomery Watson, 1999, 2000, 2001 a,b). Several investigations by the US Geological Survey (USGS) reported the chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale (Desborough et al., 1999; Herring et al., 2000a,b). Other USGS investigations reported trace element concentrations in aquatic bryophytes and terrestrial plants that were influenced by mining activities (Herring and Amacher, 2001; Herring et al., 2001). Release of toxic trace elements from waste-rock piles and accumulation in the food chain has resulted in adverse biological effects. In recent years, five of six horses in a pasture in Dry Valley were euthanized due to Se toxicosis; two horses in the Woddall area were put down due to Se poisoning, and 60-80 sheep were found dead on the old Stauffer Mine site (within the Caribou National Forest) due to Se poisoning (Caribou County Sun, 1999). Twenty-six of the sheep were found at the south end of Rasmussen Ridge Mine near a spring or seep at a waste-rock pile. Elevated concentrations of Se and other trace elements have been reported in limited samples of fish fillets and aquatic invertebrates (Montgomery Watson, 1999). A recent report suggested that Se concentrations in fish and wildlife were sufficiently elevated to cause adverse effects in sensitive fish species (Piper et al., 2000). The purpose of this study was to determine the concentrations of Se and other trace elements in water, surficial sediment, aquatic plants, aquatic invertebrates, and fish from streams in SE Idaho near phosphate mining. This information was used in a hazard assessment of the potential effects of Se and other elements on aquatic resources in areas of the Blackfoot River watershed that are potentially impacted by phosphate mining.
METHODS AND MATERIALS Samples of water, surficial sediment, aquatic plants, aquatic invertebrates, and fish were collected in June 2000 from nine sites in the Blackfoot River watershed located in SE Idaho.
Collection site description
The Angus Creek (ACM) site was located at the crossing of the creek by Forest Route 095, approximately 0.5 km above the confluence with the Blackfoot River (see Table 18-I for coordinates of all sampling sites). The land on either side of the road is
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
485
TABLE 18-1 Latitude and longitude of nine sites sampled in southeast Idaho near Soda Springs keyed to Fig. 18-1 Site name and ID
Location I
Angus Creek near mouth (ACM) Upper East Mill Creek (UEMC) Lower East Mill Creek (LEMC) Trail Creek (TC) Upper Slug Creek (USC) Lower Slug Creek (LSC) Sheep Creek (ShpC) Dry Valley Creek (DVC) Lower Blackfoot River (LBR)
N42~ W 111020' 15.04" N42~ '26. 74" W111~ N42~ " W111~ " N42~ Wl 11~ " N42~ " W111~ " N42~ '' W 111 ~ N42~ '46.92" W 111020' 00.53" N42~ W111~ " N42~ W111~ "
_
11 m
_
14 m
_
11 m
___8m • 14 m •
m
• 18 m • 14 m • 10m
IGlobal positioning system (courtesy of Phil Moyle, US Geological Survey).
.
.
managed by the Idaho Department of Fish and Game and primarily was grassland habitat with sparse forbs and very limited grazing activity (see Fig. 18-1). The upper East Mill Creek (UEMC) site was located near an unmaintained dirt road (Forest Route 309) about 3.2 km from the intersection of Forest Route 102 and Forest Route 309, about 8 km above the confluence with the Blackfoot River. The land is managed by the US Forest Service and was pine forest with some sagebrush in open areas. Sample collection was in a generally open area of forbs, grass, and sparse pine trees with very limited grazing activity. The lower East Mill Creek (LEMC) site was located at the crossing of the creek by Forest Route 102, approximately 4 km below the upper East Mill Creek site and about 4 km above the confluence with the Blackfoot River. The site is on private land (Bear Lake Grazing Company) accessed by landowner permission (Ms. Joan Bunderson). The land on either side of the road supported grass and sagebrush with moderate grazing. Sample collection was about equally distributed upstream and downstream of the road crossing.
486
S.J. Hamilton, K.J. Buhl and P.J. Lamothe
Fig. 18-1. Map of sample sites: 1, Angus Creek; 2, upper East Mill Creek; 3, lower East Mill Creek; 4, Trail Creek; 5, upper Slug Creek; 6, lower Slug Creek; 7, Sheep Creek; 8, Dry Valley Creek; 9, lower Blackfoot River. Numbers in parenthesis are total hazard assessment scores from Table 18-XV; modified from Presser et al., Chapter 16.
4.
5.
6.
The Trail Creek (TC) site was located at the crossing of the creek by Trail Creek Road (Forest Route 124), about 8 km above the confluence with the Blackfoot River. The site is on private land accessed by landowner permission (Mr. Val Bloxham). The land on either side of the road was grazed grassland. Sample collection was primarily downstream of the crossing in an area with light grazing, whereas the upstream side of the crossing had moderate to heavy grazing. The upper Slug Creek (USC) site was located about 2 km inside the US Forest Service boundary on Slug Creek Road (Forest Route 095), about 0.4 km off the gravel road that parallels the stream. The site was located above the influence of mining activities and is thus the reference site. The US Forest Service manages the land. The vegetation around the stream was primarily willow-type shrubs with sparse grass, with a substantial amount of sagebrush and quaking aspen nearby. The site had light grazing. The lower Slug Creek (LSC) site was located at the intersection of Slug Creek Road (Forest Route 095) and Old Mill Road (Forest Route 124). The site is in the road rightof-way and about 10 km above the confluence with the Blackfoot River. The land
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
7.
8.
9.
487
around the stream is primarily grassland and was heavily grazed in one area downstream and lightly grazed in two other areas upstream. Sample collection was upstream of the heavily grazed area in a stream section with little grazing. The Sheep Creek (ShpC) site was located on private land (Sheep Creek Guest Ranch) about 3 km upstream of the crossing of the creek by Forest Route 095. The site was accessed with landowner permission (Mr. Phil Baker). The land along the stream was primarily pine forest with some forbs, shrubs, and grass. Sample collection for water, sediment, and fish was upstream of most animal and human activity at the end of a private road and about 2 km above Lanes Creek, which flows into the Blackfoot River about 4 km downstream. Aquatic plants and one fish species were collected about 1.5 km downstream of the primary sample collection site and near the stream crossing with Forest Route 095 in an area of sagebrush and grass with moderate grazing. The Dry Valley Creek (DVC) site was located on private land (Hunsacker Ranch) 0.5 km from Forest Route 122 and accessed along a railroad track that parallels the creek. The site is about 0.75 km above the confluence with the Blackfoot River. The land was accessed with landowner permission (Mr. and Mrs. Keith Hunsacker). The land along the stream was primarily grassland with some shrubs nearby and moderate grazing. The lower Blackfoot River (LBR) site was located on county parkland on the upstream side of a large steel and concrete bridge located on Blackfoot River Road about 0.5 km east of State Highway 34. The site was located about 1 km above the confluence with Blackfoot Reservoir. The land was accessed by landowner permission (Caribou County Commissioner Carol Davids-Moore). The land along the river was primarily riparian with some grass and light camping activity.
S a m p l e collection
Samples of water, surficial sediment, aquatic plants, aquatic invertebrates, and fish were collected at each of the nine stream sites. Water sample bottles were conditioned by immersion in site water three times. Sample container conditioning and water sample collection followed US Geological Survey procedures. Water samples were collected using depth-integrated sampling techniques. For each sample site, one set of samples was unfiltered and a second set filtered through a 0.45 Izm polycarbonate filter. One filtered water sample was collected for measurement of major cations and anions. A 200 mL sample of filtered water was collected in an acid-cleaned polyethylene bottle for analysis of Se concentrations and a second bottle was collected for trace element anaysis concentrations (hereafter referred to as other elements). A reagent deionized-water blank was also collected. Water samples for Se analysis were acidified with ultrapure HC1 and those for other elements were acidified with ultrapure HNO3. All samples were stored frozen. Sediment was collected using a plastic scoop to acquire surficial sediments including organic detritus (hereafter called detritus), but not pebbles or plant material. The scoop and
488
S.J. Hamilton, K.J. Buhl and P.J Lamothe
sample containers were rinsed in ambient water for sufficient time to condition the equipment to ambient conditions prior to sample collection. After sediments had settled, excess water was removed and the sample stored frozen. One sample was analyzed for Se and Hg and a second sample for other elements. Submerged aquatic plants (white-water buttercup, Ranunculus longirostris) were collected from each site by hand. The sample consisted of leaf whorls removed from stems using plastic or stainless steel forceps. Two sets of plant samples were collected from each site, squeezed to remove excess water, weighed, bagged in Whirl-Pak bags, labeled, and stored frozen. One composite set was analyzed for Se concentrations and the other set analyzed for other element concentrations. Aquatic invertebrates were sieved from bed substrate materials collected either by D-frame kick nets or by removing large stones with attached invertebrates. Substrate was placed in large polypropylene trays and invertebrates separated from substrate using forceps or glass tubes with suction bulbs. No aquatic invertebrates were collected from lower East Mill Creek (due to their paucity) or lower Slug Creek (not sampled). Two sets of invertebrate samples were collected from each site, separated by taxa group, weighed, bagged in WhirlPak bags, labeled, and stored frozen. Each composite invertebrate sample contained about half the available taxa weight. One composite set was analyzed for Se concentrations and the other set analyzed for other element concentrations. An opportunistic sample of crane fly nymphs (Tipulidae) was collected from Angus Creek headwaters at the base of Wooley Valley mine Unit 4 waste pile and analyzed like the other invertebrates. Fish were collected by electrofishing using a Coffet Mark-10 electroshocker. The equipment was rinsed in ambient water for sufficient time to condition the equipment to ambient conditions. Two sets of fish samples were collected from each site, euthanized with MS-222 (tricaine methanesulfonate), identified to species, measured for total length and weight, bagged in Whirl-Pak bags, and stored frozen. No fish were collected from lower East Mill Creek in spite of substantial effort. One set of fish was analyzed for Se concentrations in whole body and the other set analyzed for other element concentrations. The species collected included cutthroat trout (Oncorhynchus clarki), brook trout (Salvelinus fontinalis), mottled sculpin (Cottus bairdi), longnose dace (Rhinichthys cataractae), speckled dace (Rhinichthys osculus), and redside shiner (Richardsonius
balteatus).
Water quality analyses and flow measurement Ambient-water samples collected at each study site were analyzed for general water quality characteristics according to standard methods (American Public Health Association and other associations, 1995). Site water was analyzed in situ for conductivity, pH, temperature, and dissolved oxygen concentration. Immediately after arrival of water samples at the mobile laboratory, conductivity, pH, alkalinity, hardness, Ca, Mg, and temperature were determined. A subsample of 200 mL was collected and stored at 4~ with no preservative, and transported to USGS Laboratory
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
489
in Yankton for analysis of sulfate and chloride. A second subsample of 125 mL was collected, acidified with 0.5 mL concentrated H2SO4, and transported to Yankton for analysis of ammonia concentrations. Ammonia was measured using ion-selective electrodes following the procedures for low concentration measurements of the electrode manufacturer (Orion Research, 1990, 1991; Analytical Technology Incorporated Orion, 1994). Chloride was measured by the mercuric nitrate titration method (Hach Company, 1997).
E l e m e n t analys&
Water, surficial sediment, aquatic plants, aquatic invertebrates, and fish were analyzed for Se concentrations by atomic absorption spectroscopy-hydride generation (AA-HG; Environmental Trace Substances Laboratory (ETSL), University of Missouri). Analyses incorporated appropriate quality assurance/quality control (QA/QC) procedures such as standardizing equipment with certified reference material, determination of limit of detection, analysis of reagent blanks, spiked samples (termed % recovery of digested spike), duplicate analysis samples (termed % relative standard deviation, RSD), certified reference materials, reference materials spiked before digestion (termed % recovery of reference material), and reference material spiked after digestion (termed % recovery of reference standard). Analysis for Se concentrations was based on Method 7000 of the US Environmental Protection Agency (1983), and results are reported on a dry weight basis. A subsample of sediment was digested using a four-acid procedure and analyzed for concentrations of Se by AA-HG and Hg by cold vapor AA (USGS lab, Denver). All analyses incorporated QA/QC including standardizing analytical instruments using certified reference materials, determination of the limit of detection, and analysis of certified reference standards. Results are reported on a dry weight basis. Water, surficial sediment, aquatic plant, aquatic invertebrate, and fish samples were analyzed for other element concentrations by inductively coupled argon plasma atomic emission spectrophotometry (ICP) for AI, As, Ba, Be, B, Cd, Cr, Cu, Fe, Pb, Mg, Mn, Mo, Ni, Sr, V, and Zn. Analyses were conducted by ETSL with QA/QC as described above. Analysis was based on Method 6010 of the US Environmental Protection Agency (1983) and results are reported on a dry weight basis. A subsample of sediment was digested using a four-acid procedure and analyzed by ICP by the USGS lab, Denver, for Ca, Mg, P, K, Na, AI, Sb, As, Ba, Be, Bi, Cd, Ce, Cs, Cr, Co, Cu, Ga, Fe, La, Pb, Li, Mn, Mo, Ni, Nb, Rb, Sc, Ag, Sr, T1, Th, Ti, U, V, Y, and Zn, with QA/QC as listed above.
Statistical analyses
Data were analyzed using computer programs (Statistical Analysis System, Inc., 1985) to determine the relation between various measures made during the study. The Pearson correlation coefficients were determined for the relations between water-quality characteristics
490
S.J. Hamilton, K.J. Buhl and PJ. Lamothe
and Se concentrations in water, sediments, aquatic plants, aquatic invertebrates, and fish. For fish residue data for each sample location, the geometric mean was used in the correlation analyses with other variables.
RESULTS AND DISCUSSION
Quality assurance~quality control o f chemical analyses In general, the limit of detection (LOD) was 0.2 txg g-l for Se in surficial sediment, aquatic plant, aquatic invertebrate, and fish matrices and 0.5 txg L-l for water (see CD, Appendix 2, Table A). The procedure blank had a concentration less than the LOD. The percent relative standard deviation (duplicate preparation and analysis) ranged from
L._.J
I',~
Lm.J
~
~
I~
C
~
,m'
('3
r~
~
r
r
.'~
0
s~
r
0
>~
e,~
0
~,,Jo
t'rl ~
~-~
,m,
,_,. r~
~
i__a
Lm.J
A
~mJ
A
,m,
,,m
A
L,._,a
~
A
A
~
A
~
A
,M
A ~
,m~m~
A
I~ ~
.1~
....I
C
l"
r~
c
.-]
S~
,.,
,.,~
(:~ ~
~o
r~
r
I~
O0
,~,,o
r L_.J
A
>
v ,.Q ,~
r
0
r~
~o
e~
492
S.J Hamilton, K.J Buhl and P.J Lamothe
TABLE 18-III Water-quality characteristics measured in situ and in the USGS Mobile, ID lab at the nine locations listed in Table 18-I Measure
ACM
UEMC
LEMC
TC
USC
LSC
DVC
ShpC
LBR
8.4 380
8.3 357
8.4 353
8.3 362
8.2 388
8.1 399
8.1 548
8.3 359
7.8 407
198
210
206
215
244
234
227
214
256
4.3
4.3
5.3
4.8
4.3
2.4
0
5.3
0
8.5
9.1
8.7
10.9
9.3
10.6
10.2
9.6
9.1
93
84
85
102
86
111
111
87
98
20
12
23
12
12
22
19
!1
19
1.63
1.48
0.78
3.08
0.92
2.96
0.80
5.03
87.5
pH Conductivity (ixmhos cm -I ) Bicarbonate (mgL -I) Carbonate (mgL -I) Dissolved oxygen (mgL -I) % Saturation dissolved oxygen Water tcmperaturc (~ Discharge (cfs)
TABLE 18-IV Se concentrations for water (txg L ~) and for sediment, aquatic plants, and aquatic invertebrates (txg g ~ dry weight); collected from nine locations listed in Table 18-I and site ACH ~ Matrix Water Sediment Aquatic plant Aquatic invertebrate
ACH NS 2 NS NS 186
ACM UEMC <0.5 1.0 0.9 7.2
29.9 32.5 30.2 35.7
LEMC
TC
USC
LSC
DVC
ShpC
LBR
14.9 45.8 74.1 NS
<0.5 0.8 0.8 2.9
<0.5 0.8 1.1 4.9
<0.5 1.7 1.5 NS
<0.5 5.2 3.8 19.5
<0.5 0.5 0.6 2.6
<0.5 2.1 4.5 5.4
(n = 1). Analyses from ESTL. IACH: crane fly nymphs collection site at Angus Creek headwaters. 2NS: not sampled.
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
493
workshop on Se aquatic toxicity and bioaccumulation discussed the technical issues underlying the freshwater aquatic life chronic criterion for Se, and concluded that water was a poor choice (US Environmental Protection Agency, 1998b). The growing body of Se literature, which emphasizes the toxicological importance of the dietary route of Se exposure, requires a reassessment o f the continued use of a water-based criterion (Hamilton, 2002).
Other elements
Concentrations of other elements in water were comparable among the nine sites (Table 18-V). Although upper and lower East Mill Creek had elevated Se concentrations, those waters were not among the highest in other element concentrations, except Sr. Dry Valley Creek, relative to the other eight sites, had elevated A1, Fe, Mg, and Mn. However, no element concentrations in water seemed elevated to concentrations of concern. Concentrations of elements are generally the basis of water quality standards issued by the US Environmental Protection Agency (1998a, 1999). However, investigations indicated that dietary routes of exposures of elements were important in discerning effects on biota
TABLE 18-V Element concentrations (l~g L ~ n -- 1) in water from nine locations listed in Table 18-I Element AI As Ba Be B Cd Cr Cu Fe Pb Mg Mn Mo Ni Sr V Zn
ACM 46 14 37 <0.8 48 <3 8 <2 103 <10 12,900 86 <5 12 199 <2 <5
UEMC
DVC
ShpC
LBR
54 26 42 50 20 89 13 11 11 16 12 13 24 26 56 67 39 42 <0.8 <0.8 <0.8 <0.8 <0.8 <0.8 48 37 43 46 40 51 <3 <3 <3 <3 <3 <3 7 8 7 8 8 7 <2 <2 <2 <2 <2 12 89 90 92 106 94 164 <10 <10 <10 <10 <10 <10 14,000 13,200 11,200 13,000 13,000 17,300 4 5 79 28 31 70 <5 <5 <5 <5 <5 <5 9 9 9 9 18 13 245 234 150 136 230 203 <2 <2 <2 <2 <2 <2 <5 <5 <5 <5 <5 <5
63 11 60 <0.8 47 <3 8 <2 90 <10 12,700 10 <5 9 111 <2 <5
98 11 54 <0.8 48 <3 8 <2 106 <10 16,600 42 <5 9 208 <2 <5
Analyses from ESTL.
LEMC
TC
USC
LSC
494
S.J. Hamilton, K.J. Buhl and P.J Lamothe
(reviewed in Hamilton and Hoffman, 2003). For example, Kiffney and Clements (1993) reported that monitoring concentrations of Cd, Cu, and Zn in aquatic invertebrates was a better indicator of element bioavailability in the Arkansas River of Colorado, which was impacted by acid-mine drainage, than element concentrations in water.
Comparison to other Idaho data
The Idaho Mining Association Selenium Subcommittee (Selenium Subcommittee) investigated concentrations of Se, Cd, Mn, Ni, V, and Zn in water from numerous sites in the Southeastern Idaho Phosphate Resource Area and concluded that Se was the major contaminant of potential concern (Montgomery Watson, 1999). In May 1998, Se concentrations in water at 12 of 37 stream sites exceeded the US Environmental Protection Agency criterion of 5 ixgL -l, whereas in September 1998 only one stream, East Mill Creek (32 txg L-l), exceeded the criterion (Montgomery Watson, 1999). In the May 1998 sampling, the stream sites exceeding the criterion included five on the Blackfoot River (5-12 ixgL- l), Trail Creek (8.7 txgL- 1), Dry Valley Creek (5.6 ixgL- l), and two on East Mill Creek (210 and 260 ixgL-l). The values reported by Montgomery Watson (1999) were higher than those measured here. Montgomery Watson (2000) continued measuring Se concentrations in waters of the Blackfoot River in 1999. Concentrations in the lower Blackfoot River near our sampling site were 6.7 txg L-! in May, 2.1 lug L- i in June, 2.4 lug L- ~ in July, and 1.5 lug L- ! in August, which shows the variability over time that can occur in the fiver (see Presser et al., Chapter 16). Montgomery Watson (2000) reported similar variability in Dry Valley Creek: 49 lug L-i in May, 6.8 tug L-i in June, 2.7 lug L-! in July, and 1 Ixg L-l in August. In May 1999, Dry Valley Creek (49 lug L-l) and Spring Creek (46 lug L-i) were major Se contributors to the Blackfoot River. Above Spring Creek the Blackfoot River had < 1 Ixg L-!, whereas below Spring Creek concentrations from upstream to downstream were 9.0, 7.4 (above Dry Valley Creek), 7.9 (below Dry Valley Creek), 9.8, 7.2, 8.2, and 6.7 Ixg L-i. Thus, substantial contamination of the Blackfoot River occurred during 1999. This contamination was evidenced in Se concentrations in sediment, aquatic plants, aquatic invertebrates, and fish measured here. Montgomery Watson (200 l a,b) reported additional Se concentrations in water sampled in September 1999 and May 2000. Most water samples in September 1999- April 2000 contained < 5 Ixg L-~, except for Dry Valley Creek, which contained 12 lug L-~ above the Blackfoot River, 270 ixgL -! downstream of Maybe Creek, and 120 ixgL -i upstream of Maybe Creek; and East Mill Creek, which contained 19 Ixg L-~ (Montgomery Watson, 200 l a). In May 2000, Se concentrations of > 5 Ixg L-l were reported in the Blackfoot River (5.5-7.1 Ixg L-i) and several creeks including State Land Creek (10 lug L-l) and two tributaries (16 txgL-i), North Fork Wooley Valley Creek (98 IxgL-l), lower Slug Creek (6.3 IxgL-l), Dry Valley Creek (8-87 txgL-i), Angus Creek (6.5 ixgL- l), Spring Creek (28 txgL-l), and East Mill Creek (400 ixgL -l) (Montgomery Watson, 200 l b). These data demonstrate continued Se contamination of the Blackfoot River watershed. The data also demonstrate that Se contamination occurs primarily during spring runoff (see Presser et al., Chapter 16).
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
495
Sediment Selenium
Selenium concentrations in surficial sediment were relatively low at five sites (<2 Ixg g-l), moderately elevated at the Dry Valley Creek and lower Blackfoot River (2.1-5.2 Ixg g-l), and very elevated at the upper and lower East Mill Creek (33-52 Ixg g-l) (Tables 18-IV and 18-VI). Sheep Creek and Trail Creek had the lowest Se concentrations in surficial sediment. Selenium concentrations in surficial sediment from upper Slug Creek, Sheep Creek, and Trail Creek were below the no hazard rating of < 1 I~g g-1, and lower Slug Creek and Angus Creek were <2 ~ g g - l , the minimal hazard rating of 1-2 Ixgg-t proposed by Lemly (I 995), and below the no effect concentration of <2 Ixg g-1 proposed by Stephens et al. (1997). However, Presser et al. (1994) and Moore et al. (1990) used 0.5 Ixg g-l as a reasonable Se concentration in sediment to represent the threshold between uncontaminated background conditions and environments with elevated Se concentrations. Selenium in surficial sediment from the lower Blackfoot River site (2.1 Ixg g-l) falls in the concentration-of-concern range of > 2 - 4 txg g-~ of Stephens et al. (1997) and the moderate hazard of 2-4 Ixg g-i of Lemly (1995). Selenium in surficial sediment from the Dry Valley Creek (5.2 ~ g g - l) and upper and lower East Mill Creek (32.5-45.8 txgg -I) sites were higher than the toxicity threshold of >4 Ixg g-~ proposed by Stephens et al. (1997) and the high hazard rating of > 4 Ixg g-l proposed by Lemly (1995). Elevated Se concentrations in surficial sediment at upper and lower East Mill Creek suggest a substantial contamination concern, as do the elevated Se concentrations in Dry Valley Creek. The peer consultation workshop on Se aquatic toxicity and bioaccumulation (US Environmental Protection Agency, 1998b) discussed the possibility of a sedimentbased criterion. However, they concluded that sediment was a poor criterion because of spatial heterogeneity of Se deposition, variable water retention time, variable volatilization rates, heterogeneity of the benthic phytoplankton community and benthic invertebrates, and variable feeding habitats of higher trophic organisms (US Environmental Protection Agency, 1998b). Here, the strong correlation for Se concentrations between surficial sediments and aquatic plants (0.96-0.97) and between surficial sediment and aquatic invertebrates (0.92-0.94) indicate ready movement of Se among aquatic ecosystem components and accumulation in the food web (Table 18-VII). Similar movements of Se through the food web were reported in two field studies of seleniferous areas of the upper Colorado River (Hamilton et al., 200 l a,b). Specific to Se, Woock (1984) demonstrated in a cage study with golden shiner (Notemigonus crysoleucas) that fish with access to bottom sediments accumulated more Se than fish held in cages suspended about 1.5 m above the sediment. That study revealed that effects in fish were linked to Se exposure via sediment, benthic invertebrates, detritus, or a combination of sediment components.
TABLE 18-VI Element concentrations ( ~ g g - dry ' weight; n = 1) in sediment from nine locations listed in Table 18-1; analyses from USGS lab Element
ACM
UEMC
LEMC
TC
USC
LSC
DVC
ShpC
LBR
498
S.J. Hamilton, K.J. Buhl and PJ. Lamothe
TABLE 18-VII Significant (P < 0.05) Pearson correlation coefficients for various aquatic ecosystem components Ecosystem component
Sediment (USGS lab)
Sediment (ESTL)
Aquatic plant
Aquatic invertebrate
Ecosystem component Sediment (ESTL)
Aquatic plant
Aquatic invertebrate
Fish
Se
0.99
Se
0.97
0.69
0.92 0.85 0.92 0.92 0.90
Se
Ba
Se A1 As Ba Cr
Be Cd Cu
0.93 0.98 0.97
Cd
0.73 Cu Fe
0.78 0.91
Pb Mg Mn Ni Sr V Zn
0.78 0.76 0.83 0.99 0.80 0.96 0.99
0.87 0.79
Mg Ni
0.71
Mn Ni
Zn
0.91
V Zn
0.93 0.83
Se
0.96
Se AI Cu Mn Ni V Zn
0.94 0.78 0.79 0.87 0.84 0.92 0.83
Se Cd Mn Mo Ni Zn
0.91 0.99 0.85 0.83 0.98 0.84
Ni
0.74
Zn
0.90
0.95
-0.66
Se
0.96
Cu
0.79
V
0.81
Se
0.95
Mn
0.71
Se Cu
0.99 0.84
Other elements
Concentrations of other elements in surficial sediments reported by USGS and ESTL are c o m p a r a b l e (Tables 18-VI and 18-VIII) with correlation coefficients o f > 0 . 9 0
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
499
TABLE 18-VIII Element concentrations (Ixg g-i dry weight; n = 1) in sediment from nine locations listed in Table 18-1; analyses from ESTL Element
ACM
UEMC
LEMC
A1 As Ba Be B Cd Cr Cu Fe Pb Mg Mn Mo Ni Sr V Zn
1650 7.7 131 0.8 12 2.8 33 8.8 184 8.1 453 280 <0.5 18 70 39 69
1020 6.6 7.8 0.7 9.0 5.2 10.2 18 161 5.6 461 117 1.9 43 78 58 175
1320 8.0 9.4 0.7 13 4.4 12 18 175 5.8 498 107 1.9 43 69 62 182
TC
USC
LSC
DVC
ShpC
LBR
1160 6.5 11.2 0.6 9.4 0.6 6.7 5.9 156 4.8 388 91 <0.5 11 40 15 41
1010 3.9 85 <0.1 7.3 1.9 20 7.0 710 5.0 258 51 <0.5 12 29 16 61
1090 7.5 10.4 0.5 8.7 1.4 7.6 5.4 130 4.3 315 173 <0.5 12 94 21 48
1330 4.5 10.9 0.7 8.3 5.5 7.7 10.7 171 6.3 363 142 <0.5 54 34 24 555
944 6.3 7.8 0.6 9.0 0.4 6.2 5.8 140 5.1 367 29 <0.5 12 20 15 34
1240 2.0 108 <0.1 9.0 0.8 16 5.7 782 3.9 516 99 <0.5 13 73 16 41
(P < 0.001-0.003) for Se, Be, Cd, Cu, Ni, V, and Zn; 0.60-0.89 (P = 0.01-0.006) for Ba, Pb, Mg, Mn, and Sr; and <0.59 (P = 0.26-0.76) for AI, As, Cr, and Fe (Table 18-VII). The difference in concentrations of some elements was probably due to the four-acid digestion used at the USGS lab and the single-acid digestion at ESTL. Concentrations of other elements in surficial sediments followed a similar trend as Se in sediments (Tables 18-VI and 18-VIII). Upper and lower East Mill Creek tended to have the highest concentrations of Sb, Cd, Cr, Cu, Mo, Ni, Ag, V, and Zn, whereas Dry Valley Creek had a moderate amount of these elements relative to the other six sites. Other sites such as lower Slug Creek had elevated Mn and Sr compared to upper Slug Creek and Angus Creek had elevated B, Mn, Sr, V, and Zn compared to upper Slug Creek, thus suggesting some element contamination in sediments at these sites. Mercury was not elevated at any of the nine sites (Table 18-VI). The magnitude of difference in concentrations at upper and lower East Mill Creek sites compared to those in the non-impacted upper Slug Creek site ranged from 1.8- to 11-fold. In contrast, the magnitude of difference for Se concentrations between upper and lower East Mill Creek and upper Slug Creek ranged from 35- to 57-fold, thus suggesting a major disparity between Se enrichment in the East Mill Creek compared to upper Slug Creek.
500
S.J. Hamilton, K.J. Buhl and PJ. Lamothe
Of all the elements measured, Se was the most bioaccumulative in the food web and the element of principal concern for effects in biota. The sediment in aquatic ecosystems is an important pathway of element movement through the food web (Seelye et al., 1982). Sediment contains the most concentrated pool of elements in the aquatic environment and many types of aquatic organisms ingest sediment during the foraging process (Luoma, 1983), for example, fish (Dallinger and Kautzky, 1985; Dallinger et al., 1987; Kirby et al., 2001a,b).
Comparison to other Idaho data
Overall, the elevated concentrations of Se and other elements in sediments from several streams in the Blackfoot River watershed that were reported by TRC Environmental (1999), Montgomery Watson (1999, 200 l a), and in the present study suggest widespread contamination of the aquatic environment. Selenium concentrations in sediment from Maybe Creek, a tributary of Dry Valley Creek, near its mouth were 261 Ixg g-i (TRC Environmental, 1999), indicating that much of the Se loading in Dry Valley Creek comes from Maybe Creek. They reported that other segments of Maybe Creek contained 12-77 txg g-~ of Se in sediment. The Selenium Subcommittee investigated concentrations of Se, Cd, Mn, Ni, V, and Zn in sediment from numerous sites in the Southeastern Idaho Phosphate Resource Area in September 1998 (Montgomery Watson, 1999). Out of 54 sites investigated, 11 had Se concentrations in sediment of 2-4 ~g g-~, thus placing them in the concentration-of-concern range proposed by Stephens et al. (1997) and the moderate hazard range of Lemly (1995). These sites included Slug Creek, Dry Valley Creek, Rasmussen Creek (tributary to Angus Creek), and East Mill Creek. Three sites had sediment values greater than 4 p~gg - I (State Land Creek, Sage Creek below Smoky Canyon Mine, North Fork Sage Creek below Pole Creek), which places them above the toxicity threshold of Stephens et al. (1997) and the high hazard of Lemly (1995). The Se concentrations in sediment reported by Montgomery Watson (1999) in East Mill Creek (2.9 I~g g-i) were substantially lower than those in the present investigation (35-52 p~gg-i), whereas their values for Dry Valley Creek (3.3 l~gg -I) were slightly lower than our values (4.3-5.2 i~gg-l). Other sediment contents reported by Montgomery Watson (1999) for similar sites were similar to those reported here. Montgomery Watson (2001a) reported elevated Se concentrations in sediment collected in September 1999 from State Land Creek (2.1 i~g g-~), Pedro Creek (3.1 i~g g- ~), Dry Valley Creek (3.9 ~gg-i), Rasmussen Creek (2.2 p~gg-i), and the Blackfoot River (2.1-3.0 ~gg-~), which were in the concentration-of-concern range of Stephens et al. (1997) and the moderate hazard of Lemly (1995). They reported high Se concentrations in sediment from Dry Valley Creek (6.2 ~zgg-1), Angus Creek (5.1 ~g g-!), and East Mill Creek (5.0 p~gg-i), which were in the toxicity threshold range of Stephens et al. (1997) and the high hazard of Lemly (1995).
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
501
Aquatic plants Selenium
Selenium concentrations in aquatic plants followed a similar pattern as in surficial sediments: low at five sites (0.6-1.5 Ixg g-l, Sheep Creek, Trail Creek, Angus Creek, upper Slug Creek, lower Slug Creek), moderately elevated at Dry Valley Creek and lower Blackfoot River (3.8-4.5 ixg g- l), and very elevated at the upper and lower East Mill Creek (30-74 i~gg -l) (Table 18-IV). Sheep Creek and Trail Creek had the lowest Se concentrations in aquatic plants and surficial sediment. The correlation coefficient between Se concentrations in sediment and aquatic plants is 0.97 (P < 0.0001; Table 18-VII). No guidelines were found in the literature proposing toxicity threshold concentrations for Se in aquatic plants that might be hazardous to aquatic organisms. However, most domestic animals exhibit signs of Se toxicosis when consuming terrestrial vegetation containing ->3-5 ixgg -! Se (National Research Council, 1980; Eisler, 1985; Olson, 1986). The low Se concentrations in aquatic plants from upper Slug Creek, Trail Creek, Sheep Creek, Angus Creek, and lower Slug Creek might be considered near background for the Blackfoot River watershed. By comparison, Se concentrations at Dry Valley Creek (3.8 Ixg g-I) and lower Blackfoot River (4.5 i~gg -l) were elevated and those at upper and lower East Mill Creek (30-74 ixgg -l) were substantially elevated. Although fish typically do not feed on macrophytes, dead macrophytes become an important contributor to detrital food. Benthic invertebrates readily accumulate Se from detritus (Alaimo et al., 1994), which in turn is bioaccumulated through the food web to predators such as fish. Saiki et al. (1993) concluded that high concentrations of Se in aquatic invertebrates and fish in Se-contaminated areas of central California were the result of food-chain transfer from Se-rich detritus rather than from other pathways. Likewise, elevated Se concentrations in aquatic plants from four of the stream sites in the Blackfoot River watershed probably contribute to Se transfer in the aquatic food web.
Other elements
Concentrations of elements in aquatic plants follow a similar trend as Se in surficial sediments (Table 18-IX). Lower East Mill Creek tended to have the highest concentrations of A1, As, Cd, Cu, Fe, Pb, Mg, Mo, Ni, Sr, V, and Zn, whereas Dry Valley Creek had an intermediate amount of these elements relative to the other seven sites. In contrast, lower Slug Creek tended to have slightly higher concentrations of A1, As, B, Ba, Cr, Fe, and Sr than Dry Valley Creek. This relative elevation of elements in aquatic plants in lower Slug Creek compared to Dry Valley Creek was not present in sediment data. In general, aquatic plants from Sheep Creek and Trail Creek tended to have low element concentrations compared to the other sites, except for Ba, which was elevated in aquatic plants from Trail
502
S.J. Hamilton, K.J Buhl and P.J Lamothe
TABLE 18-IX Element concentrations (Ixg g-i dry weight; n = 1) in aquatic plants from nine locations listed in Table 18-I Element
ACM
UEMC
LEMC
TC
USC
LSC
DVC
ShpC
LBR
A1 As Ba Be B Cd Cr Cu Fe Pb Mg Mn Mo Ni Sr V Zn
168 1.4 92 < 0.1 12 2.3 3.7 5.3 103 <2 288 888 2.2 7.4 64 3.8 32
69 1.8 14 < 0.1 10 22 8.1 4.4 64 <2 269 85 1.7 7.7 53 3.6 166
1796 7.6 94 < 0.1 28 13 13.8 18 834 4.8 575 75 2.1 41 99 79 197
465 1.9 156 < 0.1 14 0.6 7.1 4.7 267 <2 334 749 0.8 5.7 39 8.1 22
290 2.5 65 < 0.1 12 3.2 8.5 5.4 163 <2 306 544 1.6 7.6 36 6.0 40
699 3.4 281 < 0.1 38 1.1 18 4.0 408 2.4 375 266 < 0.5 8.1 61 15 34
549 1.6 106 < 0.1 17 1.9 15 5.0 315 2.3 389 419 1.4 27 53 14 300
90 1.0 32 < 0.1 10 1.7 2.4 3.5 68 <2 232 242 0.5 4.0 21 2.1 29
153 1.4 60 < 0.1 14 3.2 3.5 5.6 113 <2 416 442 0.8 6.4 45 4.0 115
Creek. Although upper Slug Creek is considered the reference site, it had higher concentrations of some elements in aquatic plants than did Sheep Creek plants. There were strong correlation between element concentrations in aquatic plants and surficial sediments for Cd, Ni, and Zn (0.71-0.91, P < 0.03) and for V (0.59, P < 0.10) using the USGS sediment data (Tables 18-VI and 18-VII) and also for Cu and Sr using the ESTL data (Tables 18-VII and 18-VIII). Elements accumulate in aquatic plants from both the water column (Bryson et al., 1984; Devi et al., 1996) and from sediment (Cherry and Guthrie, 1977; Dallinger and Kautzky, 1985; Dallinger et al., 1987). Significant correlations for several elements (Cd, Cu, Ni, Sr, V, Zn) between surficial sediments and aquatic plants suggest a strong interconnectedness in element cycles. Although few herbivores feed on aquatic plants directly, when rooted aquatic plants die, greater than 90% of their biomass comprises 90% of the detrital-food chain, whereas the remaining 10% is from algal and animal detritus (Teal, 1962; Mann, 1972). Much of the nutritional content in detritus comes from microbe enrichment and metabolic products, which add proteins and amino acids to detritus (Odum and de la Cruz, 1967; Foda et al., 1983). Plant litter and other coarse debris that enter a stream are a major source of energy that fuels higher trophic levels (Allan, 1995).
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
503
Thus, uptake of elements by aquatic plants might seem unimportant; however, elements in dead-plant material can play an important role in movement of mass and energy through the detrital food web to aquatic invertebrates and fish.
Comparison to other Idaho data
Periphyton (attached algae and detritus), plankton, and submerged macrophyte data demonstrate that aquatic plants were accumulating Se from water and sedimentary sources. Montgomery Watson (2001 a,b) acknowledged that submerged aquatic plants were efficient accumulators of Se. Their measurements are similar to data presented here. Elevated Se concentrations were found for grasses (mean 64 txg g-l), forbs (78 ~g g-l), and shrubs (11 Ixg g-I) in Maybe Creek (TRC Environmental, 1999). Montgomery Watson (1999) did not collect or analyze element concentrations in aquatic plants, however, Montgomery Watson (2001a) reported Se concentrations in periphyton collected from artificial substrates placed in streams between September and October 1999. Elevated Se concentrations were found in periphyton in the Blackfoot River (3.0 ixg g- l), Angus Creek (3.3-9.2 txgg-l), Spring Creek (4.2-7.5 ixgg-l), and very high values in East Mill Creek (12-25 ixgg-l). Montgomery Watson (2001b) reported Se concentrations in periphyton collected from artificial substrates placed in streams between May and June 2000, but fewer streams than investigated in Montgomery Watson (2001a). Elevated Se concentrations were found in the Blackfoot River (4.3 Ixg g-l) and Angus Creek (6.0 Ixg g-l). Plankton samples collected from various sites in Blackfoot Reservoir contained Se concentrations of <-1.5 Ixg g-! in September 1999 (Montgomery Watson, 2001a). However, in the May 2000 sampling, 9 of 12 samples contained a geometric mean Se concentration of 3.3 Ixg g-i (Montgomery Watson, 2001 b). Submerged macrophytes were collected in September 1999 from numerous stream sites in the Blackfoot River watershed and analyzed for Se concentrations (Montgomery Watson, 2001 a). Several samples had elevated concentrations that range from 3.2 to 4.8 txg g-i, 10 samples had high concentrations ranging from 5.1 to 8.8 ixg g-i, and one site, East Mill Creek, with very high concentrations ranging from 31 to 46 Ixg g-i. Submerged macrophytes collected in May 2000 contained similar Se concentrations to those sampled in the September 1999 (Montgomery Watson, 2001 b). A native bryophyte that was collected from a seep at the base of the Wooley Valley phosphate mine Unit 4 waste-rock pile in the headwater area of Angus Creek contained very elevated concentrations of several elements including Cd (160 txgg-1), Co (180 Ixg g-~), Cr (210 ixgg-t), Mn (33,000 txgg-~), Ni (2000 txgg-l), V (1000 ixgg-~), Zn (11,000 txgg-t), and Se (750 ixgg -l) (Herring et al., 2001). This site and others on Angus Creek were monitored for element accumulation in late spring and late summer 1999 using an introduced bryophyte, Hygrohypnum ochraceum (Herring et al., 2001). The same elements as present in the native bryophyte accumulated in the introduced bryophyte, but Se was the most enriched of the elements measured.
504
s.J. Hamilton, K.J. Buhl and P.J. Lamothe
Aquatic invertebrates Selenium
The opportunistic sample of crane fly nymphs collected at the Angus Creek headwater contained 186 ~g g-1 Se (Table 18-IV). Aquatic invertebrates collected from Sheep Creek and Trail Creek had low Se concentrations (2.6-2.9 ~ g g - l ) , upper Slug Creek, lower Blackfoot River, and Angus Creek had moderate concentrations (4.9-7.2 p~gg-l), and Dry Valley Creek and upper East Mill Creek had elevated concentrations (19-36 p~gg-l). Correlation coefficients between Se concentrations in aquatic invertebrates were strong with aquatic plants (0.91, P < 0.004), and surficial sediment (0.94, P < 0.002; Table 18-VII). In the hazard assessment protocol proposed by Lemly (1995), Se concentrations in benthic invertebrates of <2 p~gg-1 represent no identifiable hazard, 2-3 ~g g-l minimal hazard, 3-4 I~g g-i low hazard, 4-5 ~g g-! moderate hazard, and >5 p~gg-i high hazard. Using these values suggests that Se concentrations in invertebrates at upper Slug Creek, lower Blackfoot River, Angus Creek, Dry Valley Creek, and upper East Mill Creek probably were adversely affecting sensitive larval fish, but not necessarily older life stages such as subadults or adults. Selenium concentrations in aquatic invertebrates from Sheep Creek and Trail Creek (2.6-2.9 i~gg -I) were the lowest of the sites, but were close to the proposed dietary Se threshold of 3 I~g g- ~ (Lemly, 1993, 1996b; Hamilton, 2002). Selenium concentrations of 4.6 ~g g-i in zooplankton caused nearly complete mortality of razorback sucker in about 10-13 days (Hamilton et al., 2001a,b). Several other studies summarized in Hamilton (2002) reported that dietary Se concentrations of 4-6 ~ g g - ! caused adverse effects in larval fish. Consequently, the moderate concentrations in upper Slug Creek, lower Blackfoot River, and Angus Creek (4.9-7.2 i~gg-l), and the elevated concentrations in Dry Valley Creek and upper East Mill Creek (19-36 I~g g-i) were of concern to the health of fishery resources. The high Se concentrations in crane fly nymphs (186 I~g g-l) probably pose an acute toxicity dietary hazard to predators that might feed on them. Sandhill cranes (Grus canadensis) have been observed feeding in the marshy area within a few meters of the location where the nymphs were collected (L. Stillings, US Geological Survey, personal communication). Although upper Slug Creek was the reference site and contained relatively low Se concentrations in water, surficial sediments, and aquatic plants, Se concentrations in aquatic invertebrates were elevated. Benthic invertebrates can be efficient accumulators of Se and can retain elevated concentrations over a long time. For example, Maier et al. (1998) reported that aquatic invertebrates contained Se concentrations of 1.7 p~gg-~ at pretreatment of a watershed with Se fertilizer and elevated concentrations during post-treatment monitoring, 4.0-5.0 Ixg g-l for six samplings over I 1 months. The linkage between Se concentrations in invertebrates and sediment and detritus was supported by the strong correlation between aquatic invertebrates and surficial sediments
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
505
(0.92-0.94) and between aquatic invertebrates and aquatic plants (0.91). Selenium concentrations in invertebrates and its bioaccumulation through the food web to higher trophic organisms such as fish have been reported by several investigators (Finley, 1985; Hamilton et al., 200 l a,b).
Other elements
Crane fly larvae nymphs collected from the Angus Creek headwater contained elevated concentrations of Cr, Fe, Ni, Sr, V, and Zn compared to invertebrates from the other eight sites (Table 18-X). Concentrations of elements in aquatic invertebrates from seven sites (none collected at lower East Miller Creek or lower Slug Creek) followed a similar trend to that found in surficial sediments and aquatic plants (Table 18-X). Low concentrations occur in aquatic invertebrate specimens from Sheep Creek and Trail Creek, moderate concentrations in upper Slug Creek and lower Blackfoot River, and elevated concentrations in upper East Mill Creek, Angus Creek, and Dry Valley Creek specimens. At the three sites with elevated concentrations, the elements generally elevated include A1, B, Ba, Cr, Fe, V, and Zn.
TABLE 18-X Element concentrations (~gg-~ dry weight; n = 1) in composite samples of aquatic invertebrates from seven locations listed in Table 18-I and site ACH ~ Element
ACH
ACM
AI As Ba Be B Cd Cr Cu Fe Pb Mg Mn Mo Ni Sr V Zn
49 2.6 14 < 0.1 4.3 12 31 12 374 <2 308 168 3.5 46 32 46 326
459 3.0 215 < 0.1 8.9 2.7 11 21 296 <2 238 434 1.5 16 11 11 330
UEMC 257 3.6 30 < 0.1 9.5 31 31 51 251 <2 191 29 1.0 12 21 18 595
TC
USC
DVC
ShpC
LBR
72 <2 112 < 0.1 6.8 1.0 2.7 11 68 <2 135 196 0.6 1.8 8.2 1.7 87
224 3.0 145 < 0.1 6.6 3.7 14 29 174 <2 175 109 0.7 6.9 47 5.7 143
381 2.7 140 < 0.1 8.2 1.7 12 17 262 <2 227 198 1.0 50 30 11 659
59 <2 22 < 0.1 3.7 0.6 2.6 16 59 <2 129 21 0.7 2.2 3.8 1.7 138
213 2.5 30 < 0.1 7.1 0.5 5.6 27 160 <2 181 68 0.6 6.3 17 5.5 127
1Crane fly nymphs collection site at Angus Creek headwaters.
506
S.J. Hamilton, K.J Buhl and P.J Lamothe
At upper East Mill Creek, other elevated elements include Cd, Cu, and Ni, whereas in Dry Valley Creek samples, Ni and Sr are elevated. There are strong correlations between element concentrations in aquatic invertebrates and surficial sediments for A1, As, Ba, Cr, Cu, Fe, Mn, Ni, V, and Zn (0.78-0.93, P < 0.05). There also are strong correlations for elements between aquatic invertebrates and aquatic plants for Cd, Mn, Mo, Ni, and Zn (0.83-0.99, P < 0.05). Several elements elevated in aquatic invertebrates from upper East Mill Creek, Angus Creek, and Dry Valley Creek were probably linked with concentrations in aquatic surficial sediment and aquatic plants.
Comparison to other Idaho data
Elevated Se concentrations have been reported in benthic invertebrates collected from ponds (110-390 txgg -l) and a flowing section (14 txgg -l) of Maybe Creek, a tributary of Dry Valley Creek (TRC Environmental, 1999). Those Se concentrations are substantially higher than those measured here in benthic invertebrates from Dry Valley Creek. Benthic invertebrates collected from various sites in Blackfoot Reservoir in September 1999 contain <--2 Ixg g-~ Se, except for three samples, which contain Se concentrations of 3.81 Ixg g-l (Montgomery Watson, 2001 a). However, 8 of 12 samples collected in May 2000 from Blackfoot Reservoir contain a geometric mean Se concentration of 7.8 Ixg g(range 5.3-12 txgg-~; Montgomery Watson, 2001b). Benthic invertebrates collected in September 1999 from numerous stream sites in the Blackfoot River watershed had moderately elevated Se concentrations in 5 of 26 samples (3.0--4.6 i~gg-l), elevated concentrations in five samples (5.0-15 txgg-l), and very elevated concentrations at East Mill Creek (72 ixgg -~) (Montgomery Watson, 2001a). In the May 2000 sampling, moderately elevated Se concentrations occurred in 11 of 42 samples (3.0-4.9 ixgg-l), 17 samples had elevated concentrations (5.0-37 ixgg-I), and East Mill Creek had 100, 120, and 170 txg g-i (Montgomery Watson, 2001 b). Selenium concentrations reported by Montgomery Watson (2001a,b) tended to be higher than those determined here for similar collection sites. The large number of sampies with substantial Se concentrations far above the proposed toxic threshold of 3 I~g g - I suggest that benthic invertebrate populations were highly contaminated with Se.
Fish Selenium
Six fish species were collected, but no one fish species was collected at all eight sites. Mottled sculpin were collected at six sites and Se in these fish followed the same pattern as present in the geometric mean of combined fish data (Table 18-XI). Speckled dace were collected at five sites and in general had the highest whole-body Se concentrations of the species collected. Although redside shiners were collected from only four sites, they
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
507
consistently had the lowest whole-body Se concentrations of the species collected at those sites. Geometric mean Se concentrations in fish were low from upper Slug Creek (4.0 Ixg g-l), Trail Creek (5.1 ixgg-l), Sheep Creek (5.2 txgg-l), and lower Slug Creek (5.3 ixgg-1), intermediate from Angus Creek (6.4 ixgg - l ) and lower Blackfoot River (7.8 txg g- l), and elevated from Dry Valley Creek (16.1 l~gg -1) and upper East Mill Creek (32.2 Ixg g - l ) (Table 18-XI). Selenium concentrations in fish from the eight sites, based on geometric-mean values, followed the same pattern of accumulation as in surficial sediments, aquatic plants, and aquatic invertebrates. This accumulation pattern is supported by other studies, which show that fish bioaccumulate Se primarily from the dietary route of exposure (Lemly, 1993, 1996b; Maier and Knight, 1994). Here, there were consistent differences in Se concentrations between fish species at a site. The particular feeding niche can influence the residues accumulated. Ney and Van Hassel (1983) reported that the benthic species fantail darter (Etheostoma flabellare) and blacknose dace (Rhinichthys atratulus) had the highest accumulations of Cd, Pb, Ni, and Zn, bottom-dwelling northern hog sucker (Hypentelium nigricans) and white sucker (Catostomus commersoni) had intermediate accumulations, and water-column dwelling redbreast sunfish (Lepomis auritus) and rock bass (Ambloplites rupestris) had the least accumulations. Others have reported that element residues vary among fish species, but variations were not conclusively related to food habits and trophic status (summarized in Wiener and Giesy, 1979). We found that speckled dace tended to have the highest Se concentrations of the six species collected, whereas redside shiner had the lowest concentrations. Speckled dace is a bottom browser that feeds on invertebrates and plant material (Lee et al., 1980), possibly detritus, and thus seemed to accumulate elevated elements similar to bottom-feeding
TABLE 18-XI Se concentrations (Ixg g-! dry weight; n = 1) in whole-body fish from eight locations listed in Table 18-I Species
ACM
UEMC
TC
USC
LSC
Brook trout Cutthroat trout Mottled sculpin Longnose dace Speckled dace
_l 6.7 7.1 7.0 -
42.7 24.3 -
4.4 5.9 6.3
2.4 3.7 7.2
. 5.8
Redside shiner Geometric mean
5.3 6.4
32.2
4.0 5.1
4.0
4.9 5.3
~Dash means not collected.
DVC .
ShpC
16.6 15.6 16.1
.
. 3.6 5.6 7.1 -
16.1
5.2
LBR 6.7 6.7 14.1 5.8 7.8
508
S.J. Hamilton, K.J Buhl and PJ. Lamothe
redear sunfish reported by Campbell (1994). Redside shiners are omnivores (Lee et al., 1980) and thus seemed to accumulate low element concentrations similar to omnivorous bluegill reported by Campbell (1994). We conclude that feeding niche differences result in different dietary exposures and more importantly, different Se bioaccumulation in fish.
Other elements
Several elements are elevated in whole-body fish from the eight sites (Table 18-XII). Similar to Se concentrations, highest element concentrations are in speckled dace and the lowest concentrations tended to occur in redside shiner. The elements that were somewhat elevated include A1, Ba, Cu, Mn, Ni, Sr, and Zn, but no site had fish with consistently elevated values based on geometric means (Table 18-XIII). Fish from upper Slug Creek, the reference site, had the highest geometric means for AI, Ba, Cr, and Mn. Upper East Mill Creek fish had the highest geometric-mean Cu concentrations. Lower Slug Creek fish had the highest geometric-mean Sr and Zn. There were few correlations between elements in fish (using the geometric mean for all fish at each site) and surficial sediment, aquatic plants, or aquatic invertebrates. There are three correlations with surficial sediment data: Cu (0.68, P = 0.06), Mg (-0.66, P = 0.08), and V (0.72, P = 0.07). For aquatic plant data, there is one correlation: Mg (0.71, P = 0.05). For aquatic invertebrate data there are two correlations: Cu (0.84, P = 0.02) and Mn (0.73, P = 0.06). Copper and V are the only two elements in surficial sediment and aquatic invertebrates that are consistently correlated with Se concentrations in fish. There seemed to be no parallel bioaccumulation of elements with Se in fish from the eight sites. In contrast to Se concentrations in fish, concentrations of some elements were highest in upper Slug Creek, which was the reference site. Only elevated Cu occurred in fish from upper East Mill Creek, which had the highest Se concentrations in fish as well as other aquatic ecosystem components, compared to the other sites. Selenium-dominant contamination found in the Blackfoot River watershed was also found in other contaminant investigations. For example, contaminated food chains in coal ash settling basins showed elevated AI, Co, Eu, Fe, La, Lu, Sm, Se, and Ti in biota, but only Se was of concern in terms of detriment to biota (Furr et al., 1979). Lemly (1985) reviewed information in 10 studies of potential causes of fishery problems at Belews Lake (16 species eliminated, two species present as adults only, one species recolonized, and one species unaffected), and of the 16 elements of concern only Se was present at elevated concentrations in water and fish. Montgomery Watson (1999) concluded that Se was the major element of concern associated with phosphate mining activities in the Blackfoot River watershed of southeast Idaho and not other elements (Cd, Mn, Ni, V, Zn). In the two reproduction studies with endangered razorback sucker, several elements (A1, Sb, As, B, Ba, Be, Bi, Cd, Cr, Co, Cu, Fe, Pb, Li, Mg, Mn, Mo, Ni, Si, Ag, Sr, T1, Sn, Ti, V, Zn) were elevated in water, zooplankton, sediment, and fish eggs, but only Se was high enough to cause adverse effects in fish (Hamilton et al., 2001 a,b).
TABLE 18-XI1 Element concentrations (pgg-' dry weight; n = 1) in whole-body fish from eight locations listed in Table 18-1 Element
ACM Cutthroat trout
Mottled sculpin
Longnose dace
Redside shiner
UEMC
TC
Cutthroat trout
Mottled sculpin
Longnose dace
Speckled dace
Redside shiner
Continued
TABLE 18-XI1 Continued Element
USC Brook trout
LSC Mottled sculpin
Speckled dace
Speckled dace
DVC Redside shiner
Cutthroat trout
Mottled sculpin
Speckled dace
Element
Al As Ba Be B Cd Cr Cu Fe Pb Mg Mn Mo Ni Sr V Zn
ShpC
LBR
Cutthroat trout
Mottled sculpin
Longnose dace
Mottled sculpin
Speckled dace
Speckled dace
Redside shiner
24 <2 1.5
1 04 <2 3.7
195 <2 8.8
69 <2 2.9
16 <2 8.3
127 <2 6.3
16 <2 2.7 <0.1 1.6 <0.3 1.O 1.8 17 <2 102 8 <0.5 1.6 18 0.3 128
512
.=.
~0 ,~
0
0
r~
q:
0
6 0
.N
T
r.~ ~0 ,~
0
r
0
V
""
"
o V
"
~q ~
o
o
~
~176
S J . H a m i l t o n , K.J. B u h l a n d PJ. L a m o t h e
" ~4 ~
~
" c~ V
o V
c~ V
--
,~
~
V
V
~
~
l -~-
" r
~
" o V
~t.
o V
~r,~ ~ , l
~
~
~
~
o ~ V
-
~ _
" ~
r
N
c~ ~ V
~
9o,,
o,.o V
o
e~
,-~
V
"~"
~"
e~
~
~
~
~',.
r
--.
~
"-
o V
V
V
m_ ~
V
V
~4 c~ V
V
V
~
o
" ~i ~
V
" ~
" o V
--
V
o:, o V
V
" c~
" --
' ~ V N o ~ o V
V
V
r~ c~
o V
~
~
'~
V
~
~
~4 ~
~
,'4 o
V
"
"~" V
~
"
" ~
c~ r V
,5 V
0
~
:~
~
o V
~
~
~
.,~
"~" V
r
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
513
The possibility of a tissue-based criterion was discussed at the peer consultation workshop on Se aquatic toxicity and bioaccumulation (US Environmental Protection Agency, 1998b). The conclusion was that the tissue-based criterion might be the best approach because tissue residues accounted for the biogeochemical pathways of Se by integrating the route, duration, and magnitude of exposure, chemical form, metabolic transformations, and modifying biotic and abiotic factors. Hamilton (2002) proposed a national criterion of 4 p~gg-! in whole-body based on the review of several laboratory and field studies. This value was the same as the toxicity threshold for fish proposed earlier by Lemly (1993, 1996b) and similar to the threshold of 4.5 ~g g-i proposed by Maier and Knight (1994). Recent papers proposed Se toxicity thresholds of 6 p~gg-l for coldwater anadromous fish and 9 p~gg-~ for warm-water fish (DeForest et al., 1999; Brix et al., 2000), although these proposals are controversial (Hamilton, 2003).
Comparison to other Idaho data
Rich and Associates (1999) reported concentrations of elements in cutthroat trout, rainbow trout (Oncorhynchus mykiss), brook trout, sculpin species, dace species, and redside shiner collected from Dry Valley Creek immediately upstream of the Blackfoot River, and Dry Valley Creek directly below Maybe Creek. In general, their concentrations were higher than those presented here for Cd, Cr, Cu, V, and Zn, but lower for Se. They concluded that Se and other elements were probably causing stress in fish populations in Dry Valley Creek. Montgomery Watson (1999) reported that salmonid fillets from a 1998 sampling contained Se concentrations of 6 ~gg-~ wet weight (maximum 7.9 p~gg-~) from East Mill Creek, whereas fish from two other sites (Blackfoot River above Wooley Range Ridge Creek and South Fork Sage Creek) had 1.2-1.3 p~gg-~. Converting these values to a dry weight basis (dry weight = wet weight x 4; assuming 75% moisture) results in 24 ~g g-~ in fish fillets from East Mill Creek and 4.8-5.2 I~g g-I in fish fillets from the reference sites. Selenium concentrations in fillets reported by Montgomery Watson (1999) underestimate the concentrations in whole body fish, which is the dominant matrix of Se residues in fish reported in the literature. Muscle contains less Se than whole-body due to the relatively high amounts of Se in spleen, liver, kidney, heart, and other tissues, especially mature ovaries (Adams, 1976; Lemly, 1982; Hilton et al., 1982). Consequently, the actual whole-body Se concentrations in trout would be about 40 p~gg-i in fish from East Mill Creek and 8-8.7 p~gg-i in fish from the Blackfoot River and South Fork Sage Creek (based on a conversion factor of 1.667 x muscle concentration = whole-body concentration; Lemly and Smith, 1987). Other conversion factors reported in the literature were 2.355 based on data from Adams (1976) for rainbow trout and 1.745 from Lemly (1982) for bluegill and largemouth bass. Here, the Se concentrations in whole-body brook trout (42.7 p~gg-i) and cutthroat trout (24.3 p~gg-l) from East Mill Creek were comparable to those reported by Montgomery Watson (1999) after conversion to dry weight and whole body (40 p~gg-~).
514
S.J Hamilton, K.j Buhl and P.J Lamothe
Selenium concentrations in salmonids collected in September 1999 were elevated in 21 of 50 samples (4.2-9.7 i~gg -l) and high in seven samples (12-31 i~gg -1) (converted to dry weight and whole body). For salmonids collected in May 2000, Se concentrations were elevated in 13 of 27 samples (5.2-9.2 i~gg -l) and high in 12 samples (10-48 ~ g g - l ) (converted to dry weight and whole body). These Se residues in salmonids were substantially above background concentrations in fish from laboratory and field investigations. Background Se concentrations in fish are typically 1-2 I~g g-i (Maier and Knight, 1994; Hamilton et al., 2000). More importantly, the Se residues were above those reported to cause adverse effects in early life stages of fish, including salmonids (4-5 I~g g-1; Hamilton et al., 2000). In particular, Se residues of 5.2 I~g g-l in rainbow trout were associated with reduced survival (Hunn et al., 1987), and 3.8-4.9 Ixgg -l in chinook salmon (Oncorhynchus tshawytscha) were associated with reduced survival and growth (Hamilton et al., 1986; Hamilton and Wiedmeyer, 1990). Older life stages typically are more tolerant of contaminant stresses than are early life stages (Rand and Petrocelli, 1985). Consequently, effects in adults may not be as readily apparent as effects in early life stages. Based on the above discussion, Se contamination of the Blackfoot River and its tributaries is most likely adversely affecting aquatic resources, especially the early life stages of fish. Thurow et al. (1981) reported that 13 fish species used the Blackfoot River and its tributaries and that the indigenous cutthroat trout was the dominant species. They noted that cutthroat trout used several tributaries, as well as the main stem river and the Blackfoot Reservoir during their life cycle. Thurow et al. (1981 ) noted the potential for mining activities to cause negative effects on trout and others species, primarily from erosion, sedimentation, and nutrient loading from phosphorous, but did not specially mention impacts from other elements. Eggs from cutthroat trout in 1999 contained Se concentrations of 4.4 and 6.7 ~g g-! dry weight in two ripe females, 4.0 txg g-~ in two partially spawned females, and 1.4 txg gin females from the reference site, Henry's Lake (Montgomery Watson, 2000). These data demonstrate that trout were accumulating Se and depositing it in their eggs. However, the Se concentrations in eggs were less than the toxic effects threshold of 10 Ixg g-~ proposed by Lemly (1993, 1996b). The low number of egg samples in Montgomery Watson (2000) precludes further speculation on the extent of Se contamination of fish eggs. Forage fish samples collected in September 1999 from various sites in the Blackfoot River watershed contained Se concentrations of >_4 Ixg g-~ (Montgomery Watson, 2001a), which are above the generally accepted toxic threshold of 4 Ixgg-I (Lemly, 1993, 1996b; Maier and Knight, 1994; Hamilton, 2002). Nine of 13 samples had elevated Se concentrations in fish (5.2-8.3 Ixg g-~, after conversion to dry weight), and two samples had high Se concentrations of 10 and 12.9 I~g g-~ (Montgomery Watson, 200 l a). For forage fish collected from various sites in the Blackfoot River watershed in May 2000, 13 of 36 samples contained Se concentrations of 5.0-9.4 Ixgg-l, and 13 samples had concentrations of 10-37 Ixg g-i (Montgomery Watson, 2001b). Selenium concentrations in forage fish reported by Montgomery Watson (200 l a,b) were similar to those determined here.
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
515
Other considerations
One consideration is the presence of elevated Se residues in fish without readily apparent biological effects. However, it should be considered that the data in the current study and studies by others (Rich and Associates, 1999; Montgomery Watson, 1999, 2000, 2001 a,b) were contaminant surveys and not biological effect studies. No biological effects such as survival, growth, reproduction, diversity, population structure, community structure, or other biological effects were measured. A second consideration is that residues measured in fish were for adults or subadults. Fish in these life stages are generally not as sensitive to the effects of environmental contaminants as those in earlier stages (Rand and Petrocelli, 1985). Monitoring of fish populations in rivers is an insensitive measure of contaminant effects unless substantial effort is made to assess the health of the fish community. This assertion was addressed by the US Environmental Protection Agency in their guidelines for deriving water-quality criteria. Stephan et al. (1985) found that most monitoring programs are insensitive for number of taxa and individuals, which limits their usefulness in developing water-quality criteria because some relevant changes may go undetected. Limited field studies may be able to determine whether criteria are under-protective, but only high-quality field studies can clearly determine that criteria are over-protective. The claim of no biological effects in stream or river studies cannot often be confirmed without appropriate biological effects tests and often fall into the null fallacy trap: (a) there is no evidence for adverse effects, vs. (b) there is evidence for no adverse effects (J. Skorupa, US Fish and Wildlife Service, personal communication). The null fallacy occurs when statement (a) (a null finding) is given equal weight as statement (b) (a positive finding). What often is overlooked is that a null finding usually implies a lack of positive evidence in both directions- for effects or for absence of effects. Montgomery Watson (2001b) found that higher than expected Se concentrations in forage fish from a reference site on Spring Creek above influences of East Mill Creek were probably due to the mobility of fish. Forage fish in the upper Spring Creek contained Se concentrations of 10, 12, and 22 Ixg g-I. However, in spite of high Se residues in wholebody forage fish collected in May 2000, Montgomery Watson (2001b) concluded that neither Se nor Cd was impacting forage fish in the Blackfoot Reservoir. Likewise, Montgomery Watson (200 l a) reported elevated Se concentrations in forage fish collected in September 1999, yet concluded that forage fish in the reservoir show no evidence of impact. Because no biological effects were assessed in fish collections in September 1999 or May 2000, the null fallacy trap applies to those conclusions.
HAZARD ASSESSMENT Lemly (1995) presented a protocol for aquatic hazard assessment of Se, formulated primarily in terms of the potential for food-chain bioaccumulation and reproductive impairment in fish and aquatic birds. The protocol incorporated five ecosystem components
516
S.J Hamilton, K.j Buhl and PJ Lamothe
including water, sediment, benthic invertebrates, fish eggs, and bird eggs. Each component was given a numeric score based on the degree of hazard: (a) no identifiable hazard; (b) minimal hazard; (c) low hazard; (d) moderate hazard; (e) high hazard. The final hazard characterization was determined by adding the individual scores and comparing the total to the following evaluation criteria: 5, no hazard; 6-8, minimal hazard; 9-11, low hazard; 12-15, moderate hazard; 16-25, high hazard. Lemly (1996a) modified his protocol for use with four ecosystem components due to the difficulty in collecting residue information for all five components in an assessment. He adjusted the final ecosystem-level hazard assessment to the following four-component evaluation criteria: 4, no hazard; 5-7, minimal hazard; 8-10, low hazard; 11-14, moderate hazard; 15-20, high hazard. Lemly (1995) defined the five categories of hazards as follows: (a) high hazard denotes an imminent, persistent toxic threat sufficient to cause complete reproductive failure in most species of fish and aquatic birds; (b) moderate hazard indicates a persistent toxic threat of sufficient magnitude to substantially impair but not eliminate reproductive success; some species will be severely affected whereas others will be relatively unaffected; (c) low hazard denotes a periodic or ephemeral toxic threat that could marginally affect the reproductive success of some sensitive species, but most species will be unaffected; (d) minimal hazard indicates that no toxic threat identified but concentrations of Se are slightly elevated in one or more ecosystem components compared to uncontaminated reference sites; (e) no hazard denotes that no toxic threat is identified and Se concentrations are not elevated in any ecosystem component. Table 18-XIV gives the Se concentrations for each hazard level for the four-component model. The Se hazard protocols give equal weigh to each component (Lemly, 1995, 1996a). However, a critique of the protocol pointed out the need to give more weight to the biological components: benthic invertebrates, fish eggs, and bird eggs (H. Ohlendorf, 1996, written communication). Ohlendorf suggested a multiplication factor of two for benthic invertebrate data and a factor of three for fish eggs and bird eggs. Similar concerns have been raised by others (M. Sylvester, US Geological Survey, 2002; B. Osmundson, US Fish Wildlife Service, 2001, written communications). The weighting of the three biological components seems justified based on the repeated expression of their importance in the Se literature (reviews by Lemly, 1985, 1993; Maier and Knight, 1994; Presser et al., 1994; Hamilton and Lemly, 1999; Hamilton 2002, 2003). Incorporating these factors into the protocol using the offset summation approach used by Lemly (1995, 1996a) results in modified final characterizations for the four-component protocol of 7, no hazard; 8-13, minimal hazard; 14-20, low hazard; 21-27, moderate hazard; and 28-35, high hazard (Table 18-XIV). The offset summation is explained as follows: for the low hazard column, Lemly (1996a) gives a score of 3 for each of the four components being evaluated (water, sediment, benthic invertebrate, and fish eggs), which results in a summed score of 12 (Table 18-XIV). However, if in an environmental situation all measured Se concentrations of the four components fell into the "low" column, the additive effect of the combined low exposures would most likely result in a "moderate" final hazard to biota. Thus, Lemly (1996a) sets the final hazard range for a "low" final hazard at 8-10, instead of closer to the summed total of 12. This offsetting of the final hazard total
TABLE 8-XIV t~
Aquatic ecosystem components and concentrations posing various hazards (based on Lemly, 1996a) t~
Ecosystem component
Hazard
Conc. Water (ixgL -~) Sediment (ixgg -~)
Minimal
None Lemly Current score score
Conc.
Low
Lemly Current score score
Moderate
Conc.
Lemly score
Current score
Conc.
High
Lemly Current score score
Conc.
Lemly Current score score
< 1
1
1
1-2
2
2
2-3
3
3
3-5
4
4
>5
5
5
< 1
1
1
1-2
2
2
2-3
3
3
3-4
4
4
>4
5
5
Benthic invertebrate (p.gg-~)
<2
1
2
2-3
2
4
3-4
3
6
4-5
4
8
>5
5
10
Fish eggs
<3
e
~.~.
1
3
Sum
4
7
Final hazard (Lemly) Final hazard (Current)
4
3-5
2
6
8
14
5-10
3
9
12
21
10-20
4
12
16
28
>20
5
15
20
35
(t~gg -~)
5-7 7
8-10 8-13
11-14 14--20
t~
15-20 21-27
"~
28-35
".-..I
518
S.J. Hamilton, K.J. Buhl and PJ. Lamothe
TABLE 18-XV Hazard assessment of Se at nine locations listed in Table 18-I
Site and ecosystem component
Selenium ~
Evaluation by component
Totals for site
Hazard
Score
Score
Hazard
ACM Water Sediment Benthic invertebrate Fish eggs 2
<0.5 1.0 7.2 21.1
None Minimal High High
1 2 10 15
28
Moderate
UEMC Water Sediment Benthic invertebrate Fish eggs
29.9 32.5 35.7 106.3
High High High High
5 5 10 15
35
High
LEMC Water Sediment Benthic invertebrate Fish eggs
14.9 45.8 NS 3 NS
High High -
5 5 -
10
High 4
TC Water Sediment Benthic invertebrate Fish eggs
<0.5 0.8 2.9 16.9
None None Minimal Moderate
1 1 4 12
18
Low
USC Water Sediment Benthic invertebrate Fish eggs
<0.5 0.8 4.9 13.2
None None Moderate Moderate
1 1 8 12
22
Moderate
LSC Water Sediment Benthic invertebrate Fish eggs
<0.5 1.7 NS 17.5
None Minimal Moderate
1 2 12
14
Moderate 4
DVC Water Sediment Benthic invertebrate Fish eggs
0.5 5.2 19.5 53. I
None High High High
1 5 10 15
31
High
Continued
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
519
TABLE 18-XV Continued Site and ecosystem component
Selenium I
Evaluation by component
Totals for site
Hazard
Score
Score
ShpC Water Sediment Benthic invertebrate Fish eggs
<0.5 0.5 2.6 17.2
None None Minimal Moderate
1 1 4 12
LBR Water Sediment Benthic invertebrate Fish eggs
<0.5 2.1 5.4 25.7
None Low High High
1 3 10 15
Hazard
18
Low
29
Moderate
IConcentrations in Ixg L-i for water, Ixg g-! for sediment, benthic invertebrates, and fish eggs. 2Fish eggs: fish egg values converted from whole-body residues using: whole-body x 3.3 = fish egg (Lemly1995, 1996a). 3NS: not sampled. aBased on available aquatic ecosystem components. seems biologically reasonable and is referred to here as the offset summation approach. Similar offsets for other final hazards are given in Table 18-XIV. For the five-component protocol, the modified final hazard characterization would be, l 0, no hazard; l 1-19, minimal hazard; 20-28, low hazard; 29-38, moderate hazard; and 39-50, high hazard. In the present study, fish eggs were not collected. Consequently, as a proxy we converted the geometric mean whole-body concentrations of Se to fish eggs concentrations using the conversion factor based on Lemly and Smith (1987), who reported: viscera (fish eggs or liver) to whole-body = viscera x 0.33. The conversion factor to eggs only would be 3.3 (Lemly, 1995, 1996a). Trail Creek and Sheep Creek had low Se concentrations in most aquatic ecosystem components, a low hazard rating (Table 18-XV). Although Se concentrations were low in water and sediment at upper Slug Creek and lower Slug Creek, they were elevated in benthic invertebrates or whole-body residues converted to fish egg concentrations or both, resulting in a moderate hazard rating. Selenium concentrations in water or sediment were in the none, minimal, or low categories at Angus Creek and the lower Blackfoot River, but high in benthic invertebrates and whole-body residues, thus receiving high final hazards. Using the original Lemly (1996a) approach, these two sites would have received moderate final hazards in spite of the high score for benthic invertebrates and fish eggs. Upper East Mill Creek and Dry Valley Creek consistently had elevated Se concentrations in sediment, invertebrates, or whole-body residues, thus resulting in a high hazard rating. The hazard rating at lower East Mill Creek was considered high because of the elevated
520
S.J. Hamilton, K.J. Buhl and P.J Lamothe
Se concentrations in water and sediment, which no doubt would have resulted in high concentrations in invertebrates and fish had they been found. A preliminary assessment of Se hazard in the Caribou National Forest was conducted using Se residue data in water and fish collected from 1997 to 1998 (Lemly, 1999). Lemly (1999) concluded that there was a high potential for toxic impacts to fish and wildlife associated with the Blackfoot River, its tributaries, and Blackfoot Reservoir. The results of the present study add substantially more support to the premise that Se concentrations in several aquatic ecosystem components were sufficiently elevated to cause adverse effects to aquatic resources in the Blackfoot River watershed.
ACKNOWLEDGMENT The authors thank the following personnel for assistance in this study: Bill Janowski, Jeff Jones and Larry Michelson (US Forest Service), Jay Bateman, Don Cole, Mark Hardy, Jim Herring, Mark Huebner, Phil Moyle, and Theresa Presser (US Geological Survey), Peter Oberlindacher (US Bureau of Land Management), and Jim Mende (Idaho Department of Fish and Game). The authors thank John Besser and Marc Sylvester for reviewing the draft manuscript.
REFERENCES
Adams, W.J., 1976. The toxicity and residue dynamics of selenium in fish and aquatic invertebrates. PhD dissertation, Michigan State University, Lansing, 109 pp. Alaimo, J., Ogle, R.S. and Knight, A.W., 1994. Selenium uptake by larval Chironomus decorus from a Ruppia maritima-based benthic/detrital substrate. Arch. Environ. Contam. Toxicol., 27: 441-448. Allan, J.D., 1995. Stream Ecology- Structure and Function of Running Waters. Chapman and Hall, London, 388 pp. American Public Health Association, American Water Works Association, and Water Environment Federation, 1995. Standard Methods for the Examination of Water and Wastewater (19th edn.). American Public Health Association, Washington. Analytical Technology Incorporated Orion, 1994. Model 9346 Nitrite Electrode Instruction Manual. Analytical Technology Incorporated, Boston, 44 pp. Brix, K.V., DeForest, D.K., Fairbrother, A. and Adams, W.J., 2000. Critical review of tissue-based selenium toxicity thresholds for fish and birds. In: Planning for End Land Use in Mine Reclamation, Proceedings of the 24th Annual British Columbia Reclamation Symposium. BC Technical and Research Committee on Reclamation, June 19-22, Williams Lake, BC, pp. 220-230. Bryson, W.T., Garrett, W.R., Mallin, M.A., MacPherson, K.A., Partin, W.E. and Woock, S.E., 1984. Roxboro Steam Electric Plant Environmental Monitoring Studies, Vol. II, Hyco Reservoir Bioassay Studies 1982. Carolina Power and Light, New Hill, NC, 84 pp. Campbell, K.R., 1994. Concentrations of heavy metals associated with urban runoff in fish living in stormwater treatment ponds. Arch. Environ. Contam. Toxicol., 27: 352-356.
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
521
Caribou County Sun, 1999. Toxicologist and vet say dead sheep likely died from selenium. Soda Springs, ID, November 11. Cherry, D.S. and Guthrie, R.K., 1977. Toxic metals in surface waters from coal ash. Water Resour. Bull., 13: 1227-1236. Dallinger, R. and Kautzky, H., 1985. The importance of contaminated food for the uptake of heavy metals by rainbow trout (Salmo gairdneri): a field study. Oecologia, 67: 82-89. Dallinger, R., Prosi, E, Segner, H. and Back, H., 1987. Contaminated food and uptake of heavy metals by fish: a review and a proposal for further research. Oecologia, 73:91-98. DeForest, D.K., Brix, K.V. and Adams, W.J., 1999. Critical review of proposed residue-based selenium toxicity thresholds for freshwater fish. Hum. Ecol. Risk Assess., 5:1187-1228. Desborough, G., DeWitt, E., Jones, J., Meier, A. and Meeker, G., 1999. Preliminary mineralogical and chemical studies related to the potential mobility of selenium and associated elements in Phosphoria Formation strata, southeastem Idaho. US Geol. Surv., Open File Report, 99-129, 20 pp. Devi, M., Thomas, D.A., Barber, J.T. and Fingerman, M., 1996. Accumulation and physiological and biochemical effects of cadmium in a simple aquatic food chain. Ecotoxicol. Environ. Saf., 33: 38-43. Eisler, R., 1985. Selenium hazards to fish, wildlife, and invertebrates: a synoptic review. US Fish and Wildlife Service, Contaminant Hazard Reviews Report No. 5, 57 pp. Finley, K.A., 1985. Observations of bluegills fed selenium-contaminated Hexagenia nymphs collected from Belews Lake, North Carolina. Bull. Environ. Contam. Toxicol., 35: 816-825. Foda, A., Vandermeulen, J.H. and Wrench, J.J., 1983. Uptake and conversion of selenium by a marine bacterium. Can. J. Fish. Aquat. Sci., 40 (Supplement 2): 215-220. Furr, A.K., Parkinson, T.E, Youngs, W.D., Berg, C.O., Gutenmann, W.H., Pakkala, I.S. and Lisk, D.J., 1979. Elemental content of aquatic organisms inhabiting a pond contaminated with coal fly ash. NY Fish Game J., 26:154-161. Hach Company, 1997. Hach Water Analysis Handbook (3rd edn.). Hach Company, Loveland, CO., 1309 pp. Hamilton, S.J., 2002. Rationale for a tissue-based selenium criterion for aquatic life. Aquat. Toxicol., 57: 85-100. Hamilton, S.J., 2003. Review of residue-based selenium toxicity thresholds for freshwater fish. Ecotoxicol. Environ. Saf., 56:201-210. Hamilton, S.J. and Hoffman, D.J., 2003. Trace element and nutrition interactions in fish and wildlife. In: D.J. Hoffman, B.A. Rattner, G.A. Burton and J. Cairns (eds.), Handbook of Ecotoxicology (2nd edn.). CRC Press, Boca Raton, pp. 1197-1235. Hamilton, S.J. and Lemly, A.D., 1999. Water-sediment controversy in setting environmental standards for selenium. Ecotoxicol. Environ. Saf., 44: 227-235. Hamilton, S.J. and Wiedmeyer, R.H., 1990. Concentrations of boron, molybdenum, and selenium in chinook salmon. Trans. Am. Fish. Soc., 119:500-510. Hamilton, S.J., Palmisano, A.N., Wedemeyer, G.A. and Yasutake, W.T., 1986. Impacts of selenium on early life stages and smoltification of fall chinook salmon. Trans. N. Am. Wildl. Nat. Res. Conf., 51: 343-356. Hamilton, S.J., Muth, R.T., Waddell, B. and May, T.W., 2000. Hazard assessment of selenium and other trace elements in wild larval razorback sucker from the Green River, Utah. Ecotoxicol. Environ. Saf., 45: 132-147. Hamilton, S.J., Holley, K.M., Buhl, K.J., Bullard, EA., Weston, L.K. and McDonald, S.E, 2001 a. The evaluation of contaminant impacts on razorback sucker held in flooded bottomland sites near Grand Junction, Colorado - 1996. Final Report. US Geol. Surv., Yankton, SD, 302 pp.
522
S.J. Hamilton, K.J Buhl and PJ Lamothe
Hamilton, S.J., Holley, K.M., Buhl, K.J., Bullard, EA., Weston, L.K. and McDonald, S.E, 200 lb. The evaluation of contaminant impacts on razorback sucker held in flooded bottomland sites near Grand Junction, Colorado- 1997. Final Report. US Geol. Surv., Yankton, SD, 229 pp. Herring, J.R. and Amacher, M.C., 2001. Chemical composition of plants growing on the Wooley Valley phosphate mine waste pile and on similar rocks in nearby Dairy Syncline, Caribou County, southeast Idaho. US Geol. Surv., Open File Report, 01-0025, 47 pp. Herring, J.R., Wilson, S.A., Stillings, L.A., Knudsen, A.C., Gunter, M.E., Tysdal, R.G., Grauch, R.I., Desborough, G.A. and Zielinski, R.A., 2000a. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation B. Measured sections C and D, Dry Valley, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-147-B, 33 pp. Herring, J.R., Grauch, R.I., Desborough, G.A., Wilson, S.A. and Tysdal, R.G., 2000b. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation C. Measured sections E and F, Rasmussen Ridge, Caribou County, Idaho. US Geol. Surv., Open File Report, 99-147-C, 35 pp. Herring, J.R., Castle, C.J., Brown, Z.A. and Briggs, P.H., 2001. Chemical composition of deployed and indigenous aquatic bryophytes in a seep flowing from a phosphate mine waste pile and in the associated Angus Creek drainage, Caribou County, southeast Idaho. US Geol. Surv., Open File Report, 01-0026, 20 pp. Hilton, J.W., Hodson, P.V. and Slinger, S.J., 1982. Absorption, distribution, half-life and possible routes of elimination of dietary selenium in juvenile rainbow trout (Salmo gairdneri). Comp. Biochem. Physiol., 71C: 49-55. Hunn, J.B., Hamilton, S.J. and Buckler, D.R., 1987. Toxicity of sodium selenite to rainbow trout fry. Water Res., 21: 233-238. Kiffney, P.M. and Clements, W.H., 1993. Bioaccumulation of heavy metals by benthic invertebrates at the Arkansas River, Colorado. Environ. Toxicol. Chem., 12: 1507-1517. Kirby, J., Maher, W. and Harasti, D., 2001a. Changes in selenium, copper, cadmium, and zinc concentrations in mullet (Mugil cephalus) from the southern basin of Lake Macquarie, Australia, in response to alteration of coal-fired power station fly ash handling procedures. Arch. Environ. Contam. Toxicol., 41: 171-181. Kirby, J., Maher, W. and Krikowa, E, 2001b. Selenium, cadmium, copper, and zinc concentrations in sediments and mullet (Mugil cephalus) from the southern basin of Lake Macquarie, NSW, Australia. Arch. Environ. Contam. Toxicol., 40: 246-256. Lee, D.S., Gilbert, C.R., Hocutt, C.H., Jenkins, R.E., McAllister, D.E. and Stauffer, J.R. Jr., 1980. Atlas of North American Freshwater Fishes. North Carolina State Museum of Natural History, Raliegh, NC, 854 pp. Lemly, A.D., 1982. Response of juvenile centrarchids to sublethal concentrations of waterborne selenium. I. Uptake, tissue distribution, and retention. Aquat. Toxicol., 2" 235-252. Lemly, A.D., 1985. Toxicology of selenium in a freshwater reservoir: implications for environmental hazard evaluation and safety. Ecotoxicol. Environ. Saf., 10:314-338. Lemly, A.D., 1993. Guidelines for evaluating selenium data from aquatic monitoring and assessment studies. Environ. Monit. Assess., 28: 83-100. Lemly, A.D., 1995. A protocol for aquatic hazard assessment of selenium. Ecotoxicol. Environ. Saf., 32: 280-288. Lemly, A.D., 1996a. Evaluation of the hazard quotient method for risk assessment of selenium. Ecotoxicol. Environ. Saf., 35:156-162.
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
523
Lemly, A.D., 1996b. Selenium in aquatic organisms. In: W.N. Beyer, G.H. Heinz and A.W RedmonNorwood (eds.), Environmental Contaminants in Wildlife- Interpreting Tissue Concentrations. CRC Lewis Publ., Boca Raton, pp. 427-445. Lemly, A.D., 1999. Preliminary assessment of selenium hazards on Caribou National Forest, Idaho. Report, US Forest Service, Blacksburg, VA, 16 pp. Lemly, A.D. and Smith, G.J., 1987. Aquatic cycling of selenium: Implications for fish and wildlife. Fish and Wildlife Leaflet 12. US Fish and Wildlife Service, Washington, 10 pp. Luoma, S.N., 1983. Bioavailability of trace metals to aquatic organisms- a review. Sci. Total Environ., 28: 1-22. Maier, K.J. and Knight, A.W., 1994. Ecotoxicology of selenium in freshwater systems. Rev. Environ. Contam. Toxicol., 134:31-48. Maier, K.J., Nelson, C.R., Bailey, EC., Klaine, S.J. and Knight, A.W, 1998. Accumulation of selenium by the aquatic biota of a watershed treated with seleniferous fertilizer. Bull. Environ. Contam. Toxicol., 60:409-4 16. Mann, K.H., 1972. Macrophyte production and detritus food chains in coastal waters. Mem. Ist. Ital. Idrobiol., 29: 353-383. Montgomery Watson, 1999. Final 1998 Regional Investigation Report, Southeast Idaho Phosphate Resource Area Selenium Project. Report to Idaho Mining Association Selenium Subcommittee. Montgomery Watson, Steamboat Springs, CO. Montgomery Watson, 2000. 1999 Interim Investigation Data Report, Southeast Idaho Phosphate Resource Area Selenium Project. Report to Idaho Mining Association Selenium Subcommittee. Montgomery Watson, Steamboat Springs, CO. Montgomery Watson, 2001a. Draft 1999-2000 Regional Investigation Data Report for Surface Water, Sediment and Aquatic Biota Sampling Activities, September 1999. Southeast Idaho Phosphate Resource Area Selenium Project. Report to Idaho Mining Association Selenium Subcommittee. Montgomery Watson, Steamboat Springs, CO. Montgomery Watson, 200lb. Draft 1999-2000 Regional Investigation Data Report for Surface Water, Sediment and Aquatic Biota Sampling Activities, May-June 2000, Southeast Idaho Phosphate Resource Area Selenium Project. Report to Idaho Mining Association Selenium Subcommittee. Montgomery Watson, Steamboat Springs, CO. Moore, S.B., Winckel, J., Detwiler, S.J., Klasing, S.A., Gaul, P.A., Kanim, N.R., Kesser, B.E., DeBevec, A.B., Beardsley, K. and Puckett, L.K., 1990. Fish and Wildlife Resources and Agricultural Drainage in the San Joaquin Valley, California (vol. 2). San Joaquin Valley Drainage Program, Sacramento, CA. Ney, J.J. and Van Hassel, J.H., 1983. Sources of variability in accumulation of heavy metals by fishes in a roadside stream. Arch. Environ. Contam. Toxicol., 12: 701-706. National Research Council, 1980. Recommended daily allowances (9th edn.). National Research Council, Food and Nutrition Board, Committee on Dietary Allowances. National Academy Press, Washington, 187 pp. Odum, E.P. and de la Cruz, A.A., 1967. Particulate organic detritus in a Georgia salt marsh-estuarine ecosystem. In: G.H. Lanff (ed.), Estuaries. American Association for the Advancement of Science Publication 83, pp. 383-388. Olson, O.E., 1986. Selenium toxicity in animals with emphasis on man. J. Am. College Toxicol., 5: 45-70. Orion Research, 1990. Model 95-12 Ammonia Electrode Instruction Manual. Orion Research Incorporated, Boston, 36 pp.
524
S.J. Hamilton, K.J. Buhl and P.J. Lamothe
Orion Research, 1991. Model 93-07 Nitrate Electrode Instruction Manual. Orion Research Incorporated, Boston, 31 pp. Piper, D.Z., Skorupa, J.P., Presser, T.S., Hardy, M.A., Hamilton, S.J., Huebner, M., and Gulbrandsen, R.A., 2000. The Phosphoria Formation at the Hot Springs Mine in southeast Idaho: a source of selenium and other trace elements to surface water, ground water, vegetation, and biota. US Geol. Surv., Open File Report, 00-050, 73 pp. Platts, W.S. and Martin, S.B., 1978. Hydrochemical influences on the fishery within the phosphate mining area of eastern Idaho. US Department of Agriculture Forest Service Research Note INT246. Intermountain Forest and Range Experiment Station, Odgen, UT, 15 pp. Presser, T.S., Sylvester, M.A. and Low, W.H., 1994. Bioaccumulation of selenium from natural geologic sources in western states and its potential consequences. Environ. Manage., 18: 423-436. Rand, G.M. and Petrocelli, S.R. (eds.), 1985. Fundamentals of Aquatic Toxicology: Methods and Applications. Hemisphere, Washington, 666 pp. Rich and Associates, 1999. FMC phosphate mine expansion. Fishery Resources Technical Report, A.A. Rich and Associates, San Anselmo, CA, 125 pp. plus appendices. Saiki, M.K., Jennings, M.R. and Brumbaugh, W.G., 1993. Boron, molybdenum, and selenium in aquatic food chains from the lower San Joaquin River and its tributaries, California. Arch. Environ. Contam. Toxicol., 24:307-319. Statistical Analysis System, Inc., 1985. SAS Language Guide for Personal Computers (6th edn.). SAS Institute, Inc., Cary, NC. Seelye, J.G., Hesselberg, R.J. and Mac, M.J., 1982. Accumulation by fish of contaminants released from dredged sediments. Environ. Sci. Technol., 16: 459-464. Stephan, C.E., Mount, D.I., Hansen, D.J., Gentile, J.H., Chapman, G.A. and Brungs, W.A., 1985. Guidelines for deriving numerical national water quality criteria for the protection of aquatic organisms and their uses. US Environ. Protection Agency Report PB85-227049, 98 pp. Stephens, D., Waddell, B., DuBois, K. and Peterson, E., 1997. Field screening of water quality, bottom sediment, and biota associated with the Emery and Scofield Project areas, central Utah, 1994. US Geol. Surv., Water-Resources Investigations Report, 96-4298, 39 pp. Teal, J.M., 1962. Energy flow in a salt march ecosystem of Georgia. Ecology, 43: 614-624. Thurow, R., Wishard, L., Christensen, W. and Aebersold, P., 1981. Federal aid to fish and wildlife restoration, Job Completion Report Project F-73-R-3. ID Depart. Fish Game, Boise, 243 pp. TRC Environmental Corporation, 1999. Maybe Canyon site investigation, Caribou National Forest, Caribou County, Idaho. Report to US Forest Service. TRC Environmental Corporation, Englewood, CO, 122 pp. plus appendices. US Environmental Protection Agency, 1983. Methods for chemical analysis of water and wastes. Publication EPA 600/4-79-020. US Environmental Protection Agency, 1987. Ambient water quality criteria for selenium - 1987. Publication EPA 440/5-87-006. US Environmental Protection Agency, 1998a. National recommended water quality criteria; republication. Fed. Reg., 63: 68354-68364. US Environmental Protection Agency, 1998b. Report on the Peer Consultation Workshop on Selenium Aquatic Toxicity and Bioaccumulation. Publication EPA 822-R-98-007, Washington, 261 pp. US Environmental Protection Agency, 1999. National recommended water quality criteria- correction. Publication EPA 822-Z-99-001.
Selenium and other trace elements in water, sediment, aquatic plants, and invertebrates
525
Wiener, J.G. and Giesy, J.E Jr., 1979. Concentrations of Cd, Cu, Mn, Pb, and Zn in fishes in a highly organic softwater pond. J. Fish. Res. Bd. Can., 36: 270-279. Woock, S.E., 1984. Accumulation of Selenium by Golden Shiners Notemigonus crysoleucas: Hyco Reservoir N.C. cage study 1981-1982. Carolina Power and Light Company, New Hill, NC, 19 pp.
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
527
Chapter 19
UPTAKE OF SELENIUM AND OTHER CONTAMINANT ELEMENTS INTO PLANTS AND IMPLICATIONS FOR GRAZING ANIMALS IN SOUTHEAST IDAHO C.L. MACKOWIAK, M.C. AMACHER, J.O. HALL, and J.R. HERRING
ABSTRACT As part of a series of geoenvironmental studies on the mobilization and fate of selenium (Se) and other potentially toxic trace elements in southeast Idaho phosphate mining areas, trace element concentrations (mg kg-1 dry mass) in plant samples collected along transects at the Wooley Valley Unit 1, 3, and 4 waste-rock dumps were compared with samples collected from undisturbed sites at Dairy Syncline, Deer Creek, Dry Valley, Maybe Canyon, and Rasmussen Ridge. Additionally, trace-element concentrations in vegetation samples collected from wetlands associated with mine waste-rock piles were compared with samples collected from a single reference wetland. In undisturbed areas, Se in vegetation growing in soils overlying and derived from Phosphoria Formation phosphatic rocks tended to be higher than vegetation in undisturbed Wells Limestone or Rex Chert soils. Vegetation growing in highly disturbed soils, such as those comprising waste-rock dumps, had the highest tissue Se. Vegetation in a wetland at the base of Wooley Valley Unit 4 waste-rock dump accumulated decreasing concentrations of Se with increasing distance away from the waste-rock dump along the wetland flow path. Iron oxides were observed coating wetland sediment surfaces and helped control Se bioavailability. Plant uptake, as well as coprecipitation and sorption of Se by iron oxides, were key processes in the natural attenuation of Se in this wetland. Legumes at the rock dumps contained higher Se (mean = 80 mg kg-I) than trees (mean = 52 m g k g - l ) , grasses (mean = 18 mgkg-1), shrubs (mean = 6 m g k g - l ) , and forbs (mean = 3 mg kg-1). However, grasses were among the highest Se accumulators among plant lifeforms in contaminated wetlands, with a mean value of 53 mgkg -1 Se. In most places, uptake of Cd, Cr, Cu, Mn, Mo, Ni, and Zn was below critical high levels for plants. However, Se, Cd, Cr, and Zn uptake by some plants may have been large enough to affect their growth. Several plant lifeforms had Se concentrations that surpassed the acute and chronic toxicity thresholds for grazing livestock and wildlife, posing a lethal risk to these animals. Forages, particularly legumes, sampled from waste-rock dumps had increased Mo concentrations, resulting in Cu/Mo ratios below 1. High Mo (above 10mgkg -1) and Cu/Mo ratios below 2 may cause molybdenosis in ruminants. There were instances where
528
C.L. Mackowiak et al.
tree Zn content exceeded upper chronic intake for livestock/wildlife. This may pose some concern for browsing animals feeding upon trees, particularly in winter months. Based on the vegetation survey, possible remediation strategies via physical, chemical, and biological manipulations of the contaminated sites include removal of the most contaminated soils, capping contaminated soils and revegetating capping materials, application of selective herbicides to remove legumes from reclaimed waste-rock dumps, and fencing some contaminated areas to better manage grazing.
INTRODUCTION Vegetation has a major impact on the movement and bioavailability of trace elements in the environment, particularly in disturbed soils. This chapter will focus on the uptake of Se and other trace-elements (Cd, Cr, Cu, Mo, Ni, and Zn) by plants growing on phosphate mine waste-rock dumps and associated wetlands. We will also examine the potential for using plants as a geochemical exploration tool to detect phosphatic shale deposits, the implications of increased Se and other trace-element uptake on grazing livestock health, and the potential for using plants for remediation of phosphatic mining waste-rock dumps.
Locations a n d general g e o l o g y
The mining locations lie approximately 25 km northeast of Soda Springs, Idaho, in an area of southeast Idaho that has had extensive phosphate mining over the past several decades, and includes four active mines. The non-mined Dairy Syncline and Deer Creek locations are approximately 25 km southeast of Soda Springs (Fig. 19-1). Causey and Moyle (2001 ) compiled a digital database of the region's mining-related features. Hein et al. (Chapter 2) and Jasinski et al. (Chapter 3) present geologic and historic reviews of the region. A more detailed discussion of the historical Wooley Valley mine reclamation efforts can be found in Chambers et al. (1994).
Se a n d other trace elements - e n v i r o n m e n t a l c o n c e r n s
Phosphate mining in SE Idaho exploits the Meade Peak Phosphatic Shale Member of the Phosphoria Formation. During phosphate rock mining, the associated waste rock is typically stored in large piles, which historically have remained in place, have been used as backfill for mined trenches, or have been removed and placed in nearby valleys. Snowmelt, rainfall, springs, and buried streams percolating through the waste rock can leach trace-elements that are discharged into nearby streams (Amacher et al., 1995; Vance, 2000). Elevated concentrations of Se and other environmentally sensitive trace-elements (i.e. As, Cd, Cr, Cu, Mo, Ni, Ti, U, V, and Zn) within the middle-waste shale of the Meade Peak Phosphatic Shale Member of the Phosphoria Formation have raised concern about the
Uptake of selenium and other contaminant elements into plants
529
Fig. 19-1. General locations of phosphate mine waste-rock piles and undisturbed sampling areas in SE Idaho.
impact of these trace-elements in the environment (Piper et al., 2000; Hamilton et al., Chapter 18). Selenium and several other trace elements, notably Cu, Mo, and Zn, can act either as essential micronutrients at low concentrations or as toxins at elevated concentrations. Gough et al. (1979) provided summaries of element concentrations, both bioessential and toxic to plants, animals, and man. Specifically, safe and adequate Se concentrations in animal forage diets are 0.1-0.3 mg kg-1, dry mass (Mayland, 1994), with a critical threshold value of 5 mg kg-1 (National Research Council, 1980a).
Plants for assessing trace-element mobility Plants can be used as bioindicators (biological indicators) for various trace-elements in soils. Generally, plant uptake response correlates with concentrations ofbioavailable traceelements in soils or waste-rock piles. However, bioavailability is affected by many other interactive and complex factors, such as the mineral and organic host of the trace-elements,
530
C.L. Mackowiak et al.
ambient pH and Eh, and water solubility of the trace-elements in host soil phases. It further includes numerous aspects of plant physiology, toxic element tolerance, and growth factors, including climate. Plants have been loosely categorized in terms of their ability to accumulate Se in seleniferous environments. For example, plant species that accumulate high amounts (thousands of mgkg-ldry mass) of target trace-elements were classified by Rosenfeld and Beath (1964) as hyperaccumulators or primary accumulators. They reported that there are approximately 24 species and varieties of Astragalus, Xylorhiza, Oonopsis, and Stanleya that accumulate large quantities of Se; additionally these may have a metabolic Se requirement that favors Se uptake even against S competition (Bell et al., 1992). More recently, Feist and Parker (2001) examined the diversity of Stanleya pinnata (Pursh) populations for Se accumulation and their potential for Se remediation. They found Stanleya Se contents varied up to 3.6-fold among the naturally occurring populations and therefore suggested that native populations be screened for stock material to use in remediation programs. The next category is secondary accumulators or absorbers, which typically includes species that tend to passively take up Se (e.g. greater than 100mgkg -1) in seleniferous soils. Although some of these plants have a large S requirement, which also favors analog (i.e. Se) uptake, Se is excluded from proteins to a much greater extent than in nonaccumulating plants (Mayland et al., 1989). Several Brassica species fall into this classification and they hold promise for Se phytoremediation/phytoextraction in central California (Bafiuelos et al., 1997, 2000). Other secondary absorbers include Aster, Atriplex, and Castilleja (Rosenfeld and Beath, 1964). The third category is non-accumulator plants, which represents many grass and forb species that limit their uptake to approximately 50-100 mg kg-! Se in seleniferous soils (Mayland et al., 1989). For example, soils with high (40mgkg -I) total Se resulted in Festuca tissue Se of 50 mg kg-l (Bafiuelos et al., 1997). Alfalfa is considered a Se nonaccumulator, but reports suggest it accumulates much more Se than most other forbs and grasses in low Se soils. This may be attributed to alfalfa having a higher S requirement (Mayland et al., 1989) and/or deeper root penetration (Mayland et al., 1989; Stark and Redente, 1990). At issue is the need to keep undesirably high levels of Se (>--5 mg kg-l) out of vegetation growing at decommissioned phosphate mine sites. The variability in Se uptake among introduced forage and native plant species may prove useful in managing contaminated sites. Primary accumulator or even secondary Se absorber plant species might be used as bioindicators of Se or other target trace-elements. Over the past 3 years, extensive plant lifeform and species sampling was conducted at decommissioned mining sites and associated wetlands to characterize trace-element, particularly Se, biomass concentrations and to compare their compositions with vegetation growing in undisturbed soils overlying phosphatic-shale deposits. The potential effects of increased Se and other trace element concentrations in plants on the health of grazing animals and the potential for vegetation management as part of overall mine waste-rock remediation are also discussed.
Uptake of selenium and other contaminant elements into plants
531
METHODS
Experimental design The ideal study for assessing trace-element bioavailability to plants compares the same plant species in both reference (background) and disturbed sites. However, it is often not possible to find large assemblages of similar plants in both reclaimed areas and indigenous background areas. Often, reclamation species are non-native. In this study, the majority of plants found on the waste-rock piles were different from the indigenous species in the reference wetland or reference soil sites. Fortunately, there were enough samples to permit lifeform (i.e. grasses, rushes, sedges, forbs, legumes, shrubs, and trees) comparisons among sampling locations, including wetlands. The wetlands were defined as marshy, mostly or always wet areas in contrast to drier, sometimes seasonally wet areas found on the slopes and terraces of waste-rock piles. Rush and sedge species were found almost exclusively in the wetlands while shrubs were largely confined to the waste-rock piles. All willow species were categorized as a tree lifeform.
Plant sampling and preparation Transects with sampling points approximately 30-50 m apart were established for sampling the above-ground vegetation on the slopes and terraces of lifts 1 through 4 of Wooley Valley Unit 4 (WVU4), and the slopes of Wooley Valley Unit 1 (WVU1) and Unit 3 (WVU3). A lift is defined as a terrace paired with the adjoining slope below it. Additionally, an old adit waste-rock pile was sampled at Maybe Canyon (MC-WRP). Transects were also established at the reference sites, Dairy Syncline (DS), Deer Creek (DC), Dry Valley (DV), Rasmussen Ridge (RR), and Maybe Canyon (MC-TR). Transects were selected with reference to the geology and contour of the land, therefore sampling points along each transect varied between 10 and 50 m, depending on the site. The mineassociated wetlands were located at the base of the WVU 1 and WVU4 waste-rock dumps. The reference wetland exists in an adjacent, parallel drainage to the WVU4 mineassociated wetland, and was not significantly impacted by the mining wastes. Distances between sampling points along each transect were measured with a hip chain. The latitude and longitude of each sampling location and transect starting points were recorded with a GPS unit (DeLorme, Yarmouth, ME). The above-ground portion of each sampled plant species was clipped and placed in labeled paper bags and taken to the lab for pre-analysis preparation. Samples of plant species that could not be identified in the field were also collected and returned to the lab for identification using various plant identification manuals (e.g. Cronquist et al., 1977). A list of identified plant species is given in Table 19-I. The plant samples were dried in mechanical convection ovens at 60~ for 48 h and ground to < 20 mesh in a stainless steel grinding mill. All splits provided to the analysts were obtained with a riffle splitter to ensure the splits were representative of the whole sample.
532
C.L. Mackowiak et al.
TABLE 19-1 Plant species sampled at the sites Species
Common name
Lifeform
Agropyron caninum Agropyron intermeium Agropyron smithii A gropyron spicatum Bromus inermis Bromus marginatus Dactylis glomerata Deschampsia caespitosa Elymus cinereus Elymus glaucus Festuca arundinacea Glyceria striata Onobrychis viciifolia Phleum pratense Poa pratensis Stipa occidentalis Juncus ensifolius Carex haydeniana Carex hoodii Carex rossii Carex utriculata Achillea millifolium Balsamorhiza sagittata Cirsium arvense Epilobium angustifolium Equisetum arvense Lupinus alpestris Pedicularis racemosa Penstemon rydbergii Phacelia hastata Senecio crassulus Medicago sativa Melilotus officianalis Amelanchier alnifolia Artemesia tridentata Ceanothus velutinus Populus tremuloides Salix alba Salix geyeriana
Bearded wheatgrass Intermediate wheatgrass Western wheatgrass Arizona wheatgrass Smooth bromegrass Mountain bromegrass Orchardgrass Tufted hairgrass Wildrye Blue wildrye Reed fescue Fowl mannagrass Common sainfoin Timothy Kentucky bluegrass Western needlegrass Swordleaf rush Cloud sedge Hood sedge Ross sedge Beaked sedge Western yarrow Arrow leaf Canada thistle Fireweed Field horsetail Great Basin lupine Sickletop lousewort Rydberg beardtongue Scorpion weed Thickleaf groundsel Alfalfa Yellow sweetclover Saskatoon serviceberry Big sagebrush Snowbrush, buckbrush Aspen White willow Geyer willow
Grass Grass Grass Grass Grass Grass Grass Grass Grass Grass Grass Grass Grass Grass Grass Grass Rush Sedge Sedge Sedge Sedge Forb Forb Forb Forb Forb Forb Forb Forb Forb Forb Legume Legume Shrub Shrub Shrub Tree Tree Tree
Continued
533
Uptake of selenium and other contaminant elements into plants
Table 19-I Continued Species
Common name
Lifeform
Salix lasiandra Salix lutea Salix monticola Salix myrtillifolia Salix scouleriana
Pacific willow Yellow willow Mountain willow Blueberry willow Scouler willow
Tree Tree Tree Tree Tree
Analyses
Plant samples collected in 1999 and 2000 were analyzed by the US Geological Survey Laboratory in Denver, Colorado (Herring and Amacher, 2001). Selenium and As were determined by hydride generation atomic-absorption spectroscopy (HG-AAS) (Arbogast, 1996). Samples were also analyzed for 40 major, minor, and trace elements using four-acid digestion in conjunction with inductively coupled plasma-atomic emission spectrometry (ICP-AES) using the methods of Lamothe et al. (2000). For this chapter, discussion will be limited to the Se, Cd, Cr, Cu, Mn, Mo, Ni, and Zn data. Plant samples collected in 2001 were analyzed by the Soil and Water Analysis Laboratory at the Logan Forestry Sciences Laboratory of the US Department of Agriculture, Forest Service. The HNO3 + H202 method of Jones (1989) was used to digest the samples. Selenium in the digests was determined by HG-AAS. To convert oxidized Se in the digests to Se(IV) for hydride analysis, equal volumes of digest and concentrated HCI were heated for 20 min at 80~ in a graphite block digester. Near complete conversion of Se(VI) to Se(IV) was confirmed by spiking several plant-digest samples with known amounts of Se(IV) and Se(VI) and carrying these spiked samples through the Se-species conversion procedure. Recovery of spiked Se(VI) averaged 93.4%. Stat&tical analyses
Summary statistics (sample size, mean, standard deviation, minimum, and maximum value) of the vegetation data are given in the CD, Appendix 5. Since the data were nonnormally distributed, statistical analysis of data by location was performed using the Kruskal-Wallis one-way analysis of variance (ANOVA) on ranks with median comparisons by Dunn's method (Sigma Stat 2.0, Jandel Scientific, San Rafael, CA) and presented as box plots. Transect data are presented as line charts. Symbols represent median values and the 25th and 75th percentiles are shown as error bars. Plant lifeform data are presented in bar charts as means with standard error bars. Legume and grass Se concentrations at Wooley Valley Unit 4 were mapped and analyzed with the Arc-GIS 8.0 using the Spatial Analyst interpolation procedure (ESRI, Inc., Redlands, CA), with each point having a 40 m radius of influence. A 40m radius was chosen based on the distance between sampling points (approximately 30-50 m) within each transect.
534
C.L. Mackowiak et al.
RESULTS AND DISCUSSION
Selenium in vegetation Geographic effects Because of the highly variable number of plant species, data were pooled for each location to present general differences among sample locations. Data for individual plant lifeforms can be found in the CD, Appendix 5. As expected, the un-mined or reference locations, (i.e. DS, DV, and RR) had the lowest vegetation Se concentrations among the study sites, with total Se values below 2 mg kg -1 dry mass (Fig. 19-2). The DC reference location had plants with higher total Se because some samples were taken from a disturbed trench area, but even so, tissue values remained below 5 mg kg-1. For the MC-TR exploration trench, a moderately disturbed area, biomass Se was nearly twice the 5 m g k g -1 upper limit recommendation (CD, Appendix 5). The waste-rock dumps were severely
Fig. 19-2. Box plots of plant Se concentration as a function of location: Dairy Syncline (DS), Dry Valley (DV), Rasmussen Ridge (RR), Deer Creek (DC), Maybe Canyon transect (MC-TR), Maybe Canyon waste-rock pile (MC-WRP), Wooley Valley unit 1 (WVU1), Wooley Valley unit 3 (WVU3), and Wooley Valley unit 4 (WVU4). The DS, DV, RR, and DC locations are reference sites, the MC-TR was moderately disturbed, and the WVU1, WVU3, and WVU4 sites are highly disturbed waste-rock dumps. The 25th and 75th percentiles are shown as a box around the 50th percentile, the 10th and 90th percentiles are shown as error bars, and the 5th and 95th percentiles and outliers are shown as points. Sample size (N) varied (see CD, Appendix 5).
Uptake of selenium and other contaminant elements into plants
535
disturbed and the Se values reflect this (Fig. 19-2), but waste rock from WVU3 resulted in much lower biomass Se (mean 2 mg kg-1). From its appearance, this material was primarily low Se-bearing chert and other overburden material. Because of the variation in wasterock composition, it is difficult to evaluate the degree of land disturbance based solely on plant analysis for Se, but the trends are clear (Fig. 19-2). Interestingly, plants on the wasterock dumps continue to contain very high Se concentrations even though the land has been revegetated and undisturbed for several years. Although much bioavailable Se has been leaching through the waste-rock into the adjacent drainage seeps for years, bioavailable forms within the waste-rock piles have not diminished.
Geologic effects
Besides containing the Meade Peak Phosphatic Shale Member, the Phosphoria Formation typically contains a Rex Chert Member and limestone transition strata (Hein et al., 2002, Chapter 14). Since Rex Chert and Wells Limestone tend to be lower in Se and other trace-element contaminants, they are being considered for use as capping material over phosphatic shale waste-rock prior to re-vegetation. Vegetation samples were collected along transects of the undisturbed sites DS, DC, RR, and the exploration trench MC-TR to determine if the mineralogy of undisturbed areas was reflected in Se content of vegetation. The portion of the transect overlying Meade Peak Phosphatic Shale at the MC-TR site had plants with the highest Se values, whereas portions of the transect overlying either Wells Limestone or Rex Chert had much lower plant-Se values, as was also the case at the DC site (Fig. 19-3). The DS site had four transects sampled, but only transect 2 suggested a relationship between Meade Peak rocks and vegetation Se (Fig. 19-4). Additionally, the data suggest that phosphatic shale may be in closer proximity to the rooting environment at the DC and MC sites than at the RR and DS sites. Therefore, indiscriminant sampling of non-accumulating plant species cannot be used for geoexploration unless the phosphatic shale ore is near the root environment. Perhaps more deeply rooted species, such as alfalfa or trees may indicate shales that exist further (> 1 m) below the surface. Additionally, sampling primary accumulators that bio-accumulate Se several-fold more than other plant species (Rosenfeld and Beath, 1964; McGrath, 1998) may be useful in targeting deposits. However, these sites did not host primary accumulators.
Redox effects
Wetlands receiving drainage from the waste-rock dumps produced vegetation with more than ten times the amount of Se found in vegetation from the reference wetland (Fig. 19-5). All but one of the reference wetland tissue samples were below 0.5 m g k g -1 Se. Additionally, the Se content of wetland vegetation was similar to that of vegetation on associated waste-rock dumps (Fig. 19-5), suggesting the wetlands received much of their Se from drainage of the waste-rock terraced above them. Additionally, Vance (2000) found
536
C.L. M a c k o w i a k et al.
10
T ...........................
Maybe Canyon ade Peak Phosphatic Shale
Wellr Limestone
~
Rex Chert
,7 0
20
40
60 80 Distance (m)
100
120
6
140
Deer Creek
s r
" 4 E v 3 co 2
Meade Peak Phosphatic Shale
13r}
Rex Chert ....... , .......
Wells
"
t0
|
0
100
200 Distance (m)
400
300
Rasmussen Ridge
Lower Middle Below Upper middle Lower Upper ore Waste Waste Lower ore ore ~ 1 ~ ~ t
Above upper ore
. . _ . , .........
!
0
5
10
15 20 Distance (m)
I
!
25
30
35
Fig. 19-3. The influence of sedimentary rock units across transects on plant total Se concentration at the Maybe Canyon transect (MCTR), Deer Creek (DC), and Rasmussen Ridge (RR) reference locations. Symbols represent median values and the 25th and 75th percentiles are shown as error bars. The DC transect had 1-2 samples per data point so percentiles could not be created. Sample size (N) for each data point varied from 1 to 6 along each transect.
a connection between downstream surface-water Se and overburden waste-rock at another local site (Maybe Canyon). Transects were sampled along one of the WVU4 wetland seep flow paths. Plant Se was high (median ~80mg kg -l) closest to the seep but it declined almost exponentially with distance along the flow path (Fig. 19-6). Wetlands are useful for removing Se in water via immobilization, plant uptake, and volatilization (Hansen et al., 1998; Pilon-Smits et al.,
537
Uptake of selenium and other contaminant elements into plants
1.6
Transect 1
1.2 0.8 0.4
Meade Peak Phosphatic Shale Wells Limestone ................................................ .,.. . . . . . . . . . . :-.................................................................. . . _ ....... ::.......
'
t
I I
0.0 1.6
Rex Chert
~'*
I
I
Meade Peak Phosphatic Shale
Transect 2
1.2 0.8
0.4
mestone
Rex Chert {....,. ...... :
~-... --.---,
CO
= 0.0 co 1.6
t ~ - ~
I--
/
I
I ..u.
I
----.L ?
I
I
Transect 3
Meade Peak Phosphatic Shale 1.2
...................................................................................... '~...................................................................................... ...
"..
T
0.8 0.4 0.0 1.6
1
1
I
I
I
I
I
I
Transect 4
1.2 Meade Peak Phosphatic Shale .. ........................................................... "J".............................................................
0.8
'
0.4 0.0
0,
2'0
_
x ~
-
I
I
40
' I
Rex Chert
~ I
60 80 Distance (m)
I
I
100
120
140
Fig. 19-4. The influence of sedimentary rock units across transects on plant total Se concentration at the Dairy Syncline (DS) reference location. Symbols represent median values and the 25th and 75th percentiles are shown as error bars. Sample size (N) for each data point varied from 2 to 4 along each transect. 1999). Since these processes occur concurrently, their individual contributions towards removing bioavailable Se along the flow path are unknown. Hansen et al. (1998) found plant uptake dominated Se immobilization in a constructed wetland and Lemly (1999) proposed that most wetland Se becomes immobilized by a combination of chemical precipitation and detrital deposition caused by plants and other organisms. Stillings and
538
C.L. M a c k o w i a k et al.
Fig. 19-5. Box plots of plant total Se as a function of wetland or lifts location in Wooley Valley. The reference wetland was located approximately 200 m from the unit 4 wetland. The unit 1 (U 1) and unit 4 (U4) wetlands receive part of their water inflow from the unit 1 and unit 4 lifts (i.e. waste-rock dumps), respectively. Box statistics as in Fig. 19-2. Sample size (N) varied (see CD, Appendix 5). 200
150 C~
E 100 cO
-~ 9 50
I..-
0
!
|
i
|
!
|
|
20
40
60
80
1O0
120
140
160
Distance from seep (m) Fig. 19-6. The influence of unit 4 wetland sampling location on plant Se concentration, where concentrations decline as plants are located further from the seep. Symbols represent median values and the 25th and 75th percentiles are shown as error bars. Sample size (N) varied from 6 to 22 along the transect. Amacher (Chapter 17) in the WVU4 wetland, during the season and found to co-precipitate
observed that iron-oxide minerals precipitated along the flow path particularly in areas that experienced frequent wetting and drying that received upwelling of iron-rich groundwater. Selenium was or sorb to iron oxides along the wetland flow path (Figs 17-6 and
Uptake of selenium and other contaminant elements into plants
539
17-7; Stillings and Amacher, Chapter 17). Several reports describe the nature of Se/Fe oxide and hydroxide sorption interactions (Balistrieri and Chao, 1987, 1990; Peak and Sparks, 2002). Much less Fe-oxide formation was found at another Se-contaminated wetland located at South Maybe Canyon and water-Se concentrations declined to a lesser extent along its flow path (data not shown). This suggests Fe-oxide sorption processes may be important in mitigating plant Se uptake along the WVU4 wetland flow path. A more detailed characterization of the WVU4 wetland water and sediments can be found in Chapter 17 (Stillings and Amacher).
Vegetation effects Extensive vegetative sampling was conducted at the WVU4 waste-rock dump where the waste rock was deposited in a series of four lifts (terraces and slopes). Overall, mean vegetation Se from WVU4 was 38 mgkg -1. The large variability at the site masked any differences among terraces and slopes with regard to plant Se. Applying spatial mapping techniques with interpolation functions better delineated the extent of Se contamination among plant species at the dumpsite. In Fig. 19-7 (lower panels), the degree of plant Se uptake is color-coded, with dark green representing dry-mass Se below the 5 mg kg -! threshold and colors progressing to red, with dark red representing biomass Se above 600 mg kg-I. Based on the panels, it appears the geographic aspect (terrace vs. slope) had less impact on vegetation Se than did plant species or land disturbance. The legumes, primarily Medicago sativa, contained Se in the above-ground biomass in excess of 50 mg kg -I at several lift locations. The 1990 aerial photo in Fig. 19-7 (upper panel) shows a road between terrace 3 and slope 4, as well as some disturbed (brown) areas on terrace 2 and between slope 1 and terrace 1 that have since been revegetated. The historically disturbed areas coincide to some degree with higher vegetation Se in the legumes and grasses (Fig. 19-7, lower panels). However, the grasses, primarily Agropyron smithii and Bromus ineris, took up much less Se than did alfalfa, where tissue values above 50mg kg -1 were limited primarily to terrace 3/slope 4 and parts of terrace 1 (Fig. 19-7, lower right panel). Areas where grass Se values were exceedingly high may qualify for immediate remediation, while the majority of the lifts might be left untreated, after eradicating the legumes. As long as the legumes remain, their enhanced Se uptake will provide dangerously high Se food for livestock and wildlife and may actually increase grass contamination by increasing litter Se. Nearly 40 years ago it was reported that agricultural crops grown in proximity to the Se accumulator species resulted in higher tissue Se in those crops (Rosenfeld and Beath, 1964). More recently, Bafiuelos et al. (1992) showed that Medicago Se concentrations increased when Se-rich plant material was incorporated into the soil. Alfalfa has been an especially popular choice for revegetation efforts because it quickly meets ground coverage requirements and livestock preferentially graze alfalfa. However, alfalfa is a vigorous grower with deep penetrating roots that result in greater Se uptake during the growing season. Stark and Redente (1990) proposed that alfalfa's ability to
540
C.L. Mackowiak et al.
Fig. 19-7. Series of maps representing Se in above-ground plant biomass (bottom two panels). Upper panel is an aerial photograph of the Wooley Valley Unit 4 (WVU4) mine. The white points in the lower panels represent locations of individual samples collected from 1999 to 2001. The grass and legumes were sampled from the same points.
accumulate more trace elements from oil-shale deposits was because of its deeper root penetration. Selenium uptake by alfalfa grown in 1 and 3 kg pots was similar to grass species uptake (data not shown), but the Se content of alfalfa from the waste-rock dump sites averaged four times greater than Se in grass species (Fig. 19-8), supporting the claims of Stark and Redente (1990). Therefore, replacing the commonly used "Ranger" variety
Uptake of selenium and other contaminant elements into plants
541
Fig. 19-8. Bar graph of plant Se composition as a function of plant lifeform found in wetlands and waste-rock dumps. The dashed line represents a plant critical content of 5 mg kg-i. Data are presented as mean values and error bars are standard errors. Data were pooled from all locations and sample size (N) varied (see CD, Appendix 5).
with a more fibrous or lateral rooted variety may lessen Se uptake. Lamb et al. (2000) are developing fibrous rooted alfalfa populations that have even greater herbage yields than traditional tap rooted populations. Although legumes averaged some of the highest Se concentrations among the pooled waste-rock dump data, grass and tree species also had Se mean values well above the 5 mg kg -1 upper threshold (Fig. 19-8). In contrast, forb and shrub mean Se values were closer to the threshold (Fig. 19-8). Substituting native shrub and forb species for alfalfa in the revegetation mix and adjusting the revegetation coverage standards to accommodate such management changes may lessen the risk of Se toxicosis in livestock and wildlife. More extensive sampling of native and introduced forbs and shrubs, both in the field and through the use of potted plant studies using representative soils, may provide critical information for the formulation of acceptable revegetation mixes for this region. Compared with the waste-rock dumps, plants associated with the wetlands tended to contain more Se, except for legumes and trees, which took up more Se from the waste-rock dump sites (Fig. 19-8). Water deep in the waste-rock piles may provide an additional source of available Se for plants whose roots can reach it (e.g. legumes and trees), particularly under drought conditions. Near the end of the summer dry season the grasses and forbs died back while the alfalfa and trees remained green, suggesting that because of their rooting habit they are able to exploit water and bioavailable Se deep in the waste-rock dumps. However, representative sample numbers (N) of some of the plant lifeforms were low (CD, Appendix 5). Thus, additional sampling may be necessary to confirm differences between plants growing in wetlands and waste-rock soils.
C.L. Mackowiak et al.
542
Other trace elements in vegetation Plants from the reference and mine sites were analyzed for other trace elements. Some, such as Cu, Mn, Mo, Ni, and Zn, are essential to plants, whereas Cd and Cr are not. Sedimentary rock units of the Phosphoria Formation contain high trace-element concentrations (Grauch et al., Chapter 8; Herring and Grauch, Chapter 12), in places resulting in greater plant uptake, even at undisturbed locations. However, no significant correlations were found between trace elements and the occurrence of Meade Peak Phosphatic Shale along the sampled transects (data not shown). Upper critical ranges for the aforementioned trace elements in plants are provided in Table 19-11.
Cadmium Cadmium is a non-essential plant trace element that is highly mobile and is concentrated in some shales. These characteristics have caused concern regarding temporal increases in soil Cd through the application of fertilizers, particularly manure and sludge, to agricultural lands (Kabata-Pendias, 2001). Plant mean values tended to be high at all
TABLE 19-11 Approximate upper critical plant and livestock/wildlife intake ranges of some selected trace elements; all values given as dry mass Element
Cd Cr Cu Mn Mo Ni Se Zn
Plant upper critical content 1 (mg kg-l) 5-30 5-30 20-100 400-1000 10-50 10-100 5-30 100-400
Livestock/wildlife acute upper critical intake2 (ppm of diet day-l)
Livestock/wildlife Chronic upper critical intake 2 (ppm of diet day -l)
2000 to 3000 1000 to 3000 400 to > 2000 2000 to > 4000 20 to > 200 700 to > 1000 80 to > 1003 2000 to > 5000
50 to 500 50 to > 1000 30 to > 100 500 to > 2000 10 to > 200 700 to > 1000 5 to > 20ppm 3 500 to > 2000
1 Values taken from Kabata-Pendias (2001); ranges exclude highly sensitive or tolerant plant species. 2 Ranges are based on summary of data in Puls (1994); Se ranges based on data from Miller and Williams (1940), Rosenfeld and Beath (1964), Olson and Embry (1973), Mahan and Moxon (1984), and Puls (1994); the ranges are based on data from cattle, sheep, goats, and horses, taking into account susceptibility due to age and the chemical form. 3 Organic.
Uptake of selenium and other contaminant elements into plants
543
sample locations but they were particularly high in trees, with maximum values above 10 mg kg- 1 (CD, Appendix 5; Fig. 19-9). Concentrations in the range of 10-20 mg k g - 1 are considered toxic for many crop plants (MacNicol and Beckett, 1985). There was little difference in Cd content between plants grown in wetlands vs. rock piles, likely due to its
Fig. 19-9. Bar graphs of plant trace-element compositions as a function of plant lifeform in wetlands and waste-rock dumps. Data are presented as mean values and error bars are standard errors. Data were pooled from all locations and sample size (N) varied (see CD, Appendix 5).
544
C.L. Mackowiak et al.
high bioavailability in both environments (Fig. 19-9). Hutchison and Wai (1979) found alfalfa accumulated more Cd than grasses in Idaho phosphate mine wastes, similar to our observations, but our tree-Cd values were much higher than Cd in alfalfa or grasses, even at the reference locations (see CD, Appendix 5), suggesting the tree species (i.e. Populus tremuloides and Salix spp.) may hyperaccumulate Cd.
Chromium
Nearly all sample sites had some plants with Cr above 1 mg kg-1 (see CD, Appendix 5). Although the upper critical range is reported at 5 - 3 0 m g k g -1 (Table 19-II), ryegrass at 3 mg kg -1 Cr and white clover at 9 mgkg -1 Cr resulted in a 50% yield depression (Dijkshoorn et al., 1979). Our data show plants from the waste-rock piles may have Cr contents high enough, particularly in the forbs, to affect plant growth (Fig. 19-9). The reducing conditions found in wetlands converts Cr(VI) to Cr(III), which then may precipitate as hydroxides or sorb to mineral surfaces, thereby lessening Cr bioavailability (Eary and Rai, 1988; Losi et al., 1994). As expected, wetland plants had much lower Cr content than plants sampled from the waste-rock piles (Fig. 19-9).
Copper
Tissue Cu was variable across sampling sites with ranges from less than 1 to more than 20 mg kg- 1 (see CD, Appendix 5). In general, plant concentrations below 1-5 mg kg-1 may be deficient, whereas 20-30 mg kg- l can be toxic for many species (Marschner, 1995), and the upper critical range is 2 0 - 1 0 0 m g k g -1 (Table 19-II). There was little difference between plants sampled from wetlands vs. rock piles, except that the waste-rock forbs had the highest Cu values (Fig. 19-9). Pedicularis racemosa Dougl. had the highest Cu concentrations (-20mgkg-1), which weighted the forage mean value higher. Perhaps this species has a greater affinity for Cu, although we did not find support in the literature for this idea.
Manganese
Some of the highest tissue-Mn values came from the DS reference site and Mn was particularly low at the waste-rock dumps (see CD, Appendix 5). Although plant deficiencies are more common with high pH, the shales were below pH 8 and tissue values were above the deficiency level (10-20 mg kg- 1) reported by Marschner (1995). Like Cr, redox greatly affects Mn availability, where wetlands typically contain dissolved free Mn 2+ that is readily taken up by plants. Additionally, our data show much higher Mn concentrations in the
Uptake of selenium and other contaminant elements into plants
545
wetland plants (Fig. 19-9). Plant Mn toxicity is highly species dependent, where sensitive species show symptoms around 200 mg kg- 1 but ranges are typically 400-1000 mg kg- 1 (Table 19-11).
Molybdenum
Tissue-Mo values from the WVU4 waste-rock piles were greater by far than values from any of the other sites (see CD, Appendix 5). This suggests that the WVU4 soil or waste rock has inherently more bioavailable Mo than soils from the other locations. Additionally, the waste-rock piles appear to have more bioavailable Mo than the wetlands (Fig. 19-9), which may be attributed to Mo retention in soil under low redox conditions (Lindsay, 1979). Nitrogen fixation in leguminous species results in a higher Mo demand and therefore legumes accumulate more Mo than non-legume species, as is reflected by the plant lifeform results (Fig. 19-9). Plant concentrations below 5 mg kg-1 are typical in nonlegume species but legumes can accumulate more than 500 mg kg-l without showing toxic effects (Kabata-Pendias, 2001). Amounts above 5-10 mg kg-l in forage may be detrimental to livestock, particularly if plant Cu is relatively low. Further discussion of animal effects is provided later in the chapter.
Nickel
Although Ni is classified as an essential plant trace element, deficiencies are extremely rare. However, plant toxicities have been reported at approximately 10 mg kg-I for sensitive species and the typical upper critical range is approximately l 0-100 mg kg-1 Ni (Table 19-II). Two grasses, wheat and ryegrass, were found to have upper critical values of approximately 30 and 120mgkg -1, respectively (MacNicol and Beckett, 1985), reflecting how species-specific Ni tolerance is (Uren, 1992). Many of our higher (> 10mg kg-l) Ni tissue samples came from MC-WRP and WVU4 waste-rock pile sites (see CD, Appendix 5). Vegetation samples taken from the waste-rock piles tended to have greater Ni contents than samples from the wetlands, and forbs, trees, and legumes accumulated more Ni than grasses (Fig. 19-9). A review by Uren (1992) concluded that Ni will sorb to oxides under conditions of wetting and drying, and that organic matter has a high affinity for Ni. Perhaps these or other conditions decreased Ni bioavailability in the wetlands.
Zinc
Mean plant Zn content was greatest at the MC-WRP and WVU4 waste-rock dumps (see CD, Appendix 5). Critical high Zn values in vegetation were reported to be near 300 mgkg -l for several grass species (MacNicol and Beckett, 1985) and upper critical
546
C.L. Mackowiak et al.
values typically range from 100 to 400 mg kg- 1 Zn (Table 19-I1). All but the DV reference site and the Wooley Valley waste-rock sites contained some samples in this high range (see CD, Appendix 5). As with Cu, there was no appreciable difference between samples taken from wetlands vs. rock piles but there was a significant plant lifeform effect, where trees contained several times more Zn than the other lifeforms (Fig. 19-9). The trees consisted of two genera, Salix and Populus. They survive in metal-contaminated lands and therefore are being evaluated for metal remediation (Amiro and Courtin, 1981; Schnoor et al., 1995; Punshon and Dickinson, 1999). High Zn in plants tends to inhibit Cu uptake but tree Cu values were comparable to the other plant lifeform data. Deficiency concentrations are reported to be below 15-20 mg kg-1 (Marschner, 1995) so the non-tree species appear to have had adequate Zn nutrition. It is interesting to note that Cd and Zn may use the same transport systems for plant uptake (Guerinot, 2000), and once in the plant, excess Zn and Cd can accumulate in vacuoles (Brune et al., 1995). There was a positive correlation (r 2 = 0.747; n = 170) between Zn and Cd tissue levels suggesting the two elements may be linked with regards to uptake, and the unusually high concentrations found in trees may hold promise in geoexploration.
Livestock~wildlife response to selenium and other trace elements Selenium
Plant concentrations of Se exceeded the acute and chronic thresholds for potential poisoning at numerous disturbed sites, such as the waste dumps (see CD, Appendix 5; Table 19-II). Selenium occurs in both organic and inorganic forms. Selenomethionine is the primary plant form, where Se is substituted for sulfur in the amino acid methionine (Schrauzer, 2000). As such, this form of Se is absorbed and deposited in tissues in a similar manner to methionine. It must be noted that most of the literature regarding Se poisoning pertains to the inorganic form, which is commonly added to feed rations. The toxic effects of organic and inorganic forms are the same, but in some cases the toxic dose of organic Se is slightly greater than that of inorganic Se (Herigstad et al., 1973; Ammar and Couri, 1981). Acute Se poisoning is associated with ingestion in amounts that range from 2.2 (Rosenfeld and Beath, 1964) to greater than 20 mg kg-1 of body weight (BW) (Miller and Williams, 1940; Mahan and Moxon, 1984). The clinical syndrome is similar across species and is generally characterized by respiratory distress, restlessness or lethargy, anorexia, weight loss, gaunt appearance, salivation, watery diarrhea, fever, tachycardia, teeth grinding, stilted gait, and (or) death (Franke and Moxon, 1936; National Resource Center, 1983; Puls, 1994). Onset of clinical signs can be delayed for 12-36h post-ingestion, with prostration and death occurring after a short duration of clinical signs. A garlicky smell to the breath of poisoned animals is usually present, especially early in the clinical syndrome. In addition, excessive sweating is common in horses. The toxic dose and effect of various
Uptake of selenium and other contaminant elements into plants
547
forms of Se are not known in most wildlife species, but due to the similarities in diet and clinical effects among a wide range of domestic animals, Se dose and effects in wildlife are expected to be quite similar to those of domestic animals. Chronic toxicity, sometimes referred to as alkali disease, results from prolonged consumption of feeds that contain Se of 5-40 mg kg-1. This disease is associated with depression, weakness, emaciation, anemia, hair loss, anorexia, diarrhea, weight loss, and death (Rosenfeld and Beath, 1964). In addition, hoof abnormalities are frequently identified in ungulates and include swelling of the coronary band, hoof deformities, and/or separation and sloughing of the hoof wall. In cattle, horses, and mules the hair at the base of the tail and switch is frequently lost (Bobtail disease) and in pigs, goats, and horses there may be general alopecia (Franke et al., 1934). Chronic toxicity may cause death in sheep without the observation of any clinical signs. Unlike other species, wool loss and hoof lesions are not commonly identified in sheep, although wool production decreases. Diet concentrations of 5 - 1 0 m g k g -1 may reduce reproductive performance without the typical signs of alkali disease (Olson, 1978). In fact, at least some of the adverse effects on reproduction attributed to excess Se are caused by interference with absorption and retention of Cu, which results in a Cu deficiency (Amer et al., 1973). Chronic Se poisoning occurs more frequently than acute poisoning. Ingested Se bioaccumulates, thus causing adversities over longer periods of time. The National Research Council recommends that the appropriate dietary concentration is approximately 0.1-0.3 mg kg-1 Se on a dry matter basis (DM) for most classes of livestock, while the maximum tolerable dietary content is 2 mg kg-l Se (National Research Council, 1980b). Dietary concentrations of 4-5 mg kg-1 are sufficient to inhibit growth in animals fed an otherwise normal diet. Most adult large animals (cattle, horse, etc.) eat approximately 1.5-4% of their BW in DM daily, thus 1.5-4 kg DM per 100 kg BW (Osweiller et al., 1985). Typically, the larger the body size and the more mature the animal, the lower the daily dietary intake as a percent of their BW. For example, intake of adult horses will range from 1.5 to 2% of their BW, adult cattle will range from 2 to 2.5%, and adult sheep will range from 2.5 to 4%. Younger animals eat a greater percentage of their BW daily, with intakes being as much as 1-1.5% of their BW greater than that of an adult for the same species. Wildlife dietary intake is assumed to be similar to that of domestic species in terms of age and size, so a range of 1.5-4% of their BW in daily DM intake would be expected. Stowe and Herdt (1992) reported subacute toxicosis in horses fed Se at 0.4-0.6 mg kg- 1 BW day-1. This is equivalent to 13-40 mg kg -1 in the diet, assuming intakes of 1.5-3%. Another report by Harr and Muth (1972) suggested a decreased conception rate and increased fetal reabsorption rate in cattle, sheep, and horses fed natural diets containing 2 0 - 5 0 m g k g -1 Se in the diet. The dose reportedly corresponded to 0.5-1.5 m g S e k g -1 BW day-1. These Se concentrations also produced other signs of toxicosis including hair loss, lameness, and degenerative fibrosis of the heart, liver, and kidney. In addition, it must be noted that Se accumulates in the fetus at the expense of the dam. Consequently, a dietary intake of 4-5 mg kg- 1 of diet may produce toxicosis in a calf without clinical signs in the cow. Abortions, stillbirths, or weak/lethargic calves may result (Puls, 1994).
548
C.L. Mackowiak et al.
From our mine site data, forage Se concentrations (see CD, Appendix 5) are of a concern in regard to the occurrence of toxic effects in livestock and wildlife (Table 19-11). Assuming total dietary intake from vegetation at the waste rock pile "hot spots," acute poisoning would be likely with merely a single day's intake for cattle, sheep, horses, and grazing wildlife. Furthermore, if only the mean concentration was ingested of certain forages from several of these sites, it could result in either acute or chronic poisoning. The delay in onset of acute poisoning and, more specifically, chronic poisoning, might result in these animals dying in areas that are away from the primary contamination sites. In each of these grazing species one would assume movement patterns such that daily intake exclusively in one of the highly contaminated sites would not occur continuously over many days. Even with ingestion of plant material from the contaminated sites at variable intervals, chronic poisoning is likely. Organic Se, which accounts for 85-100% of total plant Se in non-accumulating species (Rosenfeld and Beath, 1964), has a long halflife in animals (Gronbaek and Thorlacius-Ussing, 1992; Human Health Fact Sheet, 2001), allowing for bioaccumulation that would likely result in chronic poisoning with sporadic, repeated exposures. Since chronic poisoning generally results in a debilitating syndrome, these animals would be more prone to predation, resulting in animal losses that were not directly attributed to Se poisoning and a loss of tissues for testing that could verify excess Se content.
Molybdenum
As with Se, there were many sites that surpassed the Mo thresholds for potential acute and chronic poisoning (see CD, Appendix 5; Table 19-II). Although ingestion of large concentrations of Mo can result in poisoning (Swan et al., 1998), sub-acute to chronic poisoning, in which Mo-Cu interaction is the primary cause, is more common (Ward, 1978; Suttle, 1991; Puls, 1994). Animal Cu/Mo ratios of between 3 and 8 are required for optimal performance. Excess Mo in the diet, Cu/Mo ratios less than 2 or primary forage Mo of > 10ppm, results in the production of non-bioavailable Cu-Mo complexes. This can occur in the digestive tract, preventing the absorption of Cu from the diet, but it can also result in the depletion of liver Cu stores. The mean and maximum forage Mo concentrations identified from several study locations (see CD, Appendix 5) exceed the threshold concentrations for acute and chronic poisoning (Table 19-II). The toxic effects of Mo are of much greater concern in ruminants than in non-ruminants. Although excess Mo can affect horses, our reported concentrations are not likely to be harmful. However, the primary clinical manifestation in ruminants would be due to the Cu-Mo interactions resulting in clinical Cu deficiency. The clinical effects of this type of syndrome would be associated with poor growth, diarrhea, infertility, loss of hair color, poor immune function, and/or bone and joint growth abnormalities in young animals (Ward, 1978; Puls, 1994). The overall potential for chronic effects due to Mo at these sites is much lower than for Se, because Mo does not bioaccumulate and is readily eliminated. With animals migrating
Uptake of selenium and other contaminant elements into plants
549
to various grazing sites, the interference with Cu absorption would only occur immediately after grazing on a contaminated site. Subsequent grazing at non-contaminated sites would allow for Cu absorption. However, it must be noted that several of the reference-test sites had marginal CtffMo ratios which would not allow for much improvement in overall Cu status when grazing off of the contaminated sites (see CD, Appendix 5). This, in conjunction with the antagonism of Cu absorption/retention by Se, could be an additive factor in causing Cu deficiency in livestock and wildlife grazing in these areas.
Other trace elements
Unlike the Se and Mo results, plant Cd, Cr, Cu, and Ni did not reach toxic thresholds for potential animal poisoning (see CD, Appendix 5; Table 19-11). Although the maximum Mn concentrations exceeded the low end of the chronic toxicity potential for DS reference site forbs, and WVU4 wetland sedges and trees, the mean values for each of these locations were well below that threshold. The maximum Zn concentrations exceeded the low end of the chronic toxicity potential in trees from WVU4 wetlands and tree mean values were also greater at the WVU4 waste-rock piles. These concentrations may affect browsers that primarily feed upon trees, but wildlife field studies would be required to test for potential effects.
Livestock protection
Livestock and wildlife may ingest trace-element contaminants from vegetation growing on waste-rock dumps or drinking from contaminated water supplies. Various remediation techniques are being considered to lessen contaminant impact on the environment.
Physical manipulations
One of the most direct methods for avoiding trace-element ingestion is to physically isolate the contaminant from the target animals. If the contaminant is located in a small area, it could be physically removed and transported to a place where it would pose less of a threat or the contaminated area could be fenced to keep livestock from grazing on the enclosed vegetation. The fencing option is being considered for a portion of the WVU4 wetland. However, fencing will not mitigate environmental effects. Another approach is to cover or cap the contaminated site with soil. Capping with soils containing chert and limestone has recently been used in southeast Idaho. Capping is a long-term solution but it is unknown if it will prevent contaminant drainage along slopes or leaching into groundwater. Studies defining an optimal capping depth that prevents root penetration into the waste-rock have not been conducted but are worth investigating.
550
C.L. Mackowiak et al.
Chemical manipulations Selenium chemistry is complex since redox reactions play an important role in determining Se bioavailability. Water-treatment investigations with zero-valent metals, such as Fe to reduce Se to a less bioavailable species are being attempted (Roberson, 1999). As mentioned previously, Se(IV) readily sorbs to oxides, particularly ferrihydrite (Balistrieri and Chao, 1990) and even Se(VI) forms outer-sphere complexes with oxides (Su and Suarez, 2000; Peak and Sparks, 2002). These iron materials are generally economical and therefore may have a role in Se remediation at the mine sites.
Vegetation manipulations Much attention has been given to phytoremediation as a tool to remove contaminants through the culture of bioaccumulating plants. Primary Se accumulators tend to have slow growth rates (Bafiuelos et al., 1997) but secondary absorbing plants such as the Brassica species or even a non-accumulating plant like alfalfa, which has a high growth rate, may be useful for Se removal. In fact, alfalfa of differing root morphologies are being developed to improve phytoremediation efficiencies of environmental contaminants (Lamb et al., 1999). Phytoremediation is a practical option for sites in southeast Idaho only if: (a) the land can be cultivated and all above-ground biomass is harvested rather than returned to the soil; (b) the trace element removal rate is great enough to remove the bioavailable form from soil in an acceptable period of time, months or years versus decades; (c) the plant or its progeny do not expand beyond the treatment area; and (d) animals are prevented from grazing the vegetation during reclamation. These stipulations can greatly limit the applicability of phytoremediation on waste-rock dump sites. Perhaps the most cost-effective, low-technology method for mitigating trace-element toxicosis in animals is to be more selective in which plant species or types are allowed to populate contaminated sites. As shown previously, deep-rooted plants, such as "Ranger" alfalfa is a poor choice for limiting Se intake by ungulates. An alfalfa-infested area might be treated with a broadleaf herbicide, such as 2,4-D, dicamba, or clopyralid and then seeded with non-accumulating plants that tend to have even lower Se uptake, such as grasses and forbs. Even among grasses there may be some species that accumulate lesser amounts of Se. In a preliminary greenhouse survey study, we found that species such as Festuca ovina and Agropyron smithii tended to accumulate less Se than species such as Stipa columbiana and Dechampsia caespitosa (data not shown). Further work is required to identify the most promising species that are also well adapted to the mine-site environment. Even many of the native forbs and shrubs appeared to take up less Se than legumes (Fig. 19-8). Another advantage of revegetating with native species is that it may restore some of these large tracts of land to a more natural state and provide a more varied food source for wildlife (Chambers et al., 1994).
Uptake of selenium and other contaminant elements into plants
551
CONCLUSIONS Surveying several sites in the southeast Idaho phosphate mine region, we found vegetation Se tended to increase with increasing land disturbance. Vegetation from unmined locations remained below the 5 mg kg-1 dry-mass toxicity threshold recommended by the National Research Council (1980a), but some vegetation from waste-rock dumps contained over 200mgkg -l Se. In wetland environments where Se bioavailability is greatly influenced by redox reactions, Se uptake was much lower if sufficient Fe oxides were present to sorb reduced Se. Plant Se dropped exponentially along a seep flow path to eventually fall to or below the 5 mg kg-l threshold 150 m from the seep. While forbs and shrubs tended to accumulate among the lowest amounts of Se, alfalfa contained some of the highest Se concentrations. Of those trace elements analyzed, Se, Cd, Cr, and Zn may have been high enough in some plants to affect their growth. In particular, the trees accumulated Cd and Zn at concentrations four to five times greater than any of the other plant lifeforms. Additionally, this may pose a health concern for browsing animals feeding on tree shoots and twigs, particularly during the winter when other vegetation is covered with snow. The identified excess forage Se found on waste-rock dumps poses both an acute and chronic risk for livestock and wildlife poisoning. In addition, the high Mo content of the forage, in conjunction with the ability of Se to interfere with Cu uptake, poses a real risk for induction of Cu deficiency in animals that graze these areas. Removing, covering, or otherwise treating soils have been considered for remediating trace-element contaminants in the environment. Plant selection as a criterion for remediating soils through phytoremediation also has been considered, but mainly with regard to utilizing the differences among primary accumulator, secondary absorber, and nonaccumulator plant groups. Our data provide a more thorough review of plant lifeform and species uptake differences within the non-accumulator group. Intelligent selection of specific non-accumulator replacement plants, such as grasses, native forbs, and shrubs might turn a potentially toxic landscape into one that is safer for livestock and wildlife and perhaps at a lower cost than other remediation methods.
ACKNOWLEDGMENTS We thank Deb Kutterer, Tracy Christopherson, Jackson Evans, Bryan Willis, Amy Lewis, and Michelle Colvin for their assistance in sample collection, preparation, and analysis. We thank the USGS lab in Denver, Colorado, Utah State University Analytical Labs, and Kay Laird for sample analysis. We thank Paul Brouha, Terry Harwood, and James Leatherwood of the US Forest Service Washington Office; Bill Burbridge and Maggie Manderbach of Region 4; and Jeff Jones, Dave Whittekiend, and John Lott of the Caribou National Forest for their support and assistance with this project. We thank Phil Moyle and James Hein of the US Geological Survey for their support of this project. We thank James Williams of Astaris Corp and Rob Squires of Agrium for providing access to
552
C.L. M a c k o w i a k et al.
sites and assistance with sample collection. Finally, we thank George Vance and Theresa Presser for providing helpful comments on an earlier version o f this chapter.
REFERENCES Amacher, M.C., Brown, R.W., Sidle, R.C. and Kotuby-Amacher, J., 1995. Effect of mine waste on element speciation in headwater streams. In: H.E Allen, C.P. Huang, G.W. Bailey and A.R. Bowers (eds.), Metal Speciation and Contamination of Soil. Lewis Publ., Boca Raton, F1, pp. 275-309. Amer, M.A., St-Laurent, G.J. and Brisson, G.J., 1973. Supplemental copper and selenium for calves: effects upon ceruloplasmin activity and liver copper concentration. Can. J. Physiol. Pharmacol., 51: 649-653. Amiro, B.D. and Courtin, G.M., 1981. Patterns of vegetation in the vicinity of an industrially disturbed ecosystem, Sudbury, Ontario. Can. J. Bot., 58: 1623-1639. Ammar, E.M. and Couri, D. 1981. Acute toxicity of sodium selenite and selenomethionine in mice after ICV or IV administration. Neurotoxicology, 2: 383-386. Arbogast, B.E, 1996. Analytical methods manual for the Mineral Resource Surveys Program. US Geol. Surv., Open File Report, 96-525, 248 pp. Balistrieri, L.S. and Chao, T.T., 1987. Selenium adsorption by goethite. Soil Sci. Soc. Am. J., 51: 1145-1151. Balistrieri, L.S. and Chao, T.T., 1990. Adsorption of selenium by amorphous iron oxyhydroxide and manganese dioxide. Geochim. Cosmochim. Acta, 54:739-751. Bafiuelos, G.S., Mead, R., Wu, L., Beuselinck, P. and Akohoue, S., 1992. Differential selenium accumulation among forage plant species grown in soils amended with selenium-enriched plant tissue. J. Soil Water Cons., 47: 338-342. Bafiuelos, G.S., Ajwa, H.A., Mackey, B., Wu, L., Cook, C., Akohoue, S. and Zambruzuski, S., 1997. Evaluation of different plant species used for phytoremediation of high soil selenium. J. Environ. Qual., 26: 639-646. Bafiuelos, G.S., Zambrzuski, S. and Mackey, B., 2000. Phytoextraction of selenium from soils irrigated with selenium-laden effluent. Plant Soil, 224:251-258. Bell, P.E, Parker, D.R. and Page, A.L., 1992. Contrasting selenate-sulfate interactions in seleniumaccumulating and nonaccumulating plant species. Soil Sci. Am. J., 56: 1818-1824. Brune, A., Urbach, W. and Dietz, K.J., 1995. Differential toxicity of heavy metals is partly related to a loss of preferential extraplasmic compartmentation: a comparison of Cd-, Mo-, Ni-, and Zn-stress. New Phytol., 129: 403-409. Causey, J.D. and Moyle, P.R., 2001. Digital database of mining-related features at selected historic and active phosphate mines, Bannock, Bear Lake, Bingham, and Caribou Counties, Idaho. US Geol. Surv., Open File Report, 01-142, 46 pp. Chambers, J.C., Brown, R.W. and Williams, B.D., 1994. An evaluation of reclamation success on Idaho's phosphate mines. Restoration Ecol., 2: 4-16. Cronquist, A., Holmgren, A.W., Holmgren, N.H., Reveal, J.L. and Holmgren, P.K., 1977. Vascular Plants of the Intermountain West (Vol. 6), The Monocotyledons. Columbia Univ. Press, New York, 584 pp. Dijkshoorn, W., van Broekhoven, L.W. and Lampe, J.E.M., 1979. Phytotoxicity of zinc, nickel, cadmium, lead, copper and chromium in three pasture plant species supplied with graduated amounts from the soil. Neth. J. Agric., 27: 241-253.
Uptake of selenium and other contaminant elements into plants
553
Eary, L.E. and Rai, D., 1988. Chromate removal from aqueous waste by reduction with ferrous iron. Environ. Sci. Technol., 22: 972-977. Feist, L.J. and Parker, D.R., 2001. Ecotypic variation in selenium accumulation among populations of Stanleya pinnata. New Phytol., 149:61-69. Franke, K.W., Rice, T.D., Johnson, A.G. and Schoening H. W. 1934. Report on a preliminary field survey of the so-called "alkali disease" of livestock. USDA, Circular No. 320, pp. 1-10. Franke, K.W. and A.L. Moxon. 1936. A comparison of the minimum fatal doses of selenium, tellurium, arsenic, and vanadium. J. Pharmacol. Exp. Ther., 58: 454-459. Gough, L.P., Shacklette, H.T. and Case, A.A., 1979. Element concentrations toxic to plants, animals, and man. US Geol. Surv., Bull., 1466, 80 pp. Gronbaek, H. and Thorlacius-Ussing, O. 1992. Selenium in the CNS of rats exposed to 75-Se L-selenomethionine and sodium selenite. Biol. Trace Element Rec., 35:119-127. Guerinot, M.L., 2000. The ZIP family of metal transporters. Biochim. Biophys. Acta, 1465: 190-198. Hansen, D., Duda, P.J., Zayed, A. and Terry, N., 1998. Selenium removal by constructed wetlands: role of biological volatilization. Environ. Sci. Technol., 32: 591-597. Hart, J.R. and Muth, O.H., 1972. Selenium poisoning in domestic animals and its relationship to man. Clin. Toxicol., 5:175-86. Hein, J.R., Mclntyre, B., Perkins, R.B., Piper, D.Z. and Evans, J., 2002. Composition of the Rex Chert and associated rocks of the Permian Phosphoria Formation: Soda Springs area, SE Idaho. US Geol. Surv., Open File Report, 02-345, 30 pp. Herigstad, R.R., Whitehair, C.K. and Olson, O.E., 1973. Inorganic and organic selenium toxicosis in young swine: comparison of pathologic changes with those in swine with vitamin E-selenium deficiency. Am. J. Vet. Res., 34: 1227-1238. Herring, J.R. and Amacher, M.C., 2001. Chemical composition of plants growing on the Wooley Valley phosphate mine waste pile and on similar rocks in nearby Dairy Syncline, Caribou County, Southeast Idaho. US Geol. Surv., Open File Report, 01-025, 47 pp. Human Health Fact Sheet, October 2001. Argonne National Labs. Hutchison, F. and Wai, C.M., 1979. Cadmium, lead, and zinc in reclaimed phosphate mine waste dumps in Idaho. Bull. Environ. Contam. Toxicol., 23: 377-380. Jones, J.B. Jr., 1989. Soil Testing and Plant Analysis Laboratory Manual. Benton Laboratories, Athens, GA, 78 pp. Kabata-Pendias, A., 2001. Trace Elements in Soils and Plants. CRC Press, Boca Raton, FL, 413 pp. Lamb, J.F.S., Barnes, D.K. and Henjum, K.I., 1999. Gain from two cycles of divergent selection for root morphology in alfalfa. Crop Sci., 39: 1026-1035. Lamb, J.ES., Samac, D.A., Barnes, D.K. and Henjum, K.I., 2000. Increased herbage yield in alfalfa associated with selection for fibrous and lateral roots. Crop Sci., 40: 693-699. Lamothe, P.J., Meier, A.L. and Wilson, S., 2000. The determination of forty-four elements in aqueous samples by inductively coupled plasma-mass spectrometry. US Geol. Surv., Open File Report, 00-514, 14 pp. Lemly, A. D., 1999. Selenium transport and bioaccumulation in aquatic ecosystems: a proposal for water quality based on hydrological units. Ecotoxicol. Environ. Safety, 42: 150-156. Lindsay, W.L. 1979. Chemical Equilibria. John Wiley and Sons, Inc., New York, 449 pp. Losi, M.E., Amrhein, C. and Frankenberger, W.T. Jr., 1994. Bioremediation of chromate-contaminated groundwater by reduction and precipitation in surface soils. J. Environ. Qual., 23:1141-1150. MacNicol, R.D. and Beckett, P.H.T., 1985. Critical tissue concentrations of potentially toxic elements. Plant Soil, 85: 107-129.
5 54
C.L. Mackowiak et al.
Mahan, D.C. and Moxon, A.L., 1984. Effect of inorganic selenium supplementation on selenosis in post-weaning swine. J. Anim. Sci., 58:1216 pp. Marschner, H., 1995. Mineral Nutrition of Higher Plants. Academic Press, New York, 889 pp. Mayland, H.E, 1994. Selenium in plant and animal nutrition. In: W.T. Frankenberger and S. Benson (eds.), Selenium in the Environment. Dekker, New York, pp. 29-45. Mayland, H.E, James, L.E, Panter, K.E. and Sonderegger, J.L., 1989. Selenium in seleniferous environments. In: L.W. Jacobs (ed.), Selenium in Agriculture and the Environment. Soil Sci. Soc. Am. Special Publication No. 23, Madison, WI, pp. 15-50. McGrath, S.P., 1998. Phytoextraction for soil remediation. In: R.R. Brooks (ed.), Plants that Hyperaccumulate Heavy Metals. CAB International, New York, pp. 261-287. Miller, W.T. and Williams. K.T., 1940. Minimum lethal dose of selenium as sodium selenite for horses, mules, cattle and swine. J. Agric. Res., 60: 153-173. National Research Council, 1980a. Recommended Dietary Allowances, Food and Nutrition Board, Committee on Dietary Allowances. National Academy of Sciences, Washington, DC, pp. 162-164. National Research Council, 1980b. Mineral tolerance of domestic animals. Subcommittee on mineral toxicity in animals. National Academy of Sciences, Washington, DC, pp. 400-401. National Research Council, 1983. Selenium in Nutrition (revised edition). National Academy Press, Washington, DC, pp. 107-109. Olson, O.E., 1978. Selenium in plants as a cause of livestock poisoning. In: R. E Keeler et al. (eds.), Effects of Poisonous Plants on Livestock. Academic Press, New York, pp. 121-134. Olson, O.E. and Embry, L.B., 1973. Chronic selenite toxicity in cattle. Proc. S. Dakota Acad. Sci., 52: 50-58. Osweiler, G.D., Carson, T.L., Buck, W.B. and Van Gelder, G.A., 1985. Clinical and Diagnostic Veterinary Toxicology (3rd edn.). Kendai/Hunt Publishing Co., Debuque, IA, p. 13. Peak, D. and Sparks, D.L., 2002. Mechanisms of selenate adsorption on iron oxides and hydroxides, Environ. Sci. Technol., 36: 1460-1466. Pilon-Smits, E.A.H., de Souza, M.P., Hong, G., Amini, A., Bravo, R.C., Payabyab, S.T. and Terry, N., 1999. Selenium volatilization and accumulation by twenty aquatic plant species. J. Environ. Qual., 28:1011-1018. Piper, D.Z., Skorupa, J.P., Presser, T.S., Hardy, M.A., Hamilton, S.J., Huebner, M., and Gulbrandsen, R.A., 2000. The Phosphoria Formation at the Hot Springs Mine in southeast Idaho: a source of selenium and other trace elements to surface water, ground water, vegetation, and biota. US Geol. Surv., Open File Report, 00-050, 73 pp. Puls, R., 1994. Mineral Levels in Animal Health. Diagnostic data. Sherpa International, British Columbia, Canada, 356 pp. Punshon, T. and Dickinson, N., 1999. Heavy metal resistance and accumulation characteristics in willows. Int. J. Phytoremed., 1: 361-385. Roberson, M.J., 1999. Removal of selenate from irrigation drainage water using zero-valent iron. Unpublished PhD thesis, Univ. California, Riverside, 128 pp. Rosenfeld, I. and Beath, O.A., 1964. Selenium: Geobotany, Biochemistry, Toxicity, and Nutrition. Academic Press, New York, 411 pp. Schnoor, J.L., Licht, L.L., McCutcheon, S.C., Wolfe, N.L., and Carreira, L.H., 1995. Phytoremediation of organic and nutrient contaminants. Environ. Sci. Technol., 29:318A-323A. Schrauzer, G.N., 2000. Selenomethionine: A review of its nutritional significance, metabolism and toxicity. J. Nutr., 130:1653-1656.
Uptake of selenium and other contaminant elements into plants
555
Stark, J.M. and Redente, E.E, 1990. Plant uptake and cycling of trace elements on retorted oil shale disposal piles. J. Environ. Qual., 19:495-501. Stowe, H.D. and Herdt, T.H., 1992. Clinical assessment of selenium status of livestock. J. Anim. Sci., 70: 3928-3933. Su, C. and Suarez, D.L., 2000. Selenate and selenite sorption on iron oxides: an infrared and electrophoretic study. Soil Sci. Soc. Am. J., 64:101-111. Suttle, N.E, 1991. The interaction between copper, molybdenum, and sulfur in ruminant nutrition. Annu. Rev. Nutr., 11: 121-140. Swan, D.A., Creeper, J.H., White, C.L., Ridings, M., Smith, G.M. and Costa, N.D., 1998. Molybdenum poisoning in feedlot cattle. Aust. Vet. J., 76: 345-349. Uren, N.C., 1992. Forms, reactions, and availability of nickel in soils. Adv. Agron., 48:141-203. Vance, G.E, 2000. Problems associated with selenium leaching from waste shale, The 2000 National Meeting of the American Society of Surface Mining and Reclamation, Tampa, Florida, June 11-15, pp. 71-82. Ward, G. M., 1978. Molybdenum toxicity and hypocuprosis in ruminants: a review. J. Anim. Sci., 46: 1078-1085.
This Page Intentionally Left Blank
PART V.
M O D E L I N G STUDIES
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
559
Chapter 20
R E V I E W OF W O R L D SEDIMENTARY PHOSPHATE DEPOSITS AND
OCCURRENCES G.J. ORRIS and C.B. CHERNOFF
INTRODUCTION Sedimentary phosphate deposits and occurrences are the most common type of phosphate deposit and are known to occur in over 80 countries (Table 20-I). More than 75% of economic phosphate production comes from these deposits and most of the production is used for fertilizers. Classification of these deposits remains elusive despite a wealth of descriptive literature, in part because many deposits have complex histories and features of more than one classification. There have been few previous global-scale compilations of phosphate deposits that have brought together diverse data on location, host rocks, style of mineralization, and other information. A comprehensive data set, such as the one presented here (see CD Appendix A) offers new opportunities to discover spatial relationships between phosphate occurrences and contributing geoenvironmental factors and paleogeographic controls.
DATA
Acquisition Data on phosphate mines, deposits, and occurrences were collected from a variety of sources including databases of the US Geological Survey and other agencies; published literature, unpublished data, and compilations; and company, institute, and government websites. Data on phosphate occurrences were included in the compilation along with mines and deposits of known economic potential in order to better reflect the geologic spectrum of phosphate concentration. Many of the phosphate occurrences have not been well studied and their economic potential is poorly known. The original compilation of over 1600 deposits and occurrences included all types of phosphate deposits and the raw data are given by Chernoff and Orris (2002) and Orris and Chernoff (2002). In this chapter and CD Appendix A, data on the sedimentary phosphate deposits, including guano and guano-related deposits, have been culled from the original data compilation, new data added, and the data edited for consistency.
G.J. Orris and C.B. Chernoff
560 TABLE 20-I Countries with sedimentary phosphate deposits and occurrences Country
Prod I
Country
Prod
Country
Prod
Afghanistan Albania Algeria Angola Argentina Australia Austria Belgium Belarus Benin Bhutan Bolivia Brazil Bulgaria Burkina Faso Cambodia Canada Central African Republic Chile China Colombia Congo (Zaire) Cuba Denmark Ecuador Egypt Estonia Ethiopia
X m X p X p X X p -
Finland France Gabon Germany Greece Guinea Bissau Hungary India Indonesia Iran lraq Ireland Israel Italy Jordan Kazakhstan Korea (North)
X p p X X X p X X X -
New Zealand Niger Norway Pakistan Panama Peru Philippines Poland Portugal Romania Russia Saudi Arabia Senegal South Africa Spain Sweden Syria
p m
p X X p X X -
Lebanon Luxembourg Mali Malta Mauritania Mexico Mongolia Morocco Namibia Nepal Netherlands
X X X -
Tanzania Thailand Togo Tunisia Turkey Ukraine United Kingdom United States Venezuela Vietnam Yugoslavia
X m p
p X X X X
X X X X p p p X X X p
~Prod = production; X - current production; m = minor or intermittent production; p = past production.
C o m p i l a t i o n o f the s e d i m e n t a r y p h o s p h a t e d e p o s i t data set p r e s e n t e d here was d o n e as t h o r o u g h l y as possible, and yet it c o n t a i n s s i g n i f i c a n t i n a d e q u a c i e s . At the m o s t general level, we k n o w that the dataset is not c o m p l e t e . We were u n a b l e to locate m a n y p u b l i c a tions w i t h data on deposits. We k n o w that there are m a n y m o r e p h o s p h a t e deposits and o c c u r r e n c e s in Iran than we were able to d o c u m e n t , and this is p r o b a b l y true for m a n y other areas. In addition, our access to the g e o l o g i c literature in a l a n g u a g e that we c o u l d read was limited for countries such as China. It is p o s s i b l e that we have m i s s e d i m p o r t a n t
Review of world sedimentary phosphate deposits and occurrences
561
Chinese deposits in our compilation. For many countries, we were able to obtain only district or regional level information. Many of the deposits are aerially extensive and a single point and generalized description is not fully representative of the deposit or as useful as a more localized description. Even within the United States, we were only able to list Tennessee phosphate deposits as "white rock", "brown rock", and "blue rock", without individual site data.
Description
Our compilation consists of over 1000 sedimentary phosphate deposits and over 220 guano and guano-related deposits (CD Appendix A). The compilation includes 19 data fields: (a) Deposit No.; (b) Country/Ocean; (c) State/Principal Administrative Area; (d) 3rd Order Area Name; (e) Basin/Formation/Region/Deposit; (I") Deposit or Site Name; (g) Commodities; (h) Deposit Type; (i) Type of Phosphorite; (j) Age of Mineralization; (k) Host/Associated Lithology; (1) Host Formation; (m) Geologic Comments or Deposit Description; (n) Latitude; (o) Longitude; (p) Production Status; (q) Resources; (r) General Comments; and (s) References. The data are sorted in the following order: country, state/principal administrative area, third order administrative area, basin/formation/region(area)/deposit, and deposit/site name. Guano and guano-derived deposits are listed at the end of CD Appendix A and sorted in the same order. Seamount deposits are listed alphabetically preceding the guano deposits. Although more complete than any other sedimentary phosphate compilation the authors are aware of, this phosphate database has all of the standard data compilation problems and compromises. Perhaps the most frustrating problem is that of names. It is commonly unclear in the literature if a given name refers to a mine or deposit, a nearby location or administrative area, a geologic formation, or a company involved in the exploration or development of a site. Frequently, multiple names may be used to refer to a single deposit or mine. For large deposits or mineralized areas, a single name might be used to refer to diverse sites, to deposits within these areas, or to the area as a whole. We tried to deal with the naming difficulties as consistently and completely as possible, especially within individual countries; however, these difficulties, as well as duplicate records, are still likely to be present in the database. Most of the data fields included in the compilation are largely self-explanatory. However, a few notational and other details will be useful for the reader to know. Thirdorder administrative areas, such as counties in the United States, are listed only if specified in the original source materials, or if easily determined from other sources we considered reliable. The information in the next two fields, Basin/Fmt/Region (area)/Deposit and Deposit or Site Name, is meant to identify the area, deposit, district, mine, and (or) occurrence to which the geologic, economic, and location data apply. While we tried to avoid duplicates, for some areas we report information for a district or area, as well as for specific deposits or mines within those larger areas. The exact level of information in these two
562
G.J Orris and C.B. Chernoff
fields may vary among countries; but we used these fields consistently within a given country. Finally, data for one or more data fields are missing or inconsistent in most of the records. Location data might be missing for one record; host rock information for a different record. There is also some inconsistency in level of information. The host rock for some deposits might be "carbonates"; for others "black shaly limestone". This can make finding patterns in the data challenging.
SEDIMENTARY PHOSPHATE DEPOSITS Sedimentary phosphate deposits are known on every continent except Antarctica (which is likely due to insufficient sampling there) and provide as much as 80% of the world's phosphate (Bartels and Gurr, 1994). They range in age from Early Proterozoic through Holocene, although some geologic ages were more phosphogenic than others. Major phosphogenic systems include, but are not limited to, Late Precambrian and Cambrian of central and southeast Asia; Permian of North America; Jurassic and Early Cretaceous of eastern Europe; Late Cretaceous through Eocene of North Africa; and Miocene of North America (Cook, 1976). Each system covers several thousands of square kilometers and is inherently complex with multiple types of sedimentary phosphate deposits. Riggs (1980) suggested that the depositional environments of sedimentary phosphate deposits could be grouped into three categories: intra-plate marine deposits, deposits associated with active-plate boundaries, and insular deposits. Intra-plate marine depositional environments include: (a) plateaus and seamounts; (b) shelves, platforms, and geosynclinal basins that include both marginal and intracratonic sedimentary troughs; and (c) unspecified environments in which destructional sedimentary processes take place. Active-plate boundary depositional environments include (d) marine environments and (e) lacustrine environments. Insular environments in this classification scheme include (f) guano and (g) secondary environments with replacement and residual deposits. More recently, Hein et al. (1993) described four geographic-tectonic settings in the present-day ocean basins in which phosphorites occur: (a) continental shelves and slopes, primarily off the west coast of continents; (b) submarine plateaus, ridges, and banks; (c) islands, atolls, and within atoll lagoons; and (d) mid-plate seamounts. The great majority of economically significant sedimentary phosphate deposits are marine deposits that formed in active margin basin, epicontinental sea, or shelf environments. Insular phosphate deposits resulting from the chemical and (or) mechanical weathering of marine phosphatic sediments and guano have historically been significant sources of phosphate ore (Cook, 1976). A few nonmarine occurrences are known, but are not typically of economic importance; these include concretionary phosphatic lenses in sedimentary rocks of the Triassic Cachueta Basin, Argentina (Leanza et al., 1989) and phosphatic sediments of the Eocene Green River Formation, western US (Love, 1964).
Review of world sedimentary phosphate deposits and occurrences
563
Marine sedimentary phosphate deposits Marine sedimentary deposits are composed of phosphate in the form of pellets, nodules, grains, cement, skeletal material, and (or) coprolites. The economically significant deposits are laterally extensive, and are associated with low to intermediate levels of tectonism. The following groupings of sedimentary phosphate deposits are meant to provide a framework for discussion of the deposits and not to necessarily support any particular classification scheme.
Active margin basin and epicontinental sea deposits For many economically significant phosphate deposits, phosphate sedimentation was localized by marine upwelling in basins along active continental margins, epicontinental seaways, or intracratonic sag basins. Phosphate deposits of this type tend to be pelletal and extensive, commonly extending over hundreds to thousands of square kilometers with a high P205 content throughout (Cook, 1976). These laterally extensive deposits are associated with organic-rich argillaceous sediments, chert, dolostone, and volcanic materials (Mosier, 1986a). Examples include the Early Cambrian Chulak Tau Suite of the Karatau district in Kazakhstan, the Permian Phosphoria Formation of the Western Phosphate Field in the United States, and the Late Cretaceous to Tertiary deposits of North Africa.
Shelf and platform deposits By comparison, phosphate mineralization on shelves and platforms of passive continental margins tend to be less extensive, occurring as enriched areas, tens to hundreds of square kilometers in size, or as nodular deposits distributed over large areas. The deposits are typically hosted by sandstone and carbonate; other common hosts are glauconite, diatomite, and coquina. Deposition tends to be focused on the flanks of structural highs (Mosier, 1986b). The Miocene phosphate deposits of Florida are commonly used as an example of this environment. Sedimentation occurred in basins off the flanks of structural highs and platforms and consequently, the distribution of phosphate mineralization was not continuous. Phosphate deposits of the Miocene Guines Pipian District of Cuba and those of the Maastrichtian Gramame Formation of Brazil also have features related to this style of phosphate deposition.
Other marine deposits Other marine sedimentary phosphate deposits include modern sea-floor deposits and deposits that are formed from the replacement of carbonate and other rock types on
564
G.J. Orris and C.B. Chernoff
seamounts and submarine plateaus and ridges. Modern sea-floor deposits, typically off the west coast of continents, formed from early diagenetic processes near the seawatersediment interface beneath zones of coastal upwelling (e.g. Glenn and Arthur, 1988). These deposits range in age from Middle Tertiary to Holocene (Harben and Kuzvart, 1996). Deposits at the Peru-Chile margin and off the coast of South Africa are representative of this type of mineralization (Loebner et al., 1987; Piper et al., 1988; Birch, 1990). Phosphorites also occur extensively on submarine plateaus, ridges, and banks. In this environment, phosphate deposits form from the cementation and replacement of carbonate in an organic matter-rich environment and reworking of the deposits is common (Hein et al., 1993). Representative phosphate deposits include those of the Blake Plateau off the southeastern US and the Chatham Rise off New Zealand (Cullen, 1989; Riggs, 1989). Phosphate mineralization on seamounts typically results from the replacement of carbonate in relatively organic-poor environments, although some seamount deposits are submerged guano deposits and hydrothermal deposits are rarely found (Hein et al., 1993). Seamount phosphorites are found at depths of up to several thousand meters at latitudes that range from 35~ to 42~ The compilation in CD Appendix A includes 20 Pacific Ocean seamounts with phosphate mineralization, although phosphate has been recovered from hundreds of seamounts in the Pacific (Hein et al., 1993). Our compilation does not include seamount phosphates or other submarine deposits in the Indian Ocean, and only one example of a deposit in the Atlantic Ocean. Similar to seamount deposits, some continental shelf and slope phosphates form in organic-poor environments (Benninger and Hein, 2000). Phosphorite deposition off the east coast of Australia falls into this class. Phosphate is concentrated during iron-redox reactions where sorbed P is acquired and released during oxidation and reduction of Fe oxyhydroxide, respectively.
Insular phosphate deposits Weathering-related residual and infiltration deposits Chemical and mechanical weathering of phosphatized rock is essential for development of economic grades of P205 for some sedimentary phosphate deposits. In these deposits, there are two variations of chemical weathering (Cook, 1976). In the first type, non-phosphates are leached from the host rock leaving an enriched residual phosphate behind. This process formed the Tennessee "brown rock" deposits in the United States. Other phosphate deposits for which residual enrichment is important include the extensive deposits of the Ganntour and Oulad Abdoun Plateaus in Morocco and those of the Thies and Taiba Plateaus in Senegal. In the second form of chemical weathering, the phosphate is leached from the rock, taken into solution by carbonic- or uric-acid containing ground or surface waters, and reprecipitated. The phosphate may be reprecipitated in the weathered rock or soils, or into karst and other cavities in underlying carbonate rocks. The Tennessee "white rock"
Review of world sedimentary phosphate deposits and occurrences
565
deposits are an example of this process in the United States. Phosphate deposits of the Battambang and Kampot provinces of Cambodia were in part formed by the infiltration of P-bearing waters into fissures and cavities. This infiltration process also had a role in the formation of some island and cave guano deposits, such as the Nauru Island deposits and the guano deposit at Wichianburi, Thailand. In addition to chemical weathering, mechanical weathering or reworking of phosphatized rock can enrich a deposit. In this process, phosphatic gravel from older weathered deposits is reworked and redeposited after a change in sea level or structural/tectonic change. The most cited example of mechanical weathering is the Pliocene "land pebble" deposits of the Bone Valley Formation, Florida (Cook, 1976; Harben and Kuzvart, 1996). Within most insular phosphate deposits, mechanical weathering takes place at a much smaller scale.
Guano and guano-related deposit
The least complete part of the data set compiled here is the guano deposits. Many of these deposits are exploited locally and, because of their small size and lack of large-scale economic viability, are ignored in many resource compilations. Historically, guano deposits were the main source of phosphatic fertilizer. Although, with time, the importance of guano as a fertilizer continues to decrease, guano remains an important source of phosphatic fertilizer in some parts of the world. These deposits accumulated from the excrement of birds or bats over long periods of time and are typically found on islands or in caves, respectively. The major constituents of excreta include uric acid, calcium, phosphate, and potassium. The excreta can infiltrate the limestone substrate and form phosphate minerals such as brushite or monetite (Bartels and Gurr, 1994). These deposits can be up to tens of meters thick. The location and presence of cave guano is primarily a function of geology (does the rock-type permit cave formation?) and the duration of the bat habitation. The distribution of island guano, however, is more complex. Island deposits are dependent on climatic and oceanographic conditions that bridge the fine balance between supplying enough nutrientrich water to support bird life, but remaining dry enough to preserve the guano (Harben and Kuzvart, 1996). There are only a handful of guano deposits that do not fit one of the above profiles. For example, Krauss et al. (1984), among others, report guano at Minjingu, Tanzania. Although classed as guano, the main mineralization at that location is massive and bedded phosphorite interbedded with lacustrine clays (Notholt et al., 1989). In this example, the guano does not appear to be either bat guano from caves or sea-bird guano from an open-ocean island.
FORMATION AND DISTRIBUTION OF DEPOSITS Major phosphate deposits do not form under typical marine conditions. Their formation requires that favorable paleoceanographic, paleogeographic, paleoclimatic, and other
566
G.J. Orris and C.B. Chernoff
factors coincide (Sheldon, 1987). Most develop when there is an influx of nutrient-rich water into a shallow marine environment followed by the development of prolific biota. Phosphorus may be taken up by these marine fauna, which can concentrate phosphate in their skeletons and tissues. Upon death of the organisms, their remains sink to the ocean floor. Phosphate is leached from some of the remains and concentrated in the pore waters of sediments. Diagenesis and sediment reworking can then lead to development of a phosphate deposit. The conditions under which this generalized process takes place control the size, style, and distribution of phosphate deposits through time. The most important condition in the formation of all major phosphate deposits is the paucity of input of diluting materials, such as fluvial input and formation of carbonates. Conditions necessary for the deposition of phosphate deposits require that there be both a source of phosphate and chemical and regional structural conditions favorable to trapping the phosphate. A slow sedimentation rate, permissive geography and climate, stable basin conditions, and overall depositional system stability permit significant amounts of phosphate to accumulate. Once the phosphate has accumulated, mechanisms for the enrichment of the deposit may need to be present- these include diagenesis, leaching, residual enrichment, and (or) reworking. Paleogeography is important to localization of phosphate deposits. Many ancient phosphorites formed on continental margins in or near the tropics. Riggs (1980) suggested that tectonic setting played a major role in phosphogenesis. Plate tectonic processes reconfigure the oceans and continents. The position of the continents is key to climate, current patterns, sedimentation, and the size, shape, and orientation of basins and craton margins (Riggs, 1980; Sheldon, 1987). Marine sedimentary basins on continental margins with connection to the open ocean are needed for phosphogenesis. Passive continental margins and continental margin arcs that face west to the open ocean are more likely to have phosphate deposition than active margins bordered by marginal basins and island arcs (Mitchell and Garson, 1981). Regional structures such as basins and continental shelves may provide depositional sites for phosphate. In addition, local structural highs may provide favorable phosphogenic settings. Thc ultimate source of phosphate to the ocean is from subaerial chemical weathering of rocks; the phosphate is then transported to the ocean by rivers (Sheldon, 1987). Other inorganic sources of phosphate such as volcanic activity are generally insignificant by comparison. The ocean is a phosphate sink and much of the dissolved phosphorus is stored in deep-ocean waters (Sheldon, 1987). Phosphate can be concentrated by currents and upwelling, locally by density plumes from rivers, and, most importantly, by biota in the photic zone. Upwelling is cited as a key concentrating mechanism for impacting phosphate deposition in many deposits. Most phosphate deposition is restricted to areas where cold ocean water rich in nutrients from several hundreds of meters water depth upwells into the surface zone and there supports prolific organic production. This type of upwelling is in response to regional seawardmoving water. Upwelling can also occur on a more local scale. Dynamic upwelling can occur when a current moves over a topographic high (Cook, 1976).
Review of world sedimentary phosphate deposits and occurrences
567
High biologic productivity is important, although not essential, to phosphate deposition. Some phosphate, such as that found on the eastern Australian shelf and on most seamounts, formed in areas of only moderate productivity (Hein et al., 1993; Benninger and Hein, 2000). It should be noted that phosphate deposits rarely form at the center of intense upwelling zones where the bottom waters are anoxic (Parrish and Barron, 1986), but are precipitated at the margins under suboxic conditions. Marine transgressions can be important for formation of major phosphate deposits (Mitchell and Garson, 1981; Cook et al., 1990; Herring, 1995). It is thought that a rising sea level leads to higher organic activity during flooding of the shelf. In addition, a marine
Fig. 20-1. Paleogeographic distribution of Wuchiapingian (early Late Permian) phosphate deposits and organic-rich sedimentary rocks. Organic-rich sedimentary rock distribution from a compilation in Chernoff (2002). Paleogeographic map after Scotese and Summerhayes (1986) and Ross and Scotese ( 1993).
568
G.J Orris and C.B. Chernoff
transgression may raise the carbonate compensation depth in the oceans, which in turn would aid phosphate formation in shallow water by removing the deep-sea phosphate sink (Mitchell and Garson, 1981). Also, fluctuating sea levels could lead to reworking and enrichment of phosphatic sediments. The major Miocene phosphogenic episode occurred during the middle part of a marine transgression (Cook et al., 1990) and evidence exists for a major marine transgression preceding deposition of other phosphate deposits (Mitchell and Garson, 1981). These factors and others play a role in the distribution of phosphate deposits. With minor exceptions, phosphate deposits tend to occur along continental margins in middle to low latitudes, and in areas of upwelling (Figs. 20-1-20-4). It has been shown that organic-rich
Fig. 20-2. Paleogeographic distribution of Maastrichtian (Late Cretaceous) phosphate deposits and organic-rich sedimentary rocks. Organic-rich sedimentary rock distribution from a compilation in Chernoff (2002). Paleogeographic map after Scotese and Summerhayes (1986) and Ross and Scotese (1993).
Review of world sedimentary phosphate deposits and occurrences
569
Fig. 20-3. Paleogeographic distribution of Lutetian (middle Eocene) phosphate deposits and organic-rich sedimentary rocks. Organic-rich sedimentary rock distribution from a compilation in Chernoff (2002). Paleogeographic map after Scotese and Summerhayes (1986) and Ross and Scotese (1993). sedimentary rocks are commonly deposited under upwelling conditions (Parrish, 1982; Chernoff, 2002) and there should therefore be a correlation in the distribution of the phosphate deposits and organic-rich rocks. Figures 20-1 and 20-2 show the distribution of phosphate deposits of early Late Permian (255 Ma) and Late Cretaceous (Maastrichtian, 69 Ma) ages from our compilation projected into paleogeographic maps. Each figure also shows the location of organic-rich sedimentary rocks deposited during the same time interval. It has been demonstrated that organic-rich sedimentary rocks are commonly deposited in areas of marine upwelling (Parrish, 1982; Chernoff, 2002). Data in Figs 20-1 and 20-2 show good spatial correlation for a number of the phosphate deposits with organic-rich sedimentary rocks. The spatial correlation is most pronounced for the basinal-type deposits of these time periods, an
570
G.J. Orris and C.B. Chernoff
Fig. 20-4. Paleogeographic distribution of Serravallian (middle Miocene) phosphate deposits and organic-rich sedimentary rocks. Organic-rich sedimentary rock distribution from a compilation in Chernoff (2002). Paleogeographic map atter Scotese and Summerhayes (1986) and Ross and Scotese (1993).
observation that fits well with the previously discussed character of these deposits. In contrast, Figs 20-3 and 20-4 show paleogeographic reconstructions for the Middle Eocene (Lutetian, 50.3 Ma) and Middle Miocene (Serravalian, 14 Ma), which demonstrate little spatial correlation between the phosphate deposits and organic-rich sedimentary rocks. It is to be expected that the shelf-type phosphate deposits that dominate those time periods should not necessarily show syndeposition with organic-rich sedimentary rocks. In addition to helping support interpretations of the phosphate depositional environment, the distribution of organic-rich sedimentary rocks combined with other data related to the occurrence of phosphate deposits may give clues to the location of undiscovered phosphate deposits.
Review of world sedimentary phosphate deposits and occurrences
571
CONCLUSIONS The compilation of sedimentary phosphate deposits presented here is the most complete compilation available. With the exception of the deposit type field, we chose to present the information as it was presented in the source with no interpretation or interpolation. We believe that this allows the widest use of the data for further work. These data combined with paleogeographic, paleoclimatic, and other knowledge about the genesis and occurrence of phosphate deposits offer the potential to help identify potential exploration targets in areas without known deposits and to further refine our understanding of the environments within which phosphatic sedimentary rocks form.
ACKNOWLEDGMENTS We would like to thank James R. Hein and Stephen M. Jasinski, USGS, and Peter W. Harben, Industrial Minerals Consultants, for reviews and comments.
REFERENCES Bartels, J.J. and Gurr, T.M., 1994. Phosphate rock. In: D.D. Carr (ed.), Industrial minerals and rocks (6th edn.). Society for Mining, Metallurgy, and Exploration, Littleton, CO, pp. 751-764. Benninger, L.M. and Hein, J.R., 2000. Diagenetic evolution of seamount phosphorite. In: C.R. Glenn, L. Pr6v6t-Lucas and J. Lucas (eds.), Marine Authigenesis: From Global to Microbial. SEPM Special Publication 66, Tulsa, OK, pp. 245-256. Birch, G., 1990. Phosphorite deposits on the South African continental margin and coastal terrace. In: W.C. Burnett and S.R. Riggs (eds.), Phosphate Deposits of the World, vol. 3 - Neogene to Modern Phosphorites. Cambridge University Press, Cambridge, pp. 153-158. Chemoff, C.B., 2002. Origin and redistribution of metals in sedimentary rocks. PhD dissertation, University of Arizona, Tucson, AZ, 1036 pp. Chernoff, C.B. and Orris, G.J., 2002. Data set of world phosphate mines, deposits, and occurrences; Part A, Geologic data. US Geological Survey, Open-File Report, OF 02-0156A, 352 pp. Cook, P.J., 1976. Sedimentary phosphate deposits. In: K.H. Wolf (ed.), Handbook of Strata-Bound and Stratiform Ore Deposits; II Regional Studies and Specific Deposits, vol. 7, Au, U, Fe, Mn, Hg, Sb, W, and P Deposits. Elsevier Publishing Co., New York, NY, pp. 505-535. Cook, P.J., Shergold, J.H., Bumett, W.C. and Rigg, S.R., 1990. Phosphorite research: a historical overview. In: A.J.G. Notholt and I. Jarvis (eds.), Phosphorite Research and Development. Geological Society of London, Special Publication, vol. 52, pp. 1-22. Cullen, D.J., 1989. The Chatham Rise Phosphorites of New Zealand. In: A.J.G. Notholt, R.P. Sheldon and D.E Davidson (eds.), Phosphate Deposits of the World, vol. 2 - Phosphate Rock Resources. Cambridge University Press, Cambridge, pp. 528-532. Glenn, C.R. and Arthur, M.A., 1988. Petrology and major element geochemistry of Peru margin phosphorites and associated diagenetic minerals: authigenesis in modem organic-rich sediment. Mar. Geol., 80:231-267.
572
G.J Orris and C.B. Chernoff
Harben, P.W. and Kuzvart, M., 1996. Phosphate rock. In: Industrial minerals- A global geology: Industrial Minerals Information Ltd., London, pp. 289-303. Hein, J.R., Yeh, H.-W., Gunn, S.H., Sliter, W.V., Benninger, L.M. and Wang, C.-H., 1993. Two major Cenozoic episodes of phosphogenesis recorded in equatorial Pacific seamount deposits. Paleoceanography, 8:293-311. Herring, J.R., 1995. Permian phosphorites: a paradox of phosphogenesis. In: P.A. Scholle, T.M. Peryt and D.S. Ulmer-Scholle (eds.), The Permian of Northern Pangea, vol. 2: Sedimentary Basins and Economic Resources. Springer-Verlag, New York, pp. 292-312. Krauss, U.H., Saam, H.G., and Schmidt, H.W., 1984. International strategic minerals inventory summary report- phosphate. US Geological Survey, Circular, 930-C, 41 pp. Leanza, H.A., Spiegelman, A.T., Hugo, C.A., Mastandrea, O.O. and Oblitas, C.J., 1989. Phanerozoic sedimentary phosphatic rocks of Argentina. In: A.J.G. Notholt, R.P. Sheldon and D.E Davidson (eds.), Phosphate Deposits of the World, vol. 2 - Phosphate Rock Resources. Cambridge University Press, Cambridge, pp. 147-158. Loebner, B.J., Piper, D.Z. and Vedder, J.G., 1987. Phosphorite deposits along the western North American continental margin. In: D.W. Scholl, A. Grantz and J.G. Vedder (eds.), Geology and resource potential of the continental margin of western North America and adjacent ocean basins, Beaufort Sea to Baja California. Circum-Pacific Council for Energy and Mineral Resources, Earth Science Series, 6, pp. 773-791. Love, J.D., 1964. Uraniferous phosphatic lake beds of Eocene age in intermontane basins of Wyoming and Utah. US Geological Survey, Professional Paper, 474-E, pp. 1-66. Mitchell, A.H.G. and Garson, M.S., 1981. Mineral Deposits and Global Tectonic Settings, (Chapter 3). Academic Press, London, pp. 87-101. Mosier, D.L., 1986a. Descriptive model of upwelling type phosphate deposits. In: D.P Cox and D.A. Singer (eds.), Mineral Deposit Models. US Geological Survey, Bulletin, 1693, pp. 234. Mosier, D.L., 1986b. Descriptive model of warm-current type phosphate deposits. In: D.P. Cox and D.A. Singer (eds.), Mineral Deposit Models. US Geological Survey, Bulletin, 1693, pp. 237. Notholt, A.J.G., Sheldon, R.P. and Davidson, D.E, 1989. Africa - introduction. In: A.J.G. Notholt, R.P Sheldon and D.E Davidson (eds.), Phosphate Deposits of the World, vol. 2 - Phosphate Rock Resources. Cambridge University Press, Cambridge, pp. 164-170. Orris, G.J. and Chernoff, C.B., 2002. Data set of world phosphate mines, deposits, and occurrences; Part B, Location and mineral economic data. US Geological Survey, Open-File Report, OF 02-0156B, 328 pp. Parrish, J.T., 1982. Upwelling and petroleum source beds with reference to Paleozoic. Am. Assoc. Pet. Geol. Bull., 66: 750-774. Parrish, J.T. and Barron, E.J., 1986. Paleoclimates and economic geology. Society of Economic Paleontologists and Mineralogists, Short Course, 18, 162 pp. Piper, D.Z., Baedecker, PA., Crock, J.G., Burnett, W.C. and Loebner, B.J., 1988. Rare earth elements in the phosphatic-enriched sediment of the Peru shelf. Mar. Geol., 80: 269-285. Riggs, S.R., 1980. Tectonic model of phosphate genesis. In: R.P. Sheldon and W.C. Burnett (eds.), Fertilizer Mineral Potential in Asia and the Pacific. East-West Resource Systems Institute, Honolulu, Hawaii, pp. 15-190. Riggs, S.R., 1989. Phosphate deposits of the North Carolina coastal plain, continental shelf, and adjacent Blake Plateau, USA. In: A.J.G. Notholt, R.P. Sheldon and D.E Davidson (eds.), Phosphate Deposits of the World, vol. 2 - Phosphate Rock Resources. Cambridge University Press, Cambridge, pp. 42-52.
Review of world sedimentary phosphate deposits and occurrences
573
Ross, M.I. and Scotese, C.R., 1993. Paleoclimate: a parametric paleoclimate modeling program for the Macintosh. Earth in Motion Technologies, Houston, TX. Scotese, C.R. and Summerhayes, C.E, 1986. A computer model of paleoclimate to predict upwelling in the Mesozoic and Cenozoic. Geobyte, 1: 28-42. Sheldon, R.P., 1987. Association of phosphatic and siliceous marine sedimentary deposits. In: J.R. Hein (ed.), Siliceous Sedimentary Rock-Hosted Ores and Petroleum. Van Nostrand Reinhold Co., New York, NY, pp. 58-80.
This Page Intentionally Left Blank
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
575
Chapter 21
WESTERN PHOSPHATE F I E L D - DEPOSITIONAL AND ECONOMIC DEPOSIT MODELS
P.R. MOYLE and D.Z. PIPER
ABSTRACT The Western Phosphate Field (WPF), composed of Permian marine sedimentary strata that cover over 300,000 km 2 in the middle Rocky Mountains of Idaho, Montana, Utah, and Wyoming in the United States, contains vast resources of phosphate mined for fertilizer and a range of other industrial applications. The richest deposits of phosphate in the WPF occur in the Meade Peak Phosphatic Shale Member of the Phosphoria Formation in southeast Idaho. Phosphate is an essential and even limiting nutrient of algal production, which occurs at the bottom of the marine food web in the oceanic photic zone. The high concentrations of phosphate and trace elements in the Phosphoria Formation reflect a low accumulation rate of diluting phases, such as terrigenous siliciclastic debris and carbonate, rather than an unusually high level of primary productivity at the time of deposition. Indeed, the mean rate of accumulation of PO43- required a continuous flux of PO43- into the basin and the photic zone of the water column, but only at a moderate level. This flux was maintained by upwelling of nutrient-rich seawater, imported at depth from the open ocean. Although only a fraction of the organic matter that hosted the PO43- and other nutrients (NOy-, Cd, Cu, Mo, Ni, and Zn) actually escaped oxidation in the water column, their rate of accumulation on the sea floor defined the basin hydrography. Rates of accumulation of Cr, U, V, and rare-earth elements by precipitation and adsorption reactions identify redox conditions in the bottom water as having been denitrifying, maintained by a balance between the rate of oxidation of organic matter settling through the water column and the flux of open-ocean seawater at depth. Atmospheric mixing maintained oxygen respiration in the uppermost several tens of meters of the water column. This hydrography and seawater chemistry are present in several sedimentary environments in the ocean today. In the WPF, there is an estimated surface mineable reserve base and subeconomic resource of 7.6 billion mt, at an average grade of 24% P205; a subeconomic undergroundmineable resource of 17 billion mt, at a grade of 28%; and 507 billion mt of subresourcegrade phosphatic material that underlie the WPF at a depth greater than 305 m. The relationship between phosphate-ore specifications and weathering suggests that significant
576
PR. Moyle and D.Z. Piper
changes in processing, with associated cost increases, will be required to extend recovery of ore below the relatively strongly weathered zone near the surface. Four open pit mines currently extract phosphate from two moderately to steeply dipping ore zones that typically contain between 20% and 35% P205. Although the shales are enriched in trace elements, especially As, Cd, Cr, Cu, Mo, Se, U, V, Zn, and rare-earth elements, the relative concentration of organic carbon and selected major element oxides determines the suitability of phosphate-rich rock for feed to processing plants and its other applications. Selected specifications from the four operating mines include the following: minimum P205 of 18-20% and average of 26-27%; maximum A1203 of 1.6-5.0%; maximum MgO of 0.3-0.6%; a CaO/P205 ratio of 1.5-1.6; and total carbon content of 4%-5%. Weathering to a depth of as much as 100 m significantly enhances ore quality by decreasing the proportions of calcite, dolomite, and organic matter relative to carbonate fluorapatite, the primary ore mineral.
INTRODUCTION The Permian Phosphoria and Park City Formations (Fig. 2 l-l), and related units such as the Shedhorn Sandstone, comprise a sequence of marine sedimentary strata composed primarily of carbonaceous and phosphatic mudstone, siltstone, phosphorite, dark Ca and Mg carbonates, black shales, and chert (McKelvey et al., 1959). These phosphate-rich deposits comprise the Western Phosphate Field (WPF) in the northwest United States (Fig. 21-2). Deposited about 250 million years ago (Wardlaw and Collinson, 1986), these strata presently occur over a 300,000km 2 area (Sheldon, 1989) of the middle Rocky Mountains. Deposition below an upwelling, nutrient-rich environment (Mosier, 1986; Piper, 2001) occurred over a period of about 10 My along the western margin of the North American craton (see Fig. 1-1 in Hein, Chapter 1). The aim of this study is to summarize (a) the seawater chemistry and basin hydrography that favored deposition and accumulation of the phosphate-rich sediments, and (b) the current physical and chemical parameters that make the phosphate deposits of the WPF so economically valuable.
HYDROGRAPHY OF THE PHOSPHORIA SEA- A DEPOSITIONAL MODEL Many geologists who have examined the WPF over the past 90 years have concluded that it was deposited under a very special set of conditions, for example, that its deposition was inextricably linked to evaporite basins immediately to the east, hydrothermal activity mostly to the west, catastrophic kills of biota in the water column within the basin, and even to the accumulation of cosmic debris (see Hein et al., Chapter 2 for a review). However, interelement relationships and the rates of accumulation of the different components in the Meade Peak Member suggest otherwise. Certainly, all properties of the basin in which the deposit accumulated are not present in any single modern marine basin. The size and high grade of the ore are not present at all, in any ocean-margin basin today.
Western Phosphate Field: depositional and economic deposit models
577
Fig. 21-1. (A) Permian correlation chart for southeast Idaho and eastern Wyoming, and (B) stratigraphic column of the Meade Peak Member in southeast Idaho, modified from Maughan (1994), Wilde (2000), and C. Spinosa (personal communication, 2000). Dolostone is the dominant carbonate in the Meade Peak Member in southeast Idaho and western Wyoming. See Hein (Chapter 1) for discussion of time scale. Nonetheless, the hydrography of the Phosphoria sea, chemistry of the water-column, level of primary productivity in the photic zone, and rate of deposition of the phosphate on the sea floor are amply present in several different areas of the modern ocean. Deciphering these different properties of the Phosphoria sea during deposition of the Meade Peak Member, the most phosphate-rich and trace element-rich member of the Phosphoria Formation, can be achieved through an examination of its major and traceelement inventory in the seawater-derived fraction of the deposit. The accumulation rates of these trace elements and of francolite, rather than merely their present concentrations in the deposit, address the dynamics of the Phosphoria sea and its geochemistry. The chemical elements are hosted by siliciclastic debris - the terrestrial-derived f r a c t i o n - and by calcite, dolomite, biogenic silica (now diagenetic quartz), residual organic matter, francolite, and several sulfide m i n e r a l s - the seawater-derived fraction.
578
PR. Moyle and D.Z. Piper
Fig. 21-2. Map of the northwest United States at approximately 270-266 Ma, showing the Late Permian paleogeography, facies of the Phosphoria Formation, paleoclimate influences, and locations of sections analyzed for their chemical composition. Map base is after Maughan (1994); Phosphoria facies after Dahl et al. (1993) and Stephens and Carroll (1999); Antler Highland after Geslin (1998); numbered localities, Medrano and Piper (1995) and Piper (2001). However, we are not so concerned with the current partitioning of elements among these phases as with the original sources of the elements. Goldberg (1963) showed that the many different components that constitute marine sediments have basically two sources: (a) the terrestrial environment, which contributes dominantly siliciclastic debris and (b) seawater, from which is derived biogenic debris and inorganic precipitated and adsorbed phases. Did these two sources account for the full complement of elements that now constitute the Meade Peak Member?
Nonmarine sediment fraction We start to address this question by first identifying the nonmarine fraction of the deposit. All elements likely had a terrigenous contribution to their total sediment inventory, reflecting the long-known and well-documented overall complex composition of this fraction of sediment (Clarke, 1924; Gulbrandsen and Krier, 1980). Within the constraints of our modeling scheme, several elements apparently had solely a terrigenous source.
579
Western Phosphate Field." depositional and economic deposit models
Aluminum, commonly reported as A1203, can be used to define uniquely the concentration of this fraction in a sedimentary deposit (Fig. 21-3A; Piper and Isaacs, 1995). Even though the mineralogy of the siliciclastic fraction is quite complex (Knudsen et al., 2002; Knudsen and Gunter, Chapter 7), its mean elemental composition in the Meade Peak Member, as opposed to its concentration in a single sample, can be ascertained from simple x - y plots. Several of the major-element oxides exhibit relations with A1203 (Fig. 21-3B) that are similar to their relation in the world-shale average (WSA) (Wedepohl, 1969-1978). That is, these major-element oxides plot along single trends that extrapolate to the origin and that have slopes equal to, or closely approaching, the major-element oxide/A1203 ratios of WSA. Concentrations of B, Ba, Co, Cs, Hf, Rb, Sc, Ta, and Th (Fig. 21-3C) exhibit a similar relation. They too reflect a solely terrigenous source (Piper, 2001). This fraction was clearly terrestrial in origin, as its overall composition closely approaches that of the WSA. The terrigenous material was largely wind transported (Stephens and Carroll, 1999) from the area to the north and northeast (Poole, 1964), the Wind River Uplift, and from redbeds that were exposed in the Goose Egg evaporite basin to the east (Fig. 21-2). The approximate duration of deposition of the Meade Peak Member of 4-6 Ma (B.R. Wardlaw, personal communication) and mean concentration of this sediment fraction (Piper, 2001) give an accumulation rate of 0.98 mg cm-2 yr -I. This assumes a duration of
(A)
AI20 3 (%) 60 50
._. 40 04 v 0 30 0 20 10
0
5
(B) 10
.,--.,
o~o~ n ~ detritus-calcite -o -~176176 ~ ( a n d / o r apatite) oo _o o o (~x...mixingcurve % Oo o o
6 I
o~"",o,om,~e~176 ~ o ..
~ 4t
E
o~_e_
~
o o +
o
I-
WSA
o ....~oo ~ o ~
~
WSA
~~176o 1
1
(D) 800
o
o
o
0
o
1
i
~
14
o
600
10 8
400
.,_,.
,'-
o
/
.- ^..~o ~ -0~ ~
~e~cS%,
o
ii
12 ,,,-..
AI203 (%) 10
~,-----_._.~ apatite, ~,~~alcite
j-or-~-~-mat~r o- o ~ : ~ o .o9 y , o ~-~'.~_ (c)
15
6
o
t ~
~'
4
oo
200
o
o
~ o
~ oo I~Oo~o~ ~ o~Oo ~ 1 7 6 1o7 6
i
i
o
~
i
0 Terrigenous fraction (%)
~
O
20
40
60
80
100
Terrigenous fraction (%)
Fig. 21-3. Inter-element relations between the terrigenous fraction defined as: A1203 x 6 (see text) and (A) CaO; (B) Fe203; (C) Th, and (D) La. WSA is World shale average (see text for discussion).
580
P.R. Moyle and D.Z. Piper
deposition of 5 Ma, a value that is used throughout this chapter. It should be recognized, however, that the sediment accumulation rate varied during the entire period of deposition of this member, even beyond the range of 0.78-1.18 mg cm -2 yr -l. Still, both the mean and range of the rate of accumulation of siliciclastic debris are comparable to their rates in the pelagic environment of the ocean, confirming speculation by others that the Phosphoria sea represented a sediment-starved basin. Several other trace elements have concentrations greater than can be attributed solely to the terrigenous fraction (Fig. 21-3D). They include Cd, Cr, Cu, Mo, Ni, Se, U, V, Zn, and the rare-earth elements (REE). Their enrichments above their terrigenous contribution can be ascertained through plots versus A1203 (terrigenous fraction; Piper and Isaacs, 1995), similar to the plots that identified a solely terrigenous contribution for Co, Hf, and Th. The terrigenous contribution of this second suite of trace elements is determined from the curve defined by those samples exhibiting the lowest concentrations versus the terrigenous fraction, the terrigenous maximum. These curves also extrapolate to WSA values, but equally important for this study, the plots allow the level of enrichment of each trace element to be determined on a sample-by-sample basis above its terrigenous contribution. The enrichment is represented by the displacement of a sample above the curve of the terrigenous maxima. The terrigenous contribution is minor for several of these trace elements. Thus, adjusting the bulk concentration of an element for its terrigenous contribution will introduce little error in the calculation of its marine contribution, even for those elements whose concentrations in the terrigenous fraction are poorly defined. The terrigenous and marine contributions of the REE can be confirmed by normalizing REE concentrations on an element-by-element basis to their concentrations in the WSA and plotting the ratios versus atomic number (Fig. 21-4). The patterns of samples most strongly enriched in siliciclastic debris show that the terrigenous fraction had a REE composition that also was similar to the composition of WSA. The patterns of these samples are flat and have values that are only slightly less than one. That is, the concentration of each of the REE in these samples closely approaches its concentration in WSA. The slightly reduced values in these samples result from dilution of the terrigenous fraction by phases relatively depleted in REE, mostly calcite, dolomite, biogenic quartz, and organic matter (Elderfield et al., 1981).
Marine sediment fraction The phosphate-rich units within the Meade Peak Member have REE concentrations that strongly reflect a seawater contribution (Fig. 21-4), in addition to the terrigenous contribution. The normalized REE concentrations of these samples are greater than one and have REE patterns that differ from the shale pattern. Rather, the patterns are similar to the seawater pattern (Elderfield et al., 1981; DeBaar et al., 1985) with a strong negative Ce anomaly and an enrichment of the heavy REE (atomic numbers 64 through 71) relative to the light REE (atomic numbers 57 through 62). This same pattern and absolute concentrations characterize pelletal francolite that currently accumulates on the Peru shelf (Piper et al., 1988).
Western Phosphate Field: depositional and economic deposit models
581
6
tO O
< co ~
.....
-a~- ~ ~
-~"~
--
--:.~5.~
......
O- . . . . . . . .
6
tO O
X
m
E0.100
0.01 t i 56La 5 8 p r 60 Ce Nd
I
I
I
I
I
62 Eu 64 Th 66 Ho 68Tm 70 Lu 72 Sm Gd Dy Er Yb Atomic number/symbol
Fig. 21-4. The REE pattern of selected samples from the Meade Peak Member, normalized on an element-by-element basis to world shale average (WSA) values. Samples with the lowest values and strong negative Ce anomaly represent carbonate-rich samples; those with intermediate concentrations and a flat pattern are mostly siliciclastic debris; those with the highest concentrations, a strong negative Ce anomaly and heavy REE enrichment, relative to the light REE represent pelletal phosphate-rich samples.
Although the seawater REE pattern is strongest for the phosphate-rich samples, it also characterizes the carbonate-rich samples (Fig. 21-4). This is expected because the carbonate phases had a seawater source as well. The absolute concentrations of REE in the carbonaterich samples are slightly greater than their concentrations in modern calcite of marine planktonic origin, but the Ce anomaly (Palmer, 1985) is similar. The slightly higher values probably reflect the presence of small amounts of apatite and siliciclastic debris. Other trace elements enriched above a terrigenous source can also be shown to have had a seawater source. The seawater depth profiles in the modern ocean of Cd, Cu, Ni, Se, Zn, and, much less so Cr and V, show that they are taken up in the photic zone by algae as trace nutrients and largely returned to solution at depth by bacterial and zooplankton respiration. They have nutrient-type profiles (Broecker and Peng, 1982) in that their profiles resemble those of the major nutrients - PO43-, NOT, and H4SiO4. The concentrations of Mo and U in seawater are high enough (10 and 3 ppb, respectively) and in algae low enough (2 ppm or less) such that their depth profiles do not reflect their cycling between the photic and aphotic zones. Elemental ratios of these trace elements (Fig. 21-5A) and PO~- in the Meade Peak Member closely approach their respective ratios in modern plankton (Piper, 2001). This correspondence shows that this suite of trace elements, like PO43-, had a marine biogenic source. It implies that plankton in the Permian ocean had a composition, or nutrient demand, that was virtually identical to that of modern plankton. These similarities in composition to modern planktonic debris would seem to rule out the need for another source
582 (A)
PR. Moyle and D.Z. Piper 104
I
+
+ ~-I-+
+ 1000
+ + + +~
v i,_
o
+
++
~ ~ r+.*Y-t++. ~ +
~ , ~ . ++#+.+~. + +++ +~,+++..+§
+
E
+
+
++
+
+
/ +
+*~
100
. e.~ +
+ + % ++
J
seawat
10
plankton I
1
(B)
I
10 4
1000 E
+
Ck
4+ + ++ %++ + .~ ....-'~ +:1: + ++++*T~lankton + ~"1" .....-v-r -
O
100 t-I..
++
10-
+-t. +
+++
1 1
++ +
+
1
+ ~-'1-+ 4~- ++ +-,Y./~ ~ /+ ' ' + + + + y -h-
+++
'o
+
+
+
1 0' 0
1000
Marine Cu (ppm)
Fig. 21-5. Inter-element relations between the marine fraction of Cu and the marine fraction of (A) Cr and (B) Mo. The plankton and seawater lines represent the elemental ratios extrapolated into the sample fields. for these elements, for example, a hydrothermal source. Taken with the REE patterns of the phosphate- and carbonate-rich samples, they further identify Permian seawater as having had a composition similar to that of modern seawater. The planktonic debris originally would have had the following stoichiometry: (CH20)lo6(NH3)IsH3PO4(Fe, Zn, Cd, ... )0.00x (Redfield et al., 1963; Copin-Montegut and Copin-Montegut, 1983). As the formula shows, other nutrients include NOs and Fe(aq). However, PO43-, NOr, and Fe[aq], as well as H4SiO4, differ from the trace nutrients. As well as being essential, their availability in the photic zone of the ocean can limit primary productivity, although PO4~- and NOs are viewed as the universally limiting nutrients (Codispoti, 1989) and probably were the only
Western Phosphate Field." depositional and economic deposit models
583
limiting nutrients in the Phosphoria sea. Redbeds in the Goose Egg basin would have served as a source of Fe that supplemented its seawater source and sponges and radiolarians, or zooplankton rather than algae, constituted the silica secreting organisms in the Phosphoria sea. Thus, biogenic debris, largely algal in origin (Froelich et al., 1979; Ingall and Jahnke, 1997), delivered PO 3- as well as the trace nutrients to the sea floor. The PO 3- precipitated as francolite in the upper few centimeters of sediment (Kolodny, 1981; Burnett et al., 1988; Burnett and Froelich, 1988; Piper and Kolodny, 1987; Schuffert et al., 1994). In the Meade Peak Member, it had the following composition (Piper, 2001): Cas(PO4)2.9 F(CO3)0.15" Its rate of accumulation of 27 mg m -2 day-l, which corresponded to a PO 3- accumulation rate of 15 mgm -2 day -l, required a rain rate of organic matter onto the sea floor of 0.55 g m -2 day- 1. Although the latter is at the high end of rates of organic matter settling through the water column currently (Tsunogai and Noriki, 1987; Thunell, 1998), it can be as great as 0.83 g m -2 day -l. The PO 3-, as organic matter, that settled onto the sea floor may have represented as much as 25% of the PO 3- taken up by algae (Piper and Link, 2002). Sediment-trap experiments of modern continental-shelf environments clearly show that only about 5-15% of organic matter produced by primary productivity settles out of the photic zone (Von Bockel, 1981; Thunell, 1998; Miiller-Karger et al., 2001). This is significantly less than the value advanced here for the Phosphoria sea. In contrast, 100% of the PO 3- introduced by rivers eventually settles onto the sea f l o o r - the ocean is in steady state. How might the seemingly high percentage have been maintained in the Phosphoria sea, somewhere between the two extremes? Sediment traps record the results of a single cycle of PO 3- through the photic zone. By contrast, sea floor sediments record PO 3- cycling many tens of times between the photic zone and aphotic zone, during its residence time in the ocean of possibly as long as 100,000yr (Broecker and Peng, 1982). The much greater width of the Phosphoria basin (350-450km) than modern shelves (10-15km) that currently exhibit significant PO43- accumulation, and a residence time of water in the Phosphoria basin in the range of 4.5 yr (Piper, 2001), favored recycling of PO4 3- and other nutrients between the photic zone and bottom water of the basin. Still, a value of 25% of the PO43- taken up by algae being delivered to the seafloor may be high. However, that percentage is favored here owing to the limit it imposes on primary productivity. Primary productivity would have been approximately 0.85 g m -2 day -l C for this ocean-margin basin, similar to its rate in modem ocean-margin basins (Berger et al., 1988). A value lower than 25% would have required a higher level of primary productivity that approached the level of primary productivity on the continental shelf of Peru of 2.1 g m -2 day-l C (Chavez and Toggweiler, 1995). Pelletal phosphate accumulates today on the Peru shelf (Bumett et al., 1988; Filippelli, 1997), but this open-shelf environment would seem to be quite different from the partially isolated Phosphoria basin (Fig. 21-2), one major difference being the intensity of the surface current regime and, thus primary productivity.
584
PR. Moyle and D.Z. Piper
The accumulation of francolite and trace nutrients on the sea floor required a continuous supply of PO43- into the basin and the photic zone. The source of PO43- was subsurface seawater that would have upwelled into the photic zone, with a PO43- concentration of 3 Ixmol L-1. The PO 3- concentration of the photic zone is otherwise maintained at a low concentration, less than 0.5 txmol L-1, by an opportunistic algal population that blooms then quickly settles out of the photic zone through mortality or zooplankton grazing. On the Peru shelf and other upwelling areas, seawater is imported from depths between 100 and 250m, only a few tens of meters below the photic zone, that extends down to about 50 m depth (Codispoti, 1980). Water, PO43-, and other nutrients are exported back to the open ocean in the surface - 1 0 0 m , the Ekman layer (Wyrtki, 1963). The amounts exported are less than the amounts imported because of the loss of water by evaporation and loss of PO 3- and other nutrients by their export as organic detritus to depths below the Ekman layer and accumulation on the sea floor. The mean accumulation rate of PO43- of 15 m g m -2 day -1 required the import of 7.8 L cm-2yr -1 of seawater from the open ocean, or 7 8 m y r -1. The import rate and an estimated seafloor depth of 350m (McKelvey et al., 1959), give a residence time of water in the basin of 4.5 yr. The flow for water across the basin was approximately 0.60 cm s-1. In the Peru system, the onshore flow is estimated at 3 cm s-1 (Wyrtki, 1963; Chavez and Toggweiler, 1995). The ascending velocity for Peru upwelling water is 84 m yr-1, remarkably close to the 78 m yr-l for the Phosphoria sea. The upwelling rate in the Phosphoria sea (Fig. 21-2) was as little as 3.2% of the total upwelling in coastal regions of the ocean today, which is estimated to be 470-790 • 1015L yr- 1 (Chavez and Toggweiler, 1995). This 3.2% assumes a uniform phosphate accumulation rate throughout the area of the Phosphoria sea. However, the area of rapid phosphate accumulation was actually smaller by about one-twelfth (Sheldon, 1989). Scaling the accumulation rate accordingly gives an upwelling rate closer to 13 • 1015Lyr -l, or 1.7%, a seemingly reasonable value. The water column was temperature stratified over most of the basin (Piper and Link, 2002), as it is in upwelling environments in the ocean today (Codispoti, 1980; Schuffert et al., 1994). For example, off the west coast of Mexico, the temperature of seawater decreases with depth from about 27~ at the surface to 10~ at 200 m and to 5~ at 800 m; salinity increases by less than 1%o (Codispoti, 1972). A cold-water fauna (Wardlaw and Collinson, 1986; Murchey, Chapter 5) suggests even shallower cold-water isotherms for the Phosphoria sea. This interpretation of the fauna has been questioned (Hiatt, 1997), but a paleoceanographic stratification similar to the modern ocean (Kidder and Worsely, 200 l) offers support. Oxidation of organic matter in the bottom water through bacterial respiration consumed oxidants dissolved in the seawater (Piper and Link, 2002). As noted above, studies of modern marine systems showed that 85-95% of the organic matter produced in the photic zone is oxidized in the upper few hundred meters of the water column; we use 75%. The level of enrichment of the trace nutrients in the Meade Peak Member and their concentration in plankton required approximately 90% of the amount settling onto the sea floor to have been oxidized. The major reactions of organic-matter oxidation in seawater
Western Phosphate Field." depositional and economic deposit models
585
and near-surface sediment can be written in simplified form as follows: Oxygen respiration: 10CH20 -I- 1002 + 10CaCO3 = 20HCO3 + 10Ca 2+ Denitrification: 10CH20 + 8NOr + 2CACO3 = 4N2 + 12HCO3 + 4H20 + 2Ca 2+ SO42- reduction: 10CH20 + 5SO 2- + 5CACO3 = 5HS- + 15HCO3 + 5Ca 2+, The absence of accumulation of Cd, Cu, Mo (Fig. 21-5A), Ni, and Zn above their terrigenous-plus-biogenic inputs (Piper, 2001) to bottom sediments indicates that sulfate (SO42-) reducing conditions were not established in the bottom waters, which otherwise would have enhanced the presence of these trace elements (Jacobs et al., 1987). Oxidation of organic matter by SO42- reduction must have been restricted to sediment pore water (Piper and Kolodny, 1987). The enrichment of a third suite of trace elements (Cr (Fig. 21-5), U, V and the REE) above that which can be attributed to the accumulation of terrigenous debris and organic matter (Figs. 21-3-21-5) defines bacterial respiration in the bottom water of the basin as having been denitrifying (Emerson et al., 1979; German et al., 1991; Landing and Lewis, 1991; Piper, 2001). This same chemical profile characterizes the water column on the shelf off Peru (Burnett and Froelich, 1988) and the Pacific shelf off Mexico (Schuffert et al., 1994), two environments where pelletal francolite is currently accumulating. A balance between the supply of oxidants and the settling of labile organic matter determined the level of bacterial respiration in the bottom water. The 02 concentration of the seawater imported at depth into the basin was required to have been in the range of 10% (the minimum concentration that precluded sulfate reduction) to 25% saturation (the maximum concentration that allowed establishment of denitrification) (Piper and Link, 2002). Water of this 02 concentration would have been derived from the 02 minimum zone of the paleo North Pacific Ocean. Water fluxing into the denitrifying water column of the Santa Barbara Basin in the California Borderland (Sholkovitz and Gieskes, 1971) is derived from the 02 minimum zone of the modern North Pacific Ocean and has an 02 concentration of 5% saturation, only slightly less than the range for the Phosphoria sea. This comparison of the accumulation rate of PO4 3- and the trace-element inventory of the Meade Peak Member with their distributions in the ocean today demonstrates that standard seawater was the sole source of the fraction of several trace nutrients in excess of their terrigenous source and that the composition of organic matter in the Phosphoria basin was similar to the composition of modern phytoplankton. The accumulation of a hydrogenous fraction of trace-elements in the Meade Peak Member defined the level of bacterial respiration in the water column at depth as having been denitrifying, similar to the geochemistry of the water column of modem phosphogenic environments, for example, the Pacific Ocean continental shelves of Mexico and Peru. The accumulation rate of PO4 3- further required an ocean upwelling system whose circulation resembled seawater circulation in modern upwelling environments of the ocean margin. The composition of the siliciclastic fraction in the Meade Peak Member closely approached the composition of WSA and had a terrigenous source.
586
P.R. Moyle and D.Z. Piper
CHARACTER AND CONTROLS OF PHOSPHATE RESOURCESAN ECONOMIC MODEL
Geologic setting Thick layers of phosphorus-rich sediments deposited in the Permian sea were affected by a succession of diagenetic processes and tectonic events. Compressional forces associated with the Jurassic to Early Cenozoic Sevier and Laramide orogenies severely deformed the core of the Phanerozoic basin strata into a complex series of folds and thrust faults (Armstrong and Oriel, 1986; Burchfiel et al., 1992) known as the overthrust belt. Subsequent Cenozoic Basin and Range high-angle normal faulting added further structural complexity to portions of the folded stratigraphic sequence (Sheldon, 1989), especially in southeastern Idaho and northern Utah. The Phosphoria Formation hosts the rich and thick phosphate deposits of the WPF (see map in Jasinski et al., Fig. 3-1, Chapter 3) and is divided into three members at its type locality of Phosphoria Gulch in southeast Idaho. From oldest to youngest, these are the Meade Peak Phosphatic Shale Member, the Rex Chert Member, and the Cherty Shale Member (Sheldon, 1989; Evans, Chapter 6). The Meade Peak Member ranges from 33 to 62 m, averaging 54-m thick in southeast Idaho, and is the principal host of phosphate deposits (Gulbrandsen and Krier, 1980). In the core of the southeast Idaho phosphate district, the Meade Peak Member unconformably overlies the Grandeur Tongue of the Park City Formation and the Pennsylvanian-Permian Wells Formation. The Cherty Shale Member is overlain by the Dinwoody Formation, but elsewhere in the WPF, it grades into the Retort Phosphatic Shale Member. The Phosphoria Formation also intertongues to the northeast and south with the Shedhorn Sandstone and Park City Formations (Sheldon, 1989). Phosphate-rich strata are generally thickest and highest grade in southeast Idaho due to the exceptional conditions of the depositional environment considered above. These strata thin or pinch out to the north, east, and southeast but exhibit considerable local variation. McKelvey et al. (1967) best illustrated this in their restored section across southeast Idaho and western Wyoming (see Hein et al., Fig. 2-2, Chapter 2 for a reproduction). The primary ore mineral is carbonate fluorapatite (Cas[PO4, CO313[F, OH]). Principal gangue minerals found in unweathered rocks include quartz, buddingtonite (NH3-feldspar), albite, orthoclase, muscovite-illite, kaolinite, montmorillonite, calcite, and dolomite (Gulbrandsen, 1974; Desborough, 1977; Gulbrandsen and Krier, 1980; Knudsen et al., 2002; Knudsen and Gunter, Chapter 7). Organic matter is abundant, greater than 20% in some strata. Both framboidal and euhedral pyrite are present in low but variable concentrations, and minor phases, including pyrite and sphalerite, host some trace elements such as Cd and Se. Black shales of the Meade Peak are especially enriched in several trace elements compared to average shale, with significant concentrations of As, Cd, Cr, Cu, Mo, Se, U, V, and Zn (Altschuler, 1980).
Western Phosphate Field: depositional and economic deposit models
587
Geological attributes related to mining Mining of phosphate was conducted primarily by underground methods until the 1940s and 1950s and then by open-pit methods, which have been employed to the present. Five mines currently produce phosphate (Jasinski et al., Chapter 3): one in northern Utah, Little Brush Creek Mine (also called Vernal Mine), and four in southeastern Idaho - Dry Valley, Enoch Valley (Fig. 21-6), Rasmussen Ridge, and Smoky Canyon Mines. Phosphate ore is mined from two moderately to steeply dipping (25 ~ to -> 60 ~ intervals of the Meade Peak Member in southeast Idaho. The ore strata range in thickness from 5 to 8 m and typically contain from 20 to 35% P205. They enclose a 25-30 m thick middle waste zone comprised of low-grade phosphatic shale that typically averages less that 16-18% P205; thinner zones of low-grade phosphatic sediments also commonly lie above and below the upper- and lower-ore zones of the Meade Peak Member. The Little Brush Creek Mine in Utah produces from a single shallow-dipping (6-12 ~ ore zone that is about 5.2 m thick. Individual phosphate beds range in grade from 16 to 21.5% P205. Rocks in the ore and waste zones in the Meade Peak range from slightly altered (or slightly weathered) at depth to intensely altered (or weathered), typical of shallower (<- 100 m) exposures in the mines. Unaltered phosphate ore is black to dark gray and has high carbon content, whereas altered or weathered ore appears finely oolitic, gray, and has a lower density.
Fig. 21-6. Oblique view north of the north and south pits at the Enoch Valley Mine in southeast Idaho (photograph taken circa 1998). The pale-yellow Grandeur Tongue on the pit footwall is overlain by the Meade Peak Phosphatic Shale Member; folded and faulted Permian marine strata strike northwest (lower right to upper left) and dip southwest (lower left).
P.R. Moyle and D.Z. Piper
588
A composite stratigraphic section (Fig. 21-7) was developed from nine measured and sampled sections, two each at the four operating mines in southeast Idaho and one drill core (Herring et al., 1999, 2000a-c; Tysdal et al., 1999, 2000a-c; Table 21-I). In general, average compositions for P205 and Ca are higher in the ore zones, and organic carbon, Fe203, A1203, MgO, and CaO/P205 ratio are higher in the waste zones. Total thickness of the composited stratigraphic sections averaged 46 m for the Meade Peak; however, thickness of ore and waste intervals varied considerably (Table 21-I). Note that Gulbrandsen and Krier (1980) reported a thickness range of 33-62 m with an average of 54 m for the Meade Peak in the Soda Springs area.
Zone
Section Thickness
(m)
Organic C (%) Fe203(%)
P205 (%)
AI203(% )
MgO (%)
CaO (%)
CaO:P205
0
10
20
30
0
5
10
0
5
10
0
5
10
0
5
10
0
20
40
0
2.5
5
0
10
20
30
0
5
10
0
5
10
0
5
10
0
5
10
0
20
40
0
2.5
5
46.1 ~" 40.5 ..Q 35.6 E
~ 13.7 eeI-1.3 0
Fig. 21-7. Composite stratigraphic section across the Meade Peak Phosphatic Shale Member from nine measured and sampled sections exposed at four active mines in SE Idaho; numbers are samplelength-weighted averages for each zone.
TABLE 21-1 Thickness of phosphate ore and waste zones of the Meade Peak Member at four mines in southeast Idaho Zone Upper waste 1 Upper phosphate ore Middle waste Lower phosphate ore Lower waste I
Range (m)
Average (m)
1.7-11.6 0.3-17.4 2.0-32.7 6.8-25.0 0.3-2.6
5.6 4.9 21.9 12.5 1.3
i Lower- and upper-waste zones either not exposed or not present at two sections.
Western Phosphate Field: depositional and economic deposit models
589
Spatial variation in stratigraphic thickness throughout the WPF results only partly from depositional processes, particularly in southeast Idaho. Folding and thrust faulting, where subparallel to bedding within phosphate-rich strata, may effectively thicken the ore zone in some places, enhancing its value. In contrast, tensional faulting subparallel to bedding can attenuate or entirely remove a zone. Obviously, it is economically desirable to exploit tectonically thickened phosphate-rich strata. The criteria for P205 content, oxides, contaminants, minimum thickness of beds, depth of mining, and other factors employed at each mine are a function of local stratigraphic, structural, and geochemical conditions, available mining equipment, plant-feed requirements, and end-product specifications. Use of sophisticated excavation equipment and skilled operators now allows individual strata as thin as 0.15 m to be cost-effectively extracted. Generally, low-grade beds less than 0.15 m thick in the upper and lower phosphate-rich ore beds are blended with the extracted ore, whereas high-grade beds of similar thickness in the middle-waste zone are commonly included as waste. Waste rock from low-grade zones has conventionally been placed in external dumps; however, for both environmental and aesthetic reasons, back filling of surface mine pits behind active mining is now common practice at most operating mines. Mine pits in southeast Idaho are typically up to 100m deep depending on dip of strata and thickness of overburden. These pits are backfilled with waste rock, capped with rock less likely to be reactive, such as Rex Chert (see Hein et al., Chapter 14), and reclaimed by application of indigenous soil and selected seed mixes (see Mackowiak et al., Chapter 19).
Mining characteristics and specifications Three types of phosphate processing plants operate in the WPE An elemental phosphorus plant in Soda Springs, Idaho uses a high-temperature furnace to volatilize phosphorus. Three phosphoric acid p l a n t s - Pocatello and Conda, Idaho, and Rock Springs, Wyoming - employ an acidulation process to produce a range of fertilizers and related products. A relatively new purified phosphoric acid (PPA) process plant in Conda, Idaho, removes impurities from phosphoric acid supplied by the fertilizer plant co-located in Conda. Ore-feed specifications vary for each type of plant and are subject to modification as both geologic conditions and product requirements change. Naturally, the primary quality specification for plant feed is P205 content. Elemental phosphorus plant feed ranges from 25 to 27%, whereas fertilizer-grade plant feed tends to be greater than 30% P205, so some upgrading is commonly required. Suitability of phosphate ore for plant feed is a function of more than P205 content, such as concentrations of other constituents (Table 21-II). Processing costs are sensitive to concentrations of several of the major-element oxides, particularly MgO and CaO, as well as other components such as moisture and organic matter. For example, the acidulation process in fertilizer manufacture is sensitive to MgO concentration, whereas it presents no problem for the elemental phosphorus plant. The presence of clay-sized debris may enhance the formation of nodules, a favorable property for manufacturing elemental phosphorus. It is common practice for some phosphate mines and plants to stockpile particular types of ore based on P205
P.R. Moyle and D.Z. Piper
590 Table 21-II
Phosphate ore quality and open pit mine parameters currently in effect for run-of-mine specifications composited from four operating mines in the southeast Idaho Parameter
Limit
P205 content
Minimum Average Maximum Maximum Maximum Maximum Maximum Maximum
18-20% 26-27% 0.3-0.6% 1.6-5.0% 1.5-1.6 4-5% 7% 2.5-4.2
Minimum
0.3 m
MgO content
AI203 content CaO/P205 Total carbon content Loss on ignition Overburden/ore (m3mt -l) Mining width (thickness of strata mined) Mining depth
Maximum
Specification
90-110 m
content as well as on the amount of several major oxides- Fe203, A1203, and M g O - and the relative concentrations of oxides such as the ratio of CaO to P205. Selected stockpiling allows for blending that maintains plant feed of a constant character. Blending can also be adjusted as needed if product specifications change. Mining of the typically moderate- to steep-dipping beds in southeast Idaho restricts open pits to a depth of approximately 90-110 m (Fig. 21-8) because of overburden volume, the competency of footwall rock, and the depth of weathering. Weathering, particularly, plays an important role in determining the suitability of phosphate rock for plant feed. As early as the 1920s, Mansfield (1927) speculated about the role of weathering in phosphate ore enrichment. He described a profound change in phosphate content in a series of samples beginning at a partially weathered surface outcrop with 38% P205 and progressing 260 m down dip on the same main bed to relatively less-weathered rock with a P205 content of 30.7%. Assuming the change can be entirely attributed to weathering, this amounts to a relative enrichment of over 20% from the deep sample to the surface.
Weathering Gulbrandsen and Krier (1980) documented the significant mineralogical and physical differences between deeper, relatively unweathered Meade Peak compared to intensely weathered rock exposed in surface and near-surface sections in southeast Idaho. Apatite maintains a relatively constant ratio of 0.51 to combined silicate minerals (quartz, buddingtonite, albite, orthoclase, muscovite-illite, kaolinite, montmorillonite) and pyrite in both unweathered and weathered sections of Meade Peak. However, the proportion of
Western Phosphate Field: depositional and economic deposit models
591
Fig. 21-8. Cross-section of open-pit phosphate mine showing composite ore and waste zones from four active operations in southeast Idaho: Enoch Valley (Monsanto); Rasmussen Ridge (Agrium); Dry Valley (Astaris); Smoky Canyon (J.R. Simplot). Note that strike, dip, and thickness of strata, local lithology and structure, and pit depth and configuration vary between sites. combined calcite, dolomite, and organic matter to the other components was significantly reduced in weathered rock (see Knudsen and Gunter, Chapter 7). Furthermore, loss of CO2 from carbonate fluorapatite has been documented to occur in phosphate deposits in Morocco, Senegal, and Israel (Lucas et al., 1980).
592
PR. Moyle and D.Z. Piper
Thus, natural weathering processes result in a relative enrichment of ore quality by mobilizing some undesirable components such as MgO, CaO, and organic matter away from near surface portions and deeper fluid pathways such as faults. Weathering also increases clay-mineral content (Knudsen et al., 2002; Knudsen and Gunter, Chapter 7), further enhancing the ability of phosphate rock to nodulize. Due to the net loss of dolomite, calcite, and organic matter, the porosity of weathered rock is generally higher than unweathered rock; consequently, the average density of weathered phosphate rock with greater than 20% P205 is lower, calculated by Gulbrandsen and Krier (1980) to be 2.39 g/cm 3, compared to 2.64 g/cm 3 for unweathered phosphate rock.
R e s o u r c e s a n d reserves
Phosphate resources for southeast Idaho, where most historic and active mining has occurred, were reported by Kirkham (1925) to be 5.6 billion metric tons (mt) and estimated by Mansfield (1927) to be nearly 5.1 billion mt, both based on USGS study. Systematic investigations of phosphate resources in Idaho and in the WPF were also made by Service (1966, 1967) and the Garrand Corporation (Garrand, 1975). Gulbrandsen and Krier (1980) estimated that the Soda Springs, Idaho, area has 4.1 billion mt of phosphate rock with greater than 20% P205. Phosphate resource and reserve estimates for the WPF (Table 21-III) made most recently by Sheldon (1989) and Cathcart (1991) were based on selected mine parameters and ore characteristics (Table 2 l-IV). They estimated that the field contains a surface mineable reserve (including inferred reserve) base of 1.6 billion mt at an average grade of 24% P20 5 with an additional surface mineable subeconomic resource of 6.0 billion mt at the same grade. Estimated subeconomic resources by underground mining at a grade of TABLE 21-III Resources of the Western Phosphate Field based on calculations of Sheldon (1989) and Cathcart (1991) Type
Strippable resources Reserve and inferred reserve base Subeconomic resources Underground resources Above entry level- subeconomic resources Below entry level to 305 m deep Westem Phosphate Field Total
Resources ( 109 metric tons)
P205 (%)
1.6 6.0
24 24
4.0 13.0 24.6
28 28
Western Phosphate Field: depositional and economic deposit models
593
TABLE 21-IV Specific mine parameters and ore characteristics on which phosphate resource estimates in Table 21-III were based Parameter
Specification
P205 content MgO content
> 18% < 1.5%
Fe203 + A1203 content
<3.0%
CaO/P205 Overburden/Ore (m3mt -i) Mining width of bed Mine floor width Deposit size
< 1.55 <3.5 1.5 m 76 m _> 20 x 106mt
28% P205 include 4.0 billion mt above entry level and an additional 13 billion mt below entry level to a depth of 305 m. An additional 507 billion mt of phosphatic material not regarded as resources underlie the WPF more than 305 m below the surface (Sheldon, 1989). Since the most recent resource estimates were made circa 1990, approximately 97 million mt of phosphate crude ore were produced (1990-2000) from the WPF (Table 3-I of Jasinski et al., Chapter 3), about 6% of the calculated reserve base. Additional geological mapping (Evans, Chapter 6) since 1997, and on-going modeling of phosphate resources using similar criteria for eight quadrangles in southeast Idaho indicate that the resource and reserve estimates of Sheldon (1989) and Cathcart (1991) remain valid.
CONCLUSIONS Many geological and geochemical processes- from Permian marine depositional conditions to Mesozoic and Cenozoic continental-scale tectonic events to recent and on-going weathering of the shallow subsurface- contributed to the formation and economic significance of the WPF, particularly the thick, high-grade deposits in southeast Idaho. Of the total calculated endowment of 524 billion mt of phosphatic materials in the field, only 24.6 billion mt, or 4.7%, are considered mineable reserves and subeconomic resources by surface and underground methods, and only 1.6 billion mt, or 0.31%, constitute the surface-mineable reserve and inferred reserve base (Sheldon, 1989). Some of the most significant physical and chemical parameters that currently define an economically mineable deposit include: (a) deposit size of at least 20,000,000mt; (b) minimum P205 content of 18%; (c) MgO content of 0.3-0.6%; (d) maximum CaO/P205 ratio of 1.5-1.6; (e) maximum organic-carbon content of 4-5%; (f) maximum overburden to ore ratio of 2.5-4.2 m 3 mt-l;
594
P.R. Moyle and D.Z. Piper
and (g) mining depth of about 100 m or less (Fig. 21-8 and Table 21-II). Clearly, weathering plays a critical role in modifying a favorable phosphate deposit into one that is currently economically viable. Significant changes in processing with consequent cost increases will be required to use phosphate-rich rock mined below the typical weathering depth of about 100 m.
ACKNOWLEDGMENTS We appreciate the mine managers and geologists who provided input on local geology and the physical and chemical parameters relevant to the economics of mining and processing phosphate in the Western Phosphate Field: Alan Haslam, Enoch Valley Mine, Agrium; Kelly Ransom, Dry Valley Mine, Astaris; Ray Petrun (Deceased; Ray made significant contributions to the understanding and development of phosphate resources in the WPE He was tragically killed in 2003 in a logging accident), Enoch Valley Mine, Monsanto; and Ken Reighard and Susan Nash, Smoky Canyon Mine, J.R. Simplot. We thank Peter Oberlindacher, Art Bookstrom, and James Hein for their valuable review comments.
REFERENCES Altschuler, Z.S., 1980. The geochemistry of trace elements in marine phosphorites, Part I, Characteristic abundances and enrichment. Soc. Econ. Paleon. Min., Spec. Pub., 29: 19-30. Armstrong, EC. and Oriel, S.S., 1986. Tectonic development of the Idaho-Wyoming thrust belt; Authors' commentary. In: J.A. Peterson (ed.), Paleotectonics and Sedimentation in the Rocky Mountain Region, United States. American Association of Petroleum Geologists Memoir 41, pp. 267-279. Berger, W.H., Fischer, K., Lai, C. and Wu, G., 1988. Ocean carbon flux - global maps of primary production and export production. In: C.R. Agegian (ed.), Biogeochemical cycling and fluxes between the deep euphotic zone and other oceanic realms. N.O.A.A. Under sea Research Program, Report 88-1, pp. 131-176. Broecker, W.S. and Peng, T.H., 1982. Tracers in the sea. Eldigio Press, Palisades, NY, 690 pp. Burchfiel, B.C., Cowan, D.S. and Davis, G.A., 1992. Tectonic overview of the Cordilleran orogen in the western United States. In: B.C. Burchfiel, EW. Lipman and M.L. Zoback (eds.), The Cordilleran Orogen: Conterminous US, vol. G-3, The Geology of North America. Boulder, CO, Geological Society of America, pp. 407-479. Burnett, W.C. and Froelich, EN., 1988. The origin of marine phosphorite- the results of the R/V Robert D. Conrad Cruise 23-06 to the Peru shelf. Mar. Geol., 80:181-343. Burnett, W.C., Baker, K.B., Chin, EA., McCabe, W. and Ditchburn, R., 1988. Uranium-series and AMS 14C studies of modern phosphate pellets from Peru shelf muds. Mar. Geol., 80: 215-230. Cathcart, J.B., 1991. Phosphate deposits of the United States - discovery, development, economic geology and outlook for the future. In: H.G. Gluskoler, D.D. Rice and R.B. Taylor (eds.), Economic Geology, US, vol. P-2, The Geology of North America. Boulder, CO, Geological Society of America, pp. 153-164.
Western Phosphate Field: depositional and economic deposit models
595
Chavez, EE and Toggweiler, J.R., 1995. Physical estimates of global new production - the up-welling contribution. In: C.E Summerhayes, K.C. Emeis, M.V. Angel, R.L. Smith and B. Zeitzschel (eds.), Upwelling in the ocean - modern processes and ancient records. Dahlem Workshop Reports, Environmental Sciences Research Report 18, John Wiley and Sons, Chichester, pp. 313-320. Clarke, EW., 1924. Data of geochemistry. US Geological Survey, Bulletin, 770, 841 pp. Codispoti, L.A., 1972. Denitrification in the eastern tropical North Pacific Ocean. PhD thesis, University of Washington, Seattle, WA, 118 pp. Codispoti, L.A., 1980. Temporal nutrient variability in three different upwelling regimes. In: EA. Richards (ed.), Coastal upwelling. Coastal and Estuarine Sciences, vol. l, American Geophysical Union, Washington, DC, pp. 209-220. Codispoti, L.A., 1989. Phosphorus vs. nitrogen limitation of new and export production. In: W.H. Berger, V.S. Smetacek and G. Wefer (eds.), Productivity of the Oceans- Present and Past. John Wiley and Sons, Chichester, UK, pp. 377-394. Copin-Montegut, C. and Copin-Montegut, G., 1983. Stoichiometry of carbon, nitrogen, and phosphorus in marine particulate matter. Deep-Sea Res., Part A, 30:31-46. Dahl, J., Moldowan, J.M. and Sundararaman, P., 1993. Relationship of biomarker distribution to depositional environments- Phosphoria Formation, Montana, USA. Organic Geochem., 20: 1001-1017. DeBaar, H.J.W., Bacon, M.E, Brewer, EG. and Bruland, K.W., 1985. Rare earth elements in the Pacific and Atlantic Oceans. Geochim. Cosmochim. Acta, 4:1943-1959. Desborough, G.A., 1977. Preliminary report on certain metals of potential economic interest in thin vanadium-rich zones in the Meade Peak Member of the Phosphoria Formation in western Wyoming and eastern Idaho. US Geological Survey, Open File Report, 77-341, 27 pp. Elderfield, H., Hawkesworth, C.J., Greaves, M.J. and Calvert, S.E., 198 I. Rare earth element geochemistry of oceanic ferromanganese nodules and associated sediments. Geochim. Cosmochim. Acta, 45: 513-528. Emerson, S.R., Cranston, R.E. and Liss, ES., 1979. Redox species in a reducing fjord- equilibrium and kinetic observations. Deep-Sea Res., 26: 859-878. Filippelli, G.M., 1997. Controls on phosphorus concentration and accumulation in oceanic sediments. Mar. Geol., 139:231-240. Froelich, EN., Klinkhammer, G.P., Bender, M.L., Luedtke, N.A., Heath, G.R., Cullen, D., Dauphin, P., Hammond, D., Hartman, B. and Maynard, V., 1979. Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic- suboxic diagenesis. Geochim. Cosmochim. Acta, 43(7): 1075-1090. Garrand, L.J., 1975. Phosphate study southeastern Idaho. Garrand Corporation, Salt Lake City, Utah, US DA, Contract No. 50-820, Report, Appendix, Map Folio, scale 1:24,000. German, C.R., Holliday, B.E and Elderfield, H., 1991. Redox cycling of rare earth elements in the suboxic zone of the Black sea. Geochim. Cosmochim. Acta, 55: 3553-3558. Geslin, J. K., 1998. Distal ancestral Rocky Mountains tectonism- evolution of the PennsylvanianPermian Oquirrh-Wood River basin, southern Idaho. Geol. Soc. Am. Bull., 110: 644-663. Goldberg, E.D., 1963. Mineralogy and chemistry of marine sedimentation. In: EE Shepard (ed.), Submarine Geology. Harper and Row, New York, NY, pp. 436-466. Gulbrandsen, R.A., 1974. Buddingtonite, ammonium feldspar, in the Phosphoria Formation, southeastern Idaho. US Geol. Surv. J. Res., 2(6): 693-697. Gulbrandsen, R.A. and Krier. D.J., 1980. Large and rich phosphorus resources in the Phosphoria Formation in the Soda Springs area, southeastern Idaho. US Geological Survey, Bulletin, 1496, 25 pp.
596
PR. Moyle and D.Z. Piper
Herring, J.R., Desborough, G.A., Wilson, S.A., Tysdal, R.G., Grauch, R.I. and Gunter, M.E., 1999. Chemical composition of weathered and unweathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation - A. Measured sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 99-147-A, 24 pp. Herring, J.R., Grauch, R.I., Desborough, G.A., Wilson, S.A. and Tysdal, R.G., 2000a, Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation - C. Measured sections E and F, Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 99-147-C, 35 pp. Herring, J.R., Wilson, S.A., Stillings, L.A., Knudsen, A.C., Gunter, M.E., Tysdal, R.G., Grauch, R.I., Desborough, G.A. and Zielinski, R.A., 2000b. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation- B. Measured sections C and D, Dry Valley, Caribou County, Idaho. US Geological Survey, Open File Report, 99-147-B, 33 pp. Herring, J.R., Grauch, R.I., Tysdal, R.G., Wilson, S.A. and Desborough, G.A., 2000c. Chemical composition of weathered and less weathered strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation- D. Measured sections G and H, Sage Creek Canyon area of the Webster Range, Caribou County, Idaho. US Geological Survey, Open File Report, 99-147-D, 38 pp. Hiatt, E.E., 1997. A paleoceanographic model for oceanic upwelling in a late Paleozoic epicontinental s e a - a chemostratigraphic analysis of the Permian Phosphoria Formation. PhD thesis, University of Colorado, Boulder, CO, 300 pp. Ingall, E. and Jahnke, R., 1997. Influence of water-column anoxia on the elemental fractionation of carbon and phosphorus during sediment diagenesis. Mar. Geol., 139: 219-229. Jacobs, L., Emerson, S.R. and Huested, S.S., 1987. Trace metal chemistry in the Cariaco Trench. Deep-Sea Res., 34(5-6A): 965-981. Kidder, D.L. and Worsely, T.R., 2001. Storms in the later Permian and Early Triassic (abstract). Geol. Soc. Am., Annual Meeting, 33(6): A444. Kirkham, V.R.D., 1925. Phosphate deposits of Idaho and their relation to world supply. Idaho Bureau Mines and Geology, reprint no. 1, 28 pp. Knudsen, A.C., Gunter, M.E, Herring, J.R. and Grauch, R.I., 2002. Mineralogical characterization of strata of the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation, channel and individual rock samples of measured section J and their relationship to measured sections A and B, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 02-125, 37 pp. Kolodny, Y., 1981. Phosphorites. In: C. Emiliani (ed.), The Sea - Ideas and Observations On Progress in the Study of the Seas, vol. 7. Wiley-Interscience, New York, pp. 981-1023. Landing, W.M. and Lewis, B.L., 1991. Thermodynamic modeling of trace metal speciation in the Black sea. In: Erol Izdar and J.W. Murray (eds.), Black Sea Oceanography. Kluwer, Dordrecht, pp. 125-160. Lucas, J., Flicoteaux, R., Nathan, T., Prevot, L. and Shahar, Y., 1980. Different aspects of phosphorite weathering. In: Y.K. Bentor (ed.), Marine phosphorites- geochemistry, occurrence, genesis. Soc. Econ. Paleon. Min., Spec. Pub. No. 29, pp. 41-51. Maughan, E.K., 1994. Phosphoria Formation (Permian) and its resources significance in the western interior, USA. In: EA. Embry, B. Beauchamp, and D.J. Glass (eds.), Pangea: Global Environments and Resources. Canadian Soc. Petrol. Geol. Memoir, 17, pp. 479-495.
Western Phosphate Field." depositional and economic deposit models
597
Mansfield, G.R., 1927. Geography, geology, and mineral resources of part of southeastern Idaho with a description of Carboniferous and Triassic fossils, by G. H. Girty. US Geological Survey, Professional Paper, 152, 453 pp. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M. and Swanson, R.W., 1959. The Phosphoria, Park City, and Shedhorn Formations in the western phosphate field. US Geological Survey, Professional Paper, 313-A, 47 pp. McKelvey, V.E., Williams, J.S., Sheldon, R.P., Cressman, E.R., Cheney, T.M. and Swanson, R.W., 1967. The Phosphoria, Park City, and Shedhorn Formations in western phosphate field. In: Anatomy of the western phosphate field, a guide to the geologic occurrence, exploration methods, mining engineering, and recovery technology. Intermountain Association of Geologists, 15th Annual Field Conference, pp. 15-33. Medrano, M.D. and D.Z. Piper, 1995. Partition of minor elements and major-element oxides between rock components and calculation of the marine-derived fraction of the minor elements in rocks of the Phosphoria Formation, Idaho and Wyoming. US Geological Survey, Open-File Report, 95-270, 79 pp. Mosier, D.L., 1986. Descriptive model of upwelling type phosphate deposits. In: D.E Cox, and D.A. Singer (eds.), Mineral Deposit Models. US Geological Survey, Bulletin, 1693, pp. 234-236. Miiller-Karger, E, Varela, R., Thunell, R., Scranton, M., Bohrer, R., Taylor, G., Capelo, J., Astor, Y., Tappa, E., Ho, T.Y. and Walsh, J.J., 2001. Annual cycle of primary productivity in the Cariaco basin-response to upwelling and implications for vertical export. J. Geophys. Res., 106(C3): 4527-4542. Palmer, M.A., 1985. Rare earth elements in foraminifera tests. Earth Planet. Sci. Lett., 73(2-4): 285-298. Piper, D.Z., 2001. Marine chemistry of the Permian Phosphoria Formation and basin - southeast Idaho. Econ. Geol., 96: 599-620. Piper, D.Z. and Isaacs, C.M., 1995. Geochemistry of minor elements in the Monterey Formation, California- seawater chemistry of deposition. US Geological Survey, Professional Paper, 1566, p. 41. Piper, D.Z. and Kolodny, Y., 1987. Stable isotopic composition of a marine phosphate deposit: Deep-Sea Res., 34:897-911. Piper, D.Z. and Link, P.K., 2002. An upwelling model for the Phosphoria sea: a Permian, oceanmargin sea in the northwest United States. Am. Assoc. Pet. Geol. Bull., 86(7): 1217-1235. Piper, D.Z., Baedecker, P.A., Crock, J.G., Burnett, W.C. and Loebner, B.J., 1988. Rare-earth elements in the phosphate-rich sediment of the Peru Shelf. Mar. Geol., 80: 269-285. Poole, EG., 1964. Palaeowinds in the western United States. In: A.P.M. Narin (ed.), Problems in Palaeoclimatology. John Wiley and Sons, London, pp. 394--405. Redfield, A.C., Ketchum, B.H. and Richards, EA., 1963. The influence of organisms on the composition of seawater. In: M.N. Hill (ed.), The Composition of seawater - Comparative and Descriptive Oceanography: the S e a - Ideas and Observations on Progress in the Study of the Sea, vol. 2. Wiley Interscience, New York, pp. 26-77. Schuffert, J.D., Jahnke, R.A., Kastner, M., Leather, J., Struz, A. and Wing, M.R., 1994. Rates of formation of modern phosphorite off western Mexico. Geochim. Cosmochim. Acta, 58:5001-5010. Service, A.L., 1966. An evaluation of the western phosphate industry and its resources (in five parts), 3. Idaho. US Bureau of Mines, Rep. Invest., 6801, 201 pp. Service, A.L., 1967. Evaluation of the phosphate reserves in southeastern Idaho. In: L.A. Hale (ed.), Anatomy of the Western Phosphate Field, a guide to the geologic occurrence, exploration methods,
598
P.R. Moyle and D.Z. Piper
mining engineering, and recovery technology: Intermountain Association of Geologists, 15th Annual Field Conference, pp. 73-96. Sheldon, R.P., 1989. Phosphorite deposits of the Phosphoria Formation, western United States. In: A.J.S. Notholt, R.P. Sheldon and D.E Davidson (eds.), Phosphate Deposits of the World, vol. II. Cambridge Press, London, pp. 53-61. Sholkovitz, E.R. and Gieskes, J.M., 1971. A physical-chemical study of the flushing of the Santa Barbara basin. Limn. Ocean., 16: 479-489. Stephens, N.P. and Carroll, A.R., 1999. Salinity stratification in the Permian Phosphoria s e a - a proposed paleoceanographic model. Geology, 27: 899-902. Thunell, R., 1998. Particle fluxes in a coastal upwelling z o n e - sediment trap results from Santa Barbara basin. Deep-Sea Res., 45: 1863-1884. Tsunogai, S. and Noriki, S., 1987. Organic matter fluxes and the sites of oxygen consumption in deep water. Deep-Sea Res., 34:755-767. Tysdal, R.G., Desborough, G.A., Herring, J.R., Grauch, R.I. and Stillings, L.A., 2000a. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Dry Valley, Caribou County, Idaho. US Geological Survey, Open File Report, 99-20-B, plate. Tysdal, R.G., Grauch, R.I., Desborough, G.A. and Herring, J.R., 2000b. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, east-central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 99-20-C, plate. Tysdal, R.G., Herring, J.R., Grauch, R.I., Desborough, G.A. and Johnson, E.A., 2000c. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, Sage Creek area of Webster Range, Caribou County, Idaho. US Geological Survey, Open File Report, 99-20-D, plate. Tysdal, R.G., Johnson, E.A., Herring, J.R. and Desborough, G.A., 1999. Stratigraphic sections and equivalent uranium (eU), Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, central part of Rasmussen Ridge, Caribou County, Idaho. US Geological Survey, Open File Report, 99-20-A, plate. Von Bockel, K., 1981. A note on short-term production and sedimentation in the upwelling region off Peru. In: EA. Richards (ed.), Coastal Upwelling, vol. 1 of Coastal and Estuarine Sciences. American Geophysical Union, Washington, DC, pp. 291-297. Wardlaw, B.R. and Collinson, J.W., 1986. Paleontology and deposition of the Phosphoria Formation. In: D.W. Boyd and J.A. Lillegraven (eds.), Western Phosphate Deposits. Contributions to Geology, University of Wyoming, vol. 24, pp. 107-142. Wedepohl, K.H. (ed.), 1969-1978. Handbook of Geochemistry. Springer-Verlag, Berlin. Wilde, G.L., 2000. Formal Middle Permian (Guadalupian) Series: a Fusulinacean perspective. In: B.R. Wardlaw, R.E. Grant and D.M. Rohr (eds.), The Guadalupian Symposium: Smith. Contribution to the Earth Science, vol. 32, pp. 89-100. Wyrtki, K., 1963. The horizontal and vertical field of motion in the Peru Current. Scripps Inst. Oceanogr. Bull., 8(4): 313-344.
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
599
Chapter 22
SOCIETAL RELEVANCE, PROCESSING, AND MATERIAL F L O W OF W E S T E R N PHOSPHATE - R E F R E S H M E N T S , FERTILIZER, AND WEED KILLER
S.M. JASINSKI
ABSTRACT Phosphate minerals are the only available resource of phosphorus, an essential element for plant, animal, and human nutrition. In addition, phosphorus compounds have important uses as food processing and soft drink additives, herbicides, detergents, and in water treatment. Phosphate rock is mined by five companies in the Western Phosphate Field (WPF) of the United States and converted into phosphoric acid for fertilizer use or elemental phosphorus for industrial applications. Since the start of modern phosphate mining in the WPF in 1948, phosphate rock use in the western United States has been nearly evenly divided between agricultural and industrial applications. The dissipative use of phosphates in many industries makes it difficult to quantify the amount of phosphorus that cycles through the environment.
INTRODUCTION Phosphorus, the sought-after component of phosphate rock, is an essential element for plant, animal, and human nutrition and has numerous industrial applications. Phosphate rock may be processed to yield either phosphoric acid or elemental phosphorus products. The principal uses for phosphorus are as a fertilizer or animal-feed supplement. Other applications include industrial chemicals, herbicides, detergents, and food additives. Approximately 95% of phosphate rock mined and beneficiated in the United States is treated with sulfuric acid to produce merchant-grade phosphoric acid for fertilizer use or for upgrading into purified phosphoric acid for industrial applications. Five mines, three fertilizer plants, and one elemental phosphorus plant are located in the Western Phosphate Field (WPF) (Table 22-I). Monsanto Co. operates the only elemental phosphorus plant in the Western Hemisphere at Soda Springs, Idaho; three other elemental phosphorus plants have been located in the region since 1948 (Table 22-1I). Astaris LLC mines phosphate rock in Idaho exclusively to produce purified merchant-grade phosphoric acid for nonfertilizer applications. In 2000, the combined processed mineral value of the four operating
600
S.M. JasinsM
TABLE 22-1 Active phosphate-rock mines in the Western Phosphate Field (after Jasinski, 2002) Company
Mine name
Location: county, state
End use for ore
Location of processing: plant city, state
Agrium US Inc.
Caribou, Idaho Caribou, Idaho
Monsanto Co. J.R. Simplot Co.
Enoch Valley Smoky Canyon
Caribou, Idaho Caribou, Idaho
SF Phosphates, Ltd Co.
Vernal(Little Brush Creek)
Uintah, Utah
Phosphoric acid Purifed Phosphoric acid Phosphorus Phosphoric acid Phosphoric acid
Conda, Idaho
Astaris LLC
Rasmussen Ridge Dry V a l l e y
Conda, Idaho
Soda Springs, Idaho Pocatello, Idaho Rock Springs, Wyoming
TABLE 22-II Elemental-phosphorus plants in the Western Phosphate Field (after Service and Popoff, 1964; Jasinski, 2002) Company
Location" (county, state)
Year opened
Year closed
FMC Corp.
Bannock, Idaho Bear Lake, Idaho Caribou, Idaho Silver Bow, Montana
1949
2001
127,000
1959
1963
22,700
1952
Active
109,000
1951
1995
34,000
Central Farmers Fertilizer Co. Monsanto Co. Rhodia, Inc.
Annual production capacity (tons)
phosphate mines and processing plants constituted the largest mining industry of Idaho in terms of value (US Geological Survey, 2002). Since the development of the WPF as a significant phosphate producer in the late 1940s, the proportion of phosphate rock used for phosphoric acid and elemental phosphorus has been nearly evenly divided (Fig. 22-1).
UTILIZATION AND SOCIETAL RELEVANCE OF PHOSPHATE The human body requires a daily intake of 0.6-0.7 g of phosphorus in the appropriate chemical form (Krauss et al., 1984). It is consumed primarily from food products that have
Societal relevance, processing, and material flow of western phosphate
601
Fig. 22-1. Western phosphate rock use distribution, 1948-2000: (A) Summary distribution; (B) Annual distribution.
been grown with nitrogen-phosphorus-potassium (NPK) fertilizers or meat products. In addition, phosphorus compounds have many important industrial and food preservative applications. Phosphate minerals comprise the only significant global resources of phosphorus. The proportion of phosphorus in the Earth's crust averages 0.10-0.12%, 1 l th in order of abundance (Krauss et al., 1984). Significant deposits in the United States are located in Florida, Idaho, North Carolina, and Utah. Historically, significant quantities of phosphate rock were also mined in Montana, South Carolina, Tennessee, and Wyoming. Economic phosphate deposits in the western United States are marine sedimentary deposits that are formed by deposition of abundant organic matter at the seafloor. Degradation of that organic matter took place within the upper few centimeters of burial, which released the phosphorus that precipitated as phosphorite. The organic matter originated as plankton in regions of upwelling or cold, nutrient-rich seawater from depths of a few hundred meters.
602
S.M. JasinsM
Phosphate content in these rocks was further enriched through weathering (Krauss et al., 1984; Knudsen and Gunter, Chapter 7; Moyle and Piper, Chapter 21). The most important mineral in marine phosphorite ores is francolite, a variety of apatite. Phosphate deposits currently being mined in the United States have phosphorus pentoxide (P205) contents ranging from about 8 to 38%; ores in the WPF have about 27-28% P205 (Moyle and Piper, Chapter 21). Western phosphate ores also may contain recoverable amounts of vanadium. Until 1999, approximately 2000t of vanadium pentoxide was produced annually by the Kerr-McGee Chemical Corporation plant in Soda Springs from ferrophosphorus slag generated from Monsanto's elemental phosphorus plant. The vanadium recovery plant closed for financial reasons in 1999, and since then, ferrophosphorus slag has been stockpiled onsite at the Monsanto facility.
MINING METHODS Mining methods in the WPF are similar for phosphate rock destined for both elemental phosphorus and fertilizer production. The major difference is the grade of ore necessary for processing. Elemental phosphorus production uses a lower-grade ore, referred to as furnace-grade phosphate rock, which has a P205 content between 23 and 28%. Furnacegrade rock is sorted by P205 content and blended at the processing facility to achieve the necessary concentration (Fig. 22-2). Phosphate rock used in phosphoric acid production (acid-grade) requires a P205 content of - 3 0 % , which can necessitate further processing of the ore by selective flotation or calcining to remove impurities and to concentrate the ore (Fig. 22-3). Typically, the phosphate ore occurs as two stratigraphic intervals separated by lowgrade center-waste shale. The upper ore is 5-6 m thick and the lower ore is 15-16 m thick. Mining is done with truck and shovels on 6 m benches. Initially, the overburden is blasted in a 6 m • 6 m pattern. Much of the waste removal is done in summer to minimize mining in subfreezing weather. During the cold winter season, waste shale can freeze to the ore, causing dilution, which in turn increases electricity cost at the furnace. After mining, the furnace-grade ore is stockpiled in two grades, sorted by P205 content. The ore is then moved by a bulldozer into a pair of hopper openings covered by a 0.45 m mesh grizzly. Ore is then moved through two tunnels onto a pair of double-deck vibrating screens. The material is sized to < 50 mm. Larger material is sent to a crusher and recycled back onto the screens. The final product is sent to two 150 t bins, which are then loaded into trucks for transport to the plant (Vranes, 1999). For acid-grade ore, mining methods and occurrence of the phosphate rock are similar for mines that produce furnace-grade ore. After mining and initial sorting, the ore is transported to the phosphoric acid plant either by truck, train, or slurry pipeline. Ore that is high in organic material may be calcined to volatilize the impurities. Selective flotation also is employed to concentrate the ore and remove inorganic impurities (Fig. 20-3). The final product is referred to as marketable phosphate rock.
Societal relevance, processing, and material flow of western phosphate
603
Fig. 22-2. Typical phosphate ore-processing sequence for elemental-phosphorus production (after Li, 1978).
Fig. 22-3. Typical phosphate ore-processing sequence for phosphoric-acid production.
S.M. Jasinski
604 PROCESSED PRODUCTS
Phosphoric acid production Wet-process phosphoric acid, H3PO4, is produced by combining ground phosphate rock with sulfuric acid and involves digestion, filtration, and concentration. Phosphate-rock concentrate with a minimum P205 content of 30% is digested in sulfuric acid. The amount of sulfuric acid required is dependent on the level of impurities contained in the phosphate rock, primarily A1, Ca, F, Fe, and MgO. The resulting reaction produces phosphoric acid and suspended calcium sulfate (phosphogypsum) slurry. The slurry is then pumped through a filtration circuit where the phosphogypsum is separated and pumped along with the acidic wastewater to a collection stack. About 5 t of phosphogypsum are generated for each ton of P205 in acid produced. Federal regulations prohibit the use of phosphogypsum in wallboard manufacturing or agricultural uses because it is slightly radioactive and the material must be stored onsite in large stacks. The efficiency of the acidulation process can be improved by grinding the rock finely and recirculating the phosphoric acid to digest as much of the phosphate rock as possible. The product acid is then concentrated to 42-54% P205, and used to produce fertilizers or further purified for industrial applications (Fig. 20-4) (Mannsville Chemical Products, 2001 a).
Elemental phosphorus production Historically, the combination of high-quality ore and low-cost electricity made the WPF region favorable for elemental-phosphorus production. Since 1949, four companies operated elemental phosphorus plants in the region, producing nearly l0 Mt over the period. This represents more than 55% of total US elemental-phosphorus production for that period (US Census Bureau 1949-1995; Mannsville Chemical Products 200 lb). At the Monsanto plant, furnace-grade ore is blended at the plant and fed into a 100 m long rotary kiln, where it is heated to 1500~ The kiln uses as fuel carbon monoxide (CO)
H2SO4 ~ H20 ~
Beneficiated Phosphate
Rock
,.~ F--
Phosphoric Acid
Waste,.. ~
Phosphogypsum, scale, and water
~
Defluorination
Triple superphosphate
Ammonium phosphates
Purified acid
Fig. 22-4. Wet-process phosphoric-acid production and use.
Animal feed
Societal relevance, processing, and material flow of western phosphate
605
gas expelled from the phosphorus furnace (see below). The heating process drives off volatile organic components and thermally agglomerates the ore into nodules. The nodules are then moved to the furnace burdening area where they are combined with crushed quartzite and coke (Table 22-111) (Vranes, 1999). This mixture then is sent to one of the three electric arc furnaces. The Monsanto plant is the largest consumer of electricity in the region, annually consuming as much electricity as a city of 500,000 people (Vranes, 1999). Inside the furnace, where the temperature can reach as high as 6000~ electricity is supplied by carbon electrodes to the mixture of phosphate rock, silica, and coke (Fig. 22-5). This reaction produces phosphorus gas carbon monoxide gas, molten ferrophosphorus slag, and calcium-silicate slag (Table 22-IV). The slags are periodically removed from the furnace at separate taps located near the bottom of the furnace. The phosphorus gas then passes through a condenser to yield solid, elemental phosphorus which, is loaded into rail cars or shipping containers. Elemental phosphorus must be covered with water during transportation because it will burn on contact with air. Most of the CO is reused in the plant as fuel for the kiln (Van Wazer, 1961). Thermal-acid production is the process of burning elemental phosphorus in air in a sealed chamber and combining with water to produce a high-purity acid, which is used in chemical and food-processing applications (Fig. 22-5). TABLE 22-III Raw materials required to produce one ton of elemental phosphorus (after Hartlapp, 1968) Material Phosphate rock Coke Silica Carbon electrodes Cooling water Electricity
Quantity 9.339 t 1.551 t 1.557 t 0.017 t 30,745 L 11,830 k Wh
TABLE 22-IV Byproducts from one ton of elemental phosphorus (after Hartlapp, 1968) Byproduct
Tons
Calcium silicate slag Ferrophosphorus slag CO gas Precipitator dust
8.072 0.356 2.824 0.127
606
S.M. Jasins~"
Fig. 22-5. Diagram of an integrated plant for manufacturing phosphorus and phosphoric acid (after Waggaman and Ruhlman, 1960).
Fig. 22-6. US elemental-phosphorus production, 1948-2000. Since the early 1990s, rising energy costs combined with environmental problems associated with elemental phosphorus production (Fig. 22-6) have resulted in the closure of two plants and a shift toward using lower-cost purified wet acid as feed material for industrial phosphates. Purified acid is manufactured from wet process acid that is processed by solvent extraction to remove impurities such as F and Fe. In 1995, Rhodia Inc. closed its Silver Bow, Montana, plant and opted to purchase phosphorus from the other manufacturers (Hume et al., 2000). In 2001, Astaris LLC, a joint venture between FMC Corp. and Solutia, Inc., permanently closed its Pocatello, Idaho, phosphorus plant and opened a purified acid plant in a joint venture with Agrium USA Inc. in Soda Springs. Ownership and responsibility for environmental remediation of the Pocatello facility reverted to FMC upon closure (Jasinski, 2002).
Societal relevance, processing, and materialflow of western phosphate
607
END PRODUCTS
Fertilizer products Phosphorus, in the form of phosphate rock, is insoluble in water and not readily available to plants. The reaction of phosphate rock and sulfuric acid produces a water-soluble form of phosphorus that is readily available to plants when added to the soil. The major fertilizer products produced in the WPF are triple superphosphate (TSP), diammonium phosphate (DAP), and monoammonium phosphate (MAP). Ammonium phosphates (DAP and MAP) have become the most important multinutrient phosphate fertilizers. TSP [Ca(H2PO4)2] is a concentrated phosphate fertilizer with a P205 content of 46%. It is produced by the acidulation of phosphate rock with phosphoric acid. This results in a solidified material that is then granulated for use as a fertilizer. The water solubility of TSP is 83% (IMC Global Inc, 2002). DAP and MAP [(NH4)2HPO4], which are chemically identical, are the most widely produced phosphate fertilizer compounds. They are manufactured by reacting ammonia with phosphoric acid to the required P205 concentration. The resultant slurry is then passed through a granulator, to produce uniform size particles, which are then dried, cooled, screened and coated with a thin film to reduce clumping and ensure proper dissolution. The P205 content of DAP and MAP are 46 and 54%, respectively. Both are 90% water-soluble (expressed as a percentage of available P205) (IMC Global Inc, 2002).
Major non-fertil&er applications Although industrial and food additive applications account for less than 10% of total P205 demand in North America, their uses are essential in a variety of applications (Fig. 22-7). Thermal and purified phosphoric acids are used primarily as chemical intermediates and as acidulants in foods and beverages, primarily cola drinks. Phosphoric acid also is used directly in water treatment, metal cleaning, and detergents. Sodium and calcium phosphates, of which sodium tripolyphosphate (STPP) is the largest end use, account for more than 75% of high-purity phosphoric acid consumption. STPP is used as a detergent builder in automatic dishwashing detergents and industrial cleaners. It was formerly used in laundry detergent, but its use has been banned in North America because of eutrophication (nutrient enrichment that leads to excessive algae growth) problems associated with accumulation of excess phosphorus from wastewater treatment plants in bodies of water. Phosphorus trichloride (PCI3), produced by reacting phosphorus and chlorine, is the second largest use of elemental phosphorus. More than 60% of PCI3 is used to produce glyphosate, the active ingredient in Roundup ~ and other non-selective herbicides. Many cereal and legume crops have been genetically modified to resist glyphosate herbicide, which allows the spraying of the herbicide to control weeds without harming the plant. This has contributed to the growth in the use of the product. Other significant chemicals
608
S.M. Jasinski
Sodium phophates Calcium phosphates Food & beverages Water treatment Metal treatment
J--
Matches Flares Fireworks Deoxidizer Copper alloys Phosphor bronze
Elemental Phosphorus
Herbicides Pesticides Intermediates r- Flame retardants Plasticizers Phosphorus acid
~1~
Flame retardants
Phosphorus[_____~ Plasticizers oxychloride I v
"
Industrial cleaners Water treatment Food additives
Pesticides Surfactants
Lubricant additives Pesticides Ore flotation
Fig. 22-7. Major applications for elemental phosphorus and compounds (Rhodia, 2002). are phosphorus pentasulfide, phosphorus oxychloride, and dicalcium phosphate (Mannsville Chemical Products, 2001 b).
MATERIAL FLOW IN THE ENVIRONMENT Phosphorus and its compounds are used in applications that are dissipative in the environment. Phosphorus is added to the soil through mechanisms such as release from soil reserves, crop residues, manure, phosphate rock, and significantly, mineral fertilizers. Most of the phosphorus contained in fertilizer is utilized by plants and then dissipated through the food chain (Fig. 22-8). The amount taken up from crops depends on the type of crop and the composition of the soil. Modern farming practices, such as fertilizer application rates based in soil analysis and no-till farming, have greatly reduced the overfertilization of crops and runoff losses of phosphorus from farmland (Laegreid et al., 1999). Phosphorus contained in non-fertilizer applications returns to the environment
Societal relevance, processing, and material flow of western phosphate
609
Fig. 22-8. General phosphorus flow in the environment (adapted from Matzner, 1996). through sewage treatment facilities, chemical applications, wastewater, and other similar paths. The amount of phosphorus contained in wastewater has been greatly reduced in the past 25 years through the elimination of phosphates from laundry detergent. Extensive use of phosphates in many industries makes it difficult to quantify the amount of phosphorus that cycles through the environment in a particular year. Phosphate fertilizers and chemical derivates are essential components of society. Increasing human population growth requires sustained crop and livestock production, which insures the long-term need for phosphate fertilizers. Likewise, phosphate compounds have many important industrial applications, for which there are no competitive substitutes.
610
S.M. Jasinski
ACKNOWLEDGMENTS I thank the reviewers for the helpful comments and suggestions on this chapter.
REFERENCES Hartlapp, G., 1968. Phosphoric acid by the furnace process. In: A.V. Slack (ed.), Chapter 12 Phosphoric Acid, vol. 1, Part 2, Dekker, New York, NY, pp. 927-982. Hume, C., Schmidt, B. and Alperowicz, N., 2000. Phosphates new line-up. Chemical Week, August, 2: 20-27. IMC Global, Inc., 2002. Fertilizer fact sheets. Accessed at http://www.imcglobal.com/products/ phosphates.htm. Jasinski, S.M., 2002. Phosphate Rock 2001 -Annual Review. US Geological Survey, Mineral Industry Surveys, August, 12 pp. Krauss, U.H., Saam, H.G. and Schmidt, H.W., 1984. International strategic minerals inventory summary report- phosphate. US Geological Survey, Circular, 930-C, 41 pp. Laegreid, M., Bockman, O.C. and Kaarstad, O., 1999. Agriculture, Fertilizers and the Environment. CABI, New York, pp. 150-157. Li, T., 1978. Southeastern Idaho phosphate mining: how an environmental impact statement distorts growth plans. Mining Eng., 30: 25-36. Mannsville Chemical Products Corp., 2001a. Phosphoric acid (wet process). Adams, NY, 2 pp. Mannsville Chemical Products Corp., 200lb. Phosphorus. Adams, NY, 2 pp. Matzner, E.A., 1996. The effect of regulation on trends in phosphorus. In: Sulphur Markets, Today and Tomorrow. The Sulphur Institute, Washington, DC, 25 pp. Rhodia Phosphorus and Performance Derivatives, 2002. Product information sheet, http:// rhodia-ppd, com/prod_in fo. asp. Service, A.L. and Popoff, C.C., 1964. An evaluation of the Western Phosphate Industry and its resources: Part 1, Introductory review. US Bureau of Mines, Report of Investigations, 6485, 86 pp. US Census Bureau, 1949-1995. Annual report on inorganic chemicals. US Census Bureau, report MA28A. U.S. Geological Survey, 2002. The mineral industry of Idaho, 2000. US Geological Survey, Minerals Yearbook, vol. II, Area reports- Domestic, pp. 141-145. Van Wazer, J.R. (ed.), 1961. Phosphorus and its Compounds- vol. 2, Technology, Biological Functions, and Application. lnterscience Publishers, New York, pp. 1149-1219. Vranes, R., 1999. Information technology aids Soda Springs phosphate mine. Mining Eng., 51: 15-20. Waggaman, W.H. and Ruhlman, E.R., 1960. Phosphate rock, Part 2, Processing and utilization. US Bureau of Mines, Information Circular, 7951, 36 pp.
Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment Edited by James R. Hein Handbook of Exploration and Environmental Geochemistry, Vol. 8 (M. Hale, Series Editor) Published by Elsevier B.V.
611
README
The CD accompanying this book is in a hybrid format that combines both ISO 9660 and Mac OS Standard (JFS) formats on a single CD and should be readable by most personal computers. This CD contains appendices for five chapters of "Life Cycle of the Phosphoria Formation: From Deposition to Post-Mining Environment":
APPENDIX 1 Global sedimentary and sediment-hosted phosphate deposits is a table of 1384 phosphate deposits to accompany Chapter 20, "Review of world sedimentary phosphate deposits and occurrences" by Orris and Chernoff. The deposit and occurrence listing is presented in two formats. The table is available as a pdf file called "Appendix 1.pdf". The data table is also available as a Microsoft Excel 2001 workbook labeled "Appendix l.xcl". References cited in the table are in a third file in pdf format, "Appendix 1 Refs. pdf".
APPENDIX 2 Appendix 2 consists of three quality assurance files for Chapter 18, "Selenium and other trace elements in water, sediment, aquatic plants, aquatic invertebrates, and fish from streams in SE Idaho near phosphate mining" by Hamilton, Buhl, and Lamothe. The files are described below and are in pdf format. The files are labeled "Appendix2TableA.pdf", "Appendix2TableB.pdf", and "Appendix2TableC.pdf". Table A. Quality assurance and quality control measures of Se analyses of water, sediment, aquatic plants, aquatic invertebrates, and fish; n = 1 for water, sediment, aquatic plants, and aquatic invertebrates; n = 2 for fish; (mean and standard error in parentheses). Table B. Quality assurance and quality control measures of analyses of elements in water (W), sediment (S), aquatic plants (P), aquatic invertebrates (I), and fish (F); n = 1 for water, sediment, aquatic plants, and aquatic invertebrates; n = 2 for fish; (mean and standard error in parentheses). Table C. Quality assurance and quality control measures of analyses of elements in sediment.
APPENDIX 3 Comparison of concentration histogram and log-scaled concentration histogram data for all channel samples of the Meade Peak. Appendix 3 is a pdf file, "Appendix3.pdf", that
612
Readme
presents support graphs for Chapter 12, "Lithogeochemistry of the Meade Peak Phosphatic Shale Member of the Phosphoria Formation, southeast Idaho" by Herring and Grauch.
APPENDIX 4 Comparison of concentration histogram and log-scaled concentration histogram data for all leachate samples of the Meade Peak. Appendix 4 is a pdf file, "Appendix4.pdf", that presents support graphs for Chapter 13, "Rock leachate geochemistry of Meade Peak Phosphatic Shale Member of the Phosphoria Formation, southeast Idaho" by J.R. Herring.
APPENDIX 5 Statistics on Se and other trace elements in above-ground plant oven-dried tissue. Appendix 5 is a pdf file, "Appendix5.pdf", that presents statistics for Chapter 19, "Uptake of selenium and other contaminant elements into plants and implications for grazing animals in southeast Idaho", by Mackowiak, Amacher, Hall, and Herring.
613
AUTHOR INDEX*
Adams, 513 Ahr, 31, 3 7 Alaimo, 501,520 Allan, 502, 520 Allison, 95, 105 Allmendinger, 143-7, 151-2, 154-5, 162 Altschuler, 171,185, 234, 240, 243,248, 356, 364, 586, 594 Amacher, 230, 248, 394, 467, 469, 472, 481, 484, 522, 527-8, 533, 538-9, 552-3 Amer, 547, 552 Amiro, 546, 552 Ammar, 546, 552 Angevine, 141, 163 Arbogast, 328, 364, 405, 426, 472,481,533, 552
Armstrong, 138, 144-5, 147, 155, 163, 192--3, 219, 223, 586, 594 Arthur, 564, 5 71 Atnipp, 192, 223 Baedecker, 80, 105, 328, 364, 405,426 Balistrieri, 270, 293, 478, 481,539, 550, 552 Ball, 189 Banks, 243,249 Bafiuelos, 530, 539, 550, 552 Barron, 6-7, 11, 13, 567, 572 Bartels, 562, 565,571 Basu, 12, 16 Baturin, 237, 248 Baud' 7, 13, 112-13, 126, 128, 130-2, 132 Bauer, 273 Baumgartner, 127 Beath, 305-7, 314, 438, 465, 530, 535,539, 542, 546-8, 552 Beauchamp, 7, 13, 112-13, 126, 128-32, 132 Becker, 12-13, 13
*Page numbers in italics refer to the reference list
Beckett, 543, 545,553 Behnken, 87, 103, 105 Belasky, 7, 10, 13-14 Bell, 530, 552 Benmore, 99, 105 Benninger, 564, 567, 571 Bentor, 32, 37, 77, 105 Berger, 76, 105, 583,594 Berner, 6, 14 Bethke, 242, 248 Birch, 564, 572 Bird, 299, 315,316 Blackstone, 191, 221,223 Blackwelder, 21-2, 27, 3 7 Blakey, 11 Blank, 192 Bodkin, 80, 105 Bouma, 123, 132 Bowring, 13, 14 Boyer, 150-1,163 Boyle, 85, 105, 213,216, 223 Branson, 23, 3 7 Breger, 21, 37 Breheret, 90, 105 Brewer, 141,163 Briggs, 79, 105, 260, 293, 328, 364, 370-1,397 Brittenham, 29, 37, 323,364, 406, 426 Brix, 513,520 Broecker, 77, 581,583,594 Brown, 472, 481 Bruland' 85, 105, 300-1,305,316 Brumsack, 85, 90, 105 Brune, 546, 552 Bryson, 502, 520 Buchanan, 444, 463 Budahn, 227, 232, 248, 250 Budd, 31-2, 3 7
614
Buhl, 483 Bunker, 257, 295 Burchette, 31, 37, 75, 79 Burchfiel, 11, 14, 586, 594 Burkhard, 157, 163 Burnett, 228, 234, 242, 248, 583,585,594 Byers, 31, 36, 38, 85, 98, 106 Calvert, 34, 3 7, 97, 108 Campbell, 27, 37, 508, 520 Carillo, 55, 60 Carroll, 32-3, 37, 41, 76-7, 86, 105, 578-9, 598
Carswell, 228, 240, 248, 354 Cathcart, 592-3,594 Causey, 45, 308, 316, 528, 552 Chafetz, 204, 223 Chai, 237, 248 Chambers, 528, 550, 552 Chano, 550 Chao, 252, 270, 288, 293-4, 478, 481,539, 552
Chavez, 583-4, 594 Cheney, 52, 323,365, 406, 426 Chernoff, 559, 567-9, 571 -2 Cherry, 502, 521 Christie, 359, 364 Christie-Blick, 146, 165 Cita, 97, 109 Clark, 216, 223 Clark, 31, 35, 37, 85, 106 Clarke, 578, 595 Claypool, 140-1, 146, 163, 192, 223, 237, 248, 300, 308, 316, 344, 364 Clements, 494, 522 Coalson, 35, 76, 107 Codispoti, 75, 106, 305,316, 582, 584, 595 Coffman, 56, 58, 69, 71, 60 Cohen, 329, 333,364 Coleman, 264, 294 Collier, 85, 91,106 Collinson, 31-2, 34, 74, 110, 113-14, 128, 131,135, 401,426, 576, 584, 598 Condit, 23, 38 Coogan, 146, 150-1, 163 Cook, 12, 14, 77, 106, 190, 205,223, 236, 248, 302, 314, 316, 562-8, 571 Copin-Montegut, 582, 595 Couri, 546, 552 Courtin, 546, 552 Craddock, 146-7, 150, 157-8, 163 Cremer, 80, 106
Author index
Cressman, 27, 29, 38, 86, 106, 122, 132, 140, 145, 151,155, 163, 236, 248, 336-8, 364, 401,426 Cronquist, 531,552 Crosby, 147, 151,163 Crowley, 6, 14, 66, 60 Cullen, 564, 571 Cuneo, 86, 107 Curry, 79, 106, 260, 294 Cutter, 258, 294-5, 300-1,305,316, 446, 463-4
Dahl, 33, 38, 578, 595 Dallinger, 500, 502, 521 d'Angelo, 371,396 Darby, 31, 38 Davis, 151,163, 173, 185 Dayton, 130, 132 de Baar, 95, 97, 106, 580, 595 De Bruin, 191, 221,223 de la Cruz, 502, 523 DeCelles, 141, 145-6, 164 Deer, 290, 294 DeForest, 513,521 Deiss, 24, 38 Delaney, 34, 38 DeLaubenfels, 122 Delevaux, 264, 294 Denison, 9-10, 14 Dequincey, 245,248 Desborough, 141, 143, 157, 164, 185, 189, 192, 197, 204, 223, 264, 290-1,370, 383, 392, 396, 396, 484, 521,586, 595 Detwiler, 299, 312, 316 Devi, 502, 521 Dickinson, 546, 554 Dickson, 9, 14, 228, 248 Dijkshoorn, 544, 552 Dixon, 145-7, 149-51, 154, 158-9, 164
Doerner, 242, 248 Dorr, 146, 166 Dumitrica, 117, 132 Dzombak, 478, 481 Eary, 544, 553 Edman, 140-2, 147, 164 Edmond, 91,106 Ehrenberg, 112, 128, 132 Eisler, 501,521 Ekart, 6, 14 Elderfield, 580, 595 Elliott, 150, 163
Author index
Embry, 542, 554 Emerson, 91,106, 585,595 Engberg, 468, 481 Erwin, 3, 7, 12, 14 Evans, 137, 146, 150, 164, 192-3,220, 223, 399, 586, 593,595 Fan, 301-2, 305,312, 316 Fawcett, 6-7, 11, 13 Feist, 530, 553 Ficklin, 371 Fiesinger, 193,223 Filby, 220, 223, 290, 294 Filippelli, 34, 38, 77, 106, 583,595 Finks, 127, 129-30, 132-3 Finley, 505, 521 Fischer, 12, 14 Fisher, 242, 248 Fishman, 445,464 Fleet, 171,186 Foda, 502, 521 Foster, 189, 192, 234, 240, 248, 251,356, 458 Franke, 546-7, 553 Franks, 85, 105 Friedlander, 231,249 Friedman, 445, 464 Froelich, 33, 38, 583,585,594 Fryberger, !92, 219, 224 Furr, 508, 521 Gaffin, 9, 14 Gala, 468, 482 Gale, 21, 38, 220, 224 Gallimore, 6, 15 Gammon, 129, 133 Gao, 400, 425, 426 Gardner, 24, 38 Garrels, 360, 364 Garson, 566-8, 5 72 George, 257, 294 Geptner, 204, 224 German, 97, 106, 585, 595 Geslin, 75, 106, 578, 595 Ghahremani, 243,249 Ghosh, 6, 14 Gieskes, 585,598 Giesy, 507, 525 Giordano, 290, 292, 294 Girty, 21 Glenn, 564, 571 Goldberg, 578, 595 Golonka, 31, 38 Gough, 529, 553
615
Govett, 417, 426 Grantz, 126, 133 Grauch, 36, 162, 171-2, 181,185, 189-90, 192, 195,210, 213,224, 227, 229, 233-4, 237, 249, 252, 291,294, 308, 321,324, 328, 340, 343,356, 358, 364, 368-9, 371-3, 379-80, 415, 424, 428, 434, 542 Gromet, 332-3,364 Gronbaek, 548, 553 Gruner, 171,185 Guerinot, 546, 553 Gulbrandsen, 77, 106, 172, 180, 182-3, 186, 190, 224, 228, 249, 253,294, 332, 336-8, 344, 360, 363,364, 578, 596, 588, 590, 592, 595
Gunter, 169, 171,186, 190, 198, 204, 224, 337, 339, 369, 379, 397, 579, 586, 591-2, 595, 602 Gurr, 562, 565,571 Guthrie, 502, 521 Haag, 9, 12, 14 Hageman, 79, 106, 260, 294, 370-1,396-7 Hagiwara, 215,224 Hall, 527 Hallam, 12, 17 Hamilton, 190, 299, 301,305,307, 316, 321, 437-9, 462, 464, 483,493-5, 504-5, 508, 513-14, 516, 521-2 Hansen, 430-1,434, 468, 481,536-7, 553 Harben, 564-5,572 Hardie, 9, 14 Hardy, 437 Harland, 4, 14 Harman, 173, 186 Harper, 451,464 Harr, 547, 553 Harris, 141, 143, 164 Hartlapp, 605, 610 Hartman, 130, 133 Harvey, 33, 38 Harwood, 115, 124, 133 Hatcher, 6, 14 Hayden, 48 Haygarth, 302, 316 Hein, 3, 19, 114, 128, 131,133, 399, 401,528, 535,553, 562, 564, 567, 571-2, 576-7, 586, 589, 595 Heller, 9, 12, 14, 144-5, 164 Henderson, 5, 15 Hendrix, 31, 36, 38, 85, 87, 98, 106 Herdt, 547, 555 Herigstad, 546, 553 Herman, 242, 249
616
Herring, 12, 15, 140, 164, 170-1,186, 189-90, 195,210, 213,228-9, 230, 242, 249, 252, 308, 321,324, 329, 339, 364-5, 367-73, 379-80, 397, 415, 424, 427-8, 434, 461, 469, 479, 484, 503, 522, 527, 533, 542, 553, 567, 572, 588, 595-6 Hess, 190, 225 Hiatt, 31-2, 35, 38, 87, 91, 98-9, 102-3, 106, 114-15, 128, 132, 133, 205, 225, 584, 596 Hills, 151,164 Hilton, 513,522 Hite, 33, 38 Hofmann, 217, 222, 225, 494, 521 Holser, 9, 12, 15 Horita, 9, 15 Horowitz, 446, 464 Hoskins, 242, 248 Hothem, 301,307, 317 Hotinski, 9, 12, 15 Howes, 57, 60 Hubbert, 146, 166 Huebner, 299, 437 Hume, 606, 610 Hunn, 514, 522 Hutchison, 544, 553 ljima, 130, 133 Imbrie, 406, 426 Inden, 35, 38, 76, 107 lngall, 583,596 Isaacs, 75, 108, 301,306, 317, 337, 344, 365, 425,426, 579, 597 lsbell, 86, 107 lsozaki, 9, 15 lvanovich, 242, 246, 249-50 lvanovskaya, 204, 224 Jackson, 79, 107, 405,426 Jacobs, 585,596 Jahnke, 33, 583,596 Janecker, 145, 166 Jarvis, 228, 249 Jasinski, 20-1, 45, 54, 60, 60, 190, 308, 483, 528, 586-7, 599-600, 606, 610 Jerden, 240, 249 John, 255 Johnson, 189, 198, 428, 435 Jones, 13, 15, 48-9, 60, 112-13, 116-19, 124-6, 130-1,133-4, 401,426, 533,553 Jordan, 145 Kabata-Pendias, 542, 545,553 Kakuwa, 9, 15 Kato, 7, 15, 130, 133
Author index
Kautzky, 500, 502, 521 Kazakov, 19, 22-4, 28, 39 Keller, 401,426 Kellogg, 145, 155, 164, 193,219, 225 Kennedy, 444, 464 Kennett, 7, 15 Ketner, 115, 122, 124, 133 Kidder, 584, 596 Kiffney, 494, 522 King, 457, 464 Kirby, 500, 522 Kirkham, 592, 596 Klemme, 302, 313-15, 316 Klovan, 406, 426 Knauer, 85, 91,107 Knight, 507, 513-14, 516, 523 Knoll, 13, 15 Knudsen, 87, 107, 169, 171,186, 190, 198, 204, 337, 339, 369, 379, 397, 579, 586, 591-2, 596, 602 Koelmel, 192, 219, 225 Koenig, 21 Koepnick, 9-10, 14 Kokaly, 196 Kolodny, 32-3, 40, 76, 107, 583, 585, 596-7 Krauskopf, 41 i, 426 Krauss, 565, 5 72, 60 I-2, 610 Krieg, 247, 249 Krier, 344, 360, 363,578, 586, 590, 592 Kuniyoshi, 33, 39 Kutzbach, 6-7, 15, 31, 39 Kuzvart, 564-5, 572 Labrenz, 218, 225 Laegreid, 608, 610 Lageson, 152, 164 Lamb, 541,550, 553 Lamerson, 138, 165 Lamothe, 260, 294, 427, 431,434, 437, 446, 464, 469, 479, 483,533,553 Landing, 585,596 Langford, 6, 16 Langmuir, 237, 242, 249 Latham, 245,249 Leanza, 562, 572 Lee, 45, 50, 62, 60, 228,250, 507, 522 Leeman, 146, 165 Lehr, 171, 183, 186 Lemly, 302, 316, 438, 464, 468, 481,495, 500, 504, 507-8, 513-17, 519-20, 521-3, 537, 553 Levinson, 228, 250 Levy, 146, 165 Lewan, 232, 234 Lewis, 585,596
Author index
Li, 603,610 Lin, 468, 481 Lindsay, 545,553 Link, 33-4, 40, 74, 108, 131,134, 583-5,596-7 Litchie, 79, 107 Littke, 79, 107 Llewellyn, 55, 60 Lo, 9, 13, 15 Loebner, 564, 5 72 Loose, 192, 221,225 Losi, 544, 553 Love, 562, 572 Lowe, 301,307 Lowenstein, 9, 15 Lowers, 189 Lucas, 180, 183-4, 186, 591,596 Luckey, 303,319 Luoma, 301,306-7, 309-10, 316, 439, 455, 457, 462, 464, 500, 523 Mabie, 190, 225 Macintyre, 126-7, 129, 133 Mackenzie, 360, 364 Mackowiak, 312, 433,439, 469, 479-80, 527, 589 MacNicol, 543,545,553 Magaritz, 9, 12, 15 Magoon, 315 Mahan, 542, 546, 554 Maier, 507, 513-14, 516, 523 Mann, 502 Manning, 277, 294 Mansfield, 20-1, 23, 39-40, 52, 60-1, 139, 145-6, 148, 151, 157, 165, 401,426, 590, 592 Marschner, 544, 546, 554 Martens, 252, 270, 294, 468, 478-9, 481 Martin, 85, 91,107, 484, 524 Martindale, 31-2, 39 Masscheleyn, 468, 482 Matzner, 609, 610 Maughan, 30, 31, 33, 39, 74-5, 87, 89, 107, 577-8 Mayland, 529-30, 554 McArthur, 33, 39, 99, 107, 182-3, 186 McClellan, 171, 182-3,186 McConnell, 171,185 McGrath, 535,554 Mclntyre, 19, 399 McKelvey, 12, 15, 19, 23-6, 29, 36, 39-40, 74-5, 86-7, 89, 98, 107, 119, 131,133, 140, 165, 190, 225, 227-8, 240, 250, 300-1,308, 317, 323,343-5,365, 401,406, 426, 458, 464, 576, 584, 586, 597
617
McManus, 33, 38 Medrano, 76, 78, 80, 87, 90-1, 97, 101,107, 237, 252, 290, 292, 294-5, 332, 578, 597 Meeker, 189, 224 Mei, 5, 9, 15 Meier, 79, 105 Meir, 260, 293 Metcalfe, 3, 16 Meyer, 446, 464 Miesch, 329, 365 Miller, 11, 15, 542, 546, 554 Mitchell, 566-8, 572 Mitchum, 35, 40 Mitra, 142-3, 146, 150, 152, 157, 160, 165-6 Momper, 192, 219, 225 Monley, 146, 165 Montgomery, 52, 60, 406, 426 Montgomery Watson, 301,309-10, 317, 323,365, 447-8, 458, 464, 483-4, 494, 500, 503,506, 508, 513-15, 523 Moore, 114-15, 123, 125, 133, 252, 295, 433, 435, 495,523 Morel, 478, 481 Morgan, 161, 166, 193, 219, 225, 472, 482 Mosier, 563,572, 576, 597 Moxon, 542, 546, 553-4 Moyle, 11, 34, 308, 316, 528,552, 575,602 Miiller-Karger, 76, 108, 583,597 Munkers, 192,225 Murata, 29, 40, 253,295, 401,426 Murchey, 111-19, 123-31,133, 401, 426, 584 Murray, 91, 108 Musashi, 9-10, 16 Muth, 547, 553 Nathan, 33, 40, 183 Newport, 117, 134 Ney, 507, 523 Nielson, 33, 40 Noriki, 76, 109, 583,598 Notholt, 302, 313-15,317, 565, 5 72 Oberlindacher, 148, 323,365, 406, 426
O'Dell, 441,444, 446, 448-9, 453, 465
Odum, 502, 523 Ohlendorf, 301,303,307, 317, 468, 482, 516 Olson, 501,523 Olson, 542, 547, 554 Oriel, 138, 140, 145, 147-8, 159-60, 163, 165, 192-3,219, 225, 586, 594 Orris, 559, 5 71
618
Osmond, 242, 246, 250 Osmundson, 516 Osweiller, 547, 554 Paces, 227 Palmer, 581, 597 Pan, 171,186 Papp, 79, 108, 260, 295 Pardee, 21-2, 40, 131,134 Parker, 530, 553 Parrish, 5, 7, 11, 16, 567, 569, 572 Parrish, 6, 16 Paul, 143, 157, 165 Paul, 157, 165 Peak, 539, 550, 554 Peale, 48, 60 Pedersen, 97, 108 Peng, 581,583,594 Perkins, 19, 73, 128, 192, 225, 234, 240, 251, 356, 399, 401,458 Pessagno, 117, 134 Peterson, 30-1, 33-5, 40, 74-6, 87, 108, 430, 435
Petrocelli, 514-15,524 Pfaltz, 273 Pickering, 252, 268, 295, 468, 482 Pidcock, 48 Pierce, 161,166, 193,219, 225 Pilon-Smits, 468, 482, 536, 554 Piper, 11-12, 16, 25, 32-4, 40, 73-8, 80, 87, 90-1, 97, 101,108, 128, 131,134, 237, 250, 252-3,260, 281,290-1,294-5, 299-301, 306, 308-10, 312, 317, 332, 337, 339, 352, 365, 399, 401,425,426, 428, 434, 437-9, 456, 458, 463,465, 484, 524, 529, 554, 564, 572, 575-6, 578-81,583-5, 597, 602 Platt, 140, 148, 159-60, 165 Platts, 484, 524 Poole, 76, 108, 579, 597 Popoff, 45, 54-5, 66, 60-1,600, 610 Powers, 221,225 Presser, 190, 252, 299-301,303, 305-7, 309-10, 315,317, 437-9, 455,457, 462-3,465, 486, 494-5, 516, 524 Protzman, 157, 166 Puls, 300-1,305,318, 542, 546-8, 554
Punshon, 546, 554 Pye, 87, 109 Radtke, 445,465 Rai, 544, 553 Ramseyer, 204, 225 Rand, 514-15,524
Author index
Rantz, 444, 465 Raup, 12, 16 Redente, 530, 539, 555 Redfield, 76, 108, 582, 597 Reichow, 9, 13, 16 Reid, 130, 134, 204, 223 Renne, 12, 16 Ressler, 257, 295 Retallack, 12, 16 Rich, 151,166 Richards, 20-1, 40, 52, 61-1, 145, 166, 401, 426
Richter, 48, 55 Rigby, 126-7, 129, 134 Riggs, 562, 564, 566, 572 Roberson, 550, 554 Roberts, 114-15, 123, 134 Rodgers, 145, 166 Rosenfeld, 530, 535,539, 542, 546-8, 554
Ross, 4, 16, 35, 41, 87, 98, 103, 108-9, 567-70, 573 Rossignol-Strick, 97, 109 Royse, 138, 146-7, 150-1,166, 192, 225 Rubey, 146, 166 Ruby, 24 Ruhlman, 606, 610 Riitzler, 126-7, 129, 134 Ryan, 97, 109 Rye, 208 Saiki, 301,307, 318, 501,524 Sakai, 33, 39 Sanford, 198, 201 Santolo, 468, 482 Sanzolone, 252 Sarnthein, 76, 109 Sayers, 257 Schmitt, 146, 152, 164, 166 Schnoor, 546, 554 Scholle, 3, 9-10, 16 Schrauzer, 546, 554 Schuffert, 76, 109, 172-3, 181,186, 583-5, 597 Schulze, 130, 134 Schwarcz, 245,249 Scotese, 567-70, 573 Scotese, 6, 16 Seelye, 500, 524 Seiler, 311, 315, 318 Sepkoski, 12, 16 Service, 45, 54-6, 58, 61, 62, 66, 71, 592, 597, 600, 610 Shannon, 273
Author index
Sheldon, 12, 16, 27-8, 32, 34, 41, 77-8, 89, 98, 109, 113, 128, 131,134, 192, 226, 228, 240, 253, 295, 300, 318, 354, 356, 365, 401, 426, 566, 573, 576, 584, 586, 592-3,598 Shergold, 302, 314-15, 316 Shi, 3, 16 Sholkovitz, 97, 102, 109, 585, 598 Siems, 329, 365 Silberling, 112, 115, 124, 134 Sillitoe, 216, 223 Simmons, 227 Sinha, 240, 249 Skorupa, 299, 301-3, 305-7, 309, 311-12, 318, 438, 465, 515 Smart, 77, 109 Smith, 371,397 Smith, 513, 519, 523 Somers, 444 Sparks, 539, 550, 554 Spinosa, 577 Stadtman, 300, 303, 318 Stanley, 12, 16, 301, 318 Stark, 530, 539, 555 Steidtman, 146, 166 Stephan, 515,524 Stephens, 33, 41,495, 500, 524, 578-9, 598 Steuber, 9, 16 Stevens, 12, 17, 114-16, 135 Stewart, 115, 135 Stillings, 394, 428-30, 435, 467, 504, 537,539 Stowasser, 56, 69, 61 Stowe, 547, 555 Strauss, 9-10, 17 Stumm, 472, 482 Su, 550, 555 Suarez, 252, 270, 294, 468, 478-9, 481,550, 555
Suess, 76, 98, 104, 109 Summerhayes, 567-70, 5 73 Suppe, 151,166 Surdam, 140-2, 147, 164 Suttle, 548, 555 Swan, 548, 555 Swanson, 29, 38, 86, 106, 122, 132, 227-8, 236, 240, 248, 336-8, 364, 401,426 Sylvester, 516 Taggart, 79, 109 Taylor, 172, 187 Teal, 502, 524 Thompson, 227-8 Thorlacius-Ussing, 548, 553 Thornburg, 35, 41
619
Thunell, 76, 109, 583, 598 Thurow, 514, 524 Tisoncik, 30-1, 41, 74, 109 Toggweiler, 583-4, 594 Tokunaga, 252, 295, 468, 482 Torres, 90, 109 Trappe, 314-15 Trelease, 305-7, 314, 318, 438, 465 Tsoar, 87, 109 Tsunogai, 76, 109, 583,598 Turekian, 85, 109, 332, 365 Turner, 142, 157, 166 Twitchett, 12-13, 17 Tysdal, 171-2, 187, 189-90, 226, 324, 366, 428, 435, 588, 598 Ulmishek, 302, 313-14, 316 Uren, 545,555 Vail, 35, 41 van der Pluijm, 158, 163 Van Hassel, 507, 523 Van Wagoner, 35, 41 Van Wazer, 605, 610 Vance, 528, 535, 555 Veeh, 228, 234, 248 Veizer, 9, 16 Velinsky, 258, 295 Venugopal, 303,319 Verhoogen, 142, 166 Von Bockel, 583, 598 Vranes, 48, 61,602, 605, 610 Waggaman, 606, 610 Wai, 544 Ward, 548, 555 Wardlaw, 4, 10, 12, 17, 30-2, 34, 41, 74, 110, 113-14, 124, 128, 131,133, 135, 401, 426, 576, 579, 584, 598 Warner, 192, 226 Webel, 221,226 Webster, 84, 110 Wedepohl, 85, 109, 332, 365-6, 579, 598 Welsch, 79, 106, 260, 294 Weres, 430, 433,435, 468, 478-9, 481
Whalen, 31, 35, 41, 85, 87, 110 Wheller, 228, 248 White, 273,295 Whiteford, 114-15, 135 Wiedmeyer, 514, 521 Wiener, 507, 525 Wignall, 12-13, 17 Wilde, 445,465, 577, 598
620 Williams, 192, 219, 225, 542, 546, 554 Wiltschko, 146, 166 Wojtal, 152, 166 Woock, 495,525 Wood, 290, 295 Worsely, 584, 596 Wright, 31, 37, 75, 79 Wyrtki, 584, 598 Yao, 400, 425,426 Yin, 3, 17 Yochelson, 29, 41, 113, 122, 135
Author index
Yonkee, 142-3, 145-6, 150, 160, 165
Zavarin, 433, 435 Zawislanski, 433,435 Zhang, 9, 17, 252, 295, 433,435 Zhou, 288, 293 Ziegler, 5-7, 17, 31, 39 Zielinski, 189, 196, 212, 227, 232, 250, 281,292, 354 Zieve, 430, 435
621
SUBJECT INDEX
Absaroka plates, 138, 142, 146, 150, 158 Acidified leachate samples, 371 Acid-volatile organics, 277 Active margin basin and epicontinental sea deposits, 563 Active phosphate-rock mines in Western Phosphate Field, 600 Aerosols, 13 Agitation of sediment, 98 Agrium USA Inc., 606 Agropyron smithii, 539, 550 Albaillellacids, 116 Albite, i 71, 197 Alfalfa, 530, 539, 544, 551 "Ranger", 550 Alkali disease, 547 Alumina-poor intervals, 89 Aluminium, 337, 379 American coot, 311 American Public Health Association, 488 Ammonitico Rosso ad Aptici of the Trento Plateau in Italy, 127 Anaconda Co. fertilizer plant, 53-4 Anadiaene spicules, 129 Ancestral Uncompahgre uplift, 31 Angayucham terrane, 126 Angus Creek, 484 Animal-feed supplement, 599 Anoxic environment, 394 Antimony (Sb), 338 Antiquities Act of 1906, 49 Antler or Humboldt highland, 75, 111, 114, 122, 128 Apatite, 96, 104, 251,261,265,277, 292, 602 Appalachian-Mauretanide-Variscan orogenic belts, 6 Aquatic birds, deformity in, 437 Aquatic ecosystem components and concentrations posing various hazards, 517 Aquatic hazard assessment of Se, protocol for, 515
Aquatic invertebrates, 488 comparison to other Idaho data, 506 other elements, 505 selenium, 504 Aquatic life, guidelines for risk, 301 Aquatic plants, 501-3 comparison to other Idaho data, 503 other elements, 501 selenium, 501 submerged, 488 Aragonite ooids, formation of, 23 Arctic Canada, 112 Arsenic (As), 79, 189, 190, 338, 358, 379, 425, 428 Asia and the Pacific, 315 Astaris LLC, 599, 606 Aster, 530 Astragalus, 530 Atmospheric oxygen, 6 Atmospheric samples, near-surface, gaseous selenium and other elements in, 427 Atriplex, 530 Authigenic/diagenetic assemblage, 201 Ba/AI20.~ and Sc/A1203 ratios, plots of, 88 Background water, 475 Backscattered electron (BSE) intensity, 260 Baedecker and Arbogast, analytical techniques, 328 Bannock thrust zone, 145, 154 Barium, 80 Barrier reef, 126 Basaltic flows, 193 Battle Mountain Mining District, 122 Bear River, 444 Beauchamp and Baud, integrated biologic and oceanographic model of, 131-2 Beauchamp, paleobathymetric model of, 129 Bechler Conglomerate, 145 Bedded phosphorites of the Phosphoria primary sedimentary origin, 21 Bedding, chaotic, 89
622 Bedding-plane-parallel faulting, 193 Benthic invertebrates, 506 Benthic macroinvertebrates, 310 Big Horn Basin, western Wyoming, 28 Bioaccumulation, 310, 439, 463,508 efficiency of, 315 Bioavailability, 529, 544 of Se in soils and sediments, 252 of trace elements, 528 methods to study, 431-3 results, 533-50 Bioclasts, 89 Biogenic (coralline-type) carbonates, development of, 22 Biogenic debris, 578, 583 Biologic productivity, 567 Biological reactions and selenium concentrations in biota, 310 Biomarker gammacerane, 33 Biomass, 502 Biosilica (opal), transformation to quartz, 401 Biotite, 142 Birds, 440 Birotule spicules, 129 Bitumen, 212 Black chert, 122 Black shale-hosted phosphorite of the Phosphoria Formation, 467 Black shales, 252, 300 Black spiculitic chert, 128 Blackfoot Reservoir watershed discharges for 2001 and 2002, 437 Blackfoot River gaging station, 453,462 quality assurance and quality control, 446 suspended-sediment concentrations, 455 Blackfoot River watershed, 439, 441-2,457, 484 ecological impacts in, 461 geohydrologic balance, 457 Idaho, 303 Regional Selenium Reservoir, 457-61 selenium loading, methods, 444-6 site location and description, 441-4 and tributary watersheds, modeling the, 463 Blacknose dace, 507 Blackwelder model for global phosphorus cycling, 22 Blackwelder's mechanism, 23 Bluegill, 508 Bone Valley Formation, Florida, 565 Boudinage, 143 Brassica species, 530 Bromus ineris, 539
Subject index Brook trout, 488 Buddingtonite (NH4A1Si308), 171,204, 212 Bulk chemistry, 382 of rocks, 368 Cadmium (Cd), 91, 189-90, 251, 281-2, 291, 358, 373,425, 428, 542, 549, 585 concentrations, 282 CaHCO3-type water, 473 Calcite, 171,205, 261 e-lamellae, 158 Calcium carbonate, precipitation of, 22 California, 118, 468 margin, 11 Cambodia, provinces of Battambang and Kampot, 565 Canada, 46 Capping with soils, 549 Carbon (C), 79 total, 329 organic carbon, 78, 96, 98, 102 Carbon dioxide levels, 6 poisoning, 13 Carbonate, 183,252, 259, 277 banks, 29 C, 79 content in CFA of selected phosphorites, 183 marker bed, 87, 410 minerals, 372, 410 ramp model, 75 substitution in CFA, 171, 180, 182 Carbonate fluorapatite (Cas[PO4,CO3]3[EOH], (CFA), 33, 171-2, 183, 185, 189, 201, 214, 411,425, 586, 591 CO~- in CFA, 173, 183, 185 ooid growth, mechanisms of, 36 solubility, 32 Caribou National Forest, southeast Idaho, 469 CaSOa-type chemistry, 473 Castilleja, 530 Catagenesis, 189, 220 Cathedralian (Kungurian)- to Roadian-age rocks, 36 Cathodoluminescence, 194 Cenozoic tectonic events, 220 Central Belt, mixed choristid demospongehexactinellid sponge assemblages, 124-5 Central Eurasia, 315 Central Pangean Mountains, 6 Chemical identification techniques, 185 Chemical manipulations of Se, 550 Chemical thinning, 193 Chemical weathering, 193
Subject index Chert, 183,252 Cherty Shale Member of the Phosphoria Formation, 140, 400 rocks, statistics of chemical data for, 415-18 samples, Q-mode factors for, 423, Composition, 399 Chevron Oil Refinery in Richmond, California, 468 China, 560 Lower Permian carbonaceous cherts of Hubei Province in, 400 Choristid demosponge, 129 Choristid sponges, 129 Chromium (Cr), 91,189-90, 281,283,290, 292, 338, 425, 544, 549, 575 Cimmeria, 6 Cisuralian Epoch, 3 Clastic-dominated Paleozoic and Mesozoic sedimentary rocks, deposition of, 137 Coastal upwelling, 24 Cobalt, 373 Colorado, 87 Colorado River, 303,307 Cominco Ltd., 55 Comprehensive Environmental Response, Compensation, and Liability Act, 301 Conda, 55 Contaminant trace elements, 322 Copper (Cu), 91, 251,286, 288, 373, 529, 542, 544, 549, 585 Coquina, 563 Countries with sedimentary phosphate deposits and occurrences, 560 Crawford Mountains in Utah, 21,401 Crawford plates, 138, 142, 146 Crinoids, 29 Cristobalite, 401 Criterion for Continuous Concentration (CCC), 468 Cross-cutting relationships among minerals, 189 Crystallinity, 179 Cu/Zn and Mo/Zn ratios, 97 Cuba, Miocene Guines Pipian District of, 563 Currents in the Phosphoria sea, schematic pattern of, 28 Cutthroat trout, 488 Cyclic sedimentation of Permian rocks, 28 Cycling of phosphorus and other nutrients, 75 Dairy Syncline, 527, 531 Darby plate, 138, Darby thrust, 146, 158
623 Data acquisition on phosphate mines, deposits, and occurrences, 559-61 Dechampsia caespitosa, 550 Deep-marine Triassic depocenters, 115 Deer Creek, 527, 531 Demosponges, 119, 126 Depositional and diagenetic processes, 169 Depositional environments and transgressive-regressive cycles, 27 Depth of burial of Phosphoria, 140 Detrital components, 339 Detrital mineral assemblage, 197 Detrital quartz, 261 Detritus, algal and animal, 502 Diammonium phosphate (DAP), 607 Diatomite, 563 Dicalcium phosphate (Mannsville Chemical Products), 608 Dinwoody Formation, 142-3 Triassic, 140 Dixon's cross-section, 150, 158 Dolomite, 29, 171, 180, 205, 261, 411 f-lamellae, 158 in Ervay Member and Shedhorn Sandstone, 30 Dolomitic carbonate, 203 Dolostone, 183, 577 Drainage in the Phosphoria Formation, Southeast Idaho, 443,467 Dry Valley, 171,527, 531 Dry Valley Creek (DVC), 487 Dry-mass toxicity threshold, 551 Duplex of Armstrong and Cressman, 145 Eastern Belt of distinctive spiculite deposits, 126 rhax-bearing, demosponge-dominated spiculite assemblages, 119 Ecological risk to avian and mammalian terrestrial receptors, 439 based on regional Se drainage, 438 threshold ranges, 305 Ecosystems in the western United States, 303 Edna Mountain Formation, 114, 122-3 Effiorescent salt, 190, 217 Electron microprobe (EMP) 185 analyses, 252 Element associations, 80, 190 concentrations, 327 distributions, geometric means and deviations, 333
624 Elemental phosphorus, 46, 605 byproducts, 605 and compounds, major applications for, 608 plants, 48, 54, 599-600 production, 604 typical phosphate ore-processing sequence for, 603 products, 599 Elements of environmental concern, methods of analysis, 402-6 in Meade Peak, compositional average of, 329 in residual solids, concentration of, 287 Elk, 440 Emeishan flood basalts in south China, 13 Energy-dispersive spectrometer (EDS), 254, 256 Energy dispersive X-ray analysis (EDXRF), 233 methodology, 329 Enlarged Homestead Act of 1909, 49 Enoch Valley mine, 77, 171 ENVI (The Environment for Visualizing Images), 194 Environmental concern, elements of, 399 Environmental Trace Substances Laboratory (ETSL), 489 Environmentally sensitive elements (ESE), ! 89-91,424-5 downstream transport of, 222 Eolian silt, 86 Epigenetic/supergene assemblage, 212 Equatorial Panthalassa, 7 Equator-to-Pole thermal gradient, 7 Erosion, 193 Ervay Member of the Park City Formation, 25, 34 ESRI, Inc., Redlands, CA, 533 Europe, 315 Eustatic changes and sequence stratigraphy, 34-6 Evaporative basins, 33, 74 Extraction experiments, selective, 232 False Cap Limestone, 87 Fantail darter, 507 Fe oxides, 477-9, 481 fraction of wetland sediments, association of Se with, 478 Fe203/A1203 values, 87 Fe203/AI203, Ba/AI203 and Sc/A1203 ratios, 103 Fe203/AI203/AI203 and TiO2/AI203 ratios, comparison of, 88
Subject index Feldspar, 261, 410 Ferrous Fe, 329 Fertilizers, 599 products, 607 use, 46 Festuca ovina, 550 Fish, 440, 488 comparison to other Idaho data, 513-15 forage, 310 health of community, 515 other elements, 508-13 population in rivers monitoring of, 515 selenium, 506 whole-body element concentrations, 509-12 Fish-scale bed, 23 basal, 75 Fission tracks density of, 234 radiography, 232 Florida phosphate, 234 Fluid (oil and brine) migration, 189 Fluorescence-based micro-digestion, 301 Fluorine, 79 Fluorite-carbonate-barite-bitumen veins, 189 Fold-and-thrust-belt processes and geometry, 162 Fontaneile Creek, 77 Fontanelle sections, 94 Forest Service, 533 Francolite, 585,602 and trace nutrients, accumulation of, 584 Franson Member of the Park City Formation, 25, 34, 85 Freeze-thaw effects, 395 Gamefish, 311 Gammacerane, 33 Gamma-ray spectrometry, 230 General Land Office, 52 General Mining Law of 1872, 48 Geochemical analyses methods, 327-9 in Rex Chert Member, 405 Geochemical exploration, P, U, and V, 24-7 Geochemical mechanism of dispersal and selenium discharges, 308 Geochemical sampling, 324 Geochemistry of rocks, 73, examination methods and data evaluation, 77 Geochron Laboratories, 208 Geoenvironmental problems, 363
Subject index
Geoenvironmental studies on mobilization and fate of selenium (Se), 527 Geoenvironmentally significant trace elements, 358 Geological and geochemical processes, 593 Geological and Geographical Survey of the Territories, 48 Geological and mineralogical characterization of rocks, 169 methods, 172-4 Geological attributes of phosphates, 587 Georgetown, 21 Geo-stationary Operations Environmental Satellite (GOES), 444 Geothermal gradient, 142 Glauconite, 204, 401, 410, 563 Global occurrence of phosphorites and petroleum, 313-15 Global selenium sources to environment in Phosphoria Formation, model for forecasting, 299 Global warming, 6, 13 Golconda thrust, 114-17 Gondwana, 3, 6 Goose Egg basin, 34, 74, 579 Goslarite (ZnSO4"7H20), 217 Gran titration, 469 Grandeur Member of the Park City Formation, 25, 34 Grasses, 539, 544 Se concentrations, 533 Greenhouse survey study, 550 Greenland, 12 Guadalupian, Epoch, 3,401 Guano deposits and guano-derived deposits, 561,565 at Wichianburi, Thailand, 565 Gypsum, 171, 217 Halides, 213 Havallah basin, 111, 114-15 Hayden Survey, 48 Hazard Assessment of Se and other trace elements, 515-20 Hematite, 217 Hexactine spicules, 129 Hexactinellid diversity, 130 spicules, 119 sponges, 125, 129 Hg see mercury Historic phosphate legislation, litigation, and policy, 50-1 Hogsback plates, 138
625 Hogsback thrusts, 146, 150 Hot Springs Mine, 77 Human body requirement phosphorus, 600 Human Health Fact Sheet, 548 Hydride generation atomic-absorption spectroscopy (HG-ASS), 533 Hydrography of sea marine sediment fraction, 580-5 nonmarine sediment fraction, 578-80 Phosphoria- a depositional model, 576-85 Hyperaccumulators or primary accumulators, 530 Idaho, 20-1, 25, 45, 48, 52, 74, 112, 118-19, 301,428 batholith, 114 bulk mineralogy, 174-7 environmental selenium concentration, case study of, 308-13 phosphate mines, active, 193 phosphate mining, 52-5 areas, trace elements concentrations, 52 nondiffracting component of rocks, 177-80 summary of historic and active phosphate-rock mines in, 62 Idaho Department of Environmental Quality, 311,440, 463 Idaho Mining Association Selenium Subcommittee, 494 lllite, 171 Inductively coupled plasma atomic emission spectroscopy (ICP-AES), 79, 328, 469, 533 Inductively coupled plasma mass spectroscopy (ICP-MS), 79, 259, 371, 431,446, 469 Initial and residual solids characterization, 260 Insular phosphate deposits, 564 Integrated plant for manufacturing phosphorus and phosphoric acid, 606 Inter-element relationships in Meade Peak Member in marine fraction, 103 in terrigenous components, 103 International Commission on Stratigraphy (ICS), 3 Periods and Stages of the Permian, recommended, 4 International Geological Congress in Leningrad, 23 International Union of Geological Science's (IUGS), 3 Inundation of source areas, 87
626 Ion chromatography (IC), 371,469 Iron-oxide minerals, 538 Iron-oxide-rich particulates, 89 Isoclinal folds, 143 Israel, 591 JEOL 8900, 256 JEOL JSM-5800 LV electron microscope, 194 Jones, HNO3 + H202 method of, 533 Jurassic Tethys Sea, 127 K20/AI203 ratios, 84, 90, 97 Kaolinite, 171 Karuskal-Wallis one-way analysis of variance (ANOVA), 533 Kerr-McGee Chemical Corporation plant, 602 Kesterson National Wildlife Refuge in the San Joaquin Valley, California, 307 Kesterson Reservoir, 433 K-feldspars, 197 Known Phosphate Leasing Areas (KPLA), 52 Lakeridge section, 77, 94 maximum flooding surface in, 98 Land Classification Board, 52 Lanthanum (La), 96 Large Igneous Provinces (LIP), 13 Latin America, 315 Laurasia, 3 Leachate concentrations of elements, 380-3 Leaching, multiple, 396 Lead, 242 Legumes, 539, 541 at rock dumps, 527 Se concentrations, 533 Limestone, 183 lenses, 122 Lithistid demosponge desmas, 125 Lithistid desmas, 127 Lithistid sponges, 127 Lithogeochemical analysis of the Meade Peak Phosphatic Shale, 321-2 summary of major oxides and minor elements in Meade Peak rocks, 330-2 Lithology and selective chemical composition, 253 in the Meade Peak Member, variable, 80 Lithostratigraphy, 406 Little Long Valley study, 427-8 Livestock death of, 437 protection, 549-50 response to selenium and other trace elements, 546-9
Subject index Location of the Phosphoria sea, 5 Logan Forestry Sciences Laboratory, 533 Longnose dace, 488 Lopingian Epoch, 3 Lower Blackfoot River (LBR), 487 Lower detection limit (LDL), 329 Lower East Mill Creek (LEMC) 485 Lower Slug Creek (LSG), 486 Maastrichtian Gramame Formation of Brazil, 563 Macrophytes, submerged, 503 Magnesium, 379 Major-element oxides, 83,337 and elements and trace elements, mean concentration of, 335 Mammals, small, 440 Manganese (Mn), 379, 542, 544 Manganese oxide, 217 Mannsville Chemical Products, 604 Mansfield's Meade thrust, 154 Marine components, 90 Marine elements, ratios Cu/Zn and Mo/Zn, 92 large shifts in, 104 Mo/Cu and Cr/Cu, 93 offsets in, 97 spatial variations in, 99 V/Cu and V/Cr, 94 Marine evaporite basins, 28 Marine sedimentary phosphate deposit, 74, 563 Marine sessile filter-feeding epifauna, 12 Marine sponge species, 126 Marine trace elements, 91 Marine transgression and regression, cycle of, 98 Marine-derived metals, 96 Mass extinction causes or mitigating circumstances for the, 12 end of Permian, 12 Mass mortality of organisms, 22 Mass spectrometry, 232 Maybe Canyon, 527, 531 McCloud fauna, I 15 Meade and other plates, shortening of, 146-7 Meade Peak Member of the Phosphoria Formation, 34, 73, 78 element concentrations, 363 faunal evidence, 87 geologic setting and history, 192-3 generalized lithology, 195 marine element concentrations general statistics for selected in, 100-1
Subject index petrogenetic history, approach and methodology, 193-6 sediments, origin of, 76 siltstone, mudstone, phosphorite, and carbonate, 191 stratigraphy of the Meade Peak, 190, 254, 324 Meade Peak Phosphatic Shale Member of Permian Phosphoria Formation, 48, 74, analytes from samples, summary statistics, 81-2 in Southeast Idaho, 189, 321-2, 368 detrital and authigenic/diagenetic minerals of, 198 diagenetic/epigenetic/supergene minerals of, 199-201 lithologic characterization of, 321-2, 334 Meade plate/thrust central part, 148 estimates of shortening and compression directions from other data, 157-8 macroscopic shortening and extension within, 153-4 northern proposed strain history with focus on Phosphoria Formation, 161 ramp and "flat" geometry, 150 sheet, 193 shortening and extension implied by folding and faulting in, 152-7 strain distribution and structural evolution of, 137 structure, 143-61 studies of, 139-40 style of deformation of, 151-2 system, thrust-model processes and geometry affecting the, 162 thermal history, 141-3 thickness of, 147-51 timing of thrusting in southeastern Idaho, 145-6 topology of, 147-51 variable displacement, 158-61 Medicago sativa, 539 Mercury (Hg) 189, 213,425 Metals accumulation in the Phosphoria sea, schematic pattern of, 28 movement in the Meade Peak, conceptual model of history of, 218-22 to Zn ratios, 97 Metamorphic minerals, low-grade, 142-3 Methionine, 546 Mica, 410 Microcrystalline quartz, 261
627 Middle East, 315 Middle Permian Phosphoria Formation, 20 Milk River uplift area of Central Montana, 86-7 Mine waste leaching of, 459 rock dump at Wooley Valley seepage, 480 Mineral formation and element mobility, chronological sequence of, 197-218 Mineral Leasing Acts of 1917, 49 of 1920, 49, 52 Mineralogical analyses, 406 of Phosphoria Formation rocks, 169 Mineralogy Meade Peak Phosphatic Shale Member, 379 Rex Chert Member, 411 weathering, effects on the Phosphoria Formation, 169 Mining characteristics and specifications, 589-90 districts in Montana, 57 of phosphate geological attributes, 587 Mining Law, 52 Minor and trace elements, 89-90 Miocene phosphate deposits of Florida, 563 Miocene siliceous Monterey Formation of California, 425 Molybdenum (Mo), 21, 91, 190, 251,286-7, 290-1,338, 379, 395, 529, 542, 545, 548, 585 Zn ratios, 91 Monoammonium phosphate (MAP), 607 Monsanto Co., 599 elemental phosphorus plant, 602, 604 Montana, 20-1, 25, 25, 45, 48, 74, 601 phosphate mining, 55 summary of historic phosphate-rock mines in, 66-8 Montmorillonite, 171 Morocco, Ganntour and Oulad Abdoun Plateaus in, 564 Mottled sculpin, 488 Mount Ichabod, 122 Mountain Copper Company, Ltd., 48 Mud Spring, 77 Mudstone, 183,252 Murchey, paleobathymetric model of, 127 Muscovite, 171 National Monuments (1906), 49 National Research Council, 501,529, 547, 551 National Resource Center, 546
628 Native elements, 213 Native metals and alloys, 216 Nauru Island deposits, 565 Nevada, 20, 118 Nickel (Ni) 21, 189-90, 251,286-7, 290-1, 373,395, 542, 545, 549, 585 N itrogen-phosphorus-potassium (NPK) fertilizers, 601 Non-fertilizer applications major, 607 Non-sulfide Se, 291 Normal-storm and fair-weather wave bases, 75 North Africa, 315 Northern basins, 125 Northern Brooks Range, 126 Northern hemisphere storm track, 7 Northern hog sucker, 507 Northwest Pangea, 11 Northwind Ridge, Chukchi Sea, 125 Norway, 112 Null fallacy trap, 515 Ocean basins in which phosphorites occur, 562 chemistry, 3 source of phosphate to, 566 see also Sea Oceanographic setting of the Phosphoria basin, 74 Oil, 22O formation, 204 production, 315 shale, 30 Oil and Gas Journal Energy Database, 315 Ooids, 260 of carbonate fluorapatite (apatite), 260 Oonopsis, 530 Opaque organic matter, 191 Open pit mine, 48, 590-1 surface mining, 483 Oregon, 118 Organic carbon, 97, 339 layers in fine-grained dolostone, 339 Organic-rich rocks, 91 marine sedimentary, 300 Origin and evolution of Phosphoria Formation, 22 outstanding issues, 36-7 Orogenic and structural terminology, 144 Orthoclase, 171 Ostwald ripeneing, 237 Oxides, 217 Oxidizing fluids, passage of, 222 Oxygen deficiency, 22
Subject index
Oxyhydroxides, 259, 277 acid-soluble, 251
P205 context, 589 Paleoclimatic and atmospheric circulation, computer modeling, 31 Paleogeographic continental reconstruction of the Roadian-Wordian Earth, 5 Paleogeographic distribution of Lutetian (middle Eocene) phosphate deposits, 569 of Maastrichtian (Late Cretaceous) phosphate deposits, 568 of Serravallian (middle Miocene) phosphate deposits, 569 of Wuchiapingian (early Late Permian) phosphate deposits, 567 Paleogeographic map of Arctic region, 113 Paleogeography of the Phosphoria basin, 30, 566 Paleogeography, phosphogenesis, and sequence stratigraphy, 30-6 Paleotethys, 6 Pangea, 3, 5, 7 Panthalassa, 3, 7 ocean water chemical composition of, 9 isotopic compositions of, 9 Pardee's glaciation, 23 Park City Formation, 21, 25 basic framework of three disconformitybounded sequences, 35 rocks biostratigraphic units, 34 2Iopb' 230 Pearson correlation coefficients, 489 for various aquatic ecosystem components, 498 Pennsylvanian age, 21 Pennsylvanian-Permian sea, 23 Pentactine spicules, 129 Permian arc terranes, 116 basins and terranes in Idaho, Nevada, California, Oregon, and Washington, present locations of, 112 Chert Event, 131 climate, 6-7 earth, oceanography, 7-10 end of, 12 eolian deposits, 87 geology, 3-6 island arc systems, 115 marine sedimentary strata, 575 organic C-rich, seleniferous chert in China, 425
Subject index paleogeography, 3-6 plate tectonics, 3-6 rocks, classification, 25 spiculites, 112 stratigraphy as proposed by McKelvey, 25 time intervals, absolute ages of, 36 Permian-Triassic biotic extinctions, 7 Permian-Triassic boundary, 7, 9 Petrogenesis, 189 history of the Meade Peak, 193 Petrography of the Meade Peak, 190 Rex Chert Member, 410 Petroleum basins, global distribution of, 302 generation and very low grade metamorphism, 141 in the Phosphoria sea, schematic pattern of, 28 global occurrence of, 313-15 source rocks, 30, 300 Phanerozoic ratio, 10 Phase associations of elements, 419 Phosphate, 201,217, 259 crystallization, 237 deposits, formation and distribution of, 565-70 global distribution of, 302 in Morocco, 591 end products, 607 exports, 46 fertilizers and chemical derivates, 609 mechanisms for enrichment of deposit, 566 minerals, 601 mines in Idaho, 53, 170, 323 in Montana, 56 in Wyoming, 59 ore, 400, 602 quality, 590 processed products, 604-6 processing plants, mining characteristics and specifications, 589 production rates, 45 and shale exposures, 308 recrystailization of, 237 resources character and controls, economic model, 586 for southeast Idaho, 592 rock, 315 production, 46
629 Phosphatic Shale Member of the Phosphoria Formation, Southeast Idaho, 367 Phosphogenesis, 32, 36 Phosphogypsum, 604 Phosphoria and Alaskan Artinskian fauna, similarities of, 23 basic framework of three disconformitybounded sequences, 35 embayment, 116 formation - forecasting model for global selenium sources, 303 rocks, biostratigraphic units, 34 Phosphoria Formation, 10, 25 evolution of thought concerning the origin of, 19 members, 27 Phosphoria Gulch, 21 Phosphoria sea (Carboniferous sea), 22 Phosphoric acid, 55, 599 plants, 589 production, 604 typical phosphate ore-processing sequence for, 603 Phosphorite, 183,252, 300, 339 deposits, 45 detritai, and carbonate components of Meade Peak rocks, 336 Global occurrence of, 313-15 Phosphorus, 483,607 bearing phases, nondiffracting, 185 material flow in the environment, 608-9 Phosphorus oxychloride, 608 Phosphorus pentasulfide, 608 Phosphorus pentoxide (P20~), 602 Phosphorus trichloride (PCI3), 607 Phytoremediation, 550-1 Pickett Act of 1910, 49, 52 Pillow basalt, 114 Pinch-and-swell structure, 143 Placospongia, 127 Plankton, 91,503, 601 debris, 582 ratios, 93 uptake and concentration of phosphorus, 33 Plants for assessing trace-element mobility, 529-30 as bioindicators (biological indicators) for various trace-elements, 529 selection as criterion for remediating soils, 441 species sampled, 532 on waste-rock piles, 531
630 po 3accumulation of, 575, 584-5 as organic matter, 583 Portneuf River, 444 Pretectonic changes in the Phosphoria Formation, model of, 143 Pretectonic structures, 143 Principal component analysis of Meade Peak rocks, 354 Prospect, 150 plates, 138 thrust, 146 Protriaene, anatriaene, and calthrop spicules, 129 Purified acid, 606 Purified phosphoric acid (PPA) process plant in Conda, Idaho, 589 Putnam-Paris, 138 plate, 14 l thrust system, 143-4, 150, 154 Pyrite, 171, 181,206, 208, 213,288, 401 Quality assurance/quality control of chemical analyses, 490 for dissolved and suspended-sediment selenium analyses, 446-8 Quantitative analyses using the electron microprobe, 264 Quantitative XRD, 185 Quartz, 158, 171,204 226Ra, 230 Radiolarian chert, 114 Radiolarians, 130 Radiolarian- and spicule-bearing argillaceous slope facies, 115 Radium, 242 Ramp-flat model of thrusting, 151 Rare earth elements (REE), 91,338, 575, 580 concentrations, 97 in Meade Peak Member, 95 La vs. Yb as surrogates for light and heavy, 102 terrigenous and marine contributions of, 580 Rasmussen Ridge, 171,527, 531 Reclamation Act of 1902, 49 Redbreast sunfish, 507 Redfield stoichiometry, 76 Redside shiners, 488, 508 Regional thickness variations of the Phosphoria, 24 Retort Phosphatic Shale Member, 30--1, 34, 85
Subject index
Rex Chert Member of the Permian Phosphoria Formation, 21, 24, 34, 119, 140, 400 analyses of outcrop samples from the, 424 composition of rocks, 399 chemical, 412-14 rocks, statistics of chemical data for, 415-18 samples, Q-mode factors for, 423 stratigraphic changes in chemical composition, 419 Rex Peak, 401 Rhax-bearing demosponge-dominated spiculites of Permian Phosphoria basin and Antler high, 130 Rietveld method, 177 Rietveld unit-cell refinements, 171 Roadian and Wordian Stages, 10, 12 Roberts Mountain allochthon, 114 Roberts Mountain thrust fault, l 14 Rock alternation, 392-4 Rock bass, 507 Rock leachate geochemistry of Meade Peak Phosphatic Shale Member, 367 methods, 370-2 discussion, 373-6 Rock samples in leachate experiments, 369 particle size, 383 individual, 338 Rocky Mountains of Idaho, Montana, Utah, and Wyoming, 575 Roscoelite (V illite), 205 Rugose corals, 29 Ruzhencevispongacea, 130 Ruzhencevispongacid radiolarians, 116, 124-5 Salt River, 444 San Francisco Bay-Delta Estuary, 303,307 San Francisco Chemical Co., 56, 58 San Joaquin Valley, California, 303,468 alluvial fans of, 306 Scanning electron microscope (SEM) techniques, 194, 252, 260 Schoonover basin, 114 Scientific surveys, early, 48 Sea-floor deposits, modern, 563 Sea level rise in, 98 shelf and platform deposits, 563 see also Ocean Sea of Japan, 90 Seamount deposits, 561
Subject index Seawater REE pattern, 581 Se concentrations, 301 Secondary accumulators or absorbers, 530 Sediment, 495-501 comparison to other Idaho data, 500 element concentrations, 496-7 other elements, 498-500 Selenium, 495 Sedimentary phosphate deposits, 561-5 compilation of, 571 Sedimentation in the Phosphoria sea, schematic pattern of, 28 rates in the Phosphoria basin, 80 Sediment-starved basins, 104 Sediment-trap experiments of modern continental-shelf environments, 76 Seep water, 475 from waste-rock dumps, 467 Selenide (CuxSey), 213 Seleniferous open-range forage plants, 307 Selenium (Se), 79, 189-90, 213-14, 251,329, 338, 354, 356, 368, 395,425,428, 432, 468, 529, 546 in agricultural drainage, 468 in air samples, 432 anthropogenic release to the environment, 468 attenuation in wetland, 467 methods, 469-73 results, 473-80 bioaccumulation, 310, 463 in fish, 508 in food webs, 439 biochemistry and guidelines, 303 biogeochemistry, 468 biological compounds, containing, 300 chemistry, 550 commodities and exploration of, 315 compounds, 431 airborne volatile, 430 concentrations, 271-2, 274-6, 453,476 airborne, 433 in birds and mammals, 31 I-13 in bird eggs, 301 comparison between seepage and background waters, 474 field case-studies and environmental, 305 in grasses, 503 plants, invertebrates, and fish, 310 in surface water, 473 contamination, 307 deposition, model, 315 effects, 303
631 elemental, 258 environmental concerns, 528 guidelines, 300 hazard in Caribou National Forest, assessment of, 520 categories of, 516 health and risk criteria, 300 K-edge X-ray absorption near edge spectra (XANES), 256 leachate, 379 load, 455 loading through the Blackfoot River Watershed, linking sources to ecosystems, 437 mineral affinities and distribution in black shale and phosphorite of the Phosphoria Formation, 251 mineralogical associations and distribution of, 252 methods, 252-60 native, 215 nutritional guidelines and national guidelines for risk, 303 ocean chemistry, 305 "organically bound", 277 pathway, bioaccumulation models, 463 poisoning in aquatic ecosystem, case of, 307 chronic, 547 pollution, 300 based on Se concentrations in the environment conceptual model of, 304 conceptual model, 302-7 methods and sources of data, 300-2 release of, 400 residues in fish, 515 retention and bioavailability, 315 in rocks of the Meade Peak, 293,428 in sediments, concentrations of, 500 source-rock weathering, model, 315 source rocks, 301,306 sources and biogeochemistry of, 305 prediction of, 313 and source drainage, 458 toxic effects, 303,437 toxicosis, 190 risk of, 541 uptake into plants, 539 and implications for grazing animals, 527 in vegetation, 527, 539 geographic effects, 534
632 geologic effects, 535 redox effects, 535 volatilization rates, 433 in water, sediment, aquatic plants, aquatic invertebrates, and fish from streams, 483 study methods and materials, 484-9 study results, 490-515 Selenium Subcommittee, 500 Selenomethionine, 546 Senegal, 591 Thies and Taiba Plateaus in, 564 Sequence stratigraphy, 37 Sequential extractions, 270 steps for study of selenium and trace element distribution summary of, 258 techniques, 257 Sevier fold-and-thrust belt, 138 orogeny, 144, 146, 152 thrusting, 31 Shallow-water restricted basin, 23 Shedhorn Sandstone, 25, 27 Sheep Creek (ShpC), 487 Siberia, 3 Siberian Traps volcanism, 13 Silica, 401 Silicates, 183, 204, 212 Siliceous microfossil assemblages, 113 Siliceous sponges, 129 spicules, 112-13 diagenesis of, 27 Silicic domes, 193 Siliciclastic debris, 577-8 Silver (Ag), 338, 358 Smoky Canyon mine, 171 Snake River Plain, 161 Societal Relevance, Processing, and Material Flow of Western PhosphateRefreshments, Fertilizer, and Weed Killer, 599 Society of Economic Paleontologists and Mineralogists (SEPM), 32 Soda Springs, 48,428 Sodium, 379 Sodium tripolyphosphate (STPP), 607 Solid characterization of, 256, 260 Solid-phase chemical analyses, 256 Sonoma orogeny, 11 South Carolina, 601 Southern hemisphere storm tract, 7 Spatial variations in terrigenous elements, 90 Speckled dace, 488, 507
Subject index Sphalerite, 171, 181,205,208-9, 213, 215, 263 disease, 208 Spicules populations, 127 within the Rex Chert, 122 Spiculite black chert, 123 faunas in the Permian Phosphoria Basin, regional analysis of implications for paleoceanography, 111 "Split-moving window" method, 84 Sponge distribution patterns, 130 migration, simple model of, 129 Sponge spicules, 111, 410 assemblages, 122 data for individual samples, 120-1 faunas, 123 morphotypes, identification of, 117 populations, Early Permian and Middle Permian, 131 quantitative comparison to radiolarians, 118 siliceous, in Permian rocks of the western Cordillera, 118 in total microfossil population, 116 Sponge spiculites, 111 from the Chukchi Sea, 128 Stanford Synchrotron Radiation Laboratory (SSRL), 256 Stanleva pinnata (Prush), 530 Statistical Analysis System, Inc., 489 Stauffer Chemical Co., 55 Stipa columhiana, 550 Stock-Raising Homestead Act of 1916, 49 Stowasser, 69 Strain regimes superimposed on map of structure contours on Meade thrust, 156 Strategic Minerals Program, 24 Stratigraphy, sequence, 37 Stringers of organic matter, 237 Strontium (Sr), 338 Student's t-tests, 84 Stylolites, 143 Subacute toxicosis, 547 Subaerial chemical weathering, 566 Subduction-related mountain building, late-Permian and Triassic, 11 Sublett basin, 30 Sublette Range, 58 Sulfates, 217
63 3
S u b j e c t index
Sulfides, 206, 213,218, 265,277 crystalline, 259 mass-balance calculations, 288 microprobe analyses of, 211 Se-bearing, 220 Sulfur (S), 79 isotopes, 10 total, 96, 329 vs. TOC in the Mead Peak Member of the Phosphoria Formation plot of, 102 Sulfur dioxide, acid rain, 13 Sulvanite ( C u 3 V S 4 ) , 209-10 Supergene minerals, 190 enrichment, 222 Superprobe EMP, 256 Suppression of carbonate-producing organisms, 131 Surface water samples, 230, 471 Sverdrup Basin, 128 Syntectonic Ephraim Conglomerate age of, 145 Syntectonic structures, 144 Synthetic precipitation leaching procedure (SPLP), 370 Tectonics relationships to various phases of mineralization, 36 transport direction of, 147 Temporal and spatial variations in sediment geochemistry, 73 Temporal variations in marine elements, 90 in terrigenous elements, 80 Tennessee, 601 "brown rock" deposits, 564 phosphate deposits, 561 "white rock", 564 Terranes, ! 0 Terrestrial organisms, 12 Tetrahymenol, 33 Textural criteria, 198 Thallium (TI), 359, 428 Thermal ionization mass spectrometry (TIMS), 232, 244 Thermohaline circulation, 7, 111 Thorium, 242 Thrust faults, major, 138 Thrust-plate structures, 24 Tight folding, 193 Time-series leachate concentrations of various elements, 391 TiO2/AI203, 89 Tin (Sn), 251 Titanium, 328
Tosi Member of the Phosphoria Formation, 34 Total organic carbon (TOC), 78, 96, 98, 102 see also Carbon Toxic elements, 400 Toxicity, chronic, 547 Trace elements, 78, 338-9, 354, 431,433, 549 associations as a function of alteration, 345 bioavailability to plants study for assessing methods results, 533-50 and carbon for Meade Peak zones, 342 concentrations, change in, alternation model, 359-63 contents in phosphorites in deeper-water facies, 37 environmental concerns, 528 individual, 354 leachate concentrations of, 368 leachate experiments for, 368 mineral affinities and distribution in black shale and phosphorite of the Phosphoria Formation, 251 in unweathered and weathered rocks, summary of, 292 mineralogical associations and distribution of, 252 methods, 252-60 plants and livestock/wildlife intake, ranges of approximate upper critical, 542 residence, 170 in sediments, concentrations of, 500 toxic, from waste-rock piles, 484 uptake in vegetation, 542 in water, sediment, aquatic plants, aquatic invertebrates, and fish from streams, 483 study methods and materials, 484-9 Transgressive-regressive cycles, 28 sea-level, 27 Transition to more toxic conditions, 104 TRC Environmental, 500, 503 Trial Creek (TC), 486 Tridymite, 401 Triple superphosphate (TSP), 607 Turbiditic sandstone, 114 23SU isotope, fission of, 232 23SU decay series, 230-1,246 determined by gamma-ray spectrometry, disequilibria in, 240-6 Union Phosphate Co., 58 United States see US Upper East Mill Creek (UEMC), 485 Upper Slug Creek (USG), 486 Uraniferous-phosphorite, 237
634 Uraninite, 212, 237 Uranium (U), 230, 292, 338, 356, 428, 575 in alteration products, redistribution of, 237 analytical methods, 230-3 behavior during ore or rock formation, 227 concentrations, 284 dissolved, 230 and its decay products, 227 measurements of, 228 decay-series disequilibria for U-deposit exploration, 228 studies seafloor phosphorites, 227 distribution, 238 extractable, 233 hotspots characterization by scanning electron microscope (SEM), 233 isotopic composition, 230 leaching of, 246 microdistribution in phosphorite, 234 sample collection and description, 228-30 (VI), 242 during weathering, 228, 247 Uranium oxide, 237 Uranium-thorium isotopic analyses of Meade Peak Phosphatic Shale Member samples, 244 US (United States) Atomic Energy Commission, 27 Bureau of Land Management (BLM), 54, 463 Census Bureau, 604 Department of Agriculture, 308, 315, 458-61,533 Department of Health and Human Services, 300-1,303,305, 311,440, 463 Department of Interior, 301,305, 308, 311, 315,438-9, 455,458-9, 460-2 elemental-phosphorus production, 604 Federal mining laws, 48 Fish and Wildlife Service, 515-16 map of northwest, with extent of Phosphoria Formation, 75 Phosphate Co., 58 Phosphate Exploration and Mining, Laws Associated with, 49 production of marketable phosphate rock, 47 Western, 301 guidelines for protection from Se toxicity, 304 phosphate production, 46 US Environmental Protection Agency (USEPA), 242, 301,305,438, 440, 462-3,468,489, 491,493, 513,515 guideline for Se content, 190 Method 1312, 370
Subject index US Geological Survey (USGS), 19-21, 25, 172, 441,444, 448, 451,472, 484, 516, 533, 559, 600 Blackfoot River gaging station, 442 Energy Assessment Team, 315 National Water Quality Laboratory, 445 Phosphoria programs, 24, 27 TRIGA research reactor, 232 Water Quality Field Supply Unit, Ocala, Florida, 445 Utah, 20-1, 25, 45, 48, 52, 74, 87, 138 phosphate mining, 55 summary of historic and active phosphate-rock mines in, 69-70 V/Cu ratios, 93 Vaesite-pyrite~,~, 206, 208, 210, 213 Vanadiferous Zone, 343 Vanadium (V), 19, 21, 52, 91, 251,286-7, 289, 292, 338, 358, 379, 425, 575 Vanadium pentoxide, 602 Vegetation selenium concentration, 440 manipulations of, 550 Vocontion Trough, 90 Volatile-Se species, 433 Volcanic Dekkas Formation, 125 Volcanic dust, 13 Volcanic events, 13 Volcanic-arc system, 31 Washington, 118 Waste-rock dumps/piles, 400, 429-30, 483 drainage from, 535 "hot spots", 548 Waste shale, 602 Water circulation in the Phosphoria sea, pattern of, 37 depth of the Phosphoria sea, 25 quality characteristics, 483 Water years (WYs), 448 Water-rock interaction alteration of Meade Peak rocks from, 343 solutions, 372 Wavelength-dispersive spectrometers (WDS), 256 Weathering, 180-2, 245,343-59, 590-2 compositional changes due to, 340 of CFA, 184 of Meade Peak Phosphatic Shale Member, 227,247 related residual and infiltration deposits, 564 in the Phosphoria Formation, 174
Subject index Wells Formation, 21 Western Belt, radiolarian-dominated assemblages, 125 Western North American Margin and the Phosphoria Sea, 10-12 Western Phosphate Field (WPF), 20, 575, 599 depositional and economic deposit models, 575 formation and economic significance of geological and geochemical processes, 593 history of production of, 45 mining, 45-6 methods in, 602-3 in twenty-first century, 59-60 origin of, 19 outline of, 20 resources of, 592 Western Phosphate Lands, delineation of, 21-3 Western Phosphate Reserve, 52 Western phosphate rock use distribution, 47, 601 Western Regional Climate Center, 473 Wetland, 430 sediments, 467 surface water, 475 Wet-process phosphoric acid, H3PO 4, 604 Wheat Creek, 77 Wildlife, response to selenium and other trace elements, 546-9 Willard thrust model, 143 Wooley Valley, 141, 471-3 Piper diagram of major ion chemistry of sample populations, 475 World sedimentary phosphate deposits and occurrences, review of, 559
63 5 World shale average (WSA), 579-80 element composition, 85 values, 76, 83,581 WPSB-26, 243 WPSC-135,243 WPSD-0.5,243 Wyoming, 20-1, 25, 27, 45, 48, 52, 74, 87, 138, 601 phosphate mining, 58 shelf, 73 summary of historic phosphate-rock mines in, 71 Xenotime, 212 X-ray absorption fine structure spectroscopy (XANES), 217 X-ray absorption near edge spectra (XANES), 268 X-ray absorption spectroscopy (XAS), 252, 255,268 X-ray diffraction (XRD), 252 analysis and Rietveld refinement, 172 mineralogy of Rex Chert, 407-8 of samples and solid residues, 278-9 X-ray fluorescence (XRF), 328 spectroscopy, 79 Xylorhiza, 530 Yonkee, thermal model of, 142 Zawislanski and Zavarin's study, Se volatilization rates, 433 Zinc (Zn), 189-90, 215,218, 281,285, 291,338, 373,395,425,529, 542, 545, 585
This Page Intentionally Left Blank