Special Paper 416
THE GEOLOGICAL SOCIETY OF AMERICA
Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates
edited by Ana María Alonso-Zarza Departamento Petrología y Geoquímica Facultad de Ciencias Geológicas Universidad Complutense de Madrid 28040 Madrid Spain Lawrence H. Tanner Department of Biological Sciences Le Moyne College Syracuse, New York 13214 USA
Special Paper 416 3300 Penrose Place, P.O. Box 9140
Boulder, Colorado 80301-9140, USA
2006
Copyright © 2006, The Geological Society of America, Inc. (GSA). All rights reserved. GSA grants permission to individual scientists to make unlimited photocopies of one or more items from this volume for noncommercial purposes advancing science or education, including classroom use. For permission to make photocopies of any item in this volume for other noncommercial, nonprofit purposes, contact the Geological Society of America. Written permission is required from GSA for all other forms of capture or reproduction of any item in the volume including, but not limited to, all types of electronic or digital scanning or other digital or manual transformation of articles or any portion thereof, such as abstracts, into computer-readable and/or transmittable form for personal or corporate use, either noncommercial or commercial, for-profit or otherwise. Send permission requests to GSA Copyright Permissions, 3300 Penrose Place, P.O. Box 9140, Boulder, Colorado 80301-9140, USA. Copyright is not claimed on any material prepared wholly by government employees within the scope of their employment. Published by The Geological Society of America, Inc. 3300 Penrose Place, P.O. Box 9140, Boulder, Colorado 80301-9140, USA www.geosociety.org Printed in U.S.A. GSA Books Science Editor: Marion E. Bickford and Abhijit Basu Library of Congress Cataloging-in-Publication Data Paleoenvironmental record and applications of calcretes and palustrine carbonates / edited by Ana María Alonso-Zarza, Lawrence H. Tanner. p. cm.--(Special paper; 416) Includes bibliographical references and index. ISBN-10 0813724163 (pbk.) ISBN-13 9780813724164 (pbk.) 1. Calcretes. 2. Rocks, Carbonate. 3. Paleopedology. I. Alonso-Zarza, Ana María, 1962-. II. Tanner, Lawrence H. III. Special papers (Geological Society of America) ; 416. QE471.15.C27.P35 2007 552/.58--dc22
2006041338
Cover: View of laterally continuous pedogenic calcretes in the Upper Triassic (Norian) Owl Rock Formation (Chinle Group), northern Arizona. Photo by L.H. Tanner. Back cover: Recent vertical calcrete formed by the penetration of tree roots on Miocene deposits of the Madrid Basin, Guadalajara, Spain. Photo by A.M. Alonso-Zarza.
10 9 8 7 6 5 4 3 2 1
ii
Contents Preface .........................................................................................................................................................v Ancient Landscapes, Climate and Sequence Boundaries 1.
Calcic pedocomplexes—Regional sequence boundary indicators in Tertiary deposits of the Great Plains and western United States ...................................................................................1 D.L. Hanneman and C.J. Wideman
2.
A Late Triassic soil catena: Landscape and climate controls on paleosol morphology and chemistry across the Carnian-age Ischigualasto–Villa Union basin, northwestern Argentina .....17 N.J. Tabor, I.P. Montañez, K.A. Kelso, B. Currie, T. Shipman, and C. Colombi
3.
Investigating paleosol completeness and preservation in mid-Paleozoic alluvial paleosols: A case study in paleosol taphonomy from the Lower Old Red Sandstone .......................................43 S.B. Marriott and V.P. Wright
4.
Calcareous paleosols of the Upper Triassic Chinle Group, Four Corners region, southwestern United States: Climatic implications .................................................................................................53 L.H. Tanner and S.G. Lucas
5.
Estimates of atmospheric CO2 levels during the mid-Turonian derived from stable isotope composition of paleosol calcite from Israel ......................................................................................75 A. Sandler
6.
Pedogenic carbonate distribution within glacial till in Taylor Valley, Southern Victoria Land, Antarctica ..................................................................................................................89 K.K. Foley, W.B. Lyons, J.E. Barrett, and R.A. Virginia
Sedimentary Environments and Facies 7.
Calcretes, oncolites, and lacustrine limestones in Upper Oligocene alluvial fans of the Montgat area (Catalan Coastal Ranges, Spain) .......................................................................105 D. Parcerisa, D. Gómez-Gras, and J.D. Martín-Martín
8.
The role of clastic sediment influx in the formation of calcrete and palustrine facies: A response to paleographic and climatic conditions in the southeastern Tertiary Duero basin (northern Spain) ..............................................................................................................................119 I. Armenteros and P. Huerta
9.
The Upper Triassic crenogenic limestones in Upper Silesia (southern Poland) and their paleoenvironmental context ............................................................................................................133 J. Szulc, M. Gradzi´nski, A. Lewandowska, and C. Heunisch iii
iv
Contents 10. A recent analogue for palustrine carbonate environments: The Quaternary deposits of Las Tablas de Daimiel wetlands, Ciudad Real, Spain ................................................................153 A.M. Alonso-Zarza, M. Dorado-Valiño, A. Valdeolmillos-Rodríguez, and M. Blanca Ruiz-Zapata 11. Depositional conditions of carbonate-dominated palustrine sedimentation around the K-T boundary (Faciès Rognacien, northeastern Pyrenean foreland, southwestern France)..............169 D. Marty and C.A. Meyer 12. Reworked Microcodium calcarenites interbedded in pelagic sedimentary rocks (Paleocene, Subbetic, southern Spain): Paleoenvironmental reconstruction ...................................................189 J.M. Molina, J.A. Vera, and R. Aguado Dating of Calcretes: Applications 13. Calcite cement stratigraphy of a nonpedogenic calcrete in the Triassic New Haven Arkose (Newark Supergroup) ......................................................................................................................203 E.T. Rasbury, E.H. Gierlowski-Kordesch, J.M. Cole, C. Sookdeo, G. Spataro, and J. Nienstedt 14. Calcrete features and age estimates from U/Th dating: Implications for the analysis of Quaternary erosion rates in the northern limb of the Sierra Nevada range (Betic Cordillera, southeast Spain) ...............................................................................................................................223 J.M. Azañón, P. Tuccimei, A. Azor, I.M. Sánchez-Almazo, A.M. Alonso-Zarza, M. Soligo, and J.V. Pérez-Peña
Preface The study of ancient soils continues at an accelerating pace as more geologists recognize the value of these ancient land surfaces as archives of important paleotopographic, paleoenvironmental, and paleoclimatic information. Indeed, a survey of one database yields over 600 citations containing the keyword “paleosol” for just the first half of this decade, compared to only one-fourth this number from the first half of the 1990s! Not all of these publications presented detailed descriptions and interpretations of paleosols, certainly, but many were broader studies that incorporated the description of ancient soil surfaces into examinations of tectonics, basin evolution, sedimentary processes, or climate change. Clearly, the variety of paleosols and their potential applications to geological problems is enormous. Given the breadth of this subject, we chose to focus this volume on the topic of calcretes and the closely related subject, palustrine carbonates. Calcretes are perhaps the most commonly described of paleosols, owing to their ready preservation in the rock record and relative ease of recognition. The term calcrete, synonymous with caliche, is widely applied, although it is neither the name of a soil order nor of a soil horizon. In a broad sense, calcretes are, as proposed by Watts (1980, p. 663; after Goudie, 1973), “terrestrial materials composed dominantly, but not exclusively, of CaCO3, which occurs in states ranging from nodular and powdery to highly indurated, and result mainly from the displacive and/or replacive introduction of vadose carbonate into greater or lesser quantities of soil, rock, or sediment within a soil profile.” This definition was restricted to calcretes of pedogenic origin, however Wright and Tucker (1991) later expanded the term calcrete to include, as initially recommended by Netterberg (1980), the effects of shallow groundwater. This broader sense suggests the importance of the interaction between sediments undergoing active pedogenesis and shallow groundwaters. Palustrine carbonates exhibit many similarities with calcretes. As described by Freytet (1984, p. 231), a palustrine limestone “must show the characteristics of the primary lacustrine deposit (organisms, sedimentary features) and characteristics due to later transformations (organisms, root traces, desiccation, pedogenic remobilizations).” Palustrine carbonates are common in alluvial sequences, often in association with calcretes, but their widespread recognition has been attained more slowly. Indeed, much of the research on alluvial carbonates has focused exclusively on either palustrine carbonates or calcretes, when in fact there is often a spatial transition from one to the other, revealing an interplay between pedogenic, sedimentary, and diagenetic processes. Indisputably, these deposits contain information that is significant to the interpretation of the sedimentary record and the evolution of the landscape in both recent and ancient settings (Alonso-Zarza, 2003). These terrestrial carbonates are widely distributed on floodplains and in the distal reaches of alluvial basins. Their presence and characteristics can be used as indicators of aggradation, subsidence or changing accommodation rates, and therefore as indicators of different tectonic regimes. Although calcretes and palustrine carbonates are both commonly associated with semiarid climates, more detailed climatic information can be obtained from the depths of the carbonate-bearing horizons within paleosol profiles and from the oxygen isotope signature of the carbonate. The carbon-isotope composition, on the other hand, has been used quite successfully to track changes in atmospheric pCO2 through the Phanerozoic. Vegetation is important to the formation of many of these types of carbonates, and data on the prevailing vegetation may be obtained sometimes from the analysis of the micro- and macrofabric of the carbonate. This volume was inspired by a technical session on the topic of calcretes and palustrine carbonates (chaired by us) that was held at the 32nd International Geological Congress in Florence in August 2004. Six of the contributions presented here were first delivered at this meeting, and the volume grew with additional v
vi
Preface
contributions that provided both a broad overview of calcretes and palustrine carbonates and the “state of the art” of their application. The papers presented here cover a wide array of environmental settings and ages of deposits where calcretes and palustrine occur. Moreover, the papers included in this monograph discuss a number of interesting applications, including: a possible modern analogue for palustrine carbonates, the interplay between palustrine, pedogenic and diagenetic processes, the utility of radio-isotopic dating of pedogenic carbonates and its application to understanding the evolution of recent landscapes, the reconstruction of a diagenetic sequence, and the climatic and geomorphic controls on calcrete formation. The papers have been arranged in three groups. Papers that apply calcretes and palustrine carbonates to the reconstruction of ancient landscapes, climate and sequence boundaries comprise the first group. Hanneman and Wideman illustrate the utility of calcic pedocomplexes in delineating regional unconformities that are large-scale sequence boundaries. Their study, focused on the Tertiary of the Great Plains, shows that these pedocomplexes have distinct physical properties that allow their identification in seismic sections and well logs. Tabor and co-authors demonstrate that the distribution of the fluvial channel sandstones and the characteristics of the paleosols are both controlled by geomorphic evolution during deposition of the Triassic Ischigualasto Formation in northwest Argentina. The preservational bias in paleosol formation is described in the contribution by Marriott and Wright. These authors analyzed mid-Paleozoic paleosols from the Lower Old Red Sandstone and show that reactivated, truncated cumulate horizons provide a means of assessing the dynamics of floodplains, including those from before the advent of rooted vascular plants in the mid-Paleozoic. The fourth paper, by Tanner and Lucas, relates the potential climatic control on the morphology of Upper Triassic paleosols in the Chinle Group of the southwestern United States. Temporal changes in the types of paleosols and the maturity of calcretes suggest a gradual aridification across the Colorado Plateau during the Late Triassic. Sandler uses the isotopic composition of Mid-Turonian paleosol carbonate to estimate the atmospheric pCO2 level for this interval. His results, which indicate high mid-Turonian pCO2, correspond with the high temperatures that prevailed at that time. The last paper of this group, by Foley and co-authors, demonstrates that the relatively low carbonate concentrations in Antarctic polar desert soils can be attributed to the shallow active layer, low rates of weathering, and the extreme aridity of the landscape. Moreover, the differences in CaCO3 concentrations in these soils correlate with landscape position with respect to elevation and distance from the coast. Six papers dealing with the sedimentary environments and facies of calcretes and palustrine carbonates comprise the second group. These papers provide an overview of the interrelationships between calcretes and palustrine carbonates in terrestrial environments, focusing on their similarities and on problems in their interpretations. Notably, some papers discuss the lack of a recent analogue for ancient palustrine carbonates. In the first contribution to this group, Parcerisa and co-authors analyze the geochemistry of calcretes, oncolites and lacustrine limestones formed during the Upper Oligocene in two coalescent alluvial fans. They find that the trace element and isotopic composition of the limestones were controlled mainly by the fluvial regime and the lithology and altitude of the catchment areas in the sedimentary basin. Armenteros and Huerta studied calcretes and associated palustrine of the southeastern Tertiary Duero basin. The characteristics of both carbonate facies indicate their accumulation in semiarid climates with scarce clastic sediment supply, and that meter-scale cyclicity of the carbonate and siliciclastic sediments was controlled mostly by climate. The interrelationship between spring, fluvial, palustrine, and pedogenic facies is discussed by Szulc and collaborators in their study of the Upper Triassic freshwater carbonates from the Upper Silesian basin. These carbonates were deposited within a shallow swampy depression, fed by springs of deep-circulating groundwater. Alonso-Zarza and co-authors focused their study on a recent core in Las Tablas de Daimiel, Spain, one of the few freshwater wetlands preserved in southern Europe. Their studies of the core, including mineralogy, petrography, stables isotopes and pollen analyses indicates that these sediments are similar of those of ancient palustrine sequences, suggesting that Las Tablas is a suitable recent analogue for freshwater palustrine sequences. Marty and Meyer analyze in detail a palustrine sequence (Faciès Rognacien) encompassing the K-T boundary in southwestern France. The facies association indicates a seasonal, palustrine wetland system, with ephemeral ponds surrounded by vegetated areas of freshwater marshes under subarid to intermediate climates. The last paper of this group, by Molina and co-authors, describes an unusual occurrence of various types of calcarenites containing reworked Microcodium prisms. Their study of Paleocene marine deposits from southern Spain indicates that the Microcodium was reworked from exposed inland areas, thus providing evidence of emersion and clarifying the palebathymetry of the adjacent pelagic deposits.
Preface
vii
The final section contains two papers on different types of calcretes to which radio-isotopic dating techniques have been applied. Rasbury and co-authors describe the importance of cement stratigraphy to the application of U-Pb dating of calcite in Triassic calcretes from the New Haven Arkose, and demonstrate further that this analysis is a useful tool for distinguishing between pedogenic versus nonpedogenic calcrete. U-Th dating of calcretes is used to constrain the evolution of the Quaternary Ranges in the Betic Cordillera by Azañón and co-authors. These authors describe how rapid incision by the rivers, and subsequent capture of the Guadix by the Guadalquivir River is constrained by initial dissection of a calcrete layer dated at 42 ka. This age is used to calculate the incision and erosion rates, demonstrating once again that calcretes play a major role in the evolution of landscape in many arid and semiarid regions. This collection of papers in its final form would not have been possible without the work of the reviewers who dedicated their time to careful reviews and revisions. We were truly lucky to have the help of the following colleagues: J. Andrews, C. Arenas, B. Barclay, J. Bockheim, G. Bowen, Ll. Cabrera, J. Casanova, E. Cheney, C. De Wet, S. Dunagan, M.A. García del Cura, P. Ghosh, R. Goldstein, A.D. Harvey, M. Joeckel, A. Kosir, J. López, G. Marion, A. Martín-Algarra, P. McCarthy, D. Nash, R. Palma, T. Peryt, N. Platt, G. Retallack, D. Royer, Y. Sánchez-Moya, P.G. Silva, A.R. Soria, R. Swennen, M. Talbot, S.K. Tandon, A. Travé, D. Valero-Garcés, D. Varrone, and J. Wilkinson. Our sincere thanks also go to our departments: Departamento de Petrología y Geoquímica de la Universidad Complutense de Madrid and the Department of Biological Sciences of Le Moyne College. We also have a special remembrance for F. Calvet, one of the pioneers in the studies of calcretes in Spain who passed away a few years ago. His ideas are tangibly present throughout this volume. We hope the reader finds this collection of papers both stimulating and informative. This collection will, ideally, constitute a base for understanding how calcretes and palustrine carbonates form an integral part of ancient and recent landscapes and contribute to the broader knowledge of continental basins and their geomorphic features. REFERENCES CITED Alonso-Zarza, A.M., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: EarthScience Reviews, v. 60, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Freytet, P., 1984, Les sédiments lacustres carbonatés et leurs transformations par émersion et pédogénèse: Importance de leur identification pour les reconstitutions paléogéographiques: Bulletin Centres Rechercher Exploration-Production Elf-Aquitaine, v. 8, no. 1, p. 223–246. Goudie, A.S., 1973, Duricrusts in Tropical and Subtropical Landscapes: Clarendon, Oxford, 174 p. Netterberg, F., 1980, Geology of southern African calcretes: 1. Terminology, description, macrofeatures and classification: Transactions of the Geological Society of South Africa, v. 83, p. 255–283. Watts, N.L., 1980, Quaternary pedogenic calcretes from the Kalahari (southern Africa): mineralogy, genesis and diagenesis: Sedimentology, v. 27, p. 661–686. Wright, V.P., and Tucker, M.E., 1991, Calcretes: an introduction, in Wright, V.P., and Tucker, M.E., eds., Calcretes: IAS Reprint series 2, Oxford, Blackwell Scientific Publications, p. 1–22.
Ana M. Alonso-Zarza Lawrence H. Tanner
Geological Society of America Special Paper 416 2006
Calcic pedocomplexes—Regional sequence boundary indicators in Tertiary deposits of the Great Plains and western United States Debra L. Hanneman Whitehall Geogroup, Inc., Whitehall, Montana 59759, USA Charles J. Wideman Professor Emeritus, Montana Tech of the University of Montana, Butte, Montana 59701, USA ABSTRACT Calcic pedocomplexes are associated with regional unconformities in the Great Plains and western United States that have approximate ages of 30 Ma, 20 Ma, and 4 Ma. In southwestern Montana, the calcic pedocomplexes are readily identifiable on the surface, and a pedocomplex typically contains several partial soil profiles. In the most complete scenario, an individual profile may contain an argillic or argillic/calcareous (Bt or Btk) horizon, a K horizon, and a C horizon. Often, however, the Bt(k) horizon is truncated or can be entirely absent from an individual profile. The K horizon contains an upper laminated zone that is underlain by an indurated carbonate sheet. Carbonate nodules and chalky micritic matrix materials underlie the sheet carbonate. The calcic paleosols display carbonate morphology ranging from stage IV to stage VI. The calcic pedocomplexes also possess distinct physical properties that aid in subsurface identification. The combined density and velocity differences between paleosols and nonpedogenic strata result in bright reflections on seismic sections and distinct well-log signatures. Although the calcic pedocomplexes and regional unconformity associations were first described within Tertiary strata of southwestern Montana, the same associations exist in numerous localities in the Great Plains and in other parts of the western United States. The extensive occurrence of the calcic paleosols and regional unconformity associations throughout this large area underscores their utility as a regional correlation tool. Moreover, the delineation of regional unconformities that are largescale sequence boundaries by pedocomplexes has broad implications for continental sequence stratigraphy. Keywords: calcic, paleosol, sequence, Tertiary, pedocomplex. RESUMEN En las Great Plains y oeste de Estados Unidos, los complejos edáficos cálcicos están asociados con las discontinuidades regionales cuyas edades aproximadas son: 30 Ma, 20 Ma, y 4 Ma. En el suroeste de Montana, estos edafocomplejos cálcicos se observan fácilmente en afloramientos de superficie y contienen varios perfiles edáficos Hanneman, D.L., and Wideman, C.J., 2006, Calcic pedocomplexes—Regional sequence boundary indicators in Tertiary deposits of the Great Plains and western United States, in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 1–15, doi: 10.1130/2006.2416(01). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
1
2
Hanneman and Wideman incompletos. En los casos en los que estos complejos edáficos están más completos, un perfil individual puede contener un horizonte argílico (Bt) o argílico/calcáreo (Bt[k]), un horizonte K, y un horizonte C. Sin embargo, a menudo el horizonte Bt(k) está truncado o puede estar ausente totalmente en un perfil determinado. El horizonte K tiene una zona superior laminada que se sitúa por encima de una capa carbonática endurecida. Por debajo de la capa dura se reconocen nódulos carbonáticos y material micrítico pulverulento. Los paleosuelos carbonáticos tienen estadios morfológicos que varían entre IV y VI. Los complejos edáficos cálcicos también presentan propiedades físicas que facilitan su identificación en el subsuelo. Las combinación de las variaciones de densidad y velocidad en paleosuelos y estratos sin paleosuelos da lugar a reflexiones importantes en los perfiles sísmicos y a rasgos distintivos en sondeos. Si bien la asociación entre complejos pedocálcicos y las discontinuidades regionales se describió por primera vez en estratos Terciarios del suroeste de Montana, estas mismas asociaciones se reconocen también en muchas otras zonas de las Great Plains y en otras partes del oeste de Estados Unidos Norteamericanos. La frecuente presencia de estas asociaciones en una zona tan amplia indica su utilidad como herramienta de correlación regional. Además, la delimitación de las discontinuidades regionales que constituyen límites de secuencias de gran escala, y que están marcados por estos complejos edáficos, tiene implicaciones importantes para aplicar en la estratigrafia secuencial de cuencas continentales. Palabras clave: paleosuelos cálcicos, secuencias, Terciario, edafocomplejos.
INTRODUCTION Tertiary continental strata of the Great Plains and western United States typically contain a multitude of various types of paleosols. In southwestern Montana, Tertiary paleosols commonly contain cambic, argillic, and calcic horizons; oxic horizons occur only within the basal portions of the Tertiary section (Hanneman, 1989). Of particular interest within these Tertiary continental deposits are calcic paleosols. Because of a marked climatic change to drying and cooling conditions within much of this area from ca. 33 Ma to ca. 4 Ma (Prothero, 1994, 1998; Wing, 1998; Retallack, 1992, 1998; Retallack et al., 2000), calcic paleosols commonly occur throughout the age equivalent part of the Tertiary section. Calcic paleosols with carbonate morphology stages IV and V occur within pedocomplexes at particular times within Tertiary basin fill of southwestern Montana. These times equate to regional unconformities in the northwestern United States that occurred at ca. 30 Ma, 20 Ma, and 4 Ma (Hanneman and Wideman, 1991; Hanneman et al., 1994, 2003). Consequently, these pedocomplexes mark sequence boundaries within continental Tertiary strata in southwestern Montana (Hanneman and Wideman, 1991; Hanneman et al., 1994); the sequence boundaries noted in southwestern Montana have recently been extended into central Washington (Hanneman et al., 2003) using criteria other than unconformity-bounding paleosols. The concept of using paleosols to define sequence boundaries in nonmarine strata has also recently been applied to other geologic settings. McCarthy et al. (1999) used interfluve paleo-
sols in the Cenomanian Dunvegan Formation of British Columbia to define sequence boundaries. Weissmann et al. (2002) marked sequence boundaries in Quaternary Kings River alluvial fan strata near Fresno, California, by laterally extensive, moderately mature paleosols and incised valley bases. Demko et al. (2004) used laterally continuous, mature paleosols to delineate regional unconformities within the Jurassic Morrison Formation of the U.S. Western Interior. Specifically for calcic paleosols, Gulbranson (2004) noted that calcretes within the Chinle Formation of the southwestern United States signify unconformities and delineate a terrestrial sequence stratigraphy for members of the Chinle Formation. Tandon and Gibling (1997) observed pedogenic nodular and underlying groundwater calcretes at sequence boundaries in Upper Carboniferous cyclothems in the Sydney Basin of Atlantic Canada. The purpose of this paper is to initially describe the calcic pedocomplexes in Tertiary basin fill of southwestern Montana. Because the calcic pedocomplexes do delineate regional unconformities, we will then detail their use as sequence boundary indicators in continental strata. The utility of using calcic pedocomplexes as sequence boundary markers will be further enhanced by documenting their existence within Tertiary strata of the Great Plains and western United States. CALCIC PEDOCOMPLEXES Calcic paleosol pedocomplexes typically occur within the Tertiary basin fill of many valleys in southwestern Montana
Calcic pedocomplexes (Fig. 1). The calcic pedocomplexes contain at least two calcic paleosols that are generally separated by small thicknesses of C horizon material. We define calcic paleosols informally as paleosols that have a large amount of secondary carbonate present in the form of calcic horizons (Machette, 1985). Although calcic paleosols have been placed into classifications such as Aridosols (Retallack, 1993), Calcisols (Mack et al., 1993), or paleo-Aridosols (Nettleton et al., 2000), we have not yet identified an A horizon within individual profiles of the southwestern Montana paleosol stacks, and there is typically, at best, only a truncated part of a B horizon within the profiles. Gardner et al. (1992) also noted the absence of the A and B horizons in Neogene calcic paleosol stacks of western Nebraska. These authors suggested that their absence may result from several factors such as: (1) the horizons generally not being well developed or very thick in some Aridosols, (2) the upward growth of the calcic horizon may
3
overprint the A and B horizon, and (3) the A and B horizons are more prone to erosion than petrocalcic horizons are. In any event, with the absence of a diagnostic surface horizon in the paleosol profile, we find that “calcic paleosols” is the most appropriate term for these paleosols. In former publications, we referred to the vertical configuration of calcic paleosols that we observed in southwestern Montana as calcic paleosol stacks (Hanneman and Wideman, 1991; Hanneman et al., 1994, 2003). However, instead of the term “paleosol stack,” we now prefer to use the term calcic “pedocomplexes” in accordance with the definition for pedocomplex as proposed to the Paleopedology Commission of International Union for Quaternary Research (INQUA). The proposed definition states that a pedocomplex is composed of two or more paleosols that are separated over large areas by a thin deposit of C horizon material, and are overlain and underlain by greater amounts
MONTANA
HELENA
BUTTE B Va ig H lle ol y e
Jefferson Valley
BOZEMAN
6
UNITED STATES
y
SCALE
n Valle
Horse Prairie
Madiso
DILLON
0
MONTANA
Gallatin Valley
Beaverhead Valley Ruby Valley
N
d
Deer Lodge Valley
Tos to Vall n-Tow ey nse n
Bitterro ot Valley
MISSOULA
Centennial Valley
KILOMETERS
Figure 1. Location map for southwestern Montana valleys, with selected valleys identified on a digital relief image of southwestern Montana (Montana State Library, NRIS data bank, 2001).
A
E
PC
PC
B
IP
IP
IP
Kl
PC
Ku
F
IP
IP
IP
D Figure 2. Calcic pedocomplexes and individual paleosols within the pedocomplexes, southwestern Montana (detailed locations for calcic pedocomplexes are given in Table 1). (A) Calcic pedocomplex located in lower Madison Valley, Madison Bluffs area. Maximum outcrop thickness of pedocomplex (PC) in photograph is ~6 m. (B) Closer view of individual calcic paleosols (IP) within the Madison pedocomplex. Note irregular boundaries on paleosols and the presence of C horizon material between the individual paleosols. Backpack at base of outcrop is 0.5 m in height. Maximum outcrop thickness of pedocomplex in photograph is ~5 m. (C) Calcic pedocomplex located in the upper Ruby Valley. (D) Individual calcic paleosols within the Ruby pedocomplex become welded when traced laterally. Global positioning system (GPS) unit at base of section is 15 cm in length. (E) Calcic pedocomplex located in the central Deer Lodge Valley. Outcrop thickness is ~5 m. (F) An individual calcic paleosol from the Deer Lodge pedocomplex that contains laminae within the upper part of the K horizon (Ku) and is underlain by a well-indurated carbonate sheet in the lower K horizon (Kl). Portion of Jacob staff in photo is 1 m in length.
IP = Individual Paleosol Ku = Upper Part of K Horizon in Individual Paleosol Kl = Lower Part of K Horizon in Individual Paleosol
PC = Pedocomplex
Legend
C
Calcic pedocomplexes of strata that contain weak to no evidence of soil development (Catt, 1998). Additionally, individual paleosols within a pedocomplex often are “…discontinuous, being in places truncated or cut out by small disconformities and/or amalgamated with other paleosols” (Morrison, 1998, p. 31). The term “pedocomplex” is synonymous with other terminology used in paleosol studies, such as compound and multistory paleosols. Figure 2 depicts calcic pedocomplexes found in some valleys of southwestern Montana. Each pedocomplex (Figs. 2A and 2C) contains at least two calcic paleosols and occurs between thick sections of nonpedogenically modified strata. And, as previously noted, individual paleosols (Figs. 2B, 2D, and 2F) may be discontinuous and/or amalgamated (amalgamated is synonymous with the terms “welded” and “composite”; see North American Commission on Stratigraphic Nomenclature, 1983; Morrison, 1998, p. 31) even when traced laterally over short distances. Nonetheless, the pedocomplex itself may be traced over a considerable distance. Surface Calcic Pedocomplex Paleosol Profiles The pedocomplexes characteristically contain several partial soil profiles. An individual profile may include in the most idealized scenario, in descending order, (1) an argillic (Bt) horizon, (2) an argillic/calcareous (Btk) horizon, (3) a K horizon, and (4) a C horizon (Fig. 3A). An argillic diagnostic subsurface B horizon (Bt) may be present in an individual soil profile of a pedocomplex. Bt horizons contain blocky structure; illuviated clays form bridges between grains and coat ped faces. The majority of Bt horizons in southwestern Montana Tertiary deposits are developed within tuffaceous mudstone, and thus their color range is very similar to pedogenically unmodified mudstone beds with very pale brown (10YR 5/4) to yellow gray (10Y 5/2) colors. In a few sections, Bt horizons are developed on sandy parent material, and the color range is more varied, from light reddish brown (5YR 6/3) to light brown (10YR 7/3). Root traces are common within the Bt horizon. Although the root casts and rhizoconcretions are typically calcareous, they may be also be composed of silica or sediments. Where these root structures are calcareous and are numerous, the horizon is better termed a Btk horizon. Root traces are from 0.1 cm to 2 cm in diameter and range up to 30 cm in length. The Bt(k) horizon is commonly truncated within the pedocomplex and can be entirely absent from a soil profile within the pedocomplex. However, where the horizon is preserved, it has a maximum observed thickness of 0.3 m. The K horizon (Fig. 3B) is the locus of secondary carbonate accumulation within the profile. As originally noted by Gile et al. (1965, p. 74) the carbonate is “present as an essentially continuous medium. It coats or engulfs, and commonly separates and cements skeletal pebbles, sand, and silt grains….” This type of carbonate is a K-fabric, and according to the definition originally set forth by Gile et al. (1965), a K horizon must have more than 90% K-fabric. Even though the K horizon has never
5
been formally accepted as a master horizon into Soil Taxonomy (Soil Survey Staff, 1975), we find it extremely helpful for use in separating the more weakly developed calcic horizons (Bk) from those horizons with major authigenic carbonate accumulations. The uppermost part of the K horizon contains laminations that range in thickness from 0.2 cm to 3 cm. The laminated part of the K horizon attains a maximum thickness of 0.3 m. A wellindurated sheet of carbonate occurs below the laminated zone. Floating skeletal grains, clasts, pisoliths, root casts, and some laminations are contained within the carbonate sheet (Figs. 3C and 3D). The hardpans are often fractured and brecciated. Maximum thickness of the hardpan part of the K horizon is 1 m. Powdery to indurated carbonate nodules are often present below the carbonate sheet (Fig. 3E). The nodular zone may also include micrite matrix material. More commonly, the chalky micritic matrix horizon underlies the nodular zone. This K horizon profile is similar to the pedogenic calcrete idealized profiles detailed by Esteban and Klappa (1983), Goudie (1983), and summarized by Alonso-Zarza (2003). Secondary silica, in the form of nodules, stringers, and silicified root traces commonly occurs in association with the K horizons (Fig. 3F). The silica nodules range from 5 to 20 cm along the long axis; the stringers vary from 1 to 5 cm in thickness. Both the nodules and stringers are usually located in the K-C horizon transition zone. The silicified root traces occur throughout the K to upper C horizon. Contact of the K horizon with the underlying C horizon is gradational. As stated already, the paleosol profile described here and shown in Figure 3 is an idealized profile. Not all features noted for the profile are typically found in every southwestern Montana calcic paleosol. The upper surface of the K horizon can be extremely irregular (Fig. 2B), and the entire paleosol can even be truncated when traced laterally. Individual paleosols become welded with other paleosols (Fig. 2D) within some pedocomplexes. However, there are usually enough profile characteristics present in field exposures to identify calcic paleosols. Subsurface Calcic Pedocomplexes Hanneman et al. (1994) documented the identification of calcic paleosol stacks, now termed calcic pedocomplexes herein, in the subsurface of the Deer Lodge Valley, southwestern Montana (Fig. 4A). Calcic pedocomplexes with accumulated thickness in excess of 10 m appeared in the subsurface as a collection of several relatively thin, high-velocity–high-density zones within the basin fill. Zone thickness ranged from 1 to 1.5 m. Density varied within the zones by as much as 0.6 g/cm3, and differed by as much as 0.9 g/cm3 from material immediately above these zones. Velocity differed by as much as 10 ft/ms (3.3 m/ms) from the overlying material and caused bright reflections on seismic sections. Synthetic seismograms were used to tie well-log and seismic data (Fig. 4B). The high-velocity–high-density zones in the Cenozoic basin fill were interpreted to be calcic paleosols based on data extracted
6
Hanneman and Wideman
B Horizons
K Horizon Laminations
C
K Horizon - Floating Clasts in Carbonate Sheet
D
K Horizon - Root Casts
Bt or Btk K
Legend Blocky Peds
K-C
transition
C
Rootlets Laminations Floating Grains Nodules Silica Stringers Mudstone
F
K - C Transition Zone Silica Stringers
E
K Horizon - Chalky/Nodular Zone Figure 3. (A) Idealized calcic paleosol profile (detailed locations for features in profile are given in Table 1). (B) Laminations in upper part of K horizon, lower Madison Valley. (C) Floating skeletal clasts in micrite of indurated sheet portion of K horizon, central Deer Lodge Valley. Lens cap is 67 mm in diameter. (D) Root casts in indurated sheet portion of K horizon, central Deer Lodge Valley. Film cap is 35 mm in diameter. (E) Chalky/nodular zone (indicated by arrow) present beneath indurated sheet of K horizon, lower Madison Valley. Quarter for scale. (F) Silica stringers (indicated by arrows) of the K horizon–C horizon transition zone, Jefferson Valley. Hammer is ~0.45 m in length.
A
from a suite of well logs that included sonic, density, resistivity, neutron, and lithology logs, and from well-cutting analyses (Fig. 4C). The pedogenic origin of the zones was shown by (1) wellcutting chips from the high-velocity–high-density zones that exhibited pedogenic features associated with calcic paleosols, (2) paleosol horizonation interpreted from well-log analysis, (3) the absence of minerals normally associated with lacustrine deposits, and (4) comparison with surface paleosols (Fig. 4D).
phologies as outlined by Machette (1985, p. 5; Table 1 therein). The stage IV morphology characteristics include laminae up to 1 cm in thickness in the upper part of the K horizon, with some laminae draped over fracture surfaces. Laminae of stage V are up to 3 cm in thickness. Fractures in the K horizon are typically coated with laminae, and pisolites are present. Thickness of the K horizon ranges from 0.5 to 1.5 m. Lateral Variation within Paleosol Stacks
Morphology of Calcic Paleosols The calcic paleosols within the calcic pedocomplexes of Tertiary basin fill in southwestern Montana have calcium carbonate morphologies consistent with the stage IV to stage V mor-
Although a calcic pedocomplex can be traced for several miles within a basin, lateral variation commonly occurs. The variance may be within individual paleosol profiles of the pedocomplex, in the vertical succession of horizons within a
Calcic pedocomplexes
A
7
B
MS 1-25 0.7
N
S
1.0
Two-Way Travel Time
1.5
0
Resistivity 70 ohms CNL 0.13 NPHI 0.62 1
Measured Surface Section Calcic Pedocomplex Deer Lodge Valley 100
915
Meters
930 10
945 Argillic Horizon K Horizon C Horizon
0
20
Depth (m)
945 MSP 1-25
100
930
Depth (m)
930
0
100 % Quartz
K Horizon Non-K Horizon
% Calcite Interpreted Pedocomplex
0 % Calcite
Depth (m)
924
Kilometers 1.0
Lithology Log MSP 1-25
D
945
0 MSP 1-25
C Figure 4. Hanneman et al. (1994) used well-log data, seismic data, and well cuttings analyses to define calcic paleosols and pedocomplexes in the subsurface of the Deer Lodge Valley, southwestern Montana. (Figure was modified from Hanneman et al., 2003.) (A) Geologic setting of the Deer Lodge Valley, southwestern Montana. Location of Montana State Prison (MSP) 1-25 well and seismic line of 1B are also shown. (B) Seismic-reflection line from the Deer Lodge Valley. Synthetic seismogram generated from well-log data of MSP 1-25 is tied to bright reflectors that occur on the seismic data at 1.0–1.1 s (~930–980 m in depth). (C) Paleosol profile delineated by resistivity and neutron log data. Argillic paleosol horizons are interpreted to have low resistivity; K horizons are interpreted where porosity is low on the neutron log. The overlay of these two logs depicts individual profiles within the mature pedocomplex. The K horizons also correspond to the interval’s high calcium content on the lithology log. CNL—compensated neutron log; NPHI—neutron porosity. (D) Matrix identification depth plot correlated with a surface pedocomplex measured in the northern Deer Lodge Valley. The thickness and frequency of increased calcite-content zones compare reasonably well with the K horizons of the surface pedocomplex.
complex, and in the overall thickness of the pedocomplexes (Fig. 5). Within individual profiles, soil descriptive features such as texture, color, root trace concentration, and horizon boundary distinctness often vary laterally, particularly within the Bt(k) horizons (Fig. 5A shows K horizon termination; Fig. 5B shows scoured K horizon top). These changes can be related to local soil-forming controls, such as topography, parent material texture, and scour events (McCarthy and Plint, 1998; McCarthy et
al., 1999). Lateral changes that affect soil horizon succession and overall pedocomplex thickness may be correlated to calcic profile initial development position and the variable deposition and/or erosion events associated with calcic profile formation. Typically, soil profile development begins on stabilized areas within a basin, such as interfluves or distal portions of alluvial fans (Alonso-Zarza et al., 1998; McCarthy et al., 1999). However, in order to generate a pedocomplex, episodic sedimentation
8
Hanneman and Wideman
needs to occur. Periodic deposition of sediment over the initial calcic soil profile provides more space for plant growth, and new cycles of calcic soil formation are begun. Given time, calcic soils will form over large parts of a basin, wherever surfaces become inactive. With episodic sedimentation, calcic pedocomplexes will eventually build over this larger area. The result of this soil-forming activity on a basin-wide scale is that (depending on a variety of factors, such as differential subsidence, climate, hydrology, parent materials) over time, thicker sections of calcic soils, and soils with somewhat different profiles, may exist in different areas. Alonso-Zarza et al. (1998) documented this pedocomplex variability on Pleistocene alluvial fan surfaces of the Campo de Cartagena–Mar Menor Basin, Murcia, southeast Spain. These authors noted the formation of different calcrete profiles in proximal and distal fan areas. These differences resulted from the interplay of erosion and deposition on the fan surfaces. In proximal fan areas, some soil horizons were stripped from a profile, while on the more stable areas, reworking and brecciation occurred, which would presumably produce a brecciated horizon. Conversely, in the distal fan areas, episodic sedimentation initially disrupted calcic soil formation, leading to another cycle of calcic soil formation. Collectively, these differences in the individual soil horizons of the various fan areas resulted in complex composite profiles being thicker in the distal fan areas than in proximal areas. CALCIC PEDOCOMPLEXES AS SEQUENCE BOUNDARY INDICATORS The calcic pedocomplexes present in the Tertiary basin fill of southwestern Montana developed over extended periods of time as evidenced by their advanced carbonate morphology stages. Soil development ceased for brief intervals because of sediment
influx, but then resumed, adding yet another soil profile to the pedocomplex. Collectively, the individual paleosol profiles contained within a pedocomplex represent significant breaks within the Tertiary basin-fill record. Consequently, the calcic pedocomplexes mark unconformities that occur between large-scale sedimentary packages. The age of each unconformity is constrained by paying strict attention to well-documented fossil vertebrate and radioisotopic age data taken from units occurring on both sides of the unconformity. The regional unconformities marked by calcic pedocomplexes occur at ca. 30 Ma, 20 Ma, and 4 Ma. The magnitude of each hiatus represented at these regional unconformities in southwestern Montana is estimated to be ~3–4 m.y. where all sequences are present. Because age data are derived not directly from the bounding surface itself but from strata that occur at some distance above and below pedocomplexes, there is yet a degree of uncertainty that exists for exact ages of the sequence bounding surfaces. Consequently, we are constantly looking for better age constraints on the regional unconformities. Montana Unconformity-Bounded Sequences Five unconformity-bounded sequences were initially delineated within continental Tertiary strata in southwestern Montana (Hanneman and Wideman, 1991; Hanneman et al., 2003). The sequences have upper and lower bounding surfaces that are unconformities of regional extent. We refer to these unconformity-bounded sequences as large-scale sequences, because they contain sizeable packages of basin-fill material. The unconformity-bounded sequences can include several hundred meters of strata, many different lithologies, and represent several million years of the geologic record. Calcic pedocomplexes mark the unconformities that separate four of these unconformity-bounded
A
B
Scoured K horizon K horizon termination Glove for scale
Figure 5. Examples of lateral variability that occurs within calcic pedocomplexes located in the lower Madison Valley of southwestern Montana (location of pedocomplex given in Table 1). (A) K horizon termination in a 20 Ma calcic pedocomplex. Terminated K horizon is ~0.3 m in thickness. (B) Scoured K horizon top (laminar zone and a part of the carbonate sheet). Glove is 24 cm in length.
Calcic pedocomplexes sequences. The regional unconformity-bounded sequences delimited by calcic paleosol stacks are informally designated as: sequence 2—middle/late Duchesnean to Whitneyan (ca. 38–30 Ma), sequence 3—Arikareean (ca. 27–20 Ma), sequence 4—Barstovian to Blancan (ca. 16–4 Ma), and sequence 5—early Quaternary (ca. 1.8 Ma) to the present (Fig. 6). Locations for examples of these calcic pedocomplexes and unconformity associations are given in Table 1. There are some differences among the calcic pedocomplexes that occur on the upper bounding surfaces of sequences 2, 3, and 4 in southwestern Montana. Where sequence 3 directly overlies sequence 2, pedocomplex development at the top of sequence 2
Epoch
Ma
Holocene
0.01 Pleistocene 1.75 Pliocene 5.3
North American Land Mammal Ages Rancholabrean Irvingtonian Blancan
Ma
0.1 1.75 4.9
Ma
9
typically contains a maximum of three paleosol profiles. The K horizon in these profiles has carbonate morphology equivalent to stage IV. However, in many locations, sequence 2 is overlain directly by sequence 4. In these areas, calcic paleosol stacks have several paleosol profiles and K horizons attain a carbonate morphology stage V. Calcic pedocomplexes at the top of sequence 3 have several paleosol profiles, and the K horizons in each profile reach a carbonate morphology stage V. It should be noted that in some past studies of southwestern Montana Tertiary basin fill, paleosols at this same stratigraphic level have been described as “red, saprolitic, and kaolinite-rich” (Thompson et al., 1982, p. 415;
Southwestern Montana
Washington
Sequence 5
High Cascade
Sequence 4
Walpapi
5
Hemphillian
9 Clarendonian
Miocene
Barstovian Hemingfordian
23.8
11.5 15.9
Calcic pedocomplexes 4 Ma
15
20 Ma
19
Arikareean
30 Ma
25
Sequence 3
Upper Kittitas
35
Sequence 2
Lower Kittitas
Oligocene 33.7
Whitneyan Orellan
30 32 33
Chadronian
Legend
Approximate time and duration of hiatus Newly defined hiatus
37 Duchesnean
40
Eocene Uintan Bridgerian
Challis
45 Sequence 1
47 50.4
Wasatchian
54.8 Clarkforkian
55.5 56.2
55
Tiffanian
Paleocene
60.5 Torrejonian
65
Puercan
63.5 65
65
Figure 6. Correlation of southwestern Montana sequences with central Washington (CW) sequences. The dashed lines within the CW Kittitas represent the newly recognized 30 Ma to 27 Ma hiatus. The gray area in between some of the wavy lines represents the estimated magnitude of the hiatus. Age estimates for the Cenozoic epochs are ones proposed by Berggren et al. (1995). Age estimates for Paleogene North American Land Mammal Ages (NALMA) are based on those given by Prothero (1995). Age estimates for Neogene NALMA are those delineated by Woodburne and Swisher (1995).
10
Hanneman and Wideman
TABLE 1. LOCATIONS FOR EXAMPLES OF TERTIARY CALCIC PEDOCOMPLEXES (PC) AND ASSOCIATED UNCONFORMITIES IN SOUTHWESTERN MONTANA Remarks Approximate age of Valley Section, township – NAD 1927, Zone 12, UTM U.S. Geological Survey 7.5ƍ unconformity location range easting/northing coordinates (Ma) quadrangle 4 Jefferson– SE ¼ section 34, T 387150 m E; 5031420 m N Beaverhead Rock, Vertebrate fossils below PC are Biltmore area 4 S, R 7 W Montana Hemphillian; mapped Quaternary gravels are above PC. 20 Upper Ruby NE ¼ section 3, T 9 406085 m E; 4992985 m N Belmont Park Vertebrate fossils below PC are Valley S, R 5 W Ranch, Montana late Arikareean; vertebrate fossils above PC are Barstovian. 20 Central Deer NW ¼ section 31, T 363500 m E; 5142162 m N Conleys Lake, Barstovian fossils are above Lodge 3 N, R 9 W Montana PC; mapped Arikareean strata are below PC. ? 20 (Lower Lower SW ¼ section 34, T 463440 m E; 5070289 m N Manhattan SW, Barstovian vertebrate fossils boundary age Madison– 1 N, R 2 E Montana are above PC; no age unconstrained) Madison constraints were found below, Bluffs area so they could range from Chadronian to Arikareean in age (37–19 Ma). 30 Jefferson– SE ¼ section 29, T 422640 m E; 5082290 m N Black Butte, Vertebrate fossils below PC are Golden 2 N, R 3 W Montana Chadronian; fragmentary Sunlight Mine oreodont fossils above PC most likely are Arikareean (W. Coppinger, July 2004, personal commun.). 30 Jefferson– NE ¼ section 28, T 413542 m E; 5073180 m N Whitehall, Vertebrate fossils below PC are Renova area 1 N, R 4 W Montana Chadronian; thin veneer of Quaternary loess is above PC.
Fields et al., 1985). Strata at the locations sampled for the oxic horizon were originally thought to be ca. 21–17 Ma. Later mapping with more-detailed age control revealed that strata at these sample localities are much older than previously thought. Consequently, recent work has shown no evidence for an oxic horizon at this stratigraphic level and that the regional unconformity at ca. 21–17 Ma is instead marked by calcic pedocomplexes (McLeod, 1987; Hanneman, 1989; Hanneman and Wideman, 1991; Portner and Hendrix, 2004). Calcic pedocomplexes that mark the upper surface of sequence 4 are similar to those at the top of sequence 3. However, these pedocomplexes are often absent in the southern areas of southwestern Montana, where there are no reported uppermost Tertiary strata and there are scant Quaternary age sediments. It may well be that much of this part of the section (including the calcic pedocomplexes) has been stripped from the basins due to recent uplift of the Yellowstone–Snake River Plain area of Idaho, Wyoming, and Montana. Washington Unconformity-Bounded Sequences The Cenozoic unconformity-bounded sequences identified in Montana extend into central Washington based upon work originally done by Cheney (1994, 2000). Hanneman et al. (2003) recognized that there are equivalent interregional unconformity-bounded sequences in this area: Lower Kittitas—ca.
36–30 Ma, Upper Kittitas—ca. 27–22 Ma, Walpapi—ca. 20–4 Ma, and High Cascade—ca. 4 Ma to present (Fig. 6). Although Cheney (1994, 2000) emphasized the importance of changes in lithology and provenance in initially delineating the Washington unconformity-bounded sequences rather than using the identification of pedocomplexes, the literature reports a caliche constraining the upper surface of the Walpapi Sequence at the Hanford Site (Pasco Basin) in south-central Washington. The caliche is developed on the Miocene-Pliocene Ringold Formation, and middle to late Pleistocene sediments overly it. The thickness of the caliche ranges from 0 to 20 m, and the unit is bounded by irregular surfaces having as much as 25 m of relief. The number of carbonate layers differs with the thickness of the deposits. Carbonate morphology of the layers varies from stage I to stage V. The caliche is interpreted to be pedogenic, although some modification to the paleosols by groundwater processes may have occurred (Slate, 1996). EXTENSION OF SEQUENCE BOUNDARIES DELINEATED BY CALCIC PEDOCOMPLEXES INTO THE GREAT PLAINS AND OTHER WESTERN U.S. AREAS The unconformity-bounded sequences cited above have been extended into the western United States and the northern Great Plains in previous studies by Hanneman and Wideman
Calcic pedocomplexes (1991), Cheney (1994, 2000), and Hanneman et al. (2003). Constenius et al. (2003, see their Fig. 19) expanded on these investigations and documented age-equivalent unconformity-bounded sequences throughout the Cordilleran orogenic belt that extends from southern Canada to Mexico. Based upon the interpretation of extensive structural data, Constenius et al. (2003) showed that the unconformity-bounded sequences record plate-tectonic interactions and continental deformation. Because age-equivalent Cenozoic unconformity-bounded sequences can be extended throughout the Great Plains and western United States, we expect that where equivalent soil-forming conditions prevailed, calcic pedocomplexes should delineate regional unconformities. A recent literature search revealed the likely identification of the paleosol–regional unconformity associations. Several occurrences of the calcic paleosol–regional unconformity associations at ca. 30 Ma, 20 Ma, and 4 Ma from
WASHINGTON
11
these areas are listed next. It is possible that many other occurrences of paleosol–regional unconformity associations are present in these areas. The locations of the paleosol–regional unconformity associations are shown in Figure 7. Regional Unconformity at ca. 30 Ma Pinnacles Lookout, Badlands National Park, Southwestern South Dakota Pinnacle Series paleosols occur in the top of the Poleside Member (early Oligocene) of the Brule Formation. The Pinnacle Series contains calcic paleosols that have prominent horizons of hard calcareous nodules at shallow depths. Elongate calcareous concretions that are interpreted as rodent burrows are also abundant in the paleosols. The calcic horizons probably only correspond to a stage II or at maximum stage III carbonate
NORTH DAKOTA
MONTANA
LEGEND Calcic Pedocomplexes/ Regional Unconformity Associations
SOUTH DAKOTA OREGON
4 Ma
IDAHO
20 Ma
WYOMING NEBRASKA
Southern High Plains with 4 Ma association
NEVADA COLORADO
UTAH
30 Ma
KANSAS
CALIFORNIA
OKLAHOMA
Southwestern Montana - All Associations Are Present
ARIZONA NEW MEXICO
TEXAS
0
500 Kilometers
N
Figure 7. Locations of calcic pedocomplexes and regional unconformities in the Great Plains and western United States. Details of locations and age constraints for the calcic paleosol pedocomplexes and regional unconformities are given in text.
12
Hanneman and Wideman
morphology. The Rockyford Ash (which has a radioisotopic age of close to 29 Ma) of the Sharps Formation unconformably overlies the Poleside Member of the Brule Formation (Retallack, 1983). Banner County, Western Nebraska The top of the upper Eocene to early Oligocene Brule Formation is locally cemented here with pedogenic nodular caliche. The Brule is overlain by gravel of the Neogene Ogallala Group (Gardner et al., 1992). Medicine Lodge Creek Valley, South-Central Idaho Nodular limestone armors the top of Medicine Lodge beds present in the Medicine Lodge Creek Valley, in Clark County, Idaho, and in sparse locations to the southwest as far as the southern Lemhi Range (Hodges and Link, 2002). The nodular limestone is ~2 m thick at the head of the south fork of Deep Creek, where it lies stratigraphically above a tuffaceous mudstone unit that has yielded a 40Ar/39Ar age of 30.23 ± 0.45 Ma. The nodular limestone is most likely pedogenic in origin (Hodges et al., 2004; M.K.V. Hodges, 2005, personal commun.). Regional Unconformity at ca. 20 Ma South Killdeer Mountains, Southwestern North Dakota (Medicine Hole Plateau, Dunn County) The Arikaree Formation contains a ledge-forming bed of carbonate/sandstone that is ~9 m in maximum thickness. This unit is known as the burrowed marker unit because it contains abundant fossil burrows (Forsman, 1986; Murphy et al., 1993). Delimata (1975) noted that this bed is an exceptional stratigraphic marker for the South Killdeer Mountains. He described the unit as containing tuffaceous limestone, nodular limestone, and banded limestone. Although Delimata interpreted the burrowed marker unit as a lacustrine deposit, its described features are more consistent with a pedogenic interpretation for the unit. The same marker bed may be present at White Butte, North Dakota, ~130 km southwest of the South Killdeer Mountains (Murphy et al., 1993). Presently, the burrowed marker unit is age constrained by: (1) a fission-track age of 25.1 ± 2.2 Ma taken from the base of the burrowed marker unit, and (2) the occurrence of two genera of oreodonts, Merychyus and Merycochoerus, located ~27 m above the stratigraphic position of the fission-track age. The range zones of these oreodonts overlap in the latest Arikareean to earliest Hemingfordian (Hoganson et al., 1998). Monroe Canyon, Nebraska The “terminal” paleosol at the head of Monroe Canyon, along the high rim, is ~4.6 m in thickness and is developed on the Harrison Formation. The paleosol appears to be a silcretecalcrete intergrade (Nash and Shaw, 1998), and it contains concentrations of rhizoliths and burrows, an upper laminar petrocalcic horizon, and a surface cemented as silcrete. Remnants of this
paleosol, the terminal Harrison paleosurface, are on flat-topped hills and buttes from Monroe Canyon west to the NebraskaWyoming state boundary (a distance of ~20 km). The Eagle Crag Ash, with a fission-track age of 19.2 ± 0.5 Ma, overlies the Harrison paleosurface by ~2 m; the Agate Ash, with a 40K/40Ar age of 21.9 Ma, occurs ~10 m below the Harrison paleosurface at Agate National Monument, in the Hoffman channel section (Hunt, 1990; MacFadden and Hunt, 1998). Regional Unconformity at ca. 4 Ma Kimball and Banner Counties, Western Nebraska Pedocomplexes of calcareous paleosols are present in the uppermost Neogene Ogallala Group, at the top of the Ash Hollow Formation, western Nebraska. The pedocomplexes are ~12 m thick, contain up to four paleosols, and each paleosol is ~1 m thick. The uppermost calcic paleosol in a pedocomplex has reached stage IV carbonate morphology, and the lower paleosols are between a stage III and IV carbonate morphology (Gardner et al., 1992). Hagerman Fossil Beds National Monument, Southwest Idaho A caliche is developed on Pleistocene-Pliocene gravels and forms a cap rock in most of the monument and the surrounding area. The caliche averages several meters in thickness, but thins to less than a meter locally. It is a very dense layer and contains vertical fractures that are often recemented (Farmer and Riedel, 2003). Southern High Plains, Texas and New Mexico The uppermost late Tertiary Ogallala Formation typically includes a stage V paleosol or up to two stage IV “caprock” calcic paleosols, and may have a stage VI calcic paleosol where the Quaternary Blackwater Draw Formation overlies it. Where the Blackwater Draw Formation is only a thin veneer or is entirely absent (as is the case in large portions of the western High Plains), the Ogallala calcic paleosol cap rock is 1.5–10 m thick, and has stage VI carbonate morphology. In these areas, it is probable that the pedogenic carbonate accumulations present within “…numerous buried calcic soils and the surface calcic soils of full sections of the Blackwater Draw have been welded onto the uppermost Ogallala calcrete” (Gustavson, 1996, p. 37). It is also possible that in certain areas, the Ogallala cap rock may range in age from late Miocene to late Quaternary. Roswell-Carlsbad, Southeastern New Mexico Stage VI calcic paleosols are developed on the top of the Ogallala Formation in this area. The age of the calcic paleosol is thought to be late Pliocene (Bachman, 1976; Machette, 1985). Morman Mesa, Southeastern Nevada The Morman Mesa calcic paleosol is ~2.5 m thick and has stage VI carbonate morphology. It is developed on red quartz sand
Calcic pedocomplexes of the Muddy Creek Formation. The age of the calcic paleosol is thought to be late Pliocene (Gardner, 1972; Machette, 1985). Vertebrate fossil remains of medial Hemphillian (late Miocene) age have been reported for the Muddy Creek Formation in the Morman Mesa area (Williams et al., 1997). Vidal Junction, Southern California A stage VI calcic paleosol is developed on the top of the Miocene–early Pliocene Muddy Creek Formation in this area. The age designated for this calcic paleosol is late Pliocene (Bull, 1974; Machette, 1985). DISCUSSION The significant areas of discussion that follow from our work on calcic pedocomplexes and their association with regional unconformities center on the usefulness of calcic pedocomplexes, or in fact, any type of mature paleosol, as sequence stratigraphic tools. Even with lateral variation of pedocomplexes, sequence boundaries can be defined when one combines other techniques for mapping unconformities. Additionally, although the primary control on the sequences described herein is tectonic, higherresolution work on the pedocomplexes, their adjacent strata, and better age constraints will help in understanding secondary controls of sequence and pedocomplex formation. Calcic pedocomplexes and calcic paleosols with stage IV to stage VI carbonate morphology are associated with regional unconformities of ca. 30 Ma, 20 Ma, and 4 Ma from the Great Plains through a large part of the western United States. These paleosol–unconformity associations mark large-scale regional sequence boundaries and consequently aid in surface and subsurface mapping of regional sequences. The calcic paleosols are easily identifiable in surface sections and have distinct physical properties that can be recognized in various types of geophysical data. Where basins contain several thousand feet of fill, and only have basin margins sections exposed, the ability to identify calcic pedocomplexes and use them to separate the subsurface geology into at least large-scale unconformity-bounded sequences is extremely advantageous in basin research. Additionally, the widespread extent of the calcic paleosols–regional unconformities associations enhances their utility as a regional correlation tool. It is important to note that the regional hiatuses recognized at ca. 30 Ma, 20 Ma, and 4 Ma are marked by many different sets of calcic pedocomplexes. Some of these pedocomplexes are laterally extensive over large areas, such as the Great Plains, but others formed within discrete depositional basins. Depositional basins began to form in the Cordilleran foreland fold-and-thrust belt by ca. 49 Ma (Hanneman, 1989; Hanneman and Wideman, 1991; Constenius, 1996; Constenius et al., 2003; O’Neill et al., 2004). Thus, pedocomplexes that formed in discrete depositional basins may be physically traced only within a particular basin. The pedocomplexes that mark these regional hiatuses were probably developed at similar times in various locations due to regional tectonic and climatic controls.
13
Because the regional unconformities defined at ca. 30 Ma, 20 Ma, and 4 Ma can be recognized throughout the Great Plains and the western United States, other types of mature paleosols should mark these same unconformities where climatic conditions differed. For example, in the Painted Hills of central Oregon, the 30 Ma regional unconformity separates the upper Big Basin Member of the John Day Formation from the overlying Turtle Cove Member of the John Day Formation. Mature ironrich paleosols are in the middle Big Basin Member and within the Big Basin Member, and the last one is located at the contact of the Big Basin Member and the overlying Turtle Cove Member of the John Day Formation (Bestland, 1997; Retallack et al., 2000). Even though a pedocomplex can be traced for up to several miles within a basin, lateral variation commonly occurs. The variance may be within individual paleosol profiles of the pedocomplex, in the vertical succession of horizons within a pedocomplex, and in the overall thickness of the pedocomplexes. The lateral variation is most likely related to factors such as the location of initial pedocomplex development within a basin, or the complex interplay of erosion and deposition rates (Tandon and Gibling, 1997; Alonso-Zarza et al., 1998; McCarthy et al., 1999; Weissmann et al., 2002). Although the lateral variance is easily recognized on the surface, the resolution of subsurface data may mask these differences. Where calcic paleosols or pedocomplexes are not present, angular stratal relationships, abrupt changes in provenance or lithologies, and the bases of incised valleys can also define sequence boundaries. These features can be mapped on the surface, and geometric patterns as indicators of unconformities can be recognized on seismic data. Collectively, these data types can be combined with paleosol information to complete the delineation of a sequence boundary. The calcic paleosols observed in southwestern Montana at the 30 Ma boundary are not as well developed (in regard to carbonate morphology and number of soil profiles within a pedocomplex) as those that mark the 20 Ma and 4 Ma regional unconformities. This appears to be a consistent feature of those boundaries throughout the Great Plains and western United States. The cause for this may be related somehow to a broad range of climate and/or tectonic controls, but presently, the actual reason for this difference in degree of paleosol development is not known. As stated previously, Constenius et al. (2003) have shown that the large-scale unconformity-bounded sequences defined in the northwestern United States by Hanneman and Wideman (1991), Cheney (1994, 2000), and Hanneman et al. (2003) are tectonically controlled sequences. However, future high-resolution work on these sequences will probably lead to an understanding of other secondary controls on their formation. The ages of the regional unconformities are given as approximate ages and are based upon currently available age constraints derived from radioisotopic age data and vertebrate faunal assemblages initially established in southwestern Montana. The ages appear to be fairly consistent across the Great Plains and western United States, but there is some range to these age designations. Historically, radioisotopic age
14
Hanneman and Wideman
data have been acquired in these geographic areas in order to place constraints on defining North American Land Mammal boundaries, the Cenozoic time scale, and the Cenozoic magnetopolarity time scale. Sparse attention has been paid to refining age constraints on regional unconformities. With additional age constraints, it may become apparent that there are timing differences among the regional unconformities. If the timing differences exist, they may be correlated to the time sweep on unconformity-bounded sequences boundaries noted by Constenius et al. (2003) and linked to regional tectonic events, or they may be indicators of timing differences in regional climatic change. In relation to the development of calcic pedocomplexes that mark regional unconformities, we find it of interest to contemplate the many Cenozoic relict calcic soils listed by Machette (1985, p. 11, Table 2 therein) for regions within the southwestern United States. These calcic soils may be young examples of the much older Tertiary calcic pedocomplexes. They may represent the different soils that could become pedocomplexes in a future geologic record. SUMMARY Calcic pedocomplexes with a maximum carbonate morphology of stage VI are associated with regional unconformities that have approximate ages of 30 Ma, 20 Ma, and 4 Ma. The calcic paleosols are easily identifiable in surface sections and have distinct physical properties that can be recognized in various types of geophysical data in the subsurface. The recognition of the calcic paleosol–unconformity association enables the separation of Cenozoic basin fill into at least large-scale unconformitybounded sequences, which can greatly enhance both surface and subsurface basin research. Although the 30 Ma, 20 Ma, and 4 Ma calcic pedocomplexes–regional unconformity associations were initially described in southwestern Montana, they can be traced throughout the Great Plains and western United States. The widespread extent of the calcic paleosols–regional unconformities associations enhances their utility as a regional correlation tool. Because the pedocomplexes delineate regional unconformities that are also large-scale sequence boundaries, the identification of the pedocomplex–unconformity association has broad implications for continental sequence stratigraphy. REFERENCES CITED Alonso-Zarza, A.M., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: Earth-Science Reviews, v. 60, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Alonso-Zarza, A.M., Silva, P.G., Goy, J.L., and Zaza, C., 1998, Fan-surface dynamics and biogenic calcrete development: Interactions during ultimate phases of fan evolution in the semiarid SE Spain (Murcia): Geomorphology, v. 24, p. 147–167, doi: 10.1016/S0169-555X(98)00022-1. Bachman, G.O., 1976, Cenozoic deposits of southeastern New Mexico and an outline of the history of evaporite dissolution: U.S. Geological Survey Journal of Research, v. 4, p. 135–149.
Berggren, W.A., Kent, D.V, Aubrey, M.P., Swisher, C., III, and Miller, K.G., 1995, A revised Cenozoic geochronology and chronostratigraphy, in Berggren, W.A., Kent, D.V, Aubrey, M.-P., and Hardenbol, J., eds., Geochronology, time scales and global stratigraphic correlations: Unified temporal framework for an historical geology: Tulsa, Oklahoma, Society for Sedimentary Geology, SEPM Special Publication 54, p. 129–212. Bestland, E.A., 1997, Alluvial terraces and paleosols as indicators of early Oligocene climate change (John Day Formation, Oregon): Journal of Sedimentary Research, v. 67, no. 5, p. 840–855. Bull, W.B., 1974, Geomorphic tectonic analysis of the Vidal region, in Information Concerning Site Characteristics, Vidal Nuclear Generating Station [California]: Los Angeles, Southern California Edison Company, Appendix 2.5B, amendment 1, 66 p. Catt, J.A., 1998, Report of working group on definitions used in paleopedology: Quaternary International, v. 51/52, p. 84, doi: 10.1016/S1040-6182(98)902324. Cheney, E.S., 1994, Cenozoic unconformity-bounded sequences of central and eastern Washington: Washington Division of Geology and Earth Resources, Bulletin, v. 80, p. 115–139. Cheney, E.S., 2000, Tertiary geology of the eastern flank of the central Cascade Range, Washington, in Woodsworth, G.J., Jackson, L.E., Nelson, J.L., and Ward, B.C., eds., Guidebook for Geological Field Trips in Southwestern British Columbia and Northern Washington: St. John’s, Newfoundland, Canada, Geological Association of Canada, Cordilleran Section, p. 205–227. Constenius, K.N., 1996, Late Paleogene extensional collapse of the Cordilleran foreland fold and thrust belt: Geological Society of America Bulletin, v. 108, p. 20–39, doi: 10.1130/0016-7606(1996)108<0020:LPECOT>2.3.CO;2. Constenius, K.N., Esser, R.P., and Layer, P.W., 2003, Extensional collapse of the Charleston-Nebo salient and its relationship to space-time variations in Cordilleran orogenic belt tectonism and continental stratigraphy, in Raynolds, R.G., and Flores, R.M., eds., Cenozoic Systems of the Western United States: Denver, Colorado, Society of Economic Paleontologists and Mineralogists, p. 303–353. Delimata, J.J., 1975, Petrology and geochemistry of the Killdeer carbonates [Ph.D. dissertation]: Grand Forks, North Dakota, University of North Dakota–Grand Forks, 256 p. Demko, T.M., Currie, B.S., and Nicoll, K.A., 2004, Regional paleoclimatic and stratigraphic implications of paleosols and fluvial/overbank architecture in the Morrison Formation (Upper Jurassic), Western Interior, USA: Sedimentary Geology, v. 167, p. 115–135, doi: 10.1016/j.sedgeo.2004.01.003. Esteban, M., and Klappa, C.F., 1983, Subaerial exposure environments, in Scholle, P.A., Bebout, D.G., and Moore, C.H., eds., Carbonate Depositional Environments: American Association of Petroleum Geologists Memoir 33, p. 1–96. Farmer, C.N., and Riedel, J., 2003, Water Resources Management Plan, Hagerman Fossil Beds National Monument: Hagerman, Idaho, National Park Service, 77 p. Fields, R.W., Rasmussen, D.L., Tabrum, A.R., and Nichols, R., 1985, Cenozoic rocks of the intermontane basins of western Montana and eastern Idaho, in Flores, R.M., and Kaplan, S.S., eds., Cenozoic Paleo-Geography of West-Central United States: Denver, Colorado, Society of Economic Paleontologists and Mineralogists, p. 9–36. Forsman, N.F., 1986, Documentation and diagenesis of tuffs in the Killdeer Mountains, Dunn County, North Dakota: North Dakota Geological Survey Report of Investigations 87, 13 p. Gardner, L.R., 1972, Origin of the Mormon Mesa caliche, Clark County, Nevada: Geological Society of America Bulletin, v. 83, no. 1, p. 143–156. Gardner, L.R., Diffendal, R.F., Jr., and Williams, D.F., 1992, Stable isotope composition of calcareous paleosols and ground-water cement from the Ogallala Group (Neogene), western Nebraska: Contributions to Geology, v. 29, no. 2, p. 97–106. Gile, L.H., Peterson, F.F., and Grossman, R.B., 1965, The K horizon: A master horizon of carbonate accumulation: Soil Science, v. 97, p. 74–82. Goudie, A.S., 1983, Calcrete, in Goudie, A.S., and Pye, K., eds., Chemical Sediments and Geomorphology: London, Academic Press, p. 93–131. Gulbranson, E.L., 2004, Calcretes and palustrine carbonates as indicators of climatic shift, Late Triassic and Early Cretaceous of the Southwestern US [abs.]: Geological Society of America Abstracts with Programs, v. 36, no. 5, p. 368. Gustavson, T.C., 1996, Fluvial and eolian depositional systems, paleosols, and paleoclimate of the Upper Cenozoic Ogallala and Blackwater Draw Formations, southern High Plains, Texas and New Mexico: Austin, Texas, Texas Bureau of Economic Geology, Report of Investigations no. 239, 62 p.
Calcic pedocomplexes Hanneman, D.L., 1989, Cenozoic basin evolution in a part of southwestern Montana [Ph.D. dissertation]: Missoula, Montana, University of Montana, 347 p. Hanneman, D.L., and Wideman, C.J., 1991, Sequence stratigraphy of Cenozoic continental rocks: Geological Society of America Bulletin, v. 103, p. 1335–1345, doi: 10.1130/0016-7606(1991)103<1335:SSOCCR>2.3.CO;2. Hanneman, D.L., Wideman, C.J., and Halvorson, J., 1994, Calcic paleosols: Their use in subsurface stratigraphy: American Association of Petroleum Geologists Bulletin, v. 78, p. 1360–1371. Hanneman, D.L., Cheney, E., and Wideman, C.J., 2003, Cenozoic sequence stratigraphy of northwestern USA, in Raynolds, R.G., and Flores, R.M., eds., Cenozoic Systems of the Western United States: Denver, Colorado, Society of Economic Paleontologists and Mineralogists, p. 135–156. Hodges, M.K.V., and Link, P.K., 2002, Oligocene-Miocene Medicine Lodge beds, northeastern Snake River Plain: Correlative to the Renova and Sixmile Creek Formations, Montana: Geological Society of America Abstracts with Programs, v. 34, no. 4, p. A48. Hodges, M.K.V., Link, P.K., and Miggins, D.P., 2004, The Medicine Lodge beds: New radiometric ages from Idaho Medicine Lodge Valley restrict deposition to the Paleogene basin-forming event: Geological Society of America Abstracts with Programs, v. 36, no. 5, p. 546. Hoganson, J.W., Murphy, E.C., and Forsman, N.F., 1998, Lithostratigraphy, paleontology, and biochronology of the Chadron, Brule, and Arikaree Formations in North Dakota, in Terry, D.O., LaGarry, H.E., and Hunt, R.M., Jr., eds., Depositional Environments, Lithostratigraphy, and Biostratigraphy of the White River and Arikaree Groups: Geological Society of America Special Paper 325, p. 185–196. Hunt, R.M., Jr., 1990, Taphonomy and sedimentology of Arikaree (lower Miocene) fluvial, eolian, and lacustrine paleoenvironments, Nebraska and Wyoming: A paleobiota entrenched in the fine-grained volcaniclastic rocks, in Lockley, M.G., and Rice, A., eds., Volcanism and Fossil Biotas: Geological Society of America Special Paper 244, p. 69–111. MacFadden, B.J., and Hunt, R.M., Jr., 1998, Magnetic polarity stratigraphy and correlation of the Arikaree Group, Arikareean (late Oligocene–early Miocene) of northwestern Nebraska, in Terry, D.O., LaGarry, H.E., and Hunt, R.M., Jr., eds., Depositional Environments, Lithostratigraphy, and Biostratigraphy of the White River and Arikaree Groups: Geological Society of America Special Paper 325, p. 143–165. Machette, M.N., 1985, Calcic soils of the southwestern United States, in Weide, D.L., and Faber, M.L., eds., Soils and Quaternary Geology of the Southwestern United States: Geological Society of America Special Paper 203, p. 1–21. Mack, G.H., James, W.C., and Monger, H.C., 1993, Classification of paleosols: Geological Society of America Bulletin, v. 105, p. 129–136, doi: 10.1130/00167606(1993)105<0129:COP>2.3.CO;2. McCarthy, P.J., and Plint, A.G., 1998, Recognition of interfluve sequence boundaries: Integrating paleopedology and sequence stratigraphy: Geology, v. 26, p. 387–390, doi: 10.1130/0091-7613(1998)026<0387:ROISBI>2.3.CO;2. McCarthy, P.J., Faccini, U.F., and Plint, A.G., 1999, Evolution of an ancient coastal plain: Paleosols, interfluves and alluvial architecture in a sequence stratigraphic framework, Cenomanian Dunvegan Formation, NE British Columbia, Canada: Sedimentology, v. 46, p. 861–891, doi: 10.1046/j.1365-3091.1999.00257.x. McLeod, P.J., 1987, The depositional history of the Deer Lodge Basin, western Montana [M.S. thesis]: Missoula, Montana, University of Montana, 61 p. Morrison, R.B., 1998, How can the treatment of pedostratigraphic units in the North American Stratigraphic Code be improved?: Quaternary International, v. 51/52, p. 30–33, doi: 10.1016/S1040-6182(98)90197-5. Murphy, E.C., Hoganson, J.W., and Forsman, N.F., 1993, The Chadron, Brule, and Arikaree Formations in North Dakota: Bismarck, North Dakota, North Dakota Geological Survey, Report of Investigation no. 96, 144 p. Nash, D.J., and Shaw, P.A., 1998, Silica and carbonate relationships in silcrete-calcrete intergrade duricrusts from the Kalahari of Botswana and Namibia: Journal of African Earth Sciences, v. 27, no. 1, p. 11–25. Nettleton, W.D., Olson, C.G., and Wysocki, D.A., 2000, Paleosol classification: Problems and solutions: CATENA, v. 41, p. 61–92, doi: 10.1016/S03418162(00)00109-0. North American Commission on Stratigraphic Nomenclature, 1983, North American Stratigraphic Code: American Association of Petroleum Geologists Bulletin, v. 67, p. 841–875. O’Neill, J.M., Lonn, J.D., Lageson, D.R., and Kunk, M.J., 2004, Early Tertiary
15
Anaconda metamorphic core complex, southwestern Montana: Canadian Journal of Earth Sciences, v. 41, p. 63–72, doi: 10.1139/e03-086. Portner, R., and Hendrix, M.S., 2004, Preliminary results of geologic mapping and sedimentologic analysis of the northeastern Flint Creek Basin, west central Montana [abs.]: Geological Society of America Abstracts with Programs, v. 36, no. 5, p. 72. Prothero, D.R., 1994, The Eocene-Oligocene transition: Paradise lost: New York, Columbia University Press, 281 p. Prothero, D.R., 1995, Geochronology and magnetostratigraphy of Paleogene North American Land Mammal “Ages”: an update, in Berggren, W.A., Kent, D.V., Aubry, M.-P., and Hardenbol, J., eds., Geochronology, time scales, and global stratigraphic correlations: Unified temporal framework for an historical geology: Tulsa, Oklahoma, Society for Sedimentary Geology, SEPM Special Publication 54, p. 305–316. Prothero, D.R., 1998, The chronological, climatic and paleogeographic background to North American mammalian evolution, in Janis, C.M., Scott, K.M., and Jacobs, L.L., eds., Evolution of Tertiary Mammals of North America, Volume 1: Terrestrial Carnivores, Ungulates and Ungulate Like Mammals: Cambridge, Massachusetts, Cambridge University Press, p. 9–34. Retallack, G.J., 1983, Late Eocene and Oligocene paleosols from Badlands National Park, South Dakota: Geological Society of America Special Paper 193, 82 p. Retallack, G.J., 1992, Paleosols and changes in climate and vegetation across the Eocene–Oligocene boundary, in Prothero, D.R., and Berggren, W.A., eds., Eocene–Oligocene Climatic and Biotic Evolution: Princeton, New Jersey, Princeton University Press, p. 383–398. Retallack, G.J., 1993, Classification of paleosols: Discussion: Geological Society of America Bulletin, v. 105, p. 1635–1637, doi: 10.1130/00167606(1993)105<1635:COPDAR>2.3.CO;2. Retallack, G.J., 1998, Fossil soils and completeness of the rock and fossil record, in Donovan, S.K., and Paul, C.R.C., eds., The Adequacy of the Fossil Record: Chichester, UK, John Wiley & Sons, p. 131–162. Retallack, G.J., Bestland, E.A., and Fremd, T.J., 2000, Eocene and Oligocene paleosols of central Oregon: Geological Society of America Special Paper 344, 192 p. Slate, J.L., 1996, Buried carbonate paleosols developed in Pliocene-Pleistocene deposits of the Pasco Basin, south-central Washington, USA: Quaternary International, v. 34–36, p. 191–196. Soil Survey Staff, 1975, Soil taxonomy; a basic system of soil classification for making and interpreting soil surveys: U.S. Department of Agriculture Handbook, v. 436, 754 p. Tandon, S.K., and Gibling, M.R., 1997, Calcretes at sequence boundaries in Upper Carboniferous cyclothems of the Sydney Basin: Atlantic Canada Sedimentary Geology, v. 112, p. 43–67, doi: 10.1016/S0037-0738(96)00092-9. Thompson, G.R., Fields, R.W., and Alt, D., 1982, Land-based evidence for Tertiary climatic variations: Northern Rockies: Geology, v. 10, p. 413–417, doi: 10.1130/0091-7613(1982)10<413:LEFTCV>2.0.CO;2. Weissmann, G.S., Mount, J.F., and Fogg, G.E., 2002, Glacially driven cycles in accumulation space and sequence stratigraphy of a stream-dominated alluvial fan, San Joaquin Valley, California, USA: Journal of Sedimentary Research, v. 72, p. 270–281. Williams, V.S., Bohannon, R.G., and Hoover, D.L., 1997, Geologic map of the Riverside quadrangle, Clark County, Nevada: U.S. Geological Survey Geologic Quadrangle Map GQ-1770, scale 1:24,000. Wing, S.L., 1998, Tertiary vegetation of North America as a context for mammalian evolution, in Janis, C.M., Scott, K.M., and Jacos, L.L., eds., Evolution of Tertiary Mammals of North America, Volume 1: Terrestrial Carnivores, Ungulates and Ungulate Like Mammals: Cambridge, Massachusetts, Cambridge University Press, p. 37–65. Woodburne, M.O., and Swisher, C.C., III, 1995, Land mammal high-resolution geochronology, intercontinental overland dispersals, sea level, climate, and vicariance, in Berggren, W.A., Kent, D.V., Aubry, M.-P., and Hardenbol, J., eds., Geochronology, time scales, and global stratigraphic correlations: Unified temporal framework for an historical geology: Tulsa, Oklahoma, Society for Sedimentary Geology, SEPM Special Publication 54, p. 337–364. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006 Printed in the USA
Geological Society of America Special Paper 416 2006
A Late Triassic soil catena: Landscape and climate controls on paleosol morphology and chemistry across the Carnian-age Ischigualasto–Villa Union basin, northwestern Argentina Neil J. Tabor Department of Geological Sciences, Southern Methodist University, Dallas, Texas 75275-0395, USA Isabel P. Montañez Kelley A. Kelso Department of Geology, One Shields Ave., University of California, Davis, California 95616, USA Brian Currie Department of Geology, Miami University, 114 Shideler Hall, Oxford, Ohio 45056, USA Todd Shipman Arizona Geological Survey, 416 West Congress, Suite 100, Tucson, Arizona 85701, USA Carina Colombi Instituto y Museo de Ciencias Naturales (CONICET), Espana 400(N) Ciudad San Juan, CP5400, Argentina ABSTRACT The thicknesses of stratigraphic sections of the Late Triassic (Carnian) Ischigualasto Formation change significantly, from ~300 to 700 m, along a 15 km transect in the Ischigualasto Provincial Park, San Juan, NW Argentina. Channel sandstone deposits dominate the thickest section, whereas pedogenically altered layers dominate the thinnest stratigraphic section. Eight paleosol types have been recognized in the study area, and they are unevenly distributed across the basin. In particular, paleosol B horizons are thinner and redoximorphic soil morphologies dominate in the thickest, whereas B horizons are thickest and argillic and calcic morphologies dominate in the thinnest stratigraphic section. These observations suggest that the geomorphic evolution of the Ischigualasto basin exerted the primary control on sediment distribution, depositional rate, soil drainage, and depth of the groundwater table through most of Late Triassic time in the Ischigualasto basin. In addition, δ18O values of paleosol calcite nodules are similar to modern soil calcites that form in frigid to cool climates between ~0 °C and 10 °C. Considering both lateral and stratigraphic distribution of paleosol morphological variability, there appears to be three different general modes of climate recorded throughout deposition of the Ischigualasto Formation: (1) Humid conditions recorded by Argillisols, Gleysols, and Vertisols in the lower quarter of the formation; (2) relatively dry conditions recorded by Calcisols, calcic Argillisols, and calcic Vertisols in the middle half of the formation; and (3) generally more humid conditions in the upper quarter of the formation recorded by Argillisols, Gleysols, and Vertisols. Keywords: paleosols, catena, paleoclimate, Triassic, Argentina. Tabor, N.J., Montañez, I.P., Kelso, K.A., Currie, B., Shipman, T., and Colombi, C., 2006, A Late Triassic soil catena: Landscape and climate controls on paleosol morphology and chemistry across the Carnian-age Ischigualasto–Villa Union basin, northwestern Argentina, in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 17–41, doi: 10.1130/2006.2416(02). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
17
18
Tabor et al. RESUMEN La Formación Ischigualasto del Triásico Superior (Carniense) presenta cambios importantes de espesor (de 300 a 700 m), a lo largo de una transversal de 15 km dentro del Parque Provincial de Ischigualasto, San Juan, en el noroeste de Argentina. En las zonas en las que la Formación es más potente dominan los canales de areniscas, mientras que en las zonas en las que el espesor de la Formación es menor dominan los niveles edáficos. Se han reconocido ocho tipos distintos de paleosuelos, que se distribuyen de forma desigual a lo largo de la cuenca. En particular, los horizontes B de los paleosuelos son menos potentes y presentan morfologías redoximórficas en las secciones estratigráficas de mayor espesor; por el contrario, en las secciones estratigráficas de menor espesor los horizontes B son más potentes y argílicos y en ellos son frecuentes los rasgos calcáreos. De forma conjunta, la distribución de los depósitos canalizados de areniscas y la morfología de los paleosuelos a lo largo de la Formación Ischigualasto indican que la evolución geomorfológica de la cuenca fue el principal factor de control sobre la distribución de los sedimentos, la tasa de sedimentación, el drenaje de los suelos y la profundidad del nivel freático durante la mayor parte del Triásico. Palabras clave: paleosuelos, catena, Triásico, paleoclima, Argentina.
INTRODUCTION Soil morphologies vary across the modern landscape in response to local and regional variations in depositional environment, soil drainage, and geomorphology (Jenny, 1941; Soil Survey Staff, 1975; Buol et al., 1997; Birkeland, 1999, and many others). Soil scientists have long recognized that landscape position is a primary factor of soil formation, and the term “soil catena” is reserved for specific instances where lateral variations in the development of a suite of contemporaneous soil profiles occur in accord with the position of those soils on the landscape (Steila, 1976). It is often difficult to document the catenary relationships of ancient sedimentary basins, because (1) accurate correlation and demonstration of contemporaneity between outcrops is often not feasible and (2) an ancient soil catena may be erased by the terrestrial sedimentary record’s bias toward preservation of low-lying areas. Several studies, however, provide persuasive evidence that variations in paleosol development in ancient sedimentary basins can reflect lateral changes in depositional rates across the paleolandscape, both of which can be related to landscape position and the location of the groundwater table (Allen, 1974; Leeder, 1976; Retallack, 1976, 1977, 1983; Bown and Kraus, 1981; Kraus, 1987; Besly and Turner, 1983; Atkinson, 1986; Tabor and Montañez, 2004; McCarthy and Plint, 2003; Demko et al., 2003). Although they are called by different names (e.g., pedofacies), these basin-scale studies of terrestrial sedimentary rocks demonstrate that catenary relationships can be recognized in ancient sedimentary records, even when the contemporaneity of individual paleosol horizons cannot be demonstrated. This recognition is important given the impact that paleosol morphology and composition have on paleoenvironmental and paleoclimatic reconstructions (e.g., Retallack, 1990; Mack and James, 1994).
In this work, we document the distribution of paleosols across the Late Triassic (Carnian) Ischigualasto Formation of the Ischigualasto–Villa Union basin, northwestern Argentina. The spatial and temporal distribution of paleosol morphologies establishes a basis for inferred lateral changes in soil drainage and depositional rates indicative of a paleosoil catena. In particular, lateral variations in the stratigraphic distribution of channel sandstones, pedogenic alteration, redoximorphic features, and pedogenic calcrete profiles appear to provide a sensitive record of changes in soil drainage during Ischigualasto Formation deposition. Overall, paleosol morphologies indicate that soils were generally better drained in areas of decreased accommodation, most likely reflecting the effects of a deeper groundwater table and generally drier conditions in the shallow subsurface away from the basin depocenter. Enhanced soil drainage and a reduction in the overall sediment-accumulation rate allowed more illuvial Ca2+ to accumulate as pedogenic calcite precipitation in paleosols developed on the alluvial landscape away from the basin center. BACKGROUND Tectonic Setting, Stratigraphy, and Age The study area is located along the southern edge of the Ischigualasto–Villa Unión basin in northeastern San Juan Province, Argentina, and lies within the boundaries of the Ischigualasto Provincial Park (Fig. 1). During Mesozoic time, oceaniccontinental plate interactions along the southwestern margin of Pangea produced a region of extensional deformation cratonward of the proto-Andean magmatic arc (Ramos and Kay, 1991; López-Gamundí et al., 1994). Extension focused along the NW-trending boundary between Paleozoic accreted terranes and
A Late Triassic soil catena
19
Figure 1. Location maps of the Ischigualasto basin. (A) Map of San Juan Province, Argentina. White rectangle shows location of Figure 1C. (B) Distribution of Triassic basins in southern South America. “I.B.” marks the position of the Ischigualasto–Villa Unión basin in NW Argentina. (C) Geologic map of the southern part of the Ischigualasto–Villa Unión basin. Ischigualasto Formation measured sections 1–3 are shown in Figure 3. Figure was modified from Alcober (1996).
the Precambrian Gondwanan craton (Uliana et al., 1989). The Ischigualasto–Villa Unión basin of northwest Argentina is one of a series of continental-rift basins that developed in the region as a result of this extension (Fig. 1B) (Uliana and Biddle, 1988). Rift-related deposition in the Ischigualasto basin began during Early Triassic time as normal displacement on the paleo– Valle Fértil fault led to the development of a structural half-graben (Milana and Alcober, 1994). Deposition in the basin continued throughout Triassic time and resulted in accumulation of up to 3.5 km of nonmarine and volcanic strata (Alcober, 1996). The Triassic System in the basin consists of the Lower Triassic Talampaya and Tarjados Formations, the Middle Triassic Chañares-Ischichuca and Los Rastros Formations, and the Upper Triassic Ischigualasto and Los Colorados Formations (Fig. 2) (Stipanícic and Bonaparte, 1979). Quaternary shortening in the Andean foreland produced reverse-displacement reactivation of Mesozoic normal faults and structural inversion of Ischigualasto basin strata (Zapata and Allmendinger, 1996). The rocks of the Ischigualasto basin are exposed in the hanging wall of the Valle Fértil and Alto faults, both of which are interpreted to be reverse-reactivated zones of Triassic normal faulting (Milana and Alcober, 1994; Zapata and Allmendinger, 1996). The Valle Fértil fault is interpreted as the main basin-bounding normal fault separating Proterozoic-Paleozoic crystalline and sedimentary rocks of the footwall from the Mesozoic sedimentary rocks of the hanging wall. The paleo–Alto fault is interpreted as a W-NW–dipping normal fault (Milana and Alcober, 1994) that merged with the Valle Fertíl fault to the south (Fig. 1).
Ma Period
Stage
Formation/Member
Norian
Los Colorados
Radiometric Age (Ma)
223.0±0.4 Carnian
Ischigualasto 227.8±0.3
Ladinian
Los Rastros
Anisian
Chanares
Olenekian Induan
Figure 2. Time-stratigraphic chart for Triassic rocks of the Ischigualasto basin including relative position of radiometrically dated horizons. Time scale was adapted from Gradstein et al. (1995) and Alcober (1996).
20
Tabor et al.
The focus of this investigation is the Ischigualasto Formation, which consists of ~300–700 m of mudstone, sandstone, conglomerate, and extrusive basalt (Fig. 3). The basal Ischigualasto Formation is defined as the first occurrence of pebble-cobble conglomerate-conglomeratic sandstone or smectitic mudstone overlying the carbonaceous sandstones and mudstones of the Los Rastros Formation, whereas the upper contact of the Ischigualasto Formation is positioned at the top of the final variegated mudstone below the dominantly red-colored sandstones and mudstones of the Los Colorados Formation. The Ischigualasto Formation appears to have conformable contacts with underlying and overlying formations (Alcober, 1996), and it is internally divided into four members (Fig. 3). In stratigraphically ascending order, these members are (1) Unit I, which consists of ~30–50 m of tan/gray pebble-cobble conglomerate, conglomeratic sandstones, and green/gray smectitic mudstone; (2) Unit II, which consists of 65–125 m of mudstone and sandstone with rare interbeds of bentonite and basalt; (3) Unit III, which consists of 250–470 m of mudstone and sandstone, the lithostratigraphy of which is dominated by smectitic mudstones in the east and channel and overbank sandstone deposits in the west portions of the study area; and (4) Unit IV, which consists of 35–65 m of variegated mudstone and sandstone. An Upper Triassic age of the Ischigualasto Formation is based on vertebrate fossils and radiometric ages of altered ash beds from the unit. Abundant vertebrate fossils from the lower two-thirds of the formation indicate a Carnian age of deposition (Rogers et al., 1993; Alcober, 1996). Altered ash beds in the Ischigualasto Formation have provided additional chronostratigraphic control. Sanidine crystals from a bentonite sampled ~80 m above the base of the formation yielded an 40Ar/39Ar date
Section 1
Section 2
Section 3
Unit IV 7 km
8 km
Unit III
Unit II
Ischigualasto Formation
Los Colorados Fm
Unit I
of 227.8 ± 0.3 Ma (Rogers et al., 1993), while plagioclase crystals from a bentonite ~70 m from the top of the formation yielded a date of 218 ± 1.7 Ma (Shipman, 2004). Collectively, these data support a Carnian age of deposition based on the Triassic time scale of Gradstein et al. (2004; Fig. 2). Paleoclimate Paleomagnetic evidence from intraformational basalt flows within the upper Chañares Formation, near the base of the Ischigualasto Formation, place the basin at 30°S during the Middle Triassic (Veevers et al., 1994; Valencio et al., 1975; López-Gamundí et al., 1994). Stipanícic and Bonaparte (1972) suggested that sediments in the Ischigualasto basin were deposited as a single climatic cycle, from relatively dry and seasonal, to more humid and back again, with temperatures ranging from moderate to hot. Stipanícic and Bonaparte’s (1972) reconstruction primarily relied on lithologic and flora/faunal evidence in the Agua de la Peña Group. They also considered the possibility that some of the change in the rock record could be a result of tectonic influences and that climate was relatively uniform. Volkheimer (1969) supported an interpretation of moderate to hot temperatures mostly based on paleofloral data. Previous researchers argued that the Ischigualasto Formation (Fig. 2) was subject to a water-limited, seasonal climate. Based on the morphology, mineralogy, and light stable isotope geochemistry of pedogenic minerals from a paleo-Vertisol, Tabor et al. (2004) suggested that the Lower Ischigualasto Formation was deposited in a seasonal, humid, and cool climate. Root sizes of rhizoliths in the Ischigualasto Formation indicate mean annual precipitation from 500 to 800 mm (Alcober et al., 1997). Bossi (1971) proposed a dominantly fluvial setting that experienced seasonal variations in water availability, while Martínez (1994) suggested an arid climate with seasonal precipitation, supported by lithological evidence such as secondary paleosol carbonate, slickensides, and blocky peds. Further, Spalletti et al. (1999) presented biozonation in relation to the chronostratigraphy of the basin that indicated a mixed forest plant assemblage for the Ischigualasto Formation. The landscape of the Ischigualasto Formation was sparsely forested with a dominance of herbaceous plants, although some large trees, represented by Rexoylon, are common in the upper half of the formation, and Spalletti et al. (1999) interpreted the deposition of the Ischigualasto Formation to have occurred in a dry, and moderate to hot, seasonal climate. METHODS
OLDER TRIASSIC
Figure 3. Schematic diagram of changing sedimentary thickness across the Ischigualasto Formation, position of informal lithostratigraphic members, and measured sections 1, 2, and 3. See Figure 1 for location.
Three detailed stratigraphic sections were measured through the Ischigualasto Formation in the eastern part of the Ischigualasto Provincial Park (Currie et al., 2001). The sections, each separated by ~7–8 km, are labeled 1, 2, and 3 from west to east.
A Late Triassic soil catena Paleosol tops were identified on the basis of the upper limit of observed pedogenic features or by the presence of an erosional contact with an overlying bed, whereas profile bases were delineated at the lowest occurrence of unaltered parent material. Field descriptions of paleosols (e.g., thickness, color, type, and distribution of mottling, soil structure, and mineralogy, size, morphology, and distribution of authigenic minerals) were completed following established methods (Tabor and Montañez, 2004). Paleosol and lithologic colors were identified from dry samples using Munsell color charts (Munsell Color, 1975). Paleosol classification followed the system defined by Mack et al. (1993), and the closest estimate of a modern soil taxonomic equivalent of the paleosol profiles is given within the context of the U.S. Department of Agriculture Soil Taxonomy (USDA) (Soil Survey Staff, 1975). For clay mineral identification, suspensions of the <2 μm fraction were split into 3 aliquots that were concentrated by filtration and prepared as oriented mounts using the filter membrane peel technique. Each aliquot was treated either with 1 M KCl, 0.5 M MgCl, or a mixture of 0.5 M MgCl and 1:4 glycerolwater solution to delineate changes in basal thickness of different clay minerals with X-ray diffraction. The 1 M KCl aliquots were analyzed at room temperature, and reanalyzed after heating to 550 °C for 2 h, to determine clay mineralogy based on temperature stability (Moore and Reynolds, 1997). Clay mineral samples were analyzed at Southern Methodist University (SMU) with a Rigaku Ultima III X-ray diffractometer using CuKα radiation between 2° and 14° 2Θ at a rate of 1° 2Θ per minute. Identification of clay mineralogies was made based upon the methods described in Moore and Reynolds (1997). Carbonate nodules and tubules were collected near the base of each carbonate-nodule–bearing horizon associated with paleosol profiles. Doubly polished thin sections were analyzed petrographically to differentiate pedogenic carbonate fabrics from groundwater and diagenetic calcite cements following the methods in Deutz et al. (2001) and Moore (2003). Microsamples were obtained from pedogenic micrite in carbonate nodules and tubules; replacement of micrite and microspar or sparry calcite cements was avoided. Samples were drilled from thin sections or matching billets with a hand-held dental drill equipped with faceted 100 μm diamond bits. Approximately 50 μg of carbonate powder was roasted at 375 °C for 3 h to remove organics. Oxygen isotope analysis was carried out on a Fisons-Optima infrared gas source mass spectrometer in the Department of Geology at the University of California–Davis and is reported in per mil notation, where ⎡R ⎤ δ 18O = ⎢ sample − 1⎥ × 1000 ⎢⎣ Rstandard ⎥⎦ and R = 13C/12C. The standard used to report δ18O values of calcite is the Peedee belemnite (PDB; Craig 1953), and interpreted meteoric water δ18O values are given relative to the standard mean ocean water (SMOW; Gonfiantini, 1978). Replicate analy-
21
sis of NBS 19 yielded δ18OPDB values of −2.07 ± 0.06 (n = 39) over the period of analysis. RESULTS Overview of Relevant Pedogenic Features Major diagnostic indicators of pedogenesis in Carnian-age strata of the Ischigualasto Formation are pedogenic horizons, soil structure, and fossil root traces. Profiles are subdivided into horizons on the basis of color, distribution of redoximorphic features, and translocated or authigenic minerals, and density of rooting structures, as well as the presence and down-profile change in paleosol structure and fossil root trace distribution. The presence of horizons indicates that soil-forming processes operated on a relatively stable substrate for sufficient time to reorganize the parent material into zones of alteration, accumulation, and removal (e.g., Buol et al. 1997; Retallack, 1990; Mack and James, 1994). The fundamental component of soil structure, the ped, and its pedogenic coatings, cutans, can be important indicators of the soil-forming environment and soil drainage conditions (Brewer 1976; Buol et al., 1997; Retallack 1988, 1990). Paleosols from the Ischigualasto basin exhibit platy, massive, angular, prismatic, and wedge-shaped peds with carbonate, iron- and/or manganeseoxide, and clay cutans. The wide range in density, morphology, and structure of root traces present within paleosols from the study area delineates both the relative position of the paleo–water table and the paleoenvironmental conditions of plant growth (e.g., Retallack, 1990). A good indicator of soil moisture regime and drainage is the presence or absence of gley colors and redoximorphic morphologies (e.g., Soil Survey Staff, 1975, 1998). Those portions of a soil profile that contain gley colors (chroma < 2 and value > 4 on the Munsell color charts) and/or mottling consistent with modern surface-water gley in modern soil systems (see Retallack, 1990; Vepraskas, 1994; PiPujol and Buurman, 1994) are referred to as redoximorphic zones (e.g., Vepraskas, 1994). In modern soils, redoximorphic features form in seasonally saturated portions of profiles through removal of Fe and Mn from areas of low Eh (redox depletions) and reprecipitation as Fe- and Mn-oxides (redox concentrations) in more oxidized areas. Gley matrix colors indicate reduced conditions, which are typical of relatively prolonged saturation (25–50% of the year), whereas yellow-brown to reddish mottles record seasonal soil drying (Daniels et al., 1971; Duchaufour, 1982). Using modern soils as an analogy, redoximorphic features in paleosols are interpreted to have formed in seasonally saturated portions of the profile that contained sufficient organic content to have undergone reducing conditions (PiPujol and Buurman, 1994). Soil Mineralogy Quartz, feldspars, and micas are the dominant detrital siliciclastic minerals in both fluvial channel sands and paleosol
22
Tabor et al.
profiles based on petrographic and X-ray diffraction analyses (Moore, 2003; Shipman, 2004). These detrital siliciclastic minerals also comprise a significant fraction of the clay mineralogy in soils that formed over sedimentary parent materials in the Ischigualasto Formation (pedotypes A–F described in next section; see also Moore, 2003). However, two different paleosol types (pedotypes G and H described in next section) apparently preserve in situ weathering products of volcanic ash and basalt and therefore have little or no detrital siliciclastic component. The mineralogical composition of each pedotype horizon is given in Table 1 based upon the response of the respective <2 μm size fractions to various chemical and thermal treatments discussed in the methods section. Figure 4A–C presents X-ray diffractograms of <2 μm clay mineral fractions from pedotypes C, G, and H (see description in next section). The majority of paleosols is dominated by 2:1 phyllosilicate mineralogies. Expansion of X-ray diffraction peaks near 12–15 Å in Mg-treated samples to 17.5–17.9 Å in Mg + glycerol treated samples corresponds to d(001) spacing of smectite-group minerals (Fig. 4A–C, Table 1), probably montmorillonite (Tabor et al., 2004). The persistence of an ~10 Å peak after Mg + glycerol treatment in pedotype D corresponds to the d(001) of a mica-like mineral (MLM in Table 1), similar to muscovite (Fig. 4A). In addition, the invariant position of the peak near 7.2 Å in pedotype H (Fig. 4C), and its disappearance after heating at 550 °C, corresponds to the d(001) peak of kandite group minerals, probably kaolinite (Tabor et al., 2004). Note that MLM, which is likely a detrital siliciclastic component not associated with pedogenic processes (Tabor et al., 2004), is an important constituent of paleosol profiles in the fluvial-alluvial association, whereas it is an insignificant constituent of paleosols in the volcanic association (see paleosol associations in next section). Paleosol Descriptions, Classifications, and Interpretations Based on the field inspection of over 500 paleosol profiles in the Ischigualasto Formation, eight major types of paleosols have been recognized that represent the majority of the observed morphological variability (e.g., horizonation, structure, fabric, color, mineralogy; Fig. 5). Table 1 presents detailed morphological and mineralogical characteristics of representative profiles from each major paleosol type. Next, we present generalized descriptions of the characteristics of each of the eight paleosol types. We also discuss the morphological variability among paleosol types, as well as their stratigraphic and lateral distribution within the study area, and classify each paleosol type using the criteria of Mack et al. (1993) along with the closest estimated soil taxon within the context of the USDA Soil Classification System (Soil Survey Staff, 1975, 1998). Paleosol types are unevenly distributed, both laterally and vertically, through the Ischigualasto Formation strata (Table 2). In terms of parent material, there are two distinct paleosol assemblages: a fluvial-alluvial association (A–F) and a volcanic association (G and H).
Fluvial-Alluvial Association Type A Paleosols Description. Type A paleosols exhibit rooting structures and partial to complete obliteration of original depositional features (Figs. 5 and 6). These paleosols exist as both muddy and sandy profiles 30–250 cm thick, with weak, tubular/vermicular green and red mottles with little to no soil structural development or horizonation. Rooting structures, if present, are typically poorly defined root halos. The clay mineral composition of the <2 μm fraction is dominated by quartz and MLM. These paleosols are typically associated with the upper surfaces of channel and crevasse-splay sandstones (Figs. 5 and 6) and have highly variable lateral continuity throughout the formation. Type A paleosols are the most common suite of observed paleosols in the Ischigualasto Formation and occur in all four members, but are most abundant in Units II and III. Interpretation. Paleosols with these characteristics are classified as Protosols (Mack et al., 1993), which also roughly covers the range of characteristics observed in the USDA Soil Taxonomy soil order Entisols and some suborders of Inceptisols. The presence of Protosols in the Ischigualasto Formation indicates that sedimentation within the basin ceased long enough for colonization by terrestrial flora or conditions for soil formation were exceedingly unfavorable (Buol et al., 1997). Based on the close stratigraphic association of type A Protosols with channel and channel-proximal deposits and a clay mineral fraction that is dominated by detrital components, it is most likely that the temporal development of these paleosols was limited due to frequent deposition along stream corridors. Type B Paleosols Description. Type B paleosols are smectite and MLMrich mudstone and claystone profiles 50 to >250 cm thick. They all have wedge-shaped aggregate structure in lower horizons (Bss horizons) that grade upward to angular blocky or prismatic horizons (Fig. 5). In addition, these wedge-shaped aggregates define large, arcuate bounding surfaces that are lined with highly oriented and foliated clay-sized material (i.e., slickensides). The upper horizons also exhibit sand or mudfilled clastic dikes that may extend downward through the profiles as much as 125 cm. Most of these paleosols have weak red to gray matrix colors with fine to coarse mottling, redoximorphic coloration, and Fe nodules. These paleosols are most commonly found in mudstone and claystone deposits within several meters above or below channel sandstones and crevasse-splay deposits. Several examples of these paleosols have been traced laterally in excess of 3 km, indicating that this soil morphology is, in many cases, laterally continuous. However, other examples of pedotype B, when traced laterally, become type C paleosols or terminate proximal to channel deposits (Fig. 7). Type B paleosols increase in relative abundance from west to east across the study area (from sections 1–3) and are most common in Unit III.
A Late Triassic soil catena Interpretation. Paleosols with these characteristics are classified as Gleyed Vertisols (Mack et al., 1993) or Aquerts (Soil Survey Staff, 1998). The dominant pedogenic process in the formation of a Vertisol is homogenization of the profile by shrinking and swelling of expansible 2:1 phyllosilicate clays (i.e., smectite) through repeated wetting and drying in climates with seasonal precipitation and/or fluctuation of the underlying water table (Wilding and Tessier, 1988). Redoximorphic or gley features in some of the type B paleosols indicate that these profiles were susceptible to water-logging for periods sufficient to produce anoxic conditions and mobilization of iron (Soil Survey Staff, 1975). Based on the close association of pedotype B paleosols with mudstones situated below, lateral to, and above channel and levee-splay sand deposits, it is likely that these paleosols experienced episodic water logging from flooding events proximal to stream corridors. Type C Paleosols Description. These are smectite and MLM-rich mudstone and claystone profiles from 75 to 175 cm thick. Similar to pedotype B, these paleosols exhibit wedge-shaped aggregates at depth grading upward to angular blocky or prismatic structure with slickensides and clastic dikes (Fig. 6). However, these paleosols contain nodular carbonate ranging in size from a few millimeters to 20 cm in diameter. Nodules typically exhibit a pattern of decreasing size upward through the profile, with noncalcareous horizons beneath the interpreted paleosol surfaces. Redoximorphic features are not common in type C paleosols, although many of these paleosols exhibit some rare to common faint color mottling in lower horizons. Similar to type B paleosols, these paleosols are associated with fine-grained overbank deposits. Type C paleosols are most common in Unit II, but also occur in Unit III. Several examples of these paleosols have been traced laterally over distances of 3.5 km. In some instances, the morphological compositions of individual type C paleosol layers undergo lateral transitions to type B and type E paleosol profiles. Interpretation. Wedge-shape aggregate structure, slickensides, and carbonate nodules in type C paleosols indicate that they are calcic Vertisols (Mack et al., 1993) and may possibly represent the soil suborder Xererts (Soil Survey Staff, 1975). The significant features that separate pedotype C from pedotype B are the presence of calcite nodules and a general absence of redoximorphic features. These differences have significant bearing on the pedogenic processes responsible for the formation of pedotype C. The absence of redoximorphic features in the A and B horizons of type C paleosols suggests that these horizons were seldom or never water-logged for significant periods of time. Therefore, it is likely that type C paleosols underwent episodic wetting and drying from seasonal precipitation, rather than an episodic rise and fall of the water table. Furthermore, the presence of nodular calcite in these paleosols suggests long periods of pedogenesis (between 1000 and 10,000 yr) in areas of the floodplain removed from frequent overbank deposition (Gile et al., 1981). As such, type C paleosols are interpreted to represent
23
relatively old clay-rich soils that were spatially removed from the depositional effects of major rivers and streams. In addition, these profiles likely formed under semiarid or arid climatic conditions characterized by seasonal rainfall. Type D Paleosols Description. These are smectite- and kaolinite-rich mudstone and claystone profiles that range from 75 to 150 cm thick. Lower horizons are usually characterized by massive to very coarse angular blocky structure, grading upward to fine to medium, subangular blocky to angular blocky structure. These profiles are noncalcareous, and subsurface horizons exhibit clay “films,” or argillans (Bt horizon, Figs. 5 and 6; Brewer, 1976), along ped surfaces and within root molds. In most instances, the concentration of argillans is so great that it results in apparent down-profile enrichment from the interpreted soil surface of clay-sized material. Many of the pedotype D paleosols also contain millimeter- to centimeter-size hematite nodules and exhibit few to many, fine to coarse, faint to prominent color mottlings. Type D paleosols are most common in mud-rich overbank sedimentary rocks, although a few examples were found above channel and crevasse-splay sandstones. Type D paleosols are most abundant in Unit III, sections 1 and 2, but are also common in Unit IV, section 3. Interpretation. Type D paleosols are classified as Argillisols (Mack et al., 1993), suggesting they were likely Alfisols (Soil Survey Staff, 1998). The paleoenvironmental significance of an Argillisol lies in the fact that clay films and argillic horizons form on generally well-drained and stable landscapes. This is because the dominant pedogenic process in these soils is translocation of layer-lattice silicate clays that can only be facilitated through leaching of carbonates and calcium ions from clay exchange sites in the soil profile (Franzmeier et al., 1985). Furthermore, these paleosols will not form in climates with excessive precipitation that results in ever-wet soil conditions (Soil Survey Staff, 1975). Birkeland (1984) has shown that formation of Bt horizons may form between 100 and 1000 yr, but the development of most subsurface argillic horizons requires longer than this period of time. Based on these considerations, type D paleosols represent soils that were relatively distant from active Ischigualasto fluvial channels and the associated effects of episodic sedimentation above the actively forming profile. However, Fe nodules in some type D paleosols may indicate seasonally poorly drained conditions (Duchaufour, 1982). Type E Paleosols Description. Type E paleosols are smectite- and MLM-rich mudstone profiles that range from 50 to 175 cm thick and are generally red to brown in color (Figs. 5 and 6). Lower horizons are typically massive or very coarse angular blocky, grading upward to coarse to medium angular blocky structure. These paleosols all contain tubular and nodular carbonate concretions (Bk horizons). Nodules range from a few millimeters to 7.5 cm in diameter and define calcareous horizons 10–110 cm thick. Where
D
C
B
A
Pedotype
Bt
CB C C2
13–32
32–44 44–85 85–95
BC
94–130
– Acc C Bc EB
BKss
52–94
+105–56 +56–27 +27–0 0–6 6–13
Abkss
IIC1 IIIC2 IVC3
140–176 176–206 206–221
0–52
BCss
90–140
–
Bss
45–90
+30–0
ABss
AC1 AC2 AC3 AC4 C
5Y 6/3 5Y 6/3 5Y 6/3
7.5YR 3/2
5Y 5/4 5Y 2.5/1 5Y 6/4 2.5Y 4/1 G1 6/10GY
2.5YR 3/1
2.5YR 3/1
5Y 5/1
–
G1 7/10Gy
G1 7/10Y
G2 5/10Bg
10R 5/3
G1 7/10GY G1 7/10Y 10YR 4/3 G1 5/10Y 5R 2.5/2
5Y 5/3 5Y 6/8 5Y 6/8
5Y 5/3
5Y 6/2
2.5YR 5/2
G2 5/10BG
5YR 3/1
–
2.5YR 3/2
2.5YR 5/4
G1 6/10Y
G1 8/N, 10R 4/3 5R 3/3 G1 6/5G, 5R 4/3 5R 3/3 G1 6/5GY
H-III-g H-III-f V
H-III-g
V G-I-d D-II-g V H-III-g
V
H-III-h
V
V
V V V
V
H-II-g
H-II-g
Coarse angular blocky Massive Platy
Medium angular blocky
Massive Platy Platy Medium angular blocky Single grain
Sphenoid with medium angular blocky Massive to medium angular blocky
Sphenoid with medium angular blocky
X-bedded
Massive Fissile Massive
Sphenoid
Sphenoid
Medium angular blocky
Massive, single grain Massive, single grain Massive, single grain Massive, single grain Massive, single grain
Clear, wavy Gradual, wavy –
Clear, wavy
Abrupt, wavy Clear, smooth Abrupt, wavy Clear, smooth Clear, smooth
-
Gradual, wavy
Clear, wavy
Abrupt, broken
Clear, wavy Clear, smooth
Clear, wavy
Clear, wavy
Gradual, wavy
Clear, smooth Clear, smooth Clear, wavy Sharp, smooth Smooth, abrupt Smectite, minor MLM Smectite, minor MLM Smectite, minor MLM Smectite, MLM Smectite, MLM Smectite, MLM
Hematite
Hematite
Calcite, Minor oxides
MLM, smectite MLM, smectite MLM, smectite Smectite, MLM Minor smectite, minor MLM Smectite, minor MLM Smectite, MLM MLM, smectite MLM, smectite
Smectite, minor MLM Smectite, MLM
Smectite, minor MLM, minor kaolinite Calcite, Minor Smectite, minor oxides MLM, minor kaolinite
Minor oxides
Minor oxides
Minor oxides
Minor oxides
MLM, smectite MLM, smectite MLM, smectite MLM, smectite MLM, smectite
TABLE 1. MORPHOLOGICAL AND CHEMICAL DESCRIPTIONS OF PEDOTYPES DISCUSSED THROUGHOUT TEXT Horizon Munsell color Root Pedogenic Contact Authigenic <2 Pm fraction † Matrix Mottles Class* structure minerals mineralogy G1 8/N G1 6/10GY Massive, single grain Abrupt, smooth MLM, smectite
0–45
0–41 41–98 98–133 133–143 143–189
Depth (cm) +30–0
(continued)
Argillisol, Aqualf?
Calcic Vertisol, Xerept?
Gleyed Vertisol, Aquert?
Paleosol § classification Protosol, Entisols, Inceptisols
5YR 3/4
Bt
Bk
Bn
0–18
18–67
67–101
Cr
Cr2 Cr3 R
34–45 45–65 65–865
AC2
159–219
9–34
IIC AC
95–140 140–159
BCss
AC2
12–95
+29–0 0–9
G1 6/10Y
AC
0–12
10R 3/2 G1 5//10Y G1 7/10Y G1 7/N 2.5Y 4/1 10YR 3/2
2.5Y 5/1 10YR 7/1
G1 5/N
10YR 3/1 2.5Y 5/1
2.5YR 5/1
G1 5/10Y
+12–0 G,H-II-H
V
IV
H-III-h
H-III-g,f
5Y 5/6 10YR 6/8
5Y 5/4
10R 3/2
5R 4/3
10YR 4/2 10R 4/3, 5R 4/6
– – –
V
V IV
G,H-II-H
V G-II-H
7.5R 4/2, 10R 5/6 H-II/III-H
5R 5/8
7.5R 5/8
10YR 4/6
7.5YR 5/6, 10R 5/1
G1 3/5GY
V
Medium angular blocky Medium angular blocky Massive
Spheroidal w/ subangular blocky
Massive Sphenoid
Massive to platy
Massive Massive to platy
Massive to platy
Platy
Massive
Coarse angular blocky
Medium angular blocky
Fine-medium angular blocky
Sphenoid
Hematite
Calcite
Clear, wavy Clear, wavy Abrupt, smooth
Clear, smooth
Abrupt, wavy Abrupt, broken
Smectite
Smectite Smectite
Smectite
Smectite
Smectite
Smectite, minor kaolinite, minor MLM Smectite, minor MLM Smectite, MLM
Smectite
Calcite Calcite Calcite
Smectite Smectite Disordered corrensite
MLM, smectite Phyllosilicate, Smectite, kaolinite calcite Phyllosilicate, Smectite calcite
Abrupt, smooth Phyllosilicate, hematite Abrupt, wavy Phyllosilicate, hematite Diffuse, smooth Clear, wavy Phyllosilicate, hematite – Phyllosilicate, hematite
Abrupt, smooth
–
Gradual, smooth
Clear, smooth
Abrupt, wavy
Vertisol
Andic Protosol, Andisol?
Argillic Calcisol, Xeralf?
Paleosol § classification Calcisol, Calcept?
*Root class determined according to the root classification scheme of Pfefferkorn and Fuchs (1991). † The mineralogy of the <2 Pm size fraction was determined from X-ray diffraction analysis. MLM is a 10Å mica-like mineral. See text for discussion. § Two taxonomical categories are given. The first entry is the paleosol taxonomic order according to the paleosol classification scheme of Mack et al. (1993). The second entry is the speculated taxonomic category of these paleosols according to the U.S. Department of Agriculture Soil Taxonomy. The question mark, “?,” after the second entry is meant to emphasize the uncertainty of applying the modern soil taxonomy to paleosol profiles.
H
G
G1 4/10GY
5R 2.5/2
G1 6/5G
+50–0
E
F
TABLE 1. MORPHOLOGICAL AND CHEMICAL DESCRIPTIONS OF PEDOTYPES DISCUSSED THROUGHOUT TEXT (continued) Horizon Munsell color Root Pedogenic Contact Authigenic <2 Pm fraction † Matrix Mottles Class* structure minerals mineralogy 10R 4/3 10R 4/2 V Abrupt, smooth MLM, smectite Abk 10R 3/1 G2 5/5B H-III-h Fine-medium angular blocky Clear, smooth Calcite Smectite, MLM Bk 10R 3/1 G2 5/5B H-III-h Medium angular blocky Clear, smooth Calcite Smectite, MLM Bk2 G1 5/N 2.5YR 4/1 H-III-h Medium angular blocky Clear, smooth Calcite Smectite, MLM Bk3 G1 5/N 2.5YR 4/1 H-III-h Coarse angular blocky – Calcite Smectite, MLM
Depth (cm) +38–0 0–11 11–21 21–30 30–96
Pedotype
A
B
XRD Analyses - Pedotype D, Section 3 Unit #2
XRD Analyses; Pedotype G, Section 3, Unit #3
Mg-Treated
Mg-Treated
17.8Å
Mg + Glycerol Treated
Mg + Glycerol Treated 17.9Å
K-Treated, 25°C
K-Treated, 25°C
14.7Å
K-Treated, 550°C
Relative Intensity
Relative intensity
K-Treated, 550°C
12.2Å
12.2Å
10.1Å
8.9Å (2nd Order peak of 17.8Å)
9.7Å
10.0Å
2
4
C
6
8 °2θ
10
12
2
14
4
6
8 °2θ
10
12
14
XRD Analyses - Pedotype H, Sect. 3, Unit #2 14.7Å
Mg-Treated Mg + Glycerol Treated K-Treated, 25°C K-Treated, 550°C
Relative Intensity
17.5Å
Figure 4. X-ray diffractograms (XRD) of the <2 μm fraction from (A) pedotype D, (B) pedotype G, and (C) pedotype H. See text for discussion.
12.3Å
9.9Å 7.2Å
2
4
6
8 °2θ
10
12
14
A Late Triassic soil catena
27
Fluvio-Alluvial A
B
C
D
1 50 2 00
ABk 1 00
2 00
1 50
ABss
Bc Bw Bt 2BC 2Bt 2
1 00
1 50
Bk
50
BC
0
50 BCss 1 00
1 00
1234 5
0
1234 5
2C 50
50
0
1 50
3C 4C
0
1234 5
E
Volcanics
G
1234 5 1 50
2C
F
H
AC 2 00
1 00
AC
T
1 00
50
0
ABk Bk Bk2
2C
1234 5
1 50
1 00 Bt 50
Bk3
50 B
Bk Bk2
1 00
Bc
50
0
B B
123 4 5
B
B
B B
B
2C
0
T T
T T T
BCss Cr1 Cr2 Cr3 R
Figure 5. Eight pedotypes defined for the Ischigualasto Formation: (A) Protosol; (B) Gleyed Vertisol; (C) calcic Vertisol; (D) Argillisol; (E) Calcisol; (F) calcic Argillisol; (G) Andisol; (H) Vertisol. Alpha-numeric symbols to the right of each pedotype correspond to U.S. Department of Agriculture soil horizon designations (Soil Survey Staff, 1998). Vertical axis is in centimeters. Horizontal axis: 1—clay, 2—mudstone, 3, 4, 5—fine, medium, and coarse sand, respectively. Pedotypes A–F comprise the fluvial-alluvial paleosol association, whereas pedotypes G and H comprise the volcanics paleosol assemblage. Pedotypes A and B are typically proximal to channel sands and crevasse deposits. Pedotypes C, D, E, and F occur in mudstone-dominated intervals distal from major channel sandstones.
1234 5
AC
0
1234 5
these paleosols have not been truncated, calcareous horizons occur beneath noncalcareous surface horizons. Type E paleosols are limited to mud-rich overbank deposits. Several examples have been traced along outcrop exposures in excess of 2 km. In addition, two examples of type E paleosols have been observed to grade laterally to type C paleosols. Type E paleosols occur in both Unit II and III, and they are most abundant in the eastern part of the study area, near section 3. Interpretation. These paleosols are classified as Calcisols (Mack et al., 1993) and may have been Inceptisols and/or Aridisols (Soil Survey Staff, 1998). The most important pedogenic process in the formation of type E paleosols is the accumulation of carbonate minerals. Furthermore, calcic horizons apparently do not form in modern soils that receive an excessive amount of mean annual precipitation (>750 mm/yr) (Royer, 1999), and they represent relatively uninterrupted and stable periods of time
required for formation of calcic horizons (Gile et al., 1981). Type E paleosols are interpreted to have formed as relatively well-drained and dry soils that were distant from active channels resulting in limited sedimentation during pedogenesis. Type F Paleosols Description. These are smectite-, MLM-, and kaolinite-rich mudstone and claystone profiles that are 50–160 cm thick. Lower horizons are typically massive to coarse angular blocky very fine sandstones or mudstones that grade upward to fine to medium angular to subangular blocky mudstones and claystones (Figs. 5 and 6). These profiles are red, orange, or brown with few to no color mottles. Morphological attributes of type F paleosols are similar to types D and E paleosols, in that these paleosols are characterized by subsurface horizons that contain both argillans (Bt horizons) and carbonate (Bk, Btk horizons). However,
28
Tabor et al.
Member
TABLE 2. DISTRIBUTION OF PALEOSOL TYPES IN THE ISCHIGUALASTO FORMATION † Rock Pedogenic Paleosol types Channels § thickness thickness (%)* A B C D E F G H (m) (m)
Section 1 Unit IV Unit III Unit II Unit I Totals # Abundance (%)
63 465 121 42 691
Section 2 Unit IV Unit III Unit II Unit I Totals # Abundance (%)
37 257 92 34 413
41 31 12 49 30
28 23 17 25 24
7 25 14 2 48 10
19 11 5 4 38 33
2 16 4
2 2
15 19
4 5
35 1
1 11
35 32
12 11
19 4 1 24 21
3 15 3 1 22 19
1 5
3
6 5
3 3
11.6 106.6 45.8 3.1 167.1
8
8 7
1 4 4 4
1 1
6
18.1 68.5 34.1 2.4 123.1
Section 3 Unit IV 37 8 18 1 4 1 22.5 Unit III 257 30 49 36 8 1 5 5 88.4 Unit II 67 0 9 7 3 1 3 6 2 34.7 Unit I 20 6 1 5 5.2 Totals 381 77 49 11 6 8 11 2 150.8 # 21 46 30 7 4 5 7 1 Abundance (%) *The % thickness of channel sandstone deposits within each lithostratigraphic member from sections 1, 2, and 3 in the Ischigualasto Formation. † The number of paleosol types described from sections 1, 2, and 3 for each informal member and total stratigraphic thickness of the Ischigualasto Formation. § Cumulative thickness of all pedogenically altered layers within each measured section. # The abundance of paleosols types, in % thickness of pedogenically altered layers and % thickness of channel sands through the entire Ischigualasto Formation, in sections 1, 2, and 3.
in all of the type F paleosols, the zone of inferred clay accumulation (Bt horizons) overlies a zone of carbonate accumulation. In addition, horizons directly beneath the interpreted soil surface are noncalcareous. Type F paleosols are closely associated with mud-rich overbank deposits and are observed to be laterally continuous over distances of 1 km. These paleosols are most common in Unit II, but are also present in Unit III, sections 1 and 3. Interpretation. These paleosols are classified as calcic Argillisols (Mack et al., 1993) and may possibly represent the USDA soil suborder Xeralfs (Soil Survey Staff, 1975, 1998). Similar to type D paleosols, the dominant pedogenic process for the formation of type F paleosols is the translocation of layer-lattice silicate clays that can only be facilitated through soil leaching of carbonates and calcium ions from clay exchange sites in the soil profiles (Franzmeier et al., 1985). Although leaching of calcium carbonate is required for formation of an argillic horizon, pedogenic carbonate may precipitate in lower horizons of soils with argillic horizons if leaching is not so great as to remove soluble minerals from the soil zone (Buol et al., 1997). In fact, accumulation of carbonate is common in the lower horizons of modern Alfisols in xeric (i.e., winter wet, summer dry) soil moisture regimes (Soil Survey Staff, 1975). Type F paleosols likely represent soil forma-
tion upon stable portions of the paleolandscape. Furthermore, the preservation of pedogenic carbonates below argillic horizons suggests that the depth of leaching was not sufficient to transport base cations to the water table, indicating a fairly deep water table in these portions of the landscape. Volcanic Association Type G Paleosols Description. These are smectite-rich mudstones and fine sandstone profiles ranging from 30 to 200 cm thick. Lower horizons and parent material are massive or single-grain structures consisting of feldspar-rich ash-fall deposits or fluvially reworked volcanogenic sediments grading upward to very weak medium angular blocky or massive structure (Figs. 5 and 6). Individual profiles exhibit variable coloration, ranging from white to buff, pink, brown, purple, and red. However, all profiles exhibit redder colors upward through the profile. In addition, rooting structures are typically vertical to subvertical, pink or red siliceous and iron-oxide–rich tubules. These paleosol profiles occur above and within regionally extensive ash-fall deposits, resulting in a broad lateral continuity that exceeds 4
A Late Triassic soil catena
29
Figure 6. Ischigualasto Formation pedotypes. (A) Series of stacked type A paleosols (Protosols) associated with crevasse-splay deposits, Unit IV, section 1. (B) Bedding-plane exposure of clastic dikes in type B paleosol (Gleyed Vertisol), Unit III, section 1. (C) Lowermost dark bands are exposures of pedotypes F (calcic Argillisol), E (Calcisol), and C (calcic Vertisol), Unit II, section 3. (D) Type D paleosol (Argillisol) with well-developed vertical root halos, Unit III, ~1 km east of section 1. (E) Vertical profile through type G paleosol (Vertisol) developed on basalt flow, Unit II, section 3. Alpha-numeric symbols correspond to U.S. Department of Agriculture soil horizon designations (Soil Survey Staff, 1998).
km along some stratigraphic horizons. They are typically associated with fine-grained overbank sediments directly above channel sandstones. These paleosols are limited to sections 1 and 2 of Unit III (Figs. 3 and 6). Interpretation. These paleosols are interpreted as eutric Protosols (Mack et al., 1993). It is also possible that type G paleosols may have been Andisols within the context of the USDA soil classification (Soil Survey Staff, 1998). However, short-range order (SRO) minerals, such as allophane and immoglolite, which are important constituents in the Andisol classification, are absent in type G paleosols, and therefore they cannot be justifiably placed within the Andisol soil order. The presence of relatively unstable parent constituents, such as ash and feldspar, indicates that pedogenesis was of a relatively short duration. Type H Paleosols Description. These smectite- and kaolinite-rich mudstone and claystone profiles formed above Triassic-age vesicular calcium plagioclase basalt flows in the lower one-third of Unit II, section 3 (Fig. 3). Type H paleosols are characterized by shrinkswell features in the upper portion, with a thin spheroidally
weathered horizon and weakly mottled horizons beneath grading downward into unweathered basalt (Figs. 5 and 6). Amygdules in the unweathered basalt are composed of sparry calcite cements, whereas calcite is missing from the upper horizon of the profile. The spheroidally weathered horizon is morphologically divided into three distinct zones: (1) a central silty gray spheroid with common sparry calcite-filled vesicles; (2) a middle gray siltyclay layer with few micritic calcite-filled vesicles and common green montmorillonite-filled vesicles; and (3) an outer red clayrich rind with common green subcentimeter-size montmorillonite nodules (Tabor et al., 2004). Interpretation. Type H paleosols are interpreted as Vertisols (Mack et al., 1993; Soil Survey Staff, 1975, 1998; Tabor et al., 2004). These paleosol profiles exhibit similar morphologic features and indicate similar pedogenic processes responsible for soil formation as type B and C paleosols. However, type H paleosols apparently formed from in situ weathering of parent basalt, whereas type B and C paleosols formed from siliciclastic sedimentary parent materials. Tabor et al. (2004) suggested that type H paleosols formed upon relatively stable landscapes in a climate characterized by extreme seasonality in precipitation.
30
Tabor et al.
Paleosol Calcite Petrography and Geochemistry The petrographic textures and δ18O values of carbonate nodules associated with Ischigualasto Formation paleosols is presented in Moore (2003). The relative abundance of petrographic fabrics and morphology of paleosol carbonate samples follow in the order of abundance: (1) micritic calcite nodules with subspherical macromorphology, followed by (2) whole nodules of radiaxial calcite fibers that accommodate botryoidal “cluster-like” macromorphology; and (3) “sandy” calcite nodules that exhibit floating detrital grains of quartz and feldspars in a micritic calcite matrix with lenticular macromorphology (Fig. 8; Appendix 1). Although not common, calcite spar-filled veins crosscut examples of all three major sorts of petrographic fabrics. In general, carbonate nodules contain between 16% and 34% acid-insoluble residue that is composed of phyllosilicate, quartz, and feldspar (Moore, 2003).
The typically lenticular macromorphology of the sandy nodules, in conjunction with their slightly more negative δ18O values, suggests a groundwater origin (Moore, 2003; see also Slate, 1996; Budd et al., 2002). Furthermore, micritic calcite nodules are normally observed in modern, well-drained soils where illuviation of Ca2+ ion is an important pedogenic process (e.g., Deutz et al., 2001), whereas radiaxial calcite has yet to be described as a product of well-drained pedogenic systems where percolating meteoric rainwater represents the sole source of H2O in the profile. Therefore, the δ18O values of (1) “sandy” micritic and (2) radiaxial fibrous calcite nodules are eliminated from further consideration in this work. This selection reflects that micritic calcite should provide the most conservative data set for paleotemperature estimates within the context of the model proposed by Dworkin et al. (2005). Micritic calcite nodules record an average intranodular δ18OVPDB variation of 1.1 ± 0.4‰ (1σ), with an absolute range of
A
4
EAST
200 m 200
400 m
1
WEST
400 m
2
200
3 150
Crevasse - Splay Sandstone 150
100
100 Greenish Gray BC Brown Bk
Pedotype F
150
50
Very Dusky Bk2 Brown
Pedotype E
Greenish Gray Bg
100
0 C ms f m c
100 Dusky Red Bk1 Greenish Gray Bk2 Greenish BC Gray Greenish C1 Gray
Pedotype B 50
50
Dusky Red Greenish Gray
Dusky Red Greenish Gray
Pedotype A 0
C ms f m c sand
sand
50
0
Greenish Gray C3
C ms f m c sand
0
C ms f m c sand
B 1
Crevasse Splay/Levee Deposits
2
Paleosol Horizon
3
4
Fluvial Channel Sandstone
V.E. ~ 7X
Figure 7. (A) Four laterally equivalent stratigraphic columns of a paleosol from Unit II (~48 m level in section 1). Along this ~1 km transect, there is a transition from fluvial sandstone deposits to weakly developed paleosol types A and B, to more mature types C and E. (B) Generalized cross section of laterally variable pedotypes showing the relative position of the measured profiles shown in A with respect to coeval proximal overbank channel deposits. Vertical exaggeration ~7×.
A Late Triassic soil catena
A
31
B
C
Figure 8. Plain, transmitted-light petrographic images of (A) micritic calcite nodule collected from 118 m above the base of the Ischigualasto Formation (Unit II) in section 1, (B) sandy micrite nodule collected from 106 m above the base of the Ischigualasto Formation (Unit II) in section 2, and (C) radiaxial fibrous calcite nodule from 118 m above the base of the Ischigualasto Formation (Unit II) in section 1. See text for discussion.
intranodular δ18O variability from 0‰ to 3.8‰. The δ18O values of micritic calcite and their stratigraphic position above the base of the Ischigualasto Formation for sections 1, 2, and 3 are presented in Appendix 1. Micritic paleosol calcite in section 1 exhibits a stratigraphic range of δ18O values from –5.4‰ to –13.1‰. The maximum range of δ18O values along a single lateral paleosol transect in section 1 is from –5.4‰ to –11.7‰ at 118 m (Fig. 9; Appendix 1). Section 2 exhibits a stratigraphic range of micrite δ18O values from –8.2‰ to –13.0‰, with a maximum range of δ18O values along a single paleosol transect ranging from –8.2‰ to –12.7‰ at 106 m above the base of the Ischigualasto Formation (Fig. 9; Appendix 1). Section 3 exhibits a range of micrite δ18O values from –9.0‰ to –14.3‰, with a maximum range of δ18O values along a singe paleosol transect ranging from –9.0‰ to –12.3‰, 43 m above the base of the Ischigualasto Formation (Fig. 9; Appendix 1). The large range of δ18O values along individual paleosol transects suggests the possibility of variable soil water δ18O values, possibly related to isotopic modification through in situ evaporation within the soil (e.g., Cerling
and Quade, 1993; Levin et al., 2004; Tabor and Montañez, 2005). DISCUSSION AND CONCLUSIONS The spatial and temporal distribution of observed Ischigualasto Formation pedotypes, in addition to the distribution of fluvial channel deposits, indicates lateral variations in depositional rates and soil drainage indicative of the existence of a paleosoil catena throughout deposition. While these variations are intimately linked to lateral changes in formation thickness and related structural controls on basin development, in terms of Ischigualasto paleosols, they are manifested by lateral variations in paleosol abundance, B-horizon thickness, redoximorphic features, and pedogenic carbonate accumulations. The most striking characteristic of the Ischigualasto Formation is a significant change in sedimentary thickness, from west to east, across the study area (Fig. 3). Currie et al. (2001) considered the thickness change to record structural controls on basin
32
Tabor et al. 400 Paleosol Micrite δ18OVPDB Section1, Basin Axis Section2, Basin Flank
350
Section3, Basin Margin
Stratigraphic Position (m)
300
250
Figure 9. δ18O values of micritic calcite from paleosol carbonate nodules in sections 1, 2, and 3 of the Ischigualasto Formation. See Appendix 1 for further information and text for discussion.
200
150
100
50
0 -14
-12
-10
δ18OVPDB
-8
accommodation due to syndepositional faulting in the western part of the study area. The relative abundance of channel sandstones in the thicker sedimentary package in section 1 (Fig. 3) suggests that major streams preferentially occupied that portion of the basin because this area apparently provided greater accommodation throughout Ischigualasto Formation deposition. The distribution of Ischigualasto Formation pedotypes in the eastern part of the basin provides additional insight into the relationships between basin structure and the history of sediment accumulation. The stratigraphic distribution of paleosol types in the study area is presented in Table 2. The cumulative thickness of pedogenically altered layers corresponds to 24%,
-6
-4
30%, and 40% of the total stratigraphic thickness in sections 1, 2, and 3, respectively (Fig. 10A). This suggests that along with a dramatic reduction in stratigraphic thickness from west to east (~300 m), areas in the eastern part of the study area also experienced longer periods of nondeposition and soil formation. This interpretation is supported by differences in the overall paleosol morphology across the study area. Generally speaking, longer periods of nondeposition and soil formation result in more pervasively developed or “mature” soil morphologies (Buol et al., 1997). For example, Bain et al. (1993) demonstrated that B-horizon thickness appears to progress linearly over the first ~10,000 yr of soil formation (Fig. 10B). The mean B-horizon thickness of paleosols
A
45
Ischigualasto Fm Pedogenic thickness
B
% paleosols
40 35 30 25
40%
20 15 10
24%
30%
5
Section 3
D
Mean B-Horizon Thickness (cm)
90 80
80
70
cm
60 50 40 30 10 0
n = 45 Section 1
n = 38
n = 40
88% 80%
70 60 50 40 30 20 10
Section 2
E
90
87%
30 20 10 0
Present
60 40
20%
12% Section 1
85%
Section 2
Section 3
80%
70 50
37%
Moisture Deficiency - Paleosol CaCO3
80
% paleosols
0
Section 3
63%
Not Present
20
Redoximorphic Features
90
Not Present
Section 2
Present
C
Section 1
% paleosols
0
13% Section 1
15% Section 2
20% Section 3
Figure 10. (A) Histogram of the stratigraphic thickness represented by pedogenically altered layers in sections 1, 2, and 3. See text for further discussion. (B) Plot of B-horizon thickness versus age of profiles in a soil chronosequence from Scotland (Bain et al., 1993). This graph illustrates the very good correlation that is observed between B-horizon thickness and duration of pedogenesis and is not meant as a quantitative indicator of the time represented by paleosol profiles in the Ischigualasto Formation. (C) Histogram of the mean B-horizon thickness of paleosols in sections 1, 2, and 3. Note that type A and B and G paleosols are not considered here because these paleosols either do not contain B horizons or represent cumulative deposition and pedogenesis. Progressively thicker B-horizon thickness from section 1 to 3 suggests longer periods of nondeposition and pedogenesis toward the eastern part of the study area. See text for discussion. (D) Distribution of paleosol profiles with redoximorphic features in sections 1, 2, and 3. See text for discussion. (E) Distribution of calcareous paleosol profiles, based on percent thickness of pedogenically altered layers, in sections 1, 2, and 3. See text for discussion.
34
Tabor et al.
in sections 1, 2, and 3 is 24 cm, 33 cm, and 40 cm, respectively, indicating that periods of active pedogenesis associated with the formation of individual soils were of progressively longer duration from the western to eastern parts of the study area. On alluvial landscapes, lateral changes in groundwater level and subsequent soil drainage can dramatically effect soil development and morphology (i.e., a soil catena; Steila, 1976; Birkeland, 1999; Buol et al., 1997). Soils that develop adjacent to fluvial channels are commonly poorly drained and experience frequent fluctuations in the level of the groundwater table that can lead to anoxia and the formation of gley and redoximorphic morphologies (Birkeland, 1999). In Ischigualasto Formation paleosols, evidence for gleyed and redoximorphic conditions is preserved in 88%, 63%, and 20% of the paleosol profiles in sections 1, 2, and 3, respectively (Fig. 10D), indicating a gradient of soil drainage conditions from generally very wet, to progressively better drained soil conditions, from west to east, during Ischigualasto Formation deposition. Thus, we conclude that the distribution of paleosol morphologies in the Ischigualasto Formation resulted not only from changing landscape stability and frequency of deposition, but also from changing soil drainage conditions across the paleolandscape. In spite of strong evidence for a catenary sequence that focused poorly drained conditions in areas of increased subsidence (toward section 1; Fig. 3), the presence of carbonate-bearing paleosols in all three measured sections suggests that there was a discrete interval of relatively well-drained conditions and net soil-moisture deficiency across the entire study area during deposition of Unit II and the lower half of Unit III (Table 2). Nevertheless, carbonate-bearing paleosols (types C, E, and F) represent 13%, 15%, and 19% of the total thickness of pedogenically altered layers in sections 1, 2, and 3, respectively (Fig. 10E). In addition, a single calcareous paleosol profile (type F) occurs near the top of section 3 in Unit IV, but there are no carbonate-bearing paleosol profiles in Unit IV of sections 1 and 2. Therefore, the distribution of calcareous soils seems to be related to landscape position, which may also be controlled by soil drainage patterns and the position of the groundwater table. We consider the distribution of pedogenic carbonate across the study area to reflect differential depths of the groundwater table beneath the surface of the paleolandscape. As meteoric precipitation percolates through well-drained soil, it will leach and translocate Ca2+ from the upper horizons through the “depth of wetting.” This depth is related to many factors, including permeability and porosity of soil matrix and the amount and intensity of rainfall (Jenny, 1941; Arkley, 1963). Whatever the depth of wetting, the soil profile is effectively dry below this position, and pedogenic carbonate can accumulate because illuvial Ca2+ is forced out of solution, into the solid phase (e.g., Jenny, 1941; Arkley, 1963). However, if the depth of wetting intersects the groundwater table, Ca2+ will not be retained (e.g., Wigley, 1978) and carbonate will not form in the soil (Fig. 11). Based upon the distribution of major channel sandstones and redoximorphic conditions, we suggest that there was a
long-term pattern of a very shallow groundwater table in the western portions of the study area and a relatively deeper groundwater table in the eastern portion of the study area (Figs. 3 and 11). Under these conditions, the depth of wetting in the soil would have been more likely to intersect the groundwater table and Ca2+ could have been leached out of the soil profiles in sections 1 and 2 (Fig. 11). However, the depth of wetting was not as likely to intersect the deeper groundwater table in section 3, and Ca2+ could have been retained within the soil as CaCO3. Note that argillic horizons will not form until Ca2+ has been thoroughly leached from the upper layers of the soil and translocated to lower horizons or entirely leached from the soil (Franzmeier et al., 1985). Under shallow groundwater conditions, Ca2+ would have been more likely to leach out of the soil, allowing Protosols or Argillisols to form, rather than Calcisols or calcic Argillisols. Therefore, a shallow groundwater table may also explain the relative abundance of Argillisols (type D paleosols) in sections 1 and 2, which represent 21% and 19% of the total thickness of pedogenically altered layers, and their rarity (only 4%) in section 3 (Table 2). Collectively, the distribution of channel sands, variations in pedogenic alteration, and paleosol B-horizon thickness provide persuasive evidence for the presence of a long-term geomorphic control on sedimentation, landscape stability, and soil formation in the Late Triassic Ischigualasto basin. Implications of Paleosol Nodule Micritic Calcite as a Monitor of Paleotemperature The oxygen isotopic composition of a particular soil mineral is determined by (1) the temperature-dependent isotopic fractionation of that mineral, (2) temperature of crystallization, and (3) the oxygen isotope composition of liquid H2O from which it crystallizes (e.g., O’Neil et al., 1969; Cerling and Quade, 1993; Yapp, 1993, 2000). The vast majority of liquid water in continental environments is meteoric in origin (Dansgaard, 1964). As a result of the controls upon the global meteoric water line, both temperature of mineral crystallization and the δ18O value of meteoric water may be parametrically correlated. Therefore, the oxygen isotopic values of micritic calcite (δ18Occ) in Ischigualasto Formation paleosol nodules may potentially provide important insights into Late Triassic climate. Dworkin et al. (2005) presented two different equations to relate the δ18O value of pedogenic calcite to temperature of crystallization. The first equation (Equation 1 below) is a thirdorder polynomial that relates the observed parametric correlation between meteoric precipitation and mean annual surface air temperature from the entire International Atomic Energy Association (IAEA) data set (Rozanski et al., 1993) to pedogenic calcite via the oxygen isotope fractionation equation between water and calcite (O’Neil et al., 1969: 103(ln α) = 2.78 × 106/ T2 – 2.89). The second equation (Equation 2 below) is derived from the empirical relationship that was measured between pedogenic calcite δ18O values and measured mean annual surface air
A Late Triassic soil catena
Section 1
Section 2 Gleyed Vertisols, calcic Vertisols, Argillisols, calcic Argillisols
~2 m
Gleyed Protosols, Gleyed Vertisols
Ca2+
Ca2+
35
Section 3 Calcisols, calcic Vertisols, calcic Argillisols
Groundwate
Ca2+
r Table
CaCO3
~15 km Figure 11. Schematic cross-section diagram representing the interpreted distribution of paleosol profiles and position of the paleo–water table across the Ischigualasto Formation paleolandscape. Based on the abundance of channel sandstones and redoximorphic features, the position of the groundwater table is interpreted to be relatively shallow in the western region of the study area and relatively deeper in the eastern region. Leaching of Ca2+ out of the soil and into the groundwater table would have been more likely in the western region, resulting in formation of less calcareous soils. See text for discussion.
temperature from soils located in interior continental sites (Cerling and Quade, 1993): 0 = −0.50T 3 + (δ18Occ (‰, SMOW) + 152.04) × T 2 – 2.78 × 106 (Dworkin et al., 2005)
(1)
from the coasts and (2), and continental sites >200 km inland. Based upon the results of Ferguson et al. (1999), the following equations describe the expected correlation between temperature and the δ18O value of pedogenic calcite that forms in equilibrium with meteoric water: Continental Sites
and T (±2 °C) = (δ18Occ (PDB) – 12.78)/0.64 (after Ferguson et al., 1999)
δ Occ (‰, PDB) = 0.49 × (T[°C]) – 12.65 (Dworkin et al., 2005, from Cerling and Quade, 1993). (2) 18
(3)
and A plot of Equations 1 and 2 is shown in Figure 12. The analytical uncertainty of the temperature estimates reported in Dworkin et al. (2005) is ~±0.5 °C, which reflects only the analytical uncertainty of the δ18O measurement for calcite, and not other potentially important effects, such as variable rainfall δ18O values among isothermal sites. Considering an analytical uncertainty of ±0.5 °C, Equations 1 and 2 provide paleotemperature estimates that are indistinguishable from one another between ~12° and 20 °C, corresponding to δ18Occ-PDB values ranging from –2.9‰ to –6.7‰. Pedogenic carbonate δ18O values above or below these values will result in significantly different temperature estimates from Equations 1 and 2 (Fig. 12). Ferguson et al. (1999) noted that a correlation exists between meteoric precipitation and temperature for IAEA stations with mean annual temperatures <15 °C, but the isotopic difference between 18O-enriched coastal maritime precipitation and 18Opoor continental interior precipitation for isothermal sites partially confounds the correlation. These workers therefore broke the IAEA data set into two groups: (1) maritime sites <200 km
Maritime Sites T (±5 °C) = (δ18Occ (PDB) – 12.65)/0.59 (after Ferguson et al., 1999).
(4)
The reported uncertainties of ±2 °C and ±5 °C in these temperature estimates reflect the range of precipitation δ18O values for isothermal sites in continental and maritime sites, respectively, and therefore probably provide a more realistic evaluation of uncertainty associated with Equations 1 and 2 than was reported in Dworkin et al. (2005). Figure 12 presents plots of Equations 1 through 4. Comparison of the different correlations makes clear that Equation 1 is heavily weighted by cool-climate maritime sites and will likely underestimate paleotemperature for soil calcites in continental sites. However, Equation 2 and Equation 3, both of which represent calcite δ18O values within continental sites, exhibit a close correspondence at relatively low temperatures (<~10 °C; see Fig. 12). As mentioned, the Ischigualasto
0
Dworkin et al (2005), Eqn (1)
Maritime
Dworkin et al (2005), Eqn (2) Ferguson et al (1999), Eqn (3) Ferguson et al (1999), Eqn (4)
-2
Uncertainty Envelope, Eqn (3)
Continental
δ 18 O(VPDB)
-4
-6
-8
-10
-12
-14 0
2
4
6
8
10
12
14
16
18
20
Temperature (°C)
A
Figure 12. Plot showing the relationship of calcite δ18OVPDB values versus temperature (°C). The trace of Equation 1 is the relationship that is predicted from the global International Atomic Energy Association (IAEA) data set presented in Dworkin et al. (2005). The trace of Equation 2 is the correlation that is defined by the measured mean annual surface air temperatures and δ18O values of pedogenic calcite from modern soils distributed across central North America (Dworkin et al., 2005, based on work in Cerling and Quade, 1993). The trace of Equation 3 is the relationship predicted from IAEA continental sites that are >200 km inland from the coastlines, whereas Equation 4 is the relationship predicted from IAEA maritime sites that are <200 km from the coast (Ferguson et al., 1999). See text for discussion.
400
Too Cold for Liquid H2O
Temp. range based on goethite δD,
350
Section 1; Eqn 2 Section 1; Eqn 3
δ18O (Tabor et al., 2004)
Stratigraphic Thickness (m)
300
250
200
150
100
50
0 -5
-3
-1
1
3
5
7
9
11
13
15
Temperature (°C) Figure 13. Stratigraphic position vs. estimated surface air temperature from Equations 2 and 3 based on the measured δ18OVPDB value of micritic paleosol calcite in (A) Section 1, (B) Section 2 and (C) Section 3. See text for discussion.
B
400
Too Cold for Liquid H2O
Temp. range based on goethite δD,
350
Section 2; Eqn (2) Section 2; Eqn (3)
δ O (Tabor et al., 2004) 18
Stratigraphic Thickness (m)
300
250
200
150
100
50
0 -5.0
-3.0
-1.0
1.0
3.0
5.0
7.0
9.0
11.0
13.0
15.0
Temperature (°C)
C 400
Too Cold for Liquid H2O
Temp. range based on goethite δD,
350
Section 3, Eqn (2) Section 3, Eqn (3)
δ O (Tabor et al., 2004) 18
Stratigraphic Thickness (m)
300
250
200
150
100
50
0 -5
-3
-1
1
3
5
Temperature (°C)
7
9
11
13
15
38
Tabor et al.
Formation was deposited in a closed continental basin that was likely far-removed from the effects of maritime precipitation. Therefore, we consider Equations 2 and 3 to provide the closest approximations of the relationship between temperature and calcite δ18O values during pedogenesis and paleosol calcite crystallization during Late Triassic time. Equations 1 and 4 will be disregarded in the following discussion. The resulting paleotemperature estimates from Equations 2 and 3, which are calculated from the oxygen isotope composition of micritic calcite in the Ischigualasto Formation (Appendix 1), are plotted in Figure 13A–C with respect to their stratigraphic positions in sections 1–3. Paleotemperature estimates range from –1° to 15 °C in section 1, whereas mean estimates from Equations 2 and 3 are 4° ± 3 °C (1σ) and 3° ± 2 °C (1σ), respectively. Paleotemperature estimates range from –3° to 9 °C in section 2, whereas mean estimates from Equations 2 and 3 are 4° ± 3 °C and 3 °C ± 2 °C, respectively. Finally, paleotemperature estimates range from –3 °C to 7 °C in section 3, whereas mean estimates from Equations 2 and 3 are 2° ± 3 °C and 2° ± 2 °C, respectively. The reported analytical uncertainty of the mean estimates reflects the 1 standard deviation of δ18O values for each stratigraphic section. The close correspondence of the mean temperature estimates in sections 1–3 suggests that Ischigualasto Formation paleosol calcites may preserve records of Late Triassic conditions in the soil-forming environment. However, it is critically important to note that these paleotemperature estimates provide equivocal estimates of temperature for any calcite that may have formed in the presence of evaporatively enriched soil waters and for nodules affected by postburial, diagenetic modification of calcite δ18O. Paleotemperature estimates derived from Equations 2 and 3 are appropriate only for calcite that forms in equilibrium (or very near equilibrium) with water that lies along (or very near to) the meteoric water line that is defined by cool-climate, continental sites in modern IAEA data stations (e.g., Ferguson et al., 1999). It is difficult, probably even impossible, to demonstrate any of these particular conditions for the micritic calcite samples from paleosol nodules in the Late Triassic Ischigualasto Formation. In this regard, it is probably not realistic to discuss the δ18O values of Ischigualasto Formation paleosol micrites in terms of absolute temperature values. Rather, it is probably more appropriate to use these estimates as a general guide to paleoenvironmental conditions (e.g., frigid, cool, warm, hot, hyperthermic). In this regard, we consider the similarity of average calcite δ18O values in sections 1, 2, and 3 to correspond to soil formation in a cold to cool climate (probably between ~0° and 11 °C) throughout the interval of soil calcite development in the Ischigualasto Formation (Units II, III, and IV). A lower limit of 0 °C is assumed here, because it is the practical limit for liquid water that
must be available for chemical reactions, whereas the upper limit is suggested only by the temperature estimates provided via Equations 2 and 3. However, an independent paleotemperature estimate of 8° ± 3 °C that is based on the oxygen and hydrogen isotope composition of goethite in a type H paleosol (Tabor et al., 2004) from Unit I, section 3 of the Ischigualasto Formation suggests that an upper limit of ~10 °C in the Late Triassic Ischigualasto Formation may be taken seriously. Summary of Inferred Paleoclimate Trend through Ischigualasto Formation Deposition Although the lithological composition and paleosol morphology of the Ischigualasto Formation was apparently strongly controlled by landscape position across the Ischigualasto–Villa Union basin, basin-wide changes in the stratigraphic distribution of paleosol morphologies suggest the possibility of relatively subtle climate changes in this region through Carnian time. The general absence of calcareous pedotypes (A, B, D, G, H) suggests that regional conditions were exceedingly unfavorable for pedogenic calcite formation during deposition of Unit I, the upper half of Unit III and Unit IV (Table 2). As mentioned earlier, unfavorable conditions for pedogenic calcite formation could be a humid climate that results in deep leaching of the soil profile, or a very shallow groundwater table, or a combination of both factors. However, the basinwide occurrence of calcareous pedotypes (C, E, F) suggests that conditions changed enough to permit retention of calcite within the soil zone during deposition of Unit II and the lower half of Unit III (Table 2). Such a change in paleosol morphology may correspond to a transition to generally drier conditions that resulted in incomplete leaching of the soil profile or a very deep groundwater table, or a combination of both factors. Therefore, the distribution of paleosol morphologies in the Ischigualasto Formation is suggestive of an environment that was humid and poorly drained during deposition of Unit I, followed by relatively drier and better-drained environments during deposition of Unit II and the lower half of Unit III, and a return to generally more humid and poorly drained environments during deposition of the upper half of Unit III and Unit IV. Based on available data, it is impossible to determine at this time if the perceived environmental changes are related to intra- or extrabasinal forces. Nevertheless, existing lithological, mineralogical, and chemical data indicate that structural controls on the basin and landscape position were the primary forces of landscape evolution in a cold to cool climate during deposition of the Carnian-age Ischigualasto Formation.
A Late Triassic soil catena
39
APPENDIX 1. STRATIGRAPHIC POSITION, TEXTURE, AND G O (VPDB) VALUE OF CARBONATE NODULES IN THE LATE TRIASSIC ISCHIGUALASTO FORMATION Section 1 Section 2 Section 3 Meters Texture G18O (‰) Meters Texture G18O (‰) Meters Texture G18O (‰) 60 Micrite –11.4 44 Micrite –9.7 14 Micrite –11.9 60 Micrite –11.1 46 Micrite –11.4 14 Micrite –11.5 60 Micrite –10.3 46 Micrite –10.4 14 Micrite –10.4 73 Micrite –11.1 46 Micrite –10.2 42 Micrite –9.8 75 Micrite –12.0 46 Micrite –10.0 43 Micrite –12.3 79 Micrite –9.6 59 Micrite –10.3 43 Micrite –12.0 79 Micrite –9.1 94 Micrite –12.0 43 Micrite –11.4 79 Micrite –9.1 96 Micrite –11.7 43 Micrite –11.1 81 Micrite –9.7 96 Micrite –11.3 43 Micrite –11.0 81 Micrite –9.5 106 Micrite –12.7 43 Micrite –11.0 95 Micrite –11.1 106 Micrite –12.2 43 Micrite –10.9 95 Micrite –10.9 106 Micrite –12.0 43 Micrite –10.7 118 Micrite –11.5 106 Micrite –11.0 43 Micrite –10.5 118 Micrite –11.2 106 Micrite –11.0 43 Micrite –10.4 118 Micrite –11.1 106 Micrite –10.9 43 Micrite –10.4 118 Micrite –9.3 106 Micrite –10.9 43 Micrite –10.4 118 Micrite –9.4 106 Micrite –10.5 43 Micrite –9.9 118 Micrite –11.7 106 Micrite –9.3 43 Micrite –9.9 118 Micrite –11.6 106 Micrite –8.4 43 Micrite –9.0 118 Micrite –11.6 106 Micrite –8.4 52 Micrite –10.7 118 Micrite –11.3 106 Micrite –8.2 57 Micrite –14.0 118 Micrite –10.9 108 Micrite –11.5 57 Micrite –13.6 118 Micrite –5.4 113 Micrite –10.5 138 Micrite –12.5 118 Micrite –8.9 122 Micrite –11.4 138 Micrite –11.9 118 Micrite –7.8 124 Micrite –11.1 138 Micrite –11.8 118 Micrite –8.9 155 Micrite –14.1 138 Micrite –11.7 146 Micrite –11.7 155 Micrite –13.0 164 Micrite –13.6 158 Micrite –11.0 164 Micrite –11.0 171 Micrite –12.1 210 Micrite –13.8 253 Micrite –12.1 210 Micrite –13.8 273 Micrite –13.1 210 Micrite –13.1 273 Micrite –10.6 213 Micrite –14.3 299 Micrite –10.0 213 Micrite –11.7 299 Micrite –9.2 251 Micrite –11.2 357 Micrite –13.2 357 Micrite –12.2 357 Micrite –11.6 357 Micrite –11.6 357 Micrite –11.1 18
ACKNOWLEDGMENTS We thank Oscar Alcober (Instituto y Museo de Ciencias Naturales) for permitting access to the field sites. This research was funded by National Science Foundation (NSF) grant EAR0447381 to Tabor and EAR-0519394 to Tabor and Montañez. REFERENCES CITED Alcober, O., 1996, Revisión de los crurotarsi, estratigrafía y tafonomía de la Formación Ischigualasto [Ph.D. thesis]: San Juan, Argentina, Universidad National de San Juan, 260 p. Alcober, O.A., Milana, J.P., and Martinez, R.N., 1997, Paleogeography of the Ischigualasto Formation and its influence on paleovertebrate communities: Journal of Vertebrate Paleontology Abstracts and Programs, v. 17, no. 3, p. 487–505, A28.
Allen, J.R.L., 1974, Studies in fluviatile sedimentation: Lateral variation in some fining-upwards cyclothems from the Red Marls, Pembrokeshire: Geological Journal, v. 9, p. 1–16. Arkley, R.J., 1963, Calculation of carbonate and water movement in soil from climate data: Soil Science, v. 96, p. 239–248. Atkinson, C.D., 1986, Tectonic control on alluvial sedimentation as revealed by an ancient catena in the Capella formation (Eocene) of northern Spain, in Wright, V.P., ed., Paleosols: Their recognition and interpretation: Princeton, New Jersey, Princeton University Press, p. 139–179. Bain, D.C., Mellor, A., Robewrson-Rintoul, M.S.E., and Buckland, S.T., 1993, Variations in weathering processes and rates with time in a chronosequence of soils from Glen Feshie, Scotland: Geoderma, v. 57, p. 275–293, doi: 10.1016/0016-7061(93)90010-I. Besly, B., and Turner, P., 1983, Origin of the red beds in a moist tropical climate (Etruria Formation, Upper Carboniferous, U.K.), in Wilson, R.C.L, ed., Residual Deposits: Surface Related Weathering Processes and Materials: Oxford, Blackwell Scientific, p. 131–147. Birkeland, P.W., 1984, Soils and Geomorphology, 2nd edition: New York, Oxford University Press, 372 p.
40
Tabor et al.
Birkeland, P.W., 1999, Soils and Geomorphology, 2nd edition: New York, Oxford University Press, 430 p. Bown, T.M., and Kraus, M.J., 1981, Lower Eocene alluvial paleosols (Willwood Formation, northwest Wyoming, USA) and their significance for palaeoecology, palaeoclimatology and basin analysis: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 43, p. 1–30, doi: 10.1016/00310182(81)90056-0. Bossi, G.E., 1971, Analisis de la cuenca de Ischigualasto-Ischichuca: Congreso Hispano-Luso-Americano de Geologia Economica, Trabajos, v. 1, p. 611–622. Brewer, R., 1976, Fabric and Mineral Analysis of Soils: New York, NY, Krieger, 482 p. Budd, D.A., Pack, S.M., and Fogel, M.L., 2002, The destruction of paleoclimatic isotopic signals in Pleistocene carbonate soil nodules of Western Australia: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 188, p. 249–273, doi: 10.1016/S0031-0182(02)00588-6. Buol, S.W., Hole, F.D., McCracken, R.J., and Southard, R.J., 1997, Soil Genesis and Classification: Ames, Iowa, Iowa State University Press, 527 p. Cerling, T.E., and Quade, J., 1993, Stable carbon and oxygen isotopes in soil carbonates, in Swart, P.K, Lohmann, K.C., McKenzie, J., and Savin, S., eds., Climate Change in Continental Isotopic Records: American Geophysical Union Geophysical Monograph 78, p. 217–231. Craig, H., 1953, The geochemistry of the stable carbon isotopes: Geochimica et Cosmochimica Acta, v. 3, p. 53–92, doi: 10.1016/0016-7037(53)90001-5. Currie, B.S., Tabor, N.J., Shipman, T.C., Montañez, I.P., and Moore, K.A., 2001, Stratigraphic architecture of the Upper Triassic Ischigualasto Formation, Ischigualasto Provincial Park, northwestern Argentina: Geological Society of America Abstracts with Programs, v. 33, no. 6, p. 75. Daniels, R.B., Gamble, E.E., and Nelson, L.A., 1971, Relations between soil morphology and water table levels on a dissected north-central Carolina coastal plain surface: Soil Science Society of America Proceedings, v. 35, p. 157–175. Dansgaard, W., 1964, Stable isotopes in precipitation: Tellus, v. 16, p. 436–468. Demko, T.M., Currie, B.S., and Nicoll, K.A., 2004, Regional paleoclimatic and stratigraphic implications of paleosols and fluvial/overbank architecture in the Morrison Formation (Upper Jurassic), Western Interior, USA: Sedimentary Geology, v. 167, p. 115–135, doi: 10.1016/ j.sedgeo.2004.01.003. Deutz, P., Montañez, I.P., Monger, H.C., and Morrison, J., 2001, Morphology and isotope heterogeneity of late Quaternary pedogenic carbonates; implications for paleosols carbonates as palaeoenvironmental proxies: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 166, p. 293–317, doi: 10.1016/S0031-0182(00)00214-5. Duchaufour, P., 1982, Pedology: Pedogenesis and Classification: London, George Allen and Unwin, 448 p. Dworkin, S.I., Nordt, L., and Atchley, S., 2005, Determining terrestrial paleotemperatures using the oxygen isotopic composition of pedogenic carbonate: Earth and Planetary Science Letters, v. 237, p. 56–68, doi: 10.1016/ j.epsl.2005.06.054. Ferguson, K.M., Gregory, R.T., and Constantine, A., 1999, Lower Cretaceous (Aptian-Albian) secular changes in the oxygen and carbon isotope record from high paleolatitude, fluvial sediments, southeast Australia: Comparisons to the marine record, in Barrera, E., and Jonson, C.C., eds., Evolution of the Cretaceous Ocean-Climate System: Geological Society of America Special Paper 332, p. 59–72. Franzmeier, F.P., Bryant, R.B., and Steinhardt, G., 1985, Characteristics of Wisconsinan glacial tills in Indiana and their influence on Argillic horizon development: Soil Science Society of America Journal, v. 49, p. 1481–1486. Gile, L.H., Hawley, J.W., and Grossman, R.B., 1981, Soils and Geomorphology in the Basin and Range Area of Southern New Mexico—Guidebook to the Desert Project: New Mexico Bureau of Mines and Mineral Resources Memoir 39, 222 p. Gonfiantini, R., 1978, Standards for stable isotope measurements in natural compounds: Nature, v. 271, p. 534–536, doi: 10.1038/271534a0. Gradstein, F.M., Agterberg, F.P., Ogg, J.G., Hardenbol, J., Van Veen, P., Thierry, J., and Huang, Z., 1995, A Triassic, Jurassic and Cretaceous time scale, in Berggren, W.A., Kent, D.V., Aubry, M.P., and Hardenbol, J., eds., Geochronology, Time Scales and Global Stratigraphic Correlation: SEPM Special Publication 54, p. 95–125. Gradstein, F., Ogg, J., and Smith, A., 2004, A Geologic Time Scale: Cambridge, UK, Cambridge University Press, 589 p. Jenny, H., 1941, Factors of Soil Formation: New York, McGraw-Hill, 281 p. Kraus, M.J., 1987, Integration of channel and floodplain suites, II. Vertical
relations of alluvial paleosols: Journal of Sedimentary Petrology, v. 57, p. 602–621. Leeder, M.R., 1976, Palaeogeographic significance of pedogenic carbonates in the topmost upper Old Red Sandstone of the Scottish border basin: Geological Journal, v. 11, p. 21–28. Levin, N.E., Quade, J., Simpson, S.W., Semaw, S., and Rogers, M., 2004, Isotopic evidence for Plio-Pleistocene environmental change at Gona, Ethiopia: Earth and Planetary Science Letters, v. 219, p. 93–110, doi: 10.1016/ S0012-821X(03)00707-6. López-Gamundí, O.R., Espejo, I.S., Conaghan, P.J., Powell, C.McA., and Veevers, J.J., 1994, Southern South America, in Powell, C.McA., and Veevers, J.J., eds., Permian-Triassic Pangean Basins and Foldbelts along the Panthalassan Margin of Gondwanaland: Geological Society of America Memoir 184, p. 281–329. Mack, G.H., and James, W.C., 1994, Paleoclimate and global distribution of paleosols: The Journal of Geology, v. 102, p. 360–366. Mack, G.H., James, W.C., and Monger, H.C., 1993, Classification of paleosols: Geological Society of America Bulletin, v. 105, p. 129–136, doi: 10.1130/0016-7606(1993)105<0129:COP>2.3.CO;2. Martínez, R., 1994, Estratigrafic del Sector Agua De la Pea Rio de La Chilca, Formacion Ischigualasto, Triasico Superior, Cuenca de Ischigualasto: [M.S. thesis]: San Juan, Argentina, Universidad Nacional de San Juan, 210 p. McCarthy, P.J., and Plint, A.G., 2003, Spatial variability of palaeosols across Cretaceous interfluves in the Dunvegan Formation, NE British Columbia, Canada; palaeohydrological, palaeogeomorphological and stratigraphic implications: Sedimentology, v. 50, p. 1187–1220, doi: 10.1111/j.13653091.2003.00600.x. Milana, J.P., and Alcober, O., 1994, Modelo tectonosedimentario de la cuenca Triásica de Ischigualasto (San Juan, Argentina): Revista de la Asociación Argentina, v. 49, p. 217–235. Moore, D.M., and Reynolds, R.C., 1997, X-ray diffraction and the identification and analysis of clay minerals: New York, Oxford University Press, 378 p. Moore, K.A., 2003, Coupled organic and carbonate isotope analysis of Late Triassic paleosols, Ischigualasto Basin, NW Argentina [M.S. thesis]: Davis, University of California, 127 p. Munsell Color, 1975, Munsell Soil Color Charts: Baltimore, Maryland, Munsell Color, 24 p. O’Neil, J.R., Clayton, R.N., and Mayeda, T.K., 1969, Oxygen isotope fractionation in divalent metal carbonates: The Journal of Chemical Physics, v. 51, p. 5547–5558, doi: 10.1063/1.1671982. Pfefferkorn, H.W., and Fuchs, K., 1991, A field classification of fossil plant substrate interactions: Neues Jahrbuch fuer Geologie und Palaeontologie, Abhandlungen, v. 183, p. 17–36. PiPujol, M.D., and Buurman, P., 1994, The distinction between groundwater glay and surface-water gley phenomena in Tertiary paleosols of the Ebro Basin, NE Spain: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 110, p. 103–113, doi: 10.1016/0031-0182(94)90112-0. Ramos, V.A., and Kay, S.M., 1991, Triassic rifting and associated basalts in the Cuyo basin, central Argentina, in Harmon, R.S., ed., Andean Magmatism and its Tectonic Setting: Geological Society of America Special Paper 265, p. 79–91. Retallack, G.J., 1976, Triassic palaeosols in the upper Narrabeen Group of New South Wales. Part I. Features of the palaeosols: Journal of the Geological Society of Australia, v. 23, p. 383–399. Retallack, G.J., 1977, Triassic palaeosols in the upper Narrabeen Group of New South Wales. Part II. Classification and reconstruction: Journal of the Geological Society of Australia, v. 24, p. 19–37. Retallack, G.J., 1983, A paleopedological approach to the interpretation of terrestrial sedimentary rocks: The mid-Tertiary paleosols of Badlands National Park, South Dakota: Geological Society of America Special Paper 193, 82 p. Retallack, G.J., 1988, Field recognition of paleosols, in Reinhardt, J., and Sigleo, W.R., eds., Paleosols and Weathering through Geologic Time: Principles and Applications: Geological Society of America Special Paper 216, p. 1–20. Retallack, G.J., 1990, Soils of the Past: An Introduction to Paleopedology: Boston, Unwin Hyman, 520 p. Rogers, R.R., Swisher, C.C., III, Sereno, P.C., Monetta, A.M., Forster, C.A., and Martínez, R.N., 1993, The Ischigualasto tetrapod assemblage (Late Triassic, Argentina), and 40Ar/39Ar dating of dinosaur origins: Science, v. 260, p. 794–797. Rozanski, K., Araguas-Aaruas, L., and Gonfiantini, R., 1993, Isotopic patterns in modern global precipitation, in Swart, P.K., Lohmann, K.C., McKenzie,
A Late Triassic soil catena J., and Savin, S., eds., Climate Change in Continental Isotopic records. Geophysical Monograph 78, p. 1–36. Royer, D.L., 1999, Depth to pedogenic carbonate horizon as a paleoprecipitation indicator?: Geology, v. 27, p. 1123–1126, doi: 10.1130/00917613(1999)027<1123:DTPCHA>2.3.CO;2. Shipman, T.C., 2004, Links between sediment accumulation rates and the development of alluvial architecture: Triassic Ischigualasto Formation, northwestern Argentina [Ph.D. dissertation]: Tucson, University of Arizona, 187 p. Slate, J.L., 1996, Buried carbonate paleosols developed in Pliocene-Pleistocene deposits of the Pasco Basin, south-central Washington, USA: Quaternary International, v. 34–36, p. 191–196, doi: 10.1016/1040-6182(95)00084-4. Soil Survey Staff, 1975, Soil Taxonomy: Washington, D.C., U.S. Department of Agriculture Handbook 436, 754 p. Soil Survey Staff, 1998, Keys to Soil Taxonomy: Washington, D.C., U.S. Department of Agriculture, Natural Resources Conservation Service, 221 p. Spalletti, L., Artabe, A., Morel, E., and Brea, M., 1999, Biozonación paleoflorística y cronoestratigrafía del Triásico Argentino: Ameghiana, v. 36, p. 419–451. Steila, D., 1976, The Geography of Soils: New Jersey, Prentice-Hall, 222 p. Stipanícic, P.N., and Bonaparte, J.F., 1972, Cuenca Triasica de Ischigualasto– Villa Unión (Provincias de San Juan y La Rioja), in Leanza, A.F., ed., Geología Regional Argentina: Córdoba, Argentina, Academia Nacional de Ciencias, p. 503–576. Stipanícic, P.N., and Bonaparte, J.F., 1979, Cuenca Triásica de Ischigualasto– Villa Union (Provincias de La Rioja y San Juan), in Turner, J.C., ed., Segundo Simposio de Geología Regional Argentina: 1: Córdoba, Argentina, Academia Nacional de Ciencias, p. 523–575. Tabor, N.J., and Montañez, I.P., 2004, Morphology and distribution of fossil soils in the Permo-Pennsylvanian Wichita and Bowie Groups, north-central Texas, USA: Implications for western equatorial Pangean palaeoclimate during icehouse-greenhouse transition: Sedimentology, v. 51, p. 851–884, doi: 10.1111/j.1365-3091.2004.00655.x. Tabor, N.J., and Montañez, I.P., 2005, Oxygen and hydrogen isotope compositions of Permian pedogenic phyllosilicates: Development of modern surface domain arrays and implications for paleotemperature reconstructions: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 223, p. 127–146, doi: 10.1016/j.palaeo.2005.04.009. Tabor, N.J., Montañez, I.P., Zierenberg, R., and Currie, B.S., 2004, Mineralogical and geochemical evolution of a basalt-hosted fossil soil (Late Triassic, Ischigualasto Formation, northwest Argentina); potential for paleoenvironmental reconstruction: Geological Society of America Bulletin, v. 116, p. 1280–1293, doi: 10.1130/B25222.1. Uliana, M.A., and Biddle, K.T., 1988, Mesozoic-Cenozoic paleogeographic
41
and geodynamic evolution of southern South America: Revista Brasileira de Geociencias, v. 18, p. 172–190. Uliana, M.A., Biddle, K.T., and Cerdan, J., 1989, Mesozoic extension and the formation of Argentine sedimentary basins, in Balkwill, H.R., and Tankard, A.J., eds., Extensional Tectonics and Stratigraphy of the North Atlantic Margins: American Association of Petroleum Geologists Memoir 46, p. 599–614. Valencio, D.A., Mendia, J.E., and Vilas, J.F., 1975, Palaeomagnetism and K-Ar ages of Triassic igneous rocks from the Ischigualasto-Ischichuca Basin and Puesto Viejo Formation, Argentina: Earth and Planetary Science Letters, v. 26, p. 319–330, doi: 10.1016/0012-821X(75)90007-2. Veevers, J.J., Powell, C.McA., Collinson, J.W., and López-Gamundí, O.R., 1994, Synthesis, in Veevers, J.J., and Power, C.McA., eds., Permian-Triassic Pangean Basin and Foldbelts along the Panthalassan Margin of Gondwanaland: Geological Society of America Memoir 184, p. 331–353. Vepraskas, M.J., 1994, Redoximorphic features for identifying aquic conditions: North Carolina Agricultural Resources Service Technical Bulletin 301. Volkheimer, W., 1969, Paleoclimatic evolution in Argentina and relationships with other regions of Gondwana: Gondwana Stratigraphy: Paris, France, Unesco p. 551–587. Wigley, T.M., 1978, Mass transfer and carbon isotope evolution in natural water systems: Geochimica et Cosmochimica Acta, v. 42, p. 1117–1140, doi: 10.1016/0016-7037(78)90108-4. Wilding, L.P., and Tessier, D., 1988, Genesis of Vertisols: Shrink-swell phenomena, in Wilding, L.P., and Puentes, R., eds., Vertisols: Their Distribution, Properties, Classification and Management: Texas A&M University Technical Monograph 18, p. 55–81. Yapp, C.J., 1993, The stable isotope geochemistry of low temperature Fe(III) and Al “oxides” with implications for continental paleoclimates, in Swart, P.K, Lohmann, K.C., McKenzie, J., and Savin, S., eds., Climate Change in Continental Isotopic Records: American Geophysical Union Geophysical Monograph 78, p. 285–294. Yapp, C.J., 2000, Climatic implications of surface domains in arrays of δD and δ18O from hydroxyl minerals: Goethite as an example: Geochimica et Cosmochimica Acta, v. 64, p. 2009–2025, doi: 10.1016/S00167037(00)00347-1. Zapata, T.R., and Allmendinger, R.W., 1996, Growth stratal records of instantaneous and progressive limb rotation in the Precordillera thrust belt and Bermejo Basin, Argentina: Tectonics, v. 15, p. 1065–1083, doi: 10.1029/96TC00431.
MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
Investigating paleosol completeness and preservation in mid-Paleozoic alluvial paleosols: A case study in paleosol taphonomy from the Lower Old Red Sandstone Susan B. Marriott School of Geography and Environmental Management, University of the West of England, Bristol, Coldharbour Lane, Bristol BS16 1QY, UK V. Paul Wright School of Earth, Ocean and Planetary Sciences, Cardiff University, Cardiff CF10 3YE, UK, and BG Group, 100 Thames Valley Park Drive, Reading RG6 1PT, UK ABSTRACT Preservational bias in paleosol formation is rarely discussed and remains a major issue in paleopedology. The relatively simple paleosol profiles of the Silurian-Devonian Old Red Sandstone alluvial successions of southwest Wales provide an opportunity to investigate the completeness of a widespread type of calcic Vertisol. Reactivated, truncated cumulate horizons provide means of assessing the dynamics of floodplain erosion and accumulation. While these distinctive profiles are not especially common, effects of low-magnitude erosion events were probably masked, affecting only the topmost part of the upper soil horizon. In the absence of a stabilizing rooted vascular plant cover in pre–mid-Paleozoic sediments, such mobile upper soil horizons were likely a common feature. Keywords: calcic Vertisols, Old Red Sandstone, floodplain development, soil development. RESUMEN La posibilidad de preservación de paleosuelos es un hecho raramente discutido, a pesar de ser un tema importante dentro de la paleoedafología. Los perfiles de paleosuelos relativamente sencillos de las sucesiones aluviales del Silúrico-Devónico de la Old Red Sandstone en el suroeste de Gales ofrecen una buena oportunidad para investigar si el registro de algunos tipos de paleosuelos muy frecuentes, los Vertisuelos cálcicos, es completo o no. Los horizontes reactivados, truncados y compuestos de estos paleosuelos nos aportan los medios para conocer la dinámica de la erosión y acumulación dentro de la llanura de inundación. En los casos en los que estos perfiles no son muy comunes, los efectos de eventos erosivos de baja magnitud quedan enmascarados afectando sólo a la parte más alta del horizonte superior del suelo. En ausencia de una cobertera vegetal de plantas vasculares enraizadas que estabilizaran la superficie, como es el caso de los sedimentos anteriores al Paleozoico medio, estos horizontes móviles del suelo fueron probablemente un rasgo característico. Palabras clave: Vertisuelos cálcicos, Old Red Sandstone, desarrollo de llanuras de inundación, desarrollo de suelos. Marriott, S.B., and Wright, V.P., 2006, Investigating paleosol completeness and preservation in mid-Paleozoic alluvial paleosols: A case study in paleosol taphonomy from the Lower Old Red Sandstone, in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 43–52, doi: 10.1130/2006.2416(03). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
43
44
Marriott and Wright
INTRODUCTION Paleosols are common in ancient alluvial successions, reflecting the relatively low rates of floodplain accretion in relation to rates of soil development (Marriott and Wright, 1993). Traditionally, paleosols in such deposits have been used to indicate a range of controls including sedimentation rate and landscape stability. In this study, we develop earlier work (Marriott and Wright, 1993) to show how the careful analysis of reactivated soil horizons can provide insights into the preservational history and completeness of paleosols and the dynamics of Silurian-Devonian floodplains, indicating the likelihood that much time is lost at subtle erosion surfaces. Such erosive processes may have been more prevalent in pre–mid-Paleozoic soils where the lack of an extensive rooted plant cover promoted instability and the formation of an active upper soil level. FLOODPLAIN DEVELOPMENT Nanson and Croke (1992) devised a comprehensive model of floodplain development related to the relationship between primary geomorphic factors, such as channel cutting and filling, and secondary geomorphic factors, such as peat accumulation. These factors are dependent on stream power and sediment load and will, therefore, result in different environments for floodplain formation. Over the long term (as in the stratigraphic record), most floodplains would probably be classified as polyphase floodplains (Nanson and Croke, 1992) because they are built up by transformations caused by river channel changes in response to extrinsic variations (such as climate and base-level change) that alter flow regime or sediment load (Schumm, 1977). The changes in channel processes are reflected in floodplain development, although there is often a hysteresis or changes occur at a slower rate. The floodplain surface is, therefore, a highly dynamic environment, and evidence in the sedimentary record of changes to this environment is often incomplete and complicated by destructive, erosive episodes (Brakenridge, 1981). Whether or not soils develop on land surfaces depends on the surface being relatively stable (Ruhe, 1956; Brakenridge, 1981; Bull, 1992). Active erosion prevents any soil development at all and soils will not develop to any degree where there is a high rate of sediment input. Rates of sedimentation on floodplains are highly variable (Marriott, 1998), although, since some characteristic features of soils such as mottling and development of pedogenic slickensides can develop relatively rapidly (<100 yr), rates of accretion of several millimeters per year (Wright and Marriott, 1996) would be required to prevent any soil processes from occurring. Soils should, then, be present in alluvial sequences and, because of the polyphase nature of floodplain development, are likely to record the dynamic nature of the floodplain surface by having a polygenetic nature themselves. Johnson and Watson-Stegner (1987) and Johnson et al. (1990) proposed an evolution model of pedogenesis that investigated dynamic soil development over time. The model suggested
that most soils are polygenetic due to development along both progressive and regressive pathways. The progressive element relates to processes or conditions that promote horizonation, the incorporation of small increments of sediment added to the surface (developmental upbuilding) and soil deepening. Regressive pedogenesis involves processes or conditions that lead to condensed or simple profiles with few horizons, soil removal by erosion or mass wasting, and retardant upbuilding. This condition occurs when the amount of sediment deposited on the soil surface is sufficient to impede or retard horizon development. The dynamic pedogenesis model (Johnson et al., 1990) is particularly relevant to the evolution of soils on floodplains since the processes of developmental and retardant upbuilding may be the most important factors in an environment where the other state factors of climate, organisms, and parent material (Jenny, 1961) do not vary substantially over time. Soil development on floodplains reflects the polygenetic nature of floodplain development with both progressive and regressive pathways (Johnson et al., 1990). Phillips (1993) suggested a chaos model for complex patterns of soil development that appears to be particularly relevant in environments where erosion (regressive) processes occur frequently. It was found that, on floodplain surfaces, the degree of soil development reflected the age and stability of the geomorphic surface (Phillips 1990). SOILS IN OLD RED SANDSTONE FLOODPLAIN SEDIMENTS The Silurian-Devonian Lower Old Red Sandstone of southwest Wales and the Welsh Borderlands is a continental basin-fill sequence that covers an area of at least 20,000 km2 (Fig. 1). Generally, it is a sequence of upward-coarsening, marginal marine, fluvial and alluvial fan sediments with some prominent tuff beds and well-developed calcrete horizons that act as regional markers (Allen and Williams, 1982; Williams et al., 2004). The lowermost part of the Lower Old Red Sandstone of the Anglo-Welsh Basin is a succession of mainly red alluvial sediments comprising relatively thin red or greenish-gray sand bodies (Williams and Hillier, 2004) separated by thick red to purple mudstone deposits in a ratio of ~1:4 (Marriott et al., 2005). The overall depositional system has been interpreted as low-gradient floodplains, perhaps on distal alluvial or terminal fans in a subhumid to semiarid environment (Allen and Williams, 1982; Marriott and Wright, 1993, 2004; Love and Williams, 2000), with very infrequent, shortlived, marine incursions (Allen, 1973b; Barclay et al., 1994). The stratigraphy of the Lower Old Red Sandstone of the Anglo-Welsh Basin was originally described by Dixon (1921), and later revised by Allen and Williams (1978), who allocated the succession in southwest Wales to the Milford Haven Group. Soils developed on floodplains are affected by sediment transport and deposition, and they are well known in the Old Red Sandstone, having been described in detail by Allen (1973a, 1974, 1986) and Marriott and Wright (1993). They take the
Paleosol completeness and preservation
45
North
Figure 1. Lower Old Red Sandstone in the Anglo-Welsh Basin, UK. Localities mentioned in text: FWW—Freshwater West; FWE—Freshwater East; M—Manorbier; LL—Llansteffan; L—Lydney.
Abergavenny
Haverfordwest
L LL FWW
M FWE
Swansea
Cardiff
Lower Old Red Sandstone outcrops Localities mentioned in text
form of calcic Vertisols that show varying stages of calcrete development in a Ck horizon (Machette, 1985), and varying degrees of development of a structural B horizon with pedogenic slickensides (wedge-shaped peds) (Fig. 2). Though rarely preserved, a complete profile would range from 1 to 3 m in thickness (Tandon and Friend, 1989) and would have three horizons (Marriott and Wright, 1993): an upper (A) horizon, which would probably have had a surface crust and (deep) desiccation cracks during the dry season; a middle structural (B) horizon with curved, intersecting slickensided surfaces; and a lower horizon (Bk or Ck), which would contain calcrete nodules in various developmental forms (Machette, 1985). The Moor Cliffs Formation is the central unit of the Milford Haven Group, and it crops out extensively on the southwest coast of Wales (Fig. 1). It varies in thickness from 120 to 365 m and is dominated by red to purple mudstones with subsidiary, thin sand bodies and intraformational conglomerates. Many of the mudstones have been pedogenically altered and contain prominent pedogenic slickensides from structural B horizons and varying stages of calcrete development in Bk or Ck horizons. The degree of development of the calcretes ranges from small (~10 mm diameter), discrete nodules and larger prismatic nodules (~50 mm in diameter, up to 150 mm long) to laminar petrocalcic horizons, though these are not common in the Moor Cliffs Formation. Development of the Bk or Ck horizons in the Vertisols on Old Red Sandstone floodplains is mainly related to residence time
50km
(i.e., the length of time sediment layers remain in pedogenically active zones before burial or erosion) (Leeder, 1975). Frequency of floods that resulted in episodes of stripping or aggradation will therefore result in different residence times and, consequently, different modes of soil development (Wright and Marriott, 1996). For example, developmental upbuilding (Johnson and WatsonStegner, 1987), i.e., frequent regular additions of small amounts of sediment to the soil surface, is likely to result in a cumulate soil profile, where slow aggradation causes gradual overprinting of soil horizons; retardant upbuilding (Johnson and WatsonStegner, 1987), or infrequent “dumping” of a large thickness of sediment, will give rise to either a composite or compound soil profile depending on whether the pedogenically active zone is partially or completely buried (Wright and Marriott, 1996). Examples of polygenetic paleosols from the Old Red Sandstone were described by Marriott and Wright (1993), where it was envisaged that the occurrence of fanned splays of prismatic nodules (Fig. 3) resulted from a progressive phase where frequent increments of small amounts of sediment had resulted in a cumulate profile that was subsequently truncated by rapid erosion of up to 3 m of sediment during a major regressive phase (e.g., a 10,000–yr–recurrence interval superflood [Marriott and Wright, 2004]). A subsequent progressive phase of slow aggradation then led to reactivation of the structural B horizon and displacement of the nodules with movement along the slickensided slip surfaces (see Fig. 9 of Marriott and Wright, 1993).
46
Marriott and Wright
Figure 2. Two simple calcic Vertisol profiles from the Raglan Mudstone Formation at Lydney. Calcic horizon in center is 0.5 m thick, and the top of the horizon is picked out by the dashed line. The slickensided surfaces on structural B horizons (1.75 m thick in top paleosol) are picked out by reduction. The A horizon is missing. The whole section shown is 2.5 m thick.
Figure 3. Distorted prismatic calcrete nodules overlain by intraformational conglomerate (indicated by black arrows) and laminated mudstone with tubules in the Moor Cliffs Formation at Manorbier. This sequence is at the top of that shown in Figure 8A (bracketed section). The pen for scale is 145 mm long.
Many of the units containing fanned, prismatic nodules are overlain by an intraformational conglomerate that contains granule- to pebble-grade reworked calcrete and mudstone clasts in a mudstone matrix (arrowed on Fig. 3) (Allen and Williams, 1979; Marriott and Wright, 1993, 1996, 2004). More recent study of
the Lower Old Red Sandstone sequences at Freshwater West, Freshwater East, and Llansteffan in South Wales (Fig. 1) has indicated that, in some cases, the later progressive phase also resulted in the formation of prismatic peds within the intraformational conglomerate and subsequent distortion of the peds
Paleosol completeness and preservation with the underlying calcrete nodules. This led us to re-examine the polygenetic nature of these particular units and to propose further models for their development. POLYPHASE DEVELOPMENT OF OLD RED SANDSTONE FLOODPLAINS Development of the calcic Vertisols in the Lower Old Red Sandstone shows both progressive and regressive phases that indicate different rates of deposition and erosion. These can be linked to episodes of aggradation and stripping that reveal the dynamic nature of the depositional environment. The starting point for the new models is a cumulate profile (as in Marriott and Wright, 1993) (Fig. 4). This occurs when slow aggradation on the floodplain surface causes the soil horizons to move upward progressively, so that the structural B horizon, with curved, slickensided surfaces (wedge-shaped peds), is overprinted by the calcrete nodules of the Bk or Ck horizon. The period of aggradation is likely to have occurred over a relatively long period of time (104 to 105 yr) so that prismatic calcrete nodules (stage II of Machette, 1985) could develop (Wright and Marriott, 1996, their Fig. 1b). As the profile aggrades, secondary prismatic structure in the structural B horizon is likely to influence vertical, prismatic nodule development in the overprinting Bk or Ck horizon. Model A (Fig. 5) assumes that the progressive phase is then followed by a series of erosive events that lower the surface by removing a few millimeters of sediment at a time or in pulses.
47
Thus, the seasonal wetting front gradually lowers so that older, buried structural B horizons are gradually reactivated and cause splaying of the prismatic calcrete nodules in the overprinted zone. Erosion is likely to cease when the Bk or Ck horizon is A Progressive or pulsed slow erosion
extreme distortion due to lowered surface and wetting front before deposition of conglomerate
Ai Slow erosion and slow burial process
product
** Cumulate profile
Key to figures 4,5 and 6 A
open cracks
distortion of
conglomerate will
conglomerate and
be less distorted
further distortion
than nodular zone
of nodular zone closed cracks Bt on A
pedogenic slickensides (wedge-shaped peds) calcrete concretions or nodules
Aii Slow erosion and rapid burial process
product
Bt calcrete clast conglomerate
Ck on Bt
lowering of surface
burial of surface
** undeformed
Ck
erosion surface
isolation below
conglomerate
wetting front highly deformed nodular zone
Figure 4. Cumulate soil profile (see also Marriott and Wright, 1993). Typical thicknesses for the different horizons would be: A, 0.5–1 m; Bt, 0.75–1 m; Ck, 0.5–1 m. The overprinted Bt on A and Ck on Bt horizons would vary in thickness depending on the amount and rate of surface aggradation.
Figure 5. Type A model for development of fanned prismatic nodules. Progressive or pulsed slow erosion is followed by slow burial (Ai) or by rapid burial (Aii). ** denotes diagnostic features for comparison between models; remainder of key and scale are as in Figure 4.
48
Marriott and Wright
reached because it represents a less erodible surface. The record of renewed deposition may then recommence with an intraformational conglomerate, either as a lag beneath muddy channel deposits or as part of inclined heterolithic strata (IHS) from muddy point bar or accretionary bench deposits (see Marriott and Wright [2004] and Marriott et al. [2005] for a description of models for the depositional environments). A unit with splayed nodules (see Figs. 8 and 10 of Marriott and Wright [1993] for additional examples) is not always overlain by an intraformational conglomerate, since these conglomerates are more likely to occur in relation to channel-related deposits (Gómez-Gras and Alonso-Zarza, 2003). The next phase may be slow deposition and burial or rapid burial by a single major deposition event (Fig. 5). During slow aggradation of the floodplain surface (model Ai) as new soil horizons form, further distortion of the nodules is likely as wetting and drying cycles take place while this zone lies within the structural B horizon. Nodules toward the top of the lower unit may be distorted to a greater degree than nodules lower down, which may not be affected by reactivation. Because the intraformational conglomerate has a mud matrix, it is likely to reflect pedogenic processes and the formation of prismatic peds that may also be distorted when they coincide with the B horizon, though to a lesser degree than the calcrete nodules. Model Aii illustrates the case of subsequent rapid burial, where the amount of sediment added is sufficient to prevent pedogenic processes acting on the buried material (retardant upbuilding). This is a regressive process (Johnson and WatsonStegner, 1987), as, although the surface is aggrading, the resulting sequence is a compound soil where profiles are separated by an undistorted intraformational conglomerate. The lower profile is truncated and contains distorted nodules. In the field, it may be difficult to distinguish between models Ai and Aii if the conglomerate is thin or absent. Model B (Fig. 6) is similar to that portrayed by Marriott and Wright (1993, their Fig. 9) and is characterized by a single, major erosion event that removes surface horizons. In this case, calcrete nodules in the Bk or Ck horizon are not distorted by progressive lowering. Slow burial following deposition of an intraformational conglomerate causes reactivation of the slickensided surfaces and distortion of both the calcrete nodules and the conglomerate as the structural B horizon moves upward. This model differs from Ai in that the nodules and conglomerate are distorted to a similar degree. If rapid erosion is followed by rapid burial (model Bii), then as with model Aii, a compound soil will develop, though it will differ from that in Aii because the prismatic calcrete nodules are not deformed (or only very weakly, depending on the thickness of sediment added) (Fig. 7). This is most commonly seen where truncated profiles are overlain by fluvial channel facies that can be mud-dominated. It must be stressed that distortion of the prismatic nodules is only likely to occur following erosion and subsequent reactivation of a cumulate profile. In a simple calcic Vertisol,
there would not be sufficient host sediment between prismatic (stage II) calcrete nodules for pedogenic slickensides in the structural B horizon to overprint the Bk or Ck horizon as the surface was lowered.
B Rapid erosion : one phase
single deep erosive event
no distortion of nodular zone at this point
Bi Rapid erosion and slow burial process
product
reactivation of
distorted
slickenside surfaces
conglomerate
and distortion of conglomerate
nodule fans
and nodules as zone comes within depth
**
of wetting
similar distortion of nodule zone and conglomerate
Bii Rapid erosion and rapid burial process
product
isolation below
** undeformed
wetting front
conglomerate
minor reactivation
weakly deformed
(if any)
(or undeformed) nodular zone
Figure 6. Type B model for development of fanned prismatic nodules. Rapid erosion is followed by slow burial (Bi) or by rapid burial (Bii). ** denotes diagnostic features for comparison between models; remainder of key and scale are as in Figure 4.
Paleosol completeness and preservation
49
Figure 7. Weakly distorted prismatic nodules underlying an undeformed intraformational conglomerate from the Moor Cliffs Formation (Upper Silurian) of Freshwater East, illustrating type Bii mode of development.
LANDSCAPE DEVELOPMENT IN THE LOWER OLD RED SANDSTONE Soil profile development is often used as an indicator of landscape development and for reconstructing geomorphic histories, particularly in Quaternary soils (Brakenridge, 1981). It is not considered to be a very reliable indicator of surface ages or deposits, however, due to complex spatial and temporal patterns of soil development (Phillips, 1993). The ideas outlined herein show how the degree of soil development preserved in the Old Red Sandstone paleosols, together with inferences on likely episodes of aggradation and erosion, can be used to gain an understanding of the dynamic nature and polyphase development of the paleoenvironment (Fig. 8A). Recently, Marriott and Wright (2004) and Marriott et al. (2005) proposed models for paleoenvironmental reconstruction of the early Lower Old Red Sandstone of the Anglo-Welsh Basin. The models envisage a multistage, multichannel system where different parts of the system were active over different time scales. Calcic Vertisols are important features of the modeled landscapes since the nature and stage of their development act as indicators of particular features on the Old Red Sandstone floodplains and of the frequency and magnitude of erosional and depositional events (Fig. 8B) (see also Wright and Marriott, 1996). The overall depositional environment is considered to be analogous to that of a low-gradient, dryland river system in a relatively sediment-starved basin possibly on a distal alluvial or
terminal fan margin. The main trunk stream occupies a wide twostage channel, with a moderately sinuous low-flow channel and muddy braidplain that becomes active during relatively frequent flood events (annual to 1 in 10 return-interval floods) as part of the proximal floodplain. More distal floodplain areas would only be affected by longer-return-interval floods (100–1000 yr return interval), and it is likely that catastrophic stripping and major landform effects would only take place during low-frequency superflood events (10,000+ yr intervals). Because the Old Red Sandstone paleosols are calcic Vertisols, the climate regime is likely to have been subhumid to semiarid, with a marked seasonality of rainfall. Vegetation in the Late Silurian to Early Devonian would not have had extensive root systems capable of binding sediment, and plants are likely to have been confined to areas near to rivers, lakes, and ponds (Edwards and Richardson, 2004). Thus, the self-mulching aspect of Vertisols is likely to have provided a ready supply of sand- and silt-sized soil aggregates that could be entrained and redeposited during flooding and incorporated in channel bedload. Additionally, during dry seasons, dust storms are likely to have acted on the land surface, creating a mosaic of cells where the soil surface could be deflated and aggraded periodically by intermittent events (Pickup, 1991). In areas where sediment accumulation occurred relatively frequently, such as near to the main stream and on the braidplain during flooding, or in aggradational cells, cumulate soils could develop (Fig. 4), though only relatively rapidly formed features, such as pedogenic
50
Marriott and Wright
A 6
}
5
Figure 3
4
3
2
1
0 preserved sequence
a
b
c
d
e
f
g
h
KEY remanent lamination
reactivated vertic horizon with distorted nodules
vertical burrows
pedogenic slickensides (wedge-shaped peds)
reactivated vertic horizon with silt infilled slickensides
desiccation cracks
stage I calcrete nodules
calcretized burrows (tubules)
lowering of surface
stage II calcrete nodules
intraformational conglomerate
Figure 8. (A) Polyphase development of a 6 m section of the Moor Cliffs Formation at Manorbier measured from 18 m above the top of the Rook’s Cave Tuff bed. For discussion on depositional environments, see Marriott and Wright (2004). Bracketed section is illustrated in Figure 3. Episodes: (a) Laminated mudstone with vertical burrows is deposited subaqueously and bioturbated in floodplain ponds. (b) Pond sediment is exposed, laminations in upper section are destroyed by pedogenesis, and burrows lower in profile act as nuclei for calcrete formation (tubules). (c) Developmental upbuilding—small increments of sediment are added by overbank deposition or aeolian input. Cumulate soil profile shows stage I calcrete, implying frequent aggradational events averaging between 0.1 and 1 mm/yr (see Figure 1 of Wright and Marriott, 1996). (d) Developmental upbuilding—small increments of sediment are added by overbank deposition or aeolian input as in episode c. Cumulate soil profile shows stage II prismatic calcrete nodules, implying slow aggradation averaging between 0.01 and 0.1 mm/yr. (e) Regressive pedogenesis—pulsed or slow erosion occurs (see Fig. 5) down to top of Bk horizon; pedogenic slickensides are reactivated; and surface cracking in dry season allows ingress of fine sediment between slip surfaces. Stage II prismatic nodules are distorted with the wedge-shaped peds as the surface is lowered. (f) Renewed progressive pedogenesis occurs with developmental upbuilding as in d, and at a similar rate. The underlying nodular zone is further distorted as the surface aggrades until the wetting front moves out of this zone. (See Figure 5, as Ai but with no intraformational conglomerate lag.) (g) Further episodes of regressive pedogenesis occur as in e, but with less-intense rotation of nodules, perhaps suggesting more rapid lowering of the floodplain surface than in e. (h) Rapid burial as in Figure 5, Aii, with a lag of intraformational conglomerate then subaqueous deposition and bioturbation as in a. Subsequent exposure and progressive pedogenesis lead to a further cumulate profile with tubules, as in b and c. Only part of the unit is shown here.
slickensides (wedge-shaped peds), will be evident if the mean rate of sediment input to the soil surface was greater than 1 mm/ yr (Wright and Marriott, 1996). In the example shown in Figure 8A, the frequent aggradational events assumed for episode c may have taken place in the proximal floodplain or braidplain, whereas the slow deposition rates envisioned for episodes d and f could
relate to more distal floodplain areas due either to avulsion or distributary abandonment during prolonged dry periods. Frequent dust storms may also have produced the progressive lowering of floodplain surfaces necessary to cause reactivation of the structural B horizons and further movement along the slickensided surfaces of the wedge-shaped peds. This would cause fanning
Paleosol completeness and preservation
B
51
6
f h
metres of sediment
5 g
d
4 3
e
2
c b
1 a
0 100
200
300
400
500
time (ka) Figure 8. (B) Time line demonstrating likely accretion and erosion episodes a–h from part A. The 6 m sedimentary sequence illustrated represents at least 500 k.y. based on soil development rates envisioned by Wright and Marriott (1996). The time periods lost during the erosional episodes e and g are the minimum. At present, there are no means of quantifying the length of time involved, merely identifying whether erosion was likely pulsed (relatively slow) or rapid.
of prismatic calcrete nodules that had overprinted the structural B horizon during an earlier aggradational phase (Fig. 3; Model A, Fig. 5). Perhaps prior to colonization of the land surface by plants with subaerial parts capable of limiting wind erosion and roots capable of binding the soil, many soils were prone to the sorts of erosional effects documented here. Such soils would not necessarily have modern analogues, and the erosion-prone upper mobile layer was possibly a common feature. The erosional episodes identified in the example used here (Figs. 8A and 8B) were likely infrequent events where the overall depth of erosion exceeded the thickness of the upper soil horizons and reached the Ck horizons. Had not these reactivated horizons been subsequently buried permanently, the erosive episodes would not be recorded. It is likely that erosive events of lower magnitude (but greater frequency) also occurred but that no evidence remains. For example, high-frequency erosional events just affecting the A horizon would not be identifiable. This masking of finer-scale events, together with the likely serendipitous preservation of major erosive events, suggests that only part of depositional (stratigraphic) time is preserved in these successions. Indeed the whole issue of soil taphonomy in terms of completeness is rarely considered. It is possible to envision situations where a soil could have a relatively young upper horizon overlying much older lower horizons.
dynamic nature of the ancient floodplain surfaces. Four different models have been presented that show polygenetic soil profile development related to either slow or rapid erosion followed by slow or rapid burial of cumulate calcic Vertisols. A particular section of the Moor Cliffs Formation has been used to illustrate the polyphase nature of floodplain development where both progressive and regressive pedogenesis occurred. It is likely that many pre–mid-Paleozoic soils were also prone to erosion due to a mobile A horizon rather than an organically stabilized one, though some may have had microbial or bryophyte cover that would have given the surface some cohesion and resistance to erosion. In the Lower Old Red Sandstone, the combination of a lack of deep-rooted vegetation, vertic processes favoring small ped formation, and a flood-prone hydrologic regime in a distal alluvial or terminal fan setting led to development of soils with particularly unstable upper horizons. ACKNOWLEDGMENTS We are grateful to Greg Retallack and Bill Barclay for their helpful comments on an earlier version of this article, and we particularly wish to thank Paul Revell for drawing the diagrams. REFERENCES CITED
CONCLUSIONS Lower Old Red Sandstone paleosols from the Anglo-Welsh basin show a variety of developmental forms that indicate the
Allen, J.R.L., 1973a, Compressional structures (patterned ground) in Devonian pedogenic limestones: Nature Physical Science, v. 243, p. 84–86. Allen, J.R.L., 1973b, A new find of bivalve molluscs in the Uppermost Downtonian (Lower Old Red Sandstone) of Lydney, Gloucestershire: Proceedings
52
Marriott and Wright
of the Geologists’ Association, v. 84, p. 27–30. Allen, J.R.L., 1974, Studies in fluviatile sedimentation: Implications of pedogenic carbonate units, Lower Old Red Sandstone, Anglo-Welsh outcrop: Geological Journal, v. 9, p. 181–208. Allen, J.R.L., 1986, Pedogenic carbonates in the Old Red Sandstone facies (Late Silurian–Early Devonian) of the Anglo-Welsh area, southern Britain, in Wright, V.P., ed., Paleosols: Their Recognition and Interpretation: Oxford, Blackwell Scientific, p. 58–86. Allen, J.R.L., and Williams, B.P.J., 1978, The sequence of the earlier Lower Old Red Sandstone, north of Milford Haven, southwest Dyfed: Geological Journal, v. 13, p. 113–136. Allen, J.R.L., and Williams, B.P.J., 1979, Interfluvial drainage on Siluro-Devonian alluvial plains in Wales and the Welsh Borders: Journal of the Geological Society of London, v. 136, p. 361–366. Allen, J.R.L., and Williams, B.P.J., 1982, The architecture of an alluvial suite: Rocks between the Townsend Tuff and Pickard Bay Tuff Beds (Early Devonian) southwest Wales: Philosophical Transactions of the Royal Society of London, v. B297, p. 51–89. Barclay, W.J., Rathbone, P.A., White, D.E., and Richardson, J.B., 1994, Brackish water faunas from the St. Maughans Formation: the Old Red Sandstone section at Ammons Hill, Hereford and Worcester, UK, re-examined: Geological Journal, v. 29, p. 369–379. Brakenridge, R.G., 1981, Late Quaternary floodplain sedimentation along the Pomme de Terre River, southern Missouri: Quaternary Research, v. 15, p. 62–76. Bull, W.B., 1992, Geomorphic Responses to Climatic Change: New York, Oxford University Press, 344 p. Dixon, E.E.L., 1921, Geology of the South Wales Coalfield. Part XIII. The Country around Pembroke and Tenby: Memoir of the Geological Survey of England and Wales, sheets 244 and 245: London, HMSO, 220 p. Edwards, D., and Richardson, J.B., 2004, Silurian and Lower Devonian plant assemblages from the Anglo-Welsh Basin: a palaeobotanical and palynological synthesis: Geological Journal, v. 39, p. 375–402, doi: 10.1002/gj.997. Gómez-Gras, D., and Alonso-Zarza, A.M., 2003, Reworked calcretes: Their significance in the reconstruction of alluvial sequences (Permian and Triassic, Minorca, Balearic Islands, Spain): Sedimentary Geology, v. 158, p. 299–319, doi: 10.1016/S0037-0738(02)00315-9. Jenny, H., 1961, Derivation of state factor equations of soils and ecosystems: Proceedings of the Soil Science Society of America, v. 25, p. 385–388. Johnson, D.L., and Watson-Stegner, D., 1987, Evolution model of pedogenesis: Soil Science, v. 143, p. 349–366. Johnson, D.L., Keller, E.A., and Rockwell, T.K., 1990, Dynamic pedogenesis: New views on some key soil concepts, and a model for interpreting Quaternary soils: Quaternary Research, v. 33, p. 306–319. Leeder, M., 1975, Pedogenic carbonates and floodplain sediment accretion rates: A quantitative model for alluvial arid-zone lithofacies: Geological Magazine, v. 112, p. 257–270. Love, S., and Williams, B.P.J., 2000, Sedimentology, cyclicity and floodplain architecture in the Lower Old Red Sandstone of SW Wales, in Friend, P.F.,
and Williams, B.P.J., eds., New Perspectives on the Old Red Sandstone: Geological Society [London] Special Publication 180, p. 371–388. Machette, M.N., 1985, Calcic soils of the southwestern United States, in Weide, D.L., ed., Soils and Quaternary Geology of the Southwestern United States: Geological Society of America Special Paper 203, p. 1–21. Marriott, S.B., 1998, Channel-floodplain interactions and deposition on floodplains, in Bailey, R.G., José, P.V., and Sherwood, B.R., eds., United Kingdom Floodplains: Yorkshire, Westbury Press, p. 43–61. Marriott, S.B., and Wright, V.P., 1993, Palaeosols as indicators of geomorphic stability in two Old Red Sandstone alluvial suites, south Wales: Journal of the Geological Society of London, v. 150, p. 1109–1120. Marriott, S.B., and Wright, V.P., 1996, Sediment recycling on Siluro-Devonian floodplains: Journal of the Geological Society of London, v. 153, p. 661–664. Marriott, S.B., and Wright, V.P., 2004, Mudrock deposition in an ancient dryland system: Moor Cliffs Formation, Lower Old Red Sandstone, southwest Wales, UK: Geological Journal, v. 39, p. 277–298, doi: 10.1002/gj.990. Marriott, S.B., Wright, V.P., and Williams, B.P.J., 2005, A new evaluation of fining-upward sequences in a mud-rock dominated succession of the Lower Old Red Sandstone of south Wales, UK, in Blum, M.D., Marriott, S.B., and Leclair, S., eds., Fluvial Sedimentology VII: International Association of Sedimentologists Special Publication 35, p. 517–529. Nanson, G.C., and Croke, J.D., 1992, A genetic classification of floodplains: Geomorphology, v. 4, p. 459–486, doi: 10.1016/0169-555X(92)90039-Q. Phillips, J.D., 1990, Relative ages of wetland and upland surfaces as indicated by pedogenic development: Physical Geography, v. 11, p. 363–378. Phillips, J.D., 1993, Progressive and regressive pedogenesis and complex soil evolution: Quaternary Research, v. 40, p. 169–176, doi: 10.1006/ qres.1993.1069. Pickup, G., 1991, Event frequency and landscape stability on the floodplain systems of arid central Australia: Quaternary Science Reviews, v. 10, p. 463–473, doi: 10.1016/0277-3791(91)90007-H. Ruhe, R.V., 1956, Geomorphic surfaces and the nature of soils: Soil Science, v. 82, p. 441–455. Schumm, S.A., 1977, The Fluvial System: New York, Wiley, 338 p. Tandon, S.K., and Friend, P.F., 1989, Near surface shrinkage and carbonate replacement processes, Arran Cornstone Formation, Scotland: Sedimentology, v. 36, p. 1113–1126. Williams, B.P.J., and Hillier, R.D., 2004, Variable alluvial sandstone architecture within the Lower Old Red Sandstone, southwest Wales: Geological Journal, v. 39, p. 257–275, doi: 10.1002/gj.993. Williams, B.P.J., Marriott, S.B., and Hillier, R.D., 2004, Preface: The Lower Old Red Sandstone of the Anglo-Welsh Basin: An introduction: Geological Journal, v. 39, p. 233–236, doi: 10.1002/gj.1003. Wright, V.P., and Marriott, S.B., 1996, A quantitative approach to soil occurrence in alluvial deposits and its applications to the Old Red Sandstone of Britain: Journal of the Geological Society of London, v. 153, p. 907–913. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
Calcareous paleosols of the Upper Triassic Chinle Group, Four Corners region, southwestern United States: Climatic implications Lawrence H. Tanner† Department of Biological Sciences, Le Moyne College, 1419 Salt Springs Road, Syracuse, New York 13214, USA Spencer G. Lucas New Mexico Museum of Natural History, 1801 Mountain Road N.W., Albuquerque, New Mexico 87104, USA ABSTRACT Paleosols are prominent features of the Upper Triassic Chinle Group. The oldest (Carnian-age) formations of the Chinle Group (Zuni Mountains and Shinarump Formations) contain kaolinitic paleosols that display gley features but generally lack calcretes. Paleosols of the (Upper Carnian) Blue Mesa Member of the Petrified Forest Formation are mostly mature Alfisols that have distinctive horizonation and commonly host stage II to III calcretes. Mudstones of the Jim Camp Wash Bed of the overlying Sonsela Member host similarly mature paleosols with abundant stage II to stage IV calcretes. The (Lower Norian) Painted Desert Member of the Petrified Forest Formation is characterized by paleosols that lack well-developed A horizons but display thick, red B horizons in which pedogenic slickensides, rhizocretions, and stage II to III calcretes are locally abundant. Immature paleosols hosting stage II to stage III calcretes characterize the lower part of the (Middle Norian) Owl Rock Formation. The upper Owl Rock Formation contains stage III to IV calcretes and laterally persistent limestone ledges that formed as palustrine limestones and groundwater calcretes. The (Norian-Rhaetian) Rock Point Formation generally lacks pedogenic features in most of the study area, but the uppermost strata in some locations host multiple pedogenic horizons that display drab root traces, desiccation cracks, and stage II to III calcretes. Interformational variations in the types of paleosols and the maturity of calcretes in Chinle Group strata reflect gradual aridification across the Colorado Plateau during the Late Triassic. This climatic change overprinted variations in basin sedimentation rate that were potentially controlled by base level and tectonics. Keywords: Chinle, pedogenic, calcretes, Late Triassic, palustrine. RESUMEN La presencia de paleosuelos es uno de los rasgos más característicos del Grupo Chinle del Triásico Superior. Las formaciones más antiguas, Zuni Mountains y Shinarump, son de edad Carniense, no tienen calcretas y los paleosuelos son caoliníticos con rasgos de gley. Los paleosuelos del Miembro Blue Mesa de la Formación Petrified Forest son sobre todo Alfisoles maduros con horizontes bien diferenciados y con E-mail:
[email protected].
†
Tanner, L.H., and Lucas, S.G., 2006, Calcareous paleosols of the Upper Triassic Chinle Group, Four Corners region, southwestern United States: Climatic implications, in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 53–74, doi: 10.1130/2006.2416(04). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
53
54
Tanner and Lucas calcretas de estadios II a III. Las lutitas de las capas Jim Camp Wash del Miembro Sonsela suprayacente también tienen paleosuelos maduros con frecuentes calcretas de estadios II a IV. El miembro Painted Desert (Noriense inferior) de la Formación Petrified Forest se caracteriza por presentar paleosuelos que carecen de horizontes A bien desarrollados, pero que presentan horizontes B rojos y muy potentes con slickensides pedogénicos, rizocreciones, y calcretas de estadios II a III localmente abundantes. Los paleosuelos inmaduros que contienen calcretas de estadio II a III caracterizan la parte inferior de la Formación Owl Rock (Noriense medio). La parte superior de dicha formación contiene calcretas estadio III a IV y lateralmente incluyen lentejones de calizas, que se han interpretado como depósitos palustres y calcretas freáticas. La formación Rock Point (Noriense-Rhetiense) no presenta rasgos pedogénicos en la mayor parte del área estudiada, pero localmente en los estratos superiores hay horizontes pedogénicos múltiples que presentan trazas de raíces, grietas de desecación y calcretas estadios II–III. Las variaciones en el tipo de paleosuelos y en los estadios de madurez de las calcretas en las distintas formaciones del Grupo Chinle reflejan una aridificación gradual a lo largo de la Meseta del Colorado durante el Triásico Superior. Este cambio climático, controló las variaciones en la tasa de sedimentación en la cuenca que también estuvieron potencialmente controladas por cambios en el nivel de base y por la tectónica. Palabras clave: Chinle, pedogénico, calcretas, Triásico Superior, palustres.
INTRODUCTION The utility and temporal resolution of paleoclimate modeling has improved dramatically in recent decades, in part through the increasing use of paleosols and pedogenic features as paleoclimate archives. Calcrete, the accumulation of CaCO3 in the subsurface environment as nodules (glaebules) or cemented horizons (see review in Wright and Tucker, 1991), has proven particularly useful because of its high preservation potential, and common association with soil-forming processes in semiarid to arid climates. In addition to climate, however, the morphology and maturity of all paleosols, including those that are calcareous, are controlled by a variety of factors, including sediment accumulation rate, which controls residence time of sediment in the soil-forming environment, vegetative cover, subsurface biotic activity, and host sediment composition (see reviews in Kraus, 1999; Retallack, 2001; Alonso-Zarza, 2003). The controls on the rate of sediment accumulation, which in general is inversely related to paleosol maturity, are particularly complex. Although climate exerts some control over the delivery of sediment to the receiving basin, tectonic activity greatly influences sediment deposition, both through enhancement of sediment production by source area uplift, and through control of basin configuration and accommodation space. Paleosols have long been recognized as prominent features in Upper Triassic Chinle Group strata, and general features of many of these paleosols have been described previously. Some previous studies have focused on specific local occurrences within an individual formation; for example, Kraus and Middleton (1987) described a history of floodplain incision and aggradation cycles during deposition of the Petrified Forest Formation
using, in part, variations in paleosol maturity on the alluvial plain; Therrien and Fastovsky (2000) documented that paleosol hydrology varied with distance from the channel in the same formation. Other studies have cited general characteristics of Chinle Group paleosols, typically in the context of regional changes in climate during the Late Triassic, but generally without detailed documentation of specific pedogenic features (Dubiel, 1987, 1994; Dubiel and Hasiotis, 1994a, 1994b; Hasiotis and Dubiel, 1994; Demko et al., 1998; Hasiotis et al., 1998). The Chinle Group outcrops cover a broad area of the Southwestern United States (Fig. 1) and record deposition during a major portion of the Late Triassic. A thorough and detailed examination of the lateral and temporal variations of the Chinle paleosols is not possible within the scope a single paper. Rather, we survey the spectrum of paleosols present in Chinle Group strata, with particular attention to calcretes, by describing the results of previous studies, as well as providing new data. From these data (old and new), we attempt to interpret the general sedimentologic and climatic conditions that existed during deposition and pedogenesis of Chinle strata. Calcretes occur in most Chinle formations, yet they have received only limited attention. Most previous studies have focused on individual formations at specific locations; examples include Blodgett’s (1988) study of the Norian-Rhaetian Dolores Formation, in the uppermost Chinle of southwest Colorado, the examination of Carnian-Norian age paleosols of the Petrified Forest Formation in the Petrified Forest National Park (PFNP; Figs. 1 and 2) by Therrien and Fastovsky (2000), and the description of pedogenic features of the Norian-aged Owl Rock Formation in the Four Corners region by Tanner (2000). The accumulation of carbonate in the B horizon is not specific to any single soil type or climatic region, however, so the conditions of calcrete
Calcareous paleosols of the Upper Triassic Chinle Group
55
Figure 1. Distribution of outcrops of Chinle Group strata (shaded) in the Four Corners region (after Stewart et al., 1972). Locations mentioned in the text are indicated; PFNP—Petrified Forest National Park, LPDCP—Little Painted Desert County Park.
formation can only be understood within the context of the soils, or paleosols, in which the calcrete occurs. GEOLOGIC SETTING The Chinle basin formed as a continental retro-arc basin on the western edge of the North American craton during the initial growth of the Cordilleran magmatic arc in the early Mesozoic (Dickinson, 1981; Lawton, 1994). This basin extended from southwestern Texas to northern Wyoming and was the site of terrestrial sedimentation from the Late Triassic until the beginning of the Early Jurassic (Lucas et al., 1997). Strata of the Chinle Group, ranging in age from Late Carnian to possibly Rhaetian, are exposed across much of the Colorado Plateau (Fig. 1). The Four Corners region (the common border of the states of Arizona, Colorado, New Mexico, and Utah), the focus of this study, was situated within the basin at near-equatorial latitudes (between 5° and 15°N) during Late Triassic time (Scotese, 1994; MolinaGarza et al., 1995; Kent and Olsen, 1997).
Deposition of the Chinle Group sediments was controlled by predominantly west- to northwest-flowing stream systems crossing broad, low-gradient alluvial plains. In the Four Corners region, proximal source areas for these sediments were the Mogollon highlands, located ~500 km to the south and southwest, and to a lesser extent the Uncompahgre highlands located 200–300 km to the east and northeast (Blakey and Gubitosa, 1983; Marzolf, 1994). Syndepositional arc volcanism contributed an appreciable amount of volcaniclastic sediment to the basin. Across most of the Four Corners region, the lowermost Chinle strata were deposited unconformably on Middle Triassic or older strata following an interval of lowered base level and incision. STRATIGRAPHY In the Four Corners region, basal Chinle Group strata rest unconformably (the Tr-3 unconformity) on strata of the Moenkopi Formation. Stewart et al. (1972) used the informal designation “mottled strata” to describe alluvial sediments at the base of
56
Tanner and Lucas
Figure 2. Stratigraphic hierarchy for the Chinle Group used in this paper (after Lucas et al., 1997).
the Chinle Group that exhibit strong pedogenic mottling. These strata underlie or are laterally equivalent to the basal strata of the Shinarump Formation (Lucas et al., 1997). Equivalent strata in the San Rafael Swell were named the Temple Mountain Formation by Robeck (1956), and more recently, Heckert and Lucas (2003) proposed the name Zuni Mountain Formation for these same strata in west-central and north-central New Mexico (Lucas et al., 2003). The Shinarump Formation consists of crossbedded conglomerates and sandstones of late Carnian age. The Temple Mountain and Shinarump Formations were deposited in paleovalleys incised into the underlying Moenkopi Formation strata (Stewart et al., 1972; Blakey and Gubitosa, 1983; Demko et al., 1998). Upper Carnian strata immediately above the Shinarump Formation are named regionally the Cameron, Bluewater Creek, and Monitor Butte Formations. Lucas (1993) and Lucas et al. (1997) demonstrated the stratigraphic equivalence of these formations. The lowermost strata of the Petrified Forest Formation in the Four Corners region are designated the Blue Mesa Member (Lucas et al., 1997). These strata, also of late Carnian age, overlie the Cameron–Bluewater Creek–Monitor Butte Formations. The Blue Mesa Member and the underlying Cameron– Bluewater Creek–Monitor Butte and Shinarump Formations
were mapped collectively as the lower bentonitic part of the Chinle Formation by Stewart et al. (1972). Thickness of these formations varies from <30 m in southeastern Utah to >450 m in northwestern New Mexico. An unconformity (Tr-4) separates the Blue Mesa Member from the overlying sandstone-dominated Sonsela Member of the Petrified Forest Formation and the equivalent Moss Back Formation (Lucas, 1993; Heckert and Lucas, 1996; Lucas et al., 1997). Heckert and Lucas (2002a) interpreted the Sonsela Member as filling erosional scours in the underlying Blue Mesa Member, which thins beneath the unconformity to the east. The Painted Desert Member, of early to middle Norian age, overlies Sonsela– Moss Back strata (Lucas et al., 1997). As mapped by Stewart et al. (1972), the thickness of the entire Petrified Forest Formation ranges from just over 30 m at its northeastern limit in eastern Utah, increasing southward to over 400 m at its southeastern extent in northwestern New Mexico. The overlying Owl Rock Formation is composed of up to 150 m of strata that crop out in northern Arizona, northwestern New Mexico, and southern Utah (Stewart et al., 1972; Lucas and Huber, 1994). Across the Four Corners area, the upper Norian to (possibly) Rhaetian Rock Point Formation is recognized as the youngest stratigraphic unit of the Chinle Group (Lucas, 1993; Lucas et al.,
Calcareous paleosols of the Upper Triassic Chinle Group
57
1997). The Rock Point Formation, termed the Rock Point Member of the Wingate Sandstone by Stewart et al. (1972), includes strata formerly assigned to the Church Rock Member of the Chinle Formation (Stewart et al., 1972; Dubiel, 1989; Lucas et al., 1997). The contact between the Rock Point Formation and underlying Owl Rock Formation is unconformable (the Tr-5 unconformity). The Rock Point Formation grades vertically to the eolian-dominated Wingate Formation of Rhaetian to Hettangian age (Harshbarger et al., 1957; Tanner et al., 2002; MolinaGarza et al., 2003).
such assignments are not possible, we use the alternative classification system specific for paleosols, which abandoned some traditional soil orders and created new paleosol orders (Mack et al., 1993). The term calcrete is used here to refer to the displacive and replacive growth of calcium carbonate in the soil-forming environment, and includes carbonate of both pedogenic and groundwater origins (see reviews in Wright and Tucker, 1991; Tandon and Kumar, 1999; Alonso-Zarza, 2003). The maturity of nodular calcrete is described using the stage concepts of Gile et al. (1966) and Machette (1985).
PALEOSOLS OF THE CHINLE GROUP
Zuni Mountains Formation (= Mottled Strata)
Many descriptions of paleosols have suffered from the lack of a single system of paleosol description applied by all researchers. In part, this is because traditional modern soil classification (i.e., Soil Conservation Service, 1999) requires knowledge of such information as vegetative cover and soil moisture that may not preserved in paleosols. Additionally, postdepositional processes (erosion, compaction, diagenesis) may obscure some primary pedogenic features and induce other nonpedogenic features (Retallack, 2001). Although the precise interpretation of paleosols in terms of modern soil classification standards is often problematic, sufficient information is often retained to allow classification at the order level (Kraus, 1999; Retallack, 2001). This study assigns order names to paleosols that are consistent with modern soil usage unless otherwise noted (Table 1). Where
Lithostratigraphy Stewart et al. (1972) used the informal designation “mottled strata” to describe alluvial sediments (mudstones, sandstones, and conglomerates) at the base of the Chinle Group that exhibit strong pedogenic mottling. These strata underlie or are laterally equivalent to the basal strata of the Shinarump Formation (Lucas et al., 1997). Equivalent strata in the San Rafael Swell in central Utah were named the Temple Mountain Formation by Robeck (1956), and more recently, Heckert and Lucas (2003) proposed the name Zuni Mountain Formation for these same strata in westcentral and north-central New Mexico. These same authors also noted that, locally, the Shinarump Formation may be absent, in which case the Zuni Mountains Formation is overlain by the Bluewater Creek Formation. At the type location, near Fort
TABLE 1. SUMMARY OF PALEOSOL TYPES AND CALCRETE FEATURES IN STRATA OF THE CHINLE GROUP Age Stratigraphic unit Depositional environment Paleosols Calcretes Late Norian–Rhaetian Rock Point Formation Eolian, ephemeral lake, Aridisol/Inceptisol, Local stage II–III minor alluvial Calcisol Middle Norian
Owl Rock Formation
Alluvial channel, floodplain, minor lacustrine and wetland
Calcisol, calcic Alfisol
Common stages II–III, local stage IV, groundwater calcretes
Early to middle Norian
Painted Desert Member, Petrified Forest Formation
Alluvial channel and floodplain
Vertisols, Alfisols
Stage II common, local stage III, common calcrete channel lag deposits
Carnian-Norian boundary
Sonsela Member, Petrified Forest Formation
Alluvial channel and minor floodplain
Calcic Alfisols
Stage II common, local stage III, common calcrete channel lag deposits
Latest Carnian
Blue Mesa Member, Petrified Forest Formation
Alluvial channel and floodplain
Alfisols, Vertisols
Abundant stage II, some local stage III
Late Carnian
Monitor Butte, Mesa Redondo and Cameron Formations
Alluvial channel, floodplain, and minor lacustrine
Vertisols, Alfisols
Stage II common locally
Late Carnian
Shinarump Formation
Alluvial channel and minor interfluve
Bioturbated with gley features
Absent
Late Carnian
Zuni Mountains and Temple Mountain Formations–mottled strata
Interfluve valley fills
Spodosols
Generally absent/rare stage III
58
Tanner and Lucas
Wingate, New Mexico, over 20 m of pedogenically modified strata overlie the Middle Triassic Moenkopi Formation (Heckert and Lucas, 2003). Pedogenic Features Strata of the Zuni Mountains Formation display, as do other correlative exposures in the region, extensive pedogenic modification, consisting of some combination of gleying (the presence of low chroma colors) and bioturbation. Typical features of these paleosols are crudely prismatic fabric, bluish to yellowish gray mottling in a dark reddish to orange brown host (Fig. 3), relict bedding, meniscate burrows, sandy horizons that are cemented almost entirely by hematite (spodic or Bs horizons), a clay mineral assemblage dominated by kaolinite, and penetration of the beds by near-vertical sandstone cylinders up to 1.5 m long (Tanner, 2003a). Indeed, the most striking characteristics of the Fort Wingate section are the prominent mottled horizons and penetration of these horizons by the vertical sandstone-filled casts. Originally, these latter structures, present in much of the Chinle Group, were interpreted as lungfish aestivation burrows (Dubiel et al., 1987); more recently, they have been reinterpreted as crayfish burrows (Hasiotis and Dubiel, 1993a) and the casts of deeply penetrating taproots of monopodial vegetation (Lucas and Hayden, 1989; Tanner, 2003a). Both deep taproots and crayfish burrowing would be possible, perhaps even likely, in regions where meterscale, water table fluctuations occur regularly. These hydrologic conditions would have been conducive to water-logged soils for humid intervals, but periodic, perhaps seasonal, drawdown of the water table would have been sufficient to allow translocation and oxidation of iron and manganese and shrinkage of expandable clays. However, the abundance of kaolinite in the clay mineral
Figure 3. Typical pedogenic features of the Zuni Mountains Formation in section near Fort Wingate, New Mexico (= mottled strata), include pale greenish-yellow (10 YR 8/2) and light greenish-gray (5 GY 8/1) mottling in moderate brown (5 YR 3/4) to dark reddish-brown (10 R 3/4) matrix. Hammer head for scale is 17 cm long.
assemblage of this formation (Tanner, 2003a; Tabor et al., 2004) suggests substantial humidity. Lower Chinle (undifferentiated) paleosols have been described previously as Gleysols (sensu Mack et al., 1993) formed in a humid but seasonal environment (Dubiel and Hasiotis, 1994b; Hasiotis et al., 1998). Demko et al. (1998) suggested that hydromorphism in lower Chinle paleosols was a consequence of stratigraphic proximity to an aquitard in the underlying Moenkopi Formation. Bown and Kraus (1987), Mack et al. (1993), and Retallack (2001) noted, however, that true hydromorphic (gleyed) soils rarely display extensive bioturbation or evidence of desiccation, and the identification of Gleysols cannot be based solely on the presence of mottled horizons. Conversely, pseudogleying, as demonstrated in these lower Chinle paleosols, indicates that hydromorphic conditions existed only periodically. The abundance of root casts in these paleosols demonstrates that the sediment surface was quite well-vegetated. The presence of hematite-cemented sandstone units in the profile at Fort Wingate documents instead the formation of a well-developed spodic (Bs) horizon. The sandy host was cemented, principally by hematite, in a zone in which clays had largely been destroyed by weathering (Birkeland, 1984). Therefore, we suggest that some of these paleosols formed as Spodosols, which typically form beneath forests (Birkeland, 1984; Retallack, 2001). Thus, these profiles may represent composite palosols in the sense that soil layers buried by subsequent increments of sediment remained in an extensively thick, active soil-forming environment (Wright and Marriott, 1996). A climate characterized by abundant but highly seasonal precipitation is consistent with these features. Alternatively, individual paleosol horizons may have formed as time-separated increments within a complex paleosol that was subjected to an overall pedogenic overprint by later conditions; e.g., long-term climate change could have imparted an overprint through downward translocation of clays or oxides formed under more humid conditions than existed during earlier soil formation. This also may account for the presence of gley features in horizons displaying evidence of desiccation. Calcretes are generally absent in lowermost Chinle Group strata (i.e., Zuni Mountains and Shinarump Formations) in New Mexico and Arizona, although Tabor et al. (2004) reported a significant accumulation of pedogenic carbonate at the base of a Chinle profile in eastern Utah. These authors described the occurrence of a 1.5-m-thick coalesced nodular (Bkm) horizon (stage III) at the base of a 3.5 m profile, overlain by noncalcareous, mottled kaolinitic mudstone with a blocky fabric and containing goethite nodules. Thick Bk horizons, as described by Tabor et al. (2004), are not typically associated with paleosols containing gley colors, but as noted by Bown and Kraus (1987), calcareous nodules may form in hydromorphic paleosols. Indeed, the formation of a clay pan in the lower B horizon is likely to enhance carbonate accumulation and nodule formation. Tabor et al. (2004) suggested, however, that the profile they described records multiple episodes of pedogenesis, potentially under varying climatic conditions. Therefore, this Bkm horizon may be a relict paleosol
Calcareous paleosols of the Upper Triassic Chinle Group formed in Moenkopi strata and unrelated to the later pedogenesis concomitant with initial Chinle deposition. Shinarump Formation Lithostratigraphy The Shinarump Formation consists of cross-bedded conglomerates and quartz arenite sandstones of late Carnian age deposited in paleovalleys incised into the underlying Moenkopi Formation strata (Stewart et al., 1972; Blakey and Gubitosa, 1983; Demko et al., 1998). The quartz arenite sandstones and extrabasinal conglomerates of the Shinarump Formation attain a maximum thickness of 76 m in the Four Corners region, although in many locations, this unit is generally thinner or absent (Fig. 2). The formation records infill of incised paleovalleys by northwest-flowing streams that generally carried a high bedload and were of low sinuosity (Stewart et al., 1972; Blakey and Gubitosa, 1984). Pedogenic Features Paleosols are rare in these sandstone-dominated fluvial channel deposits, but interfluve mudstones display extensive pedogenic mottling and bioturbation, as described by Dubiel (1994) and Tanner (2003a). The nature of this pedogenic alteration appears as purple, orange, and gray mottles in a sandy mudstone exhibiting a prismatic fabric, desiccation cracks, and relict ripple lamination. Bedding-plane exposures reveal centimeterscale concentric banding of purple and yellow zones, potentially a consequence of iron translocated downward along taproots, the casts of which are abundant in Shinarump Formation paleosols. The character of these paleosols is generally similar to that exhibited by the older Zuni Mountains Formation, although the lack of a spodic horizon makes classification more ambiguous. The gley features present here are likewise considered pseudogleying and not indicative of true soil hydromorphism; rather, these features record substantial fluctuations in the position of the water table. Cameron–Bluewater Creek–Monitor Butte Formations Lithostratigraphy The laterally equivalent Cameron, Bluewater Creek, and Monitor Butte Formations, of Upper Carnian age, consist mainly of gray bentonitic to red mudstones, and laminated to cross-bedded fine-grained sandstones. At Petrified Forest National Park, the Bluewater Creek Formation is made up of mainly interbedded sandstone, siltstone, and reddish-purple to grayish-red mudstone (Heckert and Lucas, 2002a). In the Zuni Mountains, however, the Bluewater Creek Formation consists of three distinct lithofacies assemblages: reddish-brown, bluish-gray, and grayish-purple mudstones; ripple-laminated to plane-bedded sandstones; and interbedded bentonitic mudstone and dark shale (Heckert and Lucas, 2002b). The mudstones locally contain abundant plant debris (Ash, 1987, 1989), and thin micritic limestone occurs near the base of the formation (Heckert and Lucas, 2002b).
59
Pedogenic Features Tanner (2003a) noted pedogenic features in the basal strata of the Cameron Formation near Cameron, Arizona, that are similar to those observed in the Zuni Mountains and Shinarump Formations, e.g., mottling with gley colors, a high kaolinite content, and pedogenic slickensides. Dubiel (1987) described a similarly mottled unit in the Monitor Butte Formation of southeastern Utah. Dubiel and Hasiotis (1994a, 1994b) interpreted such profiles in (undifferentiated) lower Chinle strata as Gleysols (sensu Mack et al., 1993), although as noted previously, this classification is not consistent with the strict definition of this paleosol order and the evidence for a greatly fluctuating water table. Following the usage of Mack et al. (1993), these paleosols might best be considered gleyed Vertisols. Paleosols that are stratigraphically higher in the Cameron and Monitor Butte Formations differ in that they typically consist of simple profiles with decimeter-scale light-colored horizons overlying thick (up to 8 m) reddened argillic (Bt) horizons (Dubiel and Hasiotis, 1994b; Hasiotis et al., 1998). Bluewater Creek mudstones commonly display pedogenic slickensides and scattered (stage II) centimeter-scale calcrete nodules with alpha fabrics (Heckert and Lucas, 2002b). Dubiel and Hasiotis (1994b) and Hasiotis et al. (1998) labeled these argillic paleosols with albic horizons as Alfisols. Petrified Forest Formation: Blue Mesa Member Lithostratigraphy The lowermost strata of the Petrified Forest Formation (PFF), designated the Blue Mesa Member (of late Carnian age), conformably overlie the Cameron–Bluewater Creek–Monitor Butte Formations in northeastern Arizona and southeastern Utah; this stratigraphic unit is not present in Colorado (Lucas et al., 1997). This interval consists of bentonitic mudstones with variegated hues of blue, gray, purple, and red, and interbedded thin coarse-grained to very fine-grained sandstones, and it attains a thickness of 100 m or more (Lucas, 1993; Lucas et al., 1997). Most of the sandstones are compositionally and texturally immature, with the exception of the Newspaper Rock Bed, a prominent ledge-forming quartz arenite sandstone unit within the Blue Mesa Member that may be local to the Petrified Forest National Park (Heckert and Lucas, 2002a). The alluvial architecture of the Blue Mesa Member consists of thick muddy floodplain deposits, deeply incised by meter-scale channels. The channels are filled by very fine-grained sand and mud, commonly displaying inclined heterolithic strata (Fig. 4A; lateral accretion surfaces), and are surrounded by well-developed levee complexes with splay deposits (Kraus and Middleton, 1987; Therrien and Fastovsky, 2000). Intraformational disconformity surfaces record cycles of valley incision and fill (Kraus and Middleton, 1987). Kraus and Middleton (1987) proposed that base-level changes during deposition of lower Chinle strata resulted in part from episodes of thermotectonically controlled uplift and subsidence in the Mogollon highlands. The presence of volcanic detritus in the
60
Tanner and Lucas
Figure 4. Features of the Blue Mesa Member, Petrified Forest Formation. (A) Channel fill sequence exposed in southern Petrified Forest National Park (the Tepees) displays inclined heterolithic strata set (arrows) ~6 m thick. (B) View of Blue Mesa, in southern Petrified Forest National Park, illustrates contact (at arrow) between Blue Mesa strata and overlying Sonsela Member. Laterally continuous banding in the Blue Mesa strata results from pedogenesis (translocation of oxides and clays) in individual soil profiles formed on an alluvial plain. Ch = lenticular channel-fill deposit. (C) Bk horizon containing abundant (stage II) calcrete nodules in upper Blue Mesa strata, exposed near Moab, Utah. The rule (for scale) is 17 cm. (D) Pedogenic features of the uppermost Blue Mesa Member at the contact with the overlying Sonsela Member (So; Rainbow Forest Bed) at Blue Mesa, Petrified Forest National Park. Rt—drab root traces; No—thin calcrete nodule horizon, Pe—pedogenic slickensides. Hiking staff (for scale) = 120 cm. The mudstone host darkens downward from pale purple (5 P 6/2) to grayish-purple (5 P 4/2).
Chinle, particularly in the Petrified Forest Formation, provides compelling evidence for arc-related magmatism at this time. Pedogenic Features The Blue Mesa Member contains thick, well-developed paleosols with distinctive horizonation (Fig. 4B); these are strikingly well-exposed in the strata in the southern end of the Petrified Forest National Park. Typical are composite profiles consisting of stacked, repetitive sequences of thin, light-colored, crossbedded to ripple-laminated sandstones and mudstones in beds up to 8 m thick that are greenish-gray to dark reddish-brown and mottled gray, purple, and red. Individual profiles within composite profiles commonly contain a thin, sandy ochric epipedon
(A horizon), typically overlying a well-defined pale albic (E) horizon. Thick clay-rich B (Bt) horizons (up to 8 m) are reddish gray to (more commonly) grayish purple, and host pedogenic slickensides (wedge-shaped peds) and pseudoanticlines, downward-tapering sandstone-filled fissures (desiccation fractures), sandstone-filled root casts, a variety of arthropod burrow structures, drab root traces up to 0.1 m long, centimeter-scale reduction spheroids, and calcrete nodule horizons (Kraus and Middleton, 1987; Hasiotis and Dubiel, 1993b; Therrien and Fastovsky, 2000). Calcrete consists most typically of scattered (stage II) centimeter-scale nodules, which commonly display vertical stacking (rhizocretions, sensu Blodgett, 1988), mainly in the uppermost 0.5 m of the horizon. Dubiel and Hasiotis (1994b) and Hasiotis et
Calcareous paleosols of the Upper Triassic Chinle Group al. (1998) noted these features and labeled paleosols of the Blue Mesa Member as Vertisols, where vertic features predominate, and Alfisols, where pale A and/or E horizons overlie reddened or purple argillic B horizons. Bown and Kraus (1987) noted, however, that the A horizon may be thin or absent in some Alfisols; we suggest, therefore, that the primary features of these paleosols is the thick B horizon, even where vertic features are common, and that they also should be considered Alfisols. The purple color that is characteristic of many Blue Mesa paleosols probably results from the coarse crystal size of the hematite in the B horizon, and indicates a high maturity of the profiles (Bown and Kraus, 1987). Kraus and Middleton (1987) noted a systematic variation in pedogenic development correlating with position on the alluvial plain; the paleosol maturity is significantly lower in the incised valley infill deposits than on the surrounding floodplain. Therrien and Fastovsky (2000) noted localized gleying in Blue Mesa paleosols, and interpreted it as poor drainage on low areas of the alluvial plain. Consistent with this observation, Heckert and Lucas (2002a) noted the local occurrence of sideritic nodules in Blue Mesa paleosols, demonstrating reducing conditions in the soil-forming environment. Additionally, we note the presence of thin (centimeter-scale) localized organic-rich layers in the epipedons at the top of Blue Mesa paleosols in Little Painted Desert County Park, north of Winslow (Fig. 1). These horizons are too thin to meet the accepted definition of histic epipedons (O horizon), but they suggest the local formation of Histosols on areas of the floodplain with impeded drainage. Kraus and Middleton (1987) described a catenary relationship in which the paleosol maturity correlates with distance from the channel (Platt and Keller, 1992; Mack and Madoff, 2005). Calcretes are common in Blue Mesa paleosols. Therrien and Fastovsky (2000) described the presence of centimeter-scale (up to 5 cm in diameter) nodules in most Blue Mesa paleosols. These calcretes are mainly limited to horizons in which isolated nodules are abundant (stage II), although more mature (stage III) calcretes occur locally (Fig. 4C; Heckert and Lucas, 2002a). The nodules typically display alpha fabrics; they range from 2 to 8 cm in diameter, have distinct boundaries, are subspherical to irregularly shaped, and commonly display septarian cracking. These nodules occur widely scattered (stage II), but they are locally concentrated in discrete horizons (stage II to incipient stage III) in the middle to upper part of the B horizon of profiles, where they may be associated with pedogenic slickensides; they are particularly prominent immediately below the Newspaper Rock Bed and at the top of the Blue Mesa Member beneath the Rainbow Forest Bed of the Sonsela Member. Numerous locations occur in southern Petrified Forest National Park where the uppermost Blue Mesa strata consist of dark bluish-gray mudstone displaying vertic fractures and centimeter-scale calcareous nodules, in some locations with diffuse boundaries (Fig. 4D). Drab root traces are locally abundant in the mudstone host. These observations are consistent with the interpretation that the paleosols represent mainly Alfisols, which typically form on well-vegetated (forested) surfaces (Bown and Kraus, 1987; Retallack, 2001).
61
Petrified Forest Formation: Sonsela Member and Moss Back Formation Lithostratigraphy The overlying sandstone-dominated Sonsela Member of the Petrified Forest Formation and the laterally equivalent Moss Back Formation consist of up to 50 m of ledge-forming litharenite sandstone and conglomerate, including both intrabasinal and extrabasinal clasts (Stewart et al., 1972; Lucas et al., 1997). Heckert and Lucas (2002a) interpreted the Sonsela Member as filling erosional scours in the underlying Blue Mesa Member, which thins beneath the unconformity to the east. The lower contact with the Blue Mesa Member is clearly erosional and has been interpreted as a regional unconformity (the Tr-4; Lucas et al., 1997). Arc-related tectonism, suggested as the cause of incision-infill cycles in the underlying Blue Mesa Member (Kraus and Middleton, 1987), may explain this unconformity; source area uplift and an increase in the local depositional gradient may have caused incision and reworking of the Blue Mesa strata prior to Sonsela–Moss Back deposition. In particular, the uppermost unit of the Sonsela, the Agate Bridge Bed, contains a significant extrabasinal component, along with a high load of reworked calcrete (type 3 deposit of Gómez-Gras and Alonso-Zarza, 2003). Alternatively, eustasy might have caused a significant baselevel drop; a regional unconformity coincident with sea-level fall occurs in the middle Keuper at about the Carnian-Norian boundary (Aigner and Bachman, 1992), approximately correlative with the Tr-4 unconformity (Lucas et al., 1997; Heckert and Lucas, 2002a). Heckert and Lucas (2002a) examined in detail the stratigraphy of the Sonsela Member in the Petrified Forest National Park and proposed that the Sonsela is composed of three subunits of mappable extent. The lowermost sandstone-dominated unit, which they designated the Rainbow Forest Bed, consists of up to 6 m of quartzarenite sandstone and conglomerate deposited by north-northeasterly flowing streams (Deacon, 1990), and locally contains abundant silicified logs of Araucarioxylon. The gradationally overlying Jim Camp Wash Bed consists of up to 30 m of grayish-purple to pale red bentonitic mudstone and interbedded sandstone. The uppermost unit, the Agate Bridge Bed, consists of up to almost 7 m of cross-bedded quartzarenite and sublitharenite sandstone and conglomerate. The conglomerate contains a significant proportion of intraformational clasts, including mudstone rip-ups and calcrete. Heckert and Lucas (2002a) and Lucas et al. (2003) noted that the Sonsela Member fills scours on the Blue Mesa erosional surface. Pedogenic Features The Jim Camp Wash Bed displays an abundance of pedogenic features including distinct horizonation, pedogenic slickensides, decimeter-scale sandstone-filled desiccation cracks, and abundant calcrete nodules and coalesced calcrete nodule layers (stage II to III). The sedimentology of the Jim Camp Wash Bed is similar to that of the underlying Blue Mesa Member in
62
Tanner and Lucas
Figure 5. Jim Camp Wash Bed section. (A) The top of the section is the lower portion of the Agate Bridge Bed and here consists of ~2 m of mainly planar cross-bedded sandstone with intraformational conglomerate (calcrete nodule) lag. The calcrete described in the text occurs immediately below the base of the sandstone. (B) Calcrete morphologies include rhizocretions (vertically stacked calcrete nodules; Rh) with downward-tapering arrangement. (C) Coalescing nodule horizons (No; stage III) along arcuate surfaces are typically one to two nodules thick and up to 1 m long.
that meter-scale channels are incised into muddy floodplain deposits on which mature paleosols formed. Jim Camp Wash paleosols differ from those of the Blue Mesa Member, however, in their generally redder color and the typically greater maturity of calcrete. Excellent Jim Camp Wash Bed paleosol outcrops occur in southern Petrified Forest National Park. Near the Rainbow Forest Museum, the uppermost Jim Camp Wash Bed consists of grayish-purple mudstone that hosts abundant pale nodules ranging in diameter from 0.5 to 3 cm (Fig. 5A). In the uppermost 40 cm of the mudstone, the nodules form coalesced horizons that are laterally discontinuous, extending a maximum distance of 50 cm, and botryoidal masses up to 30 cm long. Near the top of this zone, nodules form vertically stacked bodies, or rhizocretions, up to 20 cm long (Fig. 5B). Smaller nodules commonly exhibit crosscutting burrows. Notably, nodules are concentrated along arcuate curviplanar surfaces, presumably pedogenic slickensides, that extend laterally up to 1 m (Fig. 5C). These features have relatively smooth upper surfaces and are bounded below by a single layer of coalesced nodules projecting downward. The surfaces dip varying directions and form pseudoanticlinal intersections. Discrete fragments of charcoal, recognizable by a silky, fibrous luster, also occur in the uppermost 40 cm of the mudstone. The size and abundance of nodules decrease markedly in the reddened mudstone below the grayish-purple horizon. Nodules occur to a depth of 50 cm within this zone. The entire mudstone section is overlain by sandstone displaying planar cross-beds and conglomerate lags (Fig. 6). The lag deposits are composed of mudstone rip-ups and micritic nodules. The Jim Camp Wash Bed consists of thick floodplain mudstone sequences overlain by a high bedload stream deposit containing a lag of reworked calcrete nodules. The previously described mudstone represents a floodplain paleosol in which the uppermost horizon (epipedon) has been partially removed by erosion. The shallow soil was subject to pronounced biotic (rooting and burrowing) and vertic activity, the latter facilitated by the smectitic nature of the soil material. A distinct Bk horizon forms the uppermost preserved horizon, containing discrete and coalesced nodular masses of micritic mudstone. The vertical orientation of some nodular masses (rhizocretions) indicates a profound influence of plant roots in the formation of at least some nodules. Vertic fractures, which can form rapidly (Birkeland, 1984), comprise arcuate surfaces with downward-shallowing dips; these served as pathways for the downward translocation of calcium carbonate, as evidenced by the formation of laterally continuous horizons of coalesced nodules along these surfaces. Although this paleosol profile is truncated, we designate it a calcic Alfisol. We note, however, that the morphology of the calcrete displays significantly greater maturity than is generally present in the Blue Mesa paleosols. This disparity suggests that conditions for carbonate accumulation in the Bk horizon were enhanced during this depositional interval.
Calcareous paleosols of the Upper Triassic Chinle Group
63
Figure 6. Representative measured section of the truncated paleosol in the Jim Camp Wash Bed described in the text and shown in Figure 5A. The section is located in southern Petrified Forest National Park along the park road (near the Flattops).
Petrified Forest Formation: Painted Desert Member Lithostratigraphy The Painted Desert Member, of early to middle Norian age, overlies the Sonsela–Moss Back strata and consists of grayishred and reddish-brown mudstones and thin interbedded sandstones (Lucas et al., 1997). As mapped by Stewart et al. (1972), thickness of the entire Petrified Forest Formation ranges from just over 30 m at its northeastern limit in eastern Utah, increasing southward to over 400 m at its southeastern extent in northwestern New Mexico. Like the Blue Mesa Member, the Painted Desert Member consists of thick mudstone intervals incised by channels with a fine-grained fill, locally displaying prominent inclined heterolithic strata (Fig. 7A) and levee complexes. Alternating with these dominantly suspended-load channel fills are sandstones that are predominantly multistoried and characteristically display tabular to lenticular sets of trough and planar cross-beds (Espegren, 1985). At Petrified Forest National Park, Heckert and Lucas (2002a) recognized correlatable sandstone units within the Painted Desert Member. These are, in ascending order, the Flattops, Lithodendron Wash, and Black Forest Beds. The last of these has a considerable volcaniclastic content. Riggs et al. (1994) reported a U-Pb age of 207 ± 2 Ma for zircon from
Figure 7. Features of Painted Desert Member deposition. (A) Channel-fill sequence between the arrows consists of 3 m of (mostly) red mudstone with (minor) interbedded sandstone displaying inclined heterolithic strata. This section is located in the northern Petrified Forest National Park (near Lacey Point). (B) Locally, the Black Forest Bed consists of up to 3 m of intraformational (primarily calcrete nodules) conglomerate. This section is located in the northern Petrified Forest National Park (at Lacey Point). Scale (hiking staff) is 120 cm.
the Black Forest Beds, and Riggs et al. (2003) later obtained a maximum age of 213 ± 1.5 Ma, although the authors conceded a possible age as young as 209 Ma. The Painted Desert Member sandstones generally lack extrabasinal clasts, but locally the Black Forest Bed, and to a lesser extent, the Lithodendron Wash Bed, contain thick lag deposits of calcrete-dominated intraformational conglomerate (Fig. 7B; Heckert and Lucas, 2002a). Pedogenic Features Painted Desert paleosol profiles display A horizons that are thin to absent and B horizons that are typically several meters thick, brick red, and locally display pedogenic slickensides, burrows, drab root traces, and rhizoliths. Calcrete nodules with alpha fabrics are commonly scattered in these thick B horizons (stage
64
Tanner and Lucas
II), but more mature calcretes (stage III and rare stage IV) occur immediately below the high bedload stream deposits. Notably, the Black Forest Bed contains a gravel conglomerate load that is up to 2 m thick in places and consists entirely of calcrete nodules (Fig. 7B). Excellent exposures of Painted Desert paleosols occur in northern Petrified Forest National Park, particularly in exposures along Lithodendron Wash. Below the Lithodendron Wash Bed of Heckert and Lucas (2002a), the reddish-brown mudstone hosts stage II to stage III calcrete horizons and lenticular bodies of conglomerate (Fig. 8A), which consist of centimeter-scale calcrete nodules and mud chips and pedogenic slickensides. Calcrete nodules in stage II horizons are up to 5 cm in diameter; the nodules commonly exhibit a reduced (drab) interior cut by sparry calcite veins, or circumgranular cracking (crystallaria; Fig. 8B),
and some nodules are penetrated by thin (1 mm diameter) burrows. Vertical stacking of nodules (rhizocretions) occurs locally (Fig. 8C). Gray reduction spheres and small drab root traces are also common in the mudstone host. Similar features are present in numerous arroyos that cut Painted Desert strata north of Cameron, Arizona, where rare laminar (stage IV, or K) horizons occur beneath sandstone beds (Fig. 8D). Dubiel and Hasiotis (1994b) and Hasiotis et al. (1998) described Painted Desert paleosols as Vertisols, largely on the basis of abundant pedogenic slickensides and the presence of illuviated clay on ped surfaces. This designation may be appropriate in instances where no other significant pedogenic features occur, but weak horizonation is present in much of the mudstone-dominated section. Bown and Kraus (1987) noted that Alfisols may display profiles in which the A horizon may be thin or absent,
Figure 8. Painted Desert Member pedogenic features. (A) Lens of calcrete nodule and mud-chip conglomerate occurs at the level of the hiking staff handle (staff is 120 cm). The sandstone bed just above the staff handle is the Lithodendron Wash Bed. The mudstone below the calcrete lens has a wedge-shaped ped structure formed by intersecting pedogenic slickensides. (B) Detail of calcrete nodule from lens in (A) illustrating crosscutting crystallaria (sparry calcite veins; arrow). (C) Bk horizon in uppermost Painted Desert strata consists almost entirely of rhizocretions (Rh). The rhizocretion indicated by the arrow is 10 cm long. This section is located is located in the northernmost Petrified Forest National Park (Chinle Mesa). (D) Rare laminar (La; stage IV) calcrete horizon in Painted Desert. Irregular and vertically elongate nodule masses are up to 50 cm long. The base of the Bk horizon is gradational and extends to a depth of 1.5 m below the laminar horizon (hammer is 26 cm long). Location is north of Cameron, Arizona (arroyo near RR 6731).
Calcareous paleosols of the Upper Triassic Chinle Group and the B horizon, which may be thick and brick red in color, also may be calcareous and display vertic features. Therefore, we identify those Painted Desert paleosols that are not dominated by vertic features, and which display weak horizonation, as immature Alfisols. Owl Rock Formation Lithostratigraphy The overlying Owl Rock Formation consists of up to 150 m of interbedded mudstones, sandstones, and limestones of approximately middle Norian age. These strata crop out in northern Arizona, northwestern New Mexico, and southern Utah (Stewart et al., 1972; Lucas and Huber, 1994; Lucas et al., 1997). Dubiel and Good (1991) noted that the contact between the Owl Rock Formation and the underlying Painted Desert Member of the Petrified Forest Formation appears disconformable, and is marked in many places by the presence of a thick intrabasinal conglomerate composed mainly of reworked calcrete clasts and locally abundant unionid bivalves. The upper part of the formation is characterized by distinctive submeter scale beds of ledge-forming limestone. Earlier workers (Blakey and Gubitosa, 1983; Dubiel, 1989, 1993) described these as lacustrine limestones and interpreted them as deposits of a large lacustrine system centered on the Four Corners region. Other workers, however, recognized pervasive pedogenic fabrics in these beds and suggested that they represent mature (stage III and IV) calcretes and palustrine carbonates (Lucas and Anderson, 1993; Lucas et al., 1997; Tanner, 2000). Pedogenic Features Previous examination of the Owl Rock Formation, particularly at the type section near Kayenta, Arizona, revealed distinct differences between the upper and lower strata in the types of pedogenic features present (Tanner, 2000). Thick mudstone beds in the lower part of the formation lack distinctive horizonation but host meter-scale stage II to III calcrete (Bk) horizons that display alpha and beta fabrics. The upper Owl Rock Formation hosts limestones that display brecciated to peloidal fabrics, pisoliths, spar-filled circumgranular cracks, root channels, and rare calcite pseudomorphs after gypsum. These beds are laterally gradational with limestones of limited lateral extent that display rare charophyte debris, oscillation ripple lamination, desiccation polygons, and burrowing. Tanner (2000) interpreted the brecciated beds as palustrine limestones, formed by deposition of carbonates in ponds or wetlands on a sediment-starved floodplain that was subjected to intense pedogenesis during base-level fluctuations (Platt, 1992; Platt and Wright, 1992; Armenteros et al., 1997; Alonso-Zarza, 2003). Chert is locally abundant in the brecciated limestones, but lacks the fabrics associated with Magaditype chert formation and so is interpreted as a secondary replacement feature from groundwater (Schubel and Simonson, 1990; Bustillo, 2001). Examination of the formation at numerous localities (e.g., the type section near Kayenta, in the Echo Cliffs, at Little Painted
65
Desert County Park, and near Lukachukai, Arizona) has yielded additional details on Owl Rock pedogenic features. At the southern end of the Echo Cliffs, the contact with the Petrified Forest Formation is marked by 5 m of plane-bedded intrabasinal conglomerate and sandstone (Fig. 9A). The conglomerate is composed mainly of reworked calcrete nodules, with a lesser contribution of mudstone and chert pebbles. Lower Owl Rock calcretes (stage II to III Bk to Bkm horizons) are up to 5 m thick, with upper and lower gradational contacts in brown mudstone (Fig. 9B), and they display both alpha and beta fabrics (Fig. 9C). Alpha fabric calcretes comprise micritic nodules that have distinct boundaries and are crosscut by sparry veins. These calcretes are stage II to IV, and they exhibit obvious lateral gradations between stages over distances of hundreds of meters. Paleosols with Bk horizons displaying gradational tops probably represent cumulate paleosols in the sense that continual addition of sediment to the top of the profile gradually caused an upward shift in the depth of carbonate accumulation. Lateral gradations between stages of calcrete development are undoubtedly related to position on the floodplain (i.e., channel proximity), as described previously for the Blue Mesa paleosols. The lower Owl Rock paleosols lack the horizonation and obvious evidence of translocated clays that typifies the paleosols in the underlying formations, making their classification by traditional (i.e., Soil Conservation Service, 1999) soil orders problematic. The nomenclature of Mack et al. (1993), however, allows assignment of these paleosols to the order Calcisol. Some mudstones in the upper Owl Rock, however, exhibit pronounced horizonation, displaying ochric epipedons and pale albic horizons overlying reddened Bt/Bk horizons (Fig. 9B). These paleosols are interpreted as calcic Alfisols. Mudstone beds at various levels in the formation are penetrated by sandstone-filled cylinders that are up 60 cm long and up to 30 cm in diameter (Fig. 9D). These have been interpreted previously as decapod burrows (Dubiel, 1993), but the downward-branching shapes of many of these features leaves little doubt that at least some are instead the casts of deep roots, probably the tap roots of monopodial vegetation. Many of the ledge-forming calcareous beds in the upper Owl Rock Formation have abrupt contacts and scoured bases with tens of centimeters of relief (Figs. 10 and 11A). These beds commonly overlie mudstone with a platy to prismatic fabric and millimeter- to centimeter-scale calcrete nodules and rhizocretions. These ledge-forming beds are generally greenish-gray to mottled pink-green (on fresh surfaces), and they contain pisoliths, floating siliciclastic grains, root penetration structures, and locally abundant chert (Figs. 11B and 11C). Many of these ledges form multistoried bodies and contain mud-chip lag deposits that are commonly removed by weathering in outcrop. Notably, these beds generally have a massive fabric and lack the distinctly brecciated texture and extensive root penetration that is typical of the limestones at the type section (Figs. 11C and 11D; Tanner, 2000; Alonso-Zarza, 2003). The features we describe here are consistent with an origin as groundwater calcretes; they represent fluvial channel bodies that were pervasively cemented by calcite in
66
Tanner and Lucas
Figure 9. Pedogenic features of the Owl Rock Formation. (A) In the southern Echo Cliffs, the contact between the Owl Rock Formation and the underlying Petrified Forest Formation is marked by a 5 m bed of conglomerate, composed mainly of calcrete nodules and interbedded sandstone lenses. The base and top of the bed are just below and above the field of view in this photograph. The handle of the staff (scale is 120 cm) rests against a conglomerate layer overlying a sandier lens. (B) Overview of the Owl Rock Formation at the south end of the Echo Cliffs. Three Bk horizons with stage II calcrete are displayed. The lowermost has an abrupt top and gradational base, while the others have gradational tops and bases. Stratigraphically higher (to the left), Alfisol (A) profiles with A, E, and B horizons are visible. (C) Rare calcified root-cell structures are visible in thin sections prepared from stage II calcretes in B. (D) Calcareous sandstone cylinders with a twisting and branching morphology are common in the Owl Rock Formation (location at south end of Echo Cliffs). End of hammer handle is 4 cm wide.
the shallow subsurface and lack many of the features of subaerial exposure and desiccation displayed by palustrine limestones (Wright and Tucker, 1991; Alonso-Zarza, 2003). Rock Point Formation Lithostratigraphy Across the Four Corners area, the upper Norian to (possibly) Rhaetian Rock Point Formation is recognized as the youngest stratigraphic unit of the Chinle Group (Lucas, 1993; Lucas et al., 1997). The contact between the Rock Point Formation and underlying Owl Rock Formation is unconformable (the Tr5 unconformity). The Rock Point Formation, termed the Rock Point Member of the Wingate Sandstone by Stewart et al. (1972), includes strata formerly assigned to the Church Rock Member of
the Chinle Formation (Stewart et al., 1972; Dubiel, 1989; Lucas et al., 1997). Strata of this formation consist of up to 300 m of mainly interbedded brown to red, nonbentonitic mudstones and laminated to rippled sandstones (Stewart et al., 1972; Dubiel, 1989; Lucas et al., 1997). The Rock Point Formation grades vertically to the eolian-dominated Wingate Formation (Fig. 12A) of Rhaetian to Hettangian age (Harshbarger et al., 1957; Tanner et al., 2002; Molina-Garza et al., 2003). Much of the formation consists of sandstone and siltstone sheets that display low-amplitude (eolian) ripple lamination. Other lithofacies present include tabular to sheet sandstones with small-scale sets of high-angle trough cross-beds; erosive-based, wedge-shaped sandstones with planar cross-beds and trough cross-beds and ripple translatent strata; and ripple-laminated to massive mudstones. These lithofacies represent deposition on eolian sand sheets (small-scale dunes
Calcareous paleosols of the Upper Triassic Chinle Group
67
Figure 10. Measured section of paleosol features in upper Owl Rock Formation described from location near Little Painted Desert County Park.
and ripples), on mudflats, in ephemeral lakes, and in ephemeral streams (Stewart et al., 1972; Blakey and Gubitosa, 1984; Dubiel, 1989; Lucas et al., 1997). Pedogenic Features In many locations in the Four Corners, the Rock Point Formation displays abundant and various burrows and root traces, but lacks other well-developed pedogenic features. In northeastern Arizona, for example, near the type section for the formation, the red sheet sandstones and coarse mudstones facies that characterize the formation in this area display bedding-parallel burrows and shallow desiccation cracks in some beds, but lack extensive nodular horizons or vertic features. In other areas of the Colorado Plateau, however, much more extensive pedogenesis is evident. Rock Point calcretes are most mature in upland areas; for example, at Colorado National Monument (near Grand Junction), coarse mudstones and very fine-grained sandstones that are age-equivalent to the Rock Point Formation (Lucas et al., 1997; Tanner, 2003a) rest unconformably on granitic basement and are overlain by sandstones of the Wingate Formation. The strata near the top of this section host multiple pedogenic horizons that display drab root traces, desiccation cracks, and stage II to III calcretes (Tanner, 2003a). Pedogenic features in correlative strata north of Durango, Colorado, include desiccation cracks and drab root traces, both of
which extend tens of centimeters, crumb and blocky mudstone fabrics, rhizocretions, and stage II to III calcretes in which beta fabrics are common (Fig. 12B; Blodgett, 1988; Tanner, 2003a). Blodgett (1988) interpreted the nodule-bearing horizons in the sheet sandstones of the Dolores Formation as calcareous paleosols of the order Aridisol or Inceptisol, lacking epipedons and argillic horizons. These profiles also could be classified as Calcisols (sensu Mack et al., 1993), an interpretation that can be applied to the paleosols at Colorado National Monument as well. Root traces and rhizocretions are evidence that these soils were vegetated by plants with long monopodial root systems. Calcrete conglomerate lenses in Rock Point mudstones (Fig. 12C) provide evidence of local erosional reworking of the depositional surface. PALEOCLIMATE SYNTHESIS Colorado Plateau Pedogenic processes may be controlled to a large extent by climate, but soil development also depends very much on the rate of sediment accumulation, as paleosol maturity is inversely related to sedimentation rate (Bown and Kraus, 1987). Therefore, any interpretation of the paleoclimatic significance of pedogenic features also must examine changes in depositional rate. This,
68
Tanner and Lucas
Figure 11. Features of upper Owl Rock Formation ledge-forming beds. (A) Sandy limestone beds have irregular (erosional) bases, a multistory architecture, and exhibit significant lateral thickness variations. Location is north of Little Painted Desert County Park (north of Winslow, Arizona). (B) The fabric of the beds in A is massive, but with numerous coated grains and pisoliths (Pi) and fine sparry veins that may represent root tubules (Rt). Lens cap for scale is 55 mm. (C, D) Limestone beds in section at southern end of the Echo Cliffs display pronounced brecciation fabrics, root channeling (Rc), and extensive chert replacement. Hammer in C is 26 cm; lens cap in D is 55 mm.
in turn can be forced by such extrinsic factors as tectonics and eustasy, both of which may affect base level (Possamentier et al., 1988; Blum and Price, 1998; Possamentier and Allen, 1999). Initial accumulation of Chinle sediment, during the Carnian stage, was limited to paleovalley systems incised in the Moenkopi (Tr3) surface (Stewart et al., 1972; Blakey and Gubitosa, 1983). Middle Triassic base-level fall and subsequent Late Triassic rise matches the eustatic record of Haq et al. (1987), therefore a eustatic control on alluvial sedimentation is postulated here. The incised paleovalleys and associated tributaries had paleorelief of tens of meters, and so deposition of the Zuni Mountains, Shinarump, and the lowermost strata of the Cameron–Monitor Butte– Bluewater Creek Formations was limited to these topographic lows and was thin to absent between (Stewart et al., 1972; Blakey and Gubitosa, 1983; Demko et al., 1998). Previous workers (Dubiel and Hasiotis, 1994a, 1994b; Hasiotis et al., 1998) have interpreted a humid but seasonal cli-
mate during the late Carnian in the Colorado Plateau region on the basis of the gleyed (or pseudogleyed) and illuviated paleosols in the Zuni Mountains and Shinarump Formations. Demko et al. (1998), however, cautioned that the paleoclimate record of the basal Chinle is biased by deposition within paleovalleys underlain by aquicludes of the Moenkopi Formation, which resulted in artificially high water tables. Indeed, although the prominence of gley features in these paleosols suggests high humidity, the presence of pedogenic slickensides and a prismatic fabric in paleosols in these formations indicates that these soils were allowed to dry completely at times, perhaps seasonally. Numerous authors have commented on the evidence for a strongly seasonal distribution of precipitation during the Late Triassic resulting from a monsoonal effect, both from field studies and from climate models (Robinson, 1973; Parrish and Peterson, 1988; Crowley et al., 1989; Dubiel et al., 1991; Parrish, 1993; Crowley, 1994; Wilson et al., 1994; Pires et al., 2005). This effect presumably was a
Calcareous paleosols of the Upper Triassic Chinle Group
Figure 12. Features of the Rock Point Formation. (A) The section at Little Round Rock, near the type section at Rock Point, Arizona, demonstrates the lithologic transition from interbedded sheet sandstone and mudstone of the Rock Point Formation (RP) to the eolian sandstone–dominated Wingate Formation (Wi). Approximate contact is indicated by the arrow. The entire visible section is ~70 m thick. (B) Rock Point strata near Durango, Colorado, contain stage II calcrete (stage II) consisting primarily of rhizocretions. Lens cap for scale is 55 mm. (C) Calcrete conglomerate lens in mudstone host, in Rock Point strata west of Moab, Utah (San Rafael Swell area).
69
consequence of the configuration of the Pangean land mass and its position straddling the equator. We speculate that Late Triassic migration and rotation of the continent caused weakening of the monsoon in northern Pangea, resulting in regional drying. The prominence of deep tap roots and/or crayfish burrows is also consistent with a strongly fluctuating water table. The presence of spodic horizons and the dominantly kaolinitic composition of the clays, however (Tanner, 2003a; Tabor et al., 2004), which are not present in overlying formations, is a clear indication of humidity and strong weathering and translocation of soil materials. Modern Spodosols are generally (but not exclusively) associated with forested regions, typically coniferous, and humid climates. (Birkeland, 1984; Retallack, 2001). Thus, an overall humid to subhumid climate is likely during Carnian deposition of the Zuni Mountains and Shinarump Formations, with high water tables enforced seasonally by the position of the lower Chinle strata in paleovalleys and the locally impermeable nature of the underlying Moenkopi strata. Initial deposition of the Cameron–Monitor Butte–Bluewater Creek Formations took place under similarly humid but seasonal conditions. The observed gleying, or psuedogleying, and desiccation are both consistent with soil development under conditions of strongly fluctuating water tables, suggesting a greatly variable (possibly seasonal) distribution of precipitation under subhumid climate conditions, as described for the paleosols in the underlying Zuni Mountains and Shinarump Formations. Subsequent accumulation of younger Chinle strata (Cameron–Monitor Butte–Bluewater Creek Formations and the Blue Mesa Member of the Petrified Forest Formation) was widespread and shows no constraint from underlying paleotopography. Paleosols of these strata are primarily vertic Alfisols that are calcic (stage II to III calcretes) in some locations, gleyed, in others, varying by location on the alluvial plain. By analogy to modern soils, this classification implies that the soils formed in woodlands and forests in subhumid to semiarid climates (Birkeland, 1984; Buol et al., 1997; Retallack, 2001). As Bown and Kraus (1987) noted, the presence of gley features does not preclude the formation of calcrete, which can form rapidly in clay-rich paleosols where translocated clays retard the downward movement of meteoric waters. The abundance of pedogenic slickensides and pseudoanticlines in these paleosols further suggests a seasonal, semiarid climate (Therrien and Fastovsky, 2000). The maturity (i.e., horizonation) of the Blue Mesa floodplain paleosols is notable, attesting to a low rate of sediment accumulation. The long residence time of soil materials permitted the effective translocation of oxides and clays within the profiles and the formation of clearly delineated horizons. The general (but not complete) absence of spodic and histic horizons in these strata and the nonkaolinitic composition of the clays signal a decrease in precipitation near the end of the Carnian stage. Regardless of exact cause, Sonsela–Moss Back deposition marks a significant change in the pattern of alluvial sedimentation in the Chinle basin as the slowly aggrading, high-suspendedload, high-sinuosity stream systems were succeeded by mainly
70
Tanner and Lucas
high bedload, low-sinuosity streams. Paleosols in the Jim Camp Wash Bed, however, are mature Alfisols, similar in aspect to those of the Blue Mesa Member, with the exception of greater calcrete maturity (up to stage IV). Therefore, climate was likely more arid (but not greatly so) at the end of the Carnian, during Sonsela deposition, than during Blue Mesa deposition. Subsequent deposition of the Painted Desert Member during the early to middle Norian took place in a flood basin in which the fluvial style varied markedly; deposition by high-suspended load, high-sinuosity streams was punctuated by episodes of deposition by high-bedload, low-sinuosity streams. Base-level changes may be responsible, but the cause of these changes, i.e., eustasy, climate, or tectonism, is unknown. Paleosols that formed on the floodplains of the high-sinuosity channels are generally vertic Alfisols or Vertisols (sensu Mack et al., 1993) that display only limited translocation of clays and horizonation (Dubiel and Hasiotis, 1994b; Hasiotis et al., 1998). These paleosols display less maturity than do the Blue Mesa or Sonsela paleosols, probably reflecting faster rates of sediment accumulation on the floodplain. Painted Desert calcrete horizons, however, are typically more mature (stage II and III, and rare stage IV) than in the Blue Mesa Member, and so likely reflect more arid conditions, as interpreted for the Jim Camp Wash paleosols. Therrien and Fastovsky (2000) noted that gleying is much less common in the upper (Painted Desert Member) than in the lower (Blue Mesa Member) Petrified Forest Formation, and that Bk horizons are much more prominent. Zuber and Parnell (1989) noted that the clay mineral assemblage in the Painted Desert Member is dominated by mixed-layer illite-smectite, in contrast to the predominantly smectitic mudstones of the Blue Mesa Member. They interpreted this composition as the result of pedogenic illitization of smectitic clays in an alkaline environment in which precipitation was highly seasonal. Retallack (2001), however, viewed claims of illitization in the soil-forming environment with skepticism; the significantly less bentonitic composition of the Painted Desert mudstones may be explained instead by interformational differences in the original clay mineralogy of the sediment load. At least locally, initial Owl Rock deposition is marked by the infilling of lows incised into the underlying Painted Desert strata by thick sequences of intrabasinal conglomerate, mainly reworked calcrete. Subsequent Owl Rock depositional settings are composed of low- to high-sinuosity streams and floodplain muds, on which calcic Alfisols and Calcisols (sensu Mack et al., 1993) formed. Alluvial deposition alternated with sedimentation in carbonate ponds and marshes that were modified subsequently by pedogenesis. Tanner (2000) interpreted the Owl Rock sequence as consisting of alternating episodes of floodplain aggradation and degradation caused by changes in base level; incision and pedogenesis of highstand mud and carbonate deposits occurred during episodes of base-level fall that may have been climatically induced, similar to the model of climatically forced sequence boundaries of Tandon and Gibling (1997). Although the concept of Owl Rock deposition in large lakes has been dismissed, episodes of high base level are implied by the presence
of palustrine and minor lacustrine carbonates (Alonso-Zarza, 2003). Palustrine carbonates may form under climates that range from subhumid to semiarid (Platt and Wright, 1992; Tandon and Andrews, 2001), with drier conditions indicated by the presence of pronounced brecciation fabrics and coated grains, as seen in the Owl Rock Formation. The occurrence of well-developed (stage III and IV) calcrete horizons and Alfisols in the intervening mudstones is consistent with this interpretation of semiaridity. A pronounced unconformity (Tr-5) separates the Owl Rock and the overlying Rock Point Formations. As noted by Tanner (2003b), Rock Point deposition marked a change in basin configuration that appears to reflect the rise and migration of a forebulge. Rock Point sediments display only weakly developed Aridisols (or Calcisols sensu Mack et al., 1993), probably reflecting a relatively constant influx of sediment. The abundant evidence of eolian deposition and frequent desiccation, however, indicates that deposition took place in a semiarid to arid climatic setting. Dubiel et al. (1991) interpreted the interval of Rock Point deposition as the driest of the Late Triassic. The abundance of faunal bioturbation, however, indicates episodes of significant surface moisture, potentially a consequence of fluctuating water tables, and, locally, the depositional surface was well-vegetated, as indicated by rhizoliths and beta fabrics. This interpretation is consistent with the dominant sedimentary bedform of eolian sand sheets in the Rock Point Formation; sand sheets are an interdunal facies characteristic of wet eolian systems (Lancaster, 1993). Continued aridification during the Rhaetian and Hettangian is clearly indicated by the dominance of eolian and playa sedimentation during deposition of the Moenave and Wingate Formations, as the Wingate erg formed over the Four Corners area. In sum, evidence from sedimentary facies and paleosols indicates that the climate on the Colorado Plateau was drier during the Norian-Rhaetian than during the Carnian, confirming the interpretation of Blakey and Gubitosa (1984). Dubiel et al. (1991) and Parrish (1993), however, interpreted the same sedimentary evidence as indicating a moist climate until the very end of the Triassic (at least through the Norian). Notably, Parrish (1993) predicted that a strong monsoonal effect would produce abundant moisture in the western equatorial region, which included the Colorado Plateau. Presumably, weakening of the monsoon would have resulted in insufficient strength to draw moisture from the west and aridification of the western equatorial region. Therefore, we must consider the possibility that a weakening monsoon at the start of the Norian caused the observed drying in the Four Corners region. Global Climate Overall warm and dry conditions during the Late Triassic are indicated by the abundance of evaporite and carbonate deposits and the restriction of coal formation to high latitudes (Frakes et al., 1992; Lucas, 1999). Indeed, Colbert (1958) first proposed gradual aridification and associated changes in floral patterns during the Late Triassic to explain tetrapod turnover. The con-
Calcareous paleosols of the Upper Triassic Chinle Group figuration of the Pangean continent undoubtedly had a significant effect in controlling this climate (Robinson, 1973). Specifically, the arrangement of land areas likely resulted in a dry climate belt covering a broad region of western and central Pangea at low to mid-paleolatitudes, a consequence of the shrinkage of the humid intertropical convergence zone (ITCZ) and the weakening of zonal circulation. This interpretation received considerable support from early computer modeling exercises (Parrish et al., 1982; Kutzbach and Gallimore, 1989). A trend of Late Triassic aridification similar to that of the Colorado Plateau is indicated by facies changes, evaporite occurrences, and paleosols in the Upper Triassic to Lower Jurassic formations of the Newark Supergroup (Olsen, 1997; Kent and Olsen, 2000). For example, Norian-age formations contain more mature calcrete paleosols than do Carnian formations in the southern basins, as in the Deep River and Taylorsville basins (Coffey and Textoris, 1996; LeTourneau, 2000; Driese and Mora, 2003). To the north, in the Newark, Hartford, and Fundy basins, the absence of evaporite-bearing or eolian facies in formations of Carnian age and their presence in formations of Norian age demonstrate a similar trend of aridification (see Olsen, 1997, for review). This climate trend in the Newark Supergroup strata, however, has been interpreted as a consequence of the latitudinal drift of eastern North America by 5° to 10°, which carried the basins from a moist subtropical to a more northerly arid climate zone (Olsen, 1997; Kent and Olsen, 2000). Parrish (1993) postulated that aridification on the Colorado Plateau took place during the Early Jurassic as a consequence of the weakening of monsoonal circulation, but we suggest that this aridification took place earlier, during the Norian, as indicated herein. The weakening monsoon, potentially controlled by the northerly drift of Laurasia, resulted in strengthening of zonal circulation and allowed the latitudinal drift of the Colorado Plateau between climate zones (Parrish, 1993). Similar trends of Late Triassic aridification are seen in facies transitions in other locations globally, as in the succession of the Timezgadouine and Bigoudine Formations in the Argana basin, Morocco (Olsen, 1997; Hofmann et al., 2000), the facies changes in the Upper Triassic Mercia Mudstone Group of England (Talbot et al., 1994; Ruffell and Shelton, 1999), and the Keuper of the Germanic basin (Aigner and Bachmann, 1992). Simms et al. (1994) observed a Late Triassic change in clay mineral assemblages in European sedimentary successions, notably a loss of kaolinite, similar to that observed on the Colorado Plateau. These authors attributed the replacement of the pteridosperm flora by a coniferous, gingko, and fern flora at the Carnian-Norian boundary, to climate change. Not all areas of Pangea became drier during the Late Triassic, however. Extensive coal deposits formed in Australia and China, which became wetter at this time (Fawcett et al., 1994). The growth of large lakes in the Jameson Land basin of eastern Greenland during the Late Triassic is interpreted similarly as a consequence of increasing humidity caused by northward drift of the basin to a humid, temperate climate zone (Clemmensen et al., 1998).
71
CONCLUSIONS Paleosols and pedogenic features preserved in the formations of the Chinle Group record a trend of gradual aridification during the Late Triassic. The prominence of gleying in the kaolinitic, bioturbated paleosols of the mottled strata, Shinarump, and basal Cameron formations suggests that climate during the late Carnian was subhumid to humid, and that water tables fluctuated seasonally. High water tables during deposition of these formations may have resulted from their position within paleovalleys incised into the underlying Moenkopi Formation strata. Improved soil drainage during Cameron and Blue Mesa deposition is interpreted from the presence of thick argillic profiles interpreted as Alfisols. This condition may have resulted either from climatic drying or from the position of these soils stratigraphically higher above the Moenkopi Formation. Increasing aridity during early Norian deposition of the Painted Desert Member is clearly suggested by the prominence of vertic features and immature (stage II to III) calcretes. This trend of aridification continued during middle Norian Owl Rock deposition, as indicated by mature (stage III to IV) calcretes. The Norian-Rhaetian Rock Point strata lack mature paleosol profiles, but the predominance of eolian and playa facies in this formation suggests that the trend of increasing aridity continued through onset of the Wingate erg. This trend may have been controlled by the position of the Pangean continent, which led to the weakening of monsoonal flow and the strengthening of zonal circulation. ACKNOWLEDGMENTS We thank the Navajo Nation and U.S. Bureau of Land Management for access to land. Andrew Heckert, Kate Zeigler, Peter Reser, James DeAngelo, and Bryn Welker provided valuable assistance in the field. Additionally, we thank Ana M. AlonsoZarza, Jose Lopez, and S.K. Tandon for their thoughtful reviews and suggestions for improving this manuscript. REFERENCES CITED Aigner, T., and Bachmann, G.H., 1992, Sequence stratigraphic framework of the German Triassic: Sedimentary Geology, v. 80, p. 115–135. Alonso-Zarza, A.M., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: Earth-Science Reviews, v. 60, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Armenteros, I., Daley, B., and García, E., 1997, Lacustrine and palustrine facies in the Bembridge Limestone (late Eocene, Hampshire Basin) of the Isle of Wight, southern England: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 128, p. 111–132, doi: 10.1016/S0031-0182(96)00108-3. Ash, S.R., 1987, The Upper Triassic red bed flora of the Colorado Plateau, western United States: Journal of the Arizona-Nevada Academy of Science, v. 22, p. 95–105. Ash, S.R., 1989, The Upper Triassic Chinle flora of the Zuni Mountains, New Mexico: New Mexico Geological Society Guidebook, v. 40, p. 225–230. Birkeland, P.W., 1984, Soils and Geomorphology: New York, Oxford University Press, 372 p. Blakey, R.C., and Gubitosa, R., 1983, Late Triassic paleogeography and depositional history of the Chinle Formation, southern Utah and northern Arizona, in Reynolds, M.W., and Dolly, E.D., eds., Mesozoic Paleogeography of the West-Central United States: Denver, Rocky Mountain Section
72
Tanner and Lucas
Society for Sedimentary Geology (SEPM) , p. 57–76. Blakey, R.C., and Gubitosa, R., 1984, Controls of sandstone body geometry and architecture in the Chinle Formation (Upper Triassic), Colorado Plateau: Sedimentary Geology, v. 38, p. 51–86, doi: 10.1016/00370738(84)90074-5. Blodgett, R.H., 1988, Calcareous paleosols in the Triassic Dolores Formation, southwestern Colorado, in Reinhardt, J., and Sigleo, W.R., eds., Paleosols and Weathering through Time: Principles and Applications: Geological Society of America Special Paper 216, p. 103–121. Blum, M.D., and Price, D.M., 1998, Quaternary alluvial plain construction in response to glacio-eustatic and climate controls, Texas Gulf Coastal Plain, in Schanley, K.J., and McCabe, P.J., eds., Relative Role of Eustasy, Climate and Tectonism in Continental Rocks: Society for Sedimentary Geology (SEPM) Special Publication 59, p. 31–48. Bown, T.M., and Kraus, M.J., 1987, Integration of channel and floodplain suites, I. Developmental sequence and lateral relations of alluvial paleosols: Journal of Sedimentary Petrology, v. 57, p. 587–601. Buol, S.W., Hole, F.D., McCracken, R.J., and Southard, R.J., 1997, Soil Genesis and Classification: Ames, Iowa State University Press, 527 p. Bustillo, M.A., 2001, Cherts with moganite in continental Mg-clay deposits; an example of ‘false’ Magadi-type cherts, Madrid basin, Spain: Journal of Sedimentary Research, v. 71, p. 436–443. Clemmensen, L.B., Kent, D.V., and Jenkins, F.A., Jr., 1998, A Late Triassic lake system in East Greenland: Facies, depositional cycles and palaeoclimate: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 140, p. 135–159, doi: 10.1016/S0031-0182(98)00043-1. Coffey, B.P., and Textoris, D.A., 1996, Paleosols and paleoclimatic evolution, Durham sub-basin, North Carolina, in LeTourneau, P.M., and Olsen, P.E., eds., Aspects of Triassic-Jurassic Rift Basin Geoscience [abs.]: Hartford, Connecticut, State Geological and Natural History Survey of Connecticut, Miscellaneous Reports 1, p. 6. Colbert, E.H., 1958, Triassic tetrapod extinction at the end of the Triassic Period: Proceedings of the National Academy of Sciences, USA, v. 44, p. 973–977. Crowley, T.J., 1994, Pangean climates, in Klein, G.D., ed., Pangea: Paleoclimate, Tectonics and Sedimentation during Accretion, Zenith and BreakUp of a Supercontinent: Geological Society of America Special Paper 288, p. 25–39. Crowley, T.J., Hyde, W.T., and Short, D.A., 1989, Seasonal cycle variations on the supercontinent of Pangaea: Geology, v. 17, p. 457–460, doi: 10.1130/0091-7613(1989)017<0457:SCVOTS>2.3.CO;2. Deacon, M.W., 1990, Depositional analysis of the Sonsela Sandstone Bed, Chinle Formation, northeast Arizona and northwest New Mexico [M.S. thesis]: Flagstaff, Arizona, Northern Arizona University, 128 p. Demko, T.M., Dubiel, R.F., and Parrish, J.T., 1998, Plant taphonomy in incised valleys: Implications for interpreting paleoclimate from fossil plants: Geology, v. 26, p. 1119–1122, doi: 10.1130/0091-7613(1998)026<1119: PTIIVI>2.3.CO;2. Dickinson, W.R., 1981, Plate tectonic evolution of the southern Cordillera: Arizona Geological Society Digest, v. 14, p. 113–135. Driese, S.G., and Mora, C.I., 2003, Paleopedology and stable-isotope geochemistry of Late Triassic (Carnian-Norian) paleosols, Durham sub-basin, North Carolina, USA; implications for paleoclimate and paleoatmospheric pCO2, in Renault, R.W., and Ashley, G.M., eds., Sedimentation in Continental Rifts: Society for Sedimentary Geology (SEPM) Special Publication 73, p. 207–218. Dubiel, R.F., 1987, Sedimentology of the Upper Triassic Chinle Formation, southeastern Utah [Ph.D. thesis]: Boulder, Colorado, University of Colorado, 146 p. Dubiel, R.F., 1989, Depositional and paleoclimatic setting of the Upper Triassic Chinle Formation, Colorado Plateau, in Lucas, S.G., and Hunt, A.P., eds., Dawn of the Age of the Dinosaurs in the American Southwest: Albuquerque, New Mexico, New Mexico Museum of Natural History and Science, p. 171–187. Dubiel, R.F., 1993, Depositional setting of the Owl Rock Member of the Upper Triassic Chinle Formation, Petrified Forest National Park and vicinity, Arizona, in Lucas, S.G., and Morales, M., eds., The Nonmarine Triassic: Albuquerque, New Mexico, New Mexico Museum of Natural History and Science Bulletin 3, p. 117–121. Dubiel, R.F., 1994, Triassic deposystems, paleogeography, and paleoclimate of Western Interior, in Caputo, M.V., Peterson, J.A., and Franczyk, K.J., eds., Mesozoic Systems of the Rocky Mountain Region, USA: Denver, Colorado, Rocky Mountain Section, Society for Sedimentary Geology
(SEPM), p. 133–168. Dubiel, R.F., and Good, S.C., 1991, Paleovalley fills and lacustrine deposition in the Owl Rock Member of the Chinle Formation (Upper Triassic), Defiance Uplift, Arizona: Geological Society of America Abstracts with Programs, Rocky Mountain Section, v. 23, no. 4, p. 18. Dubiel, R.F., and Hasiotis, S.T., 1994a, Integration of sedimentology, paleosols, and trace fossils for paleohydrologic and paleoclimatic interpretations in a Triassic tropical alluvial system: Geological Society of America Abstracts with Programs, v. 26, no. 7, p. A-502. Dubiel, R.F., and Hasiotis, S.T., 1994b, Paleosols and rhizofacies as indicators of climate change and groundwater fluctuations: The Upper Triassic Chinle Formation: Geological Society of America Abstracts with Programs, v. 26, no. 6, p. 11–12. Dubiel, R.F., Blodgett, R.H., and Bown, T.M., 1987, Lungfish burrows in the Upper Triassic Chinle Formation, southeastern Utah and adjacent areas: Journal of Sedimentary Petrology, v. 57, p. 512–521. Dubiel, R.F., Parrish, J.T., Parrish, J.M., and Good, S.C., 1991, The Pangaean megamonsoon—Evidence from the Upper Triassic Chinle Formation, Colorado Plateau: Palaios, v. 6, p. 347–370. Espegren, W.A., 1985, Sedimentology and petrology of the upper Petrified Forest Member of the Chinle Formation, Petrified Forest National Park and vicinity, Arizona [M.S. thesis]: Flagstaff, Arizona, Northern Arizona University, 228 p. Fawcett, P.J., Barron, E.J., Robinson, V.D., and Katz, B.J., 1994, The climatic evolution of India and Australia from the Late Permian to Mid-Jurassic: A comparison of climate model results with the geologic record, in Klein, G.D., ed., Pangea: Paleoclimate, Tectonics and Sedimentation during Accretion, Zenith and Break-Up of a Supercontinent: Geological Society of America Special Paper 288, p. 139–157. Frakes, L.A., Francis, J.E., and Syktus, J.I., 1992, Climate modes of the Phanerozoic: The history of the Earth’s climate over the past 600 million years: Cambridge, UK, Cambridge University Press, 278 p. Gile, L.H., Peterson, F.F., and Grossman, R.B., 1966, Morphological and genetic sequences of accumulation in desert soils: Soil Science, v. 100, p. 347–360. Gómez-Gras, D., and Alonso-Zarza, A.M., 2003, Reworked calcretes; their significance in the reconstruction of alluvial sequences (Permian and Triassic, Minorca, Balearic Islands, Spain): Sedimentary Geology, v. 158, p. 299–319, doi: 10.1016/S0037-0738(02)00315-9. Haq, B.U., Hardenbol, J., and Vail, P.R., 1987, Chronology of fluctuating sea levels since the Triassic: Science, v. 235, p. 1156–1166. Harshbarger, J.W., Repenning, C.A., and Irwin, J.H., 1957, Stratigraphy of the uppermost Triassic and Jurassic rocks of the Navajo Country: U.S. Geological Survey Professional Paper 291, 74 p. Hasiotis, S.T., and Dubiel, R.F., 1993a, Crayfish burrows and their paleohydrologic significance; Upper Triassic Chinle Formation, Fort Wingate, New Mexico, in Lucas, S.G., and Morales, M., eds., The Nonmarine Triassic: New Mexico Museum of Natural History and Science Bulletin 3, p. G24–G26. Hasiotis, S.T., and Dubiel, R.F., 1993b, Continental trace fossils of the Upper Triassic Chinle Formation, Petrified Forest National Park, Arizona, in Lucas, S.G., and Morales, M., eds., The Nonmarine Triassic: New Mexico Museum of Natural History and Science Bulletin 3, p. 175–178. Hasiotis, S.T., and Dubiel, R.F., 1994, Ichnofossil tiering in Triassic alluvial paleosols; implications for Pangaean continental rocks and paleoclimate, in Embry, A.F., Beauchamp, B., and Glass, D.J., eds., Pangea; Global Environments and Resources: Canadian Society of Petroleum Geologists Memoir 17, p. 311–317. Hasiotis, S.T., Dubiel, R.F., and Demko, T.M., 1998, A holistic approach to reconstructing Triassic paleoecosystems: Using ichnofossil and paleosols as a basic framework: National Park Service Paleontological Research Technical Report NPS, v. NRGRDTR-98, no. 01. p. 122–124. Heckert, A.B., and Lucas, S.G., 1996, Stratigraphic description of the Tr-4 unconformity in west-central New Mexico and eastern Arizona: New Mexico Geology, v. 18, p. 61–70. Heckert, A.B., and Lucas, S.G., 2002a, Revised Upper Triassic stratigraphy of the Petrified Forest National Park, Arizona, U.S.A.: New Mexico Museum of Natural History and Science Bulletin, v. 21, p. 1–36. Heckert, A.B., and Lucas, S.G., 2002b, Lower Chinle Group (Upper Triassic, Carnian) stratigraphy in the Zuni Mountains, west-central New Mexico: New Mexico Museum of Natural History and Science Bulletin, v. 21, p. 51–72. Heckert, A.B., and Lucas, S.G., 2003, Triassic stratigraphy in the Zuni Mountains,
Calcareous paleosols of the Upper Triassic Chinle Group west-central New Mexico: New Mexico Geological Society, Guidebook 54th Field Conference, Geology of the Zuni Mountains, p. 245–262. Hofmann, A., Tourani, A., and Gaupp, R., 2000, Cyclicity of Triassic to Lower Jurassic continental red beds of the Argana Valley, Morocco: Implications for palaeoclimate and basin evolution: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 161, p. 229–266, doi: 10.1016/S00310182(00)00125-5. Kent, D.V., and Olsen, P.E., 1997, Magnetostratigraphy and paleopoles from the Late Triassic Dan River–Danville basin: Interbasin correlation of continental sediments and a test of the tectonic coherence of the Newark rift basins in eastern North America: Geological Society of America Bulletin, v. 109, p. 366–379, doi: 10.1130/0016-7606(1997)109<0366: POUTCS>2.3.CO;2. Kent, D.V., and Olsen, P.E., 2000, Magnetic polarity stratigraphy and paleolatitude of the Triassic-Jurassic Blomidon Formation in the Fundy basin (Canada): Implications for early Mesozoic tropical climate gradients: Earth and Planetary Science Letters, v. 179, p. 311–324, doi: 10.1016/ S0012-821X(00)00117-5. Kraus, M.J., 1999, Paleosols in clastic sedimentary rocks: Their geologic applications: Earth-Science Reviews, v. 47, p. 41–70, doi: 10.1016/S00128252(99)00026-4. Kraus, M.J., and Middleton, L.T., 1987, Dissected paleotopography and baselevel changes in a Triassic fluvial sequence: Geology, v. 15, p. 18–20, doi: 10.1130/0091-7613(1987)15<18:DPABCI>2.0.CO;2. Kutzbach, J.E., and Gallimore, R.G., 1989, Pangaean climates: Megamonsoons of the megacontinent: Journal of Geophysical Research, v. 94, p. 3341–3357. Lancaster, N., 1993, Origins and sedimentary features of supersurfaces in the northwestern Gran Desierto Sand Sea, in Pye, K., and Lancaster, N., eds., Aeolian Sediments—Ancient and Modern: International Association of Sedimentologists Special Publication 16, p. 71–83. Lawton, T.F., 1994, Tectonic setting of Mesozoic sedimentary basins, in Caputo, M.V., Peterson, J.A., and Franczyk, K.J., eds., Mesozoic Systems of the Rocky Mountain Region, USA: Denver, Colorado, Rocky Mountain Section Society for Sedimentary Geology (SEPM), p. 1–25. LeTourneau, P.M., 2000, From coal to caliche: The sedimentary record of Late Triassic paleoclimate from the Taylorsville rift basin, Virginia: Geological Society of America Abstracts with Programs, v. 32, no. 1, p. 30. Lucas, S.G., 1993, The Chinle Group: Revised stratigraphy and chronology of Upper Triassic nonmarine strata in the western United States: Museum of Northern Arizona Bulletin, v. 59, p. 27–50. Lucas, S.G., 1999, The epicontinental Triassic, an overview: Zentralblatt für Geologie und Paläontologie Teil 1, v. 1998, p. 475–496. Lucas, S.G., and Anderson, O.J., 1993, Calcretes of the Upper Triassic Owl Rock Formation, Colorado Plateau, in Lucas, S.G., and Morales, M., eds., The Nonmarine Triassic: New Mexico Museum of Natural History and Science Bulletin 3, p. G32–G41. Lucas, S.G., and Hayden, S.N., 1989, Triassic stratigraphy of west-central New Mexico: New Mexico Geological Society, Guidebook, v. 40, p. 191–211. Lucas, S.G., and Huber, P., 1994, Sequence stratigraphic correlation of Upper Triassic marine and nonmarine strata, western United States and Europe, in Embry, A.F., Beauchamp, B., and Glass, D.J., eds., Pangea: Global Environments and Resources: Canadian Association Petroleum Geologists Memoir 17, p. 241–254. Lucas, S.G., Heckert, A.B., Estep, J.W., and Anderson, O.J., 1997, Stratigraphy of the Upper Triassic Chinle Group, Four Corners region, in Anderson, O.J., Kues, B.S., and Lucas, S.G., Mesozoic Geology and Paleontology of the Four Corners Region: New Mexico Geological Society, Guidebook, v. 48, p. 81–108. Lucas, S.G., Zeigler, K.E., Heckert, A.B., and Hunt, A.P., 2003, Upper Triassic stratigraphy and biostratigraphy, Chama basin, north-central New Mexico: New Mexico Museum of Natural History and Science, Bulletin, v. 24, p. 15–39. Machette, M.N., 1985, Calcic soils of the southwestern United States, in Weide, D.L., ed., Soils and Quaternary Geology of the Southwest United States: Geological Society of America Special Paper 203, p. 1–21. Mack, G.H., and Madoff, R.D., 2005, A test of models of fluvial architecture and paleosol development: Camp Rice Formation (Upper Pliocene–Lower Pleistocene), southern Rio Grande Rift, New Mexico, USA: Sedimentology, v. 52, p. 191–211, doi: 10.1111/j.1365-3091.2004.00687.x. Mack, G.H., James, W.C., and Monger, H.C., 1993, Classification of paleosols: Geological Society of America Bulletin, v. 105, p. 129–136, doi: 10.1130/0016-7606(1993)105<0129:COP>2.3.CO;2. Marzolf, J.E., 1994, Reconstruction of the early Mesozoic Cordilleran cra-
73
tonic margin adjacent to the Colorado Plateau, in Caputo, M.V., Peterson, J.A., and Franczyk, K.J., eds., Mesozoic Systems of the Rocky Mountain Region, USA: Denver, Colorado, Rocky Mountain Section Society for Sedimentary Geology (SEPM), p. 181–215. Molina-Garza, R.S., Geissman, J.W., and Van der Voo, R., 1995, Paleomagnetism of the Dockum Group (Upper Triassic), northwest Texas: Further evidence for the J-1 cusp in the North American apparent polar wander path and implications for the apparent polar wander and Colorado Plateau migration: Tectonics, v. 14, p. 979–993, doi: 10.1029/95TC01456. Molina-Garza, R.S., Geissman, J.W., and Lucas, S.G., 2003, Paleomagnetism and magnetostratigraphy of the lower Glen Canyon and upper Chinle Groups, Triassic-Jurassic of northern Arizona and northeast Utah: Journal of Geophysical Research, v. 108, no. B4, 2181, p. 1–24, doi: 10.1029/2002JB001909. Olsen, P.E., 1997, Stratigraphic record of the early Mesozoic breakup of Pangea in the Laurasia-Gondwana rift system: Annual Review of Earth and Planetary Sciences, v. 25, p. 337–401, doi: 10.1146/annurev.earth.25.1.337. Parrish, J.T., 1993, Climate of the supercontinent Pangea: The Journal of Geology, v. 101, p. 215–253. Parrish, J.T., and Peterson, F., 1988, Wind direction predicted from global circulation models, and wind direction directions determined from eolian sandstones of the western United States—A comparison: Sedimentary Geology, v. 56, p. 261–282, doi: 10.1016/0037-0738(88)90056-5. Parrish, J.T., Ziegler, A.M., and Scotese, C.R., 1982, Rainfall patterns and the distribution of coals and evaporites in the Mesozoic and Cenozoic: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 40, p. 67–101, doi: 10.1016/0031-0182(82)90085-2. Pires, E.F., Sommer, M.G., and Scherer, C.M.S., 2005, Late Triassic climate in southernmost Parana basin (Brazil): Evidence from dendrochronological data: Journal of South American Earth Sciences, v. 18, p. 213–221. Platt, N.H., 1992, Fresh-water carbonates from the Lower Freshwater Molasse (Oligocene, western Switzerland); sedimentology and stable isotopes: Sedimentary Geology, v. 78, p. 81–99, doi: 10.1016/0037-0738(92)90114-7. Platt, N.H., and Keller, B., 1992, Distal alluvial deposits in a foreland basin setting; the Lower Molasse (lower Miocene), Switzerland; sedimentology, architecture and palaeosols: Sedimentology, v. 39, p. 545–565. Platt, N.H., and Wright, V.P., 1992, Palustrine carbonates and the Florida Everglades: Towards an exposure index for the fresh-water environment?: Journal of Sedimentary Petrology, v. 62, p. 1058–1071. Possamentier, H.W., and Allen, G.P., 1999, Siliciclastic sequence stratigraphy; concepts and applications: Society for Sedimentary Geology (SEPM) Concepts in Sedimentology and Paleontology no. 7, 210 p. Possamentier, H.W., Jervey, M.T., and Vail, P.R., 1988, Eustatic controls on clastic deposition, I. conceptual framework, in Wilgus, C.K., Hastings, B.S., Kendall, C.G.S.C., Possamentier, H.W., Ross, C.A., and Van Wagoner, J.C., eds., Sea-Level Research: An Integrated Approach: Society for Sedimentary Geology (SEPM) Special Publication 42, p. 109–142. Retallack, G.J., 2001, Soils of the past: An introduction to paleopedology: Malden, Massachusetts, Blackwell Science, 404 p. Riggs, N.R., Ash, S.R., and Mattinson, J.M., 1994, Isotopic dating of a nonvolcanic continental sequence, Chinle Formation, Arizona: Geological Society of America Abstracts with Programs, Rocky Mountain Section, v. 26, no. 6, p. 61. Riggs, N.R., Ash, S.R., Barth, A.P., Gehrels, G.E., and Wooden, J.L., 2003, Isotopic age of the Black Forest Bed, Petrified Forest Member, Chinle Formation, Arizona: Geological Society of America Bulletin, v. 115, p. 1315–1323, doi: 10.1130/B25254.1. Robeck, R.C., 1956, Temple Mountain member; new member of the Chinle Formation in San Rafael Swell, Utah: American Association of Petroleum Geologists Bulletin, v. 40, p. 2499–2506. Robinson, P.L., 1973, Palaeoclimatology and continental drift, in Tarling, D.H., and Runcorn, S.K., eds., Implications of Continental Drift to the Earth Sciences, Volume 1: London, Academic Press, p. 449–476. Ruffell, A., and Shelton, R., 1999, The control of sedimentary facies by climate during phases of crustal extension: Examples from the Triassic of onshore and offshore England and Northern Ireland: Journal of the Geological Society of London, v. 156, p. 779–789. Schubel, K.A., and Simonson, B.M., 1990, Petrography and diagenesis of cherts from Lake Magadi, Kenya: Journal of Sedimentary Petrology, v. 60, p. 761–776. Scotese, C.R., 1994, Late Triassic paleogeographic map, in Klein, G.D., ed., Pangea paleoclimate, Tectonics, and Sedimentation during Accretion, Zenith, and Breakup of a Supercontinent: Geological Society of America
74
Tanner and Lucas
Special Paper 288, p. 7. Simms, M.J., Ruffell, A.H., and Johnson, A.L.A., 1994, Biotic and climatic changes in the Carnian (Triassic) of Europe and adjacent areas, in Fraser, N.C., and Suess, H.-D., eds., In the Shadow of the Dinosaurs: New York, Cambridge University Press, p. 353–365. Soil Conservation Service, 1999, Keys to Soil Taxonomy: Blacksburg, Virginia, Pocahantas Press, 600 p. Stewart, J.H., Poole, F.G., and Wilson, R.F., 1972, Stratigraphy and origin of the Chinle Formation and related Upper Triassic strata in the Colorado Plateau region: U.S. Geological Survey Professional Paper 690, 336 p. Tabor, N.J., Yapp, C.J., and Montañez, I.P., 2004, Goethite, calcite and organic matter from Permian and Triassic soils; carbon isotopes and CO2 concentrations: Geochimica et Cosmochimica Acta, v. 68, p. 1503–1517, doi: 10.1016/S0016-7037(03)00497-6. Talbot, M.R., Holm, K., and Williams, M.A.J., 1994, Sedimentation in low-gradient desert margin systems: A comparison of the Late Triassic of northwest Somerset (England) and the late Quaternary of east-central Australia, in Rosen, M.R., ed., Paleoclimate and Basin Evolution of Playa Systems: Geological Society of America Special Paper 289, p. 97–117. Tandon, S.K., and Andrews, J.E., 2001, Lithofacies associations and stable isotopes of palustrine and calcrete carbonates; examples from an Indian Maastrichtian regolith: Sedimentology, v. 48, p. 339–355. Tandon, S.K., and Gibling, M.R., 1997, Calcretes at sequence boundaries in Upper Carboniferous cyclothems of the Sydney basin, Atlantic Canada: Sedimentary Geology, v. 112, p. 43–67, doi: 10.1016/S00370738(96)00092-9. Tandon, S.K., and Kumar, S., 1999, Semi-arid/arid zone calcretes: A review, in Singhvi, A.K., and Derbyshire, E., eds., Paleoenvironmental Reconstruction in Arid Lands: Rotterdam, Netherlands, A.A. Balkema, p. 109–152. Tanner, L.H., 2000, Palustrine-lacustrine and alluvial facies of the (Norian) Owl Rock Formation (Chinle Group), Four Corners Region, southwestern U.S.A.: Implications for Late Triassic paleoclimate: Journal of Sedimentary Research, v. 70, p. 1280–1289.
Tanner, L.H., 2003a, Pedogenic features of the Chinle Group, Four Corners region; evidence of Late Triassic aridification: New Mexico Geological Society, Guidebook 54th Field Conference, Geology of the Zuni Mountains, p. 269–280. Tanner, L.H., 2003b, Possible tectonic controls on Late Triassic sedimentation in the Chinle basin, Colorado Plateau: New Mexico Geologic Society, 54th Field Conference Guidebook, Geology of the Zuni Mountains, p. 261–267. Tanner, L.H., Lucas, S.G., Reser, P.K., and Chapman, M.G., 2002, Revision of stratigraphy across the Triassic-Jurassic boundary, Four Corners region, southwestern USA: Geological Society of America Abstracts with Programs, v. 34, no. 6, p. 138. Therrien, F., and Fastovsky, D.E., 2000, Paleoenvironments of early theropods, Chinle Formation (Late Triassic), Petrified Forest National Park, Arizona: Palaios, v. 15, p. 194–211. Wilson, K.M., Pollard, D., Hay, W.W., Thompson, S.L., and Wold, C.N., 1994, General circulation model simulations of Triassic climates; preliminary results, in Klein, G.D., ed., Pangea: Paleoclimate, Tectonics and Sedimentation during Accretion, Zenith and Break-Up of a Supercontinent: Geological Society of America Special Paper 288, p. 91–116. Wright, V.P., and Marriott, S.B., 1996, A quantitative approach to soil occurrence in alluvial deposits and its application to the Old Red Sandstone of Britain: Journal of the Geological Society [London], v. 153, p. 1–7. Wright, V.P., and Tucker, M.E., 1991, Calcretes: An introduction, in Wright, V.P., and Tucker, M.E., eds., Calcretes: New York, Blackwell Scientific, p. 1–22. Zuber, D.J., and Parnell, R.A., 1989, A new paleoclimatic indicator from Triassic paleosols, Chinle Formation, Petrified Forest National Park, Arizona: Geological Society of America Abstracts with Programs, v. 21, no. 5, p. 163. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
Estimates of atmospheric CO2 levels during the mid-Turonian derived from stable isotope composition of paleosol calcite from Israel Amir Sandler† Geological Survey of Israel, 30 Malkhe Yisrael St., Jerusalem 95501, Israel ABSTRACT The carbon and oxygen isotopic composition of pedogenic calcite from midTuronian paleosols in Israel was analyzed to evaluate paleoenvironmental conditions and calculate paleoatmospheric pCO2. The central area of Israel was a part of a marine carbonate shelf that emerged during the mid-Turonian stage, as evidenced by karstic phenomena, fluvial sandstone, and soil profiles. The paleosols have the characteristics of equivalent modern calcic Vertisols but are distinguished by the predominance of palygorskite, which formed as an essential part of the soil processes. The pedogenic calcite and the underlying and overlying marine limestone beds have mean δ13C (‰, Vienna Peedee belemnite [VPDB]) values of −6.15 ± 0.93, −2.82 ± 1.87, −1.33 ± 2.17, respectively, and δ18O values of −5.03 ± 1.24, −6.31 ± 0.87, and −5.81 ± 0.97, respectively. In most sections, the δ18O values of pedogenic calcite are much heavier than those of the limestone due to evaporation. Since most of the pedogenic calcite formed at >50 cm depth and did not show diagenetic modification, the δ13C values were used to calculate pCO2 according to the Cerling model (as applied by Ekart et al., 1999). This marks the first Turonian pCO2 estimate calculated from pedogenic calcite. The calculated range for the mid-Turonian is 1450–2690 ppmv CO2. This high pCO2 level is similar to or somewhat higher than other estimates for the Cretaceous and in accord with calculated high Turonian temperatures from many studies. Keywords: paleosols, carbon isotopes, oxygen isotopes, pedogenic calcite, calcrete, pCO2, Turonian, Israel. RESUMEN La zona central de Israel formó parte de una plataforma marina carbonática que emergió durante el Turoniense medio, tal como indican los fenómenos cársticos, las areniscas fluviales y los perfiles edáficos. Los paleosuelos tienen características similares a vertisuelos cálcicos recientes, y presentan como rasgo distintivo la abundancia de paligorskita, que se formó durante los procesos edáficos. El registro continental indica un aumento progresivo de la aridez. Se analizó la composición isotópica de carbono y oxígeno de las calcitas pedogénicas de estos paleosuelos del Turoniense medio con objeto de evaluar las condiciones ambientales y calcular los paleoniveles de CO2 atmosféricos.
E-mail:
[email protected].
†
Sandler, A., 2006, Estimates of atmospheric CO2 levels during the mid-Turonian derived from stable isotope composition of paleosol calcite from Israel, in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 75–88, doi: 10.1130/2006.2416(05). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
75
76
Sandler Los valores medios de las calcitas pedogénicas y de las calizas marinas infra y suprayacentes son, respectivamente, −6.15 ± 0.93, −2.82 ± 1.87, −1.33 ± 2.17 (‰, VPDB) para el δ 13C, y −5.03 ± 1.24, −6.31 ± 0.87, −5.81 ± 0.97, para el δ18O. En la mayoría de las secciones los valores de δ18O de las calcitas pedogénicas son más pesados que los de las calizas, debido a la evaporación. Puesto que la mayoría de las calcitas pedogénicas se formaron a profundidades superiores a 50 cm y no presentan rasgos de modificación diagénetica, sus valores de δ13C se usaron para calcular la pCO2 de acuerdo con el modelo de Cerling (Ekart et al., 1999). Esta es aparentemente la primera estimación realizada a partir de calcita pedogénica para el Turoniense. Los cálculos realizados asumen una baja tasa de respiración de CO2 en el suelo (Sz = 4000 ppmV), 25 °C como temperatura del suelo y valores de 2‰ para el δ13C de los carbonatos oceánicos superficiales. Los datos obtenidos para el Turoniense medio indican valores de 1450 a 2690 ppmv de pCO2. Estos valores elevados son similares o algo más altos que otras estimaciones realizadas para el Cretácico y son coherentes con las elevadas temperaturas que se han calculado en otros estudios para el Turoniense. Palabras clave: paleosuelos, isótopos de carbono y oxígeno, calcita pedogénica, calcreta, pCO2, Turoniense, Israel.
INTRODUCTION A Cretaceous thermal and CO2 maximum has been conceived both by models (e.g., Barron et al., 1995; Poulsen et al., 2003) and by such proxies as isotopic composition of foraminifera (e.g., Barrera, 1994; Huber et al., 2002) and leaf physiognomy (e.g., Herman and Spicer, 1996; Retallack, 2001). It has been commonly assumed that climate is regulated largely by changes in atmospheric CO2 (Barron et al., 1995; Royer et al., 2004) in addition to other factors such as ocean crust production (Kaiho and Saito, 1994) and ocean circulation (Poulsen et al., 2003). The geochemical model of Berner and Kothavala (2001) predicted maximum Cretaceous CO2 in the Jurassic-Cretaceous transition, whereas Wallmann’s (2001) model predicted a maximum in the Aptian-Albian. Several studies have suggested the Turonian as the time of maximum Cretaceous temperatures (Kolodny and Raab, 1988; Corfield et al., 1991; Barrera, 1994; Frakes, 1999; Wilson et al., 2002), whereas others have suggested low CO2 concentration and temperatures at the Cenomanian-Turonian boundary and for the early Turonian (Arthur et al., 1988; Jenkyns et al., 1994; Kuypers et al., 1999). One method of reconstructing past CO2 concentrations uses the carbon isotope composition of pedogenic calcite (Cerling, 1991). The revised model of Cerling (1999) was utilized by Ekart et al. (1999) to calculate atmospheric CO2 concentrations from paleosols of various ages and to produce a CO2 curve for the Silurian to the present. Their CO2 estimate for the Cretaceous between 120 and 75 Ma was in the range of 1260–2950 ppmv, whereas the estimated value was significantly lower for the latest Cretaceous. Later studies estimated high values of ~1400 ppmv (Ghosh et al., 2001) for the latest Cretaceous. However, a continuous paleosol record for that time showed pronounced short-term variations (Nordt et al., 2003). Recent estimates of Cretaceous CO2 levels include 2300 ppmv from
paleosols in Korea (Hauterivian or Aptian-Albian age; Lee, 1999) and 1700–3200 ppmv from paleosols in Japan (AptianAlbian age; Lee and Hisada, 1999). None of the studies, including the most comprehensive one of Ekart et al. (1999), included data from the Turonian. The current study is an attempt to fill this gap and reconstruct atmospheric CO2 from carbon isotopic compositions of mid-Turonian pedogenic calcite from Israel. This pedogenic calcite formed as glaebules and calcrete in siliciclastic paleosols during the emergence of the central part of the country, which until then had been part of a marine carbonate shelf at low latitude. GEOLOGICAL BACKGROUND The location of the study area (Fig. 1) had been part of a stable sedimentary platform since the Cambrian. During the Early Cretaceous, clastic material, mainly mature quartzose sandstone of the Nubian type, accumulated in both shallow-marine and continental depositional settings. A gradual sea-level rise began in the Albian and led to the onset of carbonate shelf deposition, accompanied by an abundant supply of fine siliciclastics and occasional sands from the Arabo-Nubian massif to the southeast (Garfunkel, 1988; Lewy, 1990). During the Cenomanian an intrashelf basin formed in the southern part of present-day Israel, south of the Ramon area (Fig. 1; Bartov et al., 1972). The drowning of the shelf in the late Cenomanian led to deposition of mainly marls in this basin and to nondeposition over most of the shallow shelf north of the Ramon area. Deposition of limestone resumed in the early Turonian (Buchbinder et al., 2000) but ceased during the mid-Turonian, as a sea-level drop combined with a weak tectonic pulse led to arching and emergence of the central part of the country. The maximum sea-level drop has been estimated by karst configuration to have been no more than ~30 m (Sandler, 1996). Various paleogeographic reconstructions put this part of
Mid-Turonian pedogenic calcite
ME
DIT
ER
AR
EA
RA NE AN SE A
33O
32O
EM
ER
GE
DEAD S
D
EA
JERUSALEM
31O
M NSF TRA D S EA DEA
the country between ~7 to ~15°N during the Turonian (e.g., Barron, 1987; Voigt et al., 1999; Philip et al., 2000). The exposed flat limestone terrain was first affected by vadose karst activity, which created vertical shafts down to a depth of ~20 m. Braided and low-angle meandering rivers deposited sand in channels and sandy clay in overbank environments on top of the karst relief. Soils developed directly on the exposed limestone and on the fluvial sediments (see following). Dark limestone debris (“black pebbles”) was formed early in coastal marshes and was later eroded and redeposited mixed with light-color debris on top of soils, suggesting flash floods due to enhanced climate seasonality. These continental sediments in-between the shelf carbonates were informally termed the “clastic unit” (Sandler and Zilberman, 1985). The emergence interval in central Israel has been estimated as equal to a span of time between two ammonite zones (Sandler, 1996) and less than one ammonite zone (Buchbinder et al., 2000). Assuming a similar duration for the seven ammonite zones (Robaszynski et al., 1990) and a 4.2 m.y. duration for the Turonian stage (Gradstein et al., 1994), the emergence lasted between ~1.2 to <0.6 m.y. Shallow-marine carbonate sedimentation resumed in the early late Turonian and continued to late Coniacian time, when deepening led to basinal chalk sedimentation (Lewy, 1990). Marine sedimentation continued through the late Eocene coincident with pulses of the Syrian Arc folding, which formed a series of northeast-southwest anticlines and synclines. The maximum burial of the mid-Turonian beds in the anticlines is no more than 800 m. A low geothermal gradient, 20–25 °C/km, similar to that prevailing today, is thought to have existed since the Turonian, based on combined fission-track and vitrinite reflection studies (Kohn et al., 1990). PALEOSOLS
OR
N
MO
RA
77
30O
50 km
Figure 1. Location map. Sampling sites (dots) on late Albian to Turonian outcrop map (light-gray pattern) of Israel and the surrounding area. Double dashed line bounds the assumed emerged area during the mid-Turonian; thin dashed line schematically traces the Dead Sea transform.
The mid-Turonian clastic unit in central Israel varies in thickness between ~0.1 and 10 m and consists of shale, sandstone, and occasionally small-pebble conglomerates. The thinnest sequences generally are composed of shale, which penetrates down along dissolution cracks and contains small calcite concretions. These thin sequences have been interpreted as residual soils developed from the insoluble residue of the underlying limestone during intensive karstification with possible later contribution of airborne and fluvial material. When the climate became more arid, the original clays were replaced by palygorskite, and pedogenic glaebules precipitated. The thickest sequences comprise a few meters of shale with large calcite concretions, or an irregular calcite layer, overlain by variegated massive quartzose sandstone, which in turn is locally overlain by another shale layer (Fig. 2A). The sandstone and the shale represent the channel and overbank sediments of the fluvial system. Freshly excavated shale sections clearly exhibit a blocky fabric of equidimensional primary peds (2–4 cm, Fig. 2B), or a columnar fabric, where each column (5 × 5 × 15 cm) consists of several primary peds. The surfaces of both columns and peds are coated by shiny white clay
78
Sandler
A
B Figure 2. (A) Part of a vertical outcrop of a thick type of composite paleosol (Nahal Boqer, see Fig. 6A). 1—solum of lower paleosol; 2—irregular calcrete of lower paleosol; 3—massive sandstone separating lower and upper paleosol; 4—base of upper paleosol. (B) A close look at the lower paleosol below calcrete, which exhibits blocky peds coated by whitish palygorskite argilans. Diameter of black lens cover is 6 cm.
cutan (argilan), which is a clear indicator of pedogenic origin. Slickensides are locally observed on the argilans. The calcite concretions exhibit hierarchical agglomeratic structure suggestive of a pedogenic origin and are considered as glaebules. In many cases, glaebules coalesce to form a continuous rigid layer of mature calcrete of stages IV and V (Sandler, 1996). In all cases, the glaebules are discrete and are embedded in clayey solum devoid of carbonates. Two main clay mineral assemblages occur in the clastic unit. The first consists of illite-smectite (IS) phases, discrete illite and kaolinite, whereas palygorskite is hardly present. The second assemblage consists of palygorskite (70%–95%), remnant detrital clays, and, in some cases, a minor amount of neoformed dioctahedral smectite. The first assemblage is detrital and characterizes, with some variations, the sandstone of the clastic unit and the lower limestone below. The second assemblage characterizes the soils and is authigenic. The clay fraction of the overlying marine beds contains half, or less, palygorskite, which decreases upward until disappearing at the top of the Turonian sequence. Pedogenic microfabrics are observed in thin sections of both shale and calcite glaebules (Sandler, 1996, 1997). The soil plasma consists of shiny clay clusters of high birefringence, reflecting a highly optical orientation (Fig. 3A). Different kinds of bright clay fabrics reflect increasingly extensive areas of oriented clay due to internal stresses in the soil, flushing into cracks and soil voids. Bright clay may form isolated streaks longer than an individual clay particle, which become more extensive along soil development, producing a streaky bright clay fabric (mosepic plasmic fabric). Symmetric linear arrangements occur when void-linings are closed by internal pressures. A network of intersecting bright clay fabric in two preferred orientations (clinobimasepic plasmic fabric) indicates a mature soil (Fig. 3B). The latter and the grain-lining bright clay (skelsepic plasmic fabric) are exclusive to soils. New generations of palygorskite as argilans and crack fills progressively get lighter in color and bigger in cluster size (Fig. 3B). In many cases, solum has developed into a palygorskite crust (palcrete), similar to the well-known occurrences in the Tertiary basins of central Spain (e.g., Rodas et al., 1994). Glaebules and calcrete have typical pedogenic microfabrics of circumgranular cracks, coated grains, and clotted fabric. Calcrete calcite has been replaced, at least in part, by palygorskite plasma, preserving pedogenic characteristics (Fig. 4A–D). In a few locations, silcrete has developed, replacing both calcrete and palygorskite plasma (Sandler, 1996, 1997). Microfabrics of silcrete are presented in Figure 5A–C. Several structures among these three pedogenic lithologies are better explained as related to roots. This data and interpretation suggest that during the course of the continental regime, the climate progressively changed from semihumid karst forming conditions to semiarid conditions of calcic Vertisol formation and later to palygorskitization of the soil clay and silcrete replacement of calcrete, apparently under more arid conditions (Sandler, 1996).
A
B
Figure 3. Examples of the clastic unit paleosol plasma microfabrics. The plasma clay is palygorskite, and the grains are quartz. (A) Hairy cracks (rootlets?) are filled with brighter and larger clay streaks; polarized light. (B) Clinobimasepic, skelsepic, and omnisepic microfabrics, indicating high soil maturity; polarized light.
A
B
C
D
Figure 4. Examples of pedogenic calcite microfabrics: (A) Micritic-microsparitic calcite in a glaebule, which preserves the original polygonal ped structure of the plasma; plain light. (B) Micritic-microsparitic calcite in a glaebule with hierarchical structure. A few structures are radial and apparently related to roots; plain light. (C) Micritic-microsparitic calcite in a glaebule with circumgranular cracks; plain light. (D) Micriticsparitic calcite in a glaebule with pronounced clotted fabric; plain light.
80
Sandler
A
B
C
Figure 5. Examples of silcrete microfabrics: (A) Microquartz in a porous silcrete layer. The elongated pores (dark gray) apparently follow original rootlets; polarized light. (B and C) Microquartz in a massive silcrete layer with clotted fabrics after the calcrete precursor; polarized and plain light, respectively.
MATERIALS AND METHODS
Stable Isotope Analysis
Sampling and Sample Processing
The carbon and oxygen isotope analysis of the calcite samples was performed at the Geological Survey of Israel using a SIRA II mass-spectrometer. Calibration to Vienna Peedee Belemnite (VPDB) was done via the NBS-19 standard following Coplen (1988). Instrument precision was better than ±0.1‰ for both δ13C and δ18O, and the external reproducibility based on duplicate measurements of reference standard was ≤0.05‰. All isotopic results are reported as per mil relative to VPDB.
The Turonian outcrops sampled for the current study are located on the Syrian Arc anticlines in central Israel from the Ramon area in the south to Jerusalem in the north (Fig. 1). Additional samples were taken from a few meters below the surface in a quarry and two shallow drill holes. Eight sections were sampled in detail to include samples of underlying and overlying marine limestones and pedogenic calcite from the calcic horizons across various types of the clastic unit. Additional samples were collected sporadically from other locations. Marine limestone and pedogenic calcite samples were washed well with distilled water, oven-dried at 60 °C, and ground to fine powder. Each sample powder was analyzed by X-ray diffraction for identification of the carbonate minerals. The only carbonate mineral in all samples analyzed for isotopic composition was low-magnesium calcite; minor occurrences of quartz and clays were occasionally detected.
RESULTS Carbon and Oxygen Isotope Profiles along Selected Sections Light δ13C and heavy δ18O of the pedogenic calcite relative to the lower and upper limestone are well recorded in the isotopic curves of the eight sections. Such a signature is expected from subaerial calcite in subtropical climates with
Mid-Turonian pedogenic calcite extensive evaporation (e.g., Allan and Matthews, 1982; Cerling and Quade, 1993). Multiple paleosol sections exhibit a trend of heavier δ18O values of pedogenic calcite in the upper parts of the section. Three selected sections are briefly described here and are graphically presented in Figures 6A and 6B. Nahal Boqer (Fig. 6A) The clastic unit is ~7 m thick and consists of two soil profiles and two fluvial beds. The lower paleosol consists of clayey solum with glaebules and is topped by calcrete (see also Fig. 2). The latter is overlain by lens-shaped fluvial sandstone. The second paleosol consists of shale with numerous calcite glaebules in its upper part. It is overlain by a thin conglomerate of millimeter-scale pebbles cemented by calcite. The conglomerate is overlain by a massive dolomitic marl bed, which superficially appears to be a continuation of the continental beds, but is actually of marine origin. The carbon isotope curve displays the distinctly lighter values typical of calcite of continental rather than of marine origin. The oxygen isotope curve displays the lightest values in the lowermost and uppermost samples of the marine limestone (three laterally adjacent samples were obtained from the lowest bed). The heaviest value is from a glaebule from the top of the second paleosol. The oxygen curve suggests significant evaporation at the end of the continental period. Haluqim Anticline—South (Fig. 6B) The clastic unit here is a paleosol ~4.5 m thick, locally overlain by a lens of black pebble conglomerate. Black pebbles are also dispersed in the lower part of the paleosol, whereas glaebules are dispersed in the middle part. This section may actually consist of two superimposed paleosols. Small black and white pebbles are locally embedded also in the upper limestone. The carbon isotope curve displays the lightest value in the two glaebules analyzed, whereas the oxygen isotope curve displays the lightest values in the underlying and overlying limestone samples. The black pebbles have isotopic values similar to those of the underlying limestone, from which they derived. Sha’ar HaGay (Fig. 6C) The clastic unit here is ~0.2–0.5 m thick and consists of two layers. The lower one is a few centimeters of whiteish shale containing millimeter-size glaebules. The clay penetrates down into the weathered limestone. This paleosol is overlain by a sandy clay layer with weathered limestone debris. This section represents thin paleosols like those in Nahal Zipporim, Jerusalem, and Arad road (data in Table 1). The latter section, however, is more complex and unique because thin calcrete can be recognized within the underlying limestone. The carbon isotope curve displays lightest values in the glaebule and in the limestone debris. The oxygen isotope curve displays the lightest values in the lowest and highest marine sam-
81
ples and the heaviest value in the glaebule. The isotopic curves of these three sections suggest the incorporation of soil solution within the lower limestone. Summary of Carbon and Oxygen Isotope Data from All Locations Other samples of pedogenic calcite, lower and upper limestone, and black pebbles from some other locations were collected and analyzed. A few samples of ostreid shells from two locations north and south of the Ramon anticline were analyzed as well. All data on carbon and oxygen isotopic composition are presented in Table 1 and Figure 7. The ostreids from a bed contemporaneous to the clastic unit are considered as a reference for marine limestone since their shells are made of low-magnesium calcite. Potentially, their living environment was restricted, which would have resulted in heavier-than-marine oxygen values. On the other hand, diagenetic, isotopically light calcite could have been precipitated in shell voids and cracks. The covariance of carbon and oxygen isotopic values of the six ostreid samples does suggest some addition of secondary calcite. Accordingly, the heaviest ostreid values are the best available approximation of mid-Turonian marine calcite isotopic composition, with δ13C values of 2.12 and 2.40 and δ18O values of −0.92 and −1.46. These values are similar to some published data (e.g., Paul et al., 1999; Veizer et al., 1999), although the δ18O values are somewhat heavier than other published data (e.g., Clarke and Jenkyns, 1999; Stoll and Schrag, 2000). Mean and standard deviation (1σ) values for δ13C of the pedogenic calcite, underlying, and overlying limestone are −6.15 ± 0.93, −2.82 ± 1.87, and −1.33 ± 2.17, respectively, and for δ18O, they are −5.03 ± 1.24, −6.31 ± 0.87, and −5.81 ± 0.97, respectively. The significantly light δ13C mean of pedogenic calcite is in accord with the formation from soil solutions. The underlying limestone δ13C mean is somewhat lighter than that of the upper limestone, reflecting the downward migration of soil solutions. The only slightly heavy mean value of δ18O for pedogenic calcite reflects mixtures between highly and slightly evaporated pedogenic samples, and between marine and freshwater calcite in the lower and upper limestone. The lightest mean δ18O value of the underlying limestone reflects: (1) the impact of freshwater during karst development and (2) the removal during emergence of the top of the underlying limestone beds, which were deposited under restricted marine conditions. Application of the Cerling Model Cerling (1991, 1999) developed a diffusion-reaction model expressed by a set of equations for the calculation of atmospheric CO2 concentrations (presented as pCO2 in ppm volume) from the carbon isotopic composition of pedogenic calcite. Ekart et al. (1999) used the modified Cerling model to calculate a Silurian-to-present CO2 curve from a large set of data using the following equation:
A
B
C
Figure 6. Selected stable isotope curves for carbon (circles) and oxygen (squares) across the mid-Turonian clastic unit. Black symbols stand for pedogenic calcite; open symbols stand for nonpedogenic calcite. (A) Nahal Boqer, lithology was modified after Sandler (1996); three subsamples of SA 517 were analyzed. (B) Haluqim anticline–south, lithology was modified after Sandler (1996). Sample SA 455 is of a black pebble from the lens at the top; samples SA 458 and TBF 30 are from lateral variations of the same limestone bed. (C) Sha’ar HaGay.
Mid-Turonian pedogenic calcite
83
TABLE 1. STABLE ISOTOPE CARBON AND OXYGEN DATA (‰) FROM ALL LOCALITIES ARRANGED BY LITHOLOGICAL TYPE 18 13 18 13 Sample Location Type Sample Location Type G C G O G C G O TBF33 Arad road Upper lst. –5.36 –3.51 SA229 Haluqim south Black pebble –6.32 –4.21 SA311 Haluqim south Upper lst. –6.16 –3.84 SA455 Haluqim south Black pebble –5.30 –4.93 SA456 Haluqim south Upper lst. –5.73 –4.86 SA213 Hatira anticline Black pebble –5.49 –3.97 SA458 Haluqim south Upper lst. –6.00 0.78 SA226 Nahal Neqarot Black pebble –3.16 –3.17 TBF30 Haluqim south Upper lst. –4.25 1.36 SA255 Ramla quarry Black pebble –4.54 –3.27 TBF31 Haluqim south Upper lst. –7.74 0.19 SA497 Arad road Lower lst. –5.73 –0.50 SA302* Hatira anticline Upper lst. –5.81 –3.57 SA300 Arad road Lower lst. –8.16 –3.48 SA319 Jerusalem north Upper lst. –6.27 –2.73 SA496 Arad road Lower lst. –7.54 –3.66 TBF 10 Kevuda Hills Upper lst. –3.85 –1.09 TBF32 Arad road Lower lst. –5.73 –4.27 TBF 5 Kevuda Hills Upper lst. –4.96 –1.71 TBF35 Arad road Lower lst. –7.64 –3.33 TBF 6 Kevuda Hills Upper lst. –4.75 –0.60 SA309 Haluqim south Lower lst. –7.93 –3.99 SA529 Nahal Boqer Upper lst. –6.73 1.38 TBF27 Haluqim south Lower lst. –7.97 –3.58 SA530 Nahal Boqer Upper lst. –7.28 0.12 TBF28 Haluqim south Lower lst. –6.30 –3.38 SA475 Nahal Zipporim Upper lst. –5.92 –2.69 SA209 Hatira anticline Lower lst. –6.99 –4.04 SA476 Nahal Zipporim Upper lst. –5.59 –0.26 SA301 Hatira anticline Lower lst. –7.62 –5.03 SA314 Sha’ar HaGay Upper lst. –4.99 –4.05 P 55 Ira Mts. Lower lst. –6.07 –4.52 SA315 Sha’ar HaGay Upper lst. –6.17 0.44 SA219 Jerusalem north Lower lst. –6.39 –4.06 SA316 Sha’ar HaGay Upper lst. –6.50 1.21 SA320 Jerusalem north Lower lst. –7.10 –1.57 SA317 Sha’ar HaGay Upper lst. –5.43 1.15 SA321* Jerusalem north Lower lst. –6.96 1.91 SA318 Sha’ar HaGay Upper lst. –6.80 –4.22 SA 91 Jerusalem south Lower lst. –4.88 –6.32 SA235 Arad road Pedogenic –3.08 –5.91 TBF 9 Kevuda Hills Lower lst. –5.42 –4.46 SA495 Arad road Pedogenic –4.90 –6.36 TBF 1 Kevuda Hills Lower lst. –5.46 –4.25 SA310a Haluqim south Pedogenic –5.55 –5.71 TBF 2 Kevuda Hills Lower lst. –5.44 –3.90 SA310b Haluqim south Pedogenic –5.31 –5.48 SA517a Nahal Boqer Lower lst. –6.72 –0.77 P 15 Haluqim center Pedogenic –4.38 –6.08 SA517b Nahal Boqer Lower lst. –6.45 –0.74 SA210 Hatira anticline Pedogenic –8.70 –7.31 SA517m Nahal Boqer Lower lst. –6.23 –0.46 SA212 Hatira anticline Pedogenic –5.58 –6.41 SA518 Nahal Boqer Lower lst. –5.68 –2.29 SA217 Hatira anticline Pedogenic –7.08 –7.07 SA519 Nahal Boqer Lower lst. –5.46 –2.43 SA489 Hatira anticline Pedogenic –6.30 –6.90 NB4/8 Nahal Boqer DH Lower lst. –5.65 –5.92 SA490 Hatira anticline silc. Pedogenic –4.69 –5.25 SA469 Nahal Zipporim Lower lst. –7.06 –2.29 SA492 Hatira anticline silc. Pedogenic –3.55 –4.87 SA470 Nahal Zipporim Lower lst. –6.01 –0.25 SA222 Jerusalem north Pedogenic –5.27 –7.71 SA471 Nahal Zipporim Lower lst. –4.64 –2.16 SA124a Kevuda Hills Pedogenic –5.28 –6.12 SA312 Sha’ar HaGay Lower lst. –5.88 –0.65 SA124b Kevuda Hills Pedogenic –5.45 –6.29 SA312a Sha’ar HaGay Lower lst. –5.17 –0.66 SA522 Nahal Boqer Pedogenic –5.07 –6.28 SA312b Sha’ar HaGay Lower lst. –4.57 –2.85 SA525 Nahal Boqer Pedogenic –5.18 –3.61 SA 8 Zavo’a anticline Lower lst. –6.62 –3.58 SA526 Nahal Boqer Pedogenic –3.16 –4.67 SA189a Ramon north Oyster –2.98 0.73 NB2/3 Nahal Boqer DH Pedogenic –3.83 –6.58 SA189b* Ramon north Oyster –0.92 2.12 NB4/7a Nahal Boqer DH Pedogenic –5.28 –6.10 SA502a Ramon south Oyster –2.68 1.40 NB4/7b Nahal Boqer DH Pedogenic –4.21 –7.27 SA502b Ramon south Oyster –2.34 1.76 SA474 Nahal Zipporim Pedogenic –4.19 –5.92 SA502c Ramon south Oyster –1.46 2.40 SA254* Ramla quarry Pedogenic –6.19 –7.53 SA502m Ramon south Oyster –2.41 1.65 SA313a Sha’ar HaGay Pedogenic –3.62 –5.75 SA313 Sha’ar HaGay Pedogenic –4.73 –5.91 SA46 Zavo’a anticline Pedogenic –5.08 –6.56 Note: Abbreviations: DH—drill hole; silc.—silcrete outcrop; lst—limestone. *Mean of two independent determinations.
Ca = S ( Z )
δ 13Cs − 1.0044δ 13Cφ − 4.4 , δ 13Ca − δ 13Cs
(1)
where Ca is atmospheric CO2, S(Z) is CO2 (ppmv) contributed by soil respiration, and δ13Cs, δ13CΦ, and δ13Ca are the isotopic compositions of soil CO2, soil respired CO2, and atmospheric CO2, respectively.
The values chosen for substitution in the equation are as follows: S(Z) is a function of vegetation intensity and soil depth but approaches a constant value below ~20–30 cm depth. Well-drained and aerated soils have values between 4000 and 7000 ppmv. Ekart et al. (1999) chose 5000 ppmv for all cases calculated. For the current calculation, the value of 4000 was chosen, since the paleosols studied developed in a semiarid to arid climate.
84
Sandler
Figure 7. Stable isotope plot of carbon versus oxygen of all data; lst.—limestone.
δ13Cs is calculated from the isotopic composition of pedogenic calcite using a fractionation factor (from Romanek et al., 1992), which is temperature-dependent and is −8.98 for 25 °C, the temperature used here and by Ekart et al. (1999) for all cases calculated. δ13CΦ is assumed to be equal to the composition of bulk organic matter. The latter was not directly measured, but was calculated as 26‰ from the value of mean surface ocean carbonates, whereas the atmospheric composition was calculated as 8‰ (see explanation in Ekart et al., 1999). The δ13C curve of Veizer et al. (1999) suggests ~1.5‰ for Turonian ocean carbonates, but since the ostreid composition (see above), as well as mid-Turonian data from several studies (e.g., Stoll and Schrag, 2000; Voigt, 2000) are higher, the value 2‰ was chosen. Accordingly, the composition of atmospheric CO2 was −6‰ and that of the organic matter was 24‰, a value which is typical for C3-dominated modern soils and for paleosols (Cerling et al., 1989). The application of this model requires the following conditions to be met: (1) The Cretaceous vegetation must be of the C3 type (e.g., Robinson et al., 2002). (2) The pedogenic calcite must not have been subjected to postpedogenic diagenesis, neither early nor late. (Any diagenetic process would have homogenized the isotopic composition of a sequence and would not leave such a pronounced pedogenic signal.) (3) The pedogenic calcite should be sampled at least 0.3 m below the Bk horizon. This condition was applicable to the thick paleosols with large glaebules or calcrete but not to the thin residual soils with the small glaebules.
However, it was assumed that these soils started to form at the beginning of the continental regime when the climate was less arid and calcite precipitation was minor (small glaebules) and hence were formed at least at 0.5 m depth. Later, the soils were subjected to erosion and more arid conditions, which caused palygorskitization. Since the carbon isotope composition of the small glaebules was similar to that of the large ones, they were accordingly accounted for in the calculation. (4) All pedogenic calcite, except the single unique sample from the Arad road section, was formed within a siliciclastic solum devoid of any other carbonate. Pedogenic calcite of Vertisols is not recommended for calculating pCO2, since well-developed Vertisols are highly dynamic and may mobilize atmospheric CO2 downward into the glaebules (Ekart et al., 1999). However, Vertisols have been used for such calculations by Ekart et al. (1999) and by others (e.g., Ghosh et al., 2001; Robinson et al., 2002), since it is often the only available paleosol with pedogenic carbonates The paleosols studied here did not show the typical pseudo-anticline structure of well-developed Vertisols, and, hence, internal movement was probably not severe. This is evidenced as well by the horizontal position of the large multigeneration glaebules. The pCO2 was calculated from pedogenic δ13C calcite values at plus and minus one standard deviation from the mean, which were −5.22‰ and −7.08‰, respectively. Substituting these values into Equation 1 results in a pCO2 range of 1450– 2686 ppmv, which is 5.2–9.6 times the present (pre-industrial) value of ~280 ppmv.
Mid-Turonian pedogenic calcite DISCUSSION Carbon isotopic composition of pedogenic carbonates is regarded as the best proxy-based method for estimating pre-Tertiary pCO2 (Royer et al., 2001), although the values are often higher than those obtained by geochemical models. The range of the carbon isotopic composition in the current study is quite wide and expands 3.5‰. Such wide ranges have been recorded in previous studies; standard deviation values, cited by Ekart et al. (1999), are often greater than 1.5 1σ. In the current study, the wide range could have resulted from changing conditions within the two soil types and from the impact of the gradually changing climate. However, the thin residual soils, for which the depth of glaebule formation might seem too shallow for this application, have compositions similar to the average of all samples and very similar to those of overlying soils, when present. On the other hand, diffusional mixing of atmospheric CO2 in deserts
85
can produce up to 2‰ enrichment in the δ13C of soil carbonate at 50 cm depth (Cerling et al., 1989). Thus, Royer et al. (2001) suggested that when S(Z) is lower than 3000, there is a significant contribution of atmospheric CO2, and the composition of pedogenic calcite should be avoided for pCO2 calculations. Due to the progressive aridization, indicated by geological and pedological evidence, the soil carbonates of heaviest values could have been formed under arid conditions and low respiration rates. The calculated range of mid-Turonian pCO2 values of 1450–2686 ppmv is similar to the range extrapolated for the Turonian on the curve of Ekart et al. (1999), and to that on the curve of Retallack (2001), which was based on fossil plant cuticles. Nevertheless, even if just the lightest values are used for calculating pCO2, the resulting concentrations are still ~4–5 times the present value. The latter range is still much higher than the values predicted by the geochemical model of Wallmann (2001), but is within the range of Berner and Kothavala’s (2001) model (Fig. 8). A similar match
Figure 8. Values of Cretaceous atmospheric pCO2 (R = times the pre-industrial value) comparing estimations based on pedogenic carbonate, plant cuticles, and geochemical models. The range of the current study is between plus and minus 1σ deviation of the mean value calculated from mid-Turonian paleosol calcite according to Ekart et al. (1999).
86
Sandler
to geochemical models was recently presented in a study that calculated low CO2 values in Barremian (Lower Cretaceous) paleosol calcite from England (Robinson et al., 2002). It should be noted that geochemical models average increments of a few million years each and that paleosol data is actually absent for the Cenomanian to the Coniacian age interval. Thus, the mid-Turonian high pCO2 values could be either a short-term peak, as demonstrated by Nordt et al. (2003) for the latest Cretaceous (Fig. 8), or a part of a continuous MidCretaceous maximum. Because the positive relation between CO2 levels and temperature has been recently questioned (Veizer et al., 2000; Boucot and Gray, 2001; Shaviv and Veizer, 2003; and see discussion in Royer et al., 2004), the mid-Turonian high CO2 levels were compared with available data on simultaneous temperatures. Direct calculation of mean annual temperature from the oxygen isotopic composition of the pedogenic calcite was avoided since the equations established are valid for middle and high latitudes, but not for low latitudes, as in the current case (Fricke and O’Neil, 1999; Nordt et al., 2003). Therefore, published sea-surface temperatures (SST) were compiled for comparison, because it is possible that they have similar values to mean annual temperatures that prevailed over adjacent continental regions (Russell and Paesler, 2003). One of the first seawater temperature curves for the Cretaceous and the Tertiary was constructed by Kolodny and Raab (1988), based on oxygen isotopic composition of fish phosphate from Israel and its surroundings. That curve recorded a maximum temperature of 33 °C in the early Turonian. Similar results for the Turonian were obtained later by various methods, mainly from oxygen stable isotopes of foraminifera. Huber et al. (1995) estimated a SST of ~33 °C at southern high latitudes, and Clarke and Jenkyns (1999) estimated a maximum warming during the mid-Turonian with low-latitude SST in excess of 33 °C. Huber et al. (2002) suggested that the hottest Cretaceous climate occurred during the latest Cenomanian through early Campanian when middle bathyal ocean water temperatures reached 20 °C. Norris et al. (2002) estimated subtropical temperatures to be 33–34 °C during the latest Cenomanian, and Wilson et al. (2002) estimated SSTs of 32–36 °C for the Turonian tropics. The same SST range was estimated by Schouten et al. (2003) using archaeal membrane lipids for the late Cenomanian–early Turonian. A synthesis of various climatic models for the Mid-Cretaceous continental Saharan region, which is relatively close to the study area, suggested temperatures may have exceeded 32 °C (Russell and Paesler, 2003). Accordingly, it can be concluded that the midTuronian high pCO2 values calculated here from paleosol calcite correspond with the high temperatures that prevailed at that time, as independently estimated from other proxies. ACKNOWLEDGMENTS Sincere thanks are due to Avner Ayalon and Bettina Schilman for carrying out the isotope analyses. Batsheva Cohen and Chana Netzer-Cohen prepared the figures. Prosenjit Ghosh
and an anonymous reviewer are thanked for their constructive comments. Lawrence Tanner and Ana Maria Alonso Zarza improved the final version. REFERENCES CITED Allan, J.R., and Matthews, R.K., 1982, Isotope signatures associated with early meteoric diagenesis: Sedimentology, v. 29, p. 797–817. Arthur, M.A., Dean, W.E., and Pratt, L.M., 1988, Geochemical and climatic effects of increased marine organic carbon burial at the Cenomanian/ Turonian boundary: Nature, v. 335, p. 714–717, doi: 10.1038/335714a0. Barrera, E., 1994, Global environmental changes preceding the Cretaceous-Tertiary boundary: Early-late Maastrichtian transition: Geology, v. 22, p. 877– 880, doi: 10.1130/0091-7613(1994)022<0877:GECPTC>2.3.CO;2. Barron, E.J., 1987, Cretaceous plate tectonic reconstructions: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 59, p. 3–29, doi: 10.1016/00310182(87)90071-X. Barron, E.J., Fawcett, P.J., Peterson, P.H., Pollard, D., and Thompson, S.L., 1995, A “simulation” of mid-Cretaceous climate: Paleoceanography, v. 10, p. 953–962, doi: 10.1029/95PA01624. Bartov, J., Eyal, Y., Garfunkel, Z., and Steinitz, G., 1972, Late Cretaceous and Tertiary stratigraphy and paleogeography of southern Israel: Israel Journal of Earth Sciences, v. 21, p. 69–97. Berner, R.A., and Kothavala, Z., 2001, Geocarb III: A revised model of atmospheric CO2 over Phanerozoic time: American Journal of Science, v. 301, p. 182–204. Boucot, A.J., and Gray, J., 2001, A critique of Phanerozoic climatic models involving changes in the CO2 content of the atmosphere: Earth-Science Reviews, v. 56, p. 1–159, doi: 10.1016/S0012-8252(01)00066-6. Buchbinder, B., Benjamini, C., and Lipson-Benita, S., 2000, Sequence development of Late Cenomanian-Turonian carbonate ramps, platforms and basins in Israel: Cretaceous Research, v. 21, p. 813–843, doi: 10.1006/ cres.2000.0228. Cerling, T.E., 1991, Carbon dioxide in the atmosphere: Evidence from Cenozoic and Mesozoic paleosols: American Journal of Science, v. 291, p. 377–400. Cerling, T.E., 1999, Stable carbon isotopes in paleosol carbonates, in Thiry, M., and Simm-Coincon, R., eds., Paleoweathering, Paleosurfaces, and Related Continental Deposits: Special Publication of the International Association of Sedimentologists 27, p. 43–60. Cerling, T.E., and Quade, J., 1993, Stable carbon and oxygen isotopes in soil carbonates, in Swart, P.K., Lohmann, K.C., et al., eds., Climate Change in Continental Isotopic Records: American Geophysical Union Geophysical Monograph 78, p. 217–231. Cerling, T.E., Quade, J., Wang, Y., and Bowman, J.R., 1989, Carbon and oxygen isotopes in soils and paleosols as ecology and paleoecology indicators: Nature, v. 341, p. 138–139, doi: 10.1038/341138a0. Clarke, L.J., and Jenkyns, H.C., 1999, New oxygen isotope evidence for long-term Cretaceous climatic changes in the Southern Hemisphere: Geology, v. 27, p. 699–702, doi: 10.1130/0091-7613(1999)027<0699: NOIEFL>2.3.CO;2. Coplen, T.B., 1988, Normalization of oxygen and hydrogen isotope data: Chemical Geology, v. 72, p. 293–297. Corfield, R.M., Cartlidge, J.E., Premoli-Silva, I., and Housley, R.A., 1991, Oxygen and carbon isotope stratigraphy of the Palaeocene and Cretaceous limestones of the Bottaccione Gorge and the Contessa Highway sections, Umbria, Italy: Terra Nova, v. 3, p. 414–422. Ekart, D.D., Cerling, T.E., Montañez, I.P., and Tabor, N.J., 1999, A 400 million year carbon isotope record of pedogenic carbonate: Implications for paleoatmospheric carbon dioxide: American Journal of Science, v. 299, p. 805–827. Frakes, L.A., 1999, Estimating the global thermal state from Cretaceous sea surface and continental temperature data, in Barrera, A., and Johnson, C.C., eds., Evolution of the Cretaceous Ocean-Climate System: Geological Society of America Special Paper 332, p. 49–57. Fricke, H.C., and O’Neil, J.R., 1999, The correlation between 18O/16O ratios of meteoric water and surface temperature: Its use in investigating terrestrial climate change over geologic time: Earth and Planetary Science Letters, v. 170, p. 181–196, doi: 10.1016/S0012-821X(99)00105-3. Garfunkel, Z., 1988, The pre-Quaternary geology of Israel, in Yom-Tov, Y., and
Mid-Turonian pedogenic calcite Tchernov, E., eds., The Zoogeography of Israel: Dordrecht, Dr. W. Junk Publishers, p. 7–33. 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: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 170, p. 219–236. Gradstein, F.M., Agterberg, F.P., Ogg, J.G., Hardenbol, V.P., Thierry, J., and Huang, Z.A., 1994, Mesozoic time scale: Journal of Geophysical Research, ser. B, Solid Earth and Planets, v. 99, p. 24,051–24,074, doi: 10.1029/94JB01889. Herman, A.B., and Spicer, R.A., 1996, Palaeobotanical evidence for a warm Cretaceous Arctic Ocean: Nature, v. 380, p. 330–333, doi: 10.1038/380330a0. Huber, B.T., Hödell, D.A., and Hamilton, C.P., 1995, Middle-Late Cretaceous climate of the southern high latitudes: Stable isotopic evidence for mineral equator-to-pole thermal gradients: Geological Society of America Bulletin, v. 107, p. 1164–1191, doi: 10.1130/0016-7606(1995)107<1164: MLCCOT>2.3.CO;2. Huber, B.T., Norris, R.D., and MacLeod, K.G., 2002, Deep-sea paleotemperature record of extreme warmth during the Cretaceous: Geology, v. 30, p. 123– 126, doi: 10.1130/0091-7613(2002)030<0123:DSPROE>2.0.CO;2. Jenkyns, E.C., Gale, A.S., and Coefield, R.M., 1994, Carbon- and oxygen-isotope stratigraphy of the English Chalk and Italian Scaglia and its paleoclimatic significance: Geological Magazine, v. 131, p. 1–44. Kaiho, K., and Saito, S., 1994, Oceanic crust production and climate during the last 100 Myr: Terra Nova, v. 6, p. 376–384. Kohn, B.P., Feinstein, S., and Eyal, M., 1990, Cretaceous to present paleothermal gradients, central Negev, Israel: Constraints from fission tracks dating: Nuclear Tracks and Radiation Measurements, v. 17, p. 381–388, doi: 10.1016/1359-0189(90)90061-2. Kolodny, Y., and Raab, M., 1988, Oxygen isotopes in phosphatic fish remains from Israel: Paleothermometry of tropical Cretaceous and Tertiary shelf waters: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 64, p. 59–67, doi: 10.1016/0031-0182(88)90142-3. Kuypers, M.M.M., Pankost, R.D., and Sinninghe Damsté, J.S., 1999, A large and abrupt fall in atmospheric CO2 concentration during Cretaceous times: Nature, v. 399, p. 342–345, doi: 10.1038/20659. Lee, I.Y., 1999, Stable isotopic composition of calcic paleosols of the Early Cretaceous Hasandog Formation, southern Korea: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 150, p. 123–133, doi: 10.1016/S00310182(99)00010-3. Lee, I.Y., and Hisada, K., 1999, Stable isotopic composition of pedogenic carbonates of the Early Cretaceous Shimonoseki Subgroup, western Honshu, Japan: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 153, p. 127–138, doi: 10.1016/S0031-0182(99)00069-3. Lewy, Z., 1990, Transgressions, regressions and relative sea level changes on the Cretaceous shelf of Israel and adjacent countries: A critical evaluation of Cretaceous global sea level correlations: Paleoceanography, v. 5, p. 619–637. Nordt, L., Atchley, S., and Dworkin, S., 2003, Terrestrial evidence for two greenhouse events in the latest Cretaceous: GSA Today, v. 13, no. 12, p. 4–9, doi: 10.1130/1052-5173(2003)013<4:TEFTGE>2.0.CO;2. Norris, R.D., Bice, K.L., Magno, E.A., and Wilson, P.A., 2002, Jiggling the tropical thermostat in the Cretaceous hothouse: Geology, v. 30, p. 299– 302, doi: 10.1130/0091-7613(2002)030<0299:JTTTIT>2.0.CO;2. Paul, C.R.C., Lamolda, M.A., Mitchell, S.F., Vaziri, M.R., Gorostidi, A., and Marshall, J.D., 1999, The Cenomanian-Turonian boundary at Eastbourne (Sussex, UK); a proposed European reference section: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 150, p. 83–121, doi: 10.1016/ S0031-0182(99)00009-7. Philip, J., Floquet, M., Platel, J.P., Bergerat, F., Sandulescu, M., Baraboshkin, E., Amon, E.O., Guiraud, R., Vaslet, D., Le-Nindre, Y.M., Ziegler, M., Poisson, A., Bouaziz., S., Bilotte, M., Canerot, J., Darkal, A.N. Gavrylyshyn V., Georgiev, A., Herkat, M., Hirsh, F., Hippolyte, J.C., Ivanik, M., Kuss, J., Lewy, Z., Meister, C., Mostafa, M., Mouty, M., Nikishin, A.M., Plotnikova, L., Razgallah, S., Rey, J., Romaniv, A., Sadek, A., Tarkowski, R.A., and van-Dycke, S., 2000, Late Cenomanian (94.7–93.5 Ma), in Dercourt, J., Gaetani, M., et al., eds., Peri-Tethys Atlas; Palaeogeographic Maps; Explanatory Notes: Paris, France, Commission for the Geological Map of the World, p. 129–136; 1 sheet. Poulsen, C.J., Gendaszek, A.S., and Jacob, R.L., 2003, Did the rifting of
87
the Atlantic Ocean cause the Cretaceous thermal maximum?: Geology, v. 31, p. 115–117, doi: 10.1130/0091-7613(2003)031<0115: DTROTA>2.0.CO;2. Retallack, G.J., 2001, A 300-million-year record of atmospheric carbon dioxide from fossil plant cuticles: Nature, v. 411, p. 287–290, doi: 10.1038/35077041. Robaszynski, F., Caron, M., Dupuis, C., Amedro, F., Gonzalez-Donoso, J.M., Linares, D., Hardenbol, J., Gartner, S., Calandra, F., and Deloffre, R., 1990, A tentative integrated stratigraphy in the Turonian of Central Tunisia: Formations, zones and sequential stratigraphy in the Kalaat Senan area: Bulletin des Centres de Recherches Exploration-Production Elf Aquitaine, v. 14, p. 214–389. Robinson, S.A., Andrews, J.E., Hesselbo, S.P., Radley, J.D., Dennis, P.F., Harding, I.C., and Allen, P., 2002, Atmospheric pCO2 and depositional environment from stable-isotope geochemistry of calcrete nodules (Barremian, Lower Cretaceous, Wealden Beds, England): Journal of the Geological Society of London, v. 159, p. 215–224. Rodas, M., Luque, F.J., Mas, R., and Garzon, M.G., 1994, Calcretes, palcretes, and silcretes in the Paleogene detrital sediments of the Duero and Tajo basins, central Spain: Clay Minerals, v. 29, p. 273–285. Romanek, C., Grossman, E.L., and Morse, J.H., 1992, Carbon isotope fractionation in synthetic aragonite and calcite: Effects of temperature and precipitation rate: Geochimica et Cosmochimica Acta, v. 56, p. 419–430, doi: 10.1016/0016-7037(92)90142-6. Royer, D.L., Berner, R.A., and Beerling, D.J., 2001, Phanerozoic atmospheric CO2 change: Evaluating geochemical and paleobiological approaches: Earth-Science Reviews, v. 54, p. 349–392, doi: 10.1016/S00128252(00)00042-8. Royer, D.L., Berner, R.A., Montañez, I.P., Tabor, N.J., and Beerling, D., 2004, CO2 as a primary driver of Phanerozoic climate: GSA Today, v. 14, no. 3, p. 4–10, doi: 10.1130/1052-5173(2004)014<4:CAAPDO>2.0.CO;2. Russell, D.A., and Paesler, M.A., 2003, Environments of Mid-Cretaceous Saharan dinosaurs: Cretaceous Research, v. 24, p. 569–588, doi: 10.1016/ S0195-6671(03)00072-7. Sandler, A., 1996, A Turonian subaerial event in Israel: Karst, sandstone and pedogenesis: Israel Geological Survey Bulletin, v. 85, 52 p. Sandler, A., 1997, Pedogenic microfabrics in palcrete, calcrete, and silcrete of the Turonian clastic unit, northern Negev: Geological Society of Israel, Annual Meeting Abstracts: Jerusalem, Geological Society of Israel, p. 98. Sandler, A., and Zilberman, E., 1985, Sandstone and shale in the Nezer Formation, northern Negev: Israel Geological Survey Report GSI/3/85, 59 p. Schouten, S., Hopmans, E.C., Forster, A., van Breugel, Y., Kuypers, M.M.M., and Sinninghe Damsté, J.S., 2003, Extremely high sea-surface temperatures at low latitudes during the middle Cretaceous as revealed by archaeal membrane lipids: Geology, v. 31, p. 1069–1072, doi: 10.1130/G19876.1. Shaviv, N.J., and Veizer, J., 2003, Celestial driver of Phanerozoic climate?: GSA Today, v. 13, no. 7, p. 4–10, doi: 10.1130/1052-5173(2003)013<0004: CDOPC>2.0.CO;2. Stoll, H.M., and Schrag, D.P., 2000, High-resolution stable isotope record from Upper Cretaceous rocks of Italy and Spain: Glacial episodes in greenhouse planet?: Geological Society of America Bulletin, v. 112, p. 308–319, doi: 10.1130/0016-7606(2000)112<0308:HRSIRF>2.3.CO;2. Veizer, J., Ala, D., Azmy, K., Bruckschen, P., Buhl, D., Bruhn, F., Carden, G.A.F., Diener, A., Ebneth, S., Godderis, Y., Jasper, T., Korte, C., Pawellek, F., Podlaha, O.G., and Strauss, H., 1999, 87Sr/86Sr, δ13C and δ18O evolution of Phanerozoic sea water: Chemical Geology, v. 161, p. 59–88, doi: 10.1016/S0009-2541(99)00081-9. Veizer, J., Godderis, Y., and Francois, L.M., 2000, Evidence for decoupling atmospheric CO2 and global climate during the Phanerozoic eon: Nature, v. 408, p. 698–701, doi: 10.1038/35047044. Voigt, S., 2000, Stable oxygen and carbon isotopes from brachiopods of southern England and northwestern Germany: Estimation of Upper Turonian palaeotemperatures: Geological Magazine, v. 137, p. 687–703, doi: 10.1017/S0016756800004696. Voigt, S., Hay, W.W., Höfling, R., and DeConto, R.M., 1999, Biogeographic distribution of late Early to Late Cretaceous rudist-reefs in the Mediterranean as climate indicators, in Barrera E., and Johnson, C.C., eds., Evolution of the Cretaceous Ocean-Climate System: Geological Society of America Special Paper 332, p. 91–103. Wallmann, K., 2001, Controls on the Cretaceous and Cenozoic evolution of
88
Sandler
seawater composition, atmospheric CO2 and climate: Geochimica et Cosmochimica Acta, v. 65, p. 3005–3025. Wilson, P.A., Norris, R.D., and Cooper, M.J., 2002, Testing the Cretaceous greenhouse hypothesis using glassy foraminiferal calcite from the core of
the Turonian tropics on Demerara Rise: Geology, v. 30, p. 607–610, doi: 10.1130/0091-7613(2002)030<0607:TTCGHU>2.0.CO;2. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
Pedogenic carbonate distribution within glacial till in Taylor Valley, Southern Victoria Land, Antarctica Kelly K. Foley W. Berry Lyons Byrd Polar Research Center and the Department of Geological Sciences, The Ohio State University, Columbus, Ohio 43210-1002, USA John E. Barrett Ross A. Virginia Environmental Studies Program, Dartmouth College, Hanover, New Hampshire 03755, USA ABSTRACT Pedogenic carbonate in the form of calcite (CaCO3) accumulates within the soils of hot and cold arid continental landscapes, but much less is known about the spatial pattern and amount of carbonates in polar soils. We measured the CaCO3 distribution in the McMurdo Dry Valleys of Antarctica, the largest ice-free expanse on the continent. Higher CaCO3 concentrations occur in the moist coastal soils, the younger till (younger than 50,000 yr), and in lower elevation tills within Taylor Valley in the McMurdo Dry Valleys area. The average CaCO3 in the moist coastal soils of the McMurdo Dry Valleys is 1.06%, 0.39% in intermediate-elevation soils, and only 0.02% in inland highest-elevation soils. In general, the youngest coastal tills contain the highest amounts of CaCO3 (2.0%), and the oldest tills from Taylor IV glaciation contain the least (0.54%). There is a noticeable difference in CaCO3 concentration near the elevation of the highest stand of ancient Glacial Lake Washburn (~340 m), with higher concentrations found in soils previously covered by the lake. This suggests that a portion of the CaCO3 in soils below this elevation could be lacustrine derived. The Fryxell, Hoare, and Bonney basin soils in Taylor Valley have mean inorganic C concentrations that are much lower than the average world inorganic soil C concentration of 33.2 kg C m–2. The relatively low carbonate concentrations in Antarctic polar desert soils can be attributed to the shallow active layer, low rates of weathering, and the extreme aridity of the landscape. Keywords: arid landscapes, calcite, pedogenic carbonates, McMurdo Dry Valleys, Taylor Valley, Victoria Land, Antarctica. RESUMEN Los carbonatos pedogénicos, sobre todo calcita se acumulan muy frecuentemente en zonas áridas o semiáridas. Los McMurdo Dry Valleys de la Antártida son las mayores zonas sin hielo del continente y se denominan desiertos polares. Hemos medido
Foley, K.K., Lyons, W.B., Barrett, J.E., and Virginia, R.A., 2006, Pedogenic carbonate distribution within glacial till in Taylor Valley, Southern Victoria Land, Antarctica, in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 89–103, doi: 10.1130/2006.2416(06). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
89
90
Foley et al. la distribución de CaCO3 en estos suelos con objeto de intentar comprender mejor la distribución y acumulación del carbonato en estos suelos. Las concentraciones más elevadas de CaCO3 se dan en los suelos costeros húmedos, dentro del till más reciente (<50,000 años), y en los tills de zonas topográficas más bajas en el Valle Taylor, situado dentro de los McMurdo Dry Valleys. Usando la clasificación de Marchant y Denton (1996) de los tres mayores regímenes climáticos del McMurdo Dry Valleys, el contenido medio de CaCO3 en los suelos húmedos costeros (Zona 1) del McMurdo Dry Valleys es 1.06%; en la zona 2 (suelos de altitudes intermedias) es de 0.39%, y en la zona 3 (altitudes mayores del interior) es sólo del 0.02%. En general, los tills costeros más recientes (5480 ± 56 14C) contienen un 2.0% de CaCO3, los tills de la Glaciación Ross I (12.4–23.8 ka) presentan valores de CaCO3 de 0.68%, en los de la Glaciación Taylor II /Bonney (74–98 ka) el contenido es 0.92%, en los de la Glaciación Taylor III (200–210 ka) estos valores son de 0.74% y en los tills más antiguos de la Glaciación Taylor IV (2100–3700 ka) el contenido medio de CaCO3 es de 0.54%. Sin embargo, hay una notable diferencia en la concentración de CaCO3 en el suelo a una altitud de 336 m, la situación más elevada del antiguo Glacial Lake Washburn (22,800–8500 años B.P.), por debajo de esta altitud las concentraciones son mayores. Por ello, una parte del CaCO3 en el suelo por debajo de esta altitud puede ser de origen lacustre. El promedio mundial de carbonato inorgánico en el suelo es de 33.2 Kg C m–2, mientras que los suelos de las cuentas Fryxell, Hoare, y Bonney en el Valle Taylor tienen valores medios de 0.38, 0.31 y 0.68 Kg C m–2, respectivamente. Palabras clave: carbonatos pedogénicos, calcita, ambientes áridos, Valle Taylor, Valles McMurdo Dry, Tierra Victoria, Antártida.
INTRODUCTION Carbonate minerals that have been formed in situ (henceforth pedogenic carbonate) are common in soils where mean annual precipitation is <75 cm and evaporation exceeds precipitation (Cerling, 1984). Information concerning the amount and rate of formation of pedogenic carbonate is important in the overall understanding of carbon fluxes and reservoirs within the terrestrial landscape (Lal et al., 2000; Landi et al., 2003) and in better understanding the relationship between carbonate accumulation and atmospheric CO2 and climate (Marion and Schlesinger, 1994). Although most of the previous research on pedogenic carbonate formation has been undertaken in more temperate environmental settings, it has been recognized that carbonate accumulation also occurs in polar environments (Hallet, 1976; Sletten, 1988; Fairchild and Spiro, 1990; Marion et al., 1991). For example, carbonate minerals, primarily calcite (CaCO3), occur in measurable quantities in the soils of the polar deserts of the Antarctic (Bockheim, 1982; Campbell and Claridge, 1969, 1982; Claridge, 1977; Keys and Williams, 1981; McCraw, 1967). This early work established that carbonate is widespread within the polar deserts of the McMurdo region of Victoria Land (Keys and Williams, 1981), and much of it occurs as encrustations or coatings of the undersides of large particles, such as pebbles and cobbles (Claridge, 1977; Nishiyama and Kurasawa, 1975). Globally, the formation of carbonate minerals in soils can be an important sink of atmospheric carbon (Lal et al., 2000); however, desert soils are thought not to be a significant sink (Capo and
Chadwick, 1999). Pedogenic carbonate in arid to semiarid soils in more temperate climates forms calcic horizons through the dissolution of carbonates in the soil surface and precipitation at depth. Caliche, or calcrete identified as Bk and K soil horizons, forms when the soil carbonate-bicarbonate is in near equilibrium (Lal et al., 2000). The source of Ca2+ in the soil may come from rock weathering or alluvium, or it may be brought in by eolian material or by marine aerosol (Brass, 1975; Capo and Chadwick, 1999; Jones and Faure, 1978; Landi et al., 2003; Lyons et al., 2002; Naiman et al., 2000; Quade et al., 1995). The slow development of the pedogenic salts in the McMurdo Dry Valleys may exist longer in the soil compared to their more temperate counterparts that contain vegetation, making the polar desert landscape an important research site in addressing pedogenic carbonate formation. One of the major geochemical differences between a continental-arid ecosystem (e.g., desert) and a continental-humid ecosystem is the accumulation of water-soluble salts. Although the vast amount of research related to arid soils has focused on warm deserts, soils in Antarctic polar deserts also can contain high salt concentrations (Campbell and Claridge, 1987). The cold desert soils in Antarctica are similar to other desert soils in that they have dry surface horizons that can be capped by a desert pavement of lag, which has a low moisture content and a zone of water-soluble salt accumulation (Campbell and Claridge, 1969, 1982). These salts include a combination of chlorides, nitrates, and sulfates (Claridge and Campbell, 1976; Keys and Williams, 1981). Origins of the ions that make up the salts in Antarctica soils have been suggested to include dry and wet deposition of marine aerosols and in situ
Calcium carbonate in Antarctic soils rock weathering (Bao et al., 2000; Claridge and Campbell, 1976; Keys and Williams, 1981). Although a general description of the relationship between salt distribution and geographic position in the McMurdo Dry Valleys has been developed (Keys and Williams, 1981), little is known about the process of formation. For example, sodium-based salt accumulation varies geographically, where NaCl occurs closer to the coast, NaNO3 is more abundant at the higher elevations closer to the polar plateau, and Na2SO4 is most abundant at intermediate ranges (Bockheim, 1997; Keys and Williams, 1981). Even though the exact process of formation of these sodium salts is unknown, it has been hypothesized that both salt formation and salt accumulation occur more readily from precipitation than from in situ weathering (Claridge and Campbell, 1976). Accumulations of salt can range from <0.1 kg m–2 in young soils (younger than 50,000 yr) to 100 kg m–2 in the older, drier soils (>10 m.y.) (see Claridge and Campbell, 1976; Bockheim, 1997). The formation of carbonate minerals in glacial environments was reviewed in Fairchild and Spiro (1990). Carbonate minerals in polar and/or glacial environments are a common occurrence, as they have been observed in subglacial environments (Hallet, 1976), soils and floodplains (Marion et al., 1991; McCraw, 1967; Sletten, 1988), permafrost (Clark and Lauriol, 1992), spring discharges (Omelon et al., 2001), and in ice (Papadimitriou et al., 2004). The pedogenic carbonates in the McMurdo Dry Valleys are primarily calcite (Keys and Williams, 1981). Although the previous studies cited above have provided important information on the general distribution of CaCO3 within these unusual Antarctic soil environments, there has been little attempt to explain the distribution of pedogenic carbonate in terms of either process, landscape age, or landscape position as has been done for the more soluble salts observed in Antarctic polar desert soils (e.g., Bockheim, 1997; Claridge, 1977). The objectives of this research were to: (1) determine the soil pedogenic inorganic carbonate concentrations; (2) determine the distribution relationship to elevation and distance inland; and (3) determine the distribution relationship to the soil age within the McMurdo Dry Valleys region of Victoria Land, Antarctica (~78°S, ~161°E), especially Taylor Valley. Location of the McMurdo Dry Valleys The McMurdo Dry Valleys consist of a series of ice-free valleys, which make up the largest ice-free desert on the continent. In 1993, a Long-Term Ecological Research (LTER) site was established by the National Science Foundation with Taylor Valley as the major focus region (Fig. 1) (Priscu, 1998). The McMurdo Dry Valleys (MCM) LTER program is focused on understanding the structure and function of the ecosystem and how the ecosystem is influenced by climatic changes. Geology of the McMurdo Dry Valleys Most of the Taylor Valley landscape is largely sandy gravel glacial tills (Claridge and Campbell, 1976), except for bedrock margins along the valley walls. Exposed bedrock and bedrock under the
91
sandy gravel are composed of different rock types, which include Precambrian to Cambrian metasediments such as schists, argillites, quartzites, and marbles; Paleozoic-age intrusive rocks of granites and granodiorites; and Jurassic-age Ferrar Dolerite (Haskell et al., 1965). The outcrops on Taylor Valley’s southern boundary are composed of the Ferrar Dolerite of Jurassic age (Haskell et al., 1965). Also exposed in Taylor Valley are thirty irregularly shaped cones of scoriaceous olivine basalt from Cenozoic McMurdo volcanism, located 550–1220 m above sea level in Taylor Valley (Haskell et al., 1965). Tills that make up Marble Point soils, north of Explorers Cove on the Ross Sea coast (see Fig. 2), are from gneiss, schist, and marble (Campbell et al., 1998). Lyons et al. (2002) demonstrated that Sr in the glacial meltwater streams in part originates from the weathering of many of these rock types, each of which has a rather distinctive 87Sr/86Sr signature. Climate of the McMurdo Dry Valleys The McMurdo Dry Valleys are one of the coldest and driest terrestrial landscapes on the planet, with a mean air temperature of −20 °C and snowfall of <100 mm yr –1 (Doran et al., 2002a). However, each area of the McMurdo Dry Valleys has its own microclimate. For example, Taylor Valley has wetter, colder, and cloudier conditions near the coast and warmer, drier conditions inland (Doran et al., 2002a). This climatic pattern is driven by the easterly winds (summer), which bring marine aerosols into the valley that are deposited near the coast, resulting in drier air containing lower marine aerosol components inland (Doran et al., 2002a; Fountain et al., 1999). Katabatic winds from the west, which are stronger and drier than the easterly winds, flow into the valley primarily during the winter season (Doran et al., 2002b; Fountain et al., 1999). These winds contribute much eolian, or wind-blown dust, from higher elevations to Taylor Valley soils (Doran et al., 2002a; Fountain et al., 1999). Mean annual temperatures range from –16.7 °C at Marble Point to –35 °C in Beacon Valley (Table 1) (Doran et al., 2002a; Sugden et al., 1995). Variation in these microclimates may be one of the most dominant control factors in formation and occurrence of soil CaCO3. Sample Locations Sample locations for this study varied with the microclimate, the landscape formation history, elevation, and the age of the glacial till and/or soil. The majority of samples used in this study were obtained in Taylor Valley and along the Ross Sea coast, including Marble Point (Fig. 1). Data from samples obtained from Wright Valley (77°33′S, 161°45′E), Victoria Valley (77°20′S, 161°45′E), Pearse Valley (77°43′S, 161°30′E), and two smaller, but higher-elevation valleys, Arena Valley (77°50′S, 161°E) and Beacon Valley (77°48′S, 160°40′E), are also presented. Taylor Valley (Fig. 1) is located at 77°30′–77°45′S and 162°00′–164°00′E and extends 34 km inland, terminating at the terminus of the Taylor Glacier, the easternmost extent of the East
92
Foley et al.
Figure 1. Taylor Valley, Antarctica, is 34 km long and 12 km wide.
Antarctica Ice Sheet (Figs. 1 and 2). The two other large valleys, Wright and Victoria, are located north of Taylor Valley. The smaller valleys are inland of Taylor Valley—Pearse Valley is just north of the Taylor Glacier, and Arena and Beacon Valleys are located further inland within the central Transantarctic Mountains (Fig. 2). Landscape Age in the McMurdo Dry Valleys In order to determine the relationship between CaCO3 concentration and soil age, knowledge of landscape development is important. Much work has been done to determine recent glacial chronology and soil age in the McMurdo Dry Valleys region.
Three different glaciations have been documented in Taylor Valley: the Taylor, the Ross Sea, and the Alpine glaciations (Denton et al., 1989; Hall and Denton, 2000; Hall et al., 2000). Several eastward advances and retreats of the Taylor Glacier (from the East Antarctic Ice Sheet) deposited till in western parts of the valley. It has been demonstrated that the Taylor Glacier advanced during the interglacials, or warmer time periods. The three major Taylor glaciations have been termed Taylor IV glaciation (2100– 3700 ka), Taylor III glaciation (200–210 ka), and Taylor II–Bonney glaciation (74–98 ka). The present-day profile of Taylor Valley was formed during these Taylor glaciations. During the Ross glaciations, the West Antarctic Ice Sheet blocked the Taylor Valley from the Ross Sea side, which led to the development of a
Calcium carbonate in Antarctic soils
93
Figure 2. Landsat image of the research area, which includes Taylor Valley (TV), Wright Valley (WV), Victoria Valley (VV), Beacon Valley (BV), and Arena Valley (AV). Cape Bernacchi and Marble Point are just east of Taylor Valley.
valley-wide paleolake, Glacial Lake Washburn, during the late Wisconsin, ca. 22,800–8500 yr B.P. (Stuiver et al., 1981). Multiple expansions of the Ross Sea ice sheet onto the shoreline at Marble Point and Cape Bernacchi occurred; therefore, soil ages in this area vary with elevation and distance inland. For example, a raised beach at 18 m containing an A. colbecki (an Antarctic scallop) shell dates to 5325 yr B.P., and an elephant seal at 13.4 m dates to 4227 yr B.P. (Denton et al., 1989; Hall and Denton,
2000). Figure 3 shows a reconstruction of the till remains from the Taylor and Ross Sea glaciations. Lastly, Alpine glaciations affected elevated areas in Taylor Valley. During the interglacial periods, the alpine glaciers in the Asgard Range expanded from the north, and the alpine glaciers in the Kukri Hills expanded from the south (Fig. 2). All three glaciations (Taylor, Ross, and Alpine) in the McMurdo Dry Valleys deposited tills, from which soils developed. Hence, soil ages vary from ca. 3700 ka (Taylor
94
Foley et al.
Figure 3. The glacial till sequence in Taylor Valley. Sample transects are shown as Bonney Riegel (BR), Middle Taylor (MT), Lake Hoare (LH), Lake Fryxell (LF), North Lower Taylor (NLT), and South Lower Taylor (SLT). Transects represent environmental gradients, including the distance from Ross Sea, elevation from the valley floor, and distance from the present-day dry valley lake shores. Three transects (BR, LH, and LF) are associated with major ice-covered lakes in the valley, and the three transects (MT, SLT, and NLT) have no present-day relationship to lacustrine systems. On the valley walls are present-day glaciers and glacial till from alpine glaciations. (Map is from Burkins et al. [2000] and was reconstructed from McKelvey and Webb [1972]; Kellogg et al. [1980]; Pastor and Bockheim [1980]; Stuiver et al. [1981]; and Denton et al. [1989].)
TABLE 1. MEAN ANNUAL TEMPERATURES FOR VARIOUS LOCATIONS WITHIN THE MCMURDO DRY VALLEYS Temperature Location (°C) Marble Point –16.7 Taylor Valley (average) –17.6 New Harbour Camp –19.29 Commonwealth Glacier –17.88 F6 (SE Fryxell Basin) –20.73 Hughes Glacier –17.35 Canada Glacier –17.51 Wormherder Farm (south shore Lake Hoare) –17.6 Lake Bonney camp –17.59 Taylor Glacier –17.59 Lake Brownworth –20.71 Lake Vanda –19.65 Lake Vida –27.06 † –30 to –35 Beacon Valley Arena Valley N/A † From Sugden et al. (1995). All others are from Doran et al. (2002a). Data range from 1986 to 2000.
IV glaciation) to ca. 12.4 ka or younger (Ross I glaciation) (Fig. 3). Multiple smaller ancient lakes have expanded and contracted, mostly corresponding with isotope stages 1, 2, 5, 6, 7, possibly 8, and 9, throughout late Quaternary time in Taylor Valley (Hendy, 2000). Very large lakes developed in Taylor Valley during times of low sea levels (isotope stages 2, 6, and possibly 8), when the Ross Ice Shelf moved north, causing the ice surface to rise steeply. During times of high sea level (isotope stages 1, 5, 7, and 9), much smaller lakes occupied Taylor Valley (Hendy, 2000). Lake formation is a direct result of the rate of melt from glaciers (e.g., alpine and the more established glaciers, such as Taylor Glacier) or local change in the evaporation-precipitation regime. The multiple filling and draining of lakes in Taylor Valley have left multiple layers of lacustrine sediment scattered about the floor of the valley (Hendy, 2000). These lake sediments can contain carbonate minerals produced in part via photosynthesis (Hendy, 2000; Neumann, 1999).
Calcium carbonate in Antarctic soils
95
Figure 4. Sampling method used in the 2005 field season. A plastic scoop was used to transfer ~100 g of sediment into a small Whirlpack®, and a plastic ruler was used to measure depth of sample excavated.
METHODS The soil CaCO3 data were assembled in three steps: we compiled data from past studies (Claridge, 1963), analyzed samples collected in 2002–2003 by MCM-LTER field team members, and analyzed soils collected in January 2005 by the senior authors.
the University of Colorado, Boulder, and 0.5 g of sample was analyzed on a coulometer (CM5130 acidification module) to determine the calcium carbonate concentration. The precision of the measurement was ±0.3%. The titration accuracy was better than ±0.15%, if sufficient sample was used to evolve over 1000 μg C.
Sample Collection, Preparation, and Protocol
RESULTS
Samples from Claridge (1963) were collected from depths ranging from 0 to 6 cm below the surface. These samples were collected long before the use of global positioning systems (GPS), and their locations are only generally described. Samples from the 2002–2003 MCM-LTER studies were from surface depths of 0–5 cm, plus two soil pits, which extended to 25 cm and 30 cm below the surface. Samples from the January 2005 field season were taken in Taylor Valley from near the terminus of Taylor Glacier to the Ross Sea coast and northward to Marble Point, and include three of the nine Beacon Valley samples. They were collected from surface depths of 0–3.5 cm (where soil development was minimal) to depths of 0–6 cm below the desert pavement zone (Fig. 4). Locations of both of these sets of recent samples were recorded via GPS. Seventy-six of the 2002–2003 samples and eighty-nine of the 2005 samples were prepared by sieving ~5 g of sediment through a 2.00 mm mesh and a 500 μm mesh sieve. About 1 g of the fine sediment (<500 μm) was then isolated and placed in sealable clear plastic snack bags. The samples were then double bagged to minimize contamination. A 0.5–1.0 g subsample was sent to Institute of Arctic and Alpine Research (INSTAAR) at
Surface Distribution Concentrations of calcium carbonate in Taylor Valley surface soil samples (0–6 cm) show a general geographic trend—CaCO3 decreases from the McMurdo Sound coast inland to the Taylor Glacier (Fig. 5). This trend is similar to what was observed by Keys and Williams (1981). Soil carbonate maxima occur in two areas with respect to distance from the Ross Sea (Fig. 5). The first maximum is the area along the Ross Sea coast, where carbonate concentrations range from 0.08% to 9.77% CaCO3, with most of the concentrations between 0.08% and 3.0% CaCO3. This wide range of values is only observed along the Ross Sea coast. Between the coast and ~20 km inland, the soil carbonate values are mostly between 0% and 0.67% CaCO3, with a few concentrations slightly higher, but below 1.33% CaCO3. The second soil carbonate maximum, with respect to distance from the Ross Sea coast, is in soils between ~20 km and ~30 km inland. This area is between the terminus of the Suess Glacier and the Marr Glacier, from 200 to 800 m elevation in the area of the eastern lobe of Lake Bonney basin. The area (Fig. 1) contains carbonate concentrations between 0% and 4.26% CaCO3 (with most being <0.33%
96
Foley et al.
Figure 5. Distance inland from the Ross Sea coast versus CaCO3.
CaCO3) and has an average of 0.81% CaCO3. Soils between 0 and 10 km inland (i.e., the soils between the two maxima) and soils past the second maximum, >30 km inland, contain much less carbonate (all having <2.5% CaCO3) and have an average of 0.33% CaCO3. Soils in the west lobe of Lake Bonney basin, and further inland, contain minimal amounts of soil carbonate. Moreover, soil carbonate values decrease with increasing distance from the coast (Fig. 6), excluding the area ~20–30 km inland. In addition to the trend of decreasing CaCO3 concentration with distance from the coast, soils in the youngest till near the coast contain higher amounts of CaCO3 than any of the older soils farther inland and at elevation (Fig. 7). Although Claridge and Campbell (1976) and Bockheim (1982) have shown that more soluble salts accumulate in greater amounts in older, drier soils in these valleys, it is now clear that carbonate minerals are less abundant in the older, drier soil tills than in the younger tills (Fig. 7). The distribution of carbonate minerals in Taylor Valley is also related to elevation (Fig. 8). Higher concentrations of CaCO3 are found below an elevation of 336 m. This elevation was the highest water level of Glacial Lake Washburn, which existed between 22,800 and 8500 yr B.P. when the Ross Ice Sheet blocked the mouth of Taylor Valley (Hendy, 2000). Pedogenic carbonate is found at much lower concentrations in the other, higher-elevation valleys of the McMurdo Dry Valleys. Analyses of the soils from West Dias (77°33′S, 161°10′E) and East Dias (77°33′S, 162°43′E) in the Wright Valley indicate an absence of carbonate. There also was no measurable CaCO3 in Lower Victoria (77°22′S, 162°19′E). A soil sample from the Lake Vida region, Victoria Valley, contained only 0.25% CaCO3 (77°23′S, 162°03′E), and a soil sample from Bull Pass had only 0.08% CaCO3 (77°29′S, 161°49′E). Carbonate concentrations were also low to nonexistent in Pearse (0.42%), Beacon (0.0% and 0.016%), and Arena Valleys (0.0%).
Figure 6. Distance inland versus CaCO3 averages from the three zones classified by Marchant and Denton (1996): zone 1, the coastal lowelevation zone; zone 2, the intermediate zone; and zone 3, the inland high-elevation zone.
Figure 7. Till age versus CaCO3 in Taylor Valley. Data are averaged from number of samples indicated next to each point.
Calcium carbonate in Antarctic soils
Figure 8. Elevation versus CaCO3. The solid line shows where the high-water mark of Glacial Lake Washburn stood, and, therefore the extent of possible lacustrine-derived soil carbonates.
97
of pedogenic CaCO3 occur at depth in soil profiles (Van der Hoven and Quade, 2002). Because of the location on the valley floor, the CaCO3 present at depth in these Taylor Valley soils could represent lacustrine production of CaCO3 during times of higher lake-level stands (Higgins et al., 2000) rather than true pedogenic production. Detailed isotopic analysis of the carbonate minerals would be the primary method to answer this question. In 2005 a set of soil samples was collected along a slope located 1 km inland from New Harbour Camp (i.e., the Ross Sea coast) on the edge of a medium-sized till mound that stood 15.85 m high (77°34′59.0′′S, 163°28′23.5′′E). This was done to evaluate the hypothesis that small-scale landscape gradients can lead to preferential accumulations of CaCO3 as hypothesized by McCraw (1967). The till at this location was small-grain pebbles, and scattered larger rocks and boulders, randomly distributed on the surface. These pebbles, rocks, and boulders were mostly the basement rock complex, including few small marble pieces, as well as some dolerite. The top of the mound contained the largest clusters of boulders made of basement rock. The mound was surrounded by multiple till mounds of similar size and matrix materials. Surface soils were sampled between 67 m to 83 m elevation. CaCO3 concentrations increased with decreasing elevation (Table 3). The importance of these results to CaCO3 formation will be discussed in the following section. DISCUSSION
Vertical Profile Distributions Valley-Scale Variation During the 2002–2003 season, soil pits were dug and sampled by MCM-LTER soil ecologists at sites in the Lake Hoare basin and in the Lake Bonney basin (Table 2). Neither the Lake Hoare basin soil pit nor the Lake Bonney basin soil pit indicated a consistent trend of CaCO3 distribution with depth. In the Lake Hoare soil pit, the highest concentrations are at depths of 7.5 cm (<2.8% CaCO3), 17.5 cm (<4.3% CaCO3), and 27.5 cm (<2.4% CaCO3) (Fig. 9A– 9B). In the Lake Bonney basin, the depth profiles are different, with peaks at the surface and at 27.5 cm. The soil CaCO3 concentrations are much lower in Bonney basin than in the Lake Hoare basin; the highest concentrations are less than 0.7% in the Lake Bonney soils (Fig. 9A–9B). These data agree with the surface distribution pattern in Taylor Valley previously discussed, where soil carbonate decreases with increasing distance from the coast. Previous work in warm desert systems has demonstrated that the highest amounts
Compared to hot arid regions, the CaCO3 content of the polar desert of the McMurdo Dry Valleys is quite low. Using the depth to permafrost (30 cm, 35 cm, and 40 cm for Lake Fryxell, Lake Hoare, and Lake Bonney basins, respectively) and the mean surface CaCO3 values, the Fryxell, Hoare, and Bonney basin soils have inorganic C concentrations of 0.38, 0.31, and 0.68 kg C m–2, respectively (Table 2). Using the integrated depth profiles from the soil pits, the average inorganic C in the Hoare basin is 0.69 kg C m–2, and 0.20 kg C m–2 in the Bonney basin, as there is more inorganic C with depth in the Hoare basin than in the Bonney basin. Therefore, the average inorganic C in Taylor Valley soils is estimated to be ~0.5 kg C m–2. The world average value in arid regions is 33.2 kg C m–2, while Arizona desert soils have 24.5 kg C m–2 (Schlesinger, 1985). These values from Taylor Valley,
TABLE 2. CARBON AND CARBONATE DATA FOR SOILS IN TAYLOR VALLEY –2 Permafrost depths CaCO3 Soil organic Elevation Soils location kg C m † ‡ (%) (cm) C Taylor Valley 0.47 0.72 Lake Fryxell basin 0.38 0.71 30 0.31 ± 0.02 107 m Lake Hoare basin 0.31 0.48 35 0.29 ± 0.03 110 m Lake Bonney basin 0.68 0.97 40 Lake Hoare soil pit 0.69 1.26 Lake Bonney soil pit 0.20 0.26 † From J.E. Barrett (2004, personal commun.). ‡ From Barrett et al. (2005).
98
Foley et al.
Figure 9. (A) CaCO3 concentrations in four Lake Hoare soil pits. Error bars represent the standard error for the data in each soil pit. (B) CaCO3 concentrations in four Lake Bonney soil pits. Error bars represent the standard error for the data in each soil pit.
TABLE 3. CaCO3 CONCENTRATIONS ON SLOPE OF MOUND LOCATED ON THE FLOOR OF TAYLOR VALLEY BETWEEN EXPLORERS COVE AND LAKE FRYXELL Elevation Elevation CaCO3 (m; with respect to sea level) (m; between sample locations) (%) 83 5 0.08 78 5 0.08 73 1 0.25 72 5 0.67 67 0 0.67
however, are skewed because the active layer is less than 50 cm deep (Table 2). Not unlike other arid regions of the world (Landi et al., 2003), inorganic C in the form of CaCO3 makes up a high percentage of the total C in the surface soils of the Taylor Valley soils, 55% and 53%, respectively, for the Fryxell and Hoare basins (Table 2). Marchant and Denton (1996) divided the McMurdo Dry Valleys region into three distinct climatic zones on the basis of precipitation, humidity, temperature, and the amount of moisture from permafrost and glacial melt in the soils. The coastal zone (zone 1) contains the most moisture, zone 2 is termed an intermediate zone, while zone 3 is of the highest elevation and contains more ancient surfaces with the least moisture. The coastal zone has subxeric soils, the intermediate region soils are xeric, while the zone 3 soils have been termed ultraxeric (Bockheim, 1997). Taylor Valley contains only zones 1 and 2. Easterly winds that usually flow into the eastern portion of Taylor Valley during the summer months bring in moist air from the Ross Sea (Bull, 1966; Doran et al., 2002b; Fountain et al., 1999), whereas the westerly katabatic winds that blow into western Taylor Valley off the
polar plateau are extremely arid (Doran et al., 2002b). This wind pattern is a major contributor to the soil-moisture and relative humidity within each zone and is a key parameter in the distribution of the terrestrial features, such as soil development. The Fryxell and Hoare basin soils are not only the youngest in Taylor Valley, they are also the regions in Taylor Valley with the highest relative humidity, the lowest average wind speed, and hence the potential for the highest soil moistures in the valley (Doran et al., 2002a). Moisture in the coastal soil is from the thawing of permanent ice and semipermanent snow patches, which are widespread in the coastal region in the summer (Campbell et al., 1998). Zone 2, the intermediate zone, shows some moisture-produced landforms such as modern gelifluction lobes and debris flows that occur randomly on the north-facing slopes and/or in protected areas with higher moisture content (Marchant and Denton, 1996). Temperature extremes are greater here than at the coast, but when the snow melts it rapidly evaporates, and little moisture is incorporated into the soil matrix (Campbell et al., 1998). Bull (1966) argued that the relative humidity in zone 1 was about two times that of zone 2. Zone 3, the most inland and highest-elevation zone that covers Arena Valley and Beacon Valley and the western area of Wright Valley and Victoria Valley, contains features that are of ancient microtopography with no moisture-derived landforms or melt streams (Marchant and Denton, 1996). Thus, the youngest soils are also the most likely to be the wettest. The distribution of CaCO3 in the soils of the McMurdo Dry Valleys can clearly be related to both the distance from the ocean and the landscape position and to the geomorphological climatic classification of Marchant and Denton (1996). The highest CaCO3 concentrations are in zone 1; there are lower, but in most cases measurable concentrations in zone 2, and very little
Calcium carbonate in Antarctic soils CaCO3 is present in zone 3 (Fig. 10). This distribution strongly argues for the requirement of liquid water to be present, at least for a portion of the austral summer, before CaCO3 can form in these polar desert soils. Since liquid water is initially needed as a medium in CaCO3 formation, it is important to consider moisture quantities and distribution in the McMurdo Dry Valleys when identifying CaCO3 formation. Marion (1989) demonstrated that the rate of pedogenic carbonate formation in the Southwestern United States desert soils is correlated with annual rainfall. In the boreal region of Canada, the rate of pedogenic carbon accumulation also increases with increasing annual precipitation (Landi et al., 2003). In the McMurdo Dry Valleys, annual precipitation is <100 mm, but this is an estimate based on little quantitative information (Doran et al., 2002b). The only significant present-day water source in the McMurdo Dry Valleys is from the austral summer melt of the glaciers, which feeds the streams and lakes (Fountain et al., 1998). Because there is no overland flow in the McMurdo Dry Valleys and the glacier meltwater is confined to stream channels only, there is no method for supplying moisture to the soil for the CaCO3 formation process. Therefore, much of the pedogenic CaCO3 is produced where there was a significant water source, either sufficient snow melt (Gooseff et al., 2003) or subsurface ice melt (Lyons et al., 2006). Within zone 1, there is a bimodal distribution at 1–10 km inland (i.e., the coast) and at 22–30 km inland (i.e., between the region at the snout of the Suess Glacier and the snout of Taylor Glacier). The higher concentrations found more inland, between 20 and 30 km, are near the face of the Defile (a narrow section of the valley across from Suess Glacier that separates the upper
Figure 10. Distance inland versus CaCO3 with respect to zones 1, 2, and 3. Figure is the same as Figure 5, but has Marchant and Denton’s (1996) zones highlighted. Nussbaum Riegel datum falls in zone 2, but because it is of questionable validity, it has been singled out in this graph.
99
valley and lower valley), the south shore of Lake Bonney, and near Mummy Pond just east of Lake Bonney. These locations could represent locations of former lacustrine sediment (Hall and Denton, 2000; Hendy, 2000). Our data suggest that distance from the coast and elevation are the major factors in CaCO3 distribution, whereas till age itself has little or no relationship to CaCO3 distribution (Fig. 7). The strong correlation between soil carbonate concentration in Taylor Valley and elevation is related to the maximum height of ancient Glacial Lake Washburn (Fig. 8). Soil carbonate accumulations below this elevation are greater than those above the 336 m elevation mark. However, it is probable that not all the carbonate below this elevation is of paleolacustrine material, as some is produced by in situ soil processes. The soils sampled above Lake Washburn’s maximum height in Taylor Valley are located on the south shore of Lake Bonney (n = 1), Andrew’s Ridge (n = 4), slope of Defile (n = 1), and between the Nussbaum Riegel and the terminus of the Marr Glacier (n = 11) and have a mean value of 0.53%. These locations are definitely late Pleistocene to Holocene pedogenic carbonate, and are not of lacustrine origin. Accumulation Rates The annual accumulation rate of CaCO3 in Taylor Valley soils can be calculated by dividing the concentration by the age of the soil (i.e., till). CaCO3 accumulation for the Fryxell, Hoare, and Bonney basins is 0.260, 0.034, and 0.071 g m–2 yr–1, respectively. These are maximum accumulation rates because we assume that all the CaCO3 is pedogenically produced. Using Marion’s (1989) empirical relationship between CaCO3 accumulation and mean annual precipitation for the U.S. warm desert and a maximum precipitation rate of 100 mm yr–1 for McMurdo Dry Valleys, the accumulation of CaCO3 for McMurdo Dry Valleys soils is 0.95 g m–2 yr–1. This is a maximum accumulation because the precipitation input of 10 cm yr–1 is the upper level for the McMurdo Dry Valleys (Doran et al., 2002b). Landi et al. (2003) recently determined values between 8 and 14 g m–2 yr–1 for Saskatchewan soils with a strong positive correlation between CaCO3 accumulation and precipitation. Schlesinger (1985) obtained values between 1.0 and 3.5 g m–2 yr–1 for the Mojave Desert, and Marion and Schlesinger (1994) modeling carbonate deposition in soils of the southwest desert of the United States, obtained values of 1–5 g m–2 yr–1. They argued that most of the CaCO3 accumulation occurred during cool, wet climate intervals. Therefore, the Fryxell and Hoare basin soils are accumulating CaCO3 at a much lower rate than most arid deserts. These McMurdo Dry Valleys values are low in part because of the relatively shallow permafrost depth in the Taylor Valley soils (Table 3). Regardless of basin, soils and slopes with northerly aspects in Taylor Valley are thought to have more moisture than those with southerly aspects and, therefore calcite (and gypsum) should accumulate on the moister north-facing slopes (Marchant and Denton, 1996). However, the data presented here essentially show no difference between the CaCO3 concentration on the dry
100
Foley et al.
south-facing slopes, 0.62% CaCO3 (n = 13), versus 0.59% CaCO3 (n = 54) on the more moist northern slopes. Small-Scale Variation On a smaller topographic scale, McCraw’s (1967) work on soil moisture regimes demonstrated that moisture flow and gravity dictate the distribution of water-soluble salts such as CaCO3 in McMurdo Dry Valleys soils, and therefore small-scale topography could play a very important role in pedogenic carbonate formation. To expand on this work, five soil samples were taken on the side of a glacial carved till mound ~1 km inland from New Harbour Camp on the coast of Explorers Cove (Fig. 2). According to McCraw’s (1967) conceptual model, there should be higher CaCO3 accumulations just below the surface at the top of the mound, at the crest of the mound where the slope turns downward, and very high concentrations of CaCO3 at the toe of the slope. There should be little or no CaCO3 on the steep part of the slope and on the valley floor below the hill. Data shown here do not follow this distribution pattern. Results show CaCO3 soil concentrations increase from the top of the hill to the valley floor (Fig. 11); presumably moisture and solute transport is driven by gravity acting as the controlling force. Sources of Ca2+ to the Pedogenic Carbonate A major factor controlling pedogenic carbonate formation is the amount of readily available Ca2+ in the soil. Van der Hoven and Quade (2002) demonstrated through isotopic analyses that in the desert southwest of the United States there are two Ca2+ sources available to form the pedogenic carbonates: a local parent material and a dust-deposited source. Potential readily soluble sources of Ca2+ in the Taylor Valley soils include marble and kenyite (Claridge, 1963; McCraw, 1967) and the abundance of dust transported into the valleys through wind action (Wellman, 1964). (Kenyite is an unusual type of volcanic rock produced on Ross Island and transferred into Fryxell Basin by the West Antarctic Ice Sheet during glacial periods). The marble and kenyite only exist in the Ross tills and hence are in the Fryxell basin and perhaps the Hoare basin (Hall and Denton, 2000). Therefore, the higher concentrations of CaCO3 found in the Fryxell basin soils may be due in part to the higher source potential for Ca2+ that exists there. MCM-LTER data collected over the past 11 yr indicate that streams in the Fryxell and Hoare basins have dissolved Ca2+ concentrations as high as 1.5 mM (Fig. 12). However, the source of this Ca2+ may be different in the two basins because the Hoare basin is dominated by CaCO3 dissolution, while the weathering of silicate minerals is a much more important source in the Fryxell basin (Nezat et al., 2001). The longer, glacier-fed southshore streams in the Fryxell basin (Fig. 1) are more enriched in H4SiO4 and depleted in Ca2+ compared to the shorter, glacier-fed north-shore streams (Fig. 12). This suggests, but certainly does not prove, that the weathering of kenyite, especially in the southern Fryxell basin, may be an important source of Ca2+. This idea
is supported by previous Sr2+ isotope work on the Taylor Valley streams and lakes (Lyons et al., 2002), where the streams entering Lake Fryxell have the most nonradiogenic waters in the valley. In addition to the potential till sources, dust being transported into and through Taylor Valley is relatively enriched in Ca2+. We measured the total major-element geochemistry of both eolian sediment and loess deposition at two locations in Taylor Valley using X-ray fluorescence (XRF) (Table 4). Comparison of the elemental ratios of these samples to elemental ratio values from the upper continental crust (UCC) and Holocene dust collected from the Taylor Dome ice core (~100 km west of Taylor Valley; Table 4) shows that the Taylor Valley eolian materials have slightly higher Ca/Sr and Ca/Ba ratios than UCC and very much lower K/Ca ratios, indicating Ca2+ enrichment in the dust relative to the crust. The transport, deposition, and subsequent dissolution of this dust material have an important impact on the geochemistry of the supraglacial as well as proglacial streams (Fortner et al., 2005) and may play an important role within soil environments as it could provide an important source of soluble Ca2+. Planned future investigations on the isotopic chemistry of the pedogenic carbonate as well as on the eolian materials should help to better constrain the sources of Ca2+ to the pedogenic carbonate. These isotopic data can then be compared to previous work on the individual rock types that make up the tills (Lyons et al., 2002). This work is currently ongoing and will help determine if the carbonate in these soils is truly pedogenic or if it was produced in paleolacustrine settings. CONCLUSIONS CaCO3 concentrations in soils from this polar desert environment are much lower than those observed in other regions of the world. The hyperarid character of this environment undoubtedly limits the formation of the CaCO3 in the soils. The distribution of pedogenic carbonate in the McMurdo Dry Valleys is related to landscape position with respect to distance from the Ross Sea coast and, within Taylor Valley, with elevation related to the maximum height of Glacial Lake Washburn. The higher concentrations of CaCO3 in the coastal soils may either be a direct result of increased moisture from a greater source of soluble Ca2+ from the marine source, from the marble and kenyite in the coastal tills from the Ross Sea glaciation (12.4–23.8 ka), or a combination of both. Overall, the highest CaCO3 concentrations in the soils exist where the soil moisture content is the highest and the soils are the youngest, which is near the coast. Among all of the McMurdo Dry Valleys, the Taylor Valley is the only valley that supports this type of microclimate for pedogenic carbonate accumulation. Although there is a CaCO3 surface distribution pattern in Taylor Valley, the CaCO3 distribution with depth in the soils in the Lake Hoare and the Lake Bonney basins shows no consistent pattern. The rates of Taylor Valley pedogenic carbonate formation are at the lower range of what has been observed in warm deserts. Given the locations of the highest concentrations, it is possible that a portion of the CaCO3 present in the lower elevations of
Calcium carbonate in Antarctic soils
101
Figure 11. Hill profile from McCraw (1967) with five sections separating the soil moisture and CaCO3 concentration relationship. McCraw argued that CaCO3 should be most abundant in sections 1, 2, and 4 as evidenced by the stipled areas in the figure. Values in parentheses are CaCO3 values from this work that do not follow McCraw’s theory for carbonate distribution on a hillside.
TABLE 4. ELEMENTAL RATIOS (WT/WT) OF EOLIAN MATERIALS IN TAYLOR VALLEY K/Rb Ca/Sr Ca/Ba K/Ca † Lake Hoare “loess” 279 99 72 0.39 † 250 108 54 0.60 Taylor Glacier “loess” 1 † Taylor Glacier “loess” 2 234 108 63 0.50 ‡ 350 120 130 0.45–0.59 Taylor Dome dust § UCC 260 93 44 0.97 Note: Taylor Dome dust and the upper continental crustal (UCC) average are shown for comparison. † W.B. Lyons and C. Dowling (2004, personal commun.). ‡ Hinkley and Matsumoto (2001). § Wedepohl (1995).
Taylor Valley is lacustrine, formed during the last highstand of Glacial Lake Washburn. ACKNOWLEDGMENTS
Figure 12. Ca versus Si for the Hoare and Fryxell basin streams. Ca and Si were analyzed from stream water samples taken near the mouth of the streams. These are McMurdo Dry Valley Long-Term Ecological Research site data from www.mcmlter.org. Data span the period 1993–2003.
This work was supported through National Science Foundation (NSF) grants (OPP-9813061) and is a contribution to the McMurdo Dry Valleys–Long-Term Ecological Research (LTER) Program. Sincere thanks are due to Wendy Freeman Roth and Suzanne Prestrud Anderson at Institute of Arctic and Alpine Research (INSTAAR), University of Colorado, for the CaCO3 analysis. We thank Kathy Welch and April Jacobs for their stream data analysis and Carolyn Dowling for the X-ray fluorescence (XRF) data of eolian materials. We thank Kate Harris for helping
102
Foley et al.
in the collection of the 2005 samples. Special thanks are given to the staff of the Crary Lab and at Lake Hoare and to PHI for air support in the Dry Valleys. We thank Ana M. Alonso-Zarza and Giles Marion, as well as an anonymous reviewer, for their constructive input to the original manuscript. REFERENCES CITED Bao, H., Bockheim, J.G., Campbell, D.A., and Thiemans, M.H., 2000, Origins of soleplate in Antarctic dry-valley soils as deduced from anomalous 17O compositions: Nature, v. 407, p. 499–502, doi: 10.1038/35035054. Barrett, J.E., Virginia, R.A., Parsons, A.N., and Wall, D.H., 2005, Potential carbon and nitrogen turnover in soils of the McMurdo Dry Valleys, Antarctica: Arctic, Antarctic and Alpine Research, v. 37, p. 107–116. Bockheim, J.G., 1982, Properties of a chronosequence of ultraxerous soils in the Trans-Antarctic Mountains: Geoderma, v. 28, p. 239–255, doi: 10.1016/0016-7061(82)90005-2. Bockheim, J.G., 1997, Properties and classification of cold desert soils from Antarctica: Soil Science Society of America Journal, v. 61, p. 224–231. Brass, G.W., 1975, The effect of weathering on the distribution of strontium isotopes in weathering profiles: Geochimica et Cosmochimica Acta, v. 39, p. 1647–1653, doi: 10.1016/0016-7037(75)90086-1. Bull, C.B., 1966, Climatological observations in ice-free areas of southern Victoria Land, Antarctica, in Rubin, M.J.M., ed., Studies in Antarctic Meteorology: Antarctic Research Series, v. 9, p. 177–194. Burkins, M.B., Virginia, R.A., Chamberlain, C.P., and Wall, D.H., 2000, Origin and distribution of soil organic matter in Taylor Valley, Antarctica: Ecology, v. 81, p. 2377–2391, doi: 10.2307/177461. Campbell, I.B., and Claridge, G.G.C., 1969, A classification of frigid soils— The zonal soils of the Antarctic continent: Soil Science, v. 107, p. 75–85. Campbell, I.B., and Claridge, G.G.C., 1982, The influence of moisture on the development of soils of the cold deserts of Antarctica: Geoderma, v. 28, p. 221–238, doi: 10.1016/0016-7061(82)90004-0. Campbell, I.B., and Claridge, G.G.C., 1987, Antarctica: Soils, Weathering Processes and Environment: Amsterdam, Elsevier, p. 368. Campbell, I.B., Claridge, G.G.C., Campbell, D.I., and Balks, M.R., 1998, The soil environment of the McMurdo Dry Valleys, Antarctica, in Priscu, J.C., ed., Ecosystem Dynamics in a Polar Desert: The McMurdo Dry Valleys, Antarctica: Washington, D.C., American Geophysical Union Antarctic Research Series, v. 72, p. 297–322. Capo, R.C., and Chadwick, O.A., 1999, Sources of strontium and calcium in desert soil and calcrete: Earth and Planetary Science Letters, v. 170, p. 61–72, doi: 10.1016/S0012-821X(99)00090-4. Cerling, T.E., 1984, The stable isotopic composition of modern soil carbonate and its relationship to climate: Earth and Planetary Science Letters, v. 71, no. 2, p. 229–240, doi: 10.1016/0012-821X(84)90089-X. Claridge, G.G.C., 1963, The clay mineralogy and chemistry of some soils from the Ross Dependency, Antarctica: New Zealand Journal of Geology and Geophysics, v. 8, p. 186–220. Claridge, G.G.C., 1977, Development and significance of polygenetic features in Antarctic soils: New Zealand Journal of Geology and Geophysics, v. 20, p. 919–931. Claridge, G.G.C., and Campbell, I.B., 1976, The salts in Antarctic soils, their distribution and relationship to soil processes: Soil Science, v. 123, no. 6, p. 377–384. Clark, I.D., and Lauriol, B., 1992, Kinetic enrichment of stable isotopes in cryogenic calcites: Chemical Geology, v. 102, p. 217–228, doi: 10.1016/00092541(92)90157-Z. Denton, G.H., Bockheim, J.G., Stuiver, M., and Wilson, S.C., 1989, Late Wisconsin and early Holocene glacial history, inner Ross Embayment, Antarctica: Quaternary Research, v. 31, no. 2, p. 151–182, doi: 10.1016/00335894(89)90004-5. Doran, P.T., Priscu, J.C., Lyons, W.B., Walsh, J.E., Fountain, A.G., McKnight, D.M., Moorhead, D.L., Virginia, R.A., Wall, D.H., Clow, G.D., Fritsen, C.H., McKay, C.P., and Parsons, A.N., 2002a, Antarctic climate cooling and terrestrial ecosystem response: Nature, v. 415, p. 517–520, doi: 10.1038/nature710. Doran, P.T., McKay, C.P., Clow, G.D., Dana, G.L., Fountain, A.G., Nylen, T., and Lyons, W.B., 2002b, Valley floor climate observations from the
McMurdo Dry Valleys, Antarctica, 1986–2000: Journal of Geophysical Research, v. 107, no. 1029, doi: 10.1029/2001JA900104. Fairchild, I.J., and Spiro, B., 1990, Carbonate minerals in glacial sediments: Geochemical clues to paleoenvironment: Glacimarine environments: Processes and Sediments, v. 53, p. 201–216. Fortner, S., Tranter, M., Fountain, A., Lyons, W.B., and Welch, K.A., 2005, The geochemistry of supraglacial streams of Canada Glacier, Taylor Valley, Antarctica, and their evolution into proglacial streams: Aquatic Geochemistry, v. 11, p. 391–412. Fountain, A., Dana, G.L., Lewis, K.J., Vaughn, B.H., and McKnight, D.M., 1998, Glaciers of the McMurdo Dry Valleys, Southern Victoria Land, Antarctica, in Priscu, J.C., ed., Ecosystem Dynamic in a Polar Desert, the McMurdo Dry Valleys, Antarctica: Washington, D.C., American Geophysical Union Antarctic Research Series, v. 72, p. 65–75. Fountain, A.G., Lyons, W.B., Burkins, M.B., Dana, G.L., Doran, P.T., Lewis, K.F., McKnight, D.M., Moorhead, D.L., Parsons, A.N., Priscu, J.C., Wall, D.H., Wharton, R.A., Jr., and Virginia, R.A., 1999, Physical controls on the Taylor Valley ecosystem, Antarctica: Bioscience, v. 49, no. 12, p. 961–971, doi: 10.2307/1313730. Gooseff, M.N., Barrett, J.E., Doran, P.T., Fountain, A.G., Lyons, W.B., Parsons, A.N., Porazinska, D.L., Virginia, R.A., and Wall, D.H., 2003, Snowpack influence on soil biogeochemical processes and invertebrate distribution in the McMurdo Dry Valleys, Antarctica: Arctic, Antarctic, and Alpine Research, v. 35, no. 1, p. 91–99. Hall, B.L., and Denton, G.H., 2000, Extent and chronology of the Ross Sea Ice Sheet and the Wilson Piedmont Glacier along the Scott Coast at and since the Last Glacial Maximum: Geografiska Annaler, v. 82A, no. 2–3, p. 337–362, doi: 10.1111/1468-0459.00128. Hall, B.L., Denton, G.H., and Hendy, C.H., 2000, Evidence from Taylor Valley for a grounded ice sheet in the Ross Sea, Antarctica: Geografiska Annaler, Physical Geography, v. 82(A), no. 2–3, p. 275–304. Hallet, B., 1976, Deposits formed by subglacial precipitation of CaCO3: Geological Society of America Bulletin, v. 87, p. 1003–1015, doi: 10.1130/00167606(1976)87<1003:DFBSPO>2.0.CO;2. Haskell, T.R., Kennett, J.P., Prebble, W.M., Smith, G., and Willis, I.A.G., 1965, The geology of the middle and lower Taylor Valley of south Victoria Land, Antarctica: Transactions of the Royal Society of New Zealand, Wellington: Geology, v. 2, p. 169–184. Hendy, C.H., 2000, Late Quaternary lakes in the McMurdo Sound region of Antarctica: Geografiska Annaler, v. 82(A), p. 411–430. Higgins, S.M., Denton, G.H., and Hendy, C.H., 2000, Glacial geomorphology of Bonney Drift, Taylor Valley, Antarctica: Geografiska Annaler, v. 82(A), no. 2–3, p. 365–389, doi: 10.1111/1468-0459.00129. Hinkley, T.K., and Matsumoto, A., 2001, Atmospheric regime of dust and salt through 75,000 years of the Taylor Dome ice core: Refinement by measurement of major, minor, and trace metal suites: Journal of Geophysical Research, v. 106, p. 18,487–18,493, doi: 10.1029/2000JD900550. Jones, L.M., and Faure, G., 1978, A study of strontium isotopes in lakes and surficial deposits of the ice-free valleys, southern Victoria Land, Antarctica: Chemical Geology, v. 22, p. 107–120, doi: 10.1016/00092541(78)90027-X. Kellogg, D., Stuiver, M., Kellogg, T., and Denton, G.H., 1980, Non-marine diatoms from late Wisconsin perched deltas in Taylor Valley, Antarctica: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 30, p. 157–189, doi: 10.1016/0031-0182(80)90055-3. Keys, J.R., and Williams, K., 1981, Origin of crystalline, cold desert salts in the McMurdo Region, Antarctica: Geochimica et Cosmochimica Acta, v. 45, p. 2299–2309, doi: 10.1016/0016-7037(81)90084-3. Lal, R., Kimble, J.M., Eswaran, H., and Stewart, B.A., eds., 2000, Pedogenic Carbonates and the Global Carbon Cycle in Global Climate Change and Pedogenic Carbonates: Boca Raton, Lewis Publishers, p. 1–14. Landi, A., Mermut, A.R., and Anderson, D.W., 2003, Origin and rate of pedogenic carbonate accumulation in Saskatchewan soils, Canada: Geoderma, v. 117, p. 143–156, doi: 10.1016/S0016-7061(03)00161-7. Lyons, W.B., Nezat, C.A., Benson, L.V., Bullen, T.D., Graham, E.Y., Kidd, J., Welch, K.A., and Thomas, J.M., 2002, Strontium isotopic signatures of the streams and lakes of Taylor Valley, southern Victoria Land, Antarctica: Chemical weathering in a polar climate: Aquatic Geochemistry, v. 8, p. 75–95, doi: 10.1023/A:1021339622515. Lyons, W.B., Welch, K.A., Carey, A.E., Wall, D.H., Virginia, R.A., Fountain, A., Doran, P.T., Csatho, B., and Tremper, C., 2005, Groundwater seeps
Calcium carbonate in Antarctic soils in Taylor Valley, Antarctica: An example of a subsurface melt event, 26–30 July 2004, in Papers from “The International Symposium on Ice and Water Interactions: Processes Across the Phase Boundary”: Portland, Oregon, Annals of Glaciology, v. 40, p. 200–206. Marchant, D.R., and Denton, G.H., 1996, Miocene and Pliocene paleoclimate of the Dry Valleys region, southern Victoria Land; a geomorphological approach: Marine Micropaleontology, v. 27, no. 1–4, p. 253–271, doi: 10.1016/0377-8398(95)00065-8. Marion, G.M., 1989, Correlation between long-term pedogenic CaCO3 formation rate and modern precipitation in desert of the American Southwest: Quaternary Research, v. 32, p. 291–295. Marion, G.M., and Schlesinger, W.H., 1994, Quantitative modeling of soil forming processes in deserts: The CALDEP and CALGYP models, in Quantitative Modeling of Soils Forming Processes: Soil Science Society of America Special Publication 39, p. 129–145. Marion, G.M., Introne, D.S., and Van Cleve, K., 1991, Stable isotope geochemistry of CaCO3 on the Tanana River floodplain of interior Alaska, U.S.A.: Composition and mechanisms of formation: Chemical Geology, v. 86, no. 2, p. 97–110. McCraw, J.D., 1967, Soils of Taylor dry valley, Victoria Land, Antarctica, with notes on soils from other locations in Victoria Land: New Zealand Journal of Geology and Geophysics, v. 10, no. 2, p. 498–539. McKelvey, B.C., and Webb, P.N., 1972, Geological investigations in southern Victoria Land, Antarctica: Part 3. Geology of Wright Valley: New Zealand Journal of Geology and Geophysics, v. 5, p. 143–162. Naiman, Z., Quade, J., and Patchett, P.J., 2000, Isotopic evidence for eolian recycling of pedogenic carbonate and variations in carbonate dust sources throughout the southwest United States: Geochimica et Cosmochimica Acta, v. 64, no. 18, p. 3099–3109, doi: 10.1016/S00167037(00)00410-5. Neumann, K., 1999, Carbon dynamic in lakes and streams of Taylor Valley, Antarctica [Ph.D. dissertation]: Tuscaloosa, Alabama, the University of Alabama, 176 p. Nezat, C.A., Lyons, W.B., and Welch, K.A., 2001, Chemical weathering in streams of a polar desert (Taylor Valley, Antarctica): Geological Society of America Bulletin, v. 113, p. 1401–1408, doi: 10.1130/00167606(2001)113<1401:CWISOA>2.0.CO;2. Nishiyama, T., and Kurasawa, H., 1975, Distribution of secondary minerals from Taylor Valley: Bulletin of the Dry Valley Drilling Project, v. 5, p. 120–133. Omelon, C.R., Marion, G.M., and Pollard, W.H., 2001, Seasonal formation of ikaite (CaCO3) 6H2O) in saline spring discharge at Expedition Fiord,
103
Canadian High Arctic; assessing conditional constraints for natural crystal growth: Geochimica et Cosmochimica Acta, v. 65, no. 9, p. 1429–1437, doi: 10.1016/S0016-7037(00)00620-7. Papadimitriou, S., Kennedy, H., Dattner, G., Dieckmann, S., and Thomas, D.N., 2004, Experimental evidence for carbonate precipitation and CO2 degassing during sea ice formation: Geochimica et Cosmochimica Acta, v. 68, no. 8, p. 1749–1761, doi: 10.1016/j.gca.2003.07.004. Pastor, J., and Bockheim, J.G., 1980, Soil development on moraines of Taylor Glacier, lower Taylor Valley, Antarctica: Soil Science Society of America Journal, v. 44, p. 341–348. Priscu, J.C., 1998, Ecosystem Dynamics in a Polar Desert: The McMurdo Dry Valleys, Antarctica: Washington, D.C., American Geophysical Union, 369 p. Quade, J., Chivas, A.R., and McCulloch, M.T., 1995, Strontium and carbon isotope tracers and the origins of soil carbonate in South Australia and Victoria: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 113, p. 103–117, doi: 10.1016/0031-0182(95)00065-T. Schlesinger, W.H., 1985, The formation of caliche in soils of the Mojave Desert, California: Geochimica et Cosmochimica Acta, v. 49, p. 57–66, doi: 10.1016/0016-7037(85)90191-7. Sletten, R.S., 1988, The formation of pedogenic carbonates on Svalbard: The influence of cold temperatures and freezing, in Fifth International Conference on Permafrost Proceedings, p. 467–472. Stuiver, M., Denton, G.H., Hughes, T.J., and Fastook, J.L., 1981, History of the marine ice sheet in West Antarctica during the last glaciation: A working hypothesis, in Denton, G.H., and Hughes, T.J., eds., The Last Great Ice Sheets: New York, Wiley-Interscience, p. 319–436. Sugden, D.E., Marchant, D.R., Potter, N.J., Souchez, R.A., Denton, G.H., Swisher, C.C., III, and Tison, J.L., 1995, Preservation of Miocene glacier ice in East Antarctica: Nature, v. 376, p. 412–414, doi: 10.1038/376412a0. Van der Hoven, S.J., and Quade, J., 2002, Tracing spatial and temporal variations in the sources of calcium in pedogenic carbonates in a semiarid environment: Geoderma, v. 108, p. 259–276, doi: 10.1016/S00167061(02)00134-9. Wedepohl, K.H., 1995, The composition of the continental crust: Geochimica et Cosmochimica Acta, v. 59, p. 1217–1232, doi: 10.1016/00167037(95)00038-2. Wellman, H.W., 1964, Later geological history of Hut Point Peninsula, Antarctica: Royal Society of New Zealand Geological Transactions, v. 2, no. 10, p. 147–154. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
Calcretes, oncolites, and lacustrine limestones in Upper Oligocene alluvial fans of the Montgat area (Catalan Coastal Ranges, Spain) David Parcerisa† Centre de Géosciences, École des Mines de Paris, 35, rue St. Honoré, 77305 Fontainebleau, France David Gómez-Gras‡ Departament de Geologia, Facultat de Ciències, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain Juan Diego Martín-Martín§ Department of Earth Sciences, Uppsala University, Villavägen 16, SE-752 36 Uppsala, Sweden
ABSTRACT The Chattian (Upper Oligocene) deposits of Montgat consist of continental detrital sediments deposited mainly in alluvial fan environments. Stratigraphic and petrographic data allow identification of two lithostratigraphic units: the Turó de Montgat unit and the Pla de la Concòrdia unit. These units are interpreted as two coalescent alluvial fans deposited synchronously. The catchment areas of these alluvial fans were located between the Collserola and Montnegre highs and consisted of a Mesozoic cover overlying a Paleozoic basement. Intrabasinal limestones interbedded in the two alluvial fan deposits of the Montgat area have been analyzed geochemically, and three groups are distinguishable. The first group is composed of the oncolites of the Turó de Montgat unit, which were formed in ponding zones in the channel or disconnected pools in a fluvial setting where waters remained under closed conditions. The second group is composed of the lacustrine limestones and the oncolites of the Pla de la Concordia unit, which were formed in an open and permanent fluvial setting. The third group is composed of the palustrine limestones and tufa-oncolites of the Pla de la Concordia unit; these were formed in ephemeral fluvial settings concurrent with development of calcrete soils. The geochemistry of the intrabasinal limestones deposited in the Chattian alluvial fans of the Montgat area is mainly controlled by the fluvial regime and the lithology and altitude of the catchment areas and the sedimentary basin. Keywords: calcretes, oncolites, lacustrine limestones, isotopes, Catalan Coastal Ranges. RESUMEN El Oligoceno superior (Catiense) de Mongat está constituido por sedimentos detríticos depositados en un ambiente de abanico aluvial. Los datos estratigráficos y E-mail:
[email protected]. E-mail:
[email protected]. § E-mail:
[email protected]. † ‡
Parcerisa, D., Gómez-Gras, D., and Martín-Martín, J.D., 2006, Calcretes, oncolites, and lacustrine limestones in Upper Oligocene alluvial fans of the Montgat area (Catalan Coastal Ranges, Spain), in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 105–117, doi: 10.1130/2006.2416(07). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
105
106
Parcerisa et al. petrográficos permiten identificar dos unidades litostratigráficas distintas: la Unidad del Turó de Montgat y la Unidad del Pla de la Concòrdia. Estas unidades han sido interpretadas como dos abanicos aluviales coalescentes que funcionaron sincrónicamente en el tiempo. Las áreas fuente de estos abanicos aluviales se situaban entre los altos de Collserola y de Montnegre y estaban constituidas por un basamento Paleozoico y una cobertera Mesozoica. En estos depósitos abundan los carbonatos intracuencales en forma de oncolitos, carbonatos lacustres y palustres y calcretas. El análisis isotópico de los distintos tipos de carbonatos intracuencales permite agruparlos en tres familias diferentes: (1) Los oncolitos presentes en los conglomerados de la Unidad del Turó de Montgat se formaron en un contexto fluvial en canales estancados o abandonados y sometidos a procesos de evaporación; (2) los carbonatos lacustres y los oncolitos de la Unidad del Pla de la Concòrdia se formaron en un contexto fluvial abierto y (3) los carbonatos palustres y los oncolitos-tobas de la Unidad del Pla de la Concòrdia se formaron en contextos fluviales efímeros conjuntamente con el desarrollo de calcretas. Así, la geoquímica de los carbonatos intracuencales formados en los abanicos aluviales del Oligoceno superior del área de Montgat está controlada por el régimen fluvial en el que se formaron y por la litología y la altitud de sus áreas fuente y de la cuenca sedimentaria. Palabras clave: calcretas, oncolitos, calizas lacustres, isótopos, Cordillera Costera Catalana.
INTRODUCTION Isotopic analysis of intrabasinal limestones has been used by many authors to study sedimentary basins (Brancaccio et al., 1986; Oberhänsli and Allen, 1987; Casanova and Nury, 1989; Janaway and Parnell, 1989; Bellanca et al., 1992; Platt, 1992; Anadón and Utrilla, 1993; Andrews et al., 1993; Zamarreño et al., 1997; Alonso-Zarza, 1999; Alonso-Zarza and Calvo, 2000). A review of the interpretation of isotopic data from lacustrine limestones and oncolites appeared in Talbot (1990); this paper remarked on the importance of intrabasinal limestones as a tool for basin analysis. Moreover, geochemical data from intrabasinal limestones can reveal significant information about the main features of the catchment areas. The paleoenvironmental significance of palustrine-lacustrine limestones and calcretes was presented by Alonso-Zarza (2003). This author underlined the importance of calcretes as indicators of paleoclimate, paleovegetation, and paleoconcentration of atmospheric pCO2 (see also Cerling, 1999). From a geochemical point of view, the analysis of lacustrine limestones and oncolites must be differentiated from the analysis of calcretes. Both groups can provide different but important information on the water that circulated through the sedimentary basin, for example: 1. Lacustrine limestones and oncolites precipitate directly or are catalyzed by biota (such as cyanobacteria, charophytes, ostracodes, mollusks) in meteoric water derived from the catchment area and delivered to the sedimentary basin. The main control on the geochemistry of meteoric water is the initial composition of this water in the catchment area (Talbot, 1990).
2.
In contrast, calcretes are mainly formed from meteoric water arriving at the sedimentary basin by in situ rain episodes. Thus, calcretes reflect the features of in situ meteoric water and its interactions with the soil (lithology, vegetation…) where calcrete grows (Alonso-Zarza, 2003).
In this paper, we analyze the geochemistry of calcretes, oncolites, and lacustrine limestones formed during Chattian times in two coalescent alluvial fans. The differences in trace-element and isotopic composition of intrabasinal limestones provide information about the main features of the alluvial fans and their respective catchment areas. MATERIALS AND METHODS A detailed (1:5000 scale) cartography was carried out in the Montgat area to determine the extent and geometry of the materials cropping out in the area. Six stratigraphic sections were measured and correlated to characterize Cenozoic materials, and 65 samples of mudstones, sandstones, limestones, and conglomerate pebbles were collected to be petrographically analyzed. Elemental composition of oncolites was determined in three polished, carbon-coated thin sections using a CAMECA model SX-50 microprobe equipped with four vertically displayed WD X-ray spectrometers. Operating conditions were an accelerating voltage of 20 kV, a beam current of 15 nA, and a spot size of 10 μm. The detection limits were 115 ppm for Mg, 380 ppm for Mn, 200 ppm for Fe, 130 ppm for Na, and 165 ppm for Sr. Several microdrillings were made in 12 samples of oncolites, calcretes, and lacustrine limestones. Powdered samples were reacted with 103% phosphoric acid for 10 min in a vacuum at 90 °C. The CO2 was analyzed using a VG-Isotech SIRA IITM mass spectrometer.
Oligocene lacustrine limestones in NE Spain Results were precise to ±0.05‰ for δ13C and ±0.09‰ for δ18O (precision was determined by multiple analyses of a standard). Results were corrected using standard procedures (Santrock et al., 1985) and are expressed in per mil with respect to the Vienna Peedee belemnite (VPDB) standard.
107
horsts (Fig. 1). Between the Collserola-Montnegre horst and the Barcelona half-graben, there is the Pla de Barcelona link zone, where the city of Barcelona is located. The Chattian deposits of Montgat crop out close to the Montnegre horst in the vicinity of a fault system that separates the Montnegre horst from the Pla de Barcelona link zone (Gaspar-Escribano et al., 2004). In fact, Montgat materials are located in the northern boundary of the Pla de Barcelona link zone (Fig. 1). The Montgat outcrops are arranged in a NW-SE fringe, which is divided in several blocks by N-S– and NW-SE–trending faults (Fig. 2). Paleozoic (Devonian dolomites and limestones, Silurian black shales, and Late Hercynian granitoids) and Triassic rocks (Buntsandstein, Muschelkalk, and Keuper) underlie these sediments.
GEOLOGICAL SETTING The structure of the Catalan Coastal Ranges is dominated by longitudinal, near-vertical basement faults that trend from NESW to ENE-WSW (Roca and Guimerà, 1992; Roca 1994). During the Alpine Paleogene compressive phase, these faults moved sinistrally with local transpression. In the course of Neogene extension, some of these faults (Vallès-Penedès and Camp faults) were reactivated as normal faults trending ENE-WSW (Roca et al., 1999). The Catalan Coastal Ranges are composed of a Hercynian basement that is unconformably overlain by a Mesozoic cover. The basement is made up of metamorphic and granitic Paleozoic rocks (Vaquer, 1973; Gil Ibarguchi and Julivert, 1988; Enrique, 1990; Julivert and Durán, 1990). The Mesozoic cover is composed of limestones, dolomites, and locally siliciclastic and evaporitic rocks that are Triassic (Virgili, 1958; Marzo and Calvet, 1985), Jurassic (Giner, 1980), and Cretaceous (Salas, 1987, 1989) in age. There are two Neogene half-grabens in the central part of the Catalan Coastal Ranges: the Vallès-Penedès half-graben, which is onshore, and the Barcelona half-graben, which is offshore. These are separated by the Garraf and the Collserola-Montnegre
STRATIGRAPHY Two different stratigraphic units separated by a reverse fault can be distinguished in the Chattian materials of Montgat (Parcerisa, 2002; Parcerisa et al., 2007). In the footwall block (southwestward) of the reverse fault, there is the Pla de la Concòrdia unit, and in the hanging-wall block (northeastward), there is the Turó de Montgat unit (Fig. 2). Turó de Montgat Unit Turó de Montgat unit overlies the Paleozoic basement and can be divided into three subunits (Fig. 3): the basal subunit, the middle subunit, and the upper subunit.
2°W
0°
2°E
4°E
Aquitaine Basin 43°N
Vallès-Penedès half-graben
RE
G
TNE
N
N MO
SE
41°N
L RO
en gr ab ha a
40°N
a
i nc le Va
80 km
Quaternary Paleogene & Neogene Mesozoic
rc
el
on
gh Montgat
ou Tr
Paleozoic
Ba
Ba
rce
lon
Montjuïc block
Pla de Barcelona
lf-
lt af au
R
GA
Montgat block
Hospital block
bo da bi ult Barcelona i T fa city
F RA
La Seu
Ebro Basin
A
LL CO
42°N
Mediterranean Sea
10 km
Late Hercynian granitoids Main emerged faults Main submerged faults
Figure 1. Geological and structural sketch of the Pla de Barcelona link zone with the location of the Montgat area.
108
Parcerisa et al. 14
N 58
Turó de les Bateries Turó del Mar 70 85
Pla de la Concòrdia
6
77
29
Turó de Ribes
56
5 74
3 35
39
51
Turó de Montgat
2 52
60
C32
1 53
4
ay highw
d
oa
Ir NI
way rail
100 m
Quaternary
Lower Muschelkalk
Pla de la Concòrdia U.
Buntsandstein
Fault Thrust 60
Turó de Montgat U.
Paleozoic basement
Keuper
Concordant contact
Syncline fold
Upper Muschelkalk
Discordant contact
Stratigraphic sections
Dip
Figure 2. Geological map of the Montgat area with the locations of the stratigraphic sections shown in Figures 3 and 4.
The basal subunit is made up of 4-m-thick massive breccia deposits. The middle subunit is 25 m thick and consists of massive, poorly sorted conglomerates with some thin interbedded sandstone layers. The upper subunit is 40 m thick and is made up of intercalations of conglomerates, gray sandstones, and red mudstones. Conglomerates are lithorudites, usually containing some intraformational oncolithic fragments (Fig. 4A). Sandstone and conglomerate grains of this unit derive from the erosion of the Mesozoic cover and the Paleozoic basement (Parcerisa, 2002). It has been interpreted that the basal subunit was deposited in a colluvial environment, the middle subunit was deposited in a proximal alluvial fan environment, and the upper subunit was deposited in a medium to distal alluvial fan environment (Parcerisa et al., 2007).
Pla de la Concòrdia Unit Pla de la Concòrdia unit unconformably overlies Triassic (Muschelkalk) and locally Devonian limestones and can be divided into three subunits (Fig. 4B). The basal subunit consists of breccia deposits and crops out discontinuously with a thickness always less than 2 m. The middle subunit is 5 m thick and only appears in one of the stratigraphic sections measured in this unit. It is made up of a 2-m-thick bed of gray marls overlain by a 3-m-thick bed of thin laminated brown limestones with plant debris, charophytes, and ostracodes. The upper subunit consists mainly of gray-colored conglomerates with some interbedded sandstone and mudstone layers. In one stratigraphic section, there are 1–1.5-m-thick beds
Oligocene lacustrine limestones in NE Spain
109 N II ROAD SECTION
3 30
RAIL-WAY TUNNEL SECTION
2
Fault
MG-13 (oncolite)
Oncolite Bioturbation Cross-stratification
20
20
Upper subunit
Conglomerate Breccia Basement
Middle subunit
BEACH QUARRY SECTION
MG-20 (oncolite) MG-19 (oncolite)
10
10
Basal s.
Meters
Middle subunit
1 10
?
cl
vfs ms vcs
cong
cl
vfs ms vcs
cong
cl
vfs ms vcs
cong
Figure 3. Stratigraphic sections carried out in the Turó de Montgat unit (see Fig. 2). cl—clay; vfs—very fine sandstone; ms—medium sandstone; vcs—very coarse sandstone; cong—conglomerate.
of gray limestones (Fig. 4B) that comprise a micritic matrix containing low amounts of limestone rock fragments (floatstones). A thin layer of micritic limestones and two conglomeratic beds consisting exclusively of oncolites (Fig. 4C) also appear in this subunit; additionally, there are some mudstone layers that contain dispersed oncolites. Conglomerates and sandstones of this unit are lithorudites and litharenites made up of limestone rock fragments derived exclusively from the erosion of the Mesozoic cover (Parcerisa, 2002). Fragments of small mammal teeth have been found in a mudstone layer of the upper subunit of the Pla de la Concòrdia unit (Fig. 5), indicating a Chattian age for Montgat deposits (Parcerisa et al., 2007). From a sedimentological point of view, it is interpreted that: (1) the basal subunit was deposited in a colluvial envi-
ronment; (2) the middle subunit was deposited in a lacustrine environment; and (3) the upper subunit was deposited in a medium to distal alluvial fan environment where conglomerates and sandstones were deposited in channels and mudstones were deposited on floodplains. The floatstones are interpreted as calcretes developed on the floodplain and the thin micritic limestone bed as a palustrine deposit. Sedimentologic and petrographic data show that Turó de Montgat and Pla de la Concòrdia units were two contemporary and attached alluvial fans with two different source areas (Parcerisa, 2002; Parcerisa et al., 2007). The Turó de Montgat alluvial fan was located eastward with Paleozoic and Mesozoic rocks in the source area, and the Pla de la Concòrdia alluvial fan was located westward, with a source area consisting exclusively of Mesozoic rocks (Fig. 6).
110
Parcerisa et al. PETROLOGY AND GEOCHEMISTRY OF INTRABASINAL LIMESTONES Intrabasinal limestones appearing in the Chattian Montgat materials include oncolites (Figs. 7A, 7B, and 7C), lacustrine and palustrine limestones (Fig. 7D), and calcretes (Figs. 7E and 7F). Oncolites appear in both the Turó de Montgat and the Pla de la Concòrdia units, whereas lacustrine and palustrine limestones and calcretes only appear in the Pla de la Concòrdia unit. Oncolites and Tufa-Oncolites Oncolites are abundant in the two units of the Chattian Montgat, especially in the upper subunits (Figs. 3 and 5). Oncolites of the Montgat area are individualized in two different petrographic categories: oncolites “sensu strictu,” which form pebbles inside conglomerates and sandstones, and tufa-oncolites, which are located exclusively inside mudstone layers of the Pla de la Concòrdia unit. Oncolites Oncolites display darkly colored tubular morphologies ranging from 3 mm to 20 cm thick (Fig. 7A1). They have a minute core surrounded by a thick cortex. The core consists of spar calcite cement or detrital sediment, which probably occupied a moldic porosity after decay of a vegetal fragment. The cortex consists of several submillimeter-scale micritic layers with microfabrics (Fig. 7B) similar to Phormidium or Calothrix/Dichothrix (Schäfer and Stapf, 1978; Casanova and Nury, 1989; Koban and Schweigert, 1993; Zamarreño et al., 1997). We infer that the oncolites precipitated from cyanobacteria in ponding zones in the channel or in disconnected pools that formed during low-discharge episodes. Thus, oncolite pebbles in conglomerate beds have probably been reworked. Geochemically, oncolites of the Turó de Montgat unit and the Pla de la Concòrdia unit are quite different. Oncolites of the Pla de la Concòrdia unit have high Fe and Sr contents (Table 1), with δ18O values ranging between −8.0‰ and −9.4‰ and δ13C values between −5.8‰ and −6.3‰ (Table 2). Oncolites of the Turó de Montgat unit have low trace-element contents (Table 1), with δ18O values ranging between −4.0‰ and −6.8‰ and δ13C between −4.9‰ and −6.6‰ (Table 2). δ18O and δ13C are arranged in a covariant line (δ13C = 0.39, δ18O −3.52 with R2 = 0.92; Fig. 8)
Figure 4. Field views and details of the intrabasinal limestones of the Montgat area. (A) Oncolite pebble inside a conglomerate layer of the upper subunit of the Turó de Montgat unit. Larger axis pebble: 3 cm. (B) Contact between Devonian limestones of the basement and the basal subunit of the Pla de la Concòrdia unit (white line). Above the basal subunit appears the upper subunit made up of calcrete deposits (black arrow). (C) Hand sample of a microconglomerate formed by intraformational grains (oncolites). Larger axis sample: 8 cm.
Tufa-Oncolites Tufa-oncolites are white-colored and form millimeter- to centimeter-scale tubular bodies (Fig. 7A2). Like oncolites, they consist of a core and a cortex; the core is also filled by spar calcite or detrital sediment, but the cortex consists of a succession of thinner micritic and thicker pseudoradial spar calcite layers (Fig. 7C). Micritic layers range from 100 μm to 1 mm thick, and spar calcite crystals are 1–3 mm long and 10–100 μm wide. Their location, inside mudstone layers, and petrographic features indicate a palustrine abiotic formation in floodplain fluvial settings
PLA DE LA CONCÒRDIA SECTION
Fault
6
Oncolite 50
Imbrication Calcrete Ripple mark Mammal fossils Vegetal debris Ostracoda Karst 40
Basement
?
30
Upper subunit
MG-46b (tufaoncolite)
MG-53b i 57 (palustrine)
RAIL-WAY TUNNEL SECTION NII ROAD SECTION
4
5
20
20
MG-t, 49 i 50 oncoites)
MG-40 (calcrete)
MG-38 (calcrete)
Middle subunit
MG-29 (lacustrine) 10 MG-28 (lacustrine)
?
Basal s.
10
MG-39 (calcrete)
?
Basal s.
Meters
20 Upper subunit
Upper subunit
10
MG-37 (oncolite)
cl
vfs ms vcs
cong
cl
vfs ms vcs
cong cl
svf ms vcs
cong
Figure 5. Stratigraphic sections carried out in the Pla de la Concordia unit (see Fig. 2). cl—clay; vfs—very fine sandstone; ms—medium sandstone; vcs—very coarse sandstone; cong—conglomerate.
112
Parcerisa et al.
Ebro Basin
lan
ta Ca
N
s nge
Ra
Montgat Oligocene alluvial fan sediments
Co
Barcelona st line
y coa nt-da Prese
al ast
Figure location
Proximal
Barcelona Basin
Medium to distal
Ephemeral lakes
Calcretes
Main channels
Undifferentiated Oligocene Undifferentiated Mesozoic Late Hercynian granitoids Devonian Turó de Montgat alluvial fan
Silurian and/or Cambro-Ordovician
Pla de la Concòrdia alluvial fan
Contact aureole Hydrographic boundary
Figure 6. Paleogeographical sketch of the Montgat area during Chattian times (location based on Roca et al., 1999).
(Pedley, 1990; Koban and Schweigert, 1993). A similar origin is attributed to the thin palustrine limestones, which occur stratigraphically close to the tufa-oncolites (Fig. 5). δ18O values of the tufa-oncolites range between −6.5‰ and −7.6‰, and δ13C values range between −6.7‰ and −6.8‰. Palustrine limestones have a similar composition, with δ18O values ranging between −7.2‰ and −8.1‰ and δ13C values ranging between −6.2‰ and −6.6‰. Lacustrine Limestones Lacustrine limestones occur exclusively in the middle subunit of the Pla de la Concòrdia unit (Fig. 5). In outcrop, they form a 3-m-thick layer of brown-colored thinly laminated limestones where every lamina measures 1–2 cm. The base and the top of these laminae are full of well-preserved charophyte stems. Petrologically, they are packstones consisting of charophyte, ostracode, and, rarely, quartz fragments inside a micritic matrix (Fig. 7D). Similar lacustrine deposits have been interpreted as the result of shallow lacustrine sedimentation (Platt, 1992; Alonso-Zarza and Calvo, 2000). Preservation of charophyte stems points to lowenergy settings with a depth less than 10 m (Platt, 1992). Isotopic data of lacustrine limestones are similar to those of the Pla de la Concòrdia oncolites; thus, δ18O values range between −8.2‰ and −9.6‰, and δ13C values range between −5.6‰ and −6.6‰.
Calcretes Calcretes are present only in a section of the Pla de la Concòrdia unit (Fig. 5). They form three 1–1.5-m-thick beds of gray limestones (Fig. 4B) formed by a micritic matrix that contains small amounts of limestone rock fragments. The micritic matrix is made up of: 1. Peloids and micropeloids. These are 10–100-μm-thick micritic spheres (Fig. 7E) typical of calcareous soils (Read, 1974; Harrison, 1977; Wright, 1994), which correspond to fungo-tufas (Calvet and Julià, 1983) or fecal pellets (Esteban and Klappa, 1983). 2. Pisoliths. These are 1–3-mm-thick globular bodies with a core and a cortex (Fig. 7F). The core consists of amalgamated peloids, and the cortex shows a submillimeter lamination. Pisoliths are also abundant in calcretes (James, 1972; Calvet and Julià, 1983; Jones, 1991; Wright, 1994). 3. Redissolution channels. These correspond to tubular cracks developed during dry episodes (Ward, 1975; Esteban and Klappa, 1983); they are filled by spar-calcite cement (Fig. 7F). 4. Detrital grains. These are relatively abundant (Fig. 7F), so it is inferred that calcretes developed on conglomeratic layers. Isotopically, calcretes have δ18O values ranging from −6.9‰ to −7.9‰ and δ13C values ranging from −6.4‰ to −7.0‰.
F
Figure 7. Hand samples and micrographs of the intrabasinal limestones of the Montgat area. (A) Oncolite (1) and tufa-oncolite (2) at hand sample. Note the core and the different layers of the cortex. Bar scale: 1 cm. Samples MG-19 (A1) and MG-46b (A2). (B) Microphotography of an oncolite with microstructures attributed to Phormidium or Calothrix/Dichothrix. Plane polarized light. Bar scale: 100 μm. Sample MG-19. (C) Micrograph of micritic layers and pseudoradial spar-calcite in a tufa-oncolite. Cross polarized light. Bar scale: 100 μm. Sample MG-46b. (D) Packstone with charophyte. Plane polarized light. Bar scale: 100 μm. Sample MG-29. (E) Peloids inside a calcrete sample. Plane polarized light. Bar scale: 100 μm. Sample MG-38. (F) Pisolith (white circle) with a redissolution channel inside (white arrows) and some detrital grains outside. Plane polarized light. Bar scale: 200 μm. Sample MG-38.
114
Parcerisa et al. TABLE 1. TRACE-ELEMENT CONTENT OF ONCOLITES OF THE TURÓ DE MONTGAT AND PLA DE LA CONCÒRDIA UNITS Sample Mg Mn Fe Sr (ppm) (ppm) (ppm) (ppm) Turó de Montgat Unit MG-13 (21) Mean 3759 bd 209 bd Min.-Max. 3644–3955 bd bd-647 bd-864 Pla de la Concòrdia Unit MG-27 (42) Mean 2585 bd 4142 1257 Min.-Max. 172–8501 bd-565 240–14853 bd-2060 MG-37 (33) Mean 2718 bd 1055 516 Min.-Max. 1621–9313 bd-766 292–3073 bd-968 Note: The number of analyses is inside parentheses; bd—below detection limit.
INTERPRETATION Microprobe data show that oncolites formed in the Pla de la Concòrdia unit have higher Fe and Sr contents than those of the Turó de Montgat unit (Table 1). The absorption of these elements from water to calcite is controlled by the distribution coefficient (McIntire, 1963), which is constant and characteristic for each element. Thus, we can deduce that the water that formed oncolites in the Pla de la Concòrdia unit was richer in Fe and Sr than the water that formed oncolites in the Turó de Montgat unit. The enrichment in Sr of the Pla de la Concòrdia water is interpreted as the result of the abundance of Sr in the Mesozoic limestones (Hem, 1970) that crop out in the catchment area of this alluvial fan. Thus, water was enriched in Sr by exchange reactions either with Mesozoic limestones occurring in the catchment area or with limestone pebbles during transport from the catchment area into the sedimentary basin or with both of them. The Fe enrichment of the Pla de la Concordia water can be explained via two mechanisms: (1) exchange reactions with Mesozoic limestones of the catchment area, and/or (2) dissolution of Fe oxides. To explain this differential Fe-oxide dissolution in the latter case, more reducing environments might have existed in the Pla de la Concordia alluvial fan than in the Turó de Montgat alluvial fan. Negative δ13C and δ18O values of intrabasinal limestones of the Montgat area (Table 2) are in agreement with precipitation from meteoric water (Cerling and Quade, 1993; Hoefs, 1997). The δ13C versus δ18O data plot in three different areas (Fig. 8): (1) oncolites and lacustrine limestones of the Pla de la Concòrdia unit are represented in area 1 (−6.6‰ < δ13C < –5.6‰ and −9.6‰ < δ18O < –8.0‰; Table 2); (2) oncolites of the Turó de Montgat unit fit a positive covariant line (area 2), with δ13C values ranging between −4.9‰ and −6.6‰ and δ18O values between −4.0‰ and −7.9‰ (Table 2); and (3) tufa-oncolites, calcretes, and palustrine limestones of the Pla de la Concordia unit are in area 3 (−7.0‰ < δ13C < –6.2‰ and −8.1‰ < δ18O < –6.5‰; Table 2). Assuming similar climatologic conditions in the area and excluding a possible biological fractionation process during calcite precipitation, differences in δ13C can be explained by: (1) closed conditions (Talbot, 1990) or (2) the presence of different types of vegetation cover in the alluvial fans or in their catch-
Na (ppm) bd bd-247 bd bd-246 bd bd-198
ment areas (Cerling, 1999). However, the covariant path of area 2 (Fig. 8) indicates that closed conditions are responsible for the observed differences. Thus, the oncolites of the Turó de Montgat unit were formed in a closed environment where meteoric water arrived from the catchment area and/or from direct precipitation in the sedimentary basin and remained in ponding zones in the channel or in disconnected pools undergoing evaporation processes. In these conditions, evaporative processes caused an isotopic fractionation of δ18O toward heavier values and reequilibration with the atmosphere, and photosynthetic processes caused an increase in δ13C values. The lowest isotopic values of the oncolites of the Turó de Montgat unit represent the initial isotopic composition in water (Talbot, 1990) and are in turn similar to δ13C values of the other intrabasinal limestones. In contrast, the absence of covariant paths in the other intrabasinal limestones indicates that they were formed in a relatively open or ephemeral environment where meteoric water did not evaporate significantly (Talbot, 1990). We infer that the heavier δ18O values of the Turó de Montgat oncolites are due to the evaporation effect. The area 1 lacustrine limestones and oncolites of the Pla de la Concòrdia unit have lighter δ18O values (typically δ18O < –8‰) than area 3 tufa-oncolites, palustrine limestones, and calcretes of the Pla de la Concòrdia unit (typically δ18O > –8‰). The lighter points of the area 2 oncolites of the Turó de Montgat unit (Fig. 8), which were less affected by evaporation, plot in an intermediate zone between areas 1 and 3. Oncolites, which are always enclosed within sandstones and conglomerates, and lacustrine deposits of area 1 were formed in permanent fluvial settings where water came from the catchment areas. In contrast, tufa-oncolites dispersed within mudstones, thin palustrine deposits, and calcretes plotted in area 3 were formed in ephemeral fluvial settings and soils developed during rain episodes in the sedimentary basin. We infer that differences in the altitude between the catchment areas and the sedimentary basin are responsible for the δ18O variations. Several works have demonstrated that a rise of 1000 m in altitude can cause a depletion of 2‰ or 3‰ in δ18O values (Poage and Chamberlain, 2001; Bowen and Wilkinson, 2002). This altitude dependence of δ18O also has been shown in the Catalan Coastal Ranges (Cruz-San Julián et al., 1992; Zamarreño et al., 1997).
Oligocene lacustrine limestones in NE Spain
TABLE 2. ISOTOPIC VALUES OF THE INTRABASINAL LIMESTONES OF THE MONTGAT AREA 18 13 Sample G C G O (‰, VPDB) (‰, VPDB) Turó de Montgat Unit Oncolites MG-19a –4.3 –5.3 MG-19b –4.5 –5.4 MG-19c –4.3 –4.9 MG-19d –4.8 –5.3 MG-19e –4.0 –5.1 MG-20a –6.8 –6.2 MG-20b –7.9 –6.6 MG-20c –5.5 –5.7 MG-20d –6.0 –5.8 MG-20e –4.8 –5.6 Pla de la Concòrdia Unit Oncolites MG-49a –9.4 –5.9 MG-49b –8.0 –6.0 MG-49c –9.0 –5.9 MG-49d –9.1 –5.9 MG-49e –8.6 –5.9 MG-50a –8.7 –6.2 MG-50b –8.7 –6.3 MG-50c –8.7 –5.8 Tufa-oncolites MG-46b-1 –6.8 –6.7 MG-46b-2 –7.6 –6.8 MG-46b-3 –6.5 –6.7 MG-46b-4 –6.6 –6.8 Lacustrine MG-28a –8.8 –6.3 MG-28b –8.9 –5.8 MG-28c –9.3 –5.6 MG-28d –9.2 –6.5 MG-28e –9.6 –6.6 MG-29a –8.3 –5.9 MG-29b –8.2 –5.9 MG-29c –9.0 –6.1 MG-29d –9.0 –6.0 MG-29e –9.1 –6.0 Palustrine MG-53b(a) –7.5 –6.4 MG-53b(b) –7.5 –6.6 MG-53b(c) –7.2 –6.6 MG-57a –7.8 –6.2 MG-57b –8.1 –6.4 MG-57c –7.7 –6.5 Calcretes MG-38a –7.1 –6.8 MG-38b –7.0 –7.0 MG-38c –7.3 –7.0 MG-38d –7.4 –7.0 MG-39a –6.8 –6.5 MG-39b –6.9 –6.5 MG-39c –7.3 –6.6 MG-40a –7.6 –6.6 MG-40b –7.0 –6.6 MG-40c –7.9 –6.4
115
Considering a difference of 2‰ between the means of the two differentiated groups (area 1 versus area 3), a rise of ~600 m might have existed between the alluvial fans and their catchment areas. On the other hand, oncolites of the Turó de Montgat unit, with intermediate δ18O values were formed in ponding zones in the channel or in disconnected pools with water arriving from the catchment area and from rain in the sedimentary basin. In sum, three groups have been distinguished geochemically in Montgat intrabasinal limestones. The first group is composed of the oncolites of the Turó de Montgat unit, which were formed in a closed fluvial setting. The second group is composed of the lacustrine limestones and the oncolites of the Pla de la Concordia unit, which were formed in an open and permanent fluvial setting. Finally, the third group is composed of the calcrete soils and palustrine limestones and tufa-oncolites of the Pla de la Concordia unit, which were formed in open ephemeral fluvial settings. The Fe and Sr enrichment observed in the oncolites of the Pla de la Concòrdia unit is attributed to exchange reactions between meteoric water and Mesozoic limestones that crop out in the catchment area. CONCLUSIONS Two coalescent alluvial fans formed during Chattian times in the Montgat area. The trace-element and isotopic compositions of intrabasinal limestones in these deposits are quite different depending on the alluvial fan in which they formed (Turó de Montgat or Pla de la Concòrdia) and depending on the kind of limestone they formed (i.e., calcretes, lacustrine-palustrine limestones, or oncolites). These differences are controlled primarily by the provenance of meteoric water, which distinguishes two groups of intrabasinal limestones: those formed by meteoric water coming from the catchment areas (oncolites and lacustrine limestones) and those formed by meteoric water coming from rain episodes inside the sedimentary basin (calcretes, palustrine limestones, and tufa-oncolites). The trace-element content of oncolites is controlled by exchange reactions with rocks that crop out in the catchment areas. δ18O values are controlled first by the different altitude of the catchment areas and the sedimentary basin; δ18O and δ13C of the oncolites of the Turó de Montgat unit are further controlled by evaporative processes that indicate a closed hydrological setting, whereas the isotopic values of the oncolites, tufa-oncolites, palustrine and lacustrine limestones, and calcretes of the El Pla de la Concordia unit suggest an open hydrological setting. ACKNOWLEDGMENTS This research was performed within the framework of projects BTE2002-04453-C02-01 CGL2004-05816-C02-02 supported by Dirección General de Enseñanza Superior e Investigación Científica of Spain and Grup Consolidat de Recerca “Geologia Sedimentària” 2005/SGR/890. D. Parcerisa benefited from a postdoctoral grant (EX-2003-1146 Ministerio de Educación,
116
Parcerisa et al.
Turó de Montgat U. Oncolites Oncolites Tufa-oncolites Pla de la Lacustrine limestones Concòrdia U. Palustrine limestones Calcretes
13
2
C
1 18
-10
-8
-6
-4
2
-2
O
4
-1 -2 -3
Figure 8. δ18O versus δ13C plot of the intrabasinal limestones of the Montgat area.
-4
2 y = 0.39x -3.52 R2 = 0.92
1 3
-5 -6 -7
Cultura y Deporte). We thank Ana M. Alonso-Zarza who encouraged us to write this paper. Ana Travé, Gabriel Bowen, and Lawrence Tanner strongly contributed to the improvement of the manuscript with their constructive and helpful comments. Eva Coca, Miguel Angel Caja, and Joaquim Perona provided indispensable technical support to realize this work. We also thank Frances Luttikhuizen and Serdar Korkmaz for the revision of the English version. The field work of this paper was done together with Francesc Calvet, who abruptly passed away; this paper is dedicated to him. REFERENCES CITED Alonso-Zarza, A., 1999, Initial stages of laminar calcrete formation by roots: Examples from the Neogene of central Spain: Sedimentary Geology, v. 126, p. 177–191, doi: 10.1016/S0037-0738(99)00039-1. Alonso-Zarza, A., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: Earth-Science Reviews, v. 60, no. 3–4, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Alonso-Zarza, A., and Calvo, J.P., 2000, Palustrine sedimentation in an episodically subsiding basin: The Miocene of the northern Teruel Graben (Spain): Palaeogeography, Palaeoclimatology, Palaeoecology, v. 160, p. 1–21, doi: 10.1016/S0031-0182(00)00041-9. Anadón, P., and Utrilla, R., 1993, Sedimentology and isotope geochemistry of lacustrine carbonates of the Oligocene Campins Basin, north-east Spain: Sedimentology, v. 40, p. 699–720. Andrews, J.E., Riding, R., and Dennis, P.F., 1993, Stable isotopic compositions of Recent freshwater cyanobacterial carbonates from the British Isles: Local and regional environmental controls: Sedimentology, v. 40, p. 303–314. Bellanca, A., Calvo, J.P., Censi, P., Neri, R., and Pozo, M., 1992, Recognition of lake-level changes in Miocene lacustrine units, Madrid Basin, Spain: Evidence from facies analysis, isotope geochemistry and clay
mineralogy: Sedimentary Geology, v. 76, p. 135–153, doi: 10.1016/00370738(92)90080-B. Bowen, G.J., and Wilkinson, B., 2002, Spatial distribution of 18O in meteoric precipitation: Geology, v. 30, no. 4, p. 315–318, doi: 10.1130/00917613(2002)030<0315:SDOOIM>2.0.CO;2. Brancaccio, L., D’Argenio, B., Ferreri, V., Stanzione, D., Turi, B., and Preite Martínez, M., 1986, Cartteri tessiturale e geochimici dei travertini di Rocchetta a Volturno (Molise): Bolletin de la Societa Geologica Italiana, v. 105, p. 265–277. Calvet, F., and Julià, R., 1983, Pisoids in the caliche profiles of Tarragona (NE Spain), in Peryt, T.M., ed., Coated Grains: Berlin, Springer-Verlag, p. 73–79. Casanova, J., and Nury, D., 1989, Biosédimentologie des stromatolites fluvio-lacustres du fossé Oligocène de Marseille: Bulletin de la Société Géologique de France, v. V, no. 6, p. 1173–1184. Cerling, T.E., 1999, Stable carbon isotope in paleosol carbonates, in Thiry, M., and Simon-Coinçon, R., eds., Palaeoweathering, Palaeosurfaces and Related Continental Deposits: Oxford, Blackwell Science, Special Publications of the International Association of Sedimentologists, v. 27, p. 43–60. Cerling, T.E., and Quade, J., 1993, Stable carbon and oxygen isotopes in soil carbonates, in Swart, P., McKenzie, J.A., and Lohmann, K.C., eds., Continental Indicators of Climate: Proceedings of Chapman Conference, Jackson Hole, Wyoming: American Geophysical Union Geophysical Monograph 78, p. 217–231. Cruz-San Julián, J., Araguas, L., Rozanski, K., Benavente, J., Cardenal, J., Hidalgo, M.C., García-López, S., Martínez-Garrido, J.C., Moral, F., and Olias, M., 1992, Sources of precipitation over south-eastern Spain and groundwater recharge: An isotopic study: Tellus, v. 44(B), p. 226–236. Enrique, P., 1990, The Hercynian intrusive rocks of the Catalonian Coastal Ranges (NE Spain): Acta Geológica Hispánica, v. 25, p. 39–64. Esteban, M., and Klappa, C.F., 1983, Subaerial exposure environments, in Scholle, P.A., et al., eds., Carbonate Depositional Environments: American Association of Petroleum Geologists Memoir 33, p. 1–96. Gaspar-Escribano, J.M., García-Castellanos, D., Roca, E., and Cloetingh, S., 2004, Cenozoic vertical motions of the Catalan Coastal Ranges (NE Spain): The role of tectonics, isostasy, and surface transport: Tectonics,
Oligocene lacustrine limestones in NE Spain v. 23(1), 1004, doi: 10.1029/2003TC001511. Gil Ibarguchi, J.I., and Julivert, M., 1988, Petrología de la aureola metamórfica de la granodiorita de Barcelona en la Sierra de Collserola (Tibidabo): Estudios Geológicos, v. 44, p. 353–374. Giner, J., 1980, Estudio sedimentológico y diagenético de las formaciones carbonatadas del Jurásico de los Catalánides, Maestrazgo y rama aragonesa de la Cordillera Ibérica (sector oriental) [Ph.D. thesis]: Barcelona, Universitat de Barcelona, 315 p. Harrison, R.S., 1977, Caliche profiles: Indicators of near-surface subaerial diagenesis, Barbados, West Indies: Bulletin of Canadian Petroleum Geologists, v. 25, no. 1, p. 123–173. Hem, J., 1970, Study and Interpretation of the Chemical Characteristics of Natural Water (second edition): Geological Survey Water-Supply Paper 1473: Washington, U.S. Government Printing Office, 363 p. Hoefs, J., 1997, Stable isotope geochemistry: Berlin, Springer-Verlag, 201 p. James, N.P., 1972, Holocene and Pleistocene calcareous crust (caliche) profiles: Criteria for subaerial exposure: Journal of Sedimentary Petrology, v. 42, p. 817–836. Janaway, T.M., and Parnell, J., 1989, Carbonate production within the Orcadian Basin, northern Scotland: A petrographic and geochemical study: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 70, p. 89–105, doi: 10.1016/0031-0182(89)90082-5. Jones, B., 1991, Genesis of terrestrial oncoids, Cayman Islands, British West Indies: Canadian Journal of Earth Sciences, v. 28, p. 382–397. Julivert, M., and Durán, H., 1990, Paleozoic stratigraphy of the central northern part of the Catalonian Coastal Ranges (NE Spain): Acta Geológica Hispánica, v. 25, p. 3–12. Koban, C.G., and Schweigert, G., 1993, Microbial origin of travertine fabrics— Two examples from southern Germany (Pleistocene Stuttgart travertines and Miocene Riedöschingen travertine): Facies, v. 29, p. 251–264. Marzo, M., and Calvet, F., 1985, Guía de la excursión: Triásico de los Catalánides: II coloquio de Estratigrafía y Paleogeografía del Pérmico y Triásico de España: Institut d’Estudis Ilerdencs, La Seu d’Urgell, 57 p. McIntire, W.L., 1963, Trace element partition coefficients—A review of theory and applications to geology: Geochimica et Cosmochimica Acta, v. 27, p. 1209–1264, doi: 10.1016/0016-7037(63)90049-8. Oberhänsli, H., and Allen, P.A., 1987, Stable isotopic signatures of Tertiary lake carbonates, Eastern Ebro Basin, Spain: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 60, p. 59–75, doi: 10.1016/0031-0182(87)90024-1. Parcerisa, D., 2002, Petrologia i diagènesi en sediments de l’Oligocè superior i del Miocè inferior i mitjà de la Depressió del Vallés i del Pla de Barcelona: Evolució de l’àrea font i dinàmica dels fluids [Ph.D. thesis]: Bellaterra, Universitat Autònoma de Barcelona, 288 p. Parcerisa, D., Gómez-Gras, D., Roca, E., Madurell, J., and Agustí, J., 2007, The Upper Oligocene of Montgat (Catalan Coastal Ranges, Spain): New constraints to the age of the opening of the western Mediterranean Basin: Geologica Acta (in press). Pedley, H.M., 1990, Classification and environmental models of cool freshwater tufas: Sedimentary Geology, v. 68, p. 143–154, doi: 10.1016/00370738(90)90124-C. Platt, N.H., 1992, Fresh-water carbonates from the Lower Freshwater Molasse (Oligocene, western Switzerland): Sedimentology and stable isotopes: Sed-
117
imentary Geology, v. 78, p. 81–99, doi: 10.1016/0037-0738(92)90114-7. Poage, M.A., and Chamberlain, C.P., 2001, Empirical relationships between elevation and the stable isotope composition of precipitation and surface waters: Considerations for studies of paleoelevation change: American Journal of Science, v. 301, no. 1, p. 1–15. Read, J.F., 1974, Calcrete deposits and Quaternary sediments, Edel Province, Shark Bay, Western Australia: American Association of Petroleum Geologists (AAPG) Memoir, v. 22, p. 250–282. Roca, E., 1994, La evolución geodinámica de la Cuenca Catalano-Balear y áreas adyacentes desde el Mesozoico hasta la actualidad: Acta Geológica Hispánica, v. 29, p. 3–25. Roca, E., and Guimerà, J., 1992, The Neogene structure of the eastern Iberian margin: Structural constraints on the crustal evolution of the Valencia trough (western Mediterranean): Tectonophysics, v. 203, p. 203–218, doi: 10.1016/0040-1951(92)90224-T. Roca, E., Sans, M., Cabrera, L., and Marzo, M., 1999, Oligocene to middle Miocene evolution of the central Catalan margin (northwestern Mediterranean): Tectonophysics, v. 315, p. 209–233, doi: 10.1016/S00401951(99)00289-9. Salas, R., 1987, El Malm i el Cretaci inferior entre el Massís del Garraf i la Serra d’Espadà [PhD thesis]: Barcelona, Universitat de Barcelona, 345 p. Salas, R., 1989, Evolución estratigráfica secuencial y tipos de plataformas de carbonatos del intervalo Oxfordiense-Barriesiense en las cordilleras Ibérica oriental y Costero Catalana meridional: Cuadernos de Geología Ibérica, v. 13, p. 121–157. Santrock, J., Studley, S.A., and Hayes, J.M., 1985, Isotopic analyses based on the mass spectrum of carbon dioxide: Analytical Chemistry, v. 57, p. 1444–1448, doi: 10.1021/ac00284a060. Schäfer, A., and Stapf, K.R.G., 1978, Permian Saar-Nahe Basin and Recent Lake Constance (Germany): Two environments of lacustrine algal carbonates, in Matter, A., et al., eds., Modern and Ancient Lake Sediments: Special Publications of the International Association of Sedimentologists, v. 2, p. 83–107. Talbot, M.R., 1990, A review of the palaeohydrological interpretation of carbon and oxygen isotopic ratios in primary lacustrine carbonates: Chemical Geology, v. 80, p. 261–279. Vaquer, R., 1973, El metamorfismo y las rocas plutónicas y filonianas de la Sierra de Collcerola (Tibidabo), Barcelona [Ph.D. thesis]: Barcelona, Universidat de Barcelona, 362 p. Virgili, C., 1958, El Triásico de los Catalánides: Boletín del Instituto Geológico y Minero de España, v. LXIX, 856 p. Ward, W.C., 1975, Petrology and diagenesis of carbonate eolianites of northeastern Yucatán Peninsula, Mexico, in Wantland, K.F., and Pusey, W.C., eds., Belize Shelf—Carbonate Sediments, Clastic Sediments and Ecology: Studies in Geology, v. 2, p. 500–571. Wright, V.P., 1994, Paleosols in shallow marine carbonate sequences: Earth-Science Reviews, v. 35, p. 367–395, doi: 10.1016/0012-8252(94)90002-7. Zamarreño, I., Anadón, P., and Utrilla, R., 1997, Sedimentology and isotopic composition of Upper Palaeocene to Eocene non-marine stromatolites, eastern Ebro Basin: NE Spain: Sedimentology, v. 44, p. 159–176. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
The role of clastic sediment influx in the formation of calcrete and palustrine facies: A response to paleographic and climatic conditions in the southeastern Tertiary Duero basin (northern Spain) Ildefonso Armenteros† Pedro Huerta‡ Departamento de Geología, Facultad de Ciencias, Universidad de Salamanca, 37071 Salamanca, Spain ABSTRACT During the Middle and Upper Miocene, calcrete and associated palustrine deposits formed marginal fringes adjacent to the margins of the Aranda–Burgo de Osma corridor in the southeastern Tertiary Duero basin. These environments interfingered laterally with narrow peripheral alluvial fans toward the corridor margins, whereas toward the center of the corridor, they graded into the fluvial systems transverse to the alluvial fans. Over time, the peripheral carbonate environments were replaced by fluvial systems. The calcretes form profiles with nodular grading upward to massive horizons. These profiles may be vertically stacked at the edges of the carbonate bodies. These calcretes are the product of mixed pedogenic and phreatic processes associated with the palustrine environments. Palustrine limestones were deposited in a shallow carbonate-precipitating lake that had low gradient margins and was subjected to periodic fluctuations in level. The sedimentologic characteristics of the carbonate facies indicate accumulation in a semiarid climate and conditions of scarce clastic sediment supply, which favored the development of carbonate-precipitating fringes. In contrast, their absence in parts of the sequence may have resulted from an increase in clastic sediment supply associated with a climatic change toward more humid conditions. At these times, the fluvial channels had greater lateral mobility and spread toward the corridor flanks, replacing the carbonate environments. Subsidence was greater in the central corridor than at its margins and did not change significantly during the Miocene. Thus, changes in climate and the clastic sediment input on the flanks of the Aranda–Burgo de Osma corridor were the main controls on the development of 10–20-m-thick carbonate clastic sediment sequences. Keywords: palustrine, calcrete, climate, sediment influx, Miocene, Duero basin. RESUMEN Durante el Mioceno medio y superior en los márgenes del corredor de Aranda-Burgo de Osma, sureste de la cuenca del Duero, se desarrollaron franjas de depósitos palustres, E-mail:
[email protected]. E-mail:
[email protected].
† ‡
Armenteros, I., and Huerta, P., 2006, The role of clastic sediment influx in the formation of calcrete and palustrine facies: A response to paleographic and climatic conditions in the southeastern Tertiary Duero basin (northern Spain), in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 119–132, doi: 10.1130/2006.2416(08). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
119
120
Armenteros and Huerta rodeados de niveles de caliche. Estos términos carbonatados cambian lateralmente a depósitos de abanicos aluviales enraizados en los relieves que flanquean el corredor, mientras que hacia el centro del mismo pasan a los depósitos del sistema fluvial axial que drenaba esta región hacia el centro de la cuenca, situado más al oeste. Temporalmente los caliches y medios palustres eran substituidos por el sistema fluvial. Los caliches forman perfiles constituidos por niveles nodulares en la base que pasan a masivos hacia el techo, los cuales pueden llegar a apilarse formando secuencias compuestas en los márgenes de las unidades carbonatadas. Los caliches se forman debido a la acción combinada de procesos edáficos y freáticos, asociados a los medios palustres adyacentes. Estos últimos representan una sedimentación carbonatada lacustre en lagos muy someros, de baja pendiente, y afectados por exposiciones subaéreas frecuentes. Las características sedimentológicas de las facies carbonatadas indican una acumulación en condiciones semiáridas que favorecían un escaso suministro de material terrígeno a la cuenca y el desarrollo de facies carbonatadas en sus márgenes. Por el contrario, la expansión del sistema fluvial axial hacia estas áreas estaba propiciada por un aumento del suministro terrígeno como respuesta a condiciones climáticas más húmedas. En estos periodos el sistema fluvial axial se extendía hasta los márgenes del corredor, reemplazando los medios palustres y anulando en buena medida el desarrollo de los caliches asociados. La subsidencia era mayor en el corredor central que en sus márgenes y no cambió substancialmente durante el Mioceno. Los cambios en el clima y el suministro de materiales terrígenos en los flancos del corredor fueron los controles principales en la acumulación de ciclos carbonatado-terrígenos. Palabras claves: palustre, caliche, clima, aporte sedimentario, Mioceno, cuenca del Duero.
INTRODUCTION The study of palustrine-calcrete carbonate associations in different depositional contexts has attracted increasing attention in recent years, mainly with regard to their sedimentological, paleopedogenic, and diagenetic aspects (Nickel, 1982; Arribas, 1986; Platt and Wright, 1992; Alonso-Zarza et al., 1992; Sanz et al., 1995; Armenteros and Daley, 1998; Dunagan and Driese, 1999; Tanner, 2000; Tandon and Andrews, 2001; Pimentel, 2002; and many others). Since the 1960s, it has been suggested repeatedly that lacustrine successions reflect changes in both climate and tectonics (for a review, see Van Houten, 1964; Picard and High, 1981; Talbot and Allen, 1996). In the early 1980s, after the introduction of the concepts of sequence stratigraphy, allocyclic processes, climate and tectonics in particular, began to be applied to soil development (Atkinson, 1986; Kraus and Bown, 1986). Even today, it is difficult to differentiate between the effects of climate and tectonics in controlling the features of calcrete and palustrine facies. Thus, recent research that has provided new integrated interpretations of lacustrine sequences is of great use in the analysis of basins (Platt, 1989; Cecil, 1990; Sanz et al., 1995; Drummond et al., 1996; Armenteros et al., 1997; De Wet et al., 1998; Carroll and Bohacs, 1999; Tanner, 2000; Dunagan and Turner, 2004), and the authors of these works have offered interesting proposals for integrating lacustrine sequences within general allocyclic models. In this context, the model of Cecil (1990) concludes that in most sedimentary systems, climate is a primary allogenic controller of sediment supply. A more inte-
grated vision of the allocyclic control of lacustrine basins was proposed by Carroll and Bohacs (1999), who compared the rates of potential accommodation (controlled mainly by tectonics) with sediment plus water supply (controlled mainly by climate). Recently, Alonso-Zarza (2003) emphasized the interplay between climate and tectonics and described two scenarios of accommodation space and/or sequence stratigraphy: (1) lowactivity alluvial/fluvial systems, favoring vertically stacked palustrine carbonates; and (2) alternating alluvial-fluvial and pond systems, a situation common in stages of high-accommodation space that favors the storage of floodplain sediments. In a number of nonmarine basins, carbonate bodies made up of calcretes and palustrine facies are interbedded with clastic successions. Calcretes and palustrine facies are representative of equilibrium conditions in floodplains, allowing soil development between situations of aggradation and degradation (Kraus and Bown, 1986). These carbonate facies are commonly associated with a low clastic input that has been attributed both to tectonic and climatic influence (e.g., semiarid conditions) (Nickel, 1985; Platt, 1989; Cecil, 1990; Drummond et al., 1996; De Wet et al., 1998). In the Miocene succession of the eastern Duero basin, the formation of calcrete and palustrine deposits alternated over time with deposition by siliciclastic fluvial systems. Calcretes and palustrine facies developed in a specific paleogeographical setting along the margins of the Aranda–Burgo de Osma corridor in the southeastern Duero basin. These carbonate-producing environments grade laterally to narrow peripheral alluvial fan
Clastic sediment influx and formation of calcretes and palustrine facies systems on both borders, whereas toward the center of the corridor, they change into an axial fluvial system. Cyclically, these peripheral carbonates did not develop, but instead were replaced by alluvial and fluvial systems (Armenteros, 1986; Huerta and Armenteros, 2003). This study involves sedimentological analyses of the facies associations and the whole depositional system, with special emphasis on the characterization of the carbonate environments and on the genetic relationship between palustrine facies and calcretes. The aim of this contribution is to underscore the importance of sediment influx in the formation of calcrete and
121
palustrine facies along the basin margins in the Miocene succession of the southeastern Duero basin. A further aim is to establish the relative roles of climate and tectonics on the control of the sediment supply in the location of formation of the carbonate bodies and in creating the carbonate and siliciclastic cycles. GEOLOGICAL SETTING The Tertiary Duero basin is the largest continental Tertiary basin of the Iberian Peninsula and is located in the northwest of Spain (Fig. 1). It features sedimentary deposits ranging in age
Figure 1. Map of the eastern Duero basin where the Aranda–Burgo de Osma corridor is located between the paleorelief of Honrubia, to the south, and that of Tejada, to the north. On the right: representative general succession of the Miocene, which is marked on the map with a star; symbols are explained in Figure 4. The cross sections represented in Figure 2 are indicated by black lines (A and B).
122
Armenteros and Huerta
from Paleogene to Miocene. The present study focuses on the Miocene deposits of the southeastern area, near the Aranda– Burgo de Osma corridor, which links the Almazán and the Duero basins. During the Neogene, the sedimentary framework was characterized by alluvial fan systems along the basin margins and carbonate and gypsiferous systems toward the center (which is located west of the study area). The northern basin margin is formed by the Tejada anticline, which consists mainly of Upper Cretaceous limestones, dolostones, and marls. The Lower Cretaceous Utrillas Formation, formed of sandstones, mudstones, and quartzite conglomerates, outcrops in the core of the anticline. This anticline plunges westward, and its relief decreases in the same direction. The southern margin is composed of Mesozoic limestones, dolostones, marls, and evaporites covering a Paleozoic basement, and it is made up of gneisses, quartzites, and slates. It constitutes an anticline bounded to the north by a thrust, which was overlapped by the Middle and Upper Miocene succession (De Vicente et al., 2004). Tectonic deformation during the Middle and Upper Miocene was minor, and no significant activity occurred in the source areas for the sediments of the Aranda–Burgo de Osma corridor. The position of a fluvial system at the center of the corridor and the considerable thickness of the Tertiary deposits in this area suggest active subsidence during the Miocene. The age of the outcropping Miocene succession ranges from middle Aragonian (MN 5, Mammal Neogene zone) to
Upper Vallesian (MN 10), with a possible presence of the latest (Turolian) continental stage of the Miocene (Armenteros et al., 2002). The carbonate units consist of calcretes and lacustrine/ palustrine limestones and form extensive levels (up to 20 m thick) at the transition to the marginal alluvial systems (2–10 km long) (Fig. 2). The Middle and Upper Miocene deposits of the eastern succession of the Duero basin consist of two siliciclastic-carbonate sequences capped by two extensive carbonate horizons, the Lower Páramo Limestone and the Upper Páramo Limestone, which extend toward the center of the Duero basin (Figs. 1 and 2) (Armenteros et al., 2002). This study examines the Upper Páramo near the northern border of the basin, in the Tejada anticline, and the Lower Páramo near the southern margin (Honrubia). In the Aranda–Burgo de Osma corridor, the limestones occur as fringes near the northern and southern basin margins, separated in the central area by the siliciclastic fluvial system that has paleocurrent directions westward toward the basin center (Fig. 3). SEDIMENTARY ENVIRONMENTS AND FACIES Three environments are recognized in the Aranda–Burgo de Osma corridor: alluvial fans rooted in the northern and southern paleoreliefs, fluvial systems in the axis of the corridor, and shallow carbonate-precipitating lakes with associated calcretes (Figs. 2, 3A, and 3C).
Figure 2. Cross-sections A and B from Figure 1. (A) Tejada: northern border for the Upper Miocene sequence. (B) Honrubia: southern border for the Lower Miocene sequence.
Clastic sediment influx and formation of calcretes and palustrine facies
123
Figure 3. Paleogeography of the Aranda–Burgo de Osma corridor. (A) Schematic map of the development of the calcrete and palustrine fringes. (B) Schematic map of the expansion of the fluvial system. (C) Schematic cross section (exaggerated vertical scale) perpendicular to the Aranda– Burgo de Osma corridor, between the Honrubia and Tejada paleoreliefs. Contours are in m.
Alluvial Fans Marginal alluvial fan systems consist of the following facies, from proximal to distal areas, respectively: reddishorange petromictic conglomerates (quartz and quartzite clasts from Paleozoic and Mesozoic, carbonate Mesozoic clasts), muddy sandstones, sandy mudstones, and calcretes (both nodular and massive). Coarser facies are poorly to moderately sorted and exhibit scarce or no sedimentary structures. Mudstone facies rarely show lamination, are intensively burrowed, and display scattered calcite nodules and rhizoliths, occasion-
ally coalesced into nodular calcrete beds. Some levels contain green mottling and root traces. The poorly stratified conglomerates were deposited by debris flows at the proximal area of the fan. Conglomerate deposits with a sheet geometry, moderate sorting, normal grading, and horizontal stratification show characteristics of deposition by sheetflood, possibly associated with low-topography gravel bars in the middle and distal fan environment (Dabrio et al., 1989; Huerta and Armenteros, 2003). Sedimentation was episodic, as indicated by common features of subaerial exposure. Between the episodic depositional events on the fan, the
124
Armenteros and Huerta
sediments were reworked by streams entrenched on the fan surface that deposited stream channel-fill sediments (Bull, 1972), and fine floodplain facies where the flow died due to dispersion and/or infiltration. These latter deposits are commonly replaced partially by carbonate and locally form stacked calcrete sequences, as explained in the following sections. The restricted extent of the alluvial fan fringes indicates a small catchment area (Fig. 3). Fluvial Environments The fluvial system is best developed and most extensive in the central corridor (Fig. 3). The fluvial system is composed of sandy channel fills, 2–5 m thick and 20–150 m wide, that locally include conglomeratic lenses with a preponderance of quartz and quartzite clasts (the centile thickness is generally less than 5 cm). Some channel fills contain oncoids, and mudstone and calcrete clasts. In the central parts of the corridor, the channel fills are
ribbon-like in shape and have low width/thickness (w/t) ratios, while toward the corridor margins, they are sheet-shaped bodies with higher w/t ratios. These channeled bodies are enclosed in massive mudstones and sandy mudstones that display abundant bioturbation and widespread reddening (Fig. 4). These mudstones also contain extensive single-to-composite calcrete levels 0.5–3 m thick; these are more common in the transition to carbonate units described in the following sections. Coarse-grained channel fills and finegrained facies form fining-upward sequences that are bounded by erosive surfaces and are commonly capped by calcretes. Channel cross-bed paleocurrents display a strong westward component, perpendicular to those of the alluvial fans. Calcretes Calcrete occurs in profiles from 1 to 3 m thick. The carbonate content in the calcrete increases upward; typical profiles
Figure 4. Representative stratigraphic sections related to the southern part (Milagros) and to the northern part (Cubillas and Espinosa) of the Aranda–Burgo de Osma corridor. The former corresponds to the Lower Miocene sequence and displays an alternation of carbonate (see detailed sequence of central unit) and fluvial siliciclastic units. The Cubillas and Espinosa sections correspond to the Upper Miocene sequence; the first shows several superimposed palustrine sequences, whereas the Espinosa section displays a stacking of thin calcrete sequences. Sections locations are shown in Figure 2.
Clastic sediment influx and formation of calcretes and palustrine facies display basal red or reddish-brown mudstones with diffuse calcification, grading upward to nodular calcrete facies, capped by massive calcretes (Figs. 4, 5C, and 5D). These calcretes typically represent the transition between siliciclastic and palustrine facies (Figs. 5C and 5D). The nodular calcrete facies are mottled and reddish brown-orange in color and are characterized by coalescing irregular-to-vertically elongated nodules (rhizocretions). The nodules become larger and coalesce upward within the beds, and the relative proportion of mudstone decreases. Within the massive calcrete facies, the mudstones are partially calcified. The only vestiges of the mudstones at the top of the massive calcrete facies are iron oxides. Pseudoanticline structures occur at the top of the calcrete beds, where they grade into the overlying palustrine facies. The massive calcrete makes up a calcite microsparite mosaic (5–12 μm in size), locally stained by manganese oxides. Detrital quartz grains range from 5% to 20% and show etched contours surrounded by a microsparite rim (Fig. 6A). Both peloids and pedogenic features, such as channel and alveolar porosity, are common in the calcrete facies. These facies are interpreted as mixed pedogenic and phreatic calcretes. The phreatic features, such as mottling and manganese staining, together with the lateral gradation to the palustrine facies indicate a water table close to surface. Mapping of these facies and the correlation among different sections suggest that the calcrete facies formed a fringe around the palustrine deposits (Huerta and Armenteros, 2005).
125
CONTROLS ON CARBONATE DEPOSITION: CLIMATE VERSUS TECTONISM The carbonate bodies occur as calcrete and palustrine accumulations (2–20 m thick) near the basin margins and alternate vertically with the siliciclastic horizons that resulted from the expansion of the axial fluvial system (Figs. 2, 5A, and 5B). The Upper Páramo and the Lower Páramo limestones represent expansion stages of the carbonate system toward the center of the Duero basin, located westward. Their location and stratigraphic distribution require some discussion. Initially, tectonic and climatic changes might seem to have been the main agents responsible for controlling carbonate precipitation in marginal fringes close to the basin borders. Recently, this problem was considered in a study focused on calcrete palustrine assemblages in the northern Tejada area (Huerta and Armenteros, 2005). Here, we propose local subsidence as the main factor involved in the formation of these carbonate bodies. Nevertheless, the regional paleogeographical configuration, cyclical arrangement of the carbonate and siliciclastic levels, as well as the distribution of the facies can contribute to a broader perspective. Along the Aranda–Burgo de Osma corridor, an almost permanent fluvial trunk system drained the basin toward the west during most of the Miocene (Armenteros, 1986; Armenteros et al., 2002) (Figs. 1 and 3). Calcrete-Palustrine Sequences and Clastic Input: Climate Control
Palustrine Facies The palustrine facies occur in tabular beds (10 cm to 1 m thick) that are light to dark gray in color (Figs. 4, 5C, and 5D). The fossil content includes pulmonate gastropods, ostracodes, and charophytes. The percentage of dispersed fine-grained sand to silt-sized quartz is generally between 1% and 3%. The palustrine facies constitute the most abundant facies in the carbonate bodies. The palustrine limestones are micritic and show a wide range of exposure features, including brecciated, clotted, and peloidal fabrics (Figs. 6B, 6C, and 6D). These are closely associated with channel, planar, and vesicular porosities and alveolar structures related to root activity and drying-wetting cycles (Figs. 6B and 6C). The clotted-peloidal textures are the most common of those found in the palustrine facies and represent the most evolved status of the exposure index (Wright and Platt, 1995; Armenteros and Daley, 1998). The peloids commonly have no coatings, are subequant and subrounded, and their sizes range from 60 mm to 5 mm and display no sorting (Fig. 6D). These facies are the result of repeated exposure of carbonate muds that accumulated in a shallow carbonate lake with low-gradient margins (Freytet and Plaziat, 1982; Platt and Wright, 1991). The ubiquity of palustrine facies indicates that the whole lake was subject to periodic lake-level fluctuations and exposure of the supralittoral carbonate mud.
The profile sequences of mudstone–nodular calcrete–massive calcrete indicate a decrease in fine clastic input and low floodplain aggradation, allowing more time for the development of massive calcretes (Bown and Kraus, 1987; Alonso-Zarza et al., 1992; Sanz et al., 1995; Wright and Marriot, 1996; AlonsoZarza et al., 1999; Huerta and Armenteros, 2005). Therefore, this facies sequence indicates low aggradation of the floodplain. Palustrine facies occur at the top of the calcrete sequences. The low percentages of detrital quartz in palustrine facies, the lateral gradation of a calcrete belt into palustrine environments, and the absence of deltaic deposits suggest that the groundwater table played an important role in calcrete-palustrine sequence development (Huerta and Armenteros, 2003, 2005). Thus, the formation of calcretes could have taken place by both pedogenic and phreatic carbonate precipitation in the plain—calcrete belt—encircling the palustrine environments. Massive calcrete and palustrine facies are separated by sharp contacts, or gradational contacts with features of both calcrete and palustrine facies. Carbonate precipitation may have been favored by a progressive rise of the groundwater table and a consequent increase in the evaporation and evapotranspiration of the pore waters. The gradual transition from brown-reddish mudstones to nodular to massive calcrete resulted from an upward decrease in the sedimentation rate (Bown and Kraus, 1987; Alonso-Zarza et al., 1992, 1999; Sanz et al., 1995, Wright and Marriot, 1996). If the clastic sediment
Figure 5. (A) Panoramic view of the lower sequence in the Riaza River valley. The lower sequence is capped by the Lower Páramo Limestone. Arrows indicate the carbonate-siliciclastic cycles. The dashed lines mark the top of the carbonate bodies. (B) Panoramic view of the cliffs on the right margin of the Riaza River. The lower sequence is capped by the Lower Páramo Limestone. Dashed line points to a carbonate body that wedges out laterally into the siliciclastic fluvial unit. (C) Calcrete-palustrine sequence. Note that the nodular calcrete gradually passes upward into a massive calcrete. The palustrine facies overlies the massive calcrete, with a sharp boundary between them (white line). The dashed line represents a bedding plane. Hammer for scale: 33 cm long (encircled). (D) Detail of a carbonate body. The lower half is composed by calcrete facies, while the upper half is made up of palustrine facies. The boundary between the calcrete and palustrine facies is sharp (black line). Dashed white lines depict bedding planes. Geologist for scale: 1.70 m tall.
Clastic sediment influx and formation of calcretes and palustrine facies
127
Figure 6. (A) Microtexture of a massive calcrete consisting of a micrite mosaic with dispersed silt- to sand-sized quartz (Q) grains with a rim of sparite. Dark patches retain clayey remains, and the porosity displays circumgranular and elongate sinuous voids originated by cracking. Plane polarized light (PPL). (B) Brecciated palustrine texture constituted by interconnected crack planes filled with sparite cement (C). Arrow indicates allotic nodule. PPL. (C) Typical channel palustrine texture in which concertina-like voids (arrowed) and vesicles are widespread. PPL. (D) Peloidal palustrine texture mainly made up of irregularly shaped and poorly sorted micrite peloids (P), with packing voids and some channels filled with sparite cement (C). Bioclasts of gastropods (B) can also be seen. PPL.
supply was very high, a massive calcrete horizon would not have developed because the carbonate would have been dispersed as the floodplain underwent aggradation. Ponds were generated in places where the water table intersected the surface. The seasonal fluctuation in the water table would have alternately exposed or flooded the carbonate muds precipitated in such ponds. The scarcity of clastic input together with constant subsidence could have allowed a decrease in the plain level and a consequent relative rise of the water table (Fig. 7). The calcrete-palustrine sequence grades laterally into a stack of thin calcrete profiles localized in the distal alluvial fan (Huerta and Armenteros, 2005). A similar stacking of several calcrete profiles has been attributed to a combination of climatic conditions, carbonate availability, and sediment-starved conditions (Tandon et al., 1998). In the high-flood stages, the sheetfloods would have prevented calcrete development. The sheetfloods were more episodic and supplied less sediment in the transition area (calcrete
belt) to the palustrine environments, resulting in a progressive decrease in the aggradation rate. This explains the lateral transition from the stacking of thin (10–40 cm) calcrete sequences on the distal areas of the alluvial fan into a single thick (1–2 m) calcrete profile on the calcrete belt. The high-flood events might have occurred during more humid periods in which clastic sedimentation took place. Calcrete would have developed during drier periods, as proposed for similar sequences in the Maastrichtian of India by Tandon et al. (1998). Maximum Stages with Carbonate Accumulation The carbonate bodies are close to the Honrubia and Tejada paleoreliefs, which stood at least 400 m higher than the corridor center. They occur as fringes (2–8 km wide) that are separated from the basin margins by alluvial fan deposits coming out from the paleoreliefs (Figs. 2 and 3A). Both paleoreliefs
128
Armenteros and Huerta
Figure 7. Stages of development of a calcrete-palustrine sequence on floodplain deposits. The progressive decrease in clastic input/subsidence ratio favors the rise of the groundwater table and increases carbonate precipitation. See explanation in text.
were relatively small and narrow and consequently had reduced catchment areas. Horizontal beds of the Miocene (Upper Aragonian-Vallesian) succession described here onlap the paleoreliefs, indicating that the relief has not been reactivated since then. The carbonate facies of these bodies consist of calcretes (pedogenic and phreatic) and palustrine facies in superimposed sequences, indicating their accumulation in a semiarid climate (Platt and Wright, 1992) and conditions of scarce clastic sediment supply (Fig. 8A). The calcrete facies always occur in the form of a transitional facies between the clastic (alluvial, fluvial) and the carbonate (palustrine) deposits. Local subsidence and low clastic input were associated with areas of calcrete and palustrine accumulation, for example, in troughs parallel to the northern border (Huerta and Armenteros, 2005). In this case, subsidence was essential for calcrete formation and the subsequent development of palustrine environments when the water table intersected the basin floor. However, the subsidence at the margins was lower than that of the central corridor, as manifested by the greater thickness of the Tertiary succession in this axial region where the river trunk developed (Figs. 3A and 3B). In contrast, carbonate paleosols and palustrine deposits are concentrated along the margins of the basin, where the subsidence was minor. A similar circumstance has been deduced from studies by computer simulation (Bridge and Leeder, 1979;
Bridge and Mackey, 1993) and has been observed in field studies (Mack and Madoff, 2005). Within the Miocene succession, two stages of maximum expansion of the carbonate environments took place, represented by the Lower Páramo Limestone and the Upper Páramo Limestone. These two singular stages could have been responses to dramatic decreases in alluvial and fluvial sediment supply. The carbonate accumulation in marginal fringes occurred in response to the retreat of the fluvial axial system toward the corridor center. This retreat was a consequence of the decrease in clastic influx from catchments to fans and axial rivers. Assuming that the subsidence remained constant, the decrease in the terrigenous input/ subsidence ratio resulted from a fall in the base level (defined as the lowest area in the basin). Similar relationships between base level and clastic sediment supply for nonmarine basins have been envisaged in many studies (e.g., Kraus and Bown 1986; Wright and Marriot, 1993; Shanley and McCabe, 1994; and Leeder, 1999). This situation, termed an underfilled basin (Carroll and Bohacs, 1999; Bohacs et al., 2000), could have been caused by a change to a semiarid climate, which would have caused the fluvial system to retreat to the center of the Aranda–Burgo de Osma corridor. In this case, the marginal alluvial fans became inactive and a sediment-starved area formed between these and the axial fluvial systems, allowing calcrete and palustrine devel-
Clastic sediment influx and formation of calcretes and palustrine facies
129
Figure 8. Schematic model representing two contrasting and alternating periods as the climate changed from (A) a semiarid phase (palustrine and calcrete formation) to (B) more humid conditions in which expansion of the axial fluvial system took place.
opment. The formation of calcrete and palustrine facies is generally associated with conditions of scarce rain (semiaridity) and low sedimentation rate (Wright and Tucker, 1991; Wright and Platt, 1995). Stages Dominated by Fluvial Clastic Accumulation Also notable is the gradual disappearance of the carbonate deposits from the marginal fringes. They were replaced by fluvial environments, whereas the marginal alluvial fan systems rooted in the Honrubia and Tejada borders did not undergo any significant changes (Armenteros, 1986; Huerta and Armenteros, 2003). This disappearance presumably reflects unfavorable conditions for the
accumulation of calcrete and palustrine deposits, and could have been due to an increase in the clastic sediment supply (Atkinson, 1986; Kraus and Bown, 1986; Carroll and Bohacs, 1999; AlonsoZarza et al., 1999). This increase could have been a consequence of: (1) a change toward more humid and seasonal climatic conditions (Cecil, 1990); or (2) an increase in tectonic activity. The alluvial fans, fed by small catchment areas of the Honrubia and Tejada paleoreliefs, do not display evidence of tectonic deformation, supporting the former hypothesis. Consequently, humid conditions undoubtedly were the main factor in the increase in clastic sediment in the catchment areas that nourished the fluvial system (Fig. 8B). This gave rise to a greater lateral mobility of the fluvial system, which spread toward the corridor flanks where
130
Armenteros and Huerta
carbonate deposits had accumulated at other stages. The increase in both clastic input and water supply raised the base level and caused an expansion of the fluvial system, thus inhibiting development of calcrete and palustrine deposits. Similar fluvial expansion over well-drained soils has been proposed by McCarthy and Plint (1998). Siliciclastic-Carbonate Cycles One of the most interesting features of this succession is the cyclic repetition of carbonate bodies alternating with fluvial clastic sediments. At the southern border (Honrubia), four carbonate bodies are interbedded with siliciclastic deposits (Figs. 2B and 5A) in the lower sequence, whereas at the northern border, at least three carbonate levels alternate with siliciclastic deposits in the upper sequence (Fig. 2A). This alternation must have been produced by a combination of the factors already discussed that caused variations in the clastic sediment input: a decrease promoted the formation of carbonate-producing environments and an increase favored the expansion of axial fluvial systems and the subsequent disappearance of carbonates. This cyclical pattern suggests a climatic imprint rather than periodic changes in subsidence and tectonics. The latter are usually progressive and their changes occur over long periods and tend not to result in sequences 10–20 m thick. Although this suggests the influence of climate in the alternation of clastic sediment and carbonate facies in these nonmarine successions, the importance of the paleogeographical framework developed through tectonics and subsidence should not be overlooked. The temporal extension of each 10–20-m-thick carbonate-siliciclastic sequence is difficult to establish since the chronostratigraphic resolution supplied by the fossil sites in the region does not permit comparisons with fourth- or fifth-order cycles (Milankovitch cycles). Nevertheless, the development of fluvial cycles can be related to variations in sediment supply due to climate change: incision, fan retreat, and soil formation occur during periods of low sediment supply, whereas aggradation and fan growth occur during periods of high sediment supply (Leeder, 1999).
westward component, perpendicular to that of the alluvial fans. Periodically, carbonate deposition was replaced by the expanding fluvial system. In the distal areas of the alluvial fans, thin mudstones-calcrete cycles are stacked. These sequences coalesce with increasing distance from the alluvial fans, where there was a reduced input from sheetfloods, suggesting that the calcretes were related to low clastic input. The carbonate bodies consist of, from base to top, nodular calcrete, massive calcrete, and palustrine limestone, where the latter represents the most voluminous facies. The calcretes are inferred to have formed by a combination of pedogenic and phreatic processes. The palustrine limestones represent periodic exposure of low-gradient lake margins. The scarcity of clastic input and a fall in base level due to dry climatic periods favored the development of carbonate bodies in starved areas located between the alluvial fans and the central fluvial system. By contrast, the increase in clastic input plus water supply raised the base level, favoring the expansion of axial fluvial system and the consequent disappearance of carbonate-producing environments. Tectonic uplift in the central-eastern Duero basin was minor during the Middle and Upper Miocene. Subsidence was more active in the center of the Aranda–Burgo de Osma corridor than at its margins, as suggested by the great thickness of the Tertiary succession and the presence of a fluvial system in this area. Thus, we suggest that the repeated sequences of carbonates and siliciclastic sediments in meter-scale cycles during the Miocene on both flanks of the corridor were caused by climatic rather than tectonic controls. ACKNOWLEDGMENTS This work has been supported by the research project BTE2002-04017-C02-02. The language was revised at the Foreign Languages Services of the University of Salamanca. We are grateful to M.A. García del Cura and M.R. Talbot, and to the editor, Alonso-Zarza, who contributed significantly to improving the original manuscript. L.H. Tanner helped to significantly improve the final English version.
CONCLUSIONS
REFERENCES CITED
The Miocene succession in the eastern Duero basin consists of alternating clastic-carbonate cycles in which the carbonate increases upward. These sequences range in age from the uppermost Middle Miocene to the Upper Miocene. The carbonate deposits occur as fringes (2–8 km wide) separated from the basin margins by alluvial fan deposits. The size of these alluvial fans was controlled by the size of the catchment areas developed in the Tejada and Honrubia paleoreliefs, respectively, to the north and south of the Aranda–Burgo de Osma corridor. The carbonate deposits grade laterally to a fluvial system in the center of the corridor. The paleocurrents in this latter system display a strong
Alonso-Zarza, A.M., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: Earth-Science Reviews, v. 60, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Alonso-Zarza, A.M., Wright, V.P., Calvo, J.P., and García del Cura, M.A., 1992, Soil landscape relationship in the Middle Miocene of the Madrid Basin: Sedimentology, v. 39, p. 17–35. Alonso-Zarza, A.M., Sopeña, A., and Sánchez-Moya, Y., 1999, Contrasting paleosol development in two different tectonic settings: The Upper Buntsandstein of the western Iberian Ranges, central Spain: Terra Nova, v. 11, p. 23–29, doi: 10.1046/j.1365-3121.1999.00218.x. Armenteros, I., 1986, Estratigrafía y sedimentología del Neógeno del sector suroriental de la Depresión del Duero: Salamanca, Spain, Ed. Diput. Salamanca, Serie Castilla y León, v. 1, 471 p. Armenteros, I., and Daley, B., 1998, Pedogenic modification and structure
Clastic sediment influx and formation of calcretes and palustrine facies evolution in palustrine facies as exemplified by Bembridge Limestone (late Eocene) of the Isle of Wight, southern England: Sedimentary Geology, v. 119, p. 275–295, doi: 10.1016/S0037-0738(98)00067-0. Armenteros, I., Daley, B., and García, E., 1997, Lacustrine and palustrine facies in the Bembridge Limestone (late Eocene, Hampshire Basin) of the Isle of Wight, southern England: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 128, p. 111–132, doi: 10.1016/S0031-0182(96)00108-3. Armenteros, I., Corrochano, I., Alonso-Gavilán, G., Carballeira, J., and Rodríguez, J.M., 2002, Duero basin (northern Spain), Chapter 13, in Gibbons, W., and Moreno, M.T., eds., The Geology of Spain: Geological Society [London], p. 309–315. Arribas, M.E., 1986, Petrología y análisis secuencial de los carbonatos lacustres del Paleógeno del sector N de la cuenca Terciaria del Tajo (Provincia de Guadalajara): Cuadernos de Geología Ibérica, v. 10, p. 295–334. Atkinson, C.D., 1986, Tectonic control on alluvial sedimentation as revealed by an ancient catena in the Capella Formation (Eocene) of northern Spain, in Wright, V.P., ed., Paleosols, their recognition and interpretation: Oxford, UK, Blackwell Scientific Publications, p. 139–179. Bohacs, K.M., Carrol, A.R., Neal, J.E., and Mankiewicz, P.J., 2000, Lake basin type, source potential, hydrocarbon character: An integrated-sequencestratigraphic-geochemical framework, in Gierlowski-Kordesch, E.H., and Kelts, K.R., eds., Lake Basins through Space and Time: Tulsa, Oklahoma, American Association of Petroleum Geologists (AAPG) Studies in Geology, v. 46, p. 3–34. Bown, T.M., and Kraus, M.J., 1987, Integration of channel and flood plain suites. I. Development of sequences and lateral relations of alluvial palaeosoils: Journal of Sedimentary Petrology, v. 57, p. 587–601. Bridge, J.S., and Leeder, M.R., 1979, A simulation model of alluvial overbank environments: Sedimentology, v. 26, p. 617–644. Bridge, J.S., and Mackey, S.D., 1993, A revised alluvial stratigraphy model, in Marzo, M., and Puigdefabregas, C., eds., Alluvial Sedimentation: International Association of Sedimentologists Special Publication 17, p. 319–336. Bull, W.B., 1972, Recognition of alluvial-fan deposits in the stratigraphic record: Recognition of ancient sedimentary environments: Society for Sedimentary Geology (SEPM) Special Publication 16, p. 63–83. Carroll, A.R., and Bohacs, K.M., 1999, Stratigraphic classification of ancient lakes; balancing tectonic and climatic controls: Geology, v. 27, no. 2, p. 99– 102, doi: 10.1130/0091-7613(1999)027<0099:SCOALB>2.3.CO;2. Cecil, C.B., 1990, Paleoclimate controls on stratigraphic repetition of chemical and siliciclastic rocks: Geology, v. 18, p. 533–536, doi: 10.1130/00917613(1990)018<0533:PCOSRO>2.3.CO;2. Dabrio, C.J., Alonso-Gavilán, G., Armenteros, I., and Mediavilla, R., 1989, Tertiary fluvial and fluvio-lacustrine deposits in the Duero basin (Spain), in Excursion Guidebook 4th International Conference on Fluvial Sedimentology: Barcelona-Sitges, Spain, October 2nd–4th: Servei Geològic de Catalunya, 141 p. De Vicente, G., Vegas, R., Muñoz Martín, A., González Casado, J.M., Carbó, A., Álvarez, J., Cloetingh, S., Andriessen, P., Elorza, F.J., and Olaiz, A., 2004, El Sistema Central, in Vera, J.A., ed., Geología de España: Madrid, Spain, SGE-IGME, p. 621–625. De Wet, C.B., Yocum, D.A., and Mora, C., 1998, Carbonate lakes in closed basins: Sensitive indicators of climate and tectonics: An example from the Gettysburg Basin (Triassic), Pennsylvania, USA, in Shanley, K.W., and McCabe, P.J., eds., Role of Eustasy, Climate and Tectonism in Continental Rocks: Society for Sedimentary Geology (SEPM) Special Publication 59, p. 191–209. Drummond, C.N., Wilkinson, B.H., and Lohmann, K.C., 1996, Climatic control of fluvial-lacustrine cyclicity in the Cretaceous Cordilleran foreland basin, western United States: Sedimentology, v. 43, p. 677–689. Dunagan, S.P., and Driese, S.G., 1999, Control of terrestrial stabilization on Late Devonian palustrine carbonate deposition; Catskill Magnafacies, New York, U.S.A.: Journal of Sedimentary Research, v. 69, no. 3, p. 772–786. Dunagan, S.P., and Turner, C.E., 2004, Regional paleohydrology and paleoclimatic settings of wetland/lacustrine depositional systems in the Morrison (Upper Jurassic), Western Interior, U.S.: Sedimentary Geology, v. 167, p. 269–296, doi: 10.1016/j.sedgeo.2004.01.007. Freytet, P., and Plaziat, J.C., 1982, Continental carbonate sedimentation and pedogenesis: Late Cretaceous and early Tertiary of southern France:
131
Contributions to Sedimentology, v. 12, Stuttgart, Germany, E. Schweizerbartsche Verlagsbuchhandlung Nägale u Obermiller, 213 p. Huerta, P., and Armenteros, I., 2003, Estratigrafía y sedimentología de la Unidad Detrítico–Carbonatada Superior (Mioceno Superior, Borde centrooriental de la cuenca del Duero), (Caleruega, Burgos): Studia Geologica Salmanticensia, v. 38, p. 55–86. Huerta, P., and Armenteros, I., 2005, Calcrete and palustrine assemblages on a distal alluvial-floodplain: A response to local subsidence (Miocene of the Duero basin, Spain): Sedimentary Geology, v. 177, p. 253–270, doi: 10.1016/j.sedgeo.2005.03.007. Kraus, M.J., and Bown, T.M., 1986, Paleosols and time resolution in alluvial stratigraphy, in Wright, V.P., ed., Paleosols, Their Recognition and Interpretation: Oxford, UK, Blackwell Scientific Publications, International Association of Sedimentologists, p. 180–207. Leeder, M.R., 1999, Sedimentology and Sedimentary Basins: Oxford, UK, Blackwell Science Ltd., 592 p. Mack, G.H., and Madoff, R.D., 2005, A test of models of fluvial architecture and palaeosol development: Camp Rice Formation (Upper Pliocene–Lower Pleistocene), southern Río Grande Rift, New Mexico, USA: Sedimentology, v. 52, p. 191–211, doi: 10.1111/j.1365-3091.2004.00687.x. McCarthy, P.J., and Plint, A.G., 1998, Recognition of interfluve sequence boundaries: Integrating paleopedology and sequence stratigraphy: Geology, v. 26, p. 387–390, doi: 10.1130/0091-7613(1998)026<0387: ROISBI>2.3.CO;2. Nickel, E., 1982, Alluvial-fan-carbonate facies with evaporites, Eocene Guarga Formation, southern Pyrenees, Spain: Sedimentology, v. 29, p. 761–796. Nickel, E., 1985, Carbonate in alluvial fan systems: An approach to physiography, sedimentology and diagenesis: Sedimentary Geology, v. 42, p. 83– 104, doi: 10.1016/0037-0738(85)90075-2. Picard, M.P., and High, L.R., 1981, Physical stratigraphy of ancient lacustrine deposits: Recent and ancient nonmarine depositional environments; models for exploration: Society for Sedimentary Geology (SEPM) Special Publication 31, p. 233–239. Pimentel, N.L., 2002, Pedogenic and early diagenetic processes in Paleogene alluvial fan and lacustrine deposits from the Sado Basin (S Portugal): Sedimentary Geology, v. 148, p. 123–138, doi: 10.1016/S00370738(01)00213-5. Platt, N.H., 1989, Lacustrine carbonates and pedogenesis: Sedimentology and origin of palustrine deposits from the Early Cretaceous Rupelo Formation, W. Cameros Basin: N Spain: Sedimentology, v. 36, p. 665–684. Platt, N.H., and Wright, V.P., 1991, Lacustrine carbonates: Facies models, facies distribution and hydrocarbon aspects, in Anadón, P., Cabrera, L., and Kelts, K., eds., Lacustrine Facies Analysis: International Association of Sedimentologists Special Publication 13, p. 57–74. Platt, N.H., and Wright, V.P., 1992, Palustrine carbonates at the Florida Everglades: Toward an exposure index for the freshwater environment: Journal of Sedimentary Petrology, v. 62, no. 6, p. 1058–1071. Sanz, M.E., Alonso-Zarza, A.M., and Calvo, J.P., 1995, Carbonate pond deposits related to semi-arid alluvial systems: Examples from the Tertiary Madrid Basin, Spain: Sedimentology, v. 42, p. 437–452. Shanley, K.W., and McCabe, P.J., 1994, Perspectives on the sequence stratigraphy of continental strata: American Association of Petroleum Geologists (AAPG) Bulletin, v. 78, no. 4, p. 544–568. Talbot, M.R., and Allen, P.A., 1996, Lakes, in Reading, H.G., ed., Sedimentary Environments: Processes, Facies and Stratigraphy: Oxford, UK, Blackwell Science, p. 83–124. Tandon, S.K., and Andrews, J.E., 2001, Lithofacies associations and stable isotopes of palustrine and calcrete carbonates: Examples from an Indian Maastrichtian regolith: Sedimentology, v. 48, p. 339–355, doi: 10.1046/ j.1365-3091.2001.00367.x. Tandon, S.K., Andrews, J.E., and Mittal, S., 1998, Shrinkage and sediment supply control on multiple calcrete profile development: A case study from the Maastrichtian of central India: Sedimentary Geology, v. 1–2, p. 25–45. Tanner, L.H., 2000, Palustrine-lacustrine and alluvial facies of the (Norian) Owl Rock Formation (Chinle Group), Four Corners Region, southwestern U.S.A.: Implications for Late Triassic paleoclimate: Journal of Sedimentary Research, v. 70, p. 1280–1290. Van Houten, F.B., 1964, Cyclic lacustrine sedimentation, Upper Triassic Lockatong Formation, central New Jersey and adjacent Pennsylvania, in Merrian,
132
Armenteros and Huerta
D.F., ed., Symposium on Cyclic Sedimentation: Kansas Geological Survey Bulletin, v. 169, p. 495–531. Wright, V.P., and Marriot, S.B., 1993, The sequence stratigraphy of fluvial depositional systems: The role of floodplain sediment storage: Sedimentary Geology, v. 86, p. 203–210, doi: 10.1016/0037-0738(93)90022-W. Wright, V.P., and Marriot, S.B., 1996, A quantitative approach to soil occurrence in alluvial deposits and its application to Old Red Sandstone of Britain: Journal of the Geological Society of London, v. 153, p. 907–913. Wright, V.P., and Platt, N.H., 1995, Seasonal wetland carbonate sequences and
dynamic catenas: A re-appraisal of palustrine limestone: Sedimentary Geology, v. 99, p. 65–71, doi: 10.1016/0037-0738(95)00080-R. Wright, V.P., and Tucker, M.E., 1991, Calcretes: An introduction, in Wright, V.P., and Tucker, M.E., eds., Calcretes: Oxford, UK, Blackwell Scientific Publications, International Association of Sedimentologists, Reprint Series, v. 2, p. 1–22. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
The Upper Triassic crenogenic limestones in Upper Silesia (southern Poland) and their paleoenvironmental context Joachim Szulc† Michał Gradzi´nski Anna Lewandowska Institute of Geological Sciences, Jagiellonian University, Cracow, Poland Carmen Heunisch Niedersächsisches Landesamt für Bodenforschung, Hannover, Germany ABSTRACT Upper Triassic (Norian) freshwater carbonates, up to 30 m in thick, occur in the northern part of Upper Silesian basin. These sediments, called the Wo´zniki Limestone, form a SE-NW–striking elongate (90 km) and narrow (<10 km) belt. The Wo´zniki Limestone overlies (mostly discordantly) Carnian gypsiferous red beds and underlies the uppermost Triassic–Lower Jurassic continental clastic deposits. Laterally, the carbonates are replaced by a typically red bed clastic assemblage formed under arid and semiarid climatic conditions. Several limestone types have been recognized within the freshwater facies, including travertines, and fluvial, palustrine and pedogenic carbonates. Palustrine limestones form a major component. Common tepee structures, karst breccia, silicified horizons, and weathering breccia indicate that the palustrine carbonates have undergone subaerial exposition and pedogenic alteration. Palustrine carbonate sedimentation has been interrupted and replaced by fluvial sedimentation. The fluvial sediments mark the pluvial climate episodes that inhibited carbonate deposition. The studied basin displays a striking scarcity of lacustrine sediments, which may be explained in terms of hydrological and climatic controls. We infer that the carbonates were deposited within shallow swampy depressions, fed by springs of deep-circulating groundwater, partly of hydrothermal nature, under dry and semidry paleoclimatic conditions in a fault-bounded basin. The travertines precipitated directly near the springs, whereas the remnant solutions formed a broad swamp area where palustrine carbonates formed. It seems very likely that the carbonate-bearing solutions were causally related to the hydrothermal karst that occurs within the Triassic and Paleozoic basement limestones. Keywords: freshwater limestones, travertines, paleoenvironments, Upper Triassic, Poland.
E-mail:
[email protected].
†
Szulc, J., Gradzi´nski, M., Lewandowska, A., and Heunisch, C., 2006, The Upper Triassic crenogenic limestones in Upper Silesia (southern Poland) and their paleoenvironmental context, in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 133–151, doi: 10.1130/2006.2416(09). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
133
134
Szulc et al. RESUMEN La parte norte de la cuenca alta de Silesia incluye una unidad de carbonatos de agua dulce, de 30 metros de potencia y edad Triásico Superior (Noriense). Estos sedimentos, denominados Caliza Wo´zniki forman un cinturón estrecho y alargado (<10 km) en dirección SE-NW. La Caliza Wo´zniki se apoya, casi siempre discordante, sobre las capas yesíferas rojas del carniense y sobre ella se sitúan los depósitos continentales del Triásico más alto y Jurásico inferior. Lateralmente estos carbonatos pasan a las capas rojas detríticas formadas bajo condiciones climáticas áridas y semiáridas. Las calizas están formadas por distintos tipos de facies que incluyen: facies fluviales, palustres, pedogénicas y de surgencias. Las calizas palustres son las más abundantes. Rasgos como “tepees,” brechas cársticas, horizontes silicificados y brechas de alteración indican que los carbonatos palustres sufrieron exposición subaérea y pedogénesis. La sedimentación palustre carbonática quedó interrumpida por una etapa posterior de sedimentación fluvial, que indica climas más húmedos. Los depósitos lacustres ss, que normalmente son anteriores a los palustres, son muy escasos en esta cuenca. Esto puede ser debido a los controles hidrológicos y climáticos. Así, si se tienen en cuenta las condiciones áridas y semiáridas, el hecho de que los bordes de la cuenca son fallas y la abundancia de travertinos, se puede considerar que los carbonatos se depositaron en una depresión somera y pantanosa, abastecida por surgencias de aguas freáticas profundas (en parte hidrotermales). Los travertinos se formaron por precipitación directa en las zonas cercanas a las surgencias, mientras que el resto del agua se acumuló en depresiones pantanosas amplias donde se formaron los carbonatos palustres. Parece muy probable que las soluciones ricas en carbonato tuvieran relación con el carst hidrotermal que se desarrolló en las calizas del Triásico y Paleozoico. Palabras clave: calizas de agua dulce, travertinos, paleoambientes, Triásico Superior.
INTRODUCTION This paper focuses on the Upper Triassic continental sediments that occur in the northern part of the Upper Silesian basin (Fig. 1) called the Wo´zniki Limestone. In this paper, we use the term Wo´zniki Limestone (WL) as an informal lithostratigraphical unit dominated by carbonates but which also includes subordinate clastic intercalations. The Wo´zniki Limestone includes a range of isolated carbonate bodies surrounded by carbonate-poor, variegated muddy sediments. Fossils of the Wo´zniki Limestone are very scarce and limited to ostracodes and calcified plant molds. These carbonates have long been recognized as continental sediments (Roemer, 1867; Michael, 1912); however, their age and exact sedimentary context are uncertain. The common consensus is that the Wo´zniki Limestone formed in a lacustrine environment (Gasiorowski ˛ and Piekarska, 1976, 1986), but no convincing evidence of such an origin has been provided so far. In fact, the sediments of the Wo´zniki Limestone display very few features of typical lacustrine deposits. The main goal of the present paper is to reinterpret the genesis of the Wo´zniki Limestone by means of sedimentological and geochemical examinations. GENERAL AND PALEOGEOGRAPHICAL SETTING The Wo´zniki Limestone forms a SE-NW–striking assemblage of carbonate bodies, stretching a distance of some 90 km
and occupying ~300 km2 (Fig. 1). Its thickness reaches up to 30 m. The Wo´zniki Limestone is situated between the gypsiferous Upper Gipskeuper of early Norian age and the fluvial facies assemblage of the Rhaetian (Fig. 2). The lack of fossil remnants and poorly recognized facies context has long hindered more precise age determination of the Wo´zniki Limestone. Its age has been assumed mainly to be Rhaetian (Znosko, 1960; GrodzickaSzymanko and Orłowska-Zwoli´nska, 1972; Bilan, 1976). According to our palynological examination, the Wo´zniki Limestone encompasses palynomorph taxa indicative of socalled palynostratigraphic assemblage IV (Corollina meyeriana zone) in the zonation scheme by Orłowska-Zwoli´nska (1983), Fijałkowska-Mader (1999), and Heunisch (1999). Based on the palynostratigraphical and lithostratigraphical position of the Wo´zniki Limestone, we estimate its age to be Norian. The studied area was situated during Norian times within the subtropical convergence zone. Dry climatic conditions dominated; however, several humid intervals have also been recognized during this time in central Europe (Reinhardt and Ricken, 2000). The pluvial intervals were affected by other climate-forming factors, such as changes in ocean-land configuration, volcanism, or supraregional tectonic-topographical changes (Simms and Ruffel, 1990; Szulc, 2007). It is important to note that the Wo´zniki Limestone is closely linked to the master fault dislocation in the region (Szulc et al.,
Upper Triassic freshwater limestones from Poland
135
A B
Figure 1. (A) Geological sketch map of the Wo´zniki Limestone, and (B) location of the studied outcrops and drill holes.
B
2002) (Fig. 1), called the Cracow-Lubliniec fault (cf. Morawska, 1997). Typical carbonates of the Wo´zniki Limestone occur in a belt of some 10 km in width, adjacent to the fault zone. Clearly, no similar facies occur outward from this belt. The thick carbonate succession, reaching up to 30 m (Figs. 2–5), is a fundamental feature that distinguishes the Wo´zniki Limestone from its coeval counterparts in other regions of the mid-European Basin (see Beutler et al., 1999). METHODS AND MATERIALS Outcrops of the Wo´zniki Limestone are rare. Therefore, in order to accomplish the research goals, several holes were drilled across the entire outcrop area of the Wo´zniki Limestone. Twelve
outcrop sections and six cores were studied for sedimentary fabrics and facies variability. The sedimentological observations were supplemented by petrological and geochemical analyses of mineralogical composition, stable isotopes, major and trace elements, and clay minerals. The measurements of carbonate δ13C and δ18O were conducted with a SUMY mass spectrometer at the Institute of Geochemistry and Geophysics, the Academy of Sciences of Belarus in Minsk. The isotope ratios were measured in carbon dioxide generated by reaction of the samples with 100% orthophosphoric acid. Carbon dioxide was subsequently trapped in liquid nitrogen and purified in a vacuum. The analytical error for single measurements was ±0.2‰. Stable oxygen isotope ratios are expressed relative to Peedee belemnite (PDB) standard (see Hoefs, 1997).
136
Szulc et al. Travertines
Figure 2. Lithostratigraphic log of the studied Upper Triassic interval.
We analyzed 41 samples from different localities (both drill holes and outcrops) palynologically in order to establish the age position of the studied succession by biostratigraphy. Most of the samples contained only opaque phytoclasts and occasional plant tissues in varying amounts, but were barren of palynomorphs. Only eight samples contained sufficient palynomorphs to allow a biostratigraphical classification. Preservation of the palynomorphs varied from poor to good. RESULTS OF SEDIMENTOLOGICAL AND MICROFACIES STUDIES Four basic facies types have been recognized within the described freshwater limestones: the travertine facies, the fluvial facies, the palustrine facies, and the pedogenic facies.
Travertine facies have been found at three sites in the SE part of the basin (Fig. 1) and are represented by calcite fabrics precipitated directly in the spring orifice, and by spring-margin pool sediments where carbonates were deposited at a more moderate rate. The two subfacies differ slightly in the dominant fabric type. The travertines that were formed in the spring orifices form either highly porous pure limestones composed of calcitic rafts and heavily calcified rhizomes and stalks (Fig. 6A) or are built by pisoids reaching 1 cm in diameter (Fig. 6B). The pisoids are in fact composed of microbial aggregates displaying faint concentric structure (Fig. 6C). The pisolites show common reversed grading and are interlayered with stromatolites (Fig. 6B). The latter are composed of dendritic shrubs of bacterial origin (Chafetz and Folk, 1983; Folk et al., 1985; Pentecost, 1990; Guo and Riding, 1994) or of filamentous fabrics (Fig. 6D). Trapped, calcified detritus of vascular plants is a common component of the spring travertines (Fig. 6E). Most of the encrustations are related to calcification driven by epiphytic microbial colonies covering the vascular plants. The limestones that were formed in small pools in the mar˙ ginal spring zone (Nowa Wie´s Zarecka site) are also very rich in calcified algae and reed-like, vascular plants, but in contrast to the spring-mouth travertines, they also include finely laminated peloidal limestones (Fig. 6F). A rich microbial epiphytic assemblage (bacteria, cyanobacteria) enhanced the calcification of the higher plants (Fig. 6G). Moreover, thin calcitic rafts that formed at the surface of the pool water were probably related to activity of neustonic algal colonies (Fig. 6H) (Szulc, 1997). The travertines are devoid of clastic impurities, and the only noncarbonate components are minute quartz grains found in the travertines from the Por˛eba and Ogrodzieniec sites. In addition to the sites examined in this study, travertines were found by Roemer (1867) in the central part of the basin (see Fig. 1). From this travertine, Roemer described numerous molds of ferns (Clathropteris sp.), typical for the Upper Keuper. Fluvial Facies The fluvial deposits are generally represented by fine-grained clastics, mainly greenish and red mudstones with subordinate contribution of arkosic sandy material. These deposits occur as sheets interbedded within the carbonate packages of the Wo´zniki Limestone (Fig. 7A). The fluvial deposits are mostly plane-bedded, cross-bedded, and rippled sandstones and siltstones (Figs. 7B and 7C). The primary structures are commonly obliterated due to postdepositional pedogenic processes; however, the dominant recognizable primary sedimentary structures include parallel lamination and ripple-drift cross-lamination. These structures suggest a sheet-like depositional system developed upon a low-relief mudflat area.
Upper Triassic freshwater limestones from Poland
137
Figure 3. Measured sections and facies interpretation from boreholes Ogrodzieniec (travertine) and Niwki (transition from gypsiferous playa sediments to palustrine carbonates) with the δ13C profiles. The shadowed parts of the profiles mark the gray to greenish sediment color. The nonshadowed part of the mudstones marks the red and brown color of the sediments.
LIMESTONES DOLOMITES MARLS MUDSTONES SANDSTONES AND CONGLOMERATES RHIZOIDAL NODULES WITH SPARITE CEMENT MARLY LIMESTONES POROUS TRAVERTINE UPPER GIPSKEUPER - PLAYA SEDIMENTS (NOT DISCUSSED IN THE TEXT) CALCRETE CRUST CHERTS MUDCRACKS RESIDUAL DEBRIS ALTERED MUDSTONES
PEDOGENIC CONCRETIONS ONCOIDS STROMATOLITES EVAPORITES RHIZOIDS PELECYPODS COALIFIED PLANTS CALCIFIED VASCULAR PLANTS
IVb PALYNOLOGICAL ZONE (SEE ALSO TEXT)
Some of the beds have an erosional lower bounding surface, in particular the sheet-like gray, poorly sorted conglomeratic beds reaching up to 50 cm in thickness. The conglomerates consist of oncoids (Fig. 7D), coalified wood fragments (Fig. 7E), reworked pedogenic carbonate nodules (Fig. 7F), vertebrate bones and unionid bivalve debris (Fig. 7G). Such a composition indicates that the conglomerates originated as intraformational deposits through reworking and mixing of the material derived from pond sediments and paleosols by ephemeral streams operating upon the mudflat after heavy runoff events. Similar Triassic conglomerates composed by reworked calcretes have been described also by Gómez-Gras and Alonso-Zarza (2003) from Minorca and by Szulc (2005) from Upper Silesia. The alluvium also encompasses black pebbles (Fig. 7H). The latter formed probably in small, poorly ventilated ephemeral pans where lithoclasts underwent impregnation by organic matter and were incorporated into fluvial sediments after redeposi-
tion (Strasser, 1984). Alternatively, the black pebbles might have originated due to local wildfires (Shinn and Lidz, 1987). The main clastic mineral component is quartz, with subordinate contributions of clay minerals, K-feldspars, and carbonates. Among the clay minerals, illite, kaolinite, and very subordinate, mixed-layered illite/smectite have been detected (Lewandowska et al., 2001). The lowermost parts of the studied succession contain chlorite also. The clastic intercalations often show green or red color mottling, reflecting postdepositional pedogenic processes. Palustrine Facies The palustrine facies dominates among the freshwater sediments and comprises some 80 percent of the entire carbonate succession of the Wo´zniki Limestone. This facies is particularly well developed in the central part of the basin, i.e., between Cynków
Figure 4. Measured sections and facies interpretation of the dominant palustrine carbonates from Wo´zniki and Cynków with the δ13C profiles. For legend, see Figure 3.
Upper Triassic freshwater limestones from Poland
139
Figure 5. Measured sections and facies interpretation of the profiles from Lipie ´ askie, Por˛eba, and Zawiercie. For legSl˛ end, see Figure 3.
and Psary (Fig. 1). The total thickness of the palustrine carbonates reaches 30 m (Fig. 4), but the carbonate succession is divided into two parts. This is particularly visible in the Cynków section, where the two palustrine carbonate packages are separated by 5 m of fluvial claystones and variegated mudstones (Fig. 4). Palustrine carbonates are massive and/or faintly stratified white micritic limestones and rarely marls (Figs. 8A–8C). The faint stratification is accentuated by intraformational breccias, sheet cracks, calcrete crusts, teepee structures, or paleokarst horizons (Figs. 8A, 8C, 8D, and 8E). The larger karst cavities are commonly filled with clayey material, and the smaller voids are filled with internal silt and sparry cement (Figs. 8D and 8F). Some voids, in particular those related to rhizome systems, are filled with marcasite and pyritic encrustations (Fig. 8G).
The dominant microfacies type is homogeneous micrite with microgranular and clotted texture, which displays microscopic features, similar to the automicritic peloidal muds generated by bacterial mediation (Fig. 9) (Reitner, 1993). The characteristics of the carbonate palustrine succession of the Wo´zniki Limestone show some obvious vertical changes. The lower part of the palustrine carbonates commonly contains pseudomorphs after dispersed crystals and aggregates of gypsum (Figs. 10A, 10B, and 10C). Rootlet fabrics are notably scarce in this part of the section. Upsection, the sulfates disappear, and the rhizoid fabrics become more common. The palustrine lithologies are composed of low magnesian calcite. The noncarbonate components consist of clay minerals (up to 3 wt%) and quartz (<1 wt%). Among clay minerals, illite
A
B
C
0.5 mm
D
F
E
0.8 mm
G
H
0.8 mm
2 mm
Figure 6. Spring facies. (A) Porous travertines with calcified stems of vascular plants and calcite rafts from the Ogrodzieniec borehole. Scale bar is 3 cm long. (B) Pisoidal and micropisoidal travertine from the Por˛eba site. Calcified debris of vascular plants is visible in the middle of the sample. Note the reversed grading of the pisolites. Scale bar is 3 cm long. (C) Microscopic view of the pisolitic limestones from B. The pisoids are composed of faintly laminated, clotted microbial grains. (D) Pisolitic-stromatolitic travertines from the Por˛eba site. The stromatolitic shrubs developed partly as overgrowths on the pisoids (arrow). Scale bar is 3 cm long. (E) Calcified cone mold embedded in the travertines from the Por˛eba site. Scale bar is 3 cm long. (F) Per˙ pendicular section of the calcified reed-like stems from the spring-fed pools from the Nowa Wie´s Zarecka site. (G) Microscopic view of the cyanobacterial mats building the stromatolite fabrics in the travertines from the Por˛eba site. (H) Calcite micritic rafts, partly broken and sunk, from the pool limestones, ˙ ˙ Nowa Wie´s Zarecka site. (Insert) calcified, bubble-like neustonic algae (cf. Botrydium sp.) from the Nowa Wie´s Zarecka site.
A
B
5 cm
20 cm
D
C
3 cm
E
F
5 mm
5 mm
H
G
5 mm
33 cm cm
Figure 7. Fluvial facies. (A) Sheetflood, alluvial deposits with intraformational, conglomerates from the Wo´zniki site. (B) Cross-bedded, mica´ askie site. (C) Plane-bedded, overbank muddy alluvial sediments from the Lipie Sl˛ ´ askie site. Hammer ceous alluvial sandstones from the Lipie Sl˛ handle is 32 cm long. (D) Reversely graded alluvial sediments composed of oncoids, coal debris, and lithoclasts from the Por˛eba borehole. Scale bar is 3 cm long. (E) Coalified wood fragment and small lithoclasts. Thin section is from the sample in D. (F) Reworked pedogenic nodules, lithoclasts, bone fragments, and Chara gyrogonite. Thin section is from the sample in D. (G) Fragment of oncoid enveloping unionid shell. Thin section is from the sample is D. (H) Black pebble accumulation upon eroded palustrine limestones. Scale bar is 3 cm long.
A
B
C
D C
30 cm
3 cm
E
3 cm
F
3 cm
G
3 cm
Figure 8. Palustrine facies. (A) Typical, faintly stratified palustrine carbonate deposits of the Wo´zniki Limestone from the Ligota site. Arrow indi´ askie site. Hammer for scale. cates tepee deformation. Hammer for scale. (B) Thick- and medium-bedded palustrine limestones from the Lipie Sl˛ (C) Massive palustrine limestones with karstic surfaces (white arrows) and calcrete crust at the top from the Cynków site. Note the uneven surface of the carbonate complex. (D) Planar cracks and microkarstic voids filled with internal silt and sparry cement from the Psary site. Scale bar is 3 cm long. (E) Small tepee deformation within the palustrine limestones from the Cynków borehole. Scale bar is 3 cm long. (F) Microkarstic cavities and dilatancy fissures (marked by arrows) filled with internal silt and sparry calcite, from the Por˛eba borehole. See text for further explanations. Scale bar is 3 cm long. (G) Massive palustrine limestones with root cavities encrusted by pyrite, from the Psary site. Scale bar is 3 cm long.
Upper Triassic freshwater limestones from Poland
Figure 9. Palustrine facies. Microgranular, lumpy micrite probably of microbial origin (automicrite sensu Reitner, 1993).
is dominant, with secondary kaolinite and a very subordinate amount of mixed-layered illite/smectite. The clay assemblage is of detrital origin and does not differ from that of the adjacent alluvial mudstones and claystones. It is noteworthy that the palustrine limestones underlying the clastic intervals (mostly of fluvial origin) are commonly dolomitized. Dolomitization may encompass a superficial layer of the palustrine limestone several centimeters thick (Fig. 10D), but may also extend as deep as 5 m from the top of the palustrine complex. The dolomitization created secondary porosity in the carbonates and produced yellow, vuggy horizons, called cellular dolomites (Figs. 10E and 10F), where the dolomite content can be up to 85 wt%. Microcrystalline, uniform dolomite dominates and replaces the calcite substrate pervasively. Most probably, the porosity was formed by replacement of limestone by the dolomite, followed and/or accompanied by dissolution of the nonreplaced limestones. The pores are partly filled with fine clastic quartz and clayey interstitial sediments, indicating infiltration by meteoric waters. From the stable isotopes analyses (see following section), it can be concluded that the dolomite formed from solutions not subjected to evaporation. Taking this all together, the dolomitization was closely related to the subsoil processes (dissolution and precipitation) proceeding in the vadose zone (Sherman et al., 1962). Within the palustrine facies, silicified horizons perfectly preserve the primary fabrics of the host carbonate sediments, including gypsum pseudomorphs, and encompass altered carbonates up to 50 cm thick (Fig. 10G). Pedogenic Facies The pedogenic sediments may be divided into soils developed on the clastic substrate (i.e., variegated mudstones and claystones) and carbonate soils (i.e., calcretes).
143
The first type encompasses a variety of paleosols from incipient, regolithic soils (Inceptisols) to more matured orders of Aridisols (sensu Retallack, 2001). This type of paleosol forms mainly brown, nodular and friable mudstones, reaching 1 m in thickness and passing gradationally to the underlying parent substrate. Some paleosols display relatively well-preserved root traces that diffusively penetrate the underlying host rocks. Pedogenic slickensides in the variegated mudstones are a fabric typical of semiarid Cambisol-Vertisol types of soil (de Vos and Virgo, 1969; Fitzpatrick, 1986; Retallack, 2001). The color mottling, typical for these paleosols, seems to reflect the early diagenetic oxidation of the hydrated Fe-oxides (gray colored) leading to mature Fe-oxides (hematite) and reddening of the primary gray sediments (Turner, 1980). The carbonate soils (Calcisols), which have thicknesses ranging between several centimeters to 1 m, are developed either as pedogenic nodules and vadoids (coated grains of coated origin) entombed within clastic substrate (mainly weathered mudstones) or as various pedogenic fabrics developed upon the palustrine carbonates (Figs. 11A–11D and 11F). Both types of Calcisols display a wide spectrum of features diagnostic for their pedogenic origin, such as glaebules, circumgranular and septarian cracks, cutans, and root canals (Freytet and Plaziat, 1982) (see Figs. 11D and 11E). Upsection, the soils pass mostly into the palustrine carbonates, but sometimes they are covered by alluvial material. Some of the palustrine limestone complexes are capped by massive calcrete of hardpan type (between several centimeters and 0.5 m in thickness) (Fig. 8C) or are intensively karstified. The surface of the karstified carbonates, as visible in the plane view, is jagged and features karst fabrics such as sinkholes up to 1 m in depth (Figs. 11F and 11G). RESULTS OF STABLE ISOTOPES EXAMINATION The paleoenvironmental reconstruction based on the carbon and oxygen stable isotope signals is consistent with the results of the sedimentological and petrographical studies. The stable isotope composition of the Wo´zniki Limestone generally follows the vertical and lateral variation of the sedimentary facies. Therefore, we discuss the diversity of stable isotope signals from various laterally equivalent facies and the isotopic variation as a result of the overall, secular paleoenvironmental changes in the region. Facies-Dependent Isotope Composition (Lateral Fractionation) By analyzing the spatial distribution of the stable isotopes in terms of the facies diversity, one may find a clear relationship between the facies type and stable isotope composition (Figs. 12A and 12B). The most positive values characterize the travertines from Por˛eba, (Fig. 12A), where δ13C ranges between –0.3‰ and –3.5‰, and δ18O ranges from –5.5‰ to –6.7‰.
A
B
3 mm
C D
2 mm
E
5 mm
F
G
Figure 10. Palustrine facies. (A) Gypsum nests (pseudomorphed) within palustrine limestones from the Cynków borehole. Scale bar is 3 cm long. (B) Thin section from the sample in A. Note the displacive and enterolithic form of gypsum growth. (C) Secondary porosity after dissolved gypsum crystals. The voids are geopetally filled with internal silt and sparry calcite (from the Wo´zniki borehole). (D) Dolomitized palustrine limestones. The ochre staining (dark color at the photo) comes from Fe- and Mn-oxide impregnation (from the Cynków borehole). Scale bar is 3 cm long. (E) Thin section from the sample in D. (F) Highly porous (“cellular”) dolomites with MnO-concentration (black spots). Thin section is from the sample in D. (G) Thin section of silicified, gypsum-bearing, palustrine carbonates from the Wo´zniki borehole.
A
B
3 cm
3 cm
V
C
3 cm
D
E
5 mm
F
G
25 cm
Figure 11. Pedogenic fabrics. (A) Pedogenic vadoid horizon (V) developed upon exposed palustrine limestones from the Brudzowice site. Note the planar cracks and rhizoid fabrics within the palustrine carbonates. Scale bar is 3 cm long. (B) Planar cracks and circumgranular, desiccation cracks featuring the exposed palustrine limestone from the Cynków borehole. Scale bar is 3 cm long. (C) Slab of the mottled pedogenic nodule isolated from the mudflat clastic sediments from the Zawiercie site. Scale bar is 3 cm long. (D) Thin section of paleosol carbonates with mottled ´ askie site. (F) Plane fabrics and initial vadoidal cortex from the Psary site. (E) Thin section of paleosol glaebules with septaria from the Lipie Sl˛ view of paleoweathering surface developed upon exposed palustrine limestones from the Cynków site. Lens cap is 55 mm across for scale. (G) Plane view of deeply karstified palustrine limestones from the Cynków site. Depth of the sinkhole reaches ~0.6 m.
146
Szulc et al.
A
pic composition has been found in spring-related Upper Triassic freshwater carbonates in Wales (Leslie et al., 1992). Evolution of the Stable Isotope Composition with Time
B
Figure 12. Cross-plot of δ13C and δ18O values for spring carbonates (A) and palustrine and pedogenic carbonates (B).
The carbonates that formed in the marginal ponds of the ˙ spring zone (Nowa Wie´s Zarecka site) are lighter in δ13C by some 2‰–3‰ than the travertines formed directly by the spring orifice. The range of δ18O is very narrow and fluctuates between –4.8‰ and –6.2‰. The stable isotope signals from the other facies (fluvial, palustrine, pedogenic) display a much wider range of values (Fig. 12B) since they were influenced by a broad spectrum of fluctuating environmental controls, particularly climatic. As in the travertines, the carbon isotope values exhibit a wider range (δ13C values from –11‰ to 0.5‰ versus PDB) than the oxygen isotopes (δ18O values from –8‰ to 0‰ versus PDB). Strongly negative excursions of δ18O (to –11‰ versus PDB) are related to dolomitization proceeding under the influence of meteoric waters (Figs. 12A and 12B). Notably, a similar variation in the isoto-
The δ13C profiles from the three longest sections of the Wo´zniki Limestone (Niwki, Wo´zniki , Cynków) display a common trend, which undoubtedly reflects the longer-term environmental changes in the basin area (Figs. 3 and 4). The fluvial sediments that formed during the wet, pluvial phase that preceded the deposition of the Wo´zniki Limestone carbonates show relatively negative δ13C values, which might have been caused by a significant influx of meteoric water with isotopically light soil CO2 derived from decay of rich plant debris. The subsequent abrupt positive shift toward a δ13C range of 0‰ to +1‰ (PDB) accompanied the facies change from fluvial to palustrine, carbonate and/or sulfate deposition. Such a positive shift may be interpreted as typical for evaporitic enrichment in heavier isotopes. This in turn implies climate aridification. The covariant trend in δ18O composition confirms this inference. Subsequently, the δ13C curve displays a gradual shift to more negative values. The isotopic trend is concurrent with a lack of sulfate minerals and the more common appearance of root systems. Therefore, we attribute it to a progressively more humid climate and a growing influence of the isotopically light carbon associated with meteoric water input and/or carbon derived from decayed organic matter. The δ18O does not display this apparent trend (Fig. 12). Unlike carbon isotopes, O-isotope fractionation is more sensitive to incidental factors such as short-term changes in evaporation and meteoric water influx. Pronounced negative shifts of δ13C and δ18O characterize the previously described dolomitized vuggy limestones. Such a negative shift indicates a definitive contribution of meteoric water and indicates that the dolomitization was related to the diluted water activity. Long-term evolution in the travertine profile of Ogrodzieniec can also be observed. The observed gradual decrease in δ13C values reflects changes of isotope contents in the parent solutions and/or migration of the spring orifice zone. Since the δ13C covaries with the negative shift of δ18O (Fig. 3), the observed trend most likely reflects the increasing contribution of the isotopically lighter, meteoric water to the formation of carbonates. DISCUSSION OF THE ORIGIN AND GENETIC ´ MODEL OF THE WOZNIKI LIMESTONE Paleohydrological and Tectonic Controls of the Origin of the Wo´zniki Limestone Sedimentation of thick carbonate complexes of the Wo´zniki Limestone seems to be in contradiction to the arid and semiarid climatic conditions prevailing in Norian times, since deposition of such a voluminous carbonate body requires an adequate volume of the parent solutions. The apparent contradiction may
Upper Triassic freshwater limestones from Poland be plausibly explained if one assumes a crenogenic, i.e., springrelated, alimentation model of the Wo´zniki basin as already postulated by Bogacz et al. (1970). It is very interesting that the facies assemblage of the Wo´zniki Limestone has a paucity of lacustrine sediments, which are limited to deposits formed in small and shallow pools fed by spring water. Presently, these more resistant carbonates form gentle hills, whereas the fine clastics of the mudflat sediment adjoining to the limestone underwent erosion, giving a reversed pattern of the Late Triassic paleotopography. It seems that the Wo´zniki Limestone did not form one laterally continuous carbonate body, as suggested G˛asiorowski and Piekarska (1986), but rather represents a group of more or less isolated smaller patches of limestones, deposited in local swampy depressions maintained by a spring system. The travertines would be the spring-adjacent facies, while the distal facies are represented by palustrine carbonates. As indicated by the stable isotopic data, the isotopic composition of the travertines differs from those of the other facies (Fig. 12). Lack of data on the original isotopic composition of the parent waters makes any further inferences, e.g., on the temperature of the water, uncertain. Recently, Słowakiewicz (2003) has claimed that all carbonates of the Wo´zniki Limestone are hydrothermal spring deposits. This conclusion is not reliable, since, as already discussed, the sedimentary facies context, biotic data, petrological and stable isotope signals, and the high temperature of the solution probably characterized only limited, spring-adjacent precipitated travertines. The other facies, such as the dominant palustrine one, include organisms (for instance, vascular plants and ostracodes) that do not have the ability to persist and develop in the temperature range (up to 97 °C) suggested by S⁄lowakiewicz (2003). As already noted the Wo´zniki Limestone is poor in fossils, particularly the palustrine facies, where only ostracodes have been found. Beside the ostracodes, one uncertain gastropod mold has been mentioned by Roemer (1867), who determined it as a possible Paludina sp. The main reason for the paucity of biota is that the palustrine sediments formed under very stressed environmental conditions. The elevated alkalinity and salinity hindered colonization on one hand, and led to rapid degradation of the organic matter on the other hand. Also, the paucity of palynomorphs resulted from their degradation under high alkaline conditions. Moreover, the common occurrence of sulfide concretions (Fig. 8G) indicates that dysoxic conditions dominated in some poorly drained, waterlogged sediments. Finally, since the basin was extremely shallow, the palustrine carbonates were desiccated very often. The environmental stress not only eliminated most of the organisms but also selected a very specialized group. A good illustration of such a selection is exemplified by the ostracodes. It is striking that most of the disarticulated ostracode tests have been found in voids developed within the sediments, i.e., below the sediment surface (Fig. 13). The ostracode colonies dwelt in the primary voids (i.e., root
147
0.8 mm
Figure 13. Thin section of root tube with disarticulated tests of coenobiotic ostracodes from the Psary site.
canals, sheet cracks). The subterranean, coenobiotic living strategy gave more chance for survival during the drought periods, since the wet conditions, supported by soil moisture and interstitial water, persisted there longer than in the superficial zone. On the other hand, the coenobionts may indicate contrasting hydrologic conditions between the dry surface and the subjacent, groundwater-soaked sediment column. An endogenic origin of fluids is confirmed by distribution of the Wo´zniki Limestone, which is tightly bound to a master fault zone that might have provided a conduit for the ascending solutions. Typical carbonates of the Wo´zniki Limestone lie in a belt adjacent to the fault (Fig. 1) and do not occur outside this zone. Such a distribution confirms the inference about the endogenic origin and crenogenic nature of the solutions maintained in the basin (cf. Hancock et al., 1999). The tectonic control of the spring distribution imposes some further constraints about the groundwater supply mechanisms. As a rule, fault-controlled springs are intermittently active (Sibson, 1987); hence, their efficiency fluctuates with time. In the studied case, the periods of water pumping in a given site might have alternated with weakening or even total vanishing of the source(s). This oscillation might have coincided with pulsing seismic activity. Small synsedimentary dilatancy cracks possibly record the paleoearthquake motion in the studied deposits (Fig. 8F), confirming this inference. Climatic Controls From the characteristics of the sedimentary complex of the Wo´zniki Limestone, one may infer that, aside from the endogenic factors, the sedimentary processes were also stimulated by alternating climatic conditions. The estimation of the climate significance is, however, more complex, since climatic influences may be overshadowed by endogenic ones.
148
Szulc et al.
Co-occurrence of the gypsum and limestones in the lower part of the succession indicates evaporitic precipitation under a negative precipitation/evaporation balance, which is typical for semidry climates (Figs. 3–5). The gradual withdrawal of the gypsum and the concomitant increasing occurrence of rhizoid fabrics observed upsection most likely reflect climate humidification. Additionally, the previously discussed δ13C curves indicate a similar trend, reflecting general climate pluvialization during the time under discussion. The “dry” facies of the Upper Gipskeuper were gradually replaced by more “wet” sediments typical for the Steinmergelkeuper facies in the entire Central European basin. This climatic trend was driven most probably by a drift of the mid-European block outside the subtropical dry belt, i.e., into the higher paleolatitudes (45–50°) (Szulc, 2007). The paleoclimatic conditions in the Late Triassic also fluctuated in rhythms of shorter frequencies (Simms and Ruffel, 1990; Reinhardt and Ricken, 2000). The shorter-term changes are mostly attributed to the orbitally controlled fluctuation in paleomonsoonal circulation, which played an important role for the mid-European area in Triassic times (Kutzbach, 1994; Parrish, 1999). These short-term climatic changes are manifested, first of all, by alluvial clastic intercalations enclosed within the palustrine limestones. Also the karstification, certification, and dolomitization phenomena intimately related to clastic intervals mark breaks in carbonate sedimentation and subaerial weathering on one hand and indicate an increasing influence of meteoric waters (i.e., climate pluvialization) on the other hand. The chert replacement of the calcite and sulfates most likely proceeded under fluctuating pH conditions, i.e., between the alkaline and the normal conditions when the dissolved silica was reprecipitated (Fig. 10G). This process may also be attributed to climatic fluctuations; during the dry periods, the evaporated solutions became alkaline, while the pluvialization led to a decrease in their alkalinity. If the subaerial exposure events coincided with dry climatic phases, the exposed palustrine limestones underwent pedogenesis. Genetic Model of the Wo´zniki Limestone As discussed already, the majority of the carbonates of the Wo´zniki Limestone is genetically related to solutions supplied by a huge spring system controlled by the active fault. As also noted, the fault-controlled spring activity fluctuated with time, so the history of Wo´zniki Limestone may be divided into periods of carbonate deposition and nondeposition. The lithological succession of the Wo´zniki Limestone indicates, however, that the endogenic cycles were also being modified by climatically controlled factors (clastic input, pedogenic alternation, karstification) superimposed upon the endogenic mechanism. Lithological variation (i.e., limestones vs. clastics) within the Wo´zniki Limestone rock assemblage necessitates a changing ratio of the endogenic versus meteoric solutions supply. We can envision four scenarios of this interplay:
Scenario 1. Endogenic Alimentation Active, Climate Arid During the dry periods, the carbonates (and gypsum) precipitated from the crenogenic, undiluted solutions. Scenario 2. Endogenic Alimentation Ceased, Climate Semiarid Carbonate deposition stopped and calcrete formed. Scenario 3. Endogenic Alimentation Ceased, Climate Wetter Carbonate sedimentation stopped, and karstification became a particularly important process affecting the limestones. Dolomitization progressed. The intimate association between detrital sedimentation and dolomitization processes suggests that the dolomitization was driven by an increase of the meteoric water input during humid periods. If the pluvial period was prolonged, some silicate minerals (feldspars, chlorite) underwent alteration and released (among others) Mg2+. This led to dolomitization of the karstified palustrine limestones. The inference is also supported by the stable isotopes data. The proposed dolomitization model is contrary to those reported from similar continental settings where calcite-todolomite transformation is attributed to fluids evaporated under arid climatic conditions (Richter, 1985; Spötl and Wright, 1992; Colson and Cojan, 1996; Warren, 1999; Sinha and Raymahashay, 2004). Scenario 4. Endogenic Alimentation Active, Climate Wetter In this scenario, the denudation processes prevailed. The muddy and clayey sediments derived from outside the spring zone were eroded and redeposited. This process led to clastic dilution of the carbonate-bearing source waters and hindered unconstrained precipitation of CaCO3. To summarize, as the presented data suggest, the switch between the carbonate and clastic sedimentation may be plausibly explained as an effect of climatic fluctuations between the dry and pluvial periods (Fig. 14). The dry periods favored deposition of carbonate sediments (travertines and palustrine limestones, calcretes), while the pluvialization obstructed carbonate sedimentation and promoted their denudation, dolomitization, and replacement by fine-grained, detrital deposition. This model is supported by the clay mineral composition. The clay minerals enclosed in the carbonate deposits are dominated by illite, which is characteristic of drier conditions, whereas the clastic, fluvial intercalations display increasing contribution of kaolinite, which forms preferably under humid conditions (Ruffel et al., 2002). Palustrine facies are commonly defined as subaerially transformed lake-margin deposits (see discussion in Alonso-Zarza, 2003). This definition is, however, difficult to apply for cases where the palustrine environment is not preceded by a lacustrine stage, as in this case. The most probable sedimentary environment of the Wo´zniki Limestone would be a low-relief area with swampy depressions filled with gradually evaporated water. The paucity of typical lacustrine sediments and fossils indicates
Upper Triassic freshwater limestones from Poland
149
A
B
that limnic conditions were limited to very small and ephemeral ponds. Similar palustrine basins have been interpreted either as floodplain ponds (Huerta and Armenteros, 2005) or as groundwater wetlands (Tandon and Andrews, 2001). In both cases, the basins would have been maintained by meteoric and fluvial waters. The specific character of the palustrine Wo´zniki Limestone depends on the crenogenic recharge system maintaining the basin. The endogenic nature of the water supply resulted in an unconventional relationship between the climate and continental carbonate sedimentation. In contrast to the typical situation, the crenogenic palustrine carbonates developed in arid conditions, and they vanished with climate pluvialization. CONCLUSIONS The Upper Triassic (Norian) Wo´zniki Limestone from Upper Silesia is composed of freshwater carbonate sediment formed in swampy depressions, fed by a huge, fault-bound
Figure 14. Genetic model of the Wo´zniki Limestone. (A) Deposition of the limestone during arid periods. (B) Degradation of the limestone and alluvial sedimentation during humid periods.
spring system. The travertines formed adjacent to the spring orifices, while in the more distal area, the palustrine carbonates were deposited. The crenogenic character of the solution supply imposes a very specific model of palustrine carbonate sedimentation. The pure carbonates formed mainly during dry intervals, whereas the climate pluvialization involved meteoric and clastic dilution and the final withdrawal of calcareous deposition. This model is opposite to some extent to the typical model of freshwater carbonate sedimentation under humid conditions, which ceases, in turn, under dry climatic conditions. The model presented here is supported by geochemical signals and biotic indicators. It is remarkable that the limestones are very poor in fossils (both faunal and floral), which are more common in the fluvial (humid) intervals. In addition to the shorter pluvialization episodes, a secular trend in climate humidification has been identified. This trend reflects a general climate evolution forced by drift of the central Europe block to the higher paleolatitudinal zone.
150
Szulc et al.
ACKNOWLEDGMENTS The study was financed by the State Committee for Scientific Research grant 6PO4D 023 19. The drillings were possible due to the permission of: Czes⁄law Kawalec, Lidia Kucharczyk, Maria Nakie⁄la, Stanis⁄law Proszowski, Janusz Psonka, and the ´ askie. We thank Renata Jach owners of the Brickyard in Lipie Sl˛ drawing the figures and Bogus⁄law Ko⁄lodziej for his assistance with photographic works. REFERENCES CITED Alonso-Zarza, A., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: Earth-Science Reviews, v. 60, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Beutler, G., Hauschke, N., and Nitsch, E., 1999, Faziesentwicklung des Keupers im Germanischen Becken, in Hauschke, N., and Wilde, V., eds., Trias. Eine Ganz Andere Welt: München, Verlag Dr. Friedrich Pfeil, p. 129–174. Bilan, W., 1976, Stratygrafia górnego triasu wschodniego obrze˙zenia Górno´slaskiego ˛ Zagłebia W˛eglowego: Zeszyty Naukowe AGH, Geologia Kwartalnik, v. 2, p. 1–73 (in Polish). Bogacz, K., D˙zuły´nski, S., and Hara´nczyk, C., 1970, Ore-filled hydrothermal karst features in the Triassic of the Upper Silesian Coal Basin: Acta Geologica Polonica, v. 20, p. 247–267. Chafetz, H.S., and Folk, R.L., 1983, Pisoliths (pisoids) in Quaternary travertines of Tivoli, Italy, in Peryt, T., ed., Coated Grains: Berlin, Springer, p. 474–487. Colson, J., and Cojan, J., 1996, Groundwater dolocretes in a lake-marginal environment: An alternative model for dolocrete formation in continental settings (Danian of the Provence Basin, France): Sedimentology, v. 43, p. 175–188. de Vos, J.H., and Virgo, K.J., 1969, Soil structure in Vertisols of the Blue Nile clay plains: Journal of Soil Sciences, v. 20, p. 189–206. Fijałkowska-Mader, A., 1999, Palynostratigraphy, palaeoecology and palaeoclimatology of the Triassic in south eastern Poland: Zentralblatt für Geologie und Paleontologie, T. I, 1998, p. 601–627. Fitzpatrick, E.A., 1986, An Introduction to Soil Science: London, Longman, 176 p. Folk, R.L., Chafetz, H.S., and Tiezzi, P., 1985, Bizarre form of depositional and diagenetic calcite in hot-spring travertine, central Italy, in Schneidermann, N., and Harris, P.M., eds., Carbonate Cements: Tulsa, Society of Economic Paleontologists and Mineralogists Special Publication 36, p. 349–369. Freytet, P., and Plaziat, J.-C., 1982, Continental carbonate sedimentation and pedogenesis—Late Cretaceous and Early Tertiary of southern France: Contributions to Sedimentology, v. 12, p. 1–213. G˛asiorowski, S.M., and Piekarska, E., 1976, Wo´zniki Limestone (?Lower Jurassic, Upper Silesia): Bulletin de l’Académie des Sciences, Série des Sciences de la Terre, v. 24, p. 177–182. G˛asiorowski, S.M., and Piekarska, E., 1986, Origin of the Wo´zniki Limestone, in Teisseyre, A.K., ed., 7th IAS European Regional Meeting Kraków, Poland, Excursion Guidebook: Wrocław, Ossolineum, p. 182–185. Gómez-Gras, D., and Alonso-Zarza, A.M., 2003, Reworked calcretes: Their significance in the reconstruction of alluvial sequences (Permian and Triassic, Minorca, Balearic Islands, Spain): Sedimentary Geology, v. 158, p. 299–319, doi: 10.1016/S0037-0738(02)00315-9. Grodzicka-Szymanko, W., and Orłowska-Zwoli´nska, T., 1972, Stratygrafia górnego triasu NE cz˛e s´ ci obrze˙zenia Górnosl˛askiego Zagłebia W˛eglowego: Kwartalnik Geologiczny, v. 16, p. 216–232 (in Polish). Guo, L., and Riding, R., 1994, Origin and diagenesis of Quaternary shrub facies, Rapolane Terme, central Italy: Sedimentology, v. 41, p. 499–520. Hancock, P.L., Chalmers, R.M.L., Altunel, E., and Çakir, Z., 1999, Travitonics: Using travertines in active fault studies: Journal of Structural Geology, v. 21, p. 903–916, doi: 10.1016/S0191-8141(99)00061-9. Heunisch, C., 1999, Die Bedeutung der Palynologie fuer Biostratigraphie und Fazies in der Germanischen Trias, in Hauschke, N., and Wilde, V., eds., Trias. Eine Ganz Andere Welt: München, Verlag Dr. Friedrich Pfeil, p. 207–220.
Hoefs, J., 1997, Stable Isotope Geochemistry: Berlin, Springer, 201 p. Huerta, P., and Armenteros, I., 2005, Calcrete and palustrine assemblages on a distal alluvial-floodplain: A response to local subsidence (Miocene of the Duero basin, Spain): Sedimentary Geology, v. 177, p. 253–270, doi: 10.1016/j.sedgeo.2005.03.007. Kutzbach, J.E., 1994, Idealized Pangean climates: Sensitivity to orbital change, in Klein, G.D., ed., Paleoclimate, Tectonism, and Sedimentation during Accretion, Zenith and Breakup of Supercontinents: Geological Society of America Special Paper 289, p. 41–55. Leslie, A.B., Tucker, M.E., and Spiro, B., 1992, A sedimentological and stable isotopic study of travertines and associated sediments within Upper Triassic lacustrine limestones, south Wells, U.K.: Sedimentology, v. 39, p. 613–629. Lewandowska, A., Gradzi´nski, M., and Szulc, J., 2001, Noncarbonate components of Wo´zniki Limestone: Preliminary results: Mineralogical Society of Poland Special Paper 19, p. 109–111. Michael, R., 1912, Beitraege zur Kenntnis des Keupers im noerdlichen Oberschlesien: Jahrbuch der Koeniglch-Preussischen Geologischen Landesamt B, v. 33, p. 72–97. Morawska, A., 1997, The Lubliniec fracture zone: Boundary of the Upper Silesian and Małopolska massifs, southern Poland: Annales Societatis Geologorum Poloniae, v. 67, p. 429–437. Orłowska-Zwoli´nska, T., 1983, Palinostratygrafia epikontynentalnych osadów wy˙zszego triasu w Polsce: Instytut Geologiczny, Prace, v. 104, p. 1–89 (in Polish). Parrish, J.T., 1999, Pangaea und das Klima der Trias, in Hauschke, N., and Wilde, V., eds., Trias. Eine Ganz Andere Welt: München, Verlag Dr. Friedrich Pfeil, p. 37–42. Pentecost, A., 1990, The formation of travertine shrubs: Mammoth Hot Springs, Wyoming: Geological Magazine, v. 127, p. 159–168. Reinhardt, L., and Ricken, W., 2000, The stratigraphic and geochemical record of Playa cycles: Monitoring a Pangean monsoon-like system (Triassic, Middle Keuper, S. Germany): Palaeogeography, Palaeoclimatology, Palaeoecology, v. 161, p. 205–227, doi: 10.1016/S0031-0182(00)00124-3. Reitner, J., 1993, Modern cryptic microbialite/metazoan facies from Lizard Island (Great Barrier Reef, Australia), formation and concepts: Facies, v. 29, p. 3–40. Retallack, G.J., 2001, Soils of the Past: Oxford, Blackwell, 404 p. Richter, D.K., 1985, Die Dolomite der Evaporit- und der Dolcrete-Playasequenz im mittleren Keuper bei Coburg (NE Bayern): Neues Jahrbuch für Geologie und Palaontologie, Abhandlungen, v. 170, p. 87–128. Roemer, F., 1867, Neuere Beobachtungen über die Gliederung des Keupers und der ihn zunnächst überlagernden Abtheilung der Juraformation in Oberschlesien und in den angrenzenden Theilen von Polen: Zeitschrift der Deutschen Geologischen Gesselschaft, v. 14, p. 255–269. Ruffel, A., Kinley, J.M., and Worden, R.H., 2002, Comparison of clay mineral stratigraphy to other proxy palaeoclimate indicators in the Mesozoic of NW Europe: Royal Society of London, Philosophical Transactions, ser. A, v. 360, p. 675–693. Sherman, G.D., Schultz, F., and Always, F.J., 1962, Dolomite in soils of the Red River Valley, Minnesota: Soil Science, v. 94, p. 304–313. Shinn, E.A., and Lidz, B., 1987, Blackened limestone pebbles; fire at subaerial unconformities, in James, N.P., and Choquette, P.W., eds., Paleokarst: New York, Springer, p. 117–131. Sibson, R.H., 1987, Earthquake rupturing as mineralizing agent in hydrothermal systems: Geology, v. 15, p. 701–704, doi: 10.1130/00917613(1987)15<701:ERAAMA>2.0.CO;2. Simms, M.J., and Ruffel, A.H., 1990, Climatic and biotic change in the Late Triassic: Journal of the Geological Society of London, v. 147, p. 321–327. Sinha, R., and Raymahashay, B.C., 2004, Evaporite mineralogy and geochemical evolution of the Sambhar Salt Lake, Rajasthan, India: Sedimentary Geology, v. 166, p. 59–71, doi: 10.1016/j.sedgeo.2003.11.021. Słowakiewicz, M., 2003, Fluid inclusion data in calcite from the Upper Triassic hot-spring travertines in southern Poland: Journal of Geochemical Exploration, v. 78–79, p. 123–126, doi: 10.1016/S0375-6742(03)00062-1. Spötl, C., and Wright, V.P., 1992, Groundwater dolocretes from the Upper Triassic of the Paris basin, France: A case study of an arid, continental, diagenetic facies: Sedimentology, v. 39, p. 1110–1136. Strasser, A., 1984, Black-pebble occurrence and genesis in Holocene carbonate sediments (Florida Keys, Bahamas, and Tunisia): Journal of Sedimentary Petrology, v. 54, p. 1097–1109. Szulc, J., 1997, Role of algae and microbes in formation of palustrine carbonates: An example from the Upper Triassic, southern Poland, in 3rd
Upper Triassic freshwater limestones from Poland Regional Symposium of International Fossil Algae Association, 3rd International Meeting of ICP 380: Guidebook and Abstracts: Cracow, Institute of Geological Science, Jagiellonian University, p. 86. Szulc, J., 2005, Sedimentary environments of the vertebrate-bearing Norian deposits from Krasiejow, Upper Silesia, Poland: Hallesches Jahrbuch für Geowisschenschaften, Reihe B, v. 19, p. 161–170. Szulc, J., 2007, Climate evolution in the Tethys domain during Triassic times and its controls, in McCann, T., ed., Geology of Central Europe: Triassic: Geological Society [London] (in press). Szulc, J., Gradzi´nski, M., and Lewandowska, A., 2002, The Upper Triassic spring-fed palustrine basin in Upper Silesia, southern Poland: Schriftenreihe der Deutsche Geologische Gesselschaft, Sediment 2002, Darmstadt, v. 17, p. 200–201. Tandon, S.K., and Andrews, J.E., 2001, Lithofacies associations and stable
151
isotopes of palustrine and calcrete carbonates: Examples from an India Maastrichtian regolith: Sedimentology, v. 48, p. 339–355, doi: 10.1046/ j.1365-3091.2001.00367.x. Turner, P., 1980, Continental Red Beds: Amsterdam, Elsevier, Developments in Sedimentology, v. 29, 562 p. Warren, J.K., 1999, Evaporites: Their Evolution and Economics: Oxford, Blackwell Science, 438 p. Znosko, J., 1960, Jura dolna i s´rodkowa okolic Cz˛estochowy, in Ró˙zycki, S.Z., ed., Przewodnik XXXIII Zjazdu Polskiego Towarzystwa Geologicznego, Cz˛estochowa, 4–6 wrze´snia 1960: Warszawa, Polskie Towarzystwo Geologiczne, p. 13–27 (in Polish). MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
A recent analogue for palustrine carbonate environments: The Quaternary deposits of Las Tablas de Daimiel wetlands, Ciudad Real, Spain Ana M. Alonso-Zarza Departamento de Petrología y Geoquímica, Fac. CC. Geológicas, Universidad Complutense, 28040 Madrid, Spain Miriam Dorado-Valiño Ana Valdeolmillos-Rodríguez M. Blanca Ruiz-Zapata Departamento de Geología, Universidad de Alcalá, Edificio de Ciencias, Campus Universitario, 28871 Alcalá de Henares, Madrid, Spain ABSTRACT Las Tablas de Daimiel, Spain, is one of the scarce, freshwater wetlands areas still preserved in southern Europe. The wetland is fed by surface and groundwater. We studied the Quaternary sedimentary record of Las Tablas in a drill hole that penetrated 38.5 m of shallow-lake and fluvial deposits. Differences in the dominantly micritic muds indicate three main stages of development. In the lowest stage, unit A, (Lower? to Middle Pleistocene) the slightly saline wetland developed under a relatively arid climate that favored slow flow movement of the fluvial system and the disconnection of the ponded areas. In the intermediate stage, unit B, (Middle to Upper Pleistocene) extensive peat developed during wetter conditions. Biosiliceous sediments (diatoms and sponge spicules) also accumulated in this swampy setting. In the latter stage, unit C, (Upper Pleistocene to Holocene) palustrine carbonates formed in a freshwater environment with desiccation events, followed by fluvial reworking of the lake margins. Lithification of these deposits was relatively fast (<10,000 yr). The studies of the core, including mineralogy, petrography, stables isotopes, and pollen analyses, indicate that these sediments are similar to those of ancient palustrine sequences. Therefore, Las Tablas can be considered as a recent analogue for freshwater palustrine systems that have no marine influence. These systems are very sensitive to changes in climate or base level, and their study is needed to better understand the terrestrial sedimentary record. Study is needed also to determine how to preserve these wetlands. Keywords: palustrine carbonates, wetlands, Quaternary, peat, vegetation, Spain. RESUMEN Las Tablas de Daimiel constituyen uno de los escasos humedales de agua dulce que aún se conservan en el sur de Europa, concretamente en España. El humedal está abastecido por aguas superficiales y subterráneas. El registro sedimentario Alonso-Zarza, A.M., Dorado-Valiño, M., Valdeolmillos-Rodríguez, A., and Ruiz-Zapata, M.B., 2006, A recent analogue for palustrine carbonate environments: The Quaternary deposits of Las Tablas de Daimiel wetlands, Ciudad Real, Spain, in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 153–168, doi: 10.1130/2006.2416(10). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
153
154
Alonso-Zarza et al. Cuaternario de Las Tablas, se ha estudiado mediante un sondeo en el que se cortaron 38.5 m de depósitos lacustres someros y fluviales, esencialmente carbonatos micríticos. Las características de los sedimentos permiten diferenciar tres principales etapas de sedimentación. En la primera etapa, unidad A, (Pleistoceno Inferior? a Medio) el humedal era ligeramente salino y se desarrolló bajo un clima relativamente árido que favoreció el flujo más lento del sistema fluvial, permitiendo la desconexión de las zonas encharcadas. En la etapa intermedia, unidad B, (Pleistoceno Medio a Superior) el amplio desarrollo de turberas indica condiciones más húmedas, en las que dentro de las áreas pantanosas también se depositaron sedimentos biosilíceos (diatomeas y espículas de esponjas). En la última etapa, unidad C, (Pleistoceno Superior a Holoceno) los sedimentos característicos son carbonatos palustres formados en un ambiente de agua dulce con eventos de desecación seguidos de posterior retrabajamiento de los márgenes lacustres por canales fluviales. La litificación de estos depósitos fue relativamente rápida (<10000 años). El estudio llevado a cabo en el sondeo (mineralogía, petrografía, isótopos estables, análisis polínicos) indica que estos sedimentos son similares a las secuencias palustres del registro geológico. Por tanto, Las Tablas de Daimiel pueden considerarse como un análogo reciente para sistemas palustres de agua dulce que no tengan influencia marina. Estos sistemas son muy sensibles a cambios climáticos y/o del nivel de base, por lo que su estudio es necesario para conocer mejor el registro sedimentario continental, pero también para preservar estos humedales. Palabras clave: carbonatos palustres, humedales, Cuaternario, turba, vegetación, España.
INTRODUCTION Palustrine deposits are widely recognized in the sedimentary record. They are very common in Mesozoic and Cenozoic terrestrial basins (see Gierlowski-Kordesch and Kelts, 2000), and some of the classical examples come from southern France (Freytet and Plaziat, 1982). They are also present in more ancient deposits, such as the Devonian of New York (Dunagan and Driese, 1999). In this latter case, palustrine deposits demonstrate evidence for the occupation of land by plants. The Florida Everglades are commonly cited as a recent analogue for ancient palustrine carbonate sequences (Platt and Wright, 1992); however, we suggest two additional constraints on any proposed recent analogue for palustrine sediments. The first is that the analogue should include a clear sedimentary record that allows its interpretation, and the other is that the water composition be mostly freshwater. Wetland areas of inland terrains can easily be isolated from marine influence, so if they have been accumulating for a long time, they should contain a useful sedimentary record. The relative contribution of groundwater and surface water to the inundated area seems important in distinguishing between lakes and wetlands. Following Currey (1990), it is considered that wetlands are fed primarily by groundwater, when the groundwater table intersects the landscape, but that wetlands also may receive surface water by sheet flows; conversely, lakes are primarily fed by surface water from rivers and streams that enter the basin, with secondary groundwater contributions (Dunagan and Turner, 2004). Even in modern environments this differentiation is difficult to make, because
as a consequence of climate, the feeding mechanism may vary with time. Such is the case of Las Tablas de Daimiel, which are wetlands fed by both surface and groundwater. This is also the case of many ancient palustrine deposits (Sanz et al., 1995; Gierlowski-Kordesch, 1998). Las Tablas de Daimiel, in the interior of the Iberian Peninsula, are a good example of wetland areas totally isolated from any marine influence because they are at present one of the last examples of wetlands in southern Europe and contain a unique freshwater ecosystem (Álvarez-Cobelas and Cirujano, 1996). They contain a relatively thick sedimentary record, including almost 40 m of freshwater fluvio-lacustrine carbonates, mostly chalky, but also indurated hard freshwater carbonates with some peat intervals. These sediments show many of the features as ancient palustrine deposits, such as desiccation cracks, bioturbation, root traces, and even mottling. Moreover, the fact that some of them are indurated, even though they are recent, is proof that diagenetic processes operate relatively fast in these deposits, even if their initial composition is presumed stable, mostly low-magnesium calcite. Ancient examples also indicate that palustrine deposits easily undergo diagenetic processes without significant burial (Wright et al., 1997; Anadón et al., 2000), but there are few constraints regarding the timing of lithification and induration. In this paper, we describe the recent and ancient sedimentary record of Las Tablas de Daimiel through the detailed study of a 38.5 m core by utilizing mineralogy, petrography, palynology, and isotope geochemistry. This multidisciplinary approach provides a data set that allows a better understanding and a clearer interpretation of the sedimentary and biologi-
Recent wetland-palustrine deposits in Spain cal processes, water chemistry, and diagenesis of palustrine or “wetlands” environments in Las Tablas. We suggest that Las Tablas de Daimiel wetlands are an analogue for palustrine carbonate deposits not influenced by marine processes. GEOLOGICAL AND ENVIRONMENTAL SETTING Las Tablas de Daimiel are located in the central part of Spain (Fig. 1) in the so-called “Llanura Manchega” (Mancha
155
Plain) (Pérez-González, 1982, 1996). The area is characterized by several flat surfaces, situated at different elevations. The fluvial network is scarcely incised and shows poorly defined channels with wide ponded areas and a low degree of terrace development (Rodríguez García and Pérez-González, 2002). Las Tablas de Daimiel developed on a Pliocene erosional surface, named the Lower Surface of Llanura Manchega, which has abundant dissolution features, such as dolines and uvalas, that are up to 900 m across and 10–15 m deep and contain temporal
0º
A
FRANCE SPAIN
PORTUGAL
IBERIAN
PENINSULA Las Tablas de Daimiel
40ºN
Ciudad Real Province
AT L
AN
TI
C
AN
SEA
A
RR
E DIT ME
OC E
AN NE
NORTH-AFRICA
B Gigüela River
Las Tablas de Daimiel
Guadiana River LT-199906 3°41’45’’W 39°07’58’’N
Outline of the National Park borders
0
5
Figure 1. (A) Location of Las Tablas de Daimiel in the interior of the Iberian Peninsula. (B) Aerial view of the wetland.
10 km
156
Alonso-Zarza et al.
ponds. The Pliocene surface is incised less than 20 m by the main rivers (Portero and Ramírez, 1988): Záncara, Gigüela, Guadiana, and Azuer. Las Tablas comprise the largest wetland of the southern Meseta of Spain. Initially, they formed by the flooding of Guadiana and Gigüela Rivers, together with the upwelling of groundwaters through many springs referred to as “eyes.” The relative feeding by surface versus groundwater discharge varied in time and space. The total wetland area thirty years ago was ~6000 ha, but today, it has been reduced to 1675 ha, included in the 1928 ha of Las Tablas de Daimiel National Park. All of the multidisciplinary studies carried out recently at this national park are recorded in the extensive monograph edited by ÁlvarezCobelas and Cirujano (1996), which provides much of the data presented in this environmental setting. The climatic setting of the Las Tablas is classified as cold, temperate continental with a dry season. Isotherms range from 12 to 14 °C, rainfall varies from 400 to 500 mm per year, and potential evapotranspiration averages 778 mm per year (Álvarez-Cobelas and Cirujano, 1996). At present, surface water on the Las Tablas wetlands is shallow, usually less that 1 m in depth (Fig. 2). The aquatic environment is highly turbid with high sedimentation rates. The wetlands are dominated by carbonate and sulfate ions, and the waters are hypertrophic. High levels of organic pollution come from the surrounding towns; however, pollution by pesticides and heavy metals is negligible (Álvarez-Cobelas and Cirujano, 1996; Dorado Valiño et al., 2004). MATERIAL AND METHODS A truck-mounted pneumatic drill was used to obtain the 10-cm-diameter Las Tablas de Daimiel (LT) core from the central area of Las Tablas de Daimiel National Park, on the edge of
Laguna Permanente (Fig. 1). The mineralogy of the sediments was determined using a Philips XDR system operating at 40 kV and 30 mA with monochromated CuKα radiation. We studied only 15 stained thin sections (due to the paucity of well-indurated carbonate beds) by transmission light microscopy. We performed scanning electron microscopy (SEM) with a JEOL 6.400 working at 20 kV on gold-covered surfaces. Isotope measurements were performed on powdered samples of the chalky limestones. The analyses were performed at the Stable Isotope Laboratory of the Estación Experimental del Zaidín (Consejo Superior de Investigationes Cientificas, Spain). Samples were ground to <200 mesh and treated with 100% phosphoric acid. Isotopic ratios were measured by a Finnigan MAT 252 mass spectrometer. Carrara and EEZ-1 were used as a standard previously calibrated to NBS-18 and NBS-19. Most of the analyses focused on the middle 8.6 m section of the core (14 samples), with fewer (11) samples from the remaining 29.9 m. Data are expressed relative to Peedee belemnite (PDB). Pollen was extracted from the sediment by flotation on Thoulet’s solution (Goeury and Beaulieu, 1979) without acetolysis; 315 pollen samples were analyzed. Radiocarbon ages were obtained from organic sediment samples by Beta Analytic Inc. (Miami, Florida, USA) using traditional techniques (Table 1). The Th/U analyses were performed at Jaume Almera Institute using the techniques of J.L. Bischoff (U.S. Geological Survey, Menlo Park; R. Juliá, 2002, personal commun.). Sample LT-86 (13.42 m) was dated at 180,000 yr B.P. The amino acid racemization analyses for age estimation on gastropods (Table 2) were carried out at North East Amino Acid Racemization Laboratory (NEAAR, University of York, UK). The majority of the amino acids resolvable had reached equilibrium, allowing only a minimum age. Based on temperature estimates for the region, the minimum age is estimated as older than oxygen isotope stage (OIS) 7.
Figure 2. View of the wetland. Reeds are common in the margins of the shallowwater bodies.
Recent wetland-palustrine deposits in Spain
Laboratory reference E - 135635 E - 135636 E - 135637 E - 135638 E - 135639 E - 135640 E - 132973 E - 132974
TABLE 1. RADIOCARBON DATES IN THE LAS TABLAS DE DAIMIEL CORE 14 Sample Depth Conventional C age (yr B.P.) (m) LT-22 3.30 8500 ± 50 LT-24 3.90 19,010 ± 60 LT-45 6.36 21,120 ± 60 LT-50 6.99 30,980 ± 170 LT-60 8.08 25,160 ± 100 LT-79 12.60 25,280 ± 140 LT-84 13.28 >44,940 (radiocarbon dead) LT-124 16.57 >41,850 (radiocarbon dead)
TABLE 2. GASTROPODS OF LAS TABLAS (LT) DE DAIMIEL SEQUENCE USED FOR AMINO ACID RACEMIZATION ANALYSES Sample Depth Gastropods (m) LT-105 15.39 Pseudotachea splendida (Draparnaud, 1801) LT-120 16.30 Planorbarius metidjensis (Forbes, 1838) LT-130 17.05 Planorbarius metidjensis (Forbes, 1838) Hydrobia sp. Stagnicola cf. fuscus (C. Peiffer, 1821) LT-155 18.35 Planorbarius metidjensis (Forbes, 1838) LT-188 21.18 Anisus sp.
FACIES ANALYSES OF THE SEDIMENTARY RECORD OF THE CORE The study presented in this paper is based mostly on the analyses of the sediments of Las Tablas (LT) core (Fig. 3), combined with some observations of recent sediments outcropping in the Las Tablas area. The Las Tablas core is 38.5 m and is composed mainly of carbonates, with some clays and peat horizons. Dating methods indicate that the lowest dated deposits are Middle Pleistocene, although it is difficult to know exactly the age of the oldest sediments recorded in the core. The most recent deposits are Holocene. By considering the types of sediments that may indicate differences in the characteristics of the wetland environment, the overall sedimentary succession can be subdivided in three units (Fig. 3), which are from bottom to top: Unit A Description The thickness of this lower part (unit A) is 17 m, including mostly white soft-chalky micritic carbonate layers with intercalated indurated limestone beds and gray to beige mudstones. Thickness of the different beds varies from a few centimeters to 1 m. In general, the hard limestone beds are thinner than the soft carbonates. Lamination is rare, and the beds are usually massive. Rare brown mudstones consisting of illite and smectite with minor amounts of quartz and calcite occur in centimeterscale intervals.
157
The chalky carbonate beds are white to beige and contain gastropod fragments, root traces, and occasionally oncoids. Mineralogically, they consist of calcite (15%–100%), clays (0%–45%), dolomite (0%–70%), aragonite (0%–5%), and traces of opal and gypsum. Dolomite, opal, and gypsum traces occur mostly in the lower 5 m of unit A. The clays are illite, sepiolite, and smectites. The occurrence of opal is related to minor amounts of silica phytoliths, and gypsum occurs as lenticular crystals within the lime mud. The mollusc shells, which show important dissolution features, consist of aragonite. Calcite and dolomite crystals are fine (<1 μm across) with a varied subeuhedral morphology. Spherical forms are either high-magnesium calcite (Fig. 4A) and/or dolomite; some dolomite crystals also exhibit rhombohedral morphologies. The more spherical forms of high-magnesium calcite and dolomite occur on phytoliths. SEM studies have shown the presence of calcified tubes, calcareous sponge spicules, siliceous phytoliths, euglenophyte algae, organic films, and fragments of vegetal tissues. Hard indurated limestone beds are relatively thin (5–10 cm), and these deposits display a variety of microfacies. In the lowest portion of unit A, the limestones are micrites with lenticular gypsum molds. The micrite is undergoing recrystallization to microspar and pseudospar (Fig. 4B); the gypsum molds are commonly cemented by calcite. The overall fabric is a coarse calcite crystalline mosaic with pseudomorphs of gypsum. Oncoids occur occasionally and are composed of microsparitic nuclei with gypsum molds enveloped by micritic laminae (Fig. 4C), which also show gypsum pseudomorphs, which in some places display a radial arrangement. At the top of unit A, the limestones are biomicrites with gastropods, charophytes, and ostracodes (Fig. 4D). These lack gypsum molds and evidence of recrystallization, but cementation by calcite spar is common mostly in the intraparticle porosity. Pollen analyses indicate a mean of 2000–20,000 grains of pollen per gram of sample. The main taxa are: Pinus, Cupressaceae, Chenopodiaceae-Amaranthaceae, Poaceae, Asteraceae, and Caryophyllaceae. The most represented aquatic taxa are Cyperaceae, Potamogeton, and Typha monade; these three taxa constitute almost the total of the aquatic pollen compound. Detailed studies of pollen have been carried out by Valdeolmillos Rodríguez (2005). Interpretation These sediments were deposited in a lacustrine environment in which the water changed from slightly saline to fresh. In the lowermost part, the presence of gypsum molds indicates a slightly saline lake system, probably closed, in which micrite precipitated, probably as high-magnesium calcite, induced by processes related to photosynthetic organisms. The micrite also could have formed by abiogenic processes, such as temperature or pCO2 changes, as in many other lake systems (Kelts and Hsü, 1978). The recognition of spherical morphologies in both high-magnesium calcite and the traces of dolomite indicates an organic origin of these carbonates,
0 m
8.500 ± 50 BP 19.010 ± 60 BP
5
21.120 ± 60 BP
Upper PleistoceneHolocene
30.980 ± 170 BP 25.160 ± 100 BP
10
25.280 ± 140 BP > 44.940 BP (radiocarbon dead)
Hippuris
Nuphar
Myriophyllum
Typha tetrade
Typha monade
Potamogeton
> 41.850 BP (radiocarbon dead)
Cyperaceae
180.000 BP (U/Th)
15
Middle Pleistocene 20
Legend Lime mud Intraclasts filled channels Clays
25
Organic-rich levels
Lower? Middle Pleistocene
Sepiolite-rich muds Marls Peat Limestones Silty lime mud
30
35
39
Figure 3. Log of the Las Tablas de Daimiel core. Samples used for different dating techniques are also shown.
Carbonate nodules Oncoliths Stromatoliths Plant debris Gastropods Fragments of molluscs Charophytes Ostracodes Rhizoliths Intraclasts Gypsum traces Samples for isotopic analyses Samples for amino acid racemization
Recent wetland-palustrine deposits in Spain
159
BA
A A
4 m
A C
20 m
A D
0.5 mm
0.5 mm
Figure 4. Lower part of the core (unit A). (A) Scanning electron microscope (SEM) view of spheroidal high-magnesium calcite crystals on the surface of a phytolith. (B) Anhedral calcite crystals due to recrystallization-dissolution processes. (C) Microphotograph of a thin section containing oncoids; lenticular gypsum molds are present in the nucleus and coatings of the oncoid. (D) Hard limestone beds at the top of lower part are biomicrites with ostracodes, gastropods, and charophytes.
probably associated with bacteria (Vasconcelos and McKenzie, 1997); this is confirmed by the occurrence of these spherical bodies on organic remains. One of the most striking features is the presence of gypsum molds in the oncoids, which suggests the presence of certain cyanobacteria groups, such as Synechococcus sp. These can play an important role in biomineralization by calcite, gypsum, and even magnesite, as shown experimentally in natural alkaline waters (Thompson and Ferris, 1990). The occurrence of oncoids indicates some movement of the lake waters, either due to agitation in the lake margin or to the entrance of fluvial channels. However, in general, the waters were quiet, as shown by the fine size of the micrite, the lack of sedimentary structures, and the preservation of delicate components (such as the opal phytoliths). The coarse crystalline texture of some indurated carbonates, the fact that the gypsum molds are filled by coarse spar cement, and the overall spar cementation suggest that diagenesis occurred under fresher meteoric waters, compared to the slightly saline lake waters.
The change toward fresher waters at the top of unit A is shown by the lack of gypsum molds. No clear subaerial exposure features have been recognized, so it seems likely that the lake system was relatively permanent, but shallow, as indicated by the aquatic pollen record, the aquatic gastropods, and the charophytes. Unit B Description This middle part (unit B) includes 8.60 m of soft micritic carbonates with intercalated dark-gray peat levels (Fig. 5). The thickness of the beds is ~0.5 m. There is only one indurated limestone level, but indurated carbonate nodules are common within the marls and peat, particularly at the top of unit B. Mudstone intervals are absent, but clay minerals are present in the peat and soft carbonate beds. The chalky carbonate beds are beige to brown due to staining by organic matter. The carbonates are composed of calcite
160
Alonso-Zarza et al.
Figure 5. View of part of the core (unit B) containing a number of peat beds (darker color).
(70%–100%), aragonite (0%–10%), quartz (0%–15%), clays (0%–20%), and opal (0%–10%), with only traces of gypsum and dolomite. Throughout the drill core, aragonite is found only in the mollusc shells, which show dissolution features. The calcite (low- and high-magnesium calcite) crystals are <1 μm, subeuhedral to rounded, and the crystals are difficult to observe under SEM due to coatings of organic films. The chalky carbonates contain gastropod shells, charophytes, ostracodes, siliceous diatoms (Fig. 6A), siliceous sponge spicules (Fig. 6B), euglenophyte algae, and other phytoliths. Gastropod opercula also have been found. The peat beds occur as differentiated intervals or as lenses within any of the other deposits. They are particularly common from 18 to 13 m. The inorganic component consists of calcite (35%–80%), clays (0%–55%), quartz (0%–35%), aragonite (0%–10%), and opal (0%–10%). Framboids of pyrite are also present in the peat beds (Fig. 6C). In these deposits, there is a higher amount of biosiliceous allochems, and they are commonly
B
A A
40 m
60 m
A D
A C
10 m
0.5 mm
Figure 6. Middle part of the core (unit B). (A) Scanning electron microscope (SEM) view of diatoms and euglenophyte algae (rounded). (B) SEM view of a siliceous sponge spicule. (C) Framboids of pyrite under SEM. (D) Palustrine microfacies of topmost of the middle part of the core.
Recent wetland-palustrine deposits in Spain better preserved and less fragmented than in the chalky carbonates. Fungal filaments and also bacterial microrods are commonly identified with SEM. At the top of this unit (unit B), the peat intervals and the chalky carbonates have hard biomicrite nodules with desiccation features (Fig. 6D) that include cracks (vertical, circumgranular, and horizontal), pseudomicrokarst, and root traces. The mean total organic carbon (TOC) value of this unit is 44.5%, which is related to the very high mean pollen content, which varies from 150,000 to 250,000 grains per gram. The main taxa are: Pinus, evergreen Quercus, deciduous Quercus, Salix, Cupressaceae, Artemisia, Chenopodiaceae-Amaranthaceae, Poaceae, and Asteraceae. There is greater taxonomic diversity of aquatic vegetation than in unit A, including: Cyperaceae, Potamogeton, Typha monade and tetrade, Myriophyllum, Nuphar, and Hippuris. Details on the distribution of these taxa are in Valdeolmillos Rodríguez (2005). Interpretation This middle part of the Las Tablas core (unit B), of Middle Pleistocene age, was deposited in a very shallow lacustrine system that passed from relatively oxidizing conditions at the base to more reduced conditions at the top. The main evidence for this, such as the peat beds, the richness in pollen grains, and the siliceous spicules and diatoms, indicates a swampypaludal system. The anoxic conditions required for the accumulation and preservation of the peat beds were produced in ponded areas that were probably disconnected. Reducing conditions could be established throughout the whole water body, but were probably only permanent in the lake bottom as a result of the combined effects of the high rate of accumulation of organic matter of terrestrial origin transported by slow-flowing streams and in situ growth of plants. This is the case for the marshes of the Miocene Teruel Graben (AlonsoZarza and Calvo, 2000). Although very shallow, these marshes were flooded most of the time, as indicated by the presence of numerous aquatic plants such as Potamogeton, Myriophyllum, Nuphar, and Hippuris. However, evidence of subaerial exposure is present in these deposits as suggested by root traces and desiccation features such as circumgranular cracks. All these exposure features characterize a shift to a more palustrine environment. This is the first indication of the environmental change that occurred at the top of this unit. Variations in water depth are also indicated by the type and diversity of aquatic vegetation. The pollen taxa diversity is lower in stages of subaerial exposure, with emergent plants, such as Cyperaceae, Typha monade and tetrade, dominant. Diatoms are very common in Quaternary lacustrine systems of the Iberian Peninsula (Pérez et al., 2002), and their presence together with the spicules may be related to the abundance of macrophytes, including grasses (Poaceae). These later plants have high silica content, and their high accumulation rates favor the preservation of these delicate components
161
(Reed, 1998). May et al. (1999) have also found white to lightgreen, massive or horizontally laminated diatomite sheets with ostracodes, gastropods, and rootlets within palustrine systems under fluvial influence in the Calama Basin of northern Chile. Siliceous sponges are not common in this type of aquatic system, although they have been described in some modern rivers and lakes from Brazil (Volkmer-Ribeiro and Motta, 1995), and their presence seems to indicate a lowering of lakes waters in swampy systems (Wüst and Bustin, 2003). In these conditions, framboids of pyrite formed either by oxidation of FeS by H2S (Butler and Richard, 2000) or by replacement of greigite (Fe3S4) framboids by pyrite (Wilkin and Barnes, 1997). Unit C Description The uppermost 12.9 m of the core consists exclusively of carbonates with varied textures and hardness (Fig. 3). The scale of bedding varies from decimeters to meters. Three different types of carbonates are recognized. Mineralogically, all the facies consist of low-magnesium calcite (45%–100%), quartz (0%–35%), clays (0%–20%), and aragonite (0%–5%). There are no traces of high-magnesium calcite, dolomite, opal, gypsum, or organic matter. Massive beige to white chalky carbonates are similar to those previously described and contain bioclasts of gastropods, ostracodes, and charophytes. Organic films and filaments are common. There are no diatoms present. Hard massive limestone beds vary from 10 to 100 cm in thickness. They are white and sufficiently hard to allow preparation of standard thin sections. These beds are biomicrites with gastropods, ostracodes, and charophytes. Some intervals display desiccation cracks, alveolar septal structures, pseudomicrokarst, and mottling. In some moldic porosity of molluscs and in the desiccation cracks, there is coarse calcite spar cement, mostly phreatic. The micrite crystals are up to 2 μm, particularly in the indurated beds (Fig. 7A); they are also commonly subhedral. Calcified and noncalcified organic filaments (Fig. 7B) and micritic fecal peloids (Fig. 7C) have been identified with SEM. Carbonate-filled channel deposits with erosional bases are present in the upper part of the core and are ~10 cm thick (unit C). These deposits consist of a packstone of angular micritic intraclasts (0.3–1 cm across), with up to 20% of angular quartz grains averaging 0.2 mm, and some fragments of bioclasts. The interparticle porosity is filled by coarse calcite spar (Fig. 7D). The lack of organic matter in unit C is also demonstrated by the low pollen content; the mean value is 5–10 grains per gram. The main taxa are: Pinus, Betula, Corylus, Olea, evergreen Quercus, deciduous Quercus, Cupressaceae, Ericaceae, Poaceae, Plantago, and Asteraceae. The most common aquatic pollen types are Cyperaceae and Typha monade. The diversity of aquatic pollen is low but increases at the top of the core, corresponding to the Holocene (Valdeolmillos Rodríguez, 2005).
162
Alonso-Zarza et al.
B
A A
20 m
30 m
A D
CA
4 m 40 m
0.5 mm
Figure 7. Upper part of the core (unit C). (A) Scanning electron microscope (SEM) view of the micrite crystals, which are larger and subeuhedral. (B) Calcified organic filaments. (C) Rounded micritic peloids. (D) Microphotograph of intraclastic packstone also containing quartz grains and sparry calcite cement.
Interpretation The deposition of unit C took place after a sedimentary gap of ~140,000 yr, as shown by the age data. The youngest (Upper Pleistocene–Holocene) sediments of Las Tablas de Daimiel were deposited in a fluvio-palustrine system characterized by desiccation events. The oxidizing nature of the environment, which inhibited organic matter accumulation, is reflected not only in the lack of peat beds, but also in the low pollen content. All these features, combined with the carbonate mineralogy (all low-magnesium calcite), indicate freshwater conditions of deposition. Indurated limestones show characteristics similar to the classic French palustrine limestones described by Freytet and Plaziat (1982) and Freytet and Verrecchia (2002). In both the classic French palustrine limestones and in the Las Tablas deposits, the micritic mud is interpreted to have been precipitated in the lake waters both biogenically and physico-chemically. As the water input to the palustrine system fluctuated, desiccation, bioturbation, and pedogenic processes disrupted the carbonate substrate, but also contributed to the induration of the mud (Wright et al., 1997). Therefore, we interpret the highest degree of induration
as a consequence of the shallower recent environments of Las Tablas, as is indicated by the low aquatic pollen diversity and the emergent taxa recorded (Cyperaceae and Typha monade). This caused the sediments to be affected more intensively by early diagenetic processes that could control changes in the size and shape of the initial calcite crystals, which are larger in this part of the core. Phreatic and vadose cementation is also common, providing more evidence that water table fluctuations can contribute to lithification in recent sediments. Unit C was deposited in an environment similar to many of the Tertiary palustrine deposits of the Iberian Peninsula (Alonso-Zarza, et al., 1992), southern France (Freytet, 1984), or even some Devonian sequences from the United States (Dunagan and Driese, 1999). The desiccation events probably occurred with high frequency during late Pleistocene to Holocene time. These events favored the early and incipient lithification of the carbonate beds, which were later reworked when groundwater filled the basins or when ephemeral fluvial channels drained the previous lake margins. The channels filled with intraclasts clearly record these processes. The beds situated between 3.8 and 3.3 m are of special
Recent wetland-palustrine deposits in Spain interest, as this interval tentatively corresponds to the beginning of the Holocene (10,000 yr B.P.), during which the more humid climate could have controlled the development of the channels that reworked the previously desiccated muds and, consequently, were mostly filled by intraclasts (Fig. 7D). The increase of aquatic vegetation diversity corroborates this idea. Evidently, the muds underwent a long desiccation event, probably from ca. 19,000 yr B.P. until the end of the Younger Dryas, when arid and cold climatic conditions prevailed (Dorado Valiño et al., 2002). STABLE ISOTOPE GEOCHEMISTRY Detailed petrographic examination allowed the selection of 25 samples for analysis of stable isotopes (C, O) in the carbonates, using only calcite. Most of the samples selected were beige to gray chalky carbonates. We avoided indurated limestones for two reasons; first, because they are relatively rare in the middle part of the core, the results would not be comparable, and second, to avoid the effect of mixing of the primary micrite with the later cements recognized in these indurated limestones. Thus, the results reflect only the composition of the primary lacustrine muds and not of the cements or recrystallized phases that have been observed in stained thin sections and with SEM. The isotopic composition of the carbonates studied is quite variable (Fig. 8). Values of δ13C range from −3.47‰ to −7.24‰,
163
and δ18O varies from −3.67‰ to −8.82‰. These values are in the observed range of typical freshwater lacustrine carbonates (Tanner, 2000; Tandon and Andrews, 2001; Alonso-Zarza, 2003). Two main features are observed in these data: (1) the overall covariant trend is very poor and negative (σ = −0.22); and (2) there is no clear differentiation between samples from units A and B, lower and middle parts, although their covariant trend is different (unit A = 0.42; unit B = −0.52). However, samples from unit C, upper part, are easy to differentiate, and they display relatively uniform δ18O values (between −6‰ and −7‰). The lack of covariance has been used as an indicator of groundwater input into ancient wetlands (Quade et al., 1995; Dunagan and Turner, 2004) and of open lake systems (Talbot, 1990; Alonso-Zarza and Calvo, 2000). We interpret the lack of covariance to indicate open lake systems for the Las Tablas wetlands because this system is within a fluvial network. The negative covariance of the middle part of the core (unit B) is an indicator of a system in which new water input, either surface or groundwater, is reduced. This condition favors the formation of isolated or disconnected swampy areas because the residence time is longer, which favors depletion in 16O. Similar conditions have been interpreted in some lake deposits from the Miocene of eastern Spain (Utrilla et al., 1998). This water input was coeval with a higher productivity of organic matter (OM), mostly of C3 origin, which accounts for enrichment in 12C.
0.00
-1.00
-2.00
-3.00
-4.00
δ 13C
LOWER PART -5.00
Figure 8. Plot of δ13C versus δ18O of samples of Las Tablas de Daimiel core.
MIDDLE PART -6.00 UPPER PART -7.00
-8.00
-9.00
-10.00 -8.00
-7.00
-6.00
-5.00
-4.00
δ 18O
-3.00
-2.00
-1.00
0.00
164
Alonso-Zarza et al.
Microbial respiration was associated with bacterial sulfate reduction near the sediment-water interface, as indicated by the gypsum molds and the framboidal pyrite; this also explains the lighter carbon values that are typical in these swampy systems (Dunagan and Turner, 2004). The differences in δ13C are not easy to explain because the sampled facies are very similar. In relatively small, short-residence water bodies, variations in primary biological productivity should cause larger differences in δ13C, particularly between unit B and the others, because the organic productivity was much higher in this unit. Similar isotopic differences have been also recognized in other ancient palustrine open lake systems such as in the Teruel Basin (AlonsoZarza and Calvo, 2000). Although there are only five samples from unit C (Upper Pleistocene to Holocene), they have a distinctive signature. Four of these samples have δ18O near −6‰. This is a typical value of oxidizing palustrine deposits. Anadón et al. (2000) considered that values around −6.5‰ represent the isotopic composition of the diagenetic fluids derived from meteoric waters. Tandon and Andrews (2001) examined a large set of isotope data from palustrine carbonates and found an overall narrow range of δ18O values and a wider range of δ13C. The narrow range of oxygen values that we observed in unit C is characteristic of palustrine deposits and reveals the influence of meteoric water during the very early subaerial exposure of each one of the beds. The sample that has a more positive oxygen value than the rest (1‰) may indicate slightly more evaporitic conditions. DISCUSSION Major Controls and Evolution of the System The sedimentary record of Las Tablas de Daimiel shows that this modern palustrine environment consisted of a variety of terrestrial subenvironments, including distal alluvial, fluvial, and lacustrine environments, which occupied this inland plain as far back as early to middle Pleistocene time. These subenvironments suggest that a variety of shallow-water conditions were present, and the subaerial exposure features suggest that water levels in the wetlands have fluctuated since the early Pleistocene. Las Tablas are considered floodplain areas of the fluviatile Gigüela-Záncara system, which drains an area with well-developed karstic features, such as dolines and uvalas (Pérez-González, 1996). The thickness of the Quaternary sequence in Las Tablas, particularly in the study core, may indicate karstic collapse and subsidence within Las Tablas area, which would have assisted in maintaining very shallow water bodies. The permanency of the shallow-water conditions was controlled mostly by the balance between water inputs and losses within the system. Inputs are either surface or groundwater, and in both cases are carried to the flooded areas by the rivers that have very low slopes, such as Azuer, Gigüela, and Guadiana; the latter originates from a spring called Los Ojos. The losses are due mostly to evaporation, infiltration, and, more recently, human uses.
Although all of the deposits show subaerial exposure features, the characteristics of this palustrine system evolved with time as a response to the chemistry of water and to the basin geometry (i.e., accommodation space) for the sediments. The latter was controlled by climate and by the maintained balance between the degree or lack of entrenchment of the river valley, either due to changes in the base level or karstic subsidence. Climate was important in controlling the following: the rate of surface water versus groundwater feeding of the systems; the volume and chemistry of water, which varied depending on the distance the groundwater had to flow; the lithological composition of the catchment areas; the ratio of rock-water interaction; and evaporation rates. All these factors determined the mineralogy of the primary precipitates, their possible transformation during early diagenesis, and the possible establishment of different ecological communities (Gierlowski-Kordesch and Park, 2004). In Las Tablas, the clear sedimentary and biological differences in the core allow for its subdivision into three major units that reflect the complex interplay of all the aforementioned factors. Although a single core is not definitive, because lateral relationships cannot be analyzed, the following is suggested about evolution of this palustrine system (Fig. 9): 1. From the Lower? to Middle Pleistocene (unit A), the water chemistry of the fluvio-lacustrine system evolved from slightly saline to fresh, as indicated by the presence of high-magnesium calcite, dolomite, and gypsum molds. A more arid climate caused increased rates of evaporation of either surface or groundwater, and the systems flowed slowly, leading to the formation of more isolated and/or disconnected areas. The input and movement of surface waters are inferred by the presence of oncoids. Toward the top of unit A, the hydrology of these isolated areas shifted from closed to open as climatic conditions became wetter, resulting in deposition of freshwater palustrine carbonates that lack gypsum and dolomite. It is possible that new accommodation space was created due to the entrenchment of the valley river. 2. During the Middle Pleistocene, unit B, the accommodation space was filled with palustrine carbonates that produced a relatively flat floodplain characterized by disconnected swampy areas. These areas were favorable sites for vegetation development and deposition of organic matter and biosiliceous sediments. 3. In the Upper Pleistocene to Holocene, unit C, age dating indicates a hiatus between units B and C, but this gap is not apparent in the core. The gap cannot be explained at this time due to the lack of additional cores or surface data. However, the surface karstic features suggest karstic subsidence/collapse, which could cause a distal entrenchment of the fluvial systems. Potentially, this could have resulted in the erosion of previously deposited sediments, and subsequently generated new accommodation space filled later by freshwater palustrine carbonates that formed in a very shallow and oxygenated system. Under these conditions,
Lower? -Middle Pleistocene
Middle Pleistocene
Upper Pleistocene-Holocene
Dolines Peat Gypsum Clastics Lacustrine carbonates Palustrine carbonates Figure 9. Sketch of the evolution of the fluvio-palustrine (wetland) system of Las Tablas de Daimiel during the Quaternary.
166
Alonso-Zarza et al. organic matter preservation was inhibited, as shown by the low pollen content. The depositional system had an open hydrology at this time.
Origin of the Micritic Mud and Diagenesis Carbonate deposits that accumulate in lakes may have a variety of different origins, ranging from detrital sources, inorganic precipitates, biochemically induced precipitates, and completely biogenic sources (Kelts and Hsü, 1978). The micrite in the core is primarily fine-grained low-magnesium calcite and high-magnesium calcite, and the crystal shape of the calcite varies throughout the core. Aragonite is found only in the gastropod shells and is not a significant mineral component of the micritic matrix overall. The micrite also includes minor siliceous bioclasts of sponge spicules and diatoms. Dolomite has very limited occurrence in unit A of the core, and detrital carbonates are rare. The origin of fine micritic muds is difficult to determine because of the probable loss of some primary textures during diagenesis. This together with the lack of primary sedimentary textures in the wetland deposits make it difficult to infer the mechanisms by which micritic mud accumulated in the wetlands. Organic activity is recognized throughout the core by the presence of organic filaments, oncoids, and algae. Moreover, the rounded morphology of some of high-magnesium calcite and dolomite crystals recognized in the lower parts (units A and B) of the core may be an indicator of their precipitation under bacterial influence (Castanier et al., 1999), as in some recent lagoonal environments (Vasconcelos and McKenzie, 1997). Hence, biogenically induced carbonate precipitation by bacteria, cyanobacteria, and algal activity were all possible within the water body. Inorganic processes, such as seasonal temperature variations and changes in pCO2 due to degassing (Platt and Wright, 1991), also likely contributed to calcite production. In addition, the regional groundwaters were enriched in calcium and bicarbonate by the presence of Mesozoic carbonate rocks in the drainage basin and the dissolution features observed in the nearby Tertiary carbonate deposits (Portero and Ramírez, 1988). Most of the sediments are soft-chalky carbonate in which the possible effects of diagenesis include: recrystallization, dissolution of lenticular gypsum crystals, cementation and induration of the lime mud, and some dissolution features specially in the aragonite shells, the mineralogy of which is still preserved. Recrystallization and dissolution of lenticular gypsum crystals are recognized only in the lower part of the core (unit A). Both are probably driven by the initial, less-stable mineralogy (highmagnesium calcite, some dolomite and gypsum), so fresh phreatic or vadose waters may be responsible. Cementation occurred throughout the core and is easily seen in the indurated levels; it is particularly present within gypsum molds and between the intraclasts of the top of the core (unit C). The cement phases are mostly of phreatic calcite spar that formed in different stages when base level was rising, allowing the groundwaters to occupy the porosity matrix of the previously deposited sedi-
ments. Throughout the core, but specifically within the more recent sediments (younger than ca. 10,000 yr B.P.), there are beds which are very indurated. This lithification affects mostly the shallower, less organic matter–rich sediments and is probably a response to different dry-wet cycles that favor lithification, without burial. Thus, Las Tablas are one more example that demonstrates that desiccation of very shallow lacustrine deposits favors rapid lithification, as in ancient palustrine deposits (Wright et al., 1997; Anadón et al., 2000), whereas lithification is inhibited in more permanent water bodies that have not undergone such continuous subaerial exposure stages. Recent Wetlands: Analogues for Ancient Palustrine Deposits? Micritic freshwater limestones showing subaerial exposure features are included in the term “palustrine limestones” (Freytet and Plaziat, 1982; Alonso-Zarza, 2003), and these deposits are common in the geological record (see Gierlowski-Kordesch and Kelts, 2000). However, narrowing palustrine deposits to a specific depositional setting may be difficult because they have been recognized in a broad spectrum of depositional environments including: modern and ancient coastal plain settings, such as the Florida Everglades (Platt and Wright, 1992) and the Lower Cretaceous of Croatia (Dini et al., 1998); floodplain areas, as in the Clarks Fork Basin of Wyoming (Bowen and Bloch, 2002), and in marginal areas of larger lacustrine systems (Freytet and Plaziat, 1982). Wright and Platt (1995) indicated that the abundance of palustrine limestones in the sedimentary record is a testimony to the widespread occurrence of seasonal wetlands in the past. Their suggestion to consider recent wetlands as a modern analogue for palustrine deposits prompted Quade et al. (1995) and Dunagan and Turner (2004) to reconsider ancient deposits previously interpreted as lacustrine as possible wetland in origin. The designation “wetland” is complicated by the numerous ecological, sedimentary, and hydrologic parameters used to define such an environment. Here we use a definition of wetlands that includes freshwater flat inundated areas, usually connected to rivers. The more generic definition of wetland is the one that consider that wetlands must have one or more of the following attributes: (1) at least periodically, the land supports predominantly hydrophytes, (2) the substrate is predominantly undrained hydric soil, and (3) the substrate is nonsoil and is saturated with water cover or covered by shallow water at some times during the growing season of each year (Cowardin et al., 1979). More recently, it has been considered that wetlands are landscape units that have a spatially and temporally positive hydric anomaly in relation to the adjacent drier land without being rivers, lakes, or marine environments (DGOH, 1991). Examples of such riverine-influenced wetlands of international significance include some rivers of Greece (Skoulikidis et al., 1998) and the Loboi Swamp in Kenya (Ashley et al., 2004). In spite of the wide use of the term wetland to include ancient
Recent wetland-palustrine deposits in Spain deposits such as Late Cretaceous to early Tertiary ephemeral carbonate lakes of the Andean Basin (Camoin et al., 1997), the clear relationship between wetlands and palustrine carbonates has not been properly established, mainly because of the lack of detailed studies of the carbonate deposits that are forming in modern wetlands. Our work has confirmed that classic palustrine carbonate features such as brecciation, root traces, nodulization, desiccation, and grainification (Freytet and Verrecchia, 2002; AlonsoZarza, 2003) are recognizable in the Quaternary palustrine sequence associated with the Las Tablas de Daimiel wetlands. In our opinion, Las Tablas de Daimiel should be considered as a modern analogue for freshwater palustrine carbonate deposits. CONCLUSIONS The sedimentary record of the Las Tablas core provides evidence that modern wetlands have been present since the Middle Pleistocene in this area. The core may be subdivided into three main depositional stages based on detailed facies analysis and age constraints. In the lower stage (Lower? to Middle Pleistocene, unit A), the wetland deposits included high-magnesium calcite, dolomite, and gypsum molds that accumulated in a slightly saline wetland that developed under a more arid climate in which a slow-flowing fluvial system allowed the formation of disconnected wetland areas. During the Middle Pleistocene (unit B), a shift to wetter climate conditions is indicated by the extensive peat development that was deposited in a shallow vegetated swampy area. From the Upper Pleistocene to Holocene, an open hydrology was established in the wetlands; however, desiccation events were common, which led to the typical hard palustrine limestones associated with the upper part (unit C) of the Las Tablas core. A drainage basin rich in carbonates contributed to the dominance of wetland carbonate deposits during the Quaternary as Ca2+- and CO3H–-rich groundwater and surface water fed into the system. The prevailing semiarid climate favored the inorganic precipitation of carbonates in the water bodies, which combined with biochemical and biogenic deposition. Variations in the climatic conditions, base-level changes, and/or karstic processes allowed the generation of the accommodation space needed to form this relatively thick Quaternary carbonate sequence. One of the causes of the preservation of these wetlands in the geological record is linked to the diagenetic processes that rapidly affect these carbonate sediments, causing induration of the micritic mud in less than 10,000 yr without significant burial. The initial stable low-magnesium calcite mineralogy, such as in the uppermost part of the core (unit C), also contributed to preservation. Conversely, stabilization of aragonite skeletons must be a slower process, as the shells are still aragonitic, even in the lower part of the core (unit A). The study of the Las Tablas core indicates that these carbonate sediments are similar in mineralogy, biota, texture, and isotopic composition to those of ancient palustrine sequences. Therefore, the Las Tablas de Daimiel wetlands should be considered as a modern analogue for freshwater palustrine
167
systems that have no marine influence. These systems are very sensitive to changes in climate or base level, and their study is needed not only to form a better interpretation of the sedimentary terrestrial record, but also to develop a better idea of how to preserve wetlands for the future. ACKNOWLEDGMENTS This work is part of projects DGICYT-BTE-2000-0779 and CGL2005-05953-C02-02. The authors thank Volkmer-Ribeiro for help with spicules identification. Scanning electron microscope (SEM) studies were carried out in Centro de Microscopía Luis Brú. B. Valero and S. Dunagan are thanked for their reviews, which helped to improve the previous manuscript. L. Tanner is thanked for the editorial tasks. REFERENCES CITED Alonso-Zarza, A.M., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: Earth-Science Reviews, v. 60, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Alonso-Zarza, A.M., and Calvo, J.P., 2000, Palustrine sedimentation in an episodically subsiding basin: The Miocene of the northern Teruel Graben (Spain): Palaeogeography, Palaeoclimatology, Palaeoecology, v. 160, p. 1–21, doi: 10.1016/S0031-0182(00)00041-9. Alonso-Zarza, A.M., Calvo, J.P., and García del Cura, M.A., 1992, Palustrine sedimentation and associated features—grainification and pseudomicrokarst—in the middle Miocene (intermediate unit) of the Madrid Basin, Spain: Sedimentary Geology, v. 76, p. 43–61, doi: 10.1016/00370738(92)90138-H. Álvarez-Cobelas, M., and Cirujano, S., eds., 1996, Las Tablas de Daimiel, Ecología Acuática y Sociedad: Madrid, Organismo Autónomo Parques Nacionales, 368 p. Anadón, P., Utrilla, R., and Vázquez, A., 2000, Use of charopyte carbonates as proxy indicators of subtle hydrological and chemical changes in marl lakes: Example form the Miocene Bicorb Basin, eastern Spain: Sedimentary Geology, v. 133, p. 325–347, doi: 10.1016/S0037-0738(00)00047-6. Ashley, G.M., Maitima Mworia, J., Muasya, A.M., Owens, R.B., Driese, S.G., Hover, V.C., Renaut, R.W., Goman, M.F., Mathai, S., and Blatt, S.H., 2004, Sedimentation and recent history of a freshwater wetland in a semiarid environment: Loboi Swamp, Kenya: East Africa: Sedimentology, v. 51, p. 1–21. Bowen, G.J., and Bloch, J.I., 2002, Petrography and geochemistry of floodplain limestones from the Clarks Fork Basin, Wyoming, U.S.A.: Carbonate deposition and fossil accumulation on a Paleocene-Eocene floodplain: Journal of Sedimentary Research, v. 72, p. 46–58. Butler, I.B., and Richard, D., 2000, Framboidal pyrite formation via the oxidation of iron, (II) monosulfide by hydrogen sulphide: Geochimica et Cosmochimica Acta, v. 64, p. 2665–2672, doi: 10.1016/S0016-7037(00)00387-2. Camoin, G., Casanova, J., Rouchy, J.M., Blanc-Valleron, M.M., and Deconinck, J.F., 1997, Environmental controls on perennial and ephemeral carbonate lakes: The central palaeo-Andean Basin of Bolivia during Late Cretaceous to early Tertiary times: Sedimentary Geology, v. 113, p. 1–26, doi: 10.1016/S0037-0738(97)00052-3. Castanier, S., Le Métayer-Levren, G., and Perthuisot, J.P., 1999, Ca-carbonates precipitation and limestone genesis—The microbiologist’s point of view: Sedimentary Geology, v. 126, p. 9–24, doi: 10.1016/S00370738(99)00028-7. Cowardin, L.M., Carter, V., Golet, F.C., and LaRoe, E.T., 1979, Classification of Wetlands and Deepwater Habitats of the United States: U.S. Fish and Wildlife Service FWS/OBS-79/31, 131 p. Currey, D.R., 1990, Quaternary palaeolakes in the evolution of semidesert basins, with special emphasis on Lake Bonneville and the Great Basin, U.S.A.: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 76, p. 189–214, doi: 10.1016/0031-0182(90)90113-L. DGOH, 1991, Estudio de las Zonas Húmedas de la España Peninsular, Inventario y
168
Alonso-Zarza et al.
tipificación, Documento de síntesis: Madrid, Dirección General de Obras Hidráulicas, Ministerio de Obras Públicas y Transportes, 80 p. Dini, M., Tunis, G., and Venturini, S., 1998, Continental, brackish and marine carbonates from the Lower Cretaceous of Kolone-Barbariga (Istria, Croatia): Stratigraphy, sedimentology and geochemistry: Sedimentary Geology, v. 140, p. 245–269. Dorado Valiño, M., Valdeolmillos, A., Ruiz Zapata, M.B., Gil García, M.J., and Bustamante, I., 2002, Climatic changes since the late-glacial/Holocene transition in La Mancha Plain (south-central Iberian Peninsula, Spain) and their incidence on Las Tablas de Daimiel marshlands: Quaternary International, v. 93, p. 73–94, doi: 10.1016/S1040-6182(02)00007-1. Dorado Valiño, M., Bustamante, I., Ruiz Zapata, M.B., Valdeolmillos, A., Álvarez Cobelas, M., and Cirujano, S., 2004, Las Tablas de Daimiel: Un pasado fascinante y un futuro sorprendente, in Guillén Mondéjar, F., and del Ramo Jiménez, A., eds., El Patrimonio Geológico: Cultura, Turismo y Medio Ambiente: Murcia, Sociedad Geológica de España, p. 211–220. Dunagan, S.P., and Driese, S.G., 1999, Control of terrestrial stabilization on late Devonian palustrine carbonate deposition: Catskill Magnafacies, New York, U.S.A.: Journal of Sedimentary Research, v. 69, p. 772–783. Dunagan, S.P., and Turner, C.E., 2004, Regional paleohydrologic and paleoclimatic settings of wetland/lacustrine depositional systems in the Morrison Formation (Upper Jurassic), Western Interior, USA: Sedimentary Geology, v. 167, p. 269–296, doi: 10.1016/j.sedgeo.2004.01.007. Freytet, P., 1984, Les sédiments lacustres carbonatés et leurs transformations par émersion et pédogénèse: Importance de leur identification pour les reconstitutions paléogéographiques: Bulletin Centres Rechercher Exploration, Prod. Elf-Aquitaine v. 8, no. 1, p. 223–246. Freytet, P., and Plaziat, J.C., 1982, Continental carbonate sedimentation and pedogenesis—Late Cretaceous and early Tertiary of southern France: Contributions to Sedimentology, v. 12: Stuttgart, Schweizerbartsche Verlagsbuchhandlung, 213 p. Freytet, P., and Verrecchia, E., 2002, Lacustrine and palustrine carbonate petrography: An overview: Journal of Paleolimnology, v. 27, p. 221–237, doi: 10.1023/A:1014263722766. Gierlowski-Kordesch, E.H., 1998, Carbonate deposition in an ephemeral siliciclastic alluvial system: Jurassic Shuttle Meadow Formation, Newark Supergroup, Hartford Basin, USA: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 140, p. 161–184, doi: 10.1016/S0031-0182(98)00039-X. Gierlowski-Kordesch, E.H., and Kelts, K., eds., 2000, Lake Basins through Space and Time: American Association of Petroleum Geologists Studies in Geology, v. 46, 648 p. Gierlowski-Kordesch, E.H., and Park, L.E., 2004, Comparing species diversity in the modern and fossil record of lakes: The Journal of Geology, v. 112, p. 703–717, doi: 10.1086/424578. Goeury, Cl., and Beaulieu, J.L., 1979, À propos de la concentration du pollen à l’aide de la liquer de Thoulet dans les sédiments minéraux: Pollen et Spores, v. 21, p. 239–251. Kelts, K.R., and Hsü, K., 1978, Freshwater carbonate sedimentation, in Lerman, A., ed., Lakes—Chemistry, Geology, Physics: New York, Springer, p. 295–323. May, G., Hartley, A.J., Stuart, F.M., and Chong, G., 1999, Tectonic signatures in arid continental basins: An example from the Upper Miocene–Pleistocene, Calama Basin, Andean forearc, northern Chile: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 151, p. 55–77, doi: 10.1016/S00310182(99)00016-4. Pérez, A., Luzón, A., Roc, A.C., Soria, A.R., Mayayo, M.J., and Sánchez, J.A., 2002, Sedimentary facies distribution and genesis of a recent carbonaterich saline lake: Gallocanta Lake, Iberian Chain, NE Spain: Sedimentary Geology, v. 148, p. 185–202, doi: 10.1016/S0037-0738(01)00217-2. Pérez-González, A., 1982, Neógeno y Cuaternario de la Llanura Manchega y sus relaciones con la Cuenca del Tajo [Ph.D. thesis]: Madrid, Universidad Complutense, 787 p. Pérez-González, A., 1996, Marco geológico y geomorfológico, in Álvarez Cobelas, M., and Cirujano, S., eds., Las Tablas de Daimiel, Ecología Acuática y Sociedad: Madrid, Publicaciones del Organismo Autónomo Parques Nacionales-Ministerio de Medio Ambiente, p. 31–45. Platt, N.H., and Wright, V.P., 1991, Lacustrine carbonates: Facies models, facies distribution and hydrocarbon aspects, in Anadón, P., et al., eds., Lacustrine Facies Analysis: International Association of Sedimentologists Special Publication 13, p. 57–74.
Platt, N.H., and Wright, V.P., 1992, Palustrine carbonates at the Florida Everglades: Towards an exposure index for the fresh-water environment: Journal of Sedimentary Petrology, v. 62, p. 1058–1071. Portero, J.M., and Ramírez, J.I., 1988, Memoria y Hoja Geológica nº 760 (Daimiel): Madrid, IGME (Instituto Geológico y Minero de España), 58 p. Quade, J., Mifflin, M.D., Pratt, W.L., McCoy, W., and Burckle, L., 1995, Fossil spring deposits in the southern Great Basin and their implications for changes in the water-table levels near Yucca Mountain, Nevada, during Quaternary time: Geological Society of America Bulletin, v. 107, p. 213– 230, doi: 10.1130/0016-7606(1995)107<0213:FSDITS>2.3.CO;2. Reed, J.M., 1998, Diatom preservation in the recent sediment record of Spanish saline lakes: Implications for palaeoclimate study: Journal of Paleolimnology, v. 19, p. 129–137, doi: 10.1023/A:1007948600780. Rodríguez García, J., and Pérez-González, A., 2002, Geomorfología de Las Tablas de Daimiel y su entorno, in Pérez-González, A., et al., eds., Aportaciones a la Geomorfología de España en el Inicio del Tercer Milenio: Instituto Geológico y Minero de España–Sociedad Española de Geomorfología, Madrid, p. 465–473. Sanz, M.E., Alonso-Zarza, A.M., and Calvo, J.P., 1995, Carbonate pond deposits related to semi-arid alluvial systems: Examples from the Tertiary Madrid Basin, Spain: Sedimentology, v. 42, p. 437–452. Skoulikidis, N.T., Bertahas, I., and Koussouris, T., 1998, The environmental state of freshwater resources in Greece (rivers and lakes): Environmental Geology, v. 36, p. 1–17, doi: 10.1007/s002540050315. Talbot, M.R., 1990, A review of the palaeohydrological interpretation of carbon and oxygen isotopic ratios in primary lacustrine carbonates: Chemical Geology, v. 80, p. 261–279. Tandon, S.K., and Andrews, J.E., 2001, Lithofacies associations and stable isotopes of palustrine and calcrete carbonates: Examples from an Indian Maastrichtian regolith: Sedimentology, v. 48, p. 339–355, doi: 10.1046/ j.1365-3091.2001.00367.x. Tanner, L.H., 2000, Palustrine-lacustrine and alluvial facies of the (Norian) Owl Rock Formation (Chinle Group), Four Corners Region, Southwestern U.S.A.: Implications for Late Triassic paleoclimate: Journal of Sedimentary Research, v. 70, p. 1280–1290. Thompson, J.B., and Ferris, F.G., 1990, Cyanobacterial precipitation of gypsum, calcite and magnesite from natural alkaline lake water: Geology, v. 18, p. 995–998, doi: 10.1130/0091-7613(1990)018<0995: CPOGCA>2.3.CO;2. Utrilla, R., Vázquez, A., and Anadón, P., 1998, Paleohydrology of the Upper Miocene Bicorp Lake (eastern Spain) as inferred from stable isotopic data from inorganic carbonates: Sedimentary Geology, v. 121, p. 191–206, doi: 10.1016/S0037-0738(98)00086-4. Valdeolmillos Rodríguez, A., 2005, Registro paleoclimático y paleoambiental de los últimos 350.000 años en el Parque Nacional de Las Tablas de Daimiel (Ciudad Real) [Ph.D. thesis]: Alcalá de Henares Madrid, Universidad de Alcalá, 308 p. Vasconcelos, C., and McKenzie, J.A., 1997, Microbial mediation of modern dolomite precipitation and diagenesis under anoxic conditions (Lagoa Vermelha, Rio de Janeiro, Brazil): Journal of Sedimentary Research, v. 67, p. 378–390. Volkmer-Ribeiro, V., and Motta, J.F.M., 1995, Esponjas formadoras de esponjilitos em lagoas no triângulo mineiro e adjacências, com indicaçao de preservaçao de habitat: Biociências, Porto Alegre, v. 3, p. 145–169. Wilkin, R.T., and Barnes, H.L., 1997, Formation processes of framboidal pyrite: Geochimica et Cosmochimica Acta, v. 61, p. 323–339, doi: 10.1016/ S0016-7037(96)00320-1. Wright, V.P., and Platt, N.H., 1995, Seasonal wetland carbonate sequences and dynamic catenas: A reappraisal: Sedimentary Geology, v. 99, p. 65–71, doi: 10.1016/0037-0738(95)00080-R. Wright, V.P., Alonso-Zarza, A.M., Sanz, M.E., and Calvo, J.P., 1997, Diagenesis of late Miocene micritic lacustrine carbonates, Madrid Basin, Spain: Sedimentary Geology, v. 114, p. 81–95, doi: 10.1016/S0037-0738(97)00059-6. Wüst, R.A.J., and Bustin, R.M., 2003, Opaline and Al-Si phytoliths from a tropical mire system of West Malaysia: Abundance, habit, elemental composition, preservation and significance: Chemical Geology, v. 200, p. 267–292, doi: 10.1016/S0009-2541(03)00196-7. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006 Printed in the USA
Geological Society of America Special Paper 416 2006
Depositional conditions of carbonate-dominated palustrine sedimentation around the K-T boundary (Faciès Rognacien, northeastern Pyrenean foreland, southwestern France) Daniel Marty† Section d’archéologie et paléontologie, Office de la culture OCC, Hôtel des Halles, P.O. Box 64, 2900 Porrentruy, Switzerland Christian A. Meyer Naturhistorisches Museum, Augustinergasse 2, 4001 Basel, Switzerland ABSTRACT The Faciès Rognacien is a sequence of highly bioturbated and pedogenically modified palustrine carbonates that were deposited under oxic conditions around the Cretaceous-Tertiary (K-T) boundary in the northeastern Pyrenean foreland basin (SW France). The sedimentary structures and early diagenetic features identified (mottling, nodule formation, brecciation, pseudomicrokarst, cracking, charophytes, Microcodium) suggest deposition in a palustrine environment between the subarid and intermediate climate type. Sedimentological and paleoecological analysis enables us to distinguish two facies associations, the lacustrine pond facies and the freshwater marsh facies associations. The majority of the carbonates are attributed to the freshwater marsh facies. The lacustrine pond facies occurs only in isolated paleolows, and is identified on the basis of its paleobiological content (charophytes, ostracodes). This suggests that the palustrine carbonates of the Faciès Rognacien were deposited in a seasonal wetland (carbonate-producing freshwater marsh), rather than in the marginal zone of a large, shallow lake. In this wetland paleoenvironment, all carbonates underwent widespread pedogenesis, and small, ephemeral ponds are of limited distribution, most likely recording deposition in paleolows. Keywords: palustrine carbonates, seasonal wetland, freshwater marsh, Rognacien, K-T boundary, SW France. RESUMEN La Faciès Rognacien es una secuencia formada por carbonatos palustres muy bioturbados y modificados pedogénicamente, que se depositó bajo condiciones óxicas en la cuenca de antepaís Pirenaica (SW de Francia). Su edad está entorno al límite K-T. Las estructuras sedimentarias y los rasgos diagenéticos tempranos (caráceas, Microcodium, moteado, formación de nódulos, brechificación, pseudomicrokarst, fisuración) indican que esta secuencia se depositó en un ambiente palustre de clima
E-mail:
[email protected].
†
Marty, D., and Meyer, C.A., 2006, Depositional conditions of carbonate-dominated palustrine sedimentation around the K-T boundary (Faciès Rognacien, northeastern Pyrenean foreland, southwestern France), in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 169–187, doi: 10.1130/2006.2416(11). For permission to copy, contact editing@ geosociety.org. ©2006 Geological Society of America. All rights reserved.
169
170
Marty and Meyer subárido a intermedio. Los análisis sedimentológicos y paleontológicos permiten distinguir dos asociaciones de facies: charcas lacustres y zonas pantanosas de agua dulce. Las facies de charcas aparecen aisladas en paleodepresiones y tienen un contenido paleobiológico característico (caráceas, ostrácodos). Esto sugiere que los carbonatos palustres de esta secuencia se depositaron en humedales estacionales más que en las zonas marginales de grandes lagos someros. En estos humedales todos los carbonatos sufrieron modificaciones pedogénicas importantes. Palabras clave: carbonatos palustres, humedales estacionales, pantanos de agua dulce, Rognaciense, límite K-T, SW de Francia.
INTRODUCTION Freytet (1971a, 1971b, 1973, and 1984) and Freytet and Plaziat (1982) provided good overviews of the continental deposits of southern France, including exhaustive descriptions of palustrine sedimentological features. More recently, other authors have described palustrine carbonates from the Cretaceous and Tertiary of France and Spain (e.g., Platt, 1989a; Cojan, 1999) or provided reviews on palustrine carbonates (e.g., Platt and Wright, 1991; Armenteros et al., 1997; AlonsoZarza 2003). Palustrine facies have only been recognized and studied during the last thirty years (Wright and Platt, 1995), and much less is known about them than about marine carbonates (Alonso-Zarza, 2003). Nevertheless, palustrine sequences are now known from the Carboniferous to the Neogene (Wright and Platt, 1995), mostly within continental successions and intercalated with floodplain deposits. It was Freytet (1964) who introduced the term palustrine (Latin: “paluster” = swampy, marshy), and Freytet (1984, p. 231) presented the following definition: “A palustrine limestone exhibits systematically characteristic features of the primary lacustrine deposit (organisms, sedimentological features), as well as features due to later transformations (organisms, root traces, desiccation, pedologic remobilisations).” The term palustrine is commonly applied as a nonmarine equivalent for peritidal. Freytet and Plaziat (1982) interpreted their palustrine facies as marginal lake deposits of shallow, unstratified freshwater lakes with swampy surroundings. Platt and Wright (1991, 1992) and Wright and Platt (1995) expanded on these ideas, noting that low shoreline gradients led to the extensive exposure of lake margins at times of low water level, and that many ancient palustrine successions show a dominance of pedogenically modified carbonates over those recording a primary lacustrine origin. Their analogy with modern environments of the Florida Everglades suggested that palustrine carbonates are the products of hardwater seasonal wetlands (sensu Tarnocai, 1979). Wetlands, however, are intermediaries between terrestrial and aquatic (lacustrine) ecosystems, and palustrine environments are often studied in combination with other depositional systems, such as lakes, deltas, or floodplains, rather than being treated as a distinct entity (Liutkus and Ashley, 2003). Hence, no sedimentological facies models have been developed as yet
for freshwater wetlands, as they have for other depositional environments (Liutkus and Ashley, 2003). Consequently, there is a limited understanding of their origin, how they are sustained hydrologically, and the type of sedimentary deposit that may be preserved in the geological record (Ashley et al., 2004). Only recently, following the model of Wright and Platt (1995), Flügel (2004, p. 13 and p. 742) provided a revised definition of the palustrine facies as pedogenically modified carbonates of nearshore deposits of extremely shallow lakes with oscillating water level and densely vegetated shorelines, as well as of carbonate swamps surrounding these lakes. This paper analyzes facies within the palustrine carbonates of the Faciès Rognacien (Cretaceous-Tertiary [K-T] boundary) of SW France, and reconstructs the paleoenvironment in detail. Within the research area, all carbonates of the Faciès Rognacien are pedogenically modified and can be classified as palustrine carbonates. However, carbonates clearly exhibiting evidence for a primary lacustrine origin are very scarce and occur only in isolated locations. These carbonates are described as the lacustrine pond facies association, as it is inferred that they were formed in small and ephemeral ponds. The majority of the carbonates of the Faciès Rognacien does not show any evidence of a primary lacustrine origin, and are thus attributed to the freshwater marsh facies association. This observed facies distribution pattern cannot be well explained with the “marginal lake” facies model of Freytet and Plaziat (1982), or the marginal lacustrine facies of Platt and Wright (1991), respectively. Moreover, this makes clear the assertion that the palustrine carbonates of the Faciès Rognacien precipitated in freshwater marshes within a carbonate wetland environment. This paper describes the lithofacies associations of both the freshwater marsh and the lacustrine pond facies, proposes microfacies and paleontological criteria for their recognition, and reconstructs the ancient wetland environment of the Faciès Rognacien in great detail. GEOGRAPHICAL AND GEOLOGICAL SETTING The study area is situated in the folded northern Pyrenean foreland (southwestern France, Département Aude), which is separated from the North Pyrenean zone to the south by the North Pyrenean frontal thrust and from the North Aquitaine folded foreland to the north by the sub-Pyrenean frontal thrust (Bousquet, 1997; Charrière and Durand-Delga, 2004) (Fig. 1).
Palustrine carbonates of SW France
171
Montagne Noire Fra n c e
N
Aude
nta
ure
l Lavelanet
Limoux
SPFT
Massif de Mouthoumet
Audois Espéraza NPFT
n
rranea
Nebias
43°10'N
Medite
Foix
Treziers
Corbiè re
Pla
s
Carcassonne
Perpignan Pyrenees
10 km
2°21'E
Hercynian basement + Trias
Mesozoic
Paleocene and Illerdian
Cuisian to Ludian (Série de Palassou)
Research Area
Faults Post Ludian
Figure 1. Geographical and geological setting of the research area. NPFT—North Pyrenean frontal thrust, SPFT—sub-Pyrenean frontal thrust (after Tambareau et al. [1995, 1997] and Bousquet [1997]).
A progressive and complex regression occurred across the Pyrenean area during the Cretaceous (Bilotte, 1978; Bilotte et al., 1983; Babinot et al., 1983). The orogenesis of the Pyrenees started between Late Cretaceous and Early Tertiary times (Bousquet, 1997). Erosion and continental sedimentation were widespread throughout much of the Pyrenean foreland from the Early Campanian until the Early Thanetian (Fig. 2). Within the research area, the basin was tectonically bounded to the north by the Montagne Noire Massif and to the south by the rising Pyrenean Range (Tambareau et al., 1995) (Fig. 1). The best overviews of the regional geology and paleontology are those by Freytet (1970, 1971a), Jaffrezo (1977), Plaziat (1981, 1984), Bilotte et al. (1989), Bousquet (1997), and Bousquet and Vianey-Liaud (2001). The text booklet of the geological map of Quillan (Bessière et al., 1989) provides a short lithological description of the Faciès Rognacien, but to date, only Peybernès and Combes (1999) have examined the Faciès Rognacien of the study area in greater detail.
STRATIGRAPHY Generally, the terms Rognacien and Vitrollien serve as local lithostratigraphic units in southern France (Provence) for the continental units of Late Cretaceous and Early Danian age, respectively (Babinot and Durand, 1980; Babinot et al., 1983). The “Rognacien” was introduced in Villot (1883) using Rognac near Aix-en-Provence (Provence) as the type locality (Babinot and Durand, 1980). In lithofacies terms, the Rognacien is characterized by an intimate association of lacustrine and palustrine marls and limestones with various types of “hypercalcimorph” soils (Freytet and Plaziat, 1982; Babinot et al., 1983). The lithostratigraphic units Rognacien and Vitrollien have also been used in a chronostratigraphic sense in Languedoc and the Pyrenees (Freytet, 1970; Plaziat, 1970; Bilotte, 1978), which are located far away, however, from the type locality. Thus, and in order to avoid confusion, we use the terms Faciès Rognacien and Faciès Vitrollien, respectively (following Bessière et al., 1989; Fig. 2).
172
Marty and Meyer
Thanetian
Thanétien supérieur Thanétien inférieur
Danian
Paleogene
Bessière et al. (1989)
Faciès Vitrollien
Maastrichtian
Marnes de la Maurine
C./M.
Grès des Estous
Campanian
Cretaceous
Faciès Rognacien Poudingue Fleuri
Marnes de Campagne
Grès d'Alet
Marine limestones Palustrine limestones Palustrine calcareous marls Fluviatile calcareous conglomerates Clays and marls (floodplain, paleosols) Sandy, calcareous marls (floodplain, paleosols) Fluviatile sandstones Figure 2. Generalized stratigraphic framework from the Campanian to the early Danian for the research area (not to scale; after Bessière et al., 1989). Based on charophyte biostratigraphy (Marty, 2001, 2004), the Faciès Rognacien is attributed to the Maastrichtian and early Danian.
Previous work has suggested that the K-T boundary has to be placed in the upper part of the Rognacien, and not in the Vitrollien (e.g., Bessière et al., 1980; Galbrun, 1989; Rocchia et al., 1989; Westphal and Durand, 1990; Galbrun et al., 1991; Cojan et al., 1998; Bousquet and Vianey-Liaud, 2001), although no unequivocal evidence has been provided as yet for a golden spike of the K-T boundary. Within the research area, the Faciès Rognacien is composed of a sequence of 15–25 m of palustrine carbonates without intercalated clastic material. The succession can generally be subdivided (from the base to the top) into four units: lower marl, lower limestone, upper marl, and upper limestone (Figs. 3 and 4). These correlate with the M1, C1a/b, M2, and C2 subdivisions of Peybernès and Combes (1999), who identified paleokarstic megafeatures and defined erosional and karstic paleosurfaces (discontinuities) at the base and the top of the limestone units. However, this study does not provide evidence for the existence of such paleokarstic megafeatures. The transition from the underlying Poudingue Fleuri to the lower marl unit of the Faciès Rognacien is not clearly marked. In this work, the limit has been defined where the clastic content falls to zero. The upper boundary is well defined, and the overlying Faciès Vitrollien rests directly upon the upper limestone unit (Fig. 3, section 2). The base of the Faciès Vitrollien consists of clays with only minor Microcodium compared to the rest of this unit, where Microcodium is abundant or even rock-forming. Besides abundant Microcodium, the Faciès Rognacien is almost barren, even within the marliest layers, many of which could be classified as hydromorphic paleosols of the freshwater marsh facies. This is consistent with the highly bioturbated and oxygenated nature, as well as the pedogenic overprinting, of palustrine carbonates, which are both factors that are unfavorable to fossil preservation. Nevertheless, paleontological and paleoecological data gained from screen-washed samples and the analysis of thin sections proved critical in resolving the complex history of many lithofacies, especially those of the lacustrine pond facies association, where gyrogonites and encrusted stems of charophytes, oncoids, and rare ostracodes are the only recognizable primary lacustrine features. The lower and upper marl units contain a Maastrichtian charophyte flora, whereas the upper limestone unit contains a Paleocene flora, attributing—at least within the research area—the entire Faciès Vitrollien to the Paleocene and indicating that no major depositional changes occurred at the K-T boundary (Marty, 2001, 2004) (Fig. 2). PETROLOGY AND SEDIMENTOLOGY OF THE FACIÈS ROGNACIEN Typical Palustrine Features Sedimentological Features At outcrop scale, the four units of the Faciès Rognacien are generally easy to distinguish. The lower and upper limestone units are chiefly gray in color and are highly indurated (Fig. 4). A high
Palustrine carbonates of SW France
173
Figure 3. Schematic lithological logs of the sections 2, 4, and 5. Out of 8 sections studied, these are the only sections where charophytes could be isolated out of screen-washed samples, allowing the determination and correlation of the Cretaceous-Tertiary (K-T) transition zone (Marty, 2001, 2004). Throughout the Faciès Rognacien, a pronounced lateral variation between and even within the sections in lithology, due to different degrees of macroscopic pedogenic modification (brecciation, nodule formation), is characteristic. Note, however, that this macroscopic lithological appearance of pedogenic modification is not related to the described lithofacies of the freshwater marsh and lacustrine pond facies.
degree of induration—despite limited burial or cementation—is a common feature of palustrine sediments (Alonso-Zarza, 2003) and is explained as the result of mineralogical stabilization and aggrading neomorphism (Wright et al., 1997; Anadón et al., 2000). The gray color indicates zones with reduced iron only, recording rather short subaerial exposure and at least seasonal hydromorphism (Platt and Wright, 1992). The contacts between the marl and limestone units are often very irregular, and doming-upward structures at the top of the limestone units may be observed. Other features of the limestone units include prismatic structures (“columnar limestone”), probably due to root bioturbation (Klappa, 1978a, 1978b; Esteban and Klappa, 1983; AlonsoZarza et al., 2000), brecciated limestone (Fig. 5A), and nodular limestone (Fig. 5B). The two marl units generally have a gray,
beige, or white color, and they appear structureless and homogeneous. However, mottling can locally be pronounced, especially at the base of the Faciès Rognacien. Nevertheless, lateral variation in lithology within the Faciès Rognacien is typically pronounced. Strongly brecciated limestone may laterally pass into nonbrecciated limestone, and the limestone units may contain intercalated marly layers or marly layers that cut through them (Fig. 3). Apart from Microcodium, the marl units contain very sparse microfauna, and marls bearing charophytes (gyrogonites and encrusted stems) and ostracodes are only rarely found. Throughout the Faciès Rognacien, red, yellow, and violet mottling is common, although this may be due to later transformation rather than prolonged exposure and pedogenic reddening (“rubefaction”).
174
Marty and Meyer
Figure 4. Section 4 (Ravin de Couleurs) showing the typical lithological sequence of the Faciès Rognacien within the study area. From base to top, the four units are identified as follows: lower marl, lower limestone (with intercalated marls), upper marl, and upper limestone. In this section, the upper limestone unit is more massive than the lower one.
The Faciès Rognacien displays a wide range of palustrine features (alveolar texture, brecciation, coated grains and gravels, cracking, crystallaria, mottling, nodule formation, pseudomicrokarst), which provide evidence of extensive subaerial exposure and pedogenesis. Some of these features may be observed at outcrop scale (Figs. 3 and 5). On a small microscale, several features are commonly intimately associated within the same thin section. Descriptions and interpretations of the most prominent palustrine features of the Faciès Rognacien are given in Table 1 and illustrated in Figures 5 and 6. Further, characteristic features of palustrine carbonates were described in Freytet (1971a, 1971b, 1984), Freytet and Plaziat (1982), Platt and Wright (1992), Alonso-Zarza (2003), and Freytet and Verrecchia (2002). Ginsburg (1975), Hardie (1977), Hardie and Shinn (1986), and Bain and Foos (1993), described the recognition of subaerial exposure surfaces from microfabrics
Figure 5. (A) Outcrop photograph showing strongly brecciated limestone, interpreted as monomict and autochthonous desiccation breccia produced by repeated wetting and drying (“palustrine brecciation”). Scale bar is 15 cm. (B) Base of the lower limestone unit of section 1 “L’Encantado”: nodular limestone. Hammer is shown for scale.
Microscopic: Intraclasts are irregularly shaped, more or less dark, (sub-) rounded, up to several mm in size, and exhibit sharp to diffuse boundaries. Fissures, cavities, and residual voids are generally filled with sparry calcite cement. Macroscopic: Made up of angular intraclasts with well-defined boundaries, each several mm up to several cm in size. Jigsaw fits between adjacent clasts common.
Microscopic: Coated grains include rounded to angular carbonate intraclasts. Coating appears either as micrite envelope or as dark micritic laminae consisting of iron-rich clay material. Micrite envelopes are smooth and regular; iron-rich laminae are generally irregular. Seldom, several generations of coatings may be observed.
Microscopic: Vertical, horizontal, planar horizontal, oblique, and curved cracks. Circumgranular cracking is prevalent and well developed. Cracks are commonly filled with sparry calcite cement. Macroscopic: Vertical cracks (vertical joint planes sensu Brewer, 1964), horizontal cracks (horizontal joint planes, sheet cracks), planar horizontal cracks, oblique cracks (skew planes), and curved cracks (craze planes). Microscopic: Consists of a mosaic fabric of microsparitic calcite crystals with a diameter of 5–10 microns. Often, crystallaria occur localized and patchy or in association with alveolar texture. Microscopic: Very scarce, never associated with other clastic material. Macroscopic: Frequent, 0.2 mm to several mm in diameter, (bi-) pyramidal. Frequently associated with gyrogonites and Microcodium.
Microscopic: Mottled areas form irregular, dark haloes of varying intensity with diffuse or rarely sharp boundaries. Locally outlined by circumgranular cracks, often associated with nodular limestone fabrics. Macroscopic: Red, yellow to beige and violet colored patches, each a few cm in diameter. Microscopic: Contains rounded intraclasts up to several mm in size. Intraclasts are similar to the surrounding matrix and are composed primarily of micrite. Macroscopic: Frequent at top or base of limestone units. Size of nodules varies between several cm up to 20–30 cm. If the intercalated marly matrix has already been eroded, the beds exhibit a “pseudo”-conglomeratic aspect. Microscopic: Irregular, vertically elongated cavities (“fenestrae”) commonly filled with complex, polyphased, fine-grained sediment (crystalline, vadose silt) reworked and rounded Microcodium debris, and other, larger rounded intraclasts. Some cavities exhibit a lobate shape suggesting that they may have resulted from dissolution or disintegration of Microcodium colonies. Residual voids are filled with phreatic sparite cement.
Coated grains and gravels Peryt (1983) Figure 6F
Cracking Figures 5A, 6A, 6B, 6C, and 7D
Crystallaria Figure 6E
Idiomorphic quartz
Mottling Figures 6F and 8C
Nodular limestone Freytet (1973) Figures 5B and 6C
Pseudomicrokarst Plaziat and Freytet (1978) Figures 6D and 7D
Enlarging of a complex network of root traces and horizontal cracks, with a polyphased filling of coarse and fine internal sediment and varied cements. During dry seasons, the vegetation on carbonate mud (reed or rooted aquatic plants like charophytes) may disappear, leaving root cavities behind. These cavities get enlarged through subsequent water circulation and are filled up before the water level rises again.
Mainly due to desiccation and the subsequent formation of planar and curved fissures. Might also be favored through brecciation by recrystallization. In modern soils, nodule formation of carbonate occurs in the zone of an oscillating water table due to repeated flooding (Ruellan, 1967).
Redistribution of iron and hydroxides in hydromorphic carbonate soils during conditions of fluctuating water table or Eh and Ph, if parental carbonate contains several percent of clay and if its total iron content exceeds 1.5%–2% (Bown and Kraus, 1987; Retallack, 1990). Results in cements which are slightly richer in clay and iron oxide (Valero Garcés et al., 1994).
Authigenic precipitation of quartz occurs during diagenesis if the sediment contains circulating salinar solutions and if pH conditions are slightly alkalic (Chilingarian and Wolf, 1988). Authigenic quartz is also described from oncoids and stromatolites (Winsborough et al., 1994).
Crystallaria result from complex and repetitive phases of recrystallization of nodules or mudstones during pedogenic modification, whereas the mosaic fabric clearly indicates crystal growth.
Commonly associated with roots or induced by desiccation. Circumgranular cracking (and nodule formation) is believed to be the result of repeated wetting (expansion) and drying (shrinkage). Cracking might reflect minor pedogenic modification of the original carbonate texture during lake regression. Subsequent phreatic cementation filled the cracks with sparry cement.
Grains and gravels might form through desiccation and mechanical reworking or fenestral fabrics and roots with associated fungi (Alonso-Zarza et al., 1992b) during an early diagenetic process called “grainification” (Mazullo and Birdwell, 1989; Wright, 1990). Micrite coating results from rolling during mechanical reworking, iron-coatings from remobilization of iron-rich clay material that was derived from overlying soils.
Angular intraclasts with jigsaw fit indicate in situ genesis without mechanical reworking. During periods of emersion (lowstands), carbonate mud may become partly lithified and wetting-and drying-related shrinkage creates planar voids (desiccation cracks). During flooding, fissures and cavities may be partially filled with fine carbonate material, and pseudomicrokarst structures may form. Typical of palustrine carbonates forming in a semiarid climate (Platt and Wright, 1992). Has also been called “autobrecciation,” resulting in “pseudoclast”-containing “pseudobreccias” (Armenteros et al., 1997).
TABLE 1. TYPICAL PALUSTRINE FEATURES IDENTIFIED WITHIN THE FACIÈS ROGNACIEN Appearance in the Faciès Rognacien Interpretation Microscopic: Very common. Occurs as microlaminar micrite or as a network of Steinen (1974) suggested that the fabric may result from the formation of discrete anastomosing, fine, dark micritic lines or filaments, including irregular, sometimes channelways within sediment that has been penetrated by rootlets, and Klappa cylindrical calcite spar-filled voids and vugs, resulting in a highly fenestral fabric. (1978b) and Esteban and Klappa (1983) described alveolar texture as the Voids are filled with sparry calcite, similar to crystallaria. The micritic network may product of coalesced mm-sized rhizoliths. Bain and Foos (1993) suggested that exhibit a strong biogenic fabric, which can easily be confused with microbial alveolar texture resulted from root penetration and diagenetic alteration. Flügel stromatolites or tufas (Wright et al., 1988). (2004), however, stated that dark cauliflower networks may also have originated from migration of colloidal Fe-solution in argillaceous micrite.
Brecciated limestone (desiccation breccias) Figures 5A and 6B
Feature Alveolar texture Esteban (1974) Figures 6D, 6E
Figure 6. Typical palustrine microfabric features of the Faciès Rognacien (for a more detailed description, see also Table 1): (A) Circumgranular cracking (curved planes) in a Microcodium-bearing wackestone. Cracks are filled with sparry calcite. (B) Partly brecciated limestone (“palustrine brecciation”). Residual patches of micritic material are intensely fissured, but fit and connection with the adjacent parts clearly indicate that they have not been transported. (C) Nodular limestone, here a mudstone of the freshwater marsh facies association. (D) Rhizolith exhibiting alveolar texture: dark micritic filaments outlining elongate vugs and tubiform pores filled with articulated Microcodium colonies (gray, roundish patches) and vadose silt. The pseudomicrokarst cavity (top left) shows a geopetal filling of crystalline (vadose) silt. (E) Alveolar texture: irregular and cylindrical fenestrae and tubiform pores outlined by micritic filaments. The cavities are filled with vadose silt and blocky calcite (crystallaria). (F) Micrite-coated rounded ferruginous intraclast. Intraclast and matrix is composed of Microcodium-bearing wackestone to packstone.
Palustrine carbonates of SW France in peritidal carbonates, which show many textural similarities with palustrine carbonates. Paleoecological Features Microcodium Complete, articulated colonies of Microcodium, as well as reworked, disarticulated colonies, are found in screen-washed samples and in thin sections. Microcodium always exhibits the same morphotype (type 1 sensu Bodergat, 1974 and Plaziat, 1984). In thin sections, Microcodium shows “corn-cob” colonies in longitudinal (Freytet and Plaziat, 1982; Plaziat, 1984) or spheroidal “rosette” structures in transverse sections (Freytet and Plaziat, 1982; Kosir, 2004) (Fig. 7A). In situ colonies are found in all sorts of cracks and cavities resulting from desiccation and root penetration. Here, they are commonly associated with dissolution and intense corrosion
177
of the adjacent carbonate substrate. However, articulated colonies are also commonly observed within mudstones lacking any other evidence of subaerial exposure. Microcodium structures appear to have been very fragile and readily subject to disintegration into single prisms (Fig. 7B). These prisms are a common feature in all reworked, bio- or pedoturbated lithofacies and are thus widespread in their distribution. If the prisms are slightly rounded or corroded, they become harder to identify as Microcodium (Figs. 7C and 7D). Also common is the association of reworked Microcodium prisms with articulated colonies, indicating repeated periods of Microcodium growth and reworking. The term Microcodium was introduced by Glück (1912). Microcodium is a problematic calcitic microfeature of many calcretes and paleosols, and there is considerable controversy surrounding its origin and possible relation with calcified plant roots (Esteban, 1974; Freytet and Plaziat, 1982; Freytet, 1984; Jaillard
Figure 7. Lithofacies of the freshwater marsh facies association. (A) Mudstone with Microcodium colonies (“rosettes”) in a micritic matrix. Transverse sections of Microcodium colonies show the typical “rosette” structure, associated with corrosion of the matrix at its margins. (B) Microcodium-bearing wackestone to packstone with partly disarticulated Microcodium colony. Reworked prisms exhibit marginal corrosion and rounding, suggesting multiple phases of growth and reworking. (C) Microcodium-bearing packstone containing single prisms and a disarticulated colony (lower right). (D) Microcodium-bearing wackestone showing nodule formation associated with circumgranular cracks, mottling, and pseudomicrokarst cavities (pmc), which are filled with crystalline (vadose) silt and blocky calcite.
178
Marty and Meyer
et al., 1991; Wright et al., 1995; Freytet et al., 1997; Kosir, 2004). However, despite its controversial origin, Microcodium is a clear indicator of terrestrial conditions, provides evidence for subaerial exposure, and may be used as a criterion for the recognition of paleosols (Klappa, 1978b; Alonso-Zarza et al., 1998). Recently, Kosir (2004) suggested that its formation takes place during early stages of soil development, probably reflecting nutrient-acquiring mechanisms used by certain types of specific types of vascular plants of a pioneer community that are able to rapidly colonize nutrient-poor carbonate substrates during relatively short-lived phases of subaerial exposure.
a lesser degree, ostracodes. The freshwater marsh facies, on the other hand, includes all pedogenically modified carbonates that do not contain readily identifiable primary lacustrine features. Both facies associations are characterized by the widespread presence of fabrics and textures recording emersion and pedogenesis (Table 1; Fig. 6). Several different lithofacies have been identified in each of the two facies associations, reflecting a complex spectrum of different degrees of emersion, pedogenesis, and associated reworking.
Charophytes Charophyte remains (gyrogonites, encrusted gyrogonites and stems) are common in the Faciès Rognacien and are present in both screen-washed samples and thin sections (Fig. 8). Charophyte remains occur in all four units of the Faciès Rognacien, but not in all sections studied. However, in sections where charophytes are present, they sometimes occur in several units (e.g., sections 2 and 5 in Fig. 3). The presence of cortical cells identifies fragments of charophyte stems within oncoid nuclei and other intraclasts (Figs. 8A and 8B). Charophyte stems may serve as nuclei for additional inorganic or organic carbonate precipitation. Inorganic encrustation of stems can be explained by photosynthetic removal of CO2 and HCO3–, leading to the precipitation of calcium carbonate. Some encrusted charophyte stems show laminated stromatolite-like structures formed by epiphytic cyanobacteria, which appear to develop after eutrophication of nutrient-poor environments (Martín-Closas, 1999). The diverse modes and degrees of epiphytic calcification may produce different forms of calcification and fragmentation of the charophyte stems, as illustrated in Freytet and Plaziat (1982) and Schneider et al. (1983). Modern charophytes live in shallow, littoral zones of temperate to warm, alkaline freshwater lakes, where they are the characteristic floral element. They occur in the photic zone, to a maximum depth of 15–20 m, although the presence of stems usually suggests depths of less than 10 m (Cohen and Thouin, 1987; Garcia, 1994). Charophytes commonly occur together with ostracodes in low-diversity assemblages (Dean and Fouch, 1983). Encrusted stems are delicate structures, and cannot be transported far. Preservation is therefore only possible in a low-energy environment without significant currents or water turbulence (Dean and Fouch, 1983).
Calcareous Marls Beige to yellow calcareous marls, which have thicknesses up to 3 m and exhibit mottling on a macroscopic scale, make up most of the lower and upper marl unit. Screen-washed samples yielded articulated Microcodium colonies and ferruginous nodules, but never remains of a lacustrine fauna or flora. Interpretation. The presence of abundant, generally articulated Microcodium colonies suggests that these marls were formed in the freshwater marsh facies. This is supported by the complete absence of charophytes and other lacustrine remains.
FACIES AND FACIES ASSOCIATIONS The entire Faciès Rognacien is essentially made up of two types of carbonate facies associations. These are the (1) freshwater marsh facies, and (2) lacustrine pond facies associations. As all carbonates are pedogenically modified, the lacustrine pond facies is primarily identified on the basis of lacustrine paleoecological evidence such as the occurrence of charophytes and, to
Facies Association 1: Freshwater Marsh Facies Association
Mudstones The lower and upper limestone units are primarily composed of fairly massive layers of gray to beige homogeneous micritic limestones that lack any fossil allochems, siliciclastic material such as quartz, lamination, or bedding structure. Typically, the limestones are brecciated (Figs. 5A and 6B) or exhibit nodular fabrics accentuated by circumgranular cracking (Fig. 6C). Rarely, they contain crystallaria and colonies (Fig. 7A) or reworked prisms of Microcodium. Interpretation. The mudstone texture suggests low-energy sedimentation. The general absence of any bedding structure implies bioturbation of the sediments, and the absence of siliciclastic material suggests a closed environment. Palustrine brecciation, nodular fabrics, and circumgranular cracking indicate pedogenic modification. The degree of pedogenic modification may be estimated, where Microcodium is present, by the ratio of Microcodium colonies to isolated prisms. Within the Faciès Rognacien, this is the most abundant and typical lithofacies. Microcodium-Bearing Wackestone and Packstone These massive gray limestones are notable in thin section as Microcodium wackestones to packstones with strong evidence of pedogenic modification (brecciation, circumgranular cracking, nodular fabric). Microcodium is present within and supported by the micritic matrix and is mostly disarticulated (Figs. 7B, 7C, and 7D). Detrital quartz is absent. Interpretation. The occurrence of different features again suggests a complex history. The carbonate mud was subjected to emersion and pedogenic overprinting, apparently during periods of low water table. Microcodium colonies could then have
Figure 8. Lithofacies of the lacustrine pond facies association. (A) Intraclastic packstone (granular limestone). Subrounded to angular intraclasts include encrusted charophyte stems (cs), a few poorly developed oncoids, Microcodium prisms, and colonies. Some grains exhibit ferruginous coatings. (B) Detail of A, showing encrusted charophyte stem (cs), gyrogonite (cg), and laminar Microcodium colonies in longitudinal section (m). (C) Intraclastic nodule wackestone or packstone with ferruginous intraclasts, alveolar texture (at), Microcodium colonies (m), and spar-filled vugs. Individual intraclasts are distinguished from the matrix by abundance and state of preservation of Microcodium debris and aggregates, and by delineation of ferruginous coatings. (D) Detail of C. The matrix (wackestone) between the intraclasts contains less Microcodium-debris and charophyte stems (cs), but clearly more articulated Microcodium colonies (m) than the ferruginous intraclasts. Note dissolution associated with Microcodium colonies within the matrix and inhomogeneous mottling. (E) Intraclastic rudstone: intraclasts comprise oncoids (alternating layers of thin and thicker micritic laminae around a poorly defined nucleus), encrusted charophyte stems (cs), and other carbonate gravels. The matrix contains abundant reworked and corroded Microcodium debris (white clasts). Note the pronounced mottling and ferruginous coatings. (F) Intraclastic floatstone: angular intraclasts include encrusted charophyte stems (cs) and other carbonate gravels. The matrix is a Microcodium-bearing (white clasts) wackestone to packstone.
180
Marty and Meyer
formed in desiccation cracks and around rootlets penetrating into the cracks. Through subsequent water circulation following heavy rains, or in times of raised water level, Microcodium colonies were disintegrated and reworked, leading to the redeposition of single prisms. The presence of abundant Microcodium “micro-bioclasts,” in the absence of other detrital grains of similar size suggests the reworking of Microcodium which were proliferating in a freshwater marsh environment. Such environments would have been protected from terrigenous influx by a filter of dense reed or other swamp vegetation (Freytet and Plaziat, 1982). Facies Association 2: Lacustrine Pond Facies Association Chalky Marls These marls are white, chalky, calcareous marls, which are mostly intercalated with nodular limestone (see below) (Fig. 5B), whereas the bed thickness varies laterally between several centimeters up to several tens of centimeters. Screen-washed samples yielded abundant gyrogonites and encrusted stems of charophytes, partially articulated to articulated Microcodium colonies, and authigenic idiomorphic quartz. Interpretation. The presence of charophyte remains implies a primary lacustrine origin. Microcodium indicates later pedogenic overprinting during lowstand of the water table in a vadose environment. Furthermore, articulated Microcodium colonies suggest minor reworking. The white color, the chalky nature, and the absence of mottling suggest at least seasonal hydromorphism. Intraclastic Packstones (Granular Limestones) This facies is composed of nodular limestones that contain carbonate nodules up to a diameter of ~20 cm, often intercalated with white calcareous marls (see above) (Fig. 5B), or massive gray limestones. Polished sections show that the carbonate nodules consist of angular (0.5–1 cm) intraclasts as well as of (sub-) rounded (0.2–0.5 mm) calcareous “gravels” (Figs. 8A and 8B). Many of the clasts are made up of encrusted gyrogonites and charophyte stems. The charophyte gyrogonites and stems exhibit diverse degrees of fragmentation, but are generally well preserved and not deformed. The clasts are locally outlined with circumgranular cracks or ferruginous coatings. The matrix is micritic, although large spar-filled vugs and recrystallized areas are common. Ostracodes and molluscan shell fragments are sometimes present. Interpretation. Charophytes and other macrophytes were apparently encrusted due to preferential Ca-precipitation onto a biological substrate as a result of CO2 drawdown through photosynthetic activity. Subsequent degradation of the encrusted macrophytes provided a range of intraclastic material. The low degree of deformation and fragmentation of charophyte gyrogonites and stems indicate that compaction and crushing was negligible or that it predated encrustation. This also points to a low-energy environment, probably within charophyte mead-
ows, which are commonly developed within carbonate lakes at shallow depths of less than 10 m (Murphy and Wilkinson, 1980; Cohen and Thouin, 1987; Garcia, 1994). During a phase of low water table, the “primary” lacustrine sediment was subjected to emersion, resulting in minor pedogenic modification (limited desiccation and root brecciation). A rather short time of subaerial exposure is also indicated by the general lack of Microcodium. Intraclastic (Nodule) Wackestone and Packstone This facies is composed of breccias containing (sub-) rounded ferruginous intraclasts with diameters of several millimeters to centimeters (Fig. 8C). The intraclasts are made up of Microcodium wackestones to packstones, including reworked Microcodium debris and rarely articulated colonies floating in a dark ferruginous, argillaceous micritic matrix (Fig. 8D). The intraclasts also contain rare charophyte gyrogonites and stems. The matrix between the intraclasts is micritic and contains Microcodium debris as well as abundant articulated colonies (Fig. 8D), alveolar texture, and up to 1 cm big vugs, filled with blocky calcite cement. Interpretation. This complex lithofacies can only be explained by several subsequent events of emersion, involving pervasive microkarstic and desiccation brecciation of the above described chalky marls and intraclastic packstones, followed by reworking back into the “lacustrine pond” setting. Thus, different scenarios may lead to this lithofacies. One possible means of formation might be that Microcodium formed around ponds during periods of low water table. A rise in water level reworked the colonies, and single prisms settled down, with charophytes growing during the subsequent period of high water level. During a second phase of emersion, the charophyte- and Microcodium-bearing sediment was partially indurated, and intraclasts formed through palustrine brecciation or nodule formation. At the same time, new Microcodium colonies were established. During the subsequent period of high water table or as a result of heavy rainfall, these intraclasts were reworked, slightly rounded, and redeposited within another Microcodium-bearing mud. During a further period of emersion, another generation of Microcodium colonies and alveolar texture formed to give this lithofacies its final, complex appearance. The different size and form of intraclasts indicates reworking during short periods of high-energy events and suggests that they have not been transported far. Intraclastic Floatstone and Rudstone This lithofacies occurs seldom and only at the base of the Faciès Rognacien. It is composed of up to 30 cm of sandy, marly limestones that exhibit a very pronounced lateral change in the content and size of intraclasts. Thin sections reveal that intraclasts include encrusted charophyte stems and gyrogonites, structureless carbonate intraclasts, and, locally, oncoids (Figs. 8E and 8F). The size of encrusted charophyte stems is up to 1–2 cm in length, while the gyrogonite-bearing intraclasts
Palustrine carbonates of SW France are mostly only a few millimeters across. The carbonate intraclasts are generally on the order of a few millimeters in size, although some may reach up to one centimeter in diameter. Oncoids have a spherical to elongate form and a diameter of several millimeters up to 10 cm. They constitute alternating fine-grained more or less dark micritic layers with wavy and cauliflower-like fabrics. Nuclei are poorly defined, but where present, they comprise small lithoclasts, or, more rarely, charophyte stems (Fig. 8E). These rocks have a matrix to grain-supported fabric with a wackestone-packstone matrix containing reworked and strongly corroded Microcodium prisms. Interpretation. As with the intraclastic packstones, intraclastic floatstones and rudstones are the result of brecciation due to emersion, reworking and resedimentation of lithified carbonate grains and gravels. They are a common facies in the littoral realm of modern lakes (Murphy and Wilkinson, 1980; Platt and Wright, 1991). Oncoids commonly grow in alkaline, Ca-rich waters in river channels, marshes, lakes, and floodplains (Monty, 1981). Recent freshwater oncoids generally grow in rather quiet shallow-water environments (benches or flats of lakes) with temporary turbulence during floods (Monty, 1972). Pronounced changes in clast size and the lateral strong variation from matrix- to grain-supported fabrics suggest transport over short distances only. This indicates deposition in a pond that was repeatedly subjected to emersion, resulting in the formation of abundant Microcodium colonies and carbonate intraclasts, which were in turn reworked during the next rise of the water level or by water movement (waves) due to storms. DISCUSSION The palustrine environments of the Faciès Rognacien developed within a continental succession in a tectonically bounded foreland basin. The palustrine carbonates are intercalated with floodplain deposits within a fluvial-lacustrine system. This palustrine environment is likely to have passed into a more lacustrine environment eastward along the river Aude (Peybernès and Combes, 1999), fluvial-alluvial environments to the south, and marine environments progressively toward the west. AlonsoZarza (2003) suggested that palustrine deposits mostly form during periods of strongly reduced subsidence with the limited accommodation space of overfilled basins, leading to the deposition of palustrine facies as highstand depositional systems. Also Platt and Wright (1992) stated that palustrine deposits are common in relatively stable basins, typically forming during periods of tectonic quiescence when clastic supply from inflowing alluvial-fluvial systems is reduced. Platt and Pujalte (1994) and Platt (1995) noted that the formation of palustrine carbonates in the Cretaceous of Spain was associated with the subaerial exposure and peneplanation of an underlying carapace or pediment of marine Jurassic carbonates where clastic supply was limited. A similar subcrop configuration might also be suggested for the Faciès Rognacien.
181
Paleoclimate Climate is a critical control factor in the development of lacustrine and palustrine successions (Platt, 1989b, 1989c; Platt and Wright, 1992; Camoin et al., 1997). Climate not only controls the lacustrine and palustrine environment, but also the surrounding, usually siliciclastic, depositional environments. The development of palustrine carbonates is favored in climates with seasonal aridity and environments of low clastic detrital supply or carbonate-dominated source terrains (Alonso-Zarza et al., 1992a; Alonso-Zarza, 2003). Nevertheless, Schullenberger et al. (2004) stated that the presence of a groundwaterfed regional water table is more important than climate in the formation of extensive palustrine deposits. Further, Dunagan and Turner (2004) noted that primary groundwater discharge may give the appearance of increased humidity in an otherwise semiarid climate. However, according to Platt and Wright (1992), palustrine sequences may form under three different types of climate: semiarid, intermediate, and subhumid. These three climate regimes are tied to specific palustrine features, documented from Carboniferous to Quaternary palustrine sequences (Platt and Wright, 1992; Dunagan and Driese, 1999). Platt and Wright (1992) further developed a freshwater exposure index, similar to the marine exposure index of Ginsburg et al. (1977); it links characteristic palustrine features to both hydroperiod and seasonality. The hydroperiod is defined as the mean number of days per year during which the ground surface at a given site is covered with water (Ginsburg et al., 1977; Platt and Wright, 1992). Throughout the Faciès Rognacien, neither evaporites and calcretes (typical for a semiarid setting) nor blackened pebbles, coal, and lignite horizons (typical for a subhumid setting) have been found. However, Microcodium (typical for an intermediate setting) is abundant, and evidence of desiccation (brecciation and nodule formation, pseudomicrokarst) is relatively common. Thus, the Faciès Rognacien may be placed between the subarid and intermediate types of Platt and Wright (1992). Assuming that the exposure index model of Platt and Wright (1992) is applicable to the palustrine carbonates of the Faciès Rognacien, these carbonates may have an estimated hydroperiod somewhere in between 100–320 d (Fig. 9), indicating that pond development and lake expansion was probably associated with a distinct wet season. Those sections in the Faciès Rognacien displaying stronger evidence of subaerial exposure may record deposition on paleotopographic highs or distal areas (“prairies”), where flooding was rare and pedogenic modification was prolonged. The fact that all carbonates of the lacustrine pond facies are pedogenically modified, at least to some extent, also suggests that the ponds dried out during the dry season. The presence of Fe-coatings, Fe-concretions, and mottling points to a mean annual temperature over 20 °C (Pédro, 1968). Clay mineralogy analyses also support deposition in a warm and seasonally humid climate (Groebke, 2001). In summary, an intermediate, seasonally humid, subtropic climate might be suggested.
182
Marty and Meyer
Facies
Palustrine
Lacustrine
Prairie
Freshwater marsh
Marginal lake
Ephemeral wetland
Perennial wetland
Fringing wetland + littoral
Environment
Faciès Rognacien
Permanent Lake
Topography
HWT LWT
5m
Hydroperiod
30-90 days
100-250 days
100-320 days
100-320 days
All year
Exposure Ind. 70-90%
30-70%
10-70%
10-70%
0%
Typical features
Freshwater marsh: pseudo-microkarst, pronounced pedogenic modification, articulated Microcodium colonies Ponds: little pedogenic modification, isolated occurrence, charophyte stems + gyrogonites, ostracodes, Microcodium debris
Calcretes, black pebbles, root traces
Facies associations in the Faciès Rognacien
Lacustrine pond facies association
Freshwater marsh facies association
(Carbonate freshwater marshes) (Ephemeral, shallow, small ponds)
Erosional breccias, Lamination, no pedogenic oncoids, charomodification, charophyte phytes, ostracodes, gyrogonites, fish pedogenic features HWT
Highest water table
LWT
Lowest water table
Wooded vegetation Lithofacies of the Faciès Rognacien
Calcareous marls Mudstones Microcodium-bearing wacke/ packstones
Chalky marls Intraclastic pack/grainstones Intraclastic wacke/packstones Intraclastic float/rudstones
Swamp vegetation (horsetail, reed) Marsh vegetation (sedge, grass) Water plant vegetation (charophytes)
Figure 9. Simplified block diagram showing a facies model for a palustrine-lacustrine setting. The freshwater marsh and the lacustrine pond facies associations of the Faciès Rognacien are inferred to have been deposited in a freshwater marsh within a wetland environment, where densely vegetated carbonate marshes and swamps prevented siliciclastic input. Extensive charophyte meadows developed in small, shallow, and ephemeral ponds. Open lacustrine facies are not developed, and lateral changes in paleotopography are on the order of a few meters only. The table links paleoenvironment, typical sedimentological and paleoecological features, as well as the described facies associations with the freshwater exposure index of Platt and Wright (1992). The hydroperiod represents the number of inundation days over the year. An exposure index of 100% is equivalent to a hydroperiod of zero days per year.
Paleoenvironment The carbonates of the Faciès Rognacien display indicators both of emergence and of water-saturation (hydromorphism), but clearly do not show any evidence of marine influence. Further, they do not include any deeper-water lacustrine facies (such as laminites), or distal alluvial sediments. Macro- and microfacies analysis allowed the grouping of the various lithofacies into two distinct facies associations, namely the ubiquitous freshwater marsh facies and the relatively much rarer lacustrine pond facies. Carbonate sedimentation of the lacustrine pond facies took place in shallow and ephemeral ponds, and all lithofacies show some evidence of subaerial exposure and pedogenic modification. These ephemeral ponds may have been connected by
groundwater only, a possibility which is also consistent with the absence of any significant clastic input (Alonso-Zarza and Calvo, 2000). The low-diversity invertebrate fauna assemblage and the absence of any vertebrate remains (e.g., fish) suggest elevated environmental stress (alkalinity, oligotrophy, water-level instability) or poor faunal preservation potential (diagenetic conditions preventing preservation); this is consistent with the development of individually isolated, small, very shallow, and probably shortlived ponds at paleolows. The general absence of shell lags, the rarity of oncoid-bearing facies, and the predominance of mudstone to wackestone textures indicate that wave energies were generally low due to the very small size of the ponds, but, in part, perhaps also because of the baffling effect of extensive stands of aquatic vegetation like charophyte meadows. Nevertheless, high-energy
Palustrine carbonates of SW France storm events (sheetfloods) occurring sporadically in this generally low-energy system might have been responsible for the formation of intraclastic packstones in the lacustrine pond facies and Microcodium wackestones to packstones in the freshwater marsh facies, respectively. The lack of evaporites points to low salinity. This is supported by the presence of charophytes, which are typical for environments with low salinities of less than 16‰–20‰ (Schudack, 1993; Schudack et al., 1998), even if, for example, Burne et al. (1980) and Mojon (1990) give examples of more saline charophytes. The source of the fine-grained carbonate (typically low-Mg calcite) in palustrine carbonates is poorly understood, but may be polygenetic in origin, reflecting biogenic production from charophytes, ostracodes, molluscs, and cyanobacteria, as well as inorganic and biogenically induced precipitation (Dean, 1981; Platt and Wright, 1992; Alonso-Zarza and Calvo, 2000; Anadón et al., 2000). The carbonates of the freshwater marsh facies were probably produced biogenically from microbial (blue-green algal) mats (Monty, 1972; Monty and Hardie, 1976), from inorganic or biogenically induced precipitation around charophytes and other vegetation, as well as Microcodium. The presence of a fringing vegetation zone of shallow-water marsh and land plants would have acted as an effective barrier to terrigenous clastic input. Tandon and Andrews (2001) suggested that semiarid carbonate flats typically have a low biomass, and under prolonged exposure, the dominantly herbaceous vegetation would leave little or no evidence of the larger root systems as expected from arboreal plants. This could explain why well-developed rhizolites, as well as organic matter, have not been observed in the Faciès Rognacien. The predominance of the freshwater marsh facies and the scarcity of the lacustrine pond facies cannot easily be explained with the “marginal lake facies” model of Freytet and Plaziat (1982) or the marginal lacustrine facies of Platt and Wright (1991). It is suggested that the Faciès Rognacien in the study area was deposited in an ancient wetland dominated by carbonate-producing freshwater marshes with some intervening small and shallow ephemeral ponds. Seasonal variations in water table and minor topographic variations across the depositional area could easily explain the distribution and sequence of the various facies. Nevertheless, the depositional environment of the Faciès Rognacien as described here might have been located within a wider lacustrine environment, since Peybernès and Combes (1999) noted that the Faciès Rognacien becomes more lacustrine eastward along the river Aude. Further, this wetland-lacustrine environment is likely to have passed into fluvial-alluvial environments to the south and marine environments progressively toward the west (Plaziat, 1981). Modern Analogues Platt and Wright (1992) stated that difficulties in identifying a convincing modern analogue for palustrine carbonate deposition had hindered the development of facies models for
183
palustrine carbonates. These authors went on to propose the Florida Everglades wetland as an analogue for many aspects of palustrine carbonate deposition. The Everglades, however, are not a lake, nor a lake margin, but a vast, densely vegetated, carbonate freshwater marsh complex, where lakes and ponds make up only a small proportion of the total area (Mitsch and Gosselink, 1993). A fall in water level of only a few meters can cause exposure of wide areas, whereas comparable rises in lake level are unlikely to permit development of a stratified water column (Platt and Wright, 1992). Today, wetlands are amongst the most important and sensitive ecosystems and cover 6% of the world’s surface. They are areas that are periodically flooded, and they are found in every climatic zone and on every continent except Antarctica (Mitsch and Gosselink, 1993). As modern wetlands are very diverse, their definition and classification is extremely problematical (Finlayson and van der Valk, 1995a, 1995b; Scott and Jones, 1995). Among the most widely accepted definitions for a palustrine wetland is the one of Cowardin et al. (1979) (see also Cowardin and Golet, 1995, for recent advances), which was adapted by the U.S. Fish and Wildlife Service. This definition requires “an area of less than 8 ha, a lack of wave-formed or bedrock shoreline features, water depth in the deepest part of the basin of less than 2 m at low water, and salinity stemming from ocean-derived salts of less than 0.5 ppt” (Cowardin et al., 1979, p. 10). However, despite their abundance, many wetlands are limited in areal extent and form small features relative to their sedimentary basins (Quade et al., 1995). As such, their preservational potential may be limited. Indeed, Deocampo (2002) suggested that wetlands are likely to be difficult to recognize in the sedimentary record, and Liutkus and Ashley (2003) noticed that, as yet, no sedimentological facies models have been developed for (siliciclastic) freshwater wetlands. To date, the Everglades wetland has been considered by Armenteros and Daley (1998) as a modern analogue for the palustrine Bembridge Limestone (Eocene, Isle of Wight), and by Valero Garcés et al. (1994) for parts of the Upper Freeport Formation (Pennsylvanian, Appalachian Basin). Platt and Pujalte (1994) proposed that an ancient analogue for the Early Cretaceous palustrine system of northern Spain might be represented by the extensive areas of shallow freshwater marshes found in southeastern Iraq around Basra. These environments pass laterally seaward into the marginal marine and peritidal facies of the Persian Gulf (Baltzer and Purser, 1990) and laterally landward into the fluvial and semidesert environments of central Iraq. Valero Garcés et al. (1994) also proposed that the semiarid to subarid carbonate-dominated, extensive wetland of Bahia in northeastern Brazil (Branner, 1910) might form a good recent analogue. Another modern analogue might be provided by Lake Balaton in Hungary (Müller and Wagner, 1978). Recently, Dunagan and Turner (2004) reinterpreted lacustrine sediments of the Late Jurassic Morrison Formation as deposits of groundwater-fed, perennial carbonate wetlands, similar
184
Marty and Meyer
to the Quaternary wetland deposits of the U.S. southern Great Basin (Quade et al., 1995, 2003). All of the modern analogues listed display a complete spectrum of lake, pond, and soil environments similar to those recorded within the Faciès Rognacien. It is thus suggested that the Faciès Rognacien was deposited in a seasonal, groundwaterfed, nontidal, carbonate-producing, palustrine (sensu Cowardin et al., 1979) wetland, which was probably characterized by hydrologically closed, ephemeral ponds surrounded by vast, more or less densely vegetated (cf. Hofmann and Zetter, 2005) carbonate freshwater marshes and swamps (Fig. 9). This environment could also have formed within the marginal area of a larger perennial lake complex, or within an extensive fluvial-alluvial plain. The development of these palustrine environments within a lacustrine-fluvial system across a tectonically active foreland basin margin to the north of the Pyrenean foredeep is closely equivalent to the modern setting of the present-day Iraq analogue. CONCLUDING REMARKS 1.
2.
3.
4.
The Faciès Rognacien is a very good example for palustrine facies. It exhibits a wide range of classical palustrine features including well-developed palustrine brecciation, nodule formation, horizontal, planar, and circumgranular cracking, pseudomicrokarst, alveolar texture, and Microcodium. The abundance and distribution of these features place the Faciès Rognacien somewhere between the subarid and the intermediate palustrine climate type of Platt and Wright (1992), an intermediate subtropical climate with an estimated hydroperiod between 100 and 320 d. The palustrine carbonates of the Faciès Rognacien show a range of highly varied fabrics and lithofacies that reflect primary depositional setting and subsequent subaerial exposure, brecciation, pedoturbation, and pedogenesis. The various lithofacies recognized have been grouped into two facies associations, the lacustrine pond facies and the freshwater marsh facies, which allow a reconstruction of the paleoenvironment in greater detail. Fossils are extremely scarce throughout the whole Faciès Rognacien, particularly in the freshwater marsh facies, where only Microcodium is frequently found. Nevertheless, paleontological and paleoecological data gained from screen-washed samples and the analysis of thin sections proved critical in resolving the complex history of many lithofacies, especially those of the lacustrine pond facies, where gyrogonites and encrusted stems of charophytes, oncoids, and seldom ostracodes are the only recognizable primary lacustrine features. Within the research area, the pedogenically modified mudstones of the freshwater marsh facies association are by far the most prevalent. Hence, the Faciès Rognacien is predominantly composed of the freshwater marsh facies, where most of the carbonates formed.
5.
6.
Only a minor part of the depositional area—probably representing paleolows—can be attributed to the lacustrine pond facies. Thus, the Faciès Rognacien may be better characterized as an assemblage of freshwater marsh facies laid down within a carbonate wetland setting (Wright and Platt 1995) rather than pedogenically modified carbonates of a marginal lake setting. The facies associations observed are consistent with deposition within a seasonal, possibly groundwater-fed, palustrine (sensu Cowardin et al., 1979) wetland system characterized by possibly hydrologically closed, ephemeral ponds surrounded by vast, more or less densely vegetated areas of carbonate-producing freshwater marshes. This wetland environment is likely to have passed into fluvial-alluvial environments to the south and marine environments progressively toward the west. Further detailed sedimentological studies and facies analysis of palustrine sequences are essential in any attempts to develop more precise facies models and to compare ancient successions with their still poorly studied modern analogue environments, which occur in a range of carbonate-producing wetlands distributed worldwide.
ACKNOWLEDGMENTS This paper presents data submitted as part of a master’s thesis of the first author at the Department of Geosciences of the University of Basel. J. and C. Le Loeuff, L. Cavin, and D. Viand (Musée des Dinosaures, Espéraza) are warmly thanked for accommodation and support during the field studies. The first author thanks A. Wetzel (University of Basel) and L. Hottinger (Natural History Museum, Basel) for supervision. Thin sections were produced by W. Tschudin (University of Basel). We thank M. Schudack (Freie Universität Berlin) for comments on an earlier draft of the manuscript, and the reviewers L. Cabrera and N.H. Platt for numerous suggestions and comments that greatly improved the manuscript. Finally, we thank A.M. Alonso-Zarza for helpful comments for the interpretation of the paleoenvironment and A.M. Alonso-Zarza and L.H. Tanner for editorial work and for inviting us to contribute to this volume. REFERENCES CITED Alonso-Zarza, A.M., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: Earth-Science Reviews, v. 60, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Alonso-Zarza, A.M., and Calvo, J.P., 2000, Palustrine sedimentation in an episodically subsiding basin: The Miocene of the northern Teruel Graben, Spain: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 160, p. 1– 21, doi: 10.1016/S0031-0182(00)00041-9. Alonso-Zarza, A.M., Calvo, J.P., and García Del Cura, M.A., 1992a, Palustrine sedimentation and associated features—grainification and pseudomicrokarst—in the middle Miocene (intermediate unit) of the Madrid Basin, Spain: Sedimentary Geology, v. 76, p. 43–61, doi: 10.1016/00370738(92)90138-H. Alonso-Zarza, A.M., Wright, V.P., Calvo, J.P., and García del Cura, M.A.,
Palustrine carbonates of SW France 1992b, Soil-landscape and climatic relationships in the middle Miocene of the Madrid Basin: Sedimentology, v. 39, p. 17–35. Alonso-Zarza, A.M., Sanz, M.E., Calvo, J.P., and Estévez, P., 1998, Calcified root cells in Miocene pedogenic carbonates of the Madrid Basin: Evidence for the origin of Microcodium b: Sedimentary Geology, v. 116, p. 81–97, doi: 10.1016/S0037-0738(97)00077-8. Alonso-Zarza, A.M., Calvo, J.P., Van Dam, J., and Alcalá, L., 2000, Northern Teruel Graben (Neogene), northeastern Spain, in Gierlowski-Kordesch, E.H., and Kelts, K.R., eds., Lake Basins through Space and Time: Tulsa, Oklahoma, American Association of Petroleum Geologists, Studies in Geology, v. 46, p. 491–496. Anadón, P., Utrilla, R., and Vázquez, A., 2000, Use of charophyte carbonates as proxy indicators of subtle hydrological and chemical changes in marl lakes: Example from the Miocene Bicorb Basin, eastern Spain: Sedimentary Geology, v. 133, p. 325–347, doi: 10.1016/S0037-0738(00)00047-6. Armenteros, I., and Daley, B., 1998, Pedogenic modification and structure evolution in palustrine facies as exemplified by the Bembridge Limestone (Late Eocene) of Isle of Wight, southern England: Sedimentary Geology, v. 119, p. 275–295, doi: 10.1016/S0037-0738(98)00067-0. Armenteros, I., Daley, B., and García, E., 1997, Lacustrine and palustrine facies in the Bembridge Limestone (late Eocene, Hampshire Basin) of the Isle of Wight, southern England: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 128, p. 111–132, doi: 10.1016/S0031-0182(96)00108-3. Ashley, G.M., Maitima Mworia, J., Muasya, A.M., Owen, R.B., Driese, S.G., Hover, V.C., Renaut, R.W., Goman, M.F., Mathai, S., and Blatt, S.H., 2004, Sedimentation and recent history of a freshwater wetland in a semiarid environment: Loboi Swamp, Kenya: East Africa: Sedimentology, v. 51, p. 1301–1321, doi: 10.1111/j.1365-3091.2004.00671.x. Babinot, J.-F., and Durand, J.P., 1980, Les étages Français et leurs stratotypes: Valdonnien, Fuvélien, Bégudien, Rognacien, Vitrollien, in Cavelier, C., and Roger, J., eds., Les Étages Français et leurs Stratotypes: Mémoire du Bureau des Recherches Géologiques et Minières 109, p. 171–192. Babinot, J.-F., Freytet, P., Amiot, M., Bilotte, M., de Colombo, F., Durand, J.-P., Floquet, M., Gayet, M., Lange-Badré, B., Masriera, A., Massieux, M., Medus, J., Tambareau, Y., Ullastre, J., and Vilatte, J., 1983, Le Sénonien supérieur continental de la France méridionale et de l’Espagne septentrionale: État des connaissances biostratigraphiques: Géologie Méditerranéenne, v. 10, no. 3–4, p. 245–268. Bain, R.J., and Foos, M., 1993, Carbonate microfabrics related to subaerial exposure and paleosol formation, in Rezak, R., and Lavoie, D.L, eds., Carbonate Microfabrics: Frontiers in Sedimentary Geology: Berlin and Heidelberg, Springer-Verlag, p. 17–27. Baltzer, F., and Purser, B.H., 1990, Modern alluvial fan and deltaic sedimentation in a foreland tectonic setting: The Lower Mesopotamian Plain and the Arabian Gulf: Sedimentary Geology, v. 67, p. 175–197, doi: 10.1016/0037-0738(90)90034-Q. Bessière, G., Tambareau, Y., and Villatte, J., 1980, Le Passage Crétacé-Tertiaire dans les Hautes Corbières: Extrait du Bulletin de la Société d’Histoire Naturelle de Toulouse, v. 116, no. 3–4, p. 283–303. Bessière, G., Bilotte, M., Crochet, B., Peybernès, B., Tambareau, Y., and Villatte, J., 1989, Livret explicatif de la carte géologique de la France (Feuille 1077: Quillan): Orléans, Éditions du Bureau des Recherches Géologiques et Minières, 98 p. Bilotte, M., 1978, Evolution sédimentaire et tectonique du basin sous-Pyrénéen à la fin du Crétacé, à l’Est de la Garonne: Bulletin de la Société Géologique de France, série 7, tome XX, no. 5, p. 649–655. Bilotte, M., Tambareau, Y., and Villatte, J., 1983, Le Crétacé supérieure et la limite Crétacé-Tertiaire en faciès continental dans le versant nord des Pyrénées: Géologie Méditérranéenne, v. 10, no. 3–4, p. 269–276. Bilotte, M., Capera, J.-C., Debat, P., Debroas, E.J., Duranthon, F., Marre, J., Pulou, R., Rey, J., and Vizcaino, D., 1989, Géologie du Département de l’Aude: Carcassone, Mémoire de la Société d’Études Scientifiques de l’Aude, 79 p. Bodergat, A.M., 1974, Les Microcodiums: Milieux et modes de devéloppement [Ph.D. thesis]: Lyon, Documents du Laboratoire de Géologie de la Faculté des Sciences de Lyon, v. 62, p. 137–235. Bousquet, J.-C., 1997, Géologie du Languedoc-Roussillon: Montpellier, Les Presses du Languedoc, Éditions du Bureau des Recherches Géologiques et Minières, 142 p. Bousquet, J.-C., and Vianey-Liaud, M., 2001, Dinosaures et autres reptiles du Languedoc: Montpellier, Les Presses du Languedoc, Éditions du Bureau des Recherches Géologiques et Minières, 199 p.
185
Bown, T.M., and Kraus, M.J., 1987, Integration of channel and floodplain suites in aggrading alluvial systems. I. Developmental sequences and lateral relations of lower Eocene alluvial palaeosols, Willwood Formation, Bighorn Basin, Wyoming: Journal of Sedimentary Petrology, v. 57, p. 589–601. Branner, J.C., 1910, Aggraded limestone plain of the interior of Bahia and the climatic changes suggested by them: Geological Society of America Bulletin, v. 2, no. 2, p. 187–206. Brewer, R., 1964, Fabric and Mineral Analysis of Soils: New York, John Wiley and Sons, 470 p. Burne, R.V., Bauld, J.R., and Dedeckker, R., 1980, Saline lake charophytes and their geological significance: Journal of Sedimentary Petrology, v. 50, p. 281–293. Camoin, G., Casanova, J., Rouchy, J.M., Blanc-Valleron, M.M., and Deconinck, J.F., 1997, Environmental controls on perennial and ephemeral carbonate lakes: The central palaeo-Andean Basin of Bolivia during Late Cretaceous to early Tertiary times: Sedimentary Geology, v. 113, p. 1–26, doi: 10.1016/S0037-0738(97)00052-3. Charrière, A., and Durand-Delga, M., 2004, Le jeu fini-Crétacé du front nordPyrénéen aux environs de Cucugnan (Corbières méridionales, Aude, France): Comptes Rendus Geoscience, v. 336, no. 13, p. 1199–1208, doi: 10.1016/j.crte.2004.07.001. Chilingarian, G.V., and Wolf, K.H., 1988, Diagenesis II: Developments in Sedimentology, v. 43, 268 p. Cohen, A.S., and Thouin, C., 1987, Nearshore carbonate deposits in Lake Tanganyika: Geology, v. 15, p. 414–418, doi: 10.1130/0091-7613(1987)15<414: NCDILT>2.0.CO;2. Cojan, I., 1999, Carbonate-rich paleosoils in the Late Cretaceous–early Palaeogene series of the Provence Basin (France): International Association of Sedimentologists Special Publication 27, p. 323–335. Cojan, I., Moreau, M.G., and Stott, L., 1998, Significance of the carbonaterich paleosols for building integrated δ13C and magneto-stratigraphy in continental series (early Cenozoic-Provence-France): 15th meeting of the International Sedimentological Congress, Abstract Volume: Alicante, International Association of Sedimentologists (IAS) and University of Alicante, p. 267. Cowardin, L.M., and Golet, F.C., 1995, U.S. Fish and Wildlife Service 1979 wetland classification: A review: Vegetatio, v. 118, no. 1–2, p. 139–152, doi: 10.1007/BF00045196. Cowardin, L.M., Carter, V., Golet, F.C., and La Roe, E.T., 1979, Classification of wetlands and deepwater habitats of the United States: Washington D.C., U.S. Fish and Wildlife Service, Publication FWS/OBS-79/31, 103 p. Dean, W.E., 1981, Carbonate minerals and organic matter in sediments of modern north temperate hard-water lakes, in Ethridge, F.G., and Flores, R.M., eds., Recent and Ancient Non-Marine Depositional Environments: Models for Explanation: Society for Sedimentary Geology Special Publication 31, p. 213–231. Dean, W.E., and Fouch, T.D., 1983, Lacustrine environment, in Scholle, P.A., Bebout, D.G., and Moore, C.H., eds., Carbonate Depositional Environments: Memoire of the American Association of Petroleum Geologists, v. 33, p. 96–130. Deocampo, D.M., 2002, Sedimentary processes and lithofacies in lake-margin groundwater-fed wetlands in East Africa, in Renaut, R.W., and Ashley, G.M., eds., Sedimentation in Continental Rifts: Society for Sedimentary Geology Special Publication 72, p. 295–308. Dunagan, S.P., and Driese, S.G., 1999, Controls of terrestrial stabilization on Late Devonian palustrine carbonate deposition: Catskill magnafacies, New York, U.S.A.: Journal of Sedimentary Research, v. 69, p. 772–783. Dunagan, S.P., and Turner, C.E., 2004, Regional paleohydrologic and paleoclimatic settings of wetland/lacustrine depositional systems in the Morrison Formation, Western Interior, USA: Sedimentary Geology, v. 167, p. 269–296, doi: 10.1016/j.sedgeo.2004.01.007. Esteban, M., 1974, Caliche textures and “Microcodium”: Bolletino della Società Geologica Italiana, v. 92, supplement 1973, p. 105–125. Esteban, M., and Klappa, C.F., 1983, Subaerial exposure environment, in Scholle, P.A., Bebout, D.G., and Moore, C.H., eds., Carbonate Depositional Environments: Memoire of the American Association of Petroleum Geologists, v. 33, p. 1–54. Finlayson, C.M., and van der Valk, A.G., (guest editors), 1995a, Classification and inventory of the world’s wetlands: Vegetatio, v. 118, no. 1–2, 192 p. Finlayson, C.M., and van der Valk, A.G., 1995b, Wetland classification and inventory: A summary: Vegetatio, v. 118, no. 1–2, p. 185–192, doi:
186
Marty and Meyer
10.1007/BF00045199. Flügel, E., 2004, Microfacies of Carbonate Rocks: Berlin, Springer-Verlag, 976 p. Freytet, P., 1964, Le Vitrollien des Corbières orientales; réflexions sur la sedimentation “lacustre” nord-Pyrénéenne; divagation fluviatile, biorhexistasie, pédogenèse: Revue de Géographie Physique et de Géologie Dynamique, v. 7, no. 3, p. 179–199. Freytet, P., 1970, Les dépôts continentaux et marins du Crétacé supérieur et des couches de passage à l’Eocène en Languedoc [Ph.D. thesis]: Orsay, Travaux du Laboratoire de Géologie Structurale et Appliquée, v. 91, 490 p. Freytet, P., 1971a, Les dépôts continentaux et marins du Crétacé supérieur et des couches de passage à l’Eocène en Languedoc: Bulletin du Bureau des Recherches Géologiques et Minières (2ème série): Colloque Géologie de la France, section I, v. 4, p. 1–54. Freytet, P., 1971b, Paléosols résiduels et paléosols alluviaux hydromorphes dans le Crétacé supérieur et l’Eocène basal en Languedoc: Revue de Géographie Physique et de Géologie Dynamique, série 2, v. 13, p. 245–268. Freytet, P., 1973, Petrography and paleoenvironment of carbonated continental deposits with particular reference to the Upper Cretaceous and Lower Eocene of Languedoc (southern France): Sedimentary Geology, v. 10, p. 25–60, doi: 10.1016/0037-0738(73)90009-2. Freytet, P., 1984, Carbonate lacustrine sediments and their transformations by emersion and pedogenesis: Importance of identifying them for paleogeographical reconstructions: Bulletin des Centres de Recherches Exploration–Production Elf-Aquitain, v. 8, p. 223–247. Freytet, P., and Plaziat, J.-C., 1982, Continental carbonate sedimentation and pedogenesis, Late Cretaceous and early Tertiary of southern France: Contributions to Sedimentology, v. 12, 213 p. Freytet, P., and Verrecchia, E.P., 2002, Lacustrine and palustrine carbonate petrography: An overview: Journal of Paleolimnology, v. 27, p. 221–237, doi: 10.1023/A:1014263722766. Freytet, P., Plaziat, J.C., and Verrecchia, E.P., 1997, A classification of rhizogenic (root-formed) calcretes, with examples from the Upper Jurassic– Lower Cretaceous of southern France—Discussion: Sedimentary Geology, v. 110, p. 299–303, doi: 10.1016/S0037-0738(96)00090-5. Galbrun, B., 1989, Résultats magnétostratigraphiques à la limite RognacienVitrollien: Précisions sur la limite Crétacé-Tertiaire dans le basin d’Aixen-Provence, in La limite Crétacé-Tertiaire dans le basin de l’Arc (sud-est, France): Cahiers de la Reserve Géologique de Provence, v. 1, p. 34–37. Galbrun, B., Rasplus, L., and Durand, J.P., 1991, La limite Crétacé-Tertiaire dans le domaine provençal: Étude magnétostratigraphique du passage Rognacien-Vitrollien à l’ouest du synclinal de l’Arc: Comptes Rendus de l’Académie Scientifique de Paris, série 2, v. 312, p. 1467–1473. Garcia, A., 1994, Charophyta: Their use in paleolimnology: Journal of Paleolimnology, v. 10, p. 43–52, doi: 10.1007/BF00683145. Ginsburg, R.N., 1975, Tidal Deposits: Berlin, Springer-Verlag, 428 p. Ginsburg, R.N., Hardie, L.A., Bricker, O.P., Garret, P., and Wanless, H.R., 1977, Exposure index: A quantitative approach to defining position within the tidal zone, in Hardie, L.A., ed., Sedimentation on the modern carbonate tidal flats of northwest Andros Island, Bahamas: Baltimore and London, The Johns Hopkins University Press, The Johns Hopkins University Studies in Geology, v. 22, p. 7–11. Glück, H., 1912, Eine neue gesteinsbildende Siphonee (Codiacée) aus dem marinen Tertiär von Süddeutschland: Mitteilungen der Badischen Geologischen Landesanstalt, v. 7, p. 3–24. Groebke, I., 2001, Sedimentologie und Paläoökologie der oberen Kreide und Kreide/Tertiärgrenze des Haute Vallée de l’Aude Südwestfrankreich [M.S. thesis]: Switzerland, University of Basel, 68 p. Hardie, L.A., 1977, Sedimentation on the modern carbonate tidal flats of northwest Andros Island, Bahamas: Baltimore and London, The Johns Hopkins University Press, The John Hopkins University Studies in Geology, v. 22, 202 p. Hardie, L.A., and Shinn, E.A., 1986, Carbonate depositional environments modern and ancient; Part 3. Tidal flats: Golden, Colorado, Colorado School of Mines Quarterly, v. 81, no. 1, 74 p. Hofmann, C.-C., and Zetter, R., 2005, Reconstruction of different wetland plant habitats of the Pannonian Basin system (Neogene, eastern Austria): Palaios, v. 20, p. 266–279. Jaffrezo, M., 1977, Pyrénées Orientales Corbières: Guides Géologiques Régionaux: Paris, Masson, 191 p. Jaillard, B., Guyon, A., and Maurin, A.F., 1991, Structure and composition of
calcified roots, and their identification in calcareous soils: Geoderma, v. 50, p. 197–210, doi: 10.1016/0016-7061(91)90034-Q. Klappa, C.F., 1978a, Morphology, composition and genesis of Quaternary calcretes from the western Mediterranean: A petrographic approach [Ph.D. thesis]: Liverpool, University of Liverpool, 446 p. Klappa, C.F., 1978b, Biolithogenesis of Microcodium: Elucidation: Sedimentology, v. 25, p. 489–522. Kosir, A., 2004, Microcodium revisited: Root calcification products of terrestrial plants on carbonate-rich substrates: Journal of Sedimentary Research, v. 74, p. 845–857. Liutkus, C.M., and Ashley, G.M., 2003, Facies model of a semi-arid freshwater wetland, Olduvai Gorge, Tanzania: Journal of Sedimentary Research, v. 73, p. 691–705. Martín-Closas, C., 1999, Epiphytic overgrowth of charophyte thalli by stromatolite-like structures and fungi in the Lower Cretaceous of the Iberian Ranges (Spain): Australian Journal of Botany, v. 47, p. 305–313, doi: 10.1071/BT97098. Marty, D., 2001, Sedimentology, palaeoecology and stratigraphy of the palustrine Faciès Rognacien at the K/T-boundary in southern France [M.S. thesis]: Switzerland, University of Basel, 70 p. Marty, D., 2004, Sedimentology, palaeoenvironment, and stratigraphy of the palustrine Faciès Rognacien at the K/T-boundary in southern France [abs.]: 32nd International Geological Congress, Scientific sessions: Florence, Italy, part 1, p. 110. Mazullo, S.J., and Birdwell, B.A., 1989, Syngenetic formation of grainstones and pisolites from fenestral carbonates in peritidal settings: Journal of Sedimentary Petrology, v. 59, p. 605–611. Mitsch, W.J., and Gosselink, J.G., 1993, Wetlands (2nd edition): New York, Van Nostrand Reinhold, 722 p. Mojon, P.O., 1990, Polymorphism écophénotypique et paléoécologie des porocharacées (charophytes) du Crétacé basal (Berriasien) du Jura FrancoSuisse: Révue de Paléobiologie, v. 8, p. 505–524. Monty, C.L.V., 1972, Recent algal stromatolitic deposits, Andros Island, Bahamas: Preliminary report: Geologische Rundschau, v. 61, p. 742–783, doi: 10.1007/BF01896344. Monty, C.L.V., 1981, Phanerozoic Stromatolites: Berlin, Springer-Verlag, 471 p. Monty, C.L.V., and Hardie, L.A., 1976, The geological significance of the freshwater blue green algal calcareous marsh, in Walter, M.R., ed., Stromatolites: Amsterdam, Elsevier, p. 447–477. Müller, G., and Wagner, F., 1978, Holocene evolution in Lake Balaton (Hungary): A response to climate and impact of man, in Matter, A., and Tucker, M.A., eds., Modern and Ancient Lake Sediments: International Association of Sedimentologists Special Publication 2, p. 57–81. Murphy, D.H., and Wilkinson, B.H., 1980, Carbonate deposition and facies distribution in a central Michigan marl lake: Sedimentology, v. 27, p. 123–135. Pédro, G., 1968, Distribution des principaux types d’altération chimique à la surface du globe: Présentation d’une esquisse géographique: Revue de Géographie Physique et de Géologie Dynamique, v. 10, no. 5, p. 457–470. Peryt, T., 1983, Coated Grains: Berlin, Springer-Verlag, 655 p. Peybernès, B., and Combes, P.-J., 1999, Paléosurfaces d’érosion et paléokarsts dans la série de passage Crétacé/Tertiaire des Corbières occidentales (Aude, France): Earth and Planetary Science Letters, v. 329, p. 815–822. Platt, N.H., 1989a, Lacustrine carbonates and pedogenesis: Sedimentology and origin of palustrine deposits from the Early Cretaceous Rupelo Formation, W Cameros Basin: N Spain: Sedimentology, v. 36, p. 665–684. Platt, N.H., 1989b, Climatic and tectonic controls on sedimentation of a Mesozoic lacustrine sequence: The Purbeck of the Western Cameros Basin, northern Spain: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 70, p. 187–197, doi: 10.1016/0031-0182(89)90089-8. Platt, N.H., 1989c, Continental sedimentation in an early evolving rift basin: The Lower Cretaceous of the western Cameros Basin (northern Spain): Sedimentary Geology, v. 64, p. 91–109, doi: 10.1016/0037-0738(89)90086-9. Platt, N.H., 1995, Sedimentation and tectonics of a synrift succession: Upper Jurassic alluvial fans and paleokarst at the late Cimmerian unconformity, western Cameros Basin, northern Spain, in Plint, A.J., ed., Clastic Facies Analysis: International Association of Sedimentologists Special Publication 22, p. 219–236. Platt, N.H., and Pujalte, V., 1994, Correlation of Upper Jurassic–Lower Cretaceous continental sequences from the southern Biscay margin, northern Spain: Journal of the Geological Society of London, v. 151, p. 715–726.
Palustrine carbonates of SW France Platt, N.H., and Wright, V.P., 1991, Lacustrine carbonates: Facies models, facies distributions and hydrocarbon aspects, in Anadón, P., Cabrera, L., and Kelts, K., eds., Lacustrine Facies Analysis: International Association of Sedimentologists Special Publication 13, p. 57–74. Platt, N.H., and Wright, V.P., 1992, Palustrine carbonates and the Florida Everglades: Towards an exposure index for the fresh-water environment?: Journal of Sedimentary Petrology, v. 62, p. 1058–1071. Plaziat, P., 1970, La limite Crétacé-Tertiaire en Alava méridionale (Pays basque espagnol): Le Rognacien n’y est pas l’équivalent continental du Danien: Compte-Rendu Sommaire de la Société Géologique de France, v. 3, p. 77–78. Plaziat, J.C., 1981, Late Cretaceous to late Eocene palaeogeographic evolution of southwest Europe: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 36, p. 263–320, doi: 10.1016/0031-0182(81)90110-3. Plaziat, J.C., 1984, Le Domaine Pyrénéen, de la fin du Crétacé à la fin de l’Eocène: Stratigraphie, paléoenvironments et évolution paléogéographique [Ph.D. thesis]: Paris, Université Paris-Sud Orsay, Order number 2885, tome I–III, 1362 p. Plaziat, J.C., and Freytet, P., 1978, Le pseudo-microkarst pédologique: Un aspect particulier des paléopédogenèses developées sur les dépots calcaires lacustres dans le Tertiare du Languedoc: Comptes Rendus de l’Académie Scientifique de Paris, v. 286, p. 1661–1664. Quade, J., Mifflin, M.D., Pratt, W.L., McCoy, W., and Burckle, L., 1995, Fossil spring deposits in the southern Great Basin and their implications for changes in water-table levels near Yucca Mountain, Nevada, during Quaternary time: Geological Society of America Bulletin, v. 107, p. 213–230, doi: 10.1130/0016-7606(1995)107<0213:FSDITS>2.3.CO;2. Quade, J., Forester, R.M., and Whelan, J.F., 2003, Late Quaternary paleohydrologic and paleotemperature change in southern Nevada, in Enzel, Y., Wells, S.G., and Lancaster, N., eds., Paleoenvironments and Paleohydrology of the Mojave and Southern Great Basin Deserts: Geological Society of America Special Paper 368, p. 165–188. Retallack, G.J., 1990, Soils of the Past: An Introduction to Paleopedology: London, Unwin Hyman, 520 p. Rocchia, R., Boclet, D., Bonte, P., Jehanno, C., Pagnier, R., Reys, J.L., Durand, J.P., Feist, M., Galbrun, B., Medus, J., Freytet, P., Bilotte, M., Massieux, M., Tambareau, Y., and Villatte, J., 1989, Recherche d’un niveau à iridium au voisinage de la limite Crétacé-Tertiaire dans le domaine continental du Sud de la France: Bassin d’Aix-en-Provence, Corbières, Plantaurel, Petites Pyrénées: Cahières Réserve Géologique de Haute Provence, v. 1, p. 51–59. Ruellan, A., 1967, Individualisation et accumulation du calcaire dans les sols et les dépôts Quarternaires du Maroc: Cahiers O.R.S.T.O.M. (Office de la Recherche Scientifique et Technique Outré-Mer), Série Pédologie, v. 4, p. 421–462. Schneider, J., Schröder, H.G., and Le Campion-Alsumard, Th., 1983, Algal micro-reefs—Coated grains from freshwater environments, in Peryt, T.M., ed., Coated Grains: Berlin, Springer-Verlag, p. 284–298. Schudack, M., 1993, Möglichkeiten palökologischer Aussagen mit Hilfe von fossilen Charophyten: Festschrift Prof. W. Krutzsch: Berlin, Museum für Naturkunde, p. 39–60. Schudack, M.E., Turner, C.E., and Peterson, F., 1998, Biostratigraphy, paleoecology and biogeography of charophytes and ostracodes from the Upper Jurassic Morrison Formation, Western Interior, USA: Modern Geology, v. 22, p. 379–414. Schullenberger, E.D., Rech, J.A., and Currie, B., 2004, Upper Miocene palustrine carbonates (Opache Formation) in the Calama Basin, Chile, and their paleoclimatic implications: Geological Society of America Abstracts with Programs, v. 36, no. 5, p. 77.
187
Scott, D.A., and Jones, T.A., 1995, Classification and inventory of wetlands: A global overview: Vegetatio, v. 118, p. 3–16, doi: 10.1007/BF00045186. Steinen, R.P., 1974, Phreatic and vadose diagenetic modification of Pleistocene limestone; petrographic observations from subsurface of Barbados, West Indies: American Association of Petroleum Geologists Bulletin, v. 58, p. 1008–1024. Tambareau, Y., Crochet, B., Villatte, J., and Deramond, J., 1995, Evolution tectono-sédimentaire du versant nord des Pyrénées centre-orientales au Paléocène et à l’Eocène inférieur: Bulletin de la Société Géologique de France, v. 166, no. 4, p. 375–387. Tambareau, Y., Hottinger, L., Rodriguez-Lazaro, J., Villatte, J., Babinot, J.-F., Colin, J.-P., Garcia-Zarraga, E., Rocchia, R., and Guerrero, N., 1997, Communautés fossiles benthiques aux alentours de la limite Crétacé/Tertiaire dans les Pyrénées: Bulletin de la Société Géologique de France, v. 168, no. 6, p. 795–804. Tandon, S.K., and Andrews, J.E., 2001, Lithofacies associations and stable isotopes of palustrine and calcrete carbonates: Examples from an Indian Maastrichtian regolith: Sedimentology, v. 48, p. 339–355, doi: 10.1046/ j.1365-3091.2001.00367.x. Tarnocai, C., 1979, Canadian wetland registry, in Rubec, C.D.A., and Pollett, F.C., eds., Workshop Canadian Wetlands Environments: Canadian Land Directorate, Ecological Land Classification Series, v. 12, p. 9–38. Valero Garcés, B.L., Gierlowski-Kordesch, E., and Bragonier, W.A., 1994, Lacustrine facies models for non-marine limestone within cyclothems in the Pennsylvanian (Upper Freeport Formation, Appalachian Basin) and its implications, in Lomando, A.J., Schreiber, B.C., and Harris, P.M., eds., Lacustrine Reservoirs and Depositional Systems: Society for Sedimentary Geology Core Workshop, v. 19, p. 321–381. Villot, L., 1883, Étude sur le bassin de Fuveau et sur un grand travail à y executer: Annales des Mines, v. 8, tome IV, p. 5–66, Juillet–Août 1885. Westphal, M., and Durand, J.P., 1990, Magnétostratigraphie des séries continentales fluvio-lacustres du Crétacé supérieur dans le synclinal de l’Arc (région d’Aix-en-Provence, France): Bulletin de la Société Géologique de France, v. 6, p. 609–620. Winsborough, B.M., Seeler, J.-S., Golubic, S., Folk, R.L., and Maguire, B., Jr., 1994, Recent fresh-water lacustrine stromatolites, stromatolitic mats and oncoids from northeastern Mexico, in Bertrand-Sarfati, J., and Monty, C., eds., Phanerozoic Stromatolites II: Amsterdam, Kluwer Academic Publishers, p. 71–100. Wright, V.P., 1990, Syngenetic formation of grainstones and pisolites from fenestral carbonates in peritidal settings: Discussion: Journal of Sedimentary Petrology, v. 60, p. 309–310. Wright, V.P., and Platt, N.H., 1995, Seasonal wetland carbonate sequences and dynamic catenas: A re-appraisal of palustrine limestones: Sedimentary Geology, v. 99, p. 65–71, doi: 10.1016/0037-0738(95)00080-R. Wright, V.P., Platt, N.H., and Wimbeldon, W.A., 1988, Biogenic laminar calcretes: Evidence for calcified root mat horizons in paleosols: Sedimentology, v. 35, p. 603–620. Wright, V.P., Platt, N.H., Marriott, S.B., and Beck, V.H., 1995, A classification of rhizogenic (root-formed) calcretes with examples from the Upper Jurassic–Lower Cretaceous of Spain and Upper Cretaceous of southern France: Sedimentary Geology, v. 100, p. 143–158, doi: 10.1016/00370738(95)00105-0. Wright, V.P., Alonso-Zarza, A., Sanz, M.E., and Calvo, J.P., 1997, Diagenesis of late Miocene lacustrine carbonates, Madrid Basin, Spain: Sedimentary Geology, v. 114, p. 81–95, doi: 10.1016/S0037-0738(97)00059-6. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
Reworked Microcodium calcarenites interbedded in pelagic sedimentary rocks (Paleocene, Subbetic, southern Spain): Paleoenvironmental reconstruction José M. Molina† Departamento de Geología, Universidad de Jaén, Facultad de Ciencias Experimentales, 23071 Jaén, Spain Juan A. Vera‡ Departamento de Estratigrafía y Paleontología, Universidad de Granada, Facultad de Ciencias, 18071 Granada, Spain Roque Aguado§ Departamento de Geología, Universidad de Jaén, Escuela Politécnica Superior de Linares, 23700 Linares Jaén, Spain ABSTRACT The Majalcorón Formation (Paleocene) is an unusual lithostratigraphic unit mainly made up of calcarenites with reworked Microcodium. Analysis of this unit from different localities of the Subbetic (southern Spain) shows that the formation is interbedded in pelagic sedimentary rocks. Calcareous nannofossils in the latter confirm that deposition of the Majalcorón Formation began in the early Danian and finished in the latest Danian–early Selandian. The reworked Microcodium calcarenites are grainstones mainly constituting disaggregated Microcodium prisms and sparite cement. Secondarily, they contain quartz grains, glauconite, mud clasts, benthonic and planktonic foraminifera, coal fragments, and plant remains. In some beds, hummocky cross-stratification and mound-shaped structures appear. The reworked Microcodium calcarenites are derived from the erosion of calcareous paleosols in source areas located to the north. A eustatic regressive-transgressive-regressive succession at the Cretaceous-Tertiary boundary and in the early Paleocene is responsible for the beginning of the Microcodium paleosol development. Sedimentation of the calcarenites took place in a shallow marine ramp after erosion of the paleosols, and ended with the development of paleokarstic features on the top of the formation. This interpretation of the Majalcorón Formation as a shallowmarine deposit is important to the understanding of the paleogeography and depositional paleobathymetry of the adjacent pelagic facies during the Late Cretaceous and Paleogene. Keywords: calcarenites, Microcodium, sea-level changes, nannofossils, Paleocene, Subbetic.
E-mail:
[email protected] E-mail:
[email protected] § E-mail:
[email protected] † ‡
Molina, J.M., Vera, J.A., and Aguado, R., 2006, Reworked Microcodium calcarenites interbedded in pelagic sedimentary rocks (Paleocene, Subbetic, southern Spain): Paleoenvironmental reconstruction, in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 189–202, doi: 10.1130/2006.2416(12). For permission to copy, contact editing@geosociety. org. ©2006 Geological Society of America. All rights reserved.
189
190
Molina et al. RESUMEN Se estudia una unidad estratigráfica peculiar, la Formación Majalcorón del Paleoceno en el Subbético, compuesta mayoritariamente por calcarenitas con Microcodium retrabajado. Esta unidad se encuentra intercalada entre rocas sedimentarias pelágicas con nanoplancton calcáreo que ha permitido establecer que su depósito comenzó en el Daniense inferior y finalizó en el Daniense terminal–Selandiense inferior. Las calcarenitas de Microcodium resedimentado son grainstones compuestos mayoritariamente por prismas disgregados de Microcodium y cemento esparítico. Además hay granos de cuarzo, glauconita, intraclastos micríticos, foraminíferos planctónicos y bentónicos, y fragmentos de plantas. En algunos estratos aparece estratificación cruzada de tipo hummocky y estructuras en montículo. Las calcarenitas de Microcodium proceden de la erosión de paleosuelos calcáreos. Se propone que las áreas fuente que proporcionaron los importantes volúmenes de Microcodium se localizaban al norte, en el Subbético Externo. Cambios eustáticos del nivel del mar en el límite Cretácico-Terciario y en el Paleoceno inferior, en una sucesión regresiva-transgresiva-regresiva fueron los principales responsables del comienzo del desarrollo de los paleosuelos con Microcodium, de la sedimentación de las calcarenitas después de la erosión de los paleosuelos y del final de su depósito con el desarollo de rasgos paleokársticos en el techo de la formación. La consideración de la Formación Majalcorón como un depósito marino somero es importante para interpretar la paleogeografía y paleobatimetría de las facies pelágicas adyacentes durante el Cretácico Superior y Paleógeno. Palabras clave: calcarenitas, Microcodium, cambios del nivel del mar, nanofósiles, Paleoceno, Subbético.
INTRODUCTION The Paleogene of many Alpine-Mediterranean successions is characterized by the presence of Microcodium, a typical feature of many calcretes and calcareous paleosols. Most of the accumulations of Microcodium occur within continental depositional settings in palustrine, fluvial, and, more rarely, karstic environments. The presence of important accumulations of reworked disaggregated Microcodium in marine deposits is rare but very interesting, mainly as a good criterion for recognizing subaerial exposure environments. This paper studies the Majalcorón Formation, which is composed of Microcodium calcarenites deposited in marine environments. Microcodium is a problematic calcareous structure of cylindrical or spherical shape and of millimeter to submillimeter scale; it consists of aggregates of elongated calcite crystals surrounding a small central hole. It is common in calcretes and eolianites, and most researchers agree that it is a product of the edaphic alteration of carbonates; it originates in subaerial environments and is formed by the calcification of symbiotic associations of fungi and plant roots (mycorrhiza) within paleosols (Klappa, 1978, 1980; Freytet and Plaziat, 1982; Wright and Tucker, 1991; AlonsoZarza, 2003; Kosir, 2004; and references therein). This creates an interesting problem in explaining the origin and paleogeographical significance of the huge volumes of Microcodium that make up Paleocene formations that are interbedded among pelagic sed-
imentary rocks, such as the Majalcorón Formation and equivalent deposits within and outside of the Betic Cordillera. In this paper, we study the stratigraphy of the Microcodium calcarenites of the Subbetic (Betic Cordillera) belonging to the Majalcorón Formation, we indicate similar facies present in other units of the Betic Cordillera, and finally we discuss the paleogeographical significance of this formation. THE MAJALCORÓN FORMATION: SETTING AND STRATIGRAPHY The Betic Cordillera is an Alpine mountain belt in southern Spain in which three areas are recognized (Fig. 1A): the Betic external zones and the Betic internal zones, separated by the Campo de Gibraltar units (Vera, 2004). The Betic external zones
Figure 1. (A) Geologic sketch of the Betic Cordillera with the location of the studied outcrops of the Majalcorón Formation. (B) Geographical situation of the Majalcorón Formation outcrops with the location of the holostratotype (Peñas de Majalcorón), parastratotype (Fuente de la Pileta), and other analyzed sections: 1—northwest of Venta de Agramaderos, 2—Cazuela del Pozo, 3—north of Fuente de la Pileta, 4—Rosal Bajo, 5—east of Cortijo de Santa Teresa, and 6—Pilas de la Fuente del Soto.
(Emerged)
1
?
8
7
5 6 7 8
In
al Z ter n
one
IN
SPA
2: UNDEFORMED COVER OF THE IBERIAN MASSIF ea
anean S Mediterr
s
3: External Prebetic
ÁFRICA
4: Internal Prebetic 5: Intermediate Domain
AFRICAN PLATE
?
6: External Subbetic
ALICANTE
7: Median Subbetic
u alq
ivir
si Ba
JAÉN
n
9: Olistostromic Unit
10: INTERNAL ZONES
N
GRANADA
A NE
A SE
A
0
50
100
4° 05’
R ER
GIBRALTAR UNITS
N
Volcanic Neogene rocks Post-orogenic basins
T
DI
ME
B
8: Internal Subbetic
M
ad
Gu
MURCIA
EXTERNAL ZONES
IBERIAN PLATE
2 3-4
1: HERCINIAN MASSIF
Atlantic Ocean
A
M
150 km
4° 00’
MAJALCORÓN Fm STUDIED OUTCROPS
3° 55’
MAJALCORÓN FORMATION ALCALÁ LA REAL
N STRATOTYPE 37° 25’
37° 25’ Province of Córdoba
Peñas de Majalcorón
La Pedriza
Villalobos
Province of Granada Lojilla Venta de Agramaderos
Province of Jaén
PARASTRATOTYPE 37° 20’
37° 20’
0
MONTEFRÍO 4° 05’
4° 00’
3° 55’
1
2 km
192
Molina et al.
form a complex of thrust sheets composed of thick successions of Triassic to Lower Miocene sedimentary rocks detached from a Paleozoic basement that corresponds to the southern prolongation of the Iberian Variscan Massif. These rocks were deposited on the southern Iberian continental margin. The external zones are subdivided in two tectonic zones that are roughly based on the two great paleogeographical domains of the southern Iberian continental margin: a platform area to the north, the Prebetic; and a mainly pelagic area to the south, the Subbetic. Four paleogeographical domains can be distinguished in the Subbetic area (Fig. 1A); these formed well-defined troughs and swells during Middle and Late Jurassic times. In the troughs, the Intermediate Domain (5 in Fig. 1A) and the Median Subbetic (7 in Fig. 1A), thick successions of pelagic sediments, were deposited. In contrast, reduced and/or condensed sequences were deposited in the swells; these include the External and Internal Subbetic (6 and 8 in Fig. 1A). The Majalcorón Formation was defined by Molina et al. (2003) as a new lithostratigraphic unit in the Subbetic of the Betic Cordillera (Fig. 1A). It is characterized by its lithology of calcarenites with abundant Microcodium. This formation outcrops mainly in the central sector of the Median Subbetic, in the provinces of Granada and Jaén (Fig. 1). In most of these outcrops, tectonic deformation and, additionally, intense cultivation make it difficult to measure complete detailed stratigraphic sections. Two stratigraphic sections have enabled us, however, to measure the entire formation, bed by bed: these sections are, respectively, the holostratotype (Peñas of Majalcorón, Fig. 1B) and the parastratotype (to the north of the Fuente de la Pileta, Fig. 1B). We have studied other sections outcropping to the northwest of Venta de Agramaderos, in Cazuela del Pozo, to the north of Fuente de la Pileta, in Rosal Bajo, to the east of Cortijo de Santa Teresa, and in Pilas de la Fuente de Soto. These locations, numbered 1–6 respectively, can be seen in the Figure 1B. The Majalcorón Formation takes its name from the village of Peñas de Majalcorón in Alcalá la Real (province of Jaén), very near the boundary between the provinces of Jaén, Córdoba, and Granada (Fig. 1B). This village is found at the foot of a sharp relief (Peñas of Majalcorón) that contains a magnificent outcrop of Microcodium calcarenites, where the formation holostratotype has been established. Its thickness here is 56 m, and its main characteristics are represented in Figure 2. Figures 3A and 3B show the outcrop appearance. The parastratotype of this formation is located 3 km to the north-northeast of Montefrío (province of Granada), 400 m to the north of the Fuente de la Pileta (Fig. 1B). This section was previously studied by Martínez-Gallego and Roca (1973), who dated it by its content in planktonic foraminifera, as Late Danian (Globorotalia trinidadensis zone of Bolli, equivalent to the Globorotalia compressa/Globigerina daubjergensis zone of Loeblich and Tappan). The detailed stratigraphic succession (Fig. 2) has a total thickness of 39.5 m (shown in Fig. 3C). The Majalcorón Formation is located on the Upper Cretaceous scaglia rossa-like, pinkish pelagic marly limestones of the
Capas Rojas Formation (Vera et al., 1982). Above the Majalcorón Formation, which shows paleokarst features and neptunian dikes at the top, there are marly limestones of the same Capas Rojas Formation or, in other sites, gray or yellowish marls with nummulite-bearing turbiditic bioclastic sandstones of the Eocene. Preliminary data on the age of this formation and on its sedimentological significance have been provided by Aguado et al. (2003) and Vera et al. (2003), respectively. The maximum thickness of this formation of ~60 m is reached in the sector of the Peñas de Majalcorón and to the northwest of the Venta de Agramaderos (Fig. 1B). Biostratigraphy A biostratigraphic study has been carried out by means of nannofossils that show well-diversified associations characteristic of median-low paleolatitudes. All Tertiary samples contained a variable proportion of reworked nannoflora that included Late Cretaceous (Campanian and Maastrichtian) and some Early Cretaceous taxa. We applied the zonal outline of Varol (1989) to the Tertiary samples, and that of Aguado (1993) to the Cretaceous samples for biostratigraphic assignment. Figure 4 shows the position and age of the samples and illustrates the most characteristic nannofossil assemblages. In the holostratotype (Peñas de Majalcorón) and adjacent areas (Pilas de la Fuente del Soto), the samples just below the bottom of the Majalcorón Formation (FP-1 and FP-2) provided a nannofossil assemblage typical of the NTp1B zone, of early Danian age. The samples immediately above the calcarenitic body (MJ-T-2 and MJ-T-3) contain a nannofossil assemblage characteristic of the NTp8C zone of middle Selandian age (Varol, 1989; see also Fig. 4). In the parastratotype (Fuente de la Pileta), the samples below the Majalcorón Formation (PI-M-1 and PI-M-2) are assigned to the Micula prinsii zone (NBK24) of latest Maastrichtian age. The nannofossil assemblages from samples taken in marly levels located in the median part (PI-17), and immediately above of the calcarenitic body (PI-T), belong to the interval between the zones NTp5B and NTp7, both of a late Danian or early Selandian age (Fig. 4). In Venta de Agramaderos, the nannofossil assemblage below the base of the Majalcorón Formation (sample VA-1) corresponds to the NBK24 zone of latest Maastrichtian age. In samples taken at the top of the calcarenitic body, the two lowest (VA-10R and VA-11R) belong to the interval between the zones NTp6 and NTp7 (latest Danian to early Selandian), but the highest sample (VA-12R), collected within a neptunian dike filling, was assigned to zone NTp8A of early Selandian age (Fig. 4). In Cazuela del Pozo, results for samples above the calcarenitic body (CZ-51 and CZ- 52) are consistent with these findings (Fig. 4). In the area located to the north of Villalobos, the Majalcorón Formation wedges laterally out, and the Microcodium calcarenitic body disappears. The marls and marly limestones, which, by lateral correlation (VL-KT sample), overlie it, contain an association characteristic of zone NTp10C of Varol (1989) of latest Selandian age (Fig. 4).
Thickness Bed Q-3 Q-2
55 m Q-1 P-5
PARASTRATOTYPE Fuente de la Pileta
HOLOSTRATOTYPE Peñas de Majalcorón
H-6 H-5 H-4 H-3 H-2 H-1 G-2
P-4
G-1
P-3
F-4
P-2 P-1
50 m O-3 O-2
Thickness Bed
25 m
26 25 24 23
35 m
22
20 m
21
F-3 F-2 F-1
20
E-3
30 m 19
E-2 O-1 Ñ-6 Ñ-5 Ñ-4 Ñ-3
18
E-1
15 m
25 m
D-4
17
45 m
D-3 D-2
Ñ-2
D-1
Ñ-1
C-1
N-12
B-7 B-6
13 12 11
B-5
10
N-11
40 m
N-10 N-9 N-8 N-7 N-6 N-5 N-4
15 14
20 m
10 m B-4
9
15 m
B-3 B-2
8
B-1 A-8 A-7
7
5m
N-3 N-2
A-6
N-1
A-5 A-4 A-3 A-2
35 m
16
10 m 6 5
5m
M-1
4
A-1 L-3 L-2
30 m L-1
K-1 J-3 J-2 J-1 I-1
0m
3
Hummocky cross stratification
Zoophycos
Wavy lamination
Chondrites
Parallel lamination
0m
Burrows
Wavy top bed
Coal fragments
Calcite veins and speleothems
Slump
Neptunian dikes
Mud pebbles
Iron oxides
2 1
Calcarenites Limestones Marly limestones
25 m Figure 2. Stratigraphic sections of the holostratotype (Peñas de Majalcorón) and parastratotype (Fuente de la Pileta).
194
Molina et al.
Figure 3. (A) Panoramic view of the scarp relief from the southwest of Peñas de Majalcorón. This small village appears to the right, at the foot of the scarp. (B) View of the outcrop to the north of Peñas de Majalcorón where the holostratotype was established. (C) Parastratotype outcrop in the Pileta ravine.
FACIES The Majalcorón Formation is formed by laterally continuous, medium to thick (dm to m) parallel-bedded gray to brown calcarenites, consisting mainly of medium- to fine-grained, well-cemented Microcodium fragments. Locally, these alternate with thin (mm- to cm-thick) beds of whitish to grayish, slightly silty marls and some isolated micritic limestone beds (Figs. 2, 3B, and 3C). The calcarenite beds are typically massive, although, locally, they have horizontal lamination due to the orientation of the Microcodium prisms with their long axes parallel to the bedding, and to the preferential concentration of quartz grains. Hummocky cross-stratification is present in some beds. Moundshaped structures also are present; they have a plane base and wavy top, a wavelength between 60 and 370 cm and height from 4 to 23 cm, grading locally to wedge-shaped strata (Figs. 5A, 5B, and 5C). Cross-beds indicate that the predominant paleocurrent was toward the SE. Mud intraclasts, up to 4 cm long (Fig. 5D), like those described by Molina and Vera (2001), are abundant, as are simple vertical (Skolithos) and horizontal dwelling burrows with micritic fills (Fig. 5E). Water escape structures are present, mainly sand volcanoes, burst-throughs, and flame structures (Fig. 5F). Sedimentary structures change abruptly, from very well-preserved stratification and lamination to massive forms, for example as in the outcrops of Rosal Bajo (4 in Fig. 1), and imply important
Figure 4. Position of the studied samples for nannoplankton, in relation to the top and bottom of the Majalcorón Formation, in the studied sections located in Figure 1B. Plane polarized light micrographs of some selected calcareous nannofossils are shown according to the calcareous nannofossil zones. All images were made under crossed nicols and at magnification of 3000×. In NBK24: 1—Micula murus (Martini) Bukry; 2—Micula prinsii Perch-Nielsen; 3—Cribrocorona gallica (Stradner) Perch-Nielsen; 4—Lithraphidites quadratus Bramlette and Martini. In NTp1B: 1—Neobiscutum parvulum (Romein) Varol; 2—Neobiscutum romeinii (Perch-Nielsen) Varol; 3—Cruciplacolithus primus Perch-Nielsen; 4—Neocrepidolithus neocrassus (Perch-Nielsen) Romein; 5—Neocrepidolithus sp. cf. N. dirimosus (Perch-Nielsen) Perch-Nielsen; 6—Neocrepidolithus fossus (Romein) Romein. In NTp5-NTp7: 1—Prinsius martinii (PerchNielsen) Haq; 2—Toweius pertusus (Sullivan) Romein; 3—Ericsonia subpertusa Hay and Mohler; 4—Ericsonia robusta (Bramlette and Sullivan) Perch-Nielsen; 5—Ellipsolithus macellus (Bramlette and Sullivan) Sullivan; 6—Neochiastozygus modestus Perch-Nielsen. In NTp8: 1—Fasciculithus pileatus Bukry; 2—Fasciculithus janii Perch-Nielsen; 3—Fasciculithus involutus Bramlette and Sullivan; 4—Ellipsolithus macellus (Bramlette and Sullivan) Sullivan; 5—Ellipsolithus distichus (Bramlette and Sullivan) Sullivan; 6—Sullivania danica (Brotzen) Varol; 7—Sphenolithus primus Perch-Nielsen; 8—Neochiastozygus modestus Perch-Nielsen. In NTp10C: 1—Chiasmolithus bidens (Bramlette and Sullivan) Hay and Mohler; 2 Heliolithus kleinpellii Sullivan; 3—Fasciculithus tympaniformis Hay and Mohler; 4—Fasciculithus pileatus Bukry; 5—Fasciculithus janii Perch-Nielsen.
NTp10C
Varol (1989) Aguado (1993)
AGE
NANNOFOSSIL ZONES
C
PI-17
SELANDIAN
NTp10
NTp8
NTp8
B
NTp6 NTp5
10 µm
C B A
NTp4 C
B
NTp3
NTp1
A B A D C B A
NTp5-NTp7
NBK24 NBK23
U. MAASTRICHT.
10 m
C B A
A
NTp2
FP-2 FP-1 PI-M-2 VA-1 PI-M-1
A
NTp9
NTp7
DANIAN
Cazuela del Pozo
Venta de Agramaderos
N. Villalobos
VL-KT
U. CRETACEOUS
Capas Rojas Fm.
VA-12R PI-T VA-11R CZ-52 CZ-51 VA-10R
P A L E O C E N E
MAJALCORÓN Fm.
MJ-T-3 MJ-T-2
Fuente de la Pileta
Capas Rojas Fm.
Pilas de la Fuente del Soto
Peñas de Majalcorón
B
NBK22
B A C
NBK21 B
NTp1B
A
NBK24
196
Molina et al.
Figure 5. (A) Calcarenite bed with mound shape at the top. (B) Calcarenite beds with horizontal parallel lamination and wavy tops. (C) Typical mound morphologies of one calcarenite bed in the holostratotype, with flat horizontal bottom and oblique top, and wedge shape. On the right, hummocky cross-stratification appears. (D) Level with micritic intraclasts in the holostratotype. (E) Simple vertical (Skolithos) and horizontal dwelling burrows. (F) Water-escape (sand volcano) structure in the calcarenites.
changes of differential cementation in the calcarenites in areas that are bounded by approximately vertical fractures. MICROFACIES The microscopic analysis of more than 100 thin sections shows that most calcarenitic beds are Microcodium grainstones, with quartz grains, benthonic and planktonic foraminifera, bioclasts (mainly echinoid spines and coralline algae), peloids, carbonaceous fragments, glauconite, and small mica grains. The
Microcodium fragments generally comprise between 40% and 90% of the rock. They usually are disarticulated and appear as individual prisms with maximum and minimum length from 0.3 to 0.5 mm and width from 0.03 to 0.05 mm (Fig. 6A). Some prisms are organized in aggregates with more-or-less complete rosette shapes like Microcodium (a) of Esteban (1972) or types 1 and 2 of Plaziat (1984). Most aggregates clearly resemble Microcodium type 2 of Plaziat (1984), formed by prisms bunched together around one side of the central channel or axis of growth (Figs. 6B and 6C).
Reworked Microcodium calcarenites
197
Figure 6. (A) Typical aspect of the calcarenitic microfacies with disaggregated prismatic crystals of Microcodium. (B) Calcarenitic microfacies, in the central lower part with a “rosette” aggregate of Microcodium. (C) Prismatic fragments of Microcodium, forming aggregates in some places (crossed nicols). (D) Microfacies of the micritic beds and mud clasts: wackestone-packstone with planktonic foraminifera.
The quartz grains form between the 5% and 15% of the rock. They are very angular, have a maximum diameter of 700 μm, and average between 100 and 200 μm. The most abundant and characteristic bioclasts are of corallinacean algae, up to 1.3 mm long, and echinoid spines, which are locally silicified. Coal fragments and plant remains are more than 2.5 mm long, and glauconite grains are from 100 to 400 μm in size. Discontinuous centimeter-scale beds interbedded with calcarenite are composed of wackestone and packstone with small planktonic foraminifera (Globigerina, Globorotalia), bioclasts, quartz grains, parallel lamination, abundant bioturbation, and silicified radiolarians up to 250 μm in diameter (Fig. 6D). The cements are syntaxial and poikilotopic. Significant sparry cement is associated with small veins of diagenetic dissolution. OTHER MICROCODIUM CALCARENITES IN THE BETIC CORDILLERA AND DIFFERENT ALPINE MEDITERRANEAN DOMAINS The fundamental characteristic of the Majalcorón Formation is the great abundance of Microcodium fragments, which are generally disaggregated and form isolated calcite prisms,
but locally are more-or-less complete aggregates. Microcodium calcarenites are also present in the Paleogene (mainly in the Paleocene and early Eocene) of other geological units of the Betic Cordillera (Vera, 2000, 2004) and in other Mediterranean-Alpine domains (Klappa, 1978; Kosir, 2004). Microcodium Calcarenites in the Betic Cordillera Microcodium calcarenites have been described in the Prebetic, Subbetic, Frontal units, Malaguide Complex, and Campo de Gibraltar units. In the Prebetic of northernmost Murcia and Albacete provinces, the Paleocene to early Eocene begins with white limestones with algae and Microcodium. In the Prebetic of Alicante province, the presence of Microcodium of Oligocene age has been determined. The existence of Microcodium limestones has been cited in the Montecorto unit and in the Corredor del Boyar units of the western Subbetic (Martín-Algarra, 1987). The Olivares Formation (Comas, 1978) of the Paleocene Subbetic, present mainly in the sector of the Fardes River (province of Granada), is composed of gray calcarenites with Microcodium. De Smet (1984) indicated the presence of Microcodium turbidites of Paleocene age with abundant Zoophycos in the eastern Subbetic.
198
Molina et al.
In the external Frontal units of Argüelles type (Martín-Algarra et al., 2004), one of the most characteristic formations in the Tertiary is formed by turbiditic limestones with Microcodium of Paleocene age (Martín-Algarra, 1987). These calcarenites are interpreted as a lateral facies change of the Capas Rojas Formation. The Paleocene of the Sierra of Espuña (Maláguide Complex) is represented by the 30-m-thick “Microcoditas of Mula Formation” (Martín-Martín, 1996), constituted by calcarenites rich in Microcodium alternating with beds of bluish sandy marls and of calcareous conglomerates (Martín-Martín et al., 1997, 1998). In the vicinity of Málaga, Martín-Algarra (1987) recognized Microcodium in place on Cretaceous limestones; this was interpreted as an edaphic alteration formed during the Paleocene. In the same region, Serrano et al. (1995) identified Microcodium facies that they attributed to the Eocene. In the Algeciras units (Mauritanian of the Campo de Gibraltar), the oldest Tertiary formation is the “Limestones with Microcodium Formation,” which outcrops very locally and is less than 50 m in thickness (Martín-Algarra, 1987). This formation, attributed to the Paleocene, is composed of decimeter-scale beds of turbiditic calcarenites consisting almost totally of remains of Microcodium, separated by marly strata. In the Aljibe units (Numidian of the Campo de Gibraltar), the lower part of the Paleocene is also made up of turbiditic calcarenites with Microcodium fragments (Esteras et al., 2004). Microcodium Calcarenites in Other Alpine-Mediterranean Domains There are abundant examples in the literature of calcarenites and reworked Microcodium in the peri-Tethyan realm. Here we consider only some of the most important references. In the Spanish Pyrenees, Arribas et al. (1996) and Rossi (1997) studied calcarenites with Microcodium in Paleocene lacustrine and fluvial facies. Freytet and Plaziat (1982) described continental Microcodium debris in the Paleogene of the French Pyrenees–Provençal Basin. In the French-Italian Maritime Alps, the formation with Microcodium (Faure-Mauret and Fallot, 1954; Varrone and Clari, 2003), consisting mainly of fluvial facies, has an early-middle Eocene age. In Sardinia, Mateucci and Murru (2002) studied lacustrine Microcodium calcarenites of late Thanetian–early Ypresian age. Kosir (2004) described abundant disaggregated Microcodium in calcrete profiles interbedded in Thanetian shallow-marine limestones of the Adriatic-Dinaric carbonate platform (Slovenia). PALEOGEOGRAPHICAL SIGNIFICANCE From its lithology and sedimentology (Molina et al., 2003), the Majalcorón Formation is interpreted as deposits on a distal carbonate ramp in which storm waves were the principal mechanism responsible for resedimentation. The presence of disaggregated Microcodium as the main component in these shallow-marine deposits resulted from the erosion and reworking of
a subaerially exposed surface of calcareous paleosols during a marine transgression. In the Late Cretaceous, principally in the Maastrichtian, there were high sea-level conditions, during which pelagic marly limestones of the Capas Rojas Formation were deposited, sometimes including carbonate turbidites, in wide areas of the Subbetic (Fig. 7A). Microcodium was abundantly produced in the early Paleocene, possibly as a consequence of important climatic and ecologic changes that happened during the Cretaceous-Tertiary transition. According to some authors (e.g., Hallam, 1998), an abrupt global sea-level fall of more than 150 m took place at this time; this would have caused the emergence of wide coastal areas. Based on paleogeographical reconstructions (Smith et al., 1994), the global increase in land area against marine domains was considerable, from 109 × 106 km2 in the Maastrichtian (70 Ma) to 138 × 106 km2 in the Paleocene (60 Ma). In these emergent areas, abundant paleosols with Microcodium were developed and subsequently were eroded and redeposited in marine environments during transgression, thus producing Microcodium calcarenites (Fig. 7B). According to these reconstructions (Smith et al., 1994; Cavazza et al., 2004), the proposed early Paleocene paleolatitude in the area corresponding to the Subbetic (southern Iberian continental margin) was between 20 and 25°N, near the Tropic of Cancer, in which arid or semiarid climatic conditions would be predominant. Paleoclimatic reconstructions for the Paleocene (e.g., Bolle et al., 2000; Zachos et al., 2001; Adatte et al., 2002) indicate that a warm period of maximum humidity with high rainfall characterized the Cretaceous-Tertiary boundary. Subsequently, in the Danian, arid climatic conditions evolved that persisted during the Selandian and Thanetian and reached a maximum during the latest Paleocene (late Paleocene thermal maximum). In this sense, the occurrence of a carbonate and a relatively abundant quartz fraction in the Majalcorón Formation might indicate a subaerially exposed environment with a sparse vegetation cover, possibly similar to modern scrubland vegetation in a semiarid Mediterranean climate. An interesting paleogeographical aspect is that the potential locations where the abundant calcareous paleosols were formed could not have been very far away from the site of deposition, thus allowing the preservation of Microcodium aggregates. In the Subbetic, outcrops of Paleocene paleosols have not been found; the regions of the Subbetic that lack Upper Cretaceous rocks (Capas Rojas Formation), or where their possible erosion during the Paleogene has been detected, are potential source areas for the studied Microcodium calcarenites of the Majalcorón Formation. Hypothetically, favorable areas would be those located in the Subbetic where the Capas Rojas Formation does not appear and where the development of erosive and paleokarst features on the Jurassic deposits is evident (Vera et al., 1988; Molina et al., 1999). We refer specifically to the central sector of the Betic Cordillera in the units of the northern external Subbetic (CamarenaLanchares and Grajales-Mentidero units) and parts of the southern external Subbetic (Lobatejo-Pollos unit); in these areas, there
Reworked Microcodium calcarenites
199
A LATEST CRETACEOUS
B DANIAN
Karst
Sea level
Paleosols with Microcodium Sea level
1 S.l. Swbl Majalcorón Fm
2
Capas Rojas Fm
S.l. Swbl
3 S.l. Swbl
Figure 7. Sketch showing the genesis of the Majalcorón Formation before (A) and after (B) the important sea-level fall at the Cretaceous-Tertiary boundary. Panels 1–3 show sea-level changes explaining the production of mud clasts and the reworking of Cretaceous nannoplankton and foraminifera (see explanation in the text). S.l.—sea level; Swbl—base level of the storm waves.
is little or no representation of deposits with ages between Late Cretaceous and early Oligocene (Molina, 1987; Molina and Nieto, 2003). The erosional boundary at the top of the Majalcorón Formation (Molina et al., 2003) could represent another important stage of sea-level fall that took place in the early Selandian and produced erosion and karstification, at least in some areas, of the
formation top. Comparison with records from the North Sea Basin, Western Pyrenees, the Nile Basin, and the eastern continental margin of North America suggests that sea-level changes across the Danian-Selandian boundary were primarily caused by eustatic fluctuations with an associated relative sea-level drop on the order of 50–100 m (e.g., Pujalte et al., 1998; Clemmensen and Thomsen, 2005). The general trend of sea level in the early
200
Molina et al.
Paleocene, with a regressive-transgressive-regressive pattern clearly controlling the beginning and the ending of the Majalcorón Formation deposition, is shown in the Figure 8. Other characteristic aspects of the Majalcorón Formation are the presence of soft micritic intraclasts, and of reworked foraminifera and nannoflora including, principally, Upper Cretaceous (Campanian and Maastrichtian), but also some Lower Cretaceous taxa. These can be explained by small oscillations in sea level and consequently in the storm-wave base during the deposition of the Majalcorón Formation; this is in agreement with the model presented in the Figures 7 and 8: (1) In a first stage, the Microcodium calcarenites were deposited above the storm-wave base (Swbl in the Fig. 7). (2) A rise in base level and decrease of wave energy favored the deposition of micritic facies containing planktonic foraminifera. (3) During the third stage, related to a fall of storm-wave base, the reworking and resedimentation of these micritic sediments took place, forming micritic intraclasts that mixed with the calcarenitic sediment. In relation to the stratigraphic units of the Late Cretaceous and Paleogene, the Majalcorón Formation appears to be a unit consisting of reworked sediments (Microcodium) of clear continental origin, deposited on a shallow-marine ramp inserted among pelagic facies (Capas Rojas Formation). CONCLUSIONS Lithologically, the Majalcorón Formation is characterized mainly by its abundant content of Microcodium. This formation, located between pelagic sedimentary rocks (Capas Rojas Formation), was deposited during the Lower Paleocene close to emergent areas covered with abundant calcareous paleosols with Microcodium. After, these paleosols were eroded and reworked, the Microcodium calcifications were disaggregated and their prismatic crystal debris resedimented in shallow-marine environments affected by storm waves. According to nannofossil
61.7 Ma
PALEOCENE
CAPAS ROJAS Fm. SELANDIAN
The authors express their gratitude to Agustín MartínAlgarra for the very precise, complete, and constructive revision of this article, and to Dario Varrone for many helpful suggestions and useful criticism of the paper. We thank Ana M.
Fall
SUBAERIAL EXPOSURE Paleokarst
(Shallow marine platform facies)
65.5 Ma CAPAS ROJAS Fm. MAASTRICHTIAN
ACKNOWLEDGMENTS
and other pelagic and turbiditic facies
MAJALCORÓN Fm. DANIAN
Rise
biostratigraphy, the deposition of the Majalcorón Formation started in the earliest Danian, coinciding with (or immediately after) subzone NTp1B of Varol (1989). The end of the deposition of the sediments of the Majalcorón Formation was heterochronic and occurred between the latest Danian–early Selandian (as maximum subzone NTp8A) and middle Selandian, at the top of subzone NTp8C. Eustatic sea-level changes from the Cretaceous-Tertiary boundary through the early Paleocene with a regressive-transgressive-regressive succession controlled the generation of the paleosols with Microcodium, the sedimentation of the calcarenites after erosion and reworking paleosols, and the ending of deposition with the development of paleokarstic features on the top of the formation. The Majalcorón Formation has great significance for understanding the paleogeographical evolution of the pelagic realms of the Betic external zones during the Late Cretaceous and the Tertiary. Its peculiar character is originally related to its deposition close to subaerial exposure areas with abundant calcareous paleosols that were redeposited on shallow carbonate ramps. These paleosols with abundant Microcodium must have had extensive development in the Subbetic, and probably in the External Subbetic, at the time of the Cretaceous-Tertiary boundary. The presence of the Majalcorón Formation and its interpretation are very important to understanding the paleogeography and the depositional paleobathymetry of the pelagic facies in the adjacent formations along the southern Iberian continental margin during the Late Cretaceous and the Paleogene.
Small sea-level oscillations
Reworking and redeposition of Microcodium
SUBAERIAL EXPOSURE Calcareous soils with abundant Microcodium
(Pelagic and hemipelagic facies)
Figure 8. General evolution in a regressive-transgressive-regressive pattern of early Paleocene sea level mainly in relation to the beginning and ending of the Majalcorón Formation deposition.
Reworked Microcodium calcarenites Alonso-Zarza and L.H. Tanner for the continual aid during all the stages of the manuscript preparation and editorial work. Also we extend our appreciation to Elizabeth A. Adams, who made the English text more readable. This work has been carried out as a contribution to the groups of investigation RNM200, RNM-208, 4064 (Junta of Andalusia) and of Investigation Projects BTE2000-1151 and BTE2001-2852 of the Spanish Ministry of Science and Technology. REFERENCES CITED Adatte, T., Keller, G., and Stinnesbeck, W., 2002, Late Cretaceous to early Paleocene climate and sea-level fluctuations: The Tunisian record: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 178, p. 165–196, doi: 10.1016/S0031-0182(01)00395-9. Aguado, R., 1993, Nannofósiles del Cretácico de la Cordillera Bética: Bioestratigrafía [Ph.D. thesis]: Granada, University of Granada, 413 p. Aguado, R., Molina, J.M., and Vera, J.A., 2003, La Formación Majalcorón (calcarenitas con Microcodium, Paleoceno, Subbético): Bioestratigrafía: Geotemas, v. 5, p. 13–17. Alonso-Zarza, A.M., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: Earth-Science Reviews, v. 60, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Arribas, M.E., Estrada, E., Obrador, A., and Rampone, G., 1996, Distribución y ordenación de Microcodium en la Formación Tremp: Anticlinal de Campllong (Pirineos Orientales, provincial de Barcelona): Revista de la Sociedad Geológica de España, v. 9, p. 9–18. Bolle, M.P., Pardo, A., Adatte, T., Von Salis, K., and Burns, S., 2000, Climatic evolution on the southeastern margin of the Tethys (Negev, Israel) from the Palaeocene to the early Eocene: Focus on the late Palaeocene thermal maximum: Journal of the Geological Society of London, v. 157, p. 929–941. Cavazza, W., Roure, F.M., Spakman, W., Stampfli, G.M., and Ziegler, P.A., eds., 2004, The TRANSMED Atlas: The Mediterranean Region from Crust to Mantle: Berlin, Springer, 141 p., 1 CD-ROM. Clemmensen, A., and Thomsen, E., 2005, Palaeoenvironmental changes across the Danian-Selandian boundary in the North Sea Basin: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 219, p. 351–394, doi: 10.1016/ j.palaeo.2005.01.005. Comas, M.C., 1978, Sobre la geología de los Montes orientales: Sedimentación y evolución paleogeográfica desde el Jurásico hasta el Mioceno inferior (zona Subbética, Andalucía) [Ph.D. thesis]: Bilbao, University of País Vasco, 323 p. De Smet, M.E.M., 1984, Investigations of the Crevillente fault zone and its role in the tectogenesis of the Betic Cordilleras, Southern Spain [Ph.D. thesis]: Amsterdam, University of Amsterdam, 174 p. Esteban, M., 1972, Una nueva forma de prismas de Microcodium elegans Glück 1972 y su relación con el caliche del Eoceno Inferior, Marmellá, provincia de Tarragona (España): Revista del Instituto de Investigaciones Geológicas, University of Barcelona, v. 27, p. 65–81. Esteras, M., Martín-Algarra, A., and Martín-Martín, M., 2004, Complejo del Campo de Gibraltar: Estratigrafía, in Vera, J.A., ed., Geología de España: Madrid, SGE-IGME, p. 391–393. Faure-Mauret, A., and Fallot, P., 1954, Sur le Secondaire et le Tertiaire aux abords sud-orientaux du massif de l’Argentera-Mercantour: Bulletin du Service de la Carte Géologique de France, v. 241, p. 189–198. Freytet, P., and Plaziat, J.C., 1982, Continental carbonate sedimentation and pedogenesis—Late Cretaceous and early Tertiary of southern France: Contributions to Sedimentology, v. 12, 213 p. Hallam, A., 1998, Interpreting sea level, in Doyle, P., and Bennett, M.R., eds., Unlocking the Stratigraphical Record: Advances in Modern Stratigraphy: Chichester, UK, John Wiley and Sons, p. 420–439. Klappa, C.F., 1978, Biolithogenesis of Microcodium: Elucidation: Sedimentology, v. 25, p. 489–522. Klappa, C.F., 1980, Rhizoliths in terrestrial carbonates: Classification, recognition and significance: Sedimentology, v. 27, p. 613–629. Kosir, A., 2004, Microcodium revisited: Root calcification products of terrestrial plants on carbonate-rich substrates: Journal of Sedimentary Research, v. 74, p. 845–857.
201
Martín-Algarra, A., 1987, Evolución geológica alpina del contacto entre las Zonas Internas y las Zonas Externas de la Cordillera Bética (sector central y occidental) [Ph.D. thesis]: Granada, University of Granada, 1271 p. Martín-Algarra, A., O’Dogherty, L., and López-Garrido, A.C., 2004, Unidades Frontales de las Zonas Internas: Estratigrafía, in Vera, J.A., ed., Geología de España: Madrid, SGE-IGME, p. 398–399. Martín-Martín, M., 1996, El Terciario del dominio Maláguide en Sierra Espuña (Cordillera Bética oriental, SE de España): Estratigrafía y evolución paleogeográfica [Ph.D. thesis]: Granada, University of Granada, 297 p. Martín-Martín, M., El Mamoune, B., Martín-Pérez, J.A., Serra-Kiel, J., and Martín-Algarra, A., 1997, Timing of deformation in the Malaguide of the Sierra Espuña (southeastern Spain): Geodynamic evolution of the Betic Internal Zone: Geologie en Mijnbouw, v. 75, p. 309–316. Martín-Martín, M., Serra-Kiel, J., El Mamoune, B., Martín-Algarra, A., and Serrano, F., 1998, Le Paléocène des Malaguides orientales (Cordillères Bétiques, Espagne): Stratigraphie et paléogéographie: Comptes Rendus de l’Académie des Sciences de Paris, Sciences de la Terre et des Planètes, v. 326, p. 35–41. Martínez-Gallego, J., and Roca, A., 1973, Estudio del Danés de la cuenca nummulítica de Montefrío-Alcalá la Real: Correlación con el de Alamedilla (Zona Subbética): Cuadernos de Geología, University of Granada, v. 4, p. 93–97. Mateucci, R., and Murru, M., 2002, Early Tertiary Microcodium from Sardinia, Italy: Bollettino della Societa Geologica Italiana, v. 121, p. 289–296. Molina, J.M., 1987, Análisis de facies del Mesozoico en el Subbético Externo (provincia de Córdoba y Sur de Jaén) [Ph.D. thesis]: Granada, University of Granada, 518 p. Molina, J.M., and Nieto, L.M., 2003, Calcarenitas y calizas del Oligoceno superior-Mioceno inferior discordantes sobre el Mesozoico en el Subbético al S de Jaén: Geotemas, v. 5, p. 171–174. Molina, J.M., and Vera, J.A., 2001, Cantos blandos en tempestitas con estructuras sedimentarias de deformación (Mioceno, Cuenca del Guadalquivir; Porcuna, Provincia de Jaén): Geotemas, v. 3, no. 1, p. 223–226. Molina, J.M., Ruiz-Ortiz, P.A., and Vera, J.A., 1999, A review of polyphase karstification in extensional tectonic regimes: Jurassic and Cretaceous examples, Betic Cordillera, southern Spain: Sedimentary Geology, v. 129, p. 71–84, doi: 10.1016/S0037-0738(99)00089-5. Molina, J.M., Vera, J.A., and Aguado, R., 2003, La Formación Majalcorón (calcarenitas con Microcodium, Paleoceno, Subbético): Definición y descripción: Geotemas, v. 5, p. 175–179. Plaziat, J.C., 1984, Le problème des Microcodium: Une mise au point, in Le Domain Pyrénéen de la Fin du Crétacé à la Fin de l’Eocène: Stratigraphie, Paléoenvironnements et Évolution Paléogeographique [Ph.D. thesis]: Paris, University of Paris-Sud II, p. 637–662. Pujalte, V., Baceta, J.I., Orue-Etxebarria, X., and Payros, A., 1998, Paleocene strata of the Basque country, western Pyrenees, northern Spain: Facies and sequence development in a deep-water starved basin, in De Graciansky, P.C., Hardenbol, J., Jacquin, T., and Vail, P.R., eds., Mesozoic and Cenozoic Sequence Stratigraphy of European Basins: Society for Sedimentary Geology (SEPM) Special Publication 60, p. 311–325. Rossi, C., 1997, Microcodium y trazas fósiles de invertebrados en facies continentales (Paleoceno de la cuenca de Áger, Lérida): Revista de la Sociedad Geológica de España, v. 10, p. 371–391. Serrano, F., Sanz de Galdeano, C., Delgado, F., López-Garrido, A.C., and Martín-Algarra, A., 1995, The Mesozoic and Cenozoic of the Malaguide Complex in the Málaga area: A Paleogene olistostrome-type chaotic complex (Betic Cordillera, Spain): Geologie en Mijnbouw, v. 74, p. 105–116. Smith, A.G., Smith, D.G., and Funnell, B.M., 1994, Atlas of Mesozoic and Cenozoic Coastlines: Cambridge, Cambridge University Press, 99 p. Varol, O., 1989, Palaeocene calcareous nannofossil biostratigraphy, in Crux, J.A., and Van-Heck, S.E., eds., Nannofossils and their applications: Chichester, UK, Ellis Horwood Ltd., p. 267–310. Varrone, D., and Clari, P., 2003, Évolution stratigraphique et paléo-environnementale de la formation à Microcodium et des calcaires à nummulites dans les Alpes Maritimes Franco-Italiennes: Geobios, v. 36, p. 775– 786, doi: 10.1016/j.geobios.2003.09.001. Vera, J.A., 2000, El Terciario de la Cordillera Bética: Estado actual de conocimientos: Revista Sociedad Geológica de España, v. 13, p. 345–373. Vera, J.A., ed., 2004, Geología de España: Madrid, Sociedad Geológica de España-Instituto Geológico y Minero de España (SGE-IGME), 884 p. Vera, J.A., García-Hernández, M., López-Garrido, A.C., Comas, M.C., RuizOrtiz, P.A., and Martín-Algarra, A., 1982, El Cretácico de la Cordillera
202
Molina et al.
Bética, in García, A., ed., El Cretácico de España: Madrid, Editorial Complutense, p. 515–632. Vera, J.A., Ruiz-Ortiz, P.A., García-Hernández, M., and Molina, J.M., 1988, Palaeokarst and related sediments in the Jurassic of the Subbetic zone, southern Spain, in James, N.P., and Choquette, P.W., eds., Paleokarst: New York, Springer-Verlag, p. 364–384. Vera, J.A., Molina, J.M., and Aguado, R., 2003, La Formación Majalcorón (calcarenitas con Microcodium, Paleoceno, Subbético): Situación e interpretación en el Terciario de la Cordillera Bética: Geotemas, v. 5, p. 243–246.
Wright, V.P., and Tucker, M.E., 1991, Calcretes: An introduction: Oxford, International Association of Sedimentologists, Reprint Series, v. 2, p. 1–22. Zachos, J., Pagani, M., Sloan, L., Thomas, E., and Billups, K., 2001, Trends, rhythms, and aberrations in global climate 65 Ma to present: Science, v. 292, p. 686–693, doi: 10.1126/science.1059412.
MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
Calcite cement stratigraphy of a nonpedogenic calcrete in the Triassic New Haven Arkose (Newark Supergroup) E. Troy Rasbury Department of Geosciences, State University of New York (SUNY) Stony Brook, New York 11794-2100, USA Elizabeth H. Gierlowski-Kordesch Department of Geological Sciences, Ohio University, Athens, Ohio 45701-2979, USA Jennifer M. Cole† Interdepartmental Doctoral Program in Anthropological Sciences, State University of New York (SUNY) Stony Brook, New York 11794-4364, USA Cherri Sookdeo Glenn Spataro Jessica Nienstedt Department of Geosciences, State University of New York (SUNY) Stony Brook, New York 11794-2100, USA ABSTRACT Nonpedogenic calcrete is difficult to distinguish from pedogenic calcrete in the fossil record; both alpha and beta textures have been observed from fossil and modern examples. However, a calcrete from the New Haven Arkose (Hartford Basin, Connecticut) is shown here to be of a nonpedogenic origin through sedimentologic and petrographic evidence. An accumulation of thin sheets of displacive calcite layers found in a decimeter-thick horizon of anastomosing veins within the upper portion of a red mudstone is correlated to calcite cement found in the overlying sandstone. Based on petrography, we recognize six generations of calcite in the mudstone-sandstone hosts. The first five generations are associated with rhizoliths that can be related to deep taproots and are interpreted to have formed by precipitation from shallow groundwater. There are no vadose-type cement morphologies; the calcite has luminescent zones, indicating that Mn was soluble and thus oxygen levels were low. These cements clearly formed several meters below what would have been the surface of the channel sand body. We suggest that calcite cement stratigraphy combined with redox models for the behavior of Mn (as well as Fe and U) may aid in the identification of nonpedogenic versus pedogenic carbonates in the geologic record. Additionally, the calcite from this carbonate layer has been dated using the U-Pb method. Our results provide insight into the environmental and diagenetic fluid conditions favorable for providing a spread in U/Pb ratios that are suitable for precise dating of calcites in otherwise undateable sections. Keywords: groundwater calcrete, alveolar septal fabric, taproots, cement stratigraphy.
Present address: Lamont-Doherty Earth Observatory of Columbia University, 61 Route 9W, Palisades, New York 10964-8000, USA
†
Rasbury, E.T., Gierlowski-Kordesch, E.H., Cole, J.M., Sookdeo, C., Spataro, G., and Nienstedt, J., 2006, Calcite cement stratigraphy of a nonpedogenic calcrete in the Triassic New Haven Arkose (Newark Supergroup), in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 203–221 doi: 10.1130/2006.2416(13). For permission to copy, contact editing@ geosociety.org. ©2006 Geological Society of America. All rights reserved.
203
204
Rasbury et al. RESUMEN Las calcretas no-pedogénicas son difíciles de diferenciar de las pedogénicas en el registro antiguo; los dos tipos de textura alfa y beta se reconocen en depósitos antiguos y recientes. Sin embargo, en el caso concreto de la calcreta de New Haven Arkose (Cuenca Hartford, Connecticut), los datos sedimentológicos y las evidencias petrográficas permiten demostrar su origen no-pedogénico. La presencia, dentro de lutitas rojas, de finas capas de calcita desplazativa dentro de un horizonte decimétrico de venas anastomosadas se puede correlacionar con la formación de cemento calcítico en las areniscas infrayacentes a dichas lutitas rojas. Petrográficametne se reconocen seis generaciones de calcita en las lutitas y areniscas que constituyen el sustrato. Las cinco primeras generaciones están asociadas con rizolitos de sistemas radiculares profundos y se pudieron formar por precipitación a partir de aguas freáticas someras. No hay cementos vadosos y la calcita tiene zonas luminiscentes indicando que el Mn era soluble y, por tanto, los niveles de oxígeno bajos. Estos cementos se formaron claramente varios metros por debajo de lo que fue la superficie del canal de arenas. Sugerimos que la estratigrafía de los cementos de calcita combinada con los modelos redox de comportamiento de Mn (también Fe y U) pueden ayudar en la identificación de los carbonatos pedogénicos y no-pedogénicos en el registro geológico. Además la calcita de esta capa carbonática se ha datado por el método de U-Pb. Nuestros resultados permiten una mejor caracterización de las condiciones ambientales y de los tipos de fluidos más favorables para ampliar el rango de valores U/Pb necesarios para datar calcitas en secciones que de otro modo no se podrían datar. Palabras clave: calcretas freáticas, fábricas alveolares, raíces, estratigrafía de cementos.
INTRODUCTION Calcite cement is one of the most important products of diagenesis in terrestrial sedimentary deposits. Based on assumptions about its relationship to the atmosphere, calcite that forms in soil profiles may archive pCO2 records (Cerling, 1984; Cerling and Quade, 1993; Mermut et al., 2000), where pedogenic calcite is attributed to dry conditions, at least seasonally (Semenuik and Searle, 1985; Harden et al., 1991; Tandon and Kumar, 1999; Lal and Kimble, 2000). Pedogenic calcrete contains both alpha and beta fabrics, and the extent of each type is mostly dependent on climate and hydrologic setting (Wright, 1990; Wright and Tucker, 1991; Tandon and Kumar, 1999). Alpha textures include rhombic calcite crystals, dense micritic fabric, floating sediment grains, complex cracks and crystallaria, and displacive growth features (cf. Watts, 1978; Rossinsky et al., 1992), and exhibit no preserved biogenic features. Beta textures contain features indicating biogenic influence, such as alveolar septal fabric, calcified tubules, microbial coatings, and Microcodium. Groundwater calcretes are nonpedogenic in origin, form in the phreatic groundwater zone, and are typically linear and tabular limestone bodies containing mostly alpha fabrics. These are differentiated from capillary fringe, nonpedogenic calcretes, which form in the vadose zone above the water table and contain beta fabrics that result from phreatophytic plants extending their roots down toward the water table (Carlisle, 1983; Purvis and Wright, 1991; Wright and Tucker, 1991; Williams and Krause, 1998; Tandon and Kumar, 1999; Nash and Smith, 1998; Khadki-
kar et al., 1998; Tandon and Andrews, 2001). Groundwater and capillary fringe nonpedogenic calcite precipitate in association with Ca-rich groundwaters in response to the common ion effect, degassing, and evapotranspiration through root activity with associated microbes (bacteria, fungi, and cyanobacteria) (Wright and Tucker, 1991; Chadwick and Graham, 2000). Differentiation between pedogenic and nonpedogenic calcrete is not possible through stable isotopic analysis because the controls on calcite precipitation are similar (Purvis and Wright, 1991; Quade and Roe, 1999; Mermut et al., 2000; Mack et al., 2000; Tandon and Andrews, 2001). Calcrete origin is assessed through field relationships (Nash, 1997), such as a position immediately above a relatively impermeable layer or bedrock as well as the presence/absence of alpha and beta fabrics. This still is an inexact method (Wright and Tucker, 1991; Nash and Smith, 1998; Tandon and Kumar, 1999). The juxtaposition of calcite cements with other diagenetic features in a calcrete horizon across a mudstone-sandstone boundary is presented here as a valuable technique in the establishment of relative timing of diagenesis in order to assess origin. In addition, geochemical methods such as the evaluation of redox-sensitive elements in petrographic work can clarify the position of carbonate accumulation with respect to the water table. Geochronologic dating of calcite cements through U-Pb techniques can separate early and late diagenetic cementation as well as provide an age for sedimentation (Rasbury et al., 1997, 1998, 2000; Wang et al., 1998). Because calcite has a low distribution coefficient for uranium (Chung and Swart, 1990; Reeder et al., 2000), it is advantageous to understand the
Nonpedogenic calcrete in the New Haven Arkose conditions that promote elevated U/Pb and U/Th ratios in carbonate sedimentary and diagenetic processes in order to predict which kinds of calcrete might be amenable to dating. Here we present detailed field and petrographic analyses of cements from one of several similar carbonate intervals within the Upper Triassic New Haven Arkose of the Newark Supergroup in the Hartford Basin (Connecticut). These subhorizontal calcite layers in red mudstone have been interpreted as pedogenic calcretes in meandering fluvial facies that formed in a semiarid climate (McInerney and Hubert, 2003), after the sedimentary model of Hubert (1977, 1978). Other sedimentologic work on the upper New Haven mudrocks and sandstones instead suggests a wetter seasonal (perhaps monsoonal) climate during sedimentation of a medium-energy braided floodplain or a high-energy unconfined, vertical accretion to cut-and-fill floodplain (after Krynine, 1950; Nanson and Croke, 1992; Gierlowski-Kordesch and Gibling, 2002). A detailed examination of one of the calcrete layers present along a sandstone-mudstone contact in the New Haven Arkose, using evidence from sedimentologic relationships, cement stratigraphy (after Meyers, 1974), and redox-sensitive elements, is coupled with published U-Pb dating (Wang et al., 1998). This study establishes the genesis and timing of the accumulation of carbonate in an attempt to more easily distinguish between pedogenic and nonpedogenic calcrete in the geologic record.
205
GEOLOGIC SETTING The Hartford Basin in Connecticut (Fig. 1) is one of the rift basins of the Newark Supergroup, which line the eastern margin of North America (Lorenz, 1987; Olsen et al. 1989; Olsen, 1997). These basins formed in response to crustal extension and rifting of Pangea during the early Mesozoic (Late Triassic to Early Jurassic) (Manspeizer, 1988, 1994; Schlische, 1993, 2003; Olsen, 1997). The sedimentary basin fill of the Hartford Basin is composed of more than 4000 m distributed among four continental sedimentary formations and three basaltic units (GierlowskiKordesch and Huber, 1995). The New Haven Arkose is part of the Chatham Group (Weems and Olsen, 1997; De Wet et al., 2002), the lowermost Hartford Basin fill, and is postulated to have a maximum thickness of 2250–2400 m or more (Olsen et al., 1989; GierlowskiKordesch and Huber, 1995). The formation is lower NorianRhaetian (Upper Triassic) to basal Hettangian (Lower Jurassic) in age (Fig. 1) and is interpreted as a fluvial deposit (Gierlowski-Kordesch and Gibling, 2002; McInerney and Hubert, 2003). The Triassic-Jurassic boundary is placed biostratigraphically a few meters below its upper contact with the Talcott Basalt (Cornet and Traverse, 1975; Olsen et al., 1982), and the section studied here is located ~1200 m below the Talcott
Figure 1. Geologic map and generalized stratigraphy of the southern Hartford Basin in Connecticut showing the location of the studied outcrop, southwest of the Hanging Hills near Meriden, Connecticut. The Upper Triassic–Lower Jurassic boundary is presently drawn a few meters below the New Haven–Talcott contact.
206
Rasbury et al.
Basalt. Age-equivalent basaltic sills and flows from other genetically related basins give ages of around 200–201 Ma (Dunning and Hodych, 1990; Hodych and Dunning, 1992; Hames et al., 2000), which are consistent with but not further constraining than a U-Pb age of 212 ± 2 Ma from early diagenetic carbonate (calcrete) in the underlying New Haven Arkose (Wang et al., 1998). A U-Pb age of late diagenetic cements from within this same calcrete is 81 ± 11 Ma (Wang et al., 1998). Thus, there is a record of 130 million years or more of calcite cementation history. The studied calcrete layer of the New Haven Arkose is exposed in a 130-m-thick roadcut along the eastbound side of Interstate Highway 691, just east of the intersection with Interstate Highway 84 near Meriden, Connecticut (Fig. 1). This interval is projected to be ~650 m above the base of the New Haven Arkose (Wang et al., 1998). Here the section contains fluvial deposits that alternate between variable thicknesses of mudrocks and sandstones, most with associated calcite layers. The stacked nature of these deposits represents aggradation during basin subsidence as high-energy streams deposited sand and mud sheets associated with shallow channels (Gierlowski-Kordesch and Gibling, 2002) or channel migration and climatic cycles (McInerney and Hubert, 2003). In this outcrop, many of the upper portions of mudrock units at their contact with overlying sandstones have quasihorizontal calcite vein-like bodies. Our focus is a detailed study of one such interval, interpreted as a calcrete and already dated geochronologically by Wang et al. (1998). PREVIOUS WORK ON NEW HAVEN CALCRETE Wang et al. (1998) describe and present U-Pb ages for the calcite cements forming vein-like bodies in one mudstone unit in the upper New Haven Arkose. The foundation for this work is the fact that calcium carbonate in soils is generally low-Mg calcite, which is stable and resistant to recrystallization. Most importantly, soil carbonates accumulate quickly; between 103 and 105 yr (see Machette, 1985), while the expected resolution of the U-Pb technique in low to intermediate U-Pb systems is on the order of 106 yr. Wang et al. (1998) identify three generations of calcite cement in the New Haven Arkose carbonate layers exposed at Meriden, Connecticut (Fig. 1). The first generation cement is dull-cathodoluminescent micritic calcite occurring as walls in alveolar structures, such as cylindrical linings of rhizoliths, crack fillings, and dense nodules. The second calcite generation is nonluminescent blocky calcite that fills pores of rhizoliths, alveolar features, and cracks. Both of these two cement generations are interpreted as beta textures diagnostic of pedogenic carbonates (Wright and Tucker, 1991). This fabric is used as the best line of evidence supporting a pedogenic origin (Hubert, 1977, 1978; McInerney and Hubert, 2003). A horizontal “sheet crack” containing the first two generations of cement, variably enriched in U (1–14 ppm), gives an age of 212 ± 2 Ma. Carbon and oxygen isotopic values as well as trace-element data support the conclusion that the calcite cements formed through meteoric diagenesis.
The third generation of calcite is brightly luminescent, fills pore spaces within the first two cements, and in some cases, replaces the earlier cement generations. Clearly occurring after the deposition of crack-filling cements, this third generation cement is interpreted as late diagenetic cement based on crosscutting relationships and a U-Pb age of 81 ± 11 Ma. Carbon isotopic values from this late cement mostly overlap with those of the first two calcite generations, but oxygen isotopes are mostly more negative, and trace-element data are distinctly different, with much lower U and Mg and higher Mn concentrations. Thus, carbon and oxygen isotope data of the early generation cements reported by Wang et al. (1998) are consistent with either a nonpedogenic or pedogenic history. Generally, both types of calcrete form relatively soon after sedimentation and are dateable using U-Pb techniques; thus, they provide an age for sedimentation as well as important climatic or biologic events. Additionally, U-series studies on calcite cement in soils demonstrate great potential for constraining landforms and climate change in the Quaternary (Ludwig and Paces, 2002; Sharp et al., 2003; Blisniuk and Sharp, 2003). While some studies suggest that U-series ages may record mixed histories of pedogenic and nonpedogenic calcite formation (Kelly et al., 2000; Candy et al., 2003), this is not likely to have an adverse affect on U-Pb dating, simply because the time resolution for fossil calcretes is much larger. The diagenetic history recorded in the cement stratigraphy can aid in assessing the reliability of radiometric ages for both U-series and U-Pb studies of calcrete layers. The sedimentology and petrography of the entire targeted calcrete layer in the New Haven Arkose, including the cemented sandstone in conformable contact with and above the U-Pb dated calcitic vein-like bodies of the mudstone, are now presented. Petrographic analysis includes microscopic work under transmitted light as well as cathodoluminescence. FIELD AND PETROGRAPHIC OBSERVATIONS At the Meriden section containing the New Haven Arkose, a 130-m-thick succession of red mudrock units (50 cm to 3 m in thickness), overlain by thick, commonly massive arkose units, up to 2.5 m thick, is exposed. The contact between the mudrock units and their overlying sandstones is the locus of significant displacive calcite cementation (cf. Watts, 1978; Rossinsky et al., 1992). The calcite cement is present in the uppermost portion of red mudstone units as anastomosing subhorizontal vein-like bodies that separate mudstone blocks (Fig. 2A) and as displacive micrite that separates sand grains in the overlying sandstone. No clasts of the calcrete are observed as lag deposits in the sandstone above, nor is there any other macroscopic line of evidence that the calcites formed prior to sandstone deposition. The calcite cements are particularly concentrated within the base of the sandstone (Fig. 2B). There is extensive evidence of pedogenesis in the mudstone, including randomly oriented slickensides in the massive mudstones, as well as tubular features, which are inter-
Nonpedogenic calcrete in the New Haven Arkose
207
Figure 2. Field photos from the Meriden section of the Upper Triassic New Haven Arkose. (A) Mudstone facies with horizontal layers of calcite. Scale on left has centimeter scale on the right and inch scale on the left. (B) Mudstone-sandstone contact showing preferential cementation at the base of the sandstone. Mallet hammer for scale is 10 cm wide. (C) Concentric features with a central cavity surrounded by calcite, interpreted as rhizoliths. Lens cap is 6 cm in diameter. (D) Long vertical features within the sandstone. The area near these features is bleached, suggesting that reducing fluids removed Fe. These features are interpreted as taproots. Hammer for scale is 28 cm long.
preted as rhizoliths (after Klappa, 1980). These rhizoliths range in diameter from 6 mm to 2 cm; they branch and curve around each other with orientations from horizontal through vertical. In cross section, some of these rhizoliths contain a circular inner tube filled with calcite-cemented green mudstone, which is interpreted as the location of the original root (Fig. 2C). On the other hand, sandstones are commonly massive to rarely trough crossbedded with no obvious macroscopic pedogenic features outside of rare, though prominent, 5–7-cm-diameter, vertically tapering structures filled with finer-grained material and abundant calcite cement; penetrating the entire thickness of the massive sandstone units (Fig. 2D). These vertical structures are much more green (drab) in color than the surrounding sandstone and are interpreted to have resulted from taproots of large plants or trees that penetrated the sandstone to reach the water table below. Based on these field relationships, rock samples were collected from the dated carbonate interval of Wang et al. (1998) as
well as from the sandstone above. Samples are from: (1) the topmost calcite-bearing interval of the mudstone; (2) the taproots that penetrate the sandstone; (3) the base of the overlying sandstone; and (4) the middle to top portion of the same sandstone channel unit. If the carbonate deposits in the mudstone are indeed pedogenic calcrete, then the overlying sandstone should not contain any of the early calcite cements found in the mudstone. The sandstone channel would then have migrated upon the pedogenically altered mud surface at a later time and contained genetically different calcite cements. If the carbonate unit in the mudstone is a nonpedogenic calcrete associated with the water table, then the lowermost portion of the sandstone in contact with the underlying mudstone should contain genetically related calcite cement because groundwater flow along the sand-mud contact would have been contemporaneous. Hand-specimen-scale examination of the cement fabrics in the mudstone and taproots from the overlying sandstone reveals
208
Rasbury et al.
striking similarities. The red mudstone has a horizontal fabric created by gleyed veins that are commonly filled with calcite (Fig. 3A). The taproot is drab with brecciated rock fragments separated by calcite veins at a similar scale to those in the mudstone (Fig. 3B). Individual grains display a floating texture requiring displacive calcite growth (Watts, 1978). Using
the white and black card techniques (Folk, 1987) at low magnification, one can see that the earliest generation of calcite in the mudstone is micrite (Fig. 3C), while the earliest generation of calcite in the sandstone forms a prismatic layer that is perpendicular to grain surfaces followed by micritic calcite (Fig. 3D).
Figure 3. (A) Scanned image of an entire normal-sized thin section using a slide scanner. The bright white subhorizontal features are filled with calcite. Other areas are lighter than the host red mudstone but do not have visible calcite. Bleaching is interpreted to result from reducing fluids that removed Fe. Dark spots are hematite nodules. (B) Scanned image of an entire thin section using a slide scanner. The clast at the bottom left is shattered by displacive calcite growth. The white band through the middle of the thin section is a calcite vein with a variety of crystal sizes ranging from microspar near the edges to over 350 microns in the center. (C) Incident light image of one of the calcite veins in the mudstone. The white material is micritic calcite that lines roots. Sparry calcite is seen in the center of the vein. (D) Incident light image of one of the veins that breaks up the clast in C. The first recognizable calcite cement at this scale is a bladed calcite, followed by micritic calcite, and then by sparry calcite in the center of the veins.
Nonpedogenic calcrete in the New Haven Arkose Cement Stratigraphy in Mudstone Based on petrography, five generations of calcite cements are recognized in the mudstone facies. Anastomosing microscopic vein-like bodies with alternating micrite and sparry calcite break up the host mudrock into progressively smaller “clods” (Figs. 4A and 4B). We consider these the first two calcite generations (G1, G2, Table 1). These cements are evenly isopachous on grain surfaces and are clearly followed by, but are likely also quasisynchronous with, micrite and bladed calcite (G3, Table 1), which line much larger-scale (visible in hand specimen) veins that cut the host mudstone (Figs. 4C and 4D; cf. 3A and 3B). This fringing calcite is followed by a voidfilling, mostly nonluminescent, blocky calcite (G4, Table 1), with some bright orange luminescent zones (Figs. 4C and 4D). The three generations of calcite cement identified by Wang et al. (1998) occur in rhizoliths, which are extremely prevalent in the top of the mudstone deposit. A dull-luminescent micritic calcite (G3, Table 1) forms the walls of the rhizoliths (Figs. 4E and 4F). A nonluminescent sparry calcite (G4, Table 1) fills the holes presumably left by roots (Figs. 4E and 4F). A brightly luminescent calcite (G6, Table 1) is observable as tiny veins and as a replacement of the earlier calcite generations (Figs. 4E and 4F). Cement Stratigraphy in Taproot Petrographic investigation reveals six generations of calcite cements in the sandstone associated with the taproot structure. Sand grains float in calcite cement (Fig. 5). A very thin layer of brightly orange, luminescent micritic calcite coats some grains and is identified as the first generation of cement (G1, Table 1), although it is not always seen prior to the precipitation of the second generation, bladed calcite (Fig. 5). Bladed calcite is the second-generation cement (G2, Table 1), which grew perpendicular to grain surfaces (Figs. 5 and 6E). Although the bladed calcite appears as a pervasive coating on sand grains, some but not all veins of calcite exhibit this bladed calcite (Figs. 6A and 6B), suggesting that the veins record a progressive history that is synchronous with and also postdates grain-coating cements. The first two generations of calcite appear to be equivalent to the calcite generations that exploded apart the underlying mudstone, based on similarity in crystal size, layer size, and cathodoluminescence. Dull-luminescent micritic calcite is the third generation of cement (G3, Table 1) in the taproot sandstone (Figs. 6A and 6B). The fourth cement generation is nonluminescent blocky calcite (G4, Table 1) that cores calcite crystals within large veins (Figs. 6C and 6D). The fifth cement generation is zoned under cathodoluminescence (G5, Table 1), occurs as overgrowths on G4 calcite cores (Figs. 6C and 6D), and is the sole generation observed in 350–500 micron crystals within the large veins. The fourth and fifth generations of calcite clearly cut horizontal veins defined by the first and second cement generations, but may reflect progression of the same
209
fluid. Veins in which G4 and G5 are recognized are lined by sparry calcite that has the granular mosaic texture described by Purvis and Wright (1991) in capillary fringe nonpedogenic calcretes of the Middle Triassic Otter Sandstone in England. A sixth generation of calcite (G6) is brightly luminescent (Figs. 5E and 5F). Brightly orange, luminescent, fluid inclusion–rich calcite is also present in the center of the largest oscillatoryzoned calcite crystals (G5), which make up the center of a large vein (see Fig. 3C). We also interpret this brightly luminescent cement as the sixth generation (Figs. 6A and 6B). The G6 calcite is seen replacing grains and as fine veins that crosscut all other cement generations (Figs. 6E and 6F). Cement Stratigraphy in Sandstone Based on petrography, we recognize four generations of calcite in the overlying sandstone unit not directly associated with a taproot. Both the basal sandstone and middle to upper sandstone exhibit the same cement generations and approximate proportions (Fig. 7). The first generation is dull-luminescent micrite with floating sand grains, reflecting a dramatic volume increase with displacive calcite growth (Fig. 7). This micrite is interpreted as analogous to the early micrite seen in the mudstone and taproot sandstone (G3, Table 1), because of its similarity in cathodoluminescence, and because cylindrical linings of rhizoliths similar in scale to those in the underlying mudstone are common in the basal sandstone and present throughout the sandstone (Fig. 7). The second generation of calcite observed in the sandstone is nonluminescent and forms very thin layers on the micrite coatings (Figs. 7B and 7D), equivalent to the nonluminescent G4 cement observed in the mudstone and taproots. This nonluminescent calcite is followed by a volumetrically far more important, brightly orange luminescent calcite that fills the remainder of the rhizolith voids (G5). Although this calcite is unzoned, it is likely equivalent to G4 (Table 1) based on its occurrence as pore-filling cement in rhizoliths. Crosscutting rhizoliths show that all the early generations of calcite reflect a time progression of the process of calcification around roots (Figs. 7E and 7F). Although most of the sandstone exhibits the floating grain texture with the displacive calcite, there are millimeterscale bands that are grain-supported, and, within these bands, deformed grains, such as muscovite, provide evidence of compaction that predates calcite cementation (Fig. 8A). In these compacted layers, only one calcite cement type is present, a brightly luminescent calcite that replaces many of the feldspar grains (see Saad, 1991; Hubert et al., 1992; van de Kamp and Leake, 1996). This brightly luminescent sparry calcite is interpreted as the sixth generation cement, which is always brightly luminescent and commonly replaces grains and older cement generations (G6, Table 1). Based on thin sections examination, these compacted bands of sandstone are not common, but our thin section samples may not be an accurate representation of the density of these features within the sandstone.
Late: after burial; 81 ± 11 Ma Brightly luminescent veins and replacement calcite Brightly luminescent veins and replacement calcite Brightly luminescent veins and replacement calcite G6
Brightly luminescent veins and replacement calcite
Prior to burial Brightly luminescent sparry calcite Not seen Not seen G5
Not seen
Prior to burial Nonluminescent sparry calcite Nonluminescent sparry calcite Nonluminescent sparry calcite G4
Nonluminescent sparry calcite
Early: right after deposition; 212 ± 2 Ma Dull-luminescent dense micrite calcite associated with alveolar features Dull-luminescent dense micrite calcite associated with alveolar features Dull-luminescent dense micrite calcite associated with alveolar features G3
Dull-luminescent dense micrite calcite associated with alveolar features
Dull-luminescent microspar
Dull-luminescent bladed calcite
Not seen
Not seen
Early: right after deposition
211
G2
Not seen Not seen Thin, grain-coating brightly luminescent calcite Brightly luminescent micrite calcite
Upper mudstone
Cement generation G1
Figure 4. (A) Plane-light photomicrograph of displacive calcite that brecciates the mudstone. An anastomosing network of micrite appears to have formed around rootlets. There is an intimate association with sparry calcite (G1, G2). (B) Cathodoluminescence photomicrograph of the same area shown in A. The anastomosing network of micrite has dull luminescence and alternates with sparry calcite that is both brightly luminescent and nonluminescent. While crosscutting relationships do not allow us to demonstrate that the micrite was the earliest generation, based on its probable association with roots, micrite may have formed first. (C–D) Large areas in the mudstone facies are composed only of calcite. The bottom half of these photomicrographs has a fringing interlayered micrite with bright luminescence and bladed (vaguely) dull-luminescent calcite (G3). This is followed by largely nonluminescent blocky calcite with zones of bright luminescence (G4). Although not labeled, it appears that the micritic calcite near the boundary with the fringing calcite has been replaced by brightly luminescent calcite (G6). (E) Micrite-lined rhizolith (G3) filled with a blocky calcite (G4). (F) Micrite lining the roots is dully luminescent and filled by a nonluminescent calcite. Both the micrite and sparry calcite may be replaced by brightly luminescent calcite (G6).
TABLE 1. CALCITE CEMENT GENERATIONS IN A MUDSTONE-SANDSTONE PROFILE IN THE NEW HAVEN ARKOSE (NEWARK SUPERGROUP) AT MERIDEN, CONNECTICUT Taproot sandstone Basal sandstone Top sandstone
Calcite generations G1–G3 coexist and may have formed synchronously, although G3 lines rhizoliths that clearly cut the first two generations, which break up clasts in the sandstone and “clods” in the mudstone. Generations G4 and G5 fill rhizoliths and are interpreted to be early cements. These cements (G4, G5) are considered two distinct generations because the mudstone has uniformly nonluminescent calcite cement (G4) forming only the first thin layer of sparry calcite in the sandstone rhizoliths followed by bright and zoned luminescent sparry calcite (G5). A cartoon relating the cement stratigraphy of the New Haven calcrete interval reflects our understanding of the interrelationships among the cements (Fig. 9). It is our interpretation that taproots provide a permeable pathway for meteoric water, and the mudstone layer is an impermeable barrier to the passage of groundwater. Thus, the cements are concentrated at the mudstone-sandstone interface where roots and groundwater meet. Similarities in color and texture of the calcite-cemented sandstone and the calcite deposited in subhorizontal cracks in the underlying mudstone suggest that the same fluids were responsible for precipitating calcite across the mud-sand boundary (Figs. 8, 2A, and 2B). However, all but the very latest cements formed very early based on their association with roots. There are no clay cutans or other features in the massive to trough cross-bedded sandstone that might suggest this was simply a thick soil profile, and the cements show no textures that could be considered of vadose origin. This points toward a groundwater, not a capillary fringe, nonpedogenic origin. Also, no clasts of the mudstone calcretes are incorporated into the bottom of the overlying sandstone. The calcified mud-sand interval is a groundwater calcrete with capillary fringe affinities (after
Age
Summary of Cement Stratigraphy Across MudstoneSandstone Contact
Early: right after deposition
Nonpedogenic calcrete in the New Haven Arkose
212
Rasbury et al.
Figure 5. (A) Plane-light photomicrograph of displacive calcite that separates and fractures sand grains from the taproot area (Fig. 2D). (B) Cathodoluminescence photomicrograph of the same area as A. (C) Plane-light photomicrograph of displacive calcite that separates and fractures sand grains from the taproot area. (D) Cathodoluminescence photomicrographs of the same area as C. Most grains have a thin brightly luminescent layer of calcite (G1) followed by bladed dull-luminescent calcite (G2). Micrite calcite (G3) is almost always associated with this bladed calcite, and here it can be shown to postdate the bladed calcite. Quartz grains may be fractured, but there is no evidence of replacement. Feldspar(?) grains are replaced to variable degrees by brightly luminescent calcite (G6).
Semenuik and Meagher, 1981; Carlisle, 1983; Semenuik and Searle, 1985; Purvis and Wright, 1991; Spötl and Wright, 1992; Slate et al., 1996; Nash and Smith, 1998; Tandon and Kumar, 1999; Khadkikar et al., 2000; Mack et al., 2000; Tandon and Andrews, 2001), resembling the penetrative calcrete of Rossinsky et al. (1992). Reducing groundwater fluids may be indicated because of the leaching of Fe responsible for the drab, green colors associated with the calcite cement (Retallack, 1988, 1991; Wright, 1992; Quade and Roe, 1999). The green, drab coloration (Fig. 3A) is interpreted as the preservation of gleying features (Retallack, 1991; PiPujol and Buurman, 1994). The long vertical fea-
tures in the overlying sandstone, interpreted as taproots, are also drab green (Figs. 9, 2C, and 2D), perhaps due to the local reducing conditions created by decaying organic matter from plant roots. Large taproot holes may have provided a permeable conduit for fluids responsible for the early generations of calcite cementation (Semenuik and Meagher, 1981; Purvis and Wright, 1991; Clothier and Green, 1997). These observations suggest that the groundwater table (which could have been perched) was within a few meters of the surface during the formation of the calcite cements. This would explain the alpha and beta fabrics associated with both groundwater and capillary fringe nonpedogenic calcrete, respectively.
Nonpedogenic calcrete in the New Haven Arkose Features that could be attributed to soils are not seen in the later two early calcite cement generations (G4 and G5). There is little evidence for the timing of the massive micrite cements with floating grains that are pervasive throughout the sandstone away from the taproots with respect to the grain-coating cement generations (G1, G2) found only in the taproot. However, this micrite cement in the sandstone away from the taproot is cut by rhizoliths that are lined by similarly luminescent micrite, and thus we interpret this as the third cement generation (G3). This texture is analogous to the dense micrite zones recognized by Purvis and Wright (1991). The sixth generation of cement (G6), which cuts and replaces former cement generations in the mudstone and sandstone and replaces some grains in the sandstone, is interpreted to have formed after burial and lithification based on these crosscutting relationships. This calcite was dated at 81 ± 11 Ma (Wang et al., 1998). The age suggests a possible relationship to the maximum advance of the Cretaceous seas during Upper Zuni A time (Haq et al., 1988). The Hartford Basin, at this time, is postulated to have been undergoing tectonic inversion and thermal subsidence (Schlische, 2003). Perhaps this combination of events established a large-scale fluid flow that was responsible for the late-formed calcite cements. However, we were not implying these are marine fluids, rather, that the rise in sea level would necessarily cause a change in base level and drive fluids through the system. Carbon isotope values from the latest cements are indistinguishable from those of the earlier cements (Wang et al., 1998). However, the oxygen isotopes have a much greater range (−10.7‰ to –5.0‰ versus –6.5‰ to –4.6‰), extending to more negative values, consistent with the higher temperatures that would be expected with a burial history. The lower uranium concentrations and much higher manganese concentrations (Wang et al., 1998) in this generation of calcite (G6) are consistent with far more reducing fluids, because uranium is insoluble and manganese is soluble in reducing fluids. These observations support our contention that the fluids responsible for G6 cements are not directly surface-derived. GENESIS OF CALCRETE Nonpedogenic calcretes, such as capillary fringe and groundwater calcretes (Tandon and Kumar, 1999), are carbonate accumulations in soil, sediment, or bedrock associated with the groundwater table, in vadose and phreatic conditions, respectively (Nash, 1997; Alonso-Zarza, 2003). Arid to semiarid as well as humid climate regimes may be conducive to groundwater calcrete formation (Semenuik and Searle, 1985; Tandon and Kumar, 1999), especially related to plant root influences from evapotranspiration (Lucas, 2001). Also important is a high volume of discharge of Ca-rich subsurface waters “where drainages converge, flow gradient decreases, saline waters mix, or permeabilities are low” (Wright and Tucker, 1991, p. 8). A composite origin of groundwater precipitation and pedogenic alteration by rooting has been suggested for nonpedogenic calcrete containing
213
both alpha and beta fabrics (Nash and Smith, 1998; Mack et al., 2000), though timing for the accumulation of associated groundwater and capillary fringe calcretes has been unclear. The New Haven calcrete interval contains calcrete fabrics and gley features. The presence of redox-sensitive elements such as U, Mn, and Fe, combined with detailed study of the cement relationships, strongly suggests a saturated-zone origin for most of this nonpedogenic calcrete interval. Cements G1 and G2 are interpreted as groundwater calcrete based on their isopachous cements and lack of biogenic structures, in addition to the displacive morphology. Cement G3, with its alveolar septal fabric, is interpreted as having a beta texture and is interpreted as a capillary fringe nonpedogenic calcrete that developed when the groundwater level was lower or in contact. No clear vadose textures can be found associated with the G3 cement, but this type of texture may not be preserved with the changing levels of groundwater. Perched water tables can potentially oscillate with respect to climatic stresses, such as variations in precipitation (e.g., Hunt et al., 1988; Fetter, 2001), and can be temporary or permanent, contingent on soil, sediment, and bedrock hydraulic conductivities (Davie, 2003). Groundwater movement can be directed along a sand-mud interface with the impermeable mud preventing infiltration of meteoric water down to the regional water table (Davie, 2003; Rushton, 2003). With the presence of a perched water table in a zone of phreatophytic plant growth during New Haven deposition, both kinds of nonpedogenic calcrete must have formed contemporaneously, producing both alpha and beta textures. The time period between the deposition of cements G4 and G5 and the first three cements in the New Haven Arkose is not known, but G4 may be related to groundwater processes flowing through both the sandstone and mudstone as burial proceeded. Cement G5 is limited to the uppermost part of the sandstone and clearly is not related to the groundwater processes of the first three cements. Cement G6 is a much later diagenetic cement. Most thick groundwater calcretes are interpreted as forming in arid to semiarid conditions (see previous references). However, the New Haven Arkose contains many indicators of frequent flooding (higher sedimentation rate) under a “humid to subhumid” seasonal (monsoonal?) setting, such as poor preservation of well-defined paleosol horizons, ferruginous concretions, intense rooting, multistory channels with wings, and rare preserved primary sedimentary structures in sandstones and mudrocks (Gierlowski-Kordesch and Gibling, 2002). The presence of deep taproots and pedogenic mud aggregates in the New Haven Arkose (Gierlowski-Kordesch and Gibling, 2002) points to seasonality, as would be expected in the monsoonal regime postulated for the Middle to Upper Triassic in eastern North America and Europe (Hay et al., 1982; Sims and Ruffell, 1990; Parrish, 1993; Wilson et al., 1994; Olsen and Kent, 1996; Reinhardt and Ricken, 2000; Kent and Muttoni, 2003). One possible paleoenvironmental interpretation for the studied New Haven Arkose section is a subhumid monsoonal setting with a short dry period. The projected paleolatitude of 7º to 11ºN (Olsen, 1997; Kent and Olsen, 2000) is consistent with this scenario. Ca-rich
Nonpedogenic calcrete in the New Haven Arkose groundwaters could have been sourced from the Paleozoic limestones and marbles exposed at the basin margin to the northwest (see Gierlowski-Kordesch, 1998; De Wet et al., 2002). Extensive recharge of groundwaters is possible in a rift setting (see Rosen, 1994), especially in the tectonically active, incipient Hartford rift (Smoot, 1991), where groundwater and surface water converged into an extensive braided plain undergoing high rates of subsidence. Other calcrete layers within the New Haven Arkose should be reassessed using these new criteria for recognizing pedogenic versus nonpedogenic origin. Significance for U-Pb Dating A better understanding of the conditions that promote favorable U/Pb and U/Th ratios in calcretes will encourage more exact identification of dateable calcrete zones in the geologic record. Because Pb and Th have a low solubility in most fluids (Shen and Boyle, 1988; Langmuir and Herman, 1980), the major influence on these ratios is perhaps the solubility of uranium. Uranium in the oxidized state is known to be quite soluble (Langmuir, 1978; Carlisle, 1983), and reduction should remove it from solution, although it is not clear how this would influence its co-precipitation in calcite. Pedogenic carbonates form in the unsaturated
215
vadose zone and are normally formed from oxidizing fluids. On the other hand, the groundwater table is often associated with a redox boundary (Fig. 10) and would be a zone where fluctuating redox conditions might be expected seasonally. The association of the New Haven calcite with gleying, the fairly high U concentrations (1–14 ppm) within the calcite, and the presence of Mn as evidenced by trace-element analyses and luminescence (Wang et al., 1998) are all consistent with the precipitation of these calcites in a mostly reducing fluid. Additionally, Chung and Swart (1990) concluded that U concentration is higher in the bulk carbonate of the phreatic zone than in the bulk carbonate of the vadose zone and suggested that: (1) higher pCO2 in the phreatic zone leads to more U-rich carbonate complexes; and (2) selective exclusion of U in the vadose zone leads to increased U/Ca ratios in the phreatic zone. These two mechanisms combine to force calcite precipitation from phreatic-zone waters with higher U concentrations. More work is required to examine the desirability of pedogenic calcrete for U-Pb dating. While soil calcretes archive important information about pCO2 as well as climatic data, their formation in the vadose zone may limit the availability of U, which is highly mobile in oxidized fluids. CONCLUSIONS 1.
Figure 6. (A) Plane-light photomicrograph of a brecciated clast at the bottom of Figure 3C. The metamorphic rock fragment is shattered along planes that are parallel to foliation. Fragment of the clast at the bottom of the photomicrograph has layers of micrite and bladed calcite (G1, G2) followed by micrite calcite. Fragment of the clast at the top of the photomicrograph is surrounded by micrite (G3). (B) Cathodoluminescence photomicrograph of the same area as A. Calcite that fringes the clast fragment at the bottom of the photomicrograph has alternating bright luminescent micritic calcite and dull-luminescent bladed calcite. A millimeter-scale vein separates the fragments of the metamorphic rock clast. This vein contains dull-luminescent micrite (G3) interpreted to have lined roots and is filled by a blotchy brightly luminescent sparry calcite. Although not labeled, the brightly luminescent calcite is likely replacement calcite by fluids responsible for G6. (C–D) Planelight and cathodoluminescence photomicrographs of the boundary between the brecciated metamorphic rock fragment (upper right corner) and vein that surrounds the clast. Sparry calcite of the large vein can be followed into smaller veins that break apart the clast and appear to be equivalent to the root-cast-filling sparry calcite (G4). Calcite in the large veins is similar to the granular mosaic calcite described by Purvis and Wright (1991) in that the crystals display a zoned luminescence. Interestingly, the granular mosaic nature of this calcite is not observed in the smaller veins, although petrography suggests that they are versions of the same cement generation. (E–F) Plane-light and cathodoluminescence photomicrographs of a clast with alteration at the edges that is rimmed by dull-luminescent bladed calcite followed by root-lining–type micrite cement with associated sparry calcite. Although not obvious on the plane-light image, there are numerous small veins of brightly luminescent calcite (G6) that cut the root-lining cement (G3) and sparry fill (G4). Here, the micrite calcite is brightly luminescent, suggesting alteration by the G6 fluid.
2.
3.
In addition to field relationships and calcrete fabric analysis, calcite cement stratigraphy, combined with redox models for the behavior of Mn, Fe, and U, is a useful technique to distinguish between pedogenic versus nonpedogenic calcrete, as well as between capillary fringe and groundwater calcrete timing. The carbonate accumulation associated with the boundary between one of many sandstone-mudstone units in the New Haven Arkose near Meriden, Connecticut, is interpreted as a nonpedogenic calcrete rather than one of pedogenic origin. The presence of rhizoliths within this calcitized horizon highlights the fact that the presence of rhizoliths does not by itself require a soil origin. A fluctuating perched paleo–water table, coupled with a zone containing phreatophytic vegetation with deep taproots, can produce alpha and beta fabrics in a groundwater and capillary fringe nonpedogenic calcrete. Reducing conditions are postulated for calcite precipitation because of gleying features, the presence of Mn and U, and an absence of vadose cementation structures. Cement stratigraphy establishes a diagenetic history of calcite precipitation in a calcrete interval of the Triassic New Haven Arkose. Cements G1 and G2 are interpreted as groundwater calcrete because of displacive features and alpha fabrics. Cement G3 is interpreted as having formed from phreatophytic vegetation at or just above an oscillating groundwater table of a perched aquifer. This cement contains beta fabric features, such as alveolar septal fabric. Cement G4 and G5 are interpreted as diagenetic
Figure 7. (A) Plane-light photomicrograph of the basal sandstone showing sand grains floating in a micrite matrix. This is cut by micrite-lined rhizoliths that are filled by sparry calcite. (B) Cathodoluminescence photomicrograph of the same area as A. Micrite that displaces the sand grains is dull-luminescent, like that of the micrite that lines the roots. It seems clear that this dense micrite calcite predates the roots that appear to cut it. The first layer of sparry calcite following this micrite is nonluminescent, analogous to the sparry calcite that completely fills roots in the mudstone (G4). However, this is followed by a volumetrically more significant sparry calcite that is brightly luminescent (G5). Although we have broken these into two generations, they are almost certainly variations of the same fluid event. (C–D) Plane-light and cathodoluminescence photomicrographs of the upper part of the sandstone showing the same floating grains in micrite cement and crosscutting rhizoliths with nonluminescent spar followed by brightly luminescent spar. Several clasts are replaced by brightly luminescent calcite (G6). (E–F) Plane-light and cathodoluminescence photomicrographs from the upper sandstone showing grains floating in the dull-luminescent micrite. The micrite-bearing host is then cut by several generations of rhizoliths. A branch of the rhizolith on the left clearly cuts the rhizolith on the right. The rhizoliths appear to be largely replaced by brightly luminescent calcite, interpreted as G6.
Figure 8. Photomicrographs from one of the millimeter-scale layers of compacted sandstone. (A) Plane-light photomicrograph showing interpenetrating and deformed grains. (B) Cathodoluminescence image of the area shown in A, showing only one generation of brightly luminescent calcite that replaces some of the grains. (C) Plane-light photomicrograph that shows interpenetrating and deformed grains. (D) Cathodoluminescence image of the area shown in C. We hypothesize that the replaced grains were originally feldspars.
218
4.
Rasbury et al. precipitation not associated with the calcretization. Cement G6 is a late diagenetic cement. The U-Pb age of cement G3 obtained by Wang et al. (1998) dates the time of sedimentation within the resolution of the dating technique. The reducing conditions of calcrete formation associated with phreatic groundwater conditions are conducive to the accumulation and preservation of U for geochronologic dating. This research, combined with that of
Wang et al. (1998) and Rasbury et al. (2000), shows great scope for U-Pb dating of nonpedogenic calcretes. ACKNOWLEDGMENTS Special thanks to David Nash for his insightful reviews that improved the manuscript greatly. We are grateful to the editors for their direction and patience. This research was funded
Figure 9. Cartoon summarizing the sedimentologic and petrographic relationships of calcite cement in a fluvial unit of the New Haven Arkose. At the field scale, there is an obvious concentration of carbonate cement and gleying at the sandstone-mudstone interface. Additionally, there are vertical features, interpreted as having originally been taproots, in which gleying is also prominent. Throughout the sequence on the millimeter scale, veins are composed of alveolar textures (beta fabrics), which are composed of micrite (G3). This micrite cuts sandstone and mudstone that are characterized by floating grains in calcite (alpha fabrics). In the sandstone, this displacive calcite is mostly micrite. In the mudstone, it is mixed micrite and sparry calcite with zoned cathodoluminescence. There are also millimeter-scale zones that appear not to have experienced early calcite cementation and show abundant evidence of physical compaction. These zones have only the last calcite cement generation (G5), which in this case is mostly seen as a replacement of grains that were most likely feldspars. At the submillimeter scale, sand grains are often coated by a prismatic calcite (G1). In veins, this prismatic calcite is followed by a dull-luminescent sparry calcite (G2) and micrite that forms rhizoliths (G3). Toothpick-shaped veins in the mudstone are composed of cathodoluminescence-zoned sparry calcite (G2). Not shown in this cartoon is the nonluminescent sparry calcite that fills the voids in rhizoliths (G4).
Nonpedogenic calcrete in the New Haven Arkose
Horizontal roots high organic matter preservation low Eh
219
Vertical roots low organic matter preservation high Eh
Figure 10. Morphology of roots and the geochemical conditions expected in vadose versus phreatic type soils (modified from Mount and Cohen, 1984). Often, ancient soils are recognized by the presence of fossil roots. These roots may be preserved by precipitation of calcite around the roots, forming rhizoliths. In the vadose zone, roots are vertical because trees have to penetrate to the water table. When the water table is at or near the surface, roots are often horizontal. The vadose zone is usually oxidizing, while below the water table, conditions are often reducing. Eh is a measure of the redox potential of a system, where high values tend to indicate more oxidizing environments and low values indicate more reducing environments.
through NSF grant EAR9814639 to Troy Rasbury and was largely accomplished by undergraduate researchers funded by a REU supplement to that grant. We also thank Bruce Ward and Carol DeWet for ideas and valuable input. REFERENCES CITED Alonso-Zarza, A.M., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: Earth-Science Reviews, v. 60, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Blisniuk, P.M., and Sharp, W.D., 2003, Rates of late Quaternary normal faulting in central Tibet from U-series dating of pedogenic carbonate in displaced fluvial gravel deposits: Earth and Planetary Science Letters, v. 215, p. 169–186, doi: 10.1016/S0012-821X(03)00374-1. Candy, I., Black, S., Sellwood, B.W., and Rowan, J.S., 2003, Calcrete profile development in Quaternary alluvial sequences, southeast Spain: Implications for using calcretes as a basis for landform chronologies: Earth Surface Processes and Landforms, v. 28, p. 169–185, doi: 10.1002/esp.445. Carlisle, D., 1983, Concentrations of uranium and vanadium in calcretes and gypcretes, in Wilson, R.C.L., ed., Residual Deposits: Geological Society [London] Special Publication 11, p. 185–195. Cerling, T.E., 1984, The stable isotopic composition of modern soil carbonate and its relation to climate: Earth and Planetary Science Letters, v. 71, p. 229–240, doi: 10.1016/0012-821X(84)90089-X. Cerling, T.E., and Quade, J., 1993, Stable carbon and oxygen isotopes in soil carbonates, in Swart, P., McKenzie, J.A., and Lohmann, K.C., eds., Climate Change in Continental Isotopic Records: American Geophysical Union Geophysical Monograph 78, p. 217–231. Chadwick, O.A., and Graham, R.C., 2000, Pedogenic processes, in Sumner, M.E., ed., Handbook of Soil Science: Boca Raton, CRC Press, p. E41–E75. Chung, G.S., and Swart, P.K., 1990, The concentration of uranium in freshwater vadose and phreatic cements in a Holocene ooid bay: A method of identifying ancient water tables: Journal of Sedimentary Petrology, v. 60, p. 735–746. Clothier, B.E., and Green, S.R., 1997, Roots: The big movers of water and chemical in soil: Soil Science, v. 162, p. 534–543, doi: 10.1097/00010694199708000-00002.
Cornet, B., and Traverse, A., 1975, Palynological contributions to the chronology and stratigraphy of the Hartford Basin in Connecticut and Massachusetts: Geoscience and Man, v. 11, p. 1–33. Davie, T., 2003, Fundamentals of Hydrology: London and New York, Routledge, 169 p. De Wet, C.B., Mora, C.I., Gore, P.J.W., Gierlowski-Kordesch, E., and Cucolo, S.J., 2002, Deposition and geochemistry of lacustrine and spring carbonates in Mesozoic rift basins, eastern North America, in Ashley, G., and Renaut, R.W., eds., Sedimentation in Continental Rifts: Society for Sedimentary Geology (SEPM) Special Publication 73, p. 309–325. Dunning, G.R., and Hodych, J.P., 1990, U/Pb zircon and baddeleyite ages for the Palisades and Gettysburg sills of the northeastern United States—Implications for the age of the Triassic-Jurassic boundary: Geology, v. 18, p. 795– 798, doi: 10.1130/0091-7613(1990)018<0795:UPZABA>2.3.CO;2. Fetter, C.W., 2001, Applied Hydrogeology (4th edition): Upper Saddle River, New Jersey, Prentice Hall, 598 p. Folk, R.L., 1987, Detection of organic matter in thin sections of carbonate rocks using a white card: Sedimentary Geology, v. 54, p. 193–200, doi: 10.1016/0037-0738(87)90022-4. Gierlowski-Kordesch, E.H., 1998, Carbonate deposition in an ephemeral siliciclastic alluvial system: Jurassic Shuttle Meadow Formation, Newark Supergroup, Hartford Basin, USA: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 140, p. 161–184, doi: 10.1016/S00310182(98)00039-X. Gierlowski-Kordesch, E., and Gibling, M.R., 2002, Pedogenic mud aggregates in rift sedimentation, in Ashley, G., and Renaut, R.W., eds., Sedimentation in Continental Rifts: Society for Sedimentary Geology (SEPM) Special Publication 73, p. 195–206. Gierlowski-Kordesch, E., and Huber, P., 1995, Lake sequences in the central Hartford Basin: Newark Supergroup: State Geological and Natural History Survey of Connecticut, Guidebook, v. 7, p. B1–B39. Hames, W.E., Renne, P.R., and Ruppel, C., 2000, New evidence for geologically instantaneous emplacement of earliest Jurassic Central Atlantic magmatic province basalts on the North American margin: Geology, v. 28, p. 859– 862, doi: 10.1130/0091-7613(2000)028<0859:NEFGIE>2.3.CO;2. Haq, B.U., Hardendol, J., and Vail, P.R., 1988, Mesozoic and Cenozoic chronostratigraphy and cycles of sea-level change, in Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., and Van Wagoner, J.C., eds., Sea-Level Changes: An Integrated Approach: Society for Sedimentary Geology (SEPM) Special Publication 42, p. 73–108.
220
Rasbury et al.
Harden, J.E., Taylor, E.M., Reheis, M.C., and McFadden, L.D., 1991, Calcic, gypsic, siliceous soil chronosequences in arid and semi-arid environments, in Nettleton, W.D., ed., Occurrence, Characteristics, and Genesis of Carbonate, Gypsum, and Silica Accumulations in Soils: Soil Science Society of America Special Publication 26, p. 1–16. Hay, W.W., Behensky, J.F., Jr., Barron, E.J., and Sloan, J.L., 1982, Late Triassic–Liassic palaeoclimatology of the protocentral North Atlantic rift system: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 40, p. 13– 30, doi: 10.1016/0031-0182(82)90083-9. Hodych, J.P., and Dunning, G.R., 1992, Did the Manicouagan impact trigger end-of-Triassic mass extinction: Geology, v. 20, p. 51–54, doi: 10.1130/0091-7613(1992)020<0051:DTMITE>2.3.CO;2. Hubert, J.F., 1977, Paleosol caliche in the New Haven Arkose, Connecticut: Record of semiaridity in Late Triassic–Early Jurassic time: Geology, v. 5, p. 302–304, doi: 10.1130/0091-7613(1977)5<302:PCITNH>2.0.CO;2. Hubert, J.F., 1978, Paleosol caliche in the New Haven Arkose, Newark Supergroup, Connecticut: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 24, p. 151–168, doi: 10.1016/0031-0182(78)90004-4. Hubert, J.F., Feshbach-Meriney, P.E., and Smith, M.A., 1992, The TriassicJurassic Hartford rift basin, Connecticut and Massachusetts: Evolution, sandstone diagenesis, and hydrocarbon history: American Association of Petroleum Geologists Bulletin, v. 76, p. 1710–1734. Hunt, C.D., Jr., Ewart, C.J., and Voss, C.I., 1988, Region 27, Hawaiian Islands, in Back, W., Rosenshein, J.S., and Seaber, P.R., eds., Hydrogeology: Boulder, Colorado, Geological Society of America, The Geology of North America, v. O-2, p. 255–262. Kelly, M., Black, S., and Rowan, J.S., 2000, A calcrete-based U/Th chronology for landform evolution in the Sorbas basin, southeast Spain: Quaternary Science Reviews, v. 19, p. 995–1010, doi: 10.1016/S02773791(99)00050-5. Kent, D.V., and Muttoni, G., 2003, Mobility of Pangea: Implications for late Paleozoic and early Mesozoic paleoclimate, in LeTourneau, P.M., and Olsen, P.E., eds., The Great Rift Valleys of Pangea in Eastern North America: Sedimentology, Stratigraphy, and Paleontology, Volume 1: New York, Columbia University Press, p. 11–20. Kent, D.V., and Olsen, P.E., 2000, Magnetic polarity stratigraphy and paleolatitudes of the Triassic-Jurassic Blomidon Formation in the Fundy basin (Canada): Implications for the early Mesozoic tropical climate gradients: Earth and Planetary Science Letters, v. 179, p. 311–324, doi: 10.1016/ S0012-821X(00)00117-5. Khadkikar, A.S., Merh, S.S., Malik, J.N., and Chamyal, L.S., 1998, Calcretes in semi-arid alluvial systems: Formative pathways and sinks: Sedimentary Geology, v. 116, p. 251–260, doi: 10.1016/S0037-0738(97)00103-6. Khadkikar, A.S., Chamyal, L.S., and Ramesh, R., 2000, The character and genesis of calcrete in late Quaternary alluvial deposits, Gujarat, western India, and its bearing on the interpretation of ancient climates: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 162, p. 239–261, doi: 10.1016/S0031-0182(00)00130-9. Klappa, C.F., 1980, Rhizoliths in terrestrial carbonates: Classification, recognition, genesis, and significance: Sedimentology, v. 27, p. 613–629. Krynine, P.D., 1950, Petrology, stratigraphy, and origin of the Triassic rocks of Connecticut: Connecticut State Geology and Natural History Survey Bulletin, v. 73, 247 p. Lal, R., and Kimble, J.M., 2000, Pedogenic carbonates and the global carbon cycle, in Lal, R., Kimble, J.M., Eswaran, H., and Stewart, B.A., eds., Global Climate Change and Pedogenic Carbonates: Boca Raton, Lewis Publishers, p. 1–14. Langmuir, D., 1978, Uranium solution-mineral equilibria at low temperatures with applications to sedimentary ore deposits: Geochimica et Cosmochimica Acta, v. 42, p. 547–569, doi: 10.1016/0016-7037(78)90001-7. Langmuir, D., and Herman, J.S., 1980, The mobility of thorium in naturalwaters at low-temperatures: Geochimica et Cosmochimica Acta, v. 44, p. 1 753–1766. Lorenz, J.C., 1987, Triassic-Jurassic Rift Basin Sedimentology—History and Methods: New York, Van Nostrand Rheinhold, 315 p. Lucas, Y., 2001, The role of plants in controlling rates and products of weathering: Importance of biological pumping: Annual Review of Earth and Planetary Sciences, v. 29, p. 135–163, doi: 10.1146/annurev.earth.29.1.135. Ludwig, K.R., and Paces, J.B., 2002, Uranium-series dating of pedogenic silica and carbonate, Crater Flat, Nevada: Geochimica et Cosmochimica Acta, v. 66, p. 487–506, doi: 10.1016/S0016-7037(01)00786-4. Machette, M.N., 1985, Calcic soils of the southwestern United States, in Weide, D.L., ed., Soil and Quaternary Geology of the Southwestern United States:
Geological Society of America Special Paper 203, p. 1–22. Mack, G.H., Cole, D.R., and Treviño, L., 2000, The distribution and discrimination of shallow, authigenic carbonate in the Pliocene-Pleistocene Palomas basin, southern Rio Grande rift: Geological Society of America Bulletin, v. 112, p. 643–656, doi: 10.1130/0016-7606(2000)112<0643: TDADOS>2.3.CO;2. Manspeizer, W., 1988, Triassic-Jurassic rifting and opening of the Atlantic: An overview, in Manspeizer, M., ed., Triassic-Jurassic Rifting: Continental Break-Up and the Origin of the Atlantic Ocean and Passive Margins, Part A: Amsterdam, Elsevier, p. 41–79. Manspeizer, W., 1994, The breakup of Pangea and its impact on climate: Consequences of Variscan-Alleghanide orogenic collapse, in Klein, G.D., ed., Pangea: Paleoclimate, Tectonics, and Sedimentation during Accretion, Zenith, and Breakup of a Supercontinent: Geological Society of America Special Paper 288, p. 169–185. McInerney, D.P., and Hubert, J.F., 2003, Meandering-river facies in the Upper Triassic New Haven Arkose, south-central Connecticut: Early evolution of the Hartford rift basin, in LeTourneau, P.M., and Olsen, P.E., eds., The Great Rift Valleys of Pangea in Eastern North America: Sedimentology, Stratigraphy, and Paleontology, Volume 2: New York, Columbia University Press, p. 88–107. Mermut, A.R., Amundson, R., and Cerling, T.E., 2000, The use of stable isotopes in studying carbonate dynamics in soils, in Lal, R., Kimble, J.M., Eswaran, H., and Stewart, B.A., eds., Global Climate Change and Pedogenic Carbonates: Boca Raton, Lewis Publishers, p. 65–85. Meyers, W.J., 1974, Carbonate cement stratigraphy of the Lake Valley Formation (Mississippian) Sacramento Mountains, New Mexico: Journal of Sedimentary Research, v. 44, p. 837–861. Mount, J.F., and Cohen, A.S., 1984, Petrology and geochemistry of rhizoliths from Plio-Pleistocene fluvial and marginal lacustrine deposits: East Lake Turkana, Kenya: Journal of Sedimentary Petrology, v. 54, p. 263–275. Nanson, G.C., and Croke, J.C., 1992, A genetic classification of floodplains: Geomorphology, v. 4, p. 459–486, doi: 10.1016/0169-555X(92)90039-Q. Nash, D.J., 1997, Groundwater as a geomorphological agent in drylands, in Thomas, D.S.G., ed., Arid Zone Geomorphology: Process, Form, and Change in Drylands: Chichester, John Wiley and Sons, p. 319–348. Nash, D.J., and Smith, R.F., 1998, Multiple calcrete profiles in the Tabernas basin, southeast Spain: Their origins and geomorphic implications: Earth Surface Processes and Landforms, v. 23, p. 1009–1029, doi: 10.1002/(SICI)10969837(1998110)23:11<1009::AID-ESP918>3.0.CO;2-Z. Olsen, P.E., 1997, Stratigraphic record of the early Mesozoic breakup of Pangea in the Laurasia Gondwana rift system: Annual Review of Earth and Planetary Sciences, v. 25, p. 337–401, doi: 10.1146/annurev.earth.25.1.337. Olsen, P.E., and Kent, D.V., 1996, Milankovitch climate forcing in the tropics of Pangaea during the Late Triassic: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 122, p. 1–26, doi: 10.1016/0031-0182(95)00171-9. Olsen, P.E., McCune, A.R., and Thomson, K.S., 1982, Correlations of the early Mesozoic Newark Supergroup by vertebrates, principally fishes: American Journal of Science, v. 282, p. 1–44. Olsen, P.E., Schlische, R.W., and Gore, P.J.W., eds., 1989, Tectonic, Depositional, and Paleoecological History of Early Mesozoic Rift Basins, Eastern North America: 1989 International Geological Congress Field Trip T-351: Washington, D.C., 28th International Geological Congress, 174 p. Parrish, J.T., 1993, Climate of the supercontinent Pangea: The Journal of Geology, v. 101, p. 215–233. PiPujol, M.D., and Buurman, P., 1994, The distinction between ground-water gley and surface-water gley phenomena in Tertiary paleosols of the Ebro Basin, NE Spain: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 110, p. 103–113. Purvis, K., and Wright, V.P., 1991, Calcretes related to phreatophytic vegetation from the Middle Triassic Otter Sandstone of south west England: Sedimentology, v. 38, p. 539–551. Quade, J., and Roe, L.J., 1999, The stable-isotope composition of early groundwater cements from sandstone in paleoecological reconstruction: Journal of Sedimentary Research, v. 69, p. 667–674. Rasbury, E.T., Hanson, G.N., Meyers, W.J., and Saller, A.H., 1997, Dating of the time of sedimentation using U-Pb ages for paleosol calcite: Geochimica et Cosmochimica Acta, v. 61, p. 1525–1529, doi: 10.1016/S00167037(97)00043-4. Rasbury, E.T., Hanson, G.N., Meyers, W.J., Holt, W.E., Goldstein, R.H., and Saller, A.H., 1998, U-Pb dates of paleosols: Constraints on late Paleozoic cycle durations and boundary ages: Geology, v. 26, no. 5, p. 403–406, doi: 10.1130/0091-7613(1998)026<0403:UPDOPC>2.3.CO;2.
Nonpedogenic calcrete in the New Haven Arkose Rasbury, E.T., Meyers, W.J., Hanson, G.N., Goldstein, R.H., and Saller, A.H., 2000, Relationship of uranium to petrography of caliche paleosols with application to precisely dating the time of sedimentation: Journal of Sedimentary Research, v. 70, p. 604–618. Reeder, R.J., Nugent, M., Lamble, G.M., Tait, C.D., and Morris, D.E., 2000, Uranyl incorporation into calcite and aragonite: XAFS and luminescence studies: Environmental Science & Technology, v. 34, p. 638–644, doi: 10.1021/es990981j. Reinhardt, L., and Ricken, W., 2000, The stratigraphic and geochemical record of playa cycles: Monitoring a Pangaean monsoon-like system (Triassic, Middle Keuper, s. Germany): Palaeogeography, Palaeoclimatology, Palaeoecology, v. 161, p. 205–227, doi: 10.1016/S0031-0182(00)00124-3. Retallack, G.J., 1988, Field recognition of paleosols, in Reinhardt, J., and Sigleo, W.R., eds., Paleosols and Weathering through Time: Principles and Applications: Geological Society of America Special Paper 216, p. 1–20. Retallack, G.J., 1991, Untangling the effects of burial alteration and ancient soil formation: Annual Review of Earth and Planetary Sciences, v. 19, p. 183–206, doi: 10.1146/annurev.ea.19.050191.001151. Rosen, M.R., 1994, The importance of groundwater in playas: A review of playa classifications and the sedimentology and hydrology of playas, in Rosen, M.R., ed., Paleoclimate and Basin Evolution of Playa Systems: Geological Society of America Special Paper 289, p. 1–18. Rossinsky, V., Wanless, H.R., and Swart, P.K., 1992, Penetrative calcretes and their stratigraphic implications: Geology, v. 20, p. 331–334, doi: 10.1130/0091-7613(1992)020<0331:PCATSI>2.3.CO;2. Rushton, K.R., 2003, Groundwater Hydrology: Conceptual and Computational Models: Chichester, John Wiley and Sons, 416 p. Saad, M.K., 1991, Diagenesis of aluminosilicate minerals in the New Haven Arkose and East Berlin Formation (Triassic-Jurassic), southeastern Hartford Basin [Ph.D. dissertation]: East Lansing, Michigan, Michigan State University, 241 p. Schlische, R.W., 1993, Anatomy and evolution of the Triassic-Jurassic continental rift system, eastern North America: Tectonics, v. 12, p. 1026–1042. Schlische, R.W., 2003, Progress in understanding the structural geology, basin evolution, and tectonic history of the eastern North American rift system, in LeTourneau, P.M., and Olsen, P.E., eds., The Great Rift Valleys of Pangea in Eastern North America: Sedimentology, Stratigraphy, and Paleontology, Volume 1: New York, Columbia University Press, p. 21–64. Semenuik, V., and Meagher, T.D., 1981, Calcrete in Quaternary coastal dunes in southwestern Australia: A capillary-rise phenomenon associated with plants: Journal of Sedimentary Petrology, v. 51, p. 47–68. Semenuik, V., and Searle, D.J., 1985, Distribution of calcrete in Holocene coastal sands in relationship to climate, southwestern Australia: Journal of Sedimentary Petrology, v. 55, p. 86–95. Sharp, W.D., Ludwig, K.R., Chadwick, O.A., Amundson, R., and Glaser, L.L., 2003, Dating fluvial terraces by Th-230/U on pedogenic carbonate, Wind River Basin, Wyoming: Quaternary Research, v. 59, p. 139–150, doi: 10.1016/S0033-5894(03)00003-6. Shen, G.T., and Boyle, E.A., 1988, Determination of lead, cadmium and other trace-metals in annually-banded corals: Chemical Geology, v. 67, p. 47– 62, doi: 10.1016/0009-2541(88)90005-8. Sims, M.J., and Ruffell, A.H., 1990, Climatic and biotic change in the Late Triassic: Journal of the Geological Society of London, v. 147, p. 321–327.
221
Slate, J.L., Smith, G.A., Wang, Y., and Cerling, T.E., 1996, Carbonate-paleosol genesis in the Plio-Pleistocene St. David Formation, southeastern Arizona: Journal of Sedimentary Petrology, v. 66, p. 85–94. Smoot, J.P., 1991, Sedimentary facies and depositional environments of early Mesozoic Newark Supergroup basins, eastern North America: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 84, p. 369–423, doi: 10.1016/0031-0182(91)90055-V. Spötl, C., and Wright, V.P., 1992, Groundwater dolocretes from the Upper Triassic of the Paris Basin, France: A case study of an arid, continental diagenetic facies: Sedimentology, v. 39, p. 1119–1136. Tandon, S.K., and Andrews, J.E., 2001, Lithofacies associations and stable isotopes of palustrine and calcrete carbonates: Examples from an Indian Maastrichtian regolith: Sedimentology, v. 48, p. 339–355, doi: 10.1046/ j.1365-3091.2001.00367.x. Tandon, S.K., and Kumar, S., 1999, Semi-arid/arid zone calcretes: A review, in Singhvi, A.K., and Derbyshire, E., eds., Paleoenvironmental Reconstruction in Arid Lands: Rotterdam, A.A. Balkema, p. 109–152. van de Kamp, P.C., and Leake, B.E., 1996, Petrology, geochemistry, and sodium metasomatism of Triassic-Jurassic nonmarine clastic sediments in the Newark, Hartford, and Deerfield Basins, northeastern USA: Chemical Geology, v. 133, p. 89–124, doi: 10.1016/S0009-2541(96)00071-X. Wang, Z.S., Rasbury, E.T., Hanson, G.N., and Meyers, W.J., 1998, Using the U-Pb system of calcretes to date the time of sedimentation of clastic sedimentary rocks: Geochimica et Cosmochimica Acta, v. 62, p. 2823–2835, doi: 10.1016/S0016-7037(98)00201-4. Watts, N.L., 1978, Displacive calcite: Evidence from recent and ancient calcretes: Geology, v. 6, p. 699–703, doi: 10.1130/0091-7613(1978)6<699: DCEFRA>2.0.CO;2. Weems, R.E., and Olsen, P.E., 1997, Synthesis and revision of groups within the Newark Supergroup, eastern North America: Geological Society of America Bulletin, v. 109, p. 195–209, doi: 10.1130/0016-7606(1997)109<0195: SAROGW>2.3.CO;2. Williams, C.A., and Krause, F.F., 1998, Pedogenic-phreatic carbonates on a Middle Devonian (Givetian) terrigenous alluvial-deltaic plain, Gilwood Member (Watt Mountain Formation), north-central Alberta, Canada: Sedimentology, v. 45, p. 1105–1124. Wilson, K.M., Pollard, D., Hay, W.W., Thompson, S.L., and Wold, C.N., 1994, General circulation model simulations of the Triassic climates: Preliminary results, in Klein, G.D., ed., Pangea: Paleoclimate, Tectonics, and Sedimentation during Accretion, Zenith, and Breakup of a Supercontinent: Geological Society of America Special Paper 288, p. 91–116. Wright, V.P., 1990, A micromorphological classification of fossil and recent calcic and petrocalcic microstructures, in Douglas, L.A., ed., Soil Micromorphology: A Basic and Applied Science: Developments in Soil Science, v. 19, p. 401–407. Wright, V.P., 1992, Paleopedology: Stratigraphic relationships and empirical models, in Martini, I.P., and Chesworth, W., eds., Weathering, Soils, and Paleosols: Amsterdam, Elsevier, p. 475–499. Wright, V.P., and Tucker, M.E., 1991, Calcretes: An introduction, in Wright, V.P., and Tucker, M.E., eds., Calcretes: International Association of Sedimentologists Reprint Series, v. 2, p. 1–22. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Geological Society of America Special Paper 416 2006
Calcrete features and age estimates from U/Th dating: Implications for the analysis of Quaternary erosion rates in the northern limb of the Sierra Nevada range (Betic Cordillera, southeast Spain) J.M. Azañón Departamento de Geodinámica, Universidad de Granada, 18071 Granada, Spain, and Instituto Andaluz de Ciencias de la Tierra, CSIC-Universidad de Granada, 18071 Granada, Spain P. Tuccimei Dipartimento di Scienze Geologiche, Universitá Roma Tre, 00146 Roma, Italy A. Azor Departamento de Geodinámica, Universidad de Granada, 18071 Granada, Spain I.M. Sánchez-Almazo Centro Andaluz de Medio Ambiente (CEAMA), Universidad de Granada, 18071 Granada, Spain A.M. Alonso-Zarza Departamento de Petrología y Geoquímica, Facultad de Ciencias Geológicas, Universidad Complutense, 28040 Madrid, Spain M. Soligo Dipartimento di Scienze Geologiche, Universitá Roma Tre, 00146 Roma, Italy J.V. Pérez-Peña Departamento de Geodinámica, Universidad de Granada, 18071 Granada, Spain ABSTRACT The Guadix topographic depression is a Neogene-Quaternary basin located in the central sector of the Betic Cordillera at the boundary between the South Iberian margin and the Alboran domain. This topographic depression is a plateau with an average elevation of 1000 m in the northern limb of the Sierra Nevada range. The continental deposits infilling the Guadix basin span time from the late Tortonian to the Pleistocene, when a laminar calcrete developed on fine- to coarse-grained fluvial and lacustrine deposits. The drainage pattern is strongly incised (up to 200 m) below the calcrete layer. Four coeval subsamples from the top laminae of the calcrete were collected and dated by the U/Th method. The resulting date is 42.6 ± 5.6 ka, which indicates the minimum age for the cessation of active sedimentation in the Guadix basin. Using this age, we have calculated the incision and erosion rates for the late Pleistocene to present-day time span in the Arroyo de Gor, a highly incised canyon in the eastern border of the Guadix basin. The minimum incision rates in this canyon are around 4 mm/yr. We envisage the capture of the Pliocene-Pleistocene endorheic Azañón, J.M., Tuccimei, P., Azor, A., Sánchez-Almazo, I.M., Alonso-Zarza, A.M., Soligo, M., and Pérez-Peña, J.V., 2006, Calcrete features and age estimates from U/Th dating: Implications for the analysis of Quaternary erosion rates in the northern limb of the Sierra Nevada range (Betic Cordillera, southeast Spain), in Alonso-Zarza, A.M., and Tanner, L.H., eds., Paleoenvironmental Record and Applications of Calcretes and Palustrine Carbonates: Geological Society of America Special Paper 416, p. 223–239, doi: 10.1130/2006.2416(14). For permission to copy, contact
[email protected]. ©2006 Geological Society of America. All rights reserved.
223
224
Azañón et al. Guadix basin by the Guadalquivir River after 42 ka as the main factor triggering the formation of the present-day eroded landscape. After the capture, the combination of climatic (wet periods), lithological (soft and loose sediments), and topographic (high average altitude) features allowed the development of the present-day entrenched drainage pattern. Keywords: calcretes, U/Th dating, stable isotopes, Quaternary incision rates, river capture, Guadix basin, Betic Cordillera, SE Spain. RESUMEN La depresión de Guadix es una cuenca neógeno-cuaternaria situada en el sector central de la Cordillera Bética cubriendo el contacto entre el Margen Sudibérico y el Dominio de Alborán. Esta depresión topográfica es, sin embargo, una superficie elevada (sobre unos 1000 m) desarrollada en el flanco N de Sierra Nevada. El relleno continental de la cuenca de Guadix abarca desde el Tortoniense superior al Pleistoceno, y está coronado por una calcreta laminar que se desarrolló sobre materiales detríticos lacustres y fluviales. La red de drenaje está fuertemente encajada (hasta 200 m) bajo este nivel de calcretas. Se han datado, mediante el método de U/Th, cuatro sub-muestras correspondientes a las facies laminares situadas en el techo del nivel de calcretas más alto. El resultado de la datación de la calcreta es 42.6 ± 5.6 ka, que puede interpretarse como la edad mínima para el final de la sedimentación activa en la cuenca de Guadix. Usando esta edad como referencia, hemos calculado las tasas de incisión y erosión desde el Pleistoceno superior en el Arroyo de Gor, un cañón fuertemente encajado en el borde oriental de la cuenca de Guadix. Las tasas de incisión en este cañón están alrededor de 4 mm/año. Consideramos que la captura PliocenaPleistocena (post–42 ka) de la cuenca, con carácter endorreico en ese momento, por parte del río Guadalquivir es el principal factor desencadenante del actual relieve erosivo que presenta la cuenca de Guadix. Tras la captura, la combinación de factores climáticos (periodos húmedos), litológicos (sedimentos detríticos con escasa cohesión) y topográficos (alta altitud media) han favorecido el encajamiento progresivo de la red de drenaje actual. Palabras clave: calcretas, datación U/Th, Isótopos estables, tasas de incisión cuaternaria, captura fluvial, Cuenca de Guadix, Cordilleras Béticas, SE España.
INTRODUCTION Quaternary calcretes are widespread in many ancient terrestrial basins from all over the world, including Australia (Arakel, 1986), southern Africa (Watts, 1980; Nash and McLaren, 2003), northwestern America (Machette, 1985), and southern Europe, especially Spain (Alonso-Zarza et al., 1998a). In the case of the Spanish Quaternary calcretes, they occur either in aggradational regimes interbedded with alluvial sediments (Jiménez-Espinosa and Jiménez-Millán, 2003), or, most commonly, in degradational regimes on different terrace levels (Sancho et al., 2004). In both cases, detailed studies of the calcretes have provided most valuable data that illuminate rates of fluvial aggradation, fluvial incision, climatic regime, and even tectonic activity. Additionally, Quaternary calcretes cap the sedimentary infill of some of the mostly terrestrial Cenozoic basins in Spain,
such as the Ebro basin (Sancho and Meléndez, 1992) and the Teruel basin (Alonso-Zarza and Arenas, 2004). In these cases, thick laminar calcretes constitute the last material accumulated at the top of the sedimentary sequences, which are incised by the present-day fluvial network. Further to the south and southeast, thick calcrete profiles also formed at the top of some NeogeneQuaternary basins in the Betic Cordillera (Dumas, 1969; Kelly et al., 2000; Candy et al., 2003; García et al., 2003; Nash and Smith, 2003). Calcrete formation in the Betic Neogene-Quaternary basins predates the incision of the present-day fluvial network, thus, if dated radiometrically, its presents a potentially useful geomorphic tool to establish incision rates of the main rivers. Precise dating of pedogenic carbonates either by 14C or U/Th series has proved to be a useful tool to establish both the chronology of sequences of terraces and incision rates in SE Spain (Kelly et al., 2000; Candy et al., 2004). Moreover, the radiometric date
Calcrete features and age estimates in southeast Spain of pedogenic carbonates, together with oxygen and carbon stable isotope studies, is of paramount importance in establishing Quaternary paleoclimatic regimes. In this paper, we use a multidisciplinary approach to study a calcrete layer developed at the top of the sedimentary infilling of the Guadix basin in SE Spain (Figs. 1 and 2). The calcrete constitutes a flat geomorphic surface in which the present-day drainage network is entrenched. U/Th dating and stable isotope analyses of the top laminae in this calcrete have provided us with a radiometric age and allowed us to propose a paleoclimatic setting for the development of this surface. Moreover, a detailed petrographic study of the calcrete reveals different vadose-phreatic phases previous to or coeval with the initial stages of river incision. Finally, we draw on the radiometric age obtained to estimate the incision rates of the present-day drainage network, while also addressing the possible causes behind the relatively high values calculated. A brief description of each technique or method used in this study will be provided in the appropriate context. GEOLOGICAL SETTING The area of study is located in the Betic Cordillera in southeastern Spain (Fig. 1), which represents a tectonically active region related to the collision between Africa and Iberia (DeMets et al., 1994; Morales et al., 1999; Galindo-Zaldívar et al., 1999, 2003). Despite this general compressional tectonic setting, the main tectonic and geomorphic features of the Betic Cordillera are related to extensional tectonics (e.g., Galindo-Zaldívar et al., 1989; García-Dueñas et al., 1992; Crespo-Blanc et al., 1994; Martínez-Martínez and Azañón, 1997; Martínez-Martínez et al., 2002). In this context, the present-day topography of the Betic Cordillera can be described as a succession of mountain ranges
225
and basins dissected by the main rivers, which incised both the ranges and the basins (Fig. 2A). The highest range in the Betic Cordillera is the Sierra Nevada, which has been recently interpreted as an elongated dome resulting from the interference of two orthogonal fold systems: one due to a rolling-hinge mechanism in the footwall of a WSW-directed extensional detachment and the other due to coeval N-S compression (Martínez-Martínez et al., 2004). The Sierra Nevada (and other neighboring ranges) emerged in middle Miocene times, progressively isolating different intramontane basins, such as the Granada and the GuadixBaza basins (Fig. 2A). Long-term uplift rates in this region are low to moderate (0.02–0.3 mm/yr) according to present-day altitudes of shallow-marine Miocene and Pliocene sediments (Braga et al., 2003; Silva et al., 2003; Booth-Rea et al., 2004; Sanz de Galdeano and Alfaro, 2004). The Guadix-Baza basin is one of the intramontane NeogeneQuaternary basins of the Betic Cordillera (Figs. 1 and 2), located in the central part of the orogen between the external (South Iberian margin) and the internal (Alboran domain) zones. The present-day topography of this basin corresponds to a depression bounded by ranges (Fig. 2). The continental infilling of this basin spans from the latest Tortonian to the Pleistocene (Vera, 1970; Peña, 1979; Viseras, 1991; Fernández et al., 1996). From a paleogeographical point of view, the Guadix-Baza basin can be viewed in Pliocene-Pleistocene times as an endorheic depression surrounded by mountains. The sedimentary record of PliocenePleistocene age suggests the existence at the marginal parts of the Guadix basin of alluvial systems, which flowed into a central lake (Viseras, 1991; Viseras and Fernández, 1992). In the eastern and southern borders of the Guadix subbasin (Guadix basin henceforth), the continental infill is represented by alternating poorly cemented conglomerates and sands of Pliocene-Pleistocene age
Figure 1. Geological setting of the Guadix-Baza basin in the Betic Cordillera (SE Spain).
A
B
Figure 2. (A) Topographic sketch of the eastern Betic Cordillera with the locations of the main basin and mountain ranges. (B) Digital elevation model (DEM) of the Guadix basin (see location in Figures 1 and 2A), where the main geomorphic features (elevated flat surface defined by the calcrete, badland areas, and main streams) can be observed.
Calcrete features and age estimates in southeast Spain (Viseras, 1991). The conglomeratic layers are dominant toward the upper part of the sequence and are capped by a 0.5–1-m-thick calcrete, which outcrops in a widespread area of the Guadix basin (Fig. 3). Toward the center of the basin, the conglomerates and sands grade laterally to lacustrine deposits represented by marls
227
and clays. These marly sediments are also capped by the calcrete layer. The uppermost outcropping layers of the lacustrine deposits in the Baza subbasin have been dated by amino acid racemization on ostracodes, yielding ages around 280 ka (Ortiz et al., 2004). The lacustrine layers dated in the Baza subbasin
A
B
Figure 3. (A) Oblique aerial view looking north of the surface defined by the calcrete into which canyon-shaped streams incise. Note the badlands in the upper part of the image (photograph by Javier Sanz de Galdeano). (B) Photograph of the Pliocene-Pleistocene stratigraphic sequence of the Guadix basin capped by the calcrete layer defining the flat surface.
228
Azañón et al.
occupy a stratigraphic position lower than the top calcrete layer. The calcrete and the associated geomorphic surface are less well preserved in the Baza subbasin, outcropping only at the margins, near the surrounding mountain ranges. GEOMORPHOLOGY OF THE GUADIX BASIN While forming a topographic depression, the Guadix basin has at present an external drainage to the Atlantic Ocean through the Guadalquivir River. The main mountain ranges surrounding the Guadix basin are the Sierra Nevada to the south, Sierra de Baza to the east, Sierra Cazorla to the north, and Sierra Arana to the west (Fig. 2A). The basin itself is an elevated (average altitude around 1000 m) plateau capped by the calcrete layer (Fig. 3). The flat geomorphic surface defined by the calcrete is strongly dissected by canyons and a main trunk river with a well-developed floodplain, the Fardes River (Fig. 2B). At present, most of the streams in this area have an ephemeral hydraulic regime with no discharge most of the time punctuated by episodic flooding events caused by heavy rains. This hydrology is controlled by the present-day climate in this region, which is semiarid, with an average annual rainfall between 300 and 350 mm. The only river with permanent discharge is the Fardes River. The flat surface defined by the calcrete is mostly horizontal (Fig. 3), except at the margins of the basin, where it inclines slightly basinward. This surface of regional extent represents the end of the sedimentation in the Guadix basin and was developed under a soil covering the underlying fluvial and lacustrine deposits (see next section). The calcrete layer formed prior to the present-day external drainage pattern, when the Guadix basin was still an endorheic catchment area. Thus, the calcrete marks a residual surface of an old flat area that lacked well-organized streams and extended throughout the entire Guadix basin. After the formation of the above-mentioned surface, the former Pleistocene endorheic Guadix basin must have been captured by the Guadalquivir River (Calvache and Viseras, 1997), thus starting the development of the present-day strongly entrenched drainage pattern. The capture was probably caused by headward erosion of the Guadalquivir River, favored by the topographically elevated position of the Guadix basin. Thus, the capture can be viewed, via a base-level lowering, as the triggering factor responsible for the formation of the present-day eroded landscape. Furthermore, the erosion would not have been a coeval process throughout the basin. Instead, once the Guadix basin was captured by the Guadalquivir River, an incision wave would have progressed headward along the basin, eventually reaching its southern margin. At the margins of the Guadix basin, the flat elevated surface marked by the calcrete appears dissected by a few narrow and rectilinear canyons (Fig. 3A), such as the Arroyo de Gor (Fig. 2B). Toward the north and northwest, the landscape is much more eroded and dominated by gullies and pipes (Vandekerckhove et al., 2000, 2003), with some buttes being the only remains of the flat geomorphic surface.
A second and lower flat surface corresponds to the present-day cultivated floodplain of the Fardes River, the main river draining the Guadix basin (Fig. 2B). This main axial valley does not run in a central position along the Guadix basin, but rather close to its western border. The Fardes River is the only one in the Guadix basin with terrace deposits at its margins. In one locality (Alicún de las Torres; Fig. 2B), three terrace levels made up of travertine deposits can be recognized at one margin of the Fardes River. In summary, three main geomorphic domains can be distinguished in the Guadix basin (Figs. 2 and 3): (1) the flat elevated surface cut by canyons; (2) the intermediate steep badland area; and (3) the lower surface of the Fardes floodplain. THE CALCRETE Profile and Micromorphology The calcrete constitutes the top of the Pliocene-Pleistocene sedimentary sequence, featuring a very continuous, but heterogeneous, layer along the Guadix basin. Up to three different calcrete layers can be observed, depending on the locality. The maximal thickness of each layer is around 1.5 m, and nonweathered clastic deposits are intercalated between the calcrete layers, as in the Aljibe Quebrado section (Fig. 4). This section is the most complete—it is composed of three layers of laminar calcrete up to 20 cm thick (Fig. 4). The calcrete layers are developed on top of
Figure 4. Aljibe Quebrado section showing the transition from palustrine deposits to the studied calcretes (at the top).
Calcrete features and age estimates in southeast Spain a thick gravel bed with intercalated red mudstones at the upper part. In other sections, such as in the Arroyo de Gor, the calcrete consists of a single layer around 1 m thick (Fig. 5), developed on brown silts with gravel clasts. In this locality, the single calcrete layer, in turn, includes three main horizons, which from bottom to top are: nodular, massive, and laminar.
229
At the macroscale, this three-horizon pattern is observed throughout the Guadix basin. The nodular horizon is ~30 cm thick and occurs at the base of the calcrete profiles. The host rocks are red mudstones in which spheroidal to cylindrical carbonate nodules are present. The nodules consist of homogeneous micrite with some floating sand grains, representing calcification
Figure 5. View of the top calcrete at Arroyo de Gor. Several indurated laminar horizons can be seen.
Figure 6. Hand sample of the uppermost part of the calcrete layer at the Arroyo de Gor. The laminae coat all the gravel bed. The indurated gravel constitutes, in this case, the massive horizon.
230
Azañón et al.
around roots. The massive horizon is decimeter-scale and consists of polygenic rock fragments (Fig. 6) incorporated in a dense and hard red micrite matrix. This matrix shows nonbiogenic features, such as desiccation cracks and floating etched detrital grains. Nevertheless, biogenic features are dominant and include root traces, calcified root cells, calcite spheres, vadose pisoliths, and micritic peloids. Moreover, coarse calcite mosaics, either as cement or as a result of recrystallization and displacement, are common. The laminar horizon occurs at the uppermost part of the profiles, in some cases constituting a sort of fine-grained detrital jacket around the uppermost part of the nodular horizon (Figs. 6 and 7). This horizon consists of alternating of light and dark laminae. The dark laminae contain more detrital grains and show alveolar septal structures as well as lines of calcite crystals, probably indicating calcified root structures (Alonso-Zarza et al., 1998b). The light laminae are richer in micrite and have
fewer detrital grains, although they have more clay (sepiolite or palygorskite) minerals. These light laminae also contain spherulites, calcified root spheres, and needle fiber calcite (Figs. 7 and 8). In those cases in which the laminar horizon envelops the massive one, there are also vadose-gravitational cements underlying the lowermost laminae. This fact probably indicates the progressive lowering of local-scale hanging water tables. The macro- and microfeatures described above are indicative of a pedogenic calcrete, where roots must have played an important role, as evidenced by the occurrence of calcified root traces and alveolar septal structures. The presence of spherulites may be taken to indicate that cyanobacterial mats (Verrecchia et al., 1995) developed at the top part of the profiles. The alternation of laminae with different proportions of detrital grains and biogenic features suggests successive small-scale periods of sedimentation, erosion, and soil formation in the uppermost part
Figure 7. View of the sample drilled for stable isotope characterization of the laminar horizon with locations of the different points of analysis. Massive (M) and laminar (L) horizons are easily separated. Massive horizons are very rich in etched detrital clasts. The different zones (I–V) within the laminar horizon are also indicated. See text for further explanations.
Calcrete features and age estimates in southeast Spain
231
Figure 8. Scanning electron microscope (SEM) view in which needle fiber calcite crystals and calcified filaments form most of the laminae with alveolar septal structure.
of a relatively stable surface. These sedimentation–erosion–soilforming periods might be related to climate-vegetation changes (Alonso-Zarza and Silva, 2002). The overall features of the calcrete indicate that it was formed due to the influence of a sparse vegetation cover of bushes and shrubs developed under a semiarid climate. This is the context deduced for the formation of similar calcretes in Spain (Alonso-Zarza et al., 1998a) and all over the world (Mack and James, 1994; Alonso-Zarza, 2003). Stable Isotope Geochemistry We performed a stable isotope study of 17 samples drilled from the uppermost centimeters of the laminar calcrete horizon (Figs. 7 and 9). The powder samples were baked under vacuum at 360 ºC for 30 min to remove any organic matter. The stable isotope analyses were performed at Cambridge University (UK), using a Micromass Multicarb Sample Preparation System attached to a VG Isotech PRIMS mass spectrometer. The isotope data are reported according to Vienna Peedee belemnite (VPDB) international standards. The precision of the results is better than ±0.06‰ for 12C/13C and ±0.08‰ for 16O/18O. The values of δ18O and δ13C (Figs. 9 and 10) vary from −9.17‰ to −6.28‰ and from −11.18‰ to −6.36‰ (VPDB), respectively. These values fall within the ranges described for calcretes by Alonso-Zarza (2003) in a recent and detailed review of the paleonvironmental significance of palustrine and pedogenic carbonates. As a whole, our results reveal a considerable variation of δ13C, greater than that of δ18O, which seems to be a common feature in calcretes (Alonso-Zarza, 2003; Alonso-Zarza and Arenas, 2004). Both δ18O
and δ13C show several fluctuations from the bottom to the top of the profile, although the general trend is toward heavier values upward (Fig. 9). Moreover, there is a strong positive correlation between δ13C and δ18O (r2 = 0.89; Fig. 10). The stable isotope composition of pedogenic carbonates has proved to be a powerful tool for paleoenvironmental studies (Cerling, 1984; Cerling and Quade, 1993; Alam et al., 1997) and has been used to reconstruct climate and vegetation changes through time (Ding and Yang, 2000; Fox and Koch, 2003, 2004; Alonso-Zarza and Arenas, 2004; Sanyal et al., 2004). Values of δ18O in calcretes depend on both the stable isotopic composition of soil water (Cerling, 1984; Cerling and Quade, 1993) and temperature. The δ18O of soil water, in turn, is related to the isotopic composition of local rainfall, which also is strongly controlled by temperature (Cerling and Quade, 1993). Additionally, evaporation in the uppermost horizons of the soil can also affect δ18O values in pedogenic carbonates and result in δ18O enrichment (Cerling and Quade, 1993). In the samples studied, the general δ18O trend toward heavier values higher in the profile (Fig. 9) might be taken to indicate a tendency toward aridity at the final stages of calcrete development. The values of δ18O obtained enable us to estimate the isotopic composition of the rain water during the initial stages of formation of the laminar calcrete. To do so, we have applied the equation proposed by Jiamao et al. (1997), which relates δ18O of pedogenic carbonate to δ18O of rainfall and also includes the effect of evaporation (Zanchetta et al., 2000). The resulting δ18O value for rain water at the time when the laminar calcrete started to develop is –10.12‰, i.e., a value 2.63‰ lower than that of
232
Azañón et al.
Figure 10. Relationship between δ18O and δ13C for the laminar calcrete.
Figure 9. δ18O and δ13C in the top laminar calcrete. The isotope values are referenced to the Pee Dee belemnite (PDB) standard. See location of the samples in Figure 7. The different zones (I–V) within the laminar horizon are also indicated.
present-day rain water (−7.49‰; Caballero et al., 1996). These lighter δ18O values in rain water could indicate that the climate was cooler than present during the formation of the calcrete. The δ13C of pedogenic carbonate depends on the isotopic composition of soil CO2 (Quade et al., 1989), which, in turn, is related to the composition of the local vegetation (Cerling, 1984). Thus, δ13C of pedogenic carbonate is controlled mainly by the ratio of C4/CAM (Crassulacean acid metabolism) to C3 plants. C3 plants (trees, most shrubs, and cool-season grasses) have lighter δ13C values (about −27‰) than C4 plants (about −12‰; Cerling and Quade, 1993). Consequently, when vegetation cover is dominated by C3 plants, δ13C in soil CO2 is lower
Figure 11. Calcrete isochron plots (2-dimensional versions) based on total dissolution of four coeval subsamples, with 1σ error crosses. The slopes in the (234U/232Th)-(238U/232Th) and (230Th/232Th)-(234U/232Th) diagrams represent the (234U/238U) and (230Th/234U) activity ratios, respectively, of the pure carbonate. MSWD—mean square of weighted deviates. 1d, 3b, 2b, and 3a are subsamples that have been used in the isochron calculation.
Calcrete features and age estimates in southeast Spain than when C4 plants dominate (Cerling, 1984; Alonso-Zarza, 2003). The respective distributions and abundances of C3 and C4 plants are controlled by climate. C4 plants are adapted to high water stress and elevated temperatures; their relative abundance has been used as an index of past aridity (Jiamao et al., 1997). In contrast, C3 plants prefer cool temperatures during the growing season and live at present at high latitudes (Alam et al., 1997; Ding and Yang, 2000). On these grounds, several authors have established that the δ13C values of pedogenic carbonate which formed at 25 ºC from pure C4 and C3 biomasses are approximately +2‰ and −12‰, respectively (Cerling, 1984; Alam et al., 1997). After comparing these values with δ13C in the calcrete studied here (−11.18‰ to −6.36‰), we conclude that C3 plants dominated the local ecosystem during the calcrete formation period. Similarly, the trend to higher δ13C values higher in the profile can be interpreted as a response to changing vegetation, with an increase in C4 plants due to a substantial increase in aridity. In the same way, the positive correlation between δ18O and δ13C (Fig. 10) can be related to increasing aridity during the final stages of calcrete formation. This is commonly true during the formation of pedogenic carbonates (Cerling, 1984). A comparison between the petrographic features and the stable isotope record in the laminar calcrete enables us to characterize five different zones, which are from bottom to top as follows (Figs. 7 and 9): (I) mostly laminar calcrete with alveolar structures, which have average δ18O and δ13C of −8.18‰ and −9.85‰, respectively; (II) a laminated zone characterized by the presence of laminae very rich in quartz silt, with average δ18O of −7.87‰ and δ13C of −9.03‰, where the quartz laminae correspond to the heaviest values in δ18O and δ13C within this zone; (III) massive micrite, with the heaviest isotope values in all of the profile; and (IV) and (V) two laminated zones, with average δ18O of −6.97‰ for 4 and −6.85‰ for 5. The δ13C values are −7.82‰ for 4 and −6.64‰ for 5. Zones I and II have the lighter values, indicating the development of a vegetation cover, probably under the least arid conditions in the entire profile. Quartz-rich laminae (with the heaviest isotopic composition) may indicate an increase of aridity and maybe an eolian dust influx. Zone III corresponds to the most arid conditions and probably the least biogenic control on the carbonate precipitated in the soil zone. The last two zones (IV and V) have similar δ18O values, thus pointing to a more stable climate, although the vegetation cover could be slightly different than in the lower part of the calcrete. In short, both micromorphology and stable isotope features are related to the climatic conditions prevailing during the formation of the different calcrete laminae. Generally, the massive micritic laminae with the heaviest isotope values represent the most arid conditions, which, in turn, indicate inhibited biogenic activity. Conversely, the occurrence of alveolar septal structures in zones with the lightest isotope values (I and II) is related to a wider vegetation cover under less arid climate.
233
U/Th Dating Four coeval subsamples of the very top laminar horizon of the calcrete were collected and dated by the U/Th isochron method (Bischoff and Fitzpatrick, 1991). The analytical basis for dating calcretes is the measurement of the 230Th formed by the decay of authigenic 234U and, indirectly, 238U, the uranium having been co-precipitated from solution with the carbonate (Kelly et al., 2000). Calcretes are, however, impure mixtures of calcium carbonate and incorporated detrital minerals, resulting in the same radionuclides being present in both the authigenic and detrital fractions. Our approach is to use an isochron method to determine the authigenic radionuclide component and, on the basis of this, its age (Bischoff and Fitzpatrick, 1991). The subsamples are assumed to be mixtures, in different proportions, of homogeneous detrital and authigenic carbonate end members. This method has been successfully applied to calcrete dating from alluvial terraces in the Sorbas basin (Kelly et al., 2000; Candy et al., 2004). The samples were cut with a diamond saw in order to remove the altered parts, then crushed and washed ultrasonically in deionized water. The fragments were checked with a stereoscopic microscope to identify and discard any recrystallized portions. About 20 g of each subsample were dissolved in an HCl-HNO3-HF mixture, following the standard “total dissolution” method outlined by Bischoff and Fitzpatrick (1991) and Luo and Ku (1991). After dissolution, isotopic complexes of uranium and thorium were extracted according to the procedure described by Edwards et al. (1987) and alphacounted using high-resolution ion implanted Ortec silicon surface barrier detectors at the Radiochemistry Laboratory of Roma Tre University (Italy). A standardized 232U-228Th tracer was used as a yield monitor, and a correction was made for the presence of detrital 228 Th and in-growth of 224Ra. 228Th/232Th equilibrium was checked on an unspiked subsample. The (230Th/234U) and (234U/238U) activity ratios of the authigenic fraction were calculated from the slope of a 3dimensional isochron fitted to the x-y-z data (238U/232Th, 230Th/232Th, 234 U/232Th) using the method of minimum likelihood estimation, outlined by Ludwig and Titterington (1994), in which the analytical data are weighed for analytical errors and error correlations. The corresponding age was determined from these authigenic fraction activity ratios using ISOPLOT, a plotting and regression program for radiogenic isotope data (Ludwig, 2003). The analytical data are given in Table 1, and the isochron diagrams are shown in Figure 11. The calculated age uncertainties are expressed as 1σ. The probability that the uncertainty in the age is due to the analytical errors alone is quantified through the calculation of the mean square of weighted deviates (MSWD). A high probability (or MSWD < 1; Fig. 11) indicates that, in this case, the uncertainty regarding the age is due only to analytical errors rather than to other sources of geological origin, e.g., subsamples are not coeval or mixture end members are not homogeneous. The resulting date is 42.6 ± 5.6 ka (Table 1), which indicates the age of the very top laminar part of the calcrete and thus gives a minimum time for the cessation of active sedimentation in the Guadix basin.
234
Azañón et al.
Subsample
TABLE 1. ANALYTICAL DATA FOR CALCRETE SUBSAMPLES, GUADIX, SPAIN 234 232 238 232 230 232 234 238 230 234 ( U/ Th) ( U/ Th) ( Th/ Th) ( U/ U)carb ( Th/ U)carb.
Age (ka)
1d 10.005 ± 0.867 8.641 ± 0.753 4.205 ± 0.383 2b 13.492 ± 1.577 10.912 ± 1.281 5.455 ± 0.630 1.358 ± 0.132 0.328 ± 0.036 42.6 ± 5.6 3a 14.647 ± 1.161 12.144 ± 0.971 5.684 ± 0.407 3b 12.229 ± 0.909 9.950 ± 0.748 4.776 ± 0.309 230 234 234 238 Note: Errors are quoted as 1V. ( Th/ U)carb and ( U/ U)carb are referenced to the pure carbonate authigenic fraction used in the calculation of the age.
INCISION RATES Estimating incision rates in fluvial environments is not an easy task due to the difficulties in establishing absolute ages of reference surfaces. Typically, local-scale, but not regionalscale, incision rates can be derived, since the processes causing entrenchment can be very variable throughout catchments through time and space. In this respect, it must be emphasized that river incision always progresses headward as a consequence of increasing stream power, which, in turn, can be due to baselevel lowering and/or profile steeping. Therefore, a single river can incise at different times along its different reaches, thus propagating an incision wave headward. Moreover, rock resistance can be quite variable along a river and can also affect local-scale incision rates. Taking into account the above drawbacks and using the age obtained for the calcrete as a reference, we have calculated incision and erosion rates for the late Pleistocene to present-day time span in the Guadix basin. These estimated rates can be considered as minimum values since the surface defined by the calcrete predates river entrenchment (see next section) and the process was not coeval throughout the basin, but probably progressed as an incision wave. We made calculations for the Arroyo de Gor, a stream with well-known geomorphologic features. This stream is a 30-km-long canyon highly incised (up to 200 m) into the Pliocene-Pleistocene infill (including the capping calcrete layer) of the eastern border of the Guadix basin (Figs. 2B and 12). This canyon is characterized by the absence of terrace deposits and by an abundance of large-scale rotational slides (Azañón et al., 2005). The present-day morphology of the Arroyo de Gor is the result of a combination of entrenchment, fracturing, and landsliding. The initial deep entrenchment of the stream is attributed to the base-level lowering related to the capture of the former endorheic Guadix basin by the Guadalquivir river in the late Pleistocene, i.e., after the formation of the calcrete layer at 42 ka. This river incision created a canyon with unstable subvertical walls, which, due to gravitational instability, give way to vertical open tension cracks at some distance from the canyon edge. The rotational slides are thought to have occurred during heavy rains by a combination of piping, which lengthened the tension cracks, and infiltration, thus reducing the shear strength along the subhorizontal lithological contact between conglomerates and underlying clays (Azañón et al., 2005). The rock volume remobilized by erosion in the Arroyo de Gor has been calculated from a digital elevation model (DEM)
with a resolution of 1 pixel per 20 m (Fig. 12). The canyon volume was estimated with the aid of ArcGis 8.2 by counting the number of pixels between a top level defined by the flat geomorphic surface formed by the calcrete and the topography. The resulting volume of rock remobilized by erosion in the Arroyo de Gor is 7972 m3 ha–1. With these data, the estimation of the erosion rate is 15.62 m3 ha–1 yr–1 or 28 t ha–1 yr–1 (assuming an average density of 1.8 t/m3 for the sedimentary infilling). Thus, the average minimum vertical incision rates in this canyon are around 4 mm/yr. Realistically, these average rates probably underestimate the actual values, since the Arroyo de Gor had almost reached its present morphology before 6 ka, as indicated by the presence of dolmens of that age built on the landslide bodies (Azañón et al., 2005). After the dolmens were built at around 6 ka, the canyon was 30–50 m into the landslide bodies, which yields a minimum Holocene vertical incision rate of roughly 5–7 mm/yr. This more recent rate is naturally higher than the average minimum rate calculated for the last 42 ka (around 4 mm/yr). Moreover, the Arroyo de Gor was developed by an initial vertical entrenchment of ~150 m that must have occurred prior to the large-scale landsliding, which, in turn, enlarged the initially very narrow canyon (Azañón et al., 2005). The age of the initial vertical entrenchment is unknown, having occurring sometime between the capture of the former endorheic Guadix basin by the Guadalquivir River and the large-scale landsliding, i.e., between 42 and 6 ka. With these observations in mind, we tentatively hypothesize that the river capture and the subsequent vertical entrenchment and landsliding could have occurred in a period between 38 and 28 ka, when several millennial-scale episodes (Is8 to Is3) of higher mean annual rainfall (up to 900 mm) occurred (Sánchez Goñi et al., 2002). Assuming that both the vertical entrenchment of 150 m and the large-scale landsliding were completed during this 10 k.y. period, the real rates of vertical incision for that period may have been as high as 15 mm/yr, i.e., 3–4 times higher than the average minimum rates. DISCUSSION The continental infill in the Guadix basin ended with the formation of a calcrete layer, which extends some hundreds of square kilometers and defines a flat elevated surface. Four coeval carbonate subsamples from the top laminae of the calcrete have been dated by the U/Th method, yielding an age of 42.6 ± 5.6 ka. This datum is in accordance with other ages obtained on the uppermost alluvial-lacustrine layers of the Guadix-Baza basin,
Calcrete features and age estimates in southeast Spain
235
Figure 12. Digital elevation model (DEM) of the Arroyo de Gor stream. Calculations of volume eroded have been performed on this DEM with the aid of ArcGis 8.2, using the geomorphic surface defined by the calcrete and the topography of the canyon as reference.
namely, an amino acid racemization age of 280 ka (Ortiz et al., 2004) and an age of 100 ka estimated for archaeological activity coeval to the most recent deposits (Botella et al., 1985, 1986, cited in Calvache and Viseras, 1997). These dates correspond to stratigraphic levels located below the calcrete layer investigated here. Therefore, the age of 42.6 ka is a more accurate estimation for the end of the sedimentation in the Guadix basin, since the calcrete layer is at the very top of the stratigraphic sequence. The regional extent of the calcrete and the lack of any observable relationship between lateral facies variation within the calcrete and the present-day stream distribution prove that the presentday drainage pattern formed later than calcrete formation, i.e., later than final sedimentation in the Guadix basin. Moreover, the Pliocene-Pleistocene sedimentary facies distribution shows that there is no spatial coincidence between present-day streams and Pliocene-Pleistocene paleorivers. The petrographic features and the stable isotope geochemistry of the calcrete indicate a pedogenic origin under a semiarid climate in which vegetation was sparse and dominated by bushes and shrubs. Variations in humidity and vegetation cover are indicated by changes in both micromorphology and isotopic
composition in the top laminar calcrete. In general, the betterlaminated zones that include more alveolar features are the isotopically lighter, both in carbon and oxygen, whereas the more massive zones are the heaviest. The results of the stable isotope study enable us to propose that during the initial stages of laminar calcrete formation, the climate was cooler than it is today, and aridity increased upward, when the sedimentation was ending. Some inferences can be made by considering the age obtained for the calcrete (42.6 ± 5.6 ka) in conjunction with the paleoclimatic conditions deduced from its petrographic and stable isotope geochemical features. In this regard, calcrete formation approximately coincides with the H5 and H4 Heinrich events (Dansgaard et al., 1993), which correspond to millennial-scale variations in atmospheric temperatures over Greenland. These events have been identified in the recent stratigraphic record of the Alboran Sea and have been related to sharp changes in surface water temperature (Pérez-Folgado et al., 2003), as well as to rapid vegetation shifts in Southern Iberia (Sánchez Goñi et al., 2002). According to these authors, H5 and H4 were characterized by a very arid climate with an average annual rainfall of ~300 mm and average winter temperatures 10 ºC cooler than present
236
Azañón et al.
day. We hypothesize that calcrete formation in the Guadix basin occurred during these extremely dry and cold climatic periods. Calcretes have been used as a basis for establishing relative landform chronologies in Quaternary alluvial sequences of the eastern Betic Cordillera (Harvey et al., 1995, 1999; Kelly et al., 2000). These chronologies, in turn, have been used to determine the rates of operation of geomorphic processes over the Quaternary period. However, a detailed calcrete micromorphological analysis must be carried out before using calcrete layers as chronomorphologic gauges (Candy et al., 2003). Particularly, Candy et al. (2003) highlighted that the pedogenic- or groundwater-linked character of the calcretes is of paramount importance for the geomorphic analysis. In this respect, the age derived for a typical pedogenic calcrete, such as the top layer of the Guadix basin, marks the end of the continental infill, when the basin was still endorheic and the present-day drainage pattern had not yet developed. In contrast, a groundwater calcrete would be much more ambiguous in terms of sedimentologic and geomorphic significance, because its formation would be related either to the final sedimentation or to the subsequent development of the presentday drainage pattern. Therefore, the calcrete layer studied here constrains both the age of the capture of the former endorheic Guadix basin by the Guadalquivir River and the minimal values of incision rates in this area. Using the age of the calcrete as a reference, we have calculated minimum incision and erosion rates for the late Pleistocene to present-day time span in the Guadix basin. In fact, estimations were performed for a canyon-shaped stream (the Arroyo de Gor) with well-known geomorphic features. The significance of the resulting incision rates, particularly whether they represent local (one single stream) or regional (the whole basin) values, must be discussed according to the geomorphic features of the Guadix basin. In this regard, we first highlight the homogeneous river incision throughout the Guadix basin, i.e., both the Fardes River and the Arroyo de Gor canyon have been incised to approximately the same depth in the flat surface defined by the calcrete. Nevertheless, the Fardes River attests to a more complicated evolution, including the formation of several terrace levels and a very intense lateral erosion with development of a badland landscape. Furthermore, the 30-km-long Arroyo de Gor itself, which represents half the length of the Guadix basin, is incised into a bedrock with homogeneous resistance and lacks important gradient variations. Thus, despite the fact that river incision would progress as a headward wave along the Arroyo the Gor, the values obtained, always considered minimums and subject to several uncertainties, can be viewed reasonably as regional-scale incision rates. Average minimum incision rates in the Arroyo de Gor stream are around 4 mm/yr. These values are relatively high (up to ten times higher) compared to available Pleistocene-Holocene incision rates, also minimum values, obtained in other NeogeneQuaternary basins of the Betic Cordillera: 0.1–0.4 mm/yr in the Sorbas basin (Mather and Harvey, 1995; Kelly et al., 2000), 0.3– 0.7 mm/yr in the Alpujarra Corridor (García et al., 2004), and 0.1–0.7 mm/yr in the Granada basin (Martín-Martín et al., 2001).
A number of factors, such as lithologies, stream piracy, climate, topography, and tectonics, can be invoked as being responsible for the high incision rates in the Guadix basin. A closer comparison with the other basins reduces these factors to three, namely, stream piracy, topography, and tectonics, since the lithologies and the climate are quite homogeneous in all the Neogene-Quaternary basins of the Betic Cordillera. Stream piracy, via the capture of a former endorheic basin by a river with a lower base level, namely sea level, can be assumed to have occurred during the Quaternary in all of the aforementioned basins. The timing of the capture, as well as the site where it took place, would notably influence the local-scale incision rates and the upstream progression of the incision wave. Nevertheless, drawing on the quite similar erosive state of these basins, one can reasonably consider that the late Quaternary incision waves have progressed headward comparably in all of them, reaching the surrounding mountain areas. Thus, at first glance, the high incision rates in the Guadix basin are probably related to the high average altitude (around 1000 m). This, in addition to the poorly indurated lithologies, would facilitate fast incision of the rivers during episodic heavy rains. In contrast, the Sorbas basin and the Alpujarra Corridor (Fig. 2A) have an average altitude of less than 400 m, which accounts for the low estimated incision rates. In the case of the Granada basin (Fig. 2A), which has an average altitude around 600 m, one would expect intermediate incision rates. The reason for the low values (quite similar to the ones of the Sorbas basin and the Alpujarra Corridor) cannot be justified solely on the basis of the difference in altitude between the Guadix and the Granada basins (see following discussion). Several issues must be discussed in regard to the possible contribution of tectonic activity to the high incision rates. First of all, the incision rates estimated for the Guadix basin are one or two orders of magnitude higher than the long-term regional uplift rates (0.02–0.3 mm/yr), calculated from the ages of marine deposits (Braga et al., 2003; Silva et al., 2003; BoothRea et al., 2004; Sanz de Galdeano and Alfaro, 2004). In the case of the Guadix basin, an uplift rate of 0.15 mm/yr, corresponding to the late Tortonian to present-day period, can be estimated according to the altitude of shallow-marine deposits of that age preserved at its southern border. The absence of marine deposits younger than the late Tortonian precludes the estimation of uplift rates for shorter time spans. Nevertheless, drawing on geological and seismological evidence, the possibility of very high uplift rates for the late Pleistocene to Holocene period can be reasonably discarded. In this respect, two salient aspects must be emphasized: (1) no faults with Pleistocene-Holocene activity are observed to affect the infill of the Guadix basin or its borders, and (2) both present-day and historical seismicity is concentrated to the south (Granada basin, Alpujarra Corridor) and east (Lorca basin, Fig. 1) of the Guadix basin (Morales et al., 1999; Mancilla et al., 2002; Muñoz et al., 2002; Serrano et al., 2002). Therefore, it can be concluded that the high late Pleistocene to Holocene incision rates in the Guadix basin do not represent a response to an accelerated
Calcrete features and age estimates in southeast Spain tectonic activity during the Quaternary. Nevertheless, the high average altitude of the Guadix basin is probably related to large-scale long-term isostatic uplift after the pre-Miocene crustal thickening in the Betic Cordillera. Interestingly, the only sector of the Betic Cordillera with a present-day crustal thickness exceeding 35 km is the Guadix basin and the Sierra Nevada (Banda et al., 1993). Therefore, we suggest that sustained isostatic uplift may be responsible for maintaining the high average altitude of the Guadix basin, while also facilitating the estimated high incision rates. We have argued before that the differences in elevation do not seem great enough to account for the considerably different Pleistocene-Holocene incision rates. At a greater scale, some differences in the tectonic activity could be the main reason for the different incision rates between the Guadix and the Granada basin. The Granada basin is bounded to the east by the western Sierra Nevada mountain front, which is thought to be one of the main active areas in all of the Betic Cordillera. This mountain front is marked by NW-SE–oriented, SW-dipping normal faults with Quaternary activity and present-day seismicity (Sanz de Galdeano et al., 2003; Azañón et al., 2004). The Granada basin is located in the hanging wall of this active normal-fault system; thus, it is undergoing Quaternary tectonic subsidence, which would preclude high rates of river incision. In contrast, the Guadix basin, together with the whole Sierra Nevada, constitutes a single block located in the footwall of this active normal-fault system, and thus it is subjected to tectonic and/or isostatic uplift. CONCLUSIONS The Guadix basin is a very special landscape when compared with other Neogene-Quaternary basins in the Betic Cordillera. It represents an elevated plateau strongly incised by the drainage pattern. Moreover, this landscape seems to be the result of a very fast incision process of the main rivers dissecting the geomorphic surface marked by the 42 ka calcrete layer. We envisage the capture of the Pliocene-Pleistocene endorheic Guadix basin by the Guadalquivir River during the late Pleistocene as the starting point for the formation of this landscape. This capture was probably a climatically driven process that must have occurred under wetter (and also possibly warmer) conditions than the present-day semiarid climate. Sánchez Goñi et al. (2002) described several millennial-scale episodes (Is8 to Is3) between 38 and 28 ka with a mean annual rainfall up to 900 mm during which the basin capture could have occurred. Undoubtedly, the capture process was facilitated by the high average altitude of the Guadix basin. Thus, the capture can be considered as an isostatically assisted process, since the high elevation of the Guadix basin seems to be in part the result of long-term large-scale isostatic uplift. After the capture, the combination of climatic, lithological, and topographic features would have facilitated the development of the present-day entrenched drainage pattern.
237
ACKNOWLEDGMENTS The research reported in this work has been financed by the Spanish Ministry of Education with Fondos Europeos de Desarrollo Regional funds of the European Union, through grants numbers REN2001-3378, CGL2004-03333/BTE, and CGL200404342/BTE. Comments and suggestions by Pablo Silva and an anonymous reviewer are kindly acknowledged. We thank Francisco Gonzálvez García for reviewing our English text. REFERENCES CITED Alam, M.S., Keppens, E., and Paepe, R., 1997, The use of oxygen and carbon isotope composition of pedogenic carbonates from Pleistocene palaeosols in NW Bangladesh, as palaeoclimatic indicators: Quaternary Science Reviews, v. 16, p. 161–168, doi: 10.1016/S0277-3791(96)00044-3. Alonso-Zarza, A.M., 2003, Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record: Earth-Science Reviews, v. 60, p. 261–298, doi: 10.1016/S0012-8252(02)00106-X. Alonso-Zarza, A.M., and Arenas, C., 2004, Cenozoic calcretes from the Teruel Graben, Spain: Microstructure, stable isotope geochemistry and environmental significance: Sedimentary Geology, v. 167, p. 91–108, doi: 10.1016/j.sedgeo.2004.02.001. Alonso-Zarza, A.M., and Silva, P.G., 2002, Quaternary laminar calcretes with bee nests: Evidences of small scale climatic fluctuations, eastern Canary Islands, Spain: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 178, p. 119–135, doi: 10.1016/S0031-0182(01)00405-9. Alonso-Zarza, A.M., Silva, P., Goy, J.L., and Zazo, C., 1998a, Fan-surface dynamics and biogenic calcrete development: Interactions during ultimate phases of fan evolution in the semiarid SE Spain (Murcia): Geomorphology, v. 24, p. 147–167, doi: 10.1016/S0169-555X(98)00022-1. Alonso-Zarza, A.M., Sanz, M.E., Calvo, J.P., and Estévez, P., 1998b, Calcified root cells in Miocene pedogenic carbonates of the Madrid Basin: Evidence for the origin of Microcodium b: Sedimentary Geology, v. 116, p. 81–97, doi: 10.1016/S0037-0738(97)00077-8. Arakel, A.V., 1986, Evolution of calcrete in palaeodrainages of the Lake Narpperby area, Central Australia: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 54, p. 283–303, doi: 10.1016/0031-0182(86)90129-X. Azañón, J.M., Azor, A., Booth-Rea, G., and Torcal, F., 2004, Small-scale faulting, topographic steps and seismic ruptures in the Alhambra (Granada, SE Spain): Journal of Quaternary Science, v. 19, p. 219–227, doi: 10.1002/ jqs.838. Azañón, J.M., Azor, A., Pérez-Peña, J.V., and Carrillo, J.M., 2005, Late Quaternary large-scale rotational slides induced by river incision: The Arroyo de Gor area (Guadix basin, SE Spain): Geomorphology v. 69, p. 152–168. Banda, E., Gallart, J., García-Dueñas, V., Dañobeitia, J.J., and Makris, J., 1993, Lateral variation of the crust in the Iberian Peninsula: New evidence from the Betic Cordillera: Tectonophysics, v. 221, p. 53–66, doi: 10.1016/00401951(93)90027-H. Bischoff, J.L., and Fitzpatrick, J.A., 1991, U-series dating of impure carbonates: An isochron technique using total sample dissolution: Geochimica et Cosmochimica Acta, v. 55, p. 543–554, doi: 10.1016/00167037(91)90011-S. Booth-Rea, G., Azañón, J.M., Azor, A., and García-Dueñas, V., 2004, Influence of strike-slip fault segmentation on drainage evolution and topography: A case study: The Palomares fault zone (southeastern Betics, Spain): Journal of Structural Geology, v. 26, p. 1615–1632, doi: 10.1016/ j.jsg.2004.01.007. Botella, M., Martínez, C., and Cárdenas, F.J., 1985, Las industrias paleolíticas de Cueva Horá (Darro, Granada): Antropología y Paleoecología Humana, v. 1, p. 59–74. Botella, M., Martínez, C., Cárdenas, F.J., and Cañabate, M.J., 1986, Industria musteriense y achelense de Cueva Horá (Darro, Granada): Book in Honour of Luis Siret: Granada, Spain, Junta de Andalucía, p. 79–95. Braga, J.C., Martín, J.M., and Quesada, C., 2003, Patterns and average rates of late Neogene–Recent uplift of the Betic Cordillera: SE Spain: Geomorphology, v. 50, p. 3–26, doi: 10.1016/S0169-555X(02)00205-2. Caballero, E., Jiménez de Cisnero, C., and Reyes, E., 1996, A stable isotope
238
Azañón et al.
study of cave seepage waters: Applied Geochemistry, v. 11, p. 583–587, doi: 10.1016/0883-2927(96)00026-1. Calvache, M.L., and Viseras, C., 1997, Long-term control mechanisms of stream piracy processes in southeast Spain: Earth Surface Processes and Landforms, v. 22, p. 93–105, doi: 10.1002/(SICI)1096-9837(199702)22:2<93:: AID-ESP673>3.0.CO;2-W. Candy, I., Black, S., Sellwood, B.W., and Rowan, J.S., 2003, Calcrete profile development in Quaternary alluvial sequences, southeast Spain: Implications for using calcretes as a basis for landform chronologies: Earth Surface Processes and Landforms, v. 28, p. 169–185, doi: 10.1002/esp.445. Candy, I., Black, S., and Sellwood, B.W., 2004, Quantifying time scales of pedogenic calcrete formation using U-series disequilibria: Sedimentary Geology, v. 170, p. 177–187, doi: 10.1016/j.sedgeo.2004.07.003. Cerling, T.E., 1984, The stable isotopic composition of modern soil carbonate and its relationship to climate: Earth and Planetary Science Letters, v. 71, p. 229–240, doi: 10.1016/0012-821X(84)90089-X. Cerling, T.E., and Quade, J., 1993, Stable carbon and oxygen isotopes in soil carbonates, in Swart, P.K., Lohmann, K.C., Mckenzie, J.A., and Savin, F., eds., Climate Change in Continental Isotopic Records: Washington, D.C., American Geophysical Union Geophysical Monograph 78, p. 217–231. Crespo-Blanc, A., Orozco, M., and García-Dueñas, V., 1994, Extension versus compression during the Miocene tectonic evolution of the Betic chain: Late folding of normal fault systems: Tectonics, v. 13, p. 79–88, doi: 10.1029/93TC02231. Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, D., Gunderstrup, N.S., Hammer, C.V., Hvidberg, C.S., Steffensen, J.P., Sveinbjornsdottin, A.E., Jouzel, J., and Bard, G., 1993, Evidence for general instability of past climate from 250 kyr ice-core record: Nature, v. 364, p. 218–220, doi: 10.1038/364218a0. DeMets, C., Gordon, R.G., Argus, D.F., and Stein, S., 1994, Effect of recent revisions to the geomagnetic reversal time scale on estimates of current plate motions: Geophysical Research Letters, v. 21, p. 2191–2194, doi: 10.1029/94GL02118. Ding, Z.L., and Yang, S.L., 2000, C3/C4 vegetation evolution over the last 7.0 Myr in the Chinese Loess Plateau: Evidence from pedogenic carbonate δ13C: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 160, p. 291–299, doi: 10.1016/S0031-0182(00)00076-6. Dumas, B., 1969, Glacis et croutes calcaires dans le levant Espagnol: Bulletin de l’Association de Geographes Français, v. 375–376, p. 553–561. Edwards, R.L., Chen, J.H., and Wasserburg, G.J., 1987, 238U-234U-230Th-232Th systematics and the precise measurement of time over the past 500,000 years: Earth and Planetary Science Letters, v. 81, p. 175–192, doi: 10.1016/0012-821X(87)90154-3. Fernández, J., Viseras, C., and Soria, J.M., 1996, Stratigraphic architecture of the Neogene basins in the central sector of the Betic Cordillera (Spain): Tectonic control and base-level changes, in Friend, P.F., and Dabrio, C., eds., Tertiary Basins of Spain: The Stratigraphic Record of Crustal Kinematics: Cambridge, UK, Cambridge University Press, p. 353–365. Fox, D.L., and Koch, P.L., 2003, Tertiary history of C4 biomass in the Great Plains, USA: Geology, v. 31, p. 809–812, doi: 10.1130/G19580.1. Fox, D.L., and Koch, P.L., 2004, Carbon and oxygen isotopic variability in Neogene paleosol carbonates: Constraints on the evolution of the C4-grasslands of the Great Plains, USA: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 207, p. 305–329, doi: 10.1016/S00310182(04)00045-8. Galindo-Zaldívar, J., González-Lodeiro, F., and Jabaloy, A., 1989, Progressive extensional shear structures in a detachment contact in the western Sierra Nevada (Betic Cordilleras, Spain): Geodinamica Acta, v. 3, p. 73–85. Galindo-Zaldívar, J., Jabaloy, A., Serrano, I., Morales, J., González-Lodeiro, F., and Torcal, F., 1999, Recent and present-day stresses in the Granada basin (Betic Cordilleras): Example of a late Miocene–present-day extensional basin in a convergent plate boundary: Tectonics, v. 18, p. 686–702, doi: 10.1029/1999TC900016. Galindo-Zaldívar, J., Gil, A.J., Borque, M.J., González Lodeiro, F., Jabaloy, A., Marín Lechado, C., Ruano, P., and Sanz de Galdeano, C., 2003, Active faulting in the internal zones of the central Betic Cordilleras (SE, Spain): Journal of Geodynamics, v. 36, p. 239–250, doi: 10.1016/S02643707(03)00049-8. García, A.F., Zhu, Z., Ku, T.L., Sanz de Galdeano, C., Chadwick, O.A., and Chacón Montero, J., 2003, Tectonically driven landscape development within the eastern Alpujarran Corridor, Betic Cordillera, SE Spain (Almería): Geomorphology, v. 50, p. 83–110, doi: 10.1016/S0169555X(02)00209-X.
García, A.F., Zhu, Z., Ku, T.L., Chadwick, O.A., and Chacón Montero, J., 2004, An incision wave in the geologic record, Alpujarran Corridor, southern Spain (Almería): Geomorphology, v. 60, p. 37–72, doi: 10.1016/ j.geomorph.2003.07.012. García-Dueñas, V., Balanyá, J.C., and Martínez-Martínez, J.M., 1992, Miocene extensional detachments in the outcropping basement of the northern Alboran basin (Betics) and their tectonic implications: Geo-Marine Letters, v. 12, p. 88–95, doi: 10.1007/BF02084917. Harvey, A.M., Miller, S.Y., and Wells, S.G., 1995, Quaternary soil and river terrace sequences in the Aguas/Feos River systems: Sorbas basin, southeast Spain, in Lewin, J., Macklin, M.G., and Woodward, J.C., eds., Mediterranean Quaternary River Environments: Rotterdam, Balkema, p. 263–281. Harvey, A.M., Goy, J.L., Stokes, M., Zazo, C., Silva, P.G., and Mather, A.E., 1999, The impact of Quaternary sea-level and climatic change on coastal alluvial fans in the Cabo de Gata ranges, southeast Spain: Geomorphology, v. 28, p. 1–22, doi: 10.1016/S0169-555X(98)00100-7. Jiamao, H., Keppens, E., Tungsheng, L., Paepe, R., and Wenying, J., 1997, Stable isotope composition of the carbonate concretion in loess and climate change: Quaternary International, v. 37, p. 37–43, doi: 10.1016/10406182(96)00005-5. Jiménez-Espinosa, R., and Jiménez-Millán, J., 2003, Calcrete development in Mediterranean colluvial carbonates systems from SE Spain: Journal of Arid Environments, v. 53, p. 479–489, doi: 10.1006/jare.2002.1061. Kelly, M., Black, S., and Rowan, J.S., 2000, A calcrete-based U/Th chronology for landform evolution in the Sorbas basin, southeast Spain: Quaternary Science Reviews, v. 19, p. 995–1010, doi: 10.1016/S02773791(99)00050-5. Ludwig, K.R., 2003, Mathematical-statistical treatment of data and errors for 230 Th/U geochronology, in Bourdon, B., Henderson, G.M., Lundstrom, C.C., and Turner, S.P., eds., Uranium-Series Geochemistry: Reviews in Mineralogy and Geochemistry, v. 52, p. 631–656. Ludwig, K.R., and Titterington, D.M., 1994, Calculation of 230Th/U isochrons, ages, and errors: Geochimica et Cosmochimica Acta, v. 58, p. 5031–5042, doi: 10.1016/0016-7037(94)90229-1. Luo, S., and Ku, T.L., 1991, U-series dating: A generalised method employing total sample dissolution: Geochimica et Cosmochimica Acta, v. 55, p. 555–564, doi: 10.1016/0016-7037(91)90012-T. Machette, M.N., 1985, Calcic soils of southwestern United States, in Weide, D.L., ed., Soil and Quaternary Geology of the Southwestern United States: Geological Society of America Special Paper 203, p. 1–21. Mack, G.H., and James, W.C., 1994, Paleoclimate and the global distribution of paleosols: The Journal of Geology, v. 102, p. 360–366. Mancilla, F.L., Ammona, C.J., Herrmann, R.B., and Morales, J., 2002, Faulting parameters of the 1999 Mula earthquake, southeastern Spain: Tectonophysics, v. 354, p. 139–155, doi: 10.1016/S0040-1951(02)00340-2. Martín-Martín, M., Andreo, B., Martín-Algarra, A., and Julià, R., 2001, Travertinos del borde noreste de la Depresión de Granada, in Sanz de Galdeano, C., Peláez-Montilla, J.A., and López-Garrido, A.C., eds., La Cuenca de Granda: Estructura, Tectónica Activa, Geomorfología y Dataciones Existentes: Granada, Editorial Universidad de Granada, p. 22–28. Martínez-Martínez, J.M., and Azañón, J.M., 1997, Mode of extensional tectonics in the southeastern Betics (SE Spain). Implications for the tectonic evolution of the peri-Alborán orogenic system: Tectonics, v. 16, p. 205– 225, doi: 10.1029/97TC00157. Martínez-Martínez, J.M., Soto, J.I., and Balanyá, J.C., 2002, Orthogonal folding of extensional detachments: Structure and origin of the Sierra Nevada elongated dome (Betics, SE Spain): Tectonics, v. 21, 1012, doi: 10.1029/2001TC001283. Martínez-Martínez, J.M., Soto, J.I., and Balanyá, J.C., 2004, Elongated domes in extended orogens: A mode of mountain uplift in the Betics (southeast Spain), in Whitney, D.L., Teyssier, C., and Siddoway, C.S., eds., Gneiss Domes in Orogeny: Geological Society of America Special Paper 380, p. 243–266. Mather, A.E., and Harvey, A.M., 1995, Controls on drainage evolution in the Sorbas basin, southeast Spain, in Lewin, J., Macklin, M.G., and Woodward, J.C., eds., Mediterranean Quaternary River Environments: Rotterdam, Balkema, p. 65–76. Morales, J., Serrano, I., Jabaloy, A., Galindo-Zaldívar, J., Zhao, D., Torcal, F., Vidal, F., and González-Lodeiro, F., 1999, Active continental subduction beneath the Betic Cordillera and the Alborán Sea: Geology, v. 27, p. 735– 738, doi: 10.1130/0091-7613(1999)027<0735:ACSBTB>2.3.CO;2. Muñoz, D., Cisternas, A., Udías, A., Mezcua, J., Sanz de Galdeano, C., Morales, J., Sánchez-Venero, M., Haessler, H., Ibañez, J., Buforn, E., Pascual,
Calcrete features and age estimates in southeast Spain G., and Rivera, L., 2002, Microseismicity and tectonics in the Granada Basin (Spain): Tectonophysics, v. 356, p. 233–252, doi: 10.1016/S00401951(02)00338-4. Nash, D.J., and McLaren, S.J., 2003, Kalahari valley calcretes: Their nature, origins, and environmental significance: Quaternary International, v. 111, p. 3–22, doi: 10.1016/S1040-6182(03)00011-9. Nash, D.J., and Smith, R.F., 2003, Properties and development of channel calcretes in a mountain catchment, Tabernas Basin, southeast Spain: Geomorphology, v. 50, p. 227–250, doi: 10.1016/S0169-555X(02)00216-7. Ortiz, J.E., Torres, T., Julià, R., Delgado, A., Lamas, F.J., Soler, V., and Delgado, J., 2004, Numerical dating algorithms of amino acid racemization ratios from continental ostracodes: Application to the Guadix-Baza basin (southern Spain): Quaternary Science Reviews, v. 23, p. 717–730, doi: 10.1016/j.quascirev.2003.06.001. Peña, J., 1979, La depresión de Guadix-Baza: Estratigrafía del Plioceno-Pleistoceno [Ph.D. thesis]: Granada, Universidad de Granada, 160 p. Pérez-Folgado, M., Sierro, F.J., Flores, J.A., Cacho, I., Grimalt, J.O., Zahn, R., and Shackleton, N.J., 2003, Western Mediterranean planktonic foraminifera events and millennial variability during the last 70 kyr: Marine Micropaleontology, v. 48, p. 49–70, doi: 10.1016/S0377-8398(02)00160-3. Quade, J., Cerling, T.E., and Bowman, J.R., 1989, Systematic variations in the carbon and oxygen isotopic composition of pedogenic carbonates along elevation transects in the southern Great Basin, United States: Geological Society of America Bulletin, v. 101, p. 464–475, doi: 10.1130/00167606(1989)101<0464:SVITCA>2.3.CO;2. Sánchez Goñi, M.F., Cacho, I., Turon, J.L., Guiot, J., Sierro, F.J., Peypouquet, J.P., Grimalt, J.O., and Shackleton, N.J., 2002, Synchroneity between marine and terrestrial responses to millennial scale climatic variability during the last glacial period in the Mediterranean region: Climate Dynamics, v. 19, p. 95–105, doi: 10.1007/s00382-001-0212-x. Sancho, C., and Meléndez, A., 1992, Génesis y significado ambiental de los caliches Pleistocenos de la región del Cinca (Depresión del Ebro): Revista de la Sociedad Geológica de España, v. 5, p. 81–93. Sancho, C., Peña, J.L., Lewis, C., McDonald, E., and Rodes, E., 2004, Registros fluviales y glaciares Cuaternarios en las cuencas de los Ríos Cinca y Gállego (Pirineos y depresión del Ebro): Geo-Guías, v. 1, p. 181–205. Sanyal, P., Bhattacharya, S.K., Kumar, R., Ghosh, S.K., and Sangode, S.J., 2004, Mio-Pliocene monsoonal record from Himalayan foreland basin (Indian Siwalik) and its relation to vegetational change: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 205, p. 23–41, doi: 10.1016/ j.palaeo.2003.11.013. Sanz de Galdeano, C., and Alfaro, P., 2004, Tectonic significance of the present relief of the Betic Cordillera: Geomorphology, v. 63, p. 175–190, doi: 10.1016/j.geomorph.2004.04.002.
239
Sanz de Galdeano, C., Peláez-Montilla, J.A., and López-Casado, C., 2003, Seismic potential of the main active faults in the Granada Basin (southern Spain): Pure and Applied Geophysics, v. 160, p. 1537–1556, doi: 10.1007/s00024-003-2359-3. Serrano, I., Zhao, D., and Morales, J., 2002, 3-D crustal structure of the extensional Granada basin in the convergent boundary between the Eurasian and African plates: Tectonophysics, v. 344, p. 61–79, doi: 10.1016/S00401951(01)00201-3. Silva, P.G., Goy, J.L., Zazo, C., and Bardají, T., 2003, Fault-generated mountain fronts in southeast Spain: Geomorphologic assessment of tectonic and seismic activity: Geomorphology, v. 50, p. 203–225, doi: 10.1016/S0169555X(02)00215-5. Vandekerckhove, L., Poesen, J., Oostwoud Wijdenes, D., Gyssels, G., Beuselinck, L., and de Luna, E., 2000, Characteristics and controlling factors of bank gullies in two semi-arid Mediterranean environments: Geomorphology, v. 33, p. 37–58, doi: 10.1016/S0169-555X(99)00109-9. Vandekerckhove, L., Poesen, J., and Govers, G., 2003, Medium-term gully headcut retreat rates in southeast Spain determined from aerial photographs and ground measurements: CATENA, v. 50, p. 329–352, doi: 10.1016/S0341-8162(02)00132-7. Vera, J.A., 1970, Estudio estratigráfico de la depresión de Guadix-Baza: Boletín Geológico y Minero, v. 81, p. 429–462. Verrecchia, E.P., Freytet, P., Verrecchia, K.E., and Dumont, J.L., 1995, Spherulites in calcrete laminar crusts: Biogenic CaCO3 precipitation as a major contributor to crust formation: Journal of Sedimentary Research, v. 65, p. 690–700. Viseras, C., 1991, Estratigrafía y sedimentología del relleno aluvial de la Cuenca de Guadix (Cordilleras Béticas) [Ph.D. thesis]: Granada, Universidad de Granada, 327 p. Viseras, C., and Fernández, J., 1992, Sedimentary basin destruction inferred from the evolution of drainage systems in the Betic Cordillera, southern Spain: Journal of the Geological Society of London, v. 149, p. 1021–1029. Watts, N.L., 1980, Quaternary pedogenic calcretes from the Kalahari (southern Africa): Mineralogy, genesis and diagenesis: Sedimentology, v. 27, p. 661–686. Zanchetta, G., Di Vito, M., Fallick, A.E., and Sulpizio, R., 2000, Stable isotopes of pedogenic carbonates from the Somma-Vesuvius area, southern Italy, over the past 18 kyr: Palaeoclimatic implications: Journal of Quaternary Science, v. 15, p. 813–824, doi: 10.1002/1099-1417(200012)15:8<813:: AID-JQS566>3.0.CO;2-Z. MANUSCRIPT ACCEPTED BY THE SOCIETY 17 MAY 2006
Printed in the USA
Contents Preface Ancient Landscapes, Climate and Sequence Boundaries
1.
Calcic pedocomplexes-Regional sequence boundary indicators in Tertiary deposits of the Great Plains and western United States
D.L. Hanneman and C.J. Wideman 2.
A Late Triassic soil catena: Landscape and climate controls on paleosol morphology and chemistry across the Carnian-age /schigua/asto-Vil/a Union basin, northwestern Argentina
N.J. Tabor, I.P. Montanez, K.A. Kelso, B. Currie, T. Shipma n, and C. Colombi 3.
Investigating paleosol completeness and preservation in mid-Paleozoic alluvial paleosols: A case study in paleosol taphonomy from the Lower Old Red Sandstone
S.B. Marriott and V.P. Wright 4.
Calcareous paleosols of the Upper Triassic Chinle Group, Four Corners region, southwestern United States: Climatic implications
9.
The Upper Triassic crenogenic limestones in Upper Si/esia (southern Poland) and their paleoenvironmental context
J. Szulc, M. Gradzifl ski, A. Lewandowska, and C. Heu nisch 10. A recent analogue for palustrine carbonate environments: The Quaternary deposits of Las Tab/as de Oaimie/ wetlands, Ciudad Real, Spain
A.M. Alonso-Za rza, M. Dorado-Valifi o, A. Valdeo lmillos-Rodrfguez, an d M. Blan ca Ruiz-Zapata 11. Depositional conditions of carbonate-dominated palustrine sedimentation around the K-T boundary (Facies Rognacien, northeastern Pyrenean foreland, southwestern France)
D. Marty and C.A. Meyer 12. ReworkedMicrocodium calcarenites interbedded in pelagic sedimentary rocks (Paleocene, Subbetic, southern Spain): Paleoenvironmental reconstruction
J.M . Molina, J.A. Vera, and R. Aguad o Dating of Calcretes: Applications
L.H. Tanner and S.G. Lucas 5.
Estimates of atmospheric C02 levels during the mid-Turonian derived from stable isotope composition of paleosol calcite from Israel
13. Calcite cement stratigraphy of a nonpedogenic calcrete in the Triassic New Haven Arkose (Newark Supergroup)
E.T. Rasbury, E. H. Gierlowski-Kordesch, J. M. Cole, C. Sookdeo, G. Spat aro, an d J. Nienst edt
A. Sandier 6.
Pedogenic carbonate distribution within glacial till in Taylor Valley, Southern Victoria Land, Antarctica
K.K. Fol ey, W.B . Lyons, J.E. Barrett, and R.A. Virgin ia Sedimentary Environments and Facies
7.
14. Calcrete features and age estimates from U/Th dating: Implications for the analysis of Quaternary erosion rates in the northern limb ofthe Sierra Nevada range (Betic Cordillera, southeast Spain)
J.M . Azafi6n, P. Tucc ime i, A. Azo r, I.M. Sanchez-Aimazo, A.M. Alon so-Zarza, M. Solig o, an d J.V. Perez- Pefia
Calcretes, oncolites, and lacustrine limestones in Upper Oligocene alluvial fans of the Montgat area (Catalan Coastal Ranges, Spain)
D. Parc erisa, D. Gom ez-Gras, and J.D. Martfn -Martfn 8.
The role of clastic sediment influx in the formation of calcrete and palustrine facies: A response to paleographic and climatic conditions in the southeastern Tertiary Ouero basin (northern Spain)
I. Arm enteros and P. Huerta
~THE
.
GEOLOGICAL SOCIETY OF AMERICA
3300 Penrose Pl ace • P.O. Box 9140 • Bould er, CO 80301 -9140, US A
ISBN-100-8137-2416-3 ISBN-13978-0-81 37-2416-4