From the Shield to the Sea: Geological Field Trips from the 2011 Joint Meeting of the GSA Northeastern and North-Central Sections edited by Richard M. Ruffolo and Charles N. Ciampaglio
Field Guide 20 THE GEOLOGICAL SOCIETY OF AMERICA®
From the Shield to the Sea: Geological Field Trips from the 2011 Joint Meeting of the GSA Northeastern and North-Central Sections
edited by
Richard M. Ruffolo GAI Consultants, Inc. Pittsburgh Office 385 East Waterfront Drive Homestead, Pennsylvania 15120-5005 USA Charles N. Ciampaglio Wright State University–Lake Campus 7600 Lake Campus Drive Celina, Ohio 45822 USA
Field Guide 20 3300 Penrose Place, P.O. Box 9140
Boulder, Colorado 80301-9140, USA
2011
Copyright © 2011, The Geological Society of America (GSA), Inc. 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. In addition, an author has the right to use his or her article or a portion of the article in a thesis or dissertation without requesting permission from GSA, provided the bibliographic citation and the GSA copyright credit line are given on the appropriate pages. 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. GSA provides this and other forums for the presentation of diverse opinions and positions by scientists worldwide, regardless of their race, citizenship, gender, religion, sexual orientation, or political viewpoint. Opinions presented in this publication do not reflect official positions of the Society. 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. Cataloging-in-Publication data for this volume is available from the Library of Congress. ISBN: 9780813700205 Cover, front: Fifteen-foot-high foreset beds, overlain by thin, dark-brown topset bed in a kame delta near West Liberty, Butler County, Pennsylvania. The topset-foreset sequence is the lower of two sets exposed vertically in the exposure. The other set is on the bench above the topset bed. The delta was deposited in an ice-dammed, pro-glacial lake at the end of the Jacksville Esker. The exposure in the Glacial Sand and Gravel Co. Mine 31 is part of Stop 1 of Field Trip 5 (see Fleeger, G.M., et al., “Quaternary geology of northwestern Pennsylvania,” p. 87). Bill Bragonier (formerly of the Pennsylvania Geological Survey; currently with Rosebud Mining Co.) for scale (6′ 4″; 1.98 m). Photograph by Gary M. Fleeger, Pennsylvania Geological Survey, July 2009. Back: Group examining a Late Devonian paleontological site in the Catskill Formation along Interstate 99 in Lycoming County, Pennsylvania. (Field Trip 1; see Daeschler, E.B., and Cressler, W.L., III, “Late Devonian paleontology and paleoenvironments at Red Hill and other fossil sites in the Catskill Formation of north-central Pennsylvania,” p. 1.) Photo by Fred Mullison, Academy of Natural Sciences, Philadelphia.
10 9 8 7 6 5 4 3 2 1
Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v 1. Late Devonian paleontology and paleoenvironments at Red Hill and other fossil sites in the Catskill Formation of north-central Pennsylvania . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Edward B. Daeschler and Walter L. Cressler III 2. An introduction to structures and stratigraphy in the proximal portion of the Middle Devonian Marcellus and Burket/Geneseo black shales in the Central Appalachian Valley and Ridge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Terry Engelder, Rudy Slingerland, Michael Arthur, Gary Lash, Daniel Kohl, and D.P. Gold 3. Pennsylvanian climatic events and their congruent biotic responses in the central Appalachian Basin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 David K. Brezinski and Albert D. Kollar 4. Landslides in the vicinity of Pittsburgh, Pennsylvania . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Richard E. Gray, James V. Hamel, and William R. Adams Jr. 5. Quaternary geology of northwestern Pennsylvania . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Gary M. Fleeger, Todd Grote, Eric Straffin, and John P. Szabo 6. The history and geology of the Allegheny Portage Railroad, Blair and Cambria Counties, Pennsylvania . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 John A. Harper 7. Early industrial geology of western Pennsylvania and eastern Ohio: Early gristmills and iron furnaces west of the Alleghenies and their geologic contexts . . . . . . . . . . . . . . . . . . . . . . . . 143 Joseph T. Hannibal, Tammie L. Gerke, Mary K. McGuire, Harry M. Edenborn, Ann L. Holstein, and David Parker 8. The old, the crude, and the muddy: Oil history in western Pennsylvania . . . . . . . . . . . . . . . . . . 169 Kristin M. Carter and Kathy J. Flaherty
iii
v
Preface From the Shield to the Sea features field guides and descriptions of eight of the geological field trips offered during the Joint Meeting of the Geological Society of America (GSA) Northeastern and NorthCentral Sections held in Pittsburgh, Pennsylvania, in March 2011. The trips described herein are not only timely and topical, but also highlight the region’s geology from eastern Ohio to the Central Appalachian Valley and Ridge. The regional geology and forces that have shaped the topography of the area are highlighted by examining landslides (Gray et al.), glacial sedimentation and stratigraphy (Fleeger et al.), and examination of the structure and stratigraphy of the prevalent Middle Devonian black shales (Engelder et al.). The geologic history of the area from a biotic perspective are covered by Daeschler and Cressler with the examination of sites that illustrate the evolution of early terrestrial vertebrates and terrestrial ecosystems from the Upper Devonian Catskill Formation, and Brezinski and Kollar’s examination of how Pennsylvanian climate changes impacted the biota of the region. In addition, three of the trip descriptions illustrate how regional geology has helped shape the industrial history of our region: Carter and Flaherty examine western Pennsylvania’s rich history of oil and gas exploration; Harper provides an opportunity to see the immense effort needed to take the Pennsylvania Mainline Canal system over the imposing Allegheny front; and Hannibal et al.’s trip visits early gristmills and iron furnaces. The editors would like to thank all the authors, trip leaders, and organizers for their hard work and time devoted to this guidebook. Special thanks to all the reviewers who helped improve the field guide. We would also like to thank the staff in the GSA Publications Department for their tremendous job of guiding us through the production process. The editors hope readers will enjoy this field guide and find its contents valuable. Richard M. Ruffolo and Charles N. Ciampaglio
The Geological Society of America Field Guide 20 2011
Late Devonian paleontology and paleoenvironments at Red Hill and other fossil sites in the Catskill Formation of north-central Pennsylvania Edward B. Daeschler Vertebrate Paleontology, Academy of Natural Sciences, 1900 Benjamin Franklin Parkway, Philadelphia, Pennsylvania 19103, USA Walter L. Cressler III Francis Harvey Green Library, 25 West Rosedale Avenue, West Chester University, West Chester, Pennsylvania 19383, USA
ABSTRACT The stratified red beds of the Catskill Formation are conspicuous in road cut exposures on the Allegheny Plateau of north-central Pennsylvania. During this field trip we will visit and explore several fossil localities within the Catskill Formation. These sites have been central to recent investigations into the nature of Late Devonian continental ecosystems. By the Late Devonian, forests were widespread within seasonally wellwatered depositional basins and the spread of plants on land from the late Silurian through the Devonian set the stage for the radiation of animals in both freshwater and terrestrial settings. A diverse assemblage of flora and fauna has been recovered from the Catskill Formation including progymnosperms, lycopsids, spermatophytes, zygopterid and stauripterid ferns, barinophytes, invertebrates and invertebrate traces, and vertebrates such as placoderms, acanthodians, chondrichthyans, actinopterygians, and a variety of sarcopterygians including early tetrapods. Since the early 1990s, highway construction projects along the Route 15 (Interstate 99) have provided a new opportunity for exploration of the Catskill Formation in Lycoming and Tioga counties. The faunas along Route 15 are dominated by Bothriolepis sp. and Holoptychius sp. and also include Sauripterus taylori and an assortment of other interesting records. The most productive Catskill site, and the source of early tetrapod remains, is Red Hill in Clinton County. Red Hill presents a diverse and unique flora and fauna that is distinct from Route 15 sites, and also provides a spectacular section of the alluvial plain deposits of the Duncannon Member of the Catskill Formation.
*
[email protected];
[email protected] Daeschler, E.B., and Cressler, W.L., III, 2011, Late Devonian paleontology and paleoenvironments at Red Hill and other fossil sites in the Catskill Formation of north-central Pennsylvania, in Ruffolo, R.M., and Ciampaglio, C.N., eds., From the Shield to the Sea: Geological Field Trips from the 2011 Joint Meeting of the GSA Northeastern and North-Central Sections: Geological Society of America Field Guide 20, p. 1–16, doi: 10.1130/2011.0020(01). For permission to copy, contact
[email protected]. ©2011 The Geological Society of America. All rights reserved.
1
2
Daeschler and Cressler
OVERVIEW The earliest paleontological investigations of the Catskill Formation in Pennsylvania date to the 1830s and 1840s, when James Hall described the pectoral fin of the rhizodontid sarcopterygian Sauripterus taylori from a fossil discovered during construction of a railroad line near Blossburg, Tioga County (Hall, 1843). Charles Lyell passed through Blossburg in 1840 and examined the Catskill Formation during one of his two North American tours (Leviton and Aldrich, 1992; Davis et al., 2004). Additional fossil material from the Upper Devonian red beds of Pennsylvania was described by Leidy, Newberry, and Eastman during the nineteenth century and early twentieth century. During the 1960s, Keith S. Thomson described the tristichopterid sarcopterygian Hyneria lindae (Thomson, 1968) from material found in the vicinity of North Bend, Clinton County, and Sterropterygion brandei (Thomson, 1972; Rackoff, 1980) from Tioga County. Thomson became aware of Catskill Formation localities from Donald Baird and Alfred Sherwood Romer of Harvard University who had done prospecting and collecting in the region during the 1950s. Paleobotanical investigations in the Catskill Formation began with Leo Lesquereux, who described two species of Archaeopteris from Meshoppen, Wyoming County (Lesquereux, 1884) in his pioneering report on Pennsylvania’s coal flora for the state’s Second Geological Survey. Chester Arnold figured and further described these specimens (Arnold, 1936), as well as other plants from the Oswayo Formation in northern Pennsylvania (Arnold, 1939), a lateral marine equivalent of the Catskill. The most productive sites in this unit were near Port Allegany in McKean County, and when John Pettitt and Charles Beck reinvestigated Arnold’s material, they were able to describe the first confirmed seed from the Late Devonian, Archaeosperma arnol-
dii (Pettitt and Beck, 1968). Recent paleobiological investigations into the Catskill Formation were initiated in 1993 as field crews from the Academy of Natural Sciences of Philadelphia– University of Pennsylvania began systematic exploration of exposures of the Catskill Formation principally in Clinton, Lycoming, and Tioga Counties. Natural, unweathered exposures of the Catskill Formation are rare in Pennsylvania, so paleontological investigations have been closely linked with the construction of the railway system in the nineteenth century and the highway system in the twentieth and twenty-first centuries. We are indebted to the Pennsylvania Department of Transportation and its contractors for access to the sites where most of our fossil discoveries have been made. The Catskill Formation is composed of sand, silt, and mud deposited in a series of prograding deltas, known as the Catskill Delta Complex (Sevon, 1985) (Fig. 1). This clastic wedge was derived from the middle to late Devonian Acadian Mountains shedding sediment westward and northwestward toward a shallow epicontinental sea in the foreland basin of the orogenic zone (Faill, 1985). The Acadian Orogeny was part of the Middle to Late Paleozoic assembly of the Euramerican landmass (a.k.a. Laurussia or the Old Red Sandstone Continent) and subsequent assembly of the supercontinent Pangea. The deposits of the Catskill Delta Complex grade upward from basinal black shales to nearshore marine facies through transitional facies and into delta plain and alluvial plain depositional settings. Our paleontological studies have focused on the deltaic and alluvial plain facies at the top of the Catskill Formation. Palynological analysis has placed all but one of the productive field sites visited on this trip within the Fa2c part of the Fammenian Stage (Traverse, 2003). This corresponds to the poorly calibrated VH palynozone and is perhaps less ambiguously attributed to the
Figure 1. Block diagram of depositional systems of the Catskill Delta Complex during the Late Devonian in Pennsylvania. Reprinted from Cressler et al. (2010a).
Red Hill and other fossil sites in the Catskill Formation VCO palynozone (sensu Streel et al., 1987). The Tioga Welcome Center locality in Tioga County (Stop 4) is slightly older. Where differentiated, the uppermost subdivisions of the Catskill Formation are the Sherman Creek Member and the overlying Duncannon Member (Fig. 2). At Powys Curve (Stop 1) in Lycoming County, the Sherman Creek Member is well exposed. The Sherman Creek Member is typically composed of fining-upward cycles on the order of two to three meters thick. These cycles are often laterally continuous and are dominated by fine to medium sandstones and siltstones. Poorly developed paleosols and root traces are seen in the fine sandstones and siltstones in the upper part of each cycle (Harvey, 1998). The lithology suggests that the Sherman Creek Member formed in a delta plain setting characterized by low-gradient, high-sinuosity, shallow channels near the coastline (Sevon, 1985). In contrast, the Duncannon Member generally consists of 5–10-m-thick fining-upward cycles dominated by thick, crossbedded, basal sandstones. The contact between the channel sandstone and the mudstone at the top of the underlying cycle is sharp and irregular. Mature paleosols and root traces are often recognized in the mudstones. The Duncannon Member formed in meandering stream channels and overbank deposition on an alluvial plain and is very well exposed at Red Hill (Stop 5), the source of abundant fossil material including plants, arthropods, and a diverse vertebrate fauna including early tetrapods. As recorded in the rocks observed on this field trip, the Late Devonian was a time of major transitions in flora, fauna, and the geobiological system. By the Late Devonian, forests were widespread within the seasonally well-watered depositional basins.
Figure 2. Stratigraphic relationships of the Catskill Formation in north-central Pennsylvania.
3
The dominant tree was the progymnosperm Archaeopteris, growing up to 18 m tall. This was the first plant known with a bifacial cambium as in modern wood, but it reproduced through spores (Meyer-Berthaud et al., 1999). Seed plants are first known from the Late Devonian (Rothwell, et al., 1989). Evidence at Red Hill indicates that opportunistic seed plants, growing in areas disturbed by fire, took advantage of the destruction of the widespread fern Rhacophyton (Cressler, 2006; Cressler et al., 2010a). Lycopsids were important swamp plants in the Late Devonian, attaining the stature of small trees and, along with Rhacophyton, contributing to the thin coal seams known from this time. These small Late Devonian lycopsids are the precursors of the immense lycopsids that were the primary components of Carboniferous coal swamps (Cressler and Pfefferkorn, 2005). The increased stature of the plants in the Late Devonian was accompanied by a concomitant increase in root zone depth, which led to increased paleosol development (Driese and Mora, 1993). The development of paleosols has been linked to increased nutrient flow to the adjacent marine basins, a factor that may also explain increased anoxia, black shale formation, and marine extinctions (Algeo and Scheckler, 1998). The spread of plants on land from late Silurian through the Devonian set the stage for the radiation of animals in both freshwater and terrestrial settings. Large, suspension-feeding bivalves, Archanodon catskillensis, are recorded sporadically from throughout the Catskill Formation. Fossil terrestrial arthropods (first known from the Silurian) include millipedes (Wilson et al., 2005), scorpions, and a trigonotarbid arachnid (Shear, 2000) from the Catskill Formation. All of these are either detritivores or predators, and there is no evidence of herbivory on living plant tissue during the Late Devonian. It appears that the increased contribution of organic detritus by land plants to terrestrial and freshwater ecosystems during this time provided the primary productivity for these increasingly complex and diverse ecosystems. A diverse vertebrate assemblage, including placoderms, acanthodians, chondrichthyans, actinopterygians, and a range of sarcopterygians including at least three species of tetrapods (Daeschler et al., 1994, 2009; Daeschler, 2000b; Shubin et al., 2004) is known from the Catskill Formation. There are two distinct faunas that characterize the vertebrates from the Catskill Formation. The fauna dominated by the porolepiform sarcopterygian Holoptychius sp. and antiarch placoderm Bothriolepis sp. is characteristic of the Sherman Creek Member and much of the undifferentiated Catskill. This assemblage also includes rare occurrences of the rhizodont sarcopterygian Sauripterus taylori. These appear to be organisms mainly restricted to lower alluvial plain–deltaic habitats. Red Hill is characteristic of the other distinct fauna, which includes groenlandaspidid placoderms, a gyracanthid acanthodian, megalichthyid and tristochopterid sarcopterygians, and tetrapods, all from alluvial plain depositional facies. Some of the sarcopterygians (particularly the tetrapods) are lineages adapted for mobility in stream channels and shallow, obstructed waters, a habitat that the well-vegetated alluvial plains of the Catskill Delta Complex seem to have provided. Red Hill
4
Daeschler and Cressler
has been the subject of recent analyses of plant distribution in a Late Devonian landscape (Cressler, 2006) and the paleoecological conditions associated with terrestrialization during the Late Devonian (Cressler et al., 2010a). DAY 1 Road Log Cumulative miles (km) 0.0 (0.0)
3.5 (5.6)
30.8 (49.6)
39.4 (63.4) to 40.4 (65.0) 43.6 (70.2) 44.1 (71.0) 47.8 (76.9)
Directions and notes Take Exit 178 off Interstate 80 onto Route 220 North toward Lock Haven. The first few miles on Route 220 cross an anticlinal valley of Ordovician units. The ridges to north and south are composed of resistant sandstones of the Reedsville, Bald Eagle, and Juniata formations. Crossing through the Ordovician sandstones of Bald Eagle Mountain into Silurian units flanking the north side of the ridge (Tuscarora Formation). Route 220 continues to the northeast with Bald Eagle Mountain to the southeast (this is the last ridge of the Valley and Ridge Province) and the Allegheny Plateau to the northwest. Take Exit 29 for Route 15 North (also called Interstate 99). Route 15 North cuts into the Allegheny Plateau along Lycoming Creek. Between road log miles 34.5 and 40.5, a series of road cuts begin with Middle Devonian near-shore marine sediments of the Lock Haven Formation and after crossing the structural boundary of the Allegheny Front (Beautys Fault in this region) begins to move up-section. Excellent section exposed in road cuts through the lower part of the Catskill Formation (Irish Valley and Sherman Creek Members). Take exit for Route 14 North (Trout Run). Sharp U-turn onto access road (SR-17). Follow access road south, parallel to Route 15. Pull over onto right shoulder of road to examine blocks along roadbed.
Creek Member of the Catskill Formation exposed in the large vertical cut on the west side of Route 15. The Sherman Creek Member at this site is composed of laterally continuous siltstones and fine sandstones in 2–3-m-thick fining-upward cycles. The depositional setting for this section is a delta plain with lowgradient, shallow channels. Since 1994, vertebrate material has been collected from the debris slopes all along the raised portion of the highway roadbed. The fauna is dominated by the antiarch placoderm Bothriolepis sp. and the porolepiform sarcopterygian Holoptychius sp. A single pectoral fin and shoulder girdle of Sauripterus taylori was recovered from this site in 1995 (Fig. 3). The new specimen of Sauripterus taylori is quite informative and, remarkably, preserves the same portion of the skeleton, a right pectoral girdle and fin in ventral view, as the classic specimen described by Hall in 1843 (AMNH 3341) from ~40 km north of Powys Curve. The new specimen is described and discussed by Daeschler and Shubin (1998) and Davis et al. (2004). During construction of the roadway between Powys Curve and Trout Run, partially articulated skull material of a large tristichopterid, cf. Eusthenodon sp. was recovered. This material is
Stop 1. Powys Curve Coordinates: N41°20.840′ W77°05.614′ Construction along this stretch of Route 15 (Interstate 99) was completed in 1994. We will spend time looking at rocks and fossils along the debris slope between the highway road bed and the access road that runs adjacent to the highway along the bend in Lycoming Creek. At this stop, we can observe the Sherman
Figure 3. Ventral view of right pectoral fin of Sauripterus taylori discovered at Powys Curve, Lycoming County, Pennsylvania, in 1995. Note that the distal fin rays (lepidotrichia) are not preserved. Abbreviation: leps—lepidotrichia. Scale bar equals 5 cm.
Red Hill and other fossil sites in the Catskill Formation easily distinguished from Hyneria lindae, the large tristichopterid common at the Red Hill site in Clinton County (Stop 5). Both are derived members of the tristichopterid clade and typical of late Famennian large tristichopterids with cosmopolitan distribution. The cf. Eusthenodon sp. material from Powys Curve–Trout Run includes derived features such as no contact between the posterior supraorbital and intertemporal, a long parietal shield with the pineal foramen located in the posterior portion of the shield (well behind the orbits), and a lacrimal with a postorbital bar excluding the jugal from the orbital margin. The “Bothriolepis Problem” Thomson and Thomas (2001) and Weems (2004) reviewed the taxonomic status of Bothriolepis from the Catskill Formation. It is clear from these papers that the diagnosis of specieslevel features in Bothriolepis from the Catskill Formation is a microcosm of the issues concerning species-level taxonomy of this cosmopolitan Late Devonian genus more broadly. Bothriolepis material from the Catskill Formation was first described by Leidy (1856) as Stenacanthus nitidus based on a distal portion of a pectoral appendage from Blossburg, Tioga County, Pennsylvania. Newberry (1889) described a new bothriolepid species from the Catskill Formation, Bothriolepis leidyi, from Mansfield, Tioga County, Pennsylvania, acknowledging a synonymy with Stenacanthus nitidus. Newberry (1889) also described a new species, Bothriolepis minor from the Upper Devonian of Bradford County, Pennsylvania, that is distinguished by smaller size and finer ornament than Bothriolepis leidyi. Several authors have questioned or rejected the validity of Bothriolepis minor (e.g., Stensiö, 1948; Weems et al., 1981; Thomson and Thomas, 2001). Cope (1892) recognized the priority of Bothriolepis nitidus as a senior synonym of Bothriolepis leidyi, although Eastman (1907) changed the name to Bothriolepis nitida to correct the gender disagreement of the binomen. Weems et al. (1981) described Bothriolepis virginiensis from a Catskill Formation equivalent in Virginia. Thomson and Thomas (2001) suggested that Bothriolepis nitida be used to refer to all Bothriolepis material from the United States, mostly due to poor material and lack of diagnostic characters, although Weems (2004) maintained that B. virginiensis is a distinct species. The situation regarding the species-level taxonomy of Catskill Formation Bothriolepis suggests the need for further work, especially in light of the large sample of well-preserved Bothriolepis sp. material collected recently along the Route 15 corridor (Interstate 99) in Lycoming and Tioga counties (see Figure 6D for an example of recently collected material). Cumulative miles (km) 47.8 (76.9) 48.8 (78.5)
55.9 (90.0)
section into the Duncannon Member of the Catskill Formation. Pull over onto shoulder just before guardrail and sign for the Cogan House Exit.
Stop 2. Steam Valley Coordinates: N41°26.421″ W77°05.994′ Road work that was completed in 2010 created significant new exposures of the upper part of the Catskill Formation in the southbound lanes of Route 15 in Steam Valley, and in the area where Steam Valley Road / Green Valley Road intersects the northbound lanes of Route 15. Scattered vertebrate remains, particularly Bothriolepis sp. and Holoptychius sp., have been found in rock fall from the series of older exposures along the previously improved northbound lanes. An interesting horizon with freshwater bivalves (Archanodon catskillensis) can be observed in situ at this site. The dense packing of these infaunal suspension feeders speaks to the productivity of the Late Devonian continental ecosystem. Chamberlain et al. (2004) reviewed the paleoecological and biogeographic implications of archanodonts, suggesting that they inhabited freshwater floodplain and brackish estuarine environments of the Old Red Continent. Remington et al. (2008) studied the Archanodon-bearing sediments at this exposure and indicated that the Archanodon horizon occurs in the point bar facies within an incised channel body. Palynomorphs from near the base of the exposure suggest a Famennian 2c age (Traverse, 2003) for the strata at this site. Vertebrate collections from the recent construction work on the southbound lanes of Route 15 in Steam Valley include the typical Holoptychius sp. material, as well as a small phyllolepid placoderm trunk plate. Cumulative miles (km) 55.9 (90.0 58.3 (93.8)
63.3 (101.9) to 65.3 (105.1)
Directions and notes Carefully make U-turn to head north on access road. Stay left and follow short ramp onto Route 15 North. After passing Trout Run, Route 15 North climbs into Steam Valley and moves up
5
71.3 (114.7)
Directions and notes Continue north on Route 15. Take exit for Route 184 (Steam Valley). Go left at top of ramp. Follow across overpass and to entrance for the Turkey Ranch Restaurant. Lunch stop. Return to overpass and get back on Route 15 North. A series of low road cuts on west side of the highway expose upper parts of the Catskill Formation. These cuts have produced a skull roof of the cosmopolitan rhynchodipterid lungfish Soederberghia groenlandica (discussed in Ahlberg et al., 2001), and an articulated body of the megalichthyid osteolepiform Sterropterygion brandei, which as described by Thomson (1972) and endochondral fin anatomy discussed by Rackoff (1980). Cresting Bloss Mountain, southern limb of the Blossburg Syncline.
6 75.0 (120.7)
77.0 (123.9) 78.0 (125.5) to 80.0 (128.7)
81.2 (130.7) 83.0 (133.6)
87.8 (141.3)
Daeschler and Cressler Exposures of Carboniferous units Burgoon Sandstone and Pottsville Group in center of syncline. Beginning to move down-section through the northern limb of the Blossburg Syncline. Huntley Mountain Formation grading downward into persistent red beds of the Catskill Formation. The DevonianCarboniferous boundary is within Huntley Mountain Formation. Exposure of grey silty sands of the Lock Haven Formation. Take exit for Covington/Canoe Camp. Go left at bottom of ramp, continue under overpass, and take immediate left onto ramp for Route 15 South. Pull over onto right shoulder just before guardrail. Please be very careful and aware of traffic, as the shoulder of the road is quite narrow.
and informative skull material of a tristichopterid sarcopterygian. The skull material of the tristichopterid is associated with scales (Fig. 6B) that are identical to Holoptychius radiatus Newberry 1889 and discussed below (Stop 4). Alongside the Tioga River, which runs northward on the valley floor, is an old railroad grade that was built to transport coal from Blossburg, an early source of high quality coal called “bloss.” This was the Blossburg-Corning Railroad and was traveled by James Hall in July of 1839 as he sought to clarify some New York geological boundary issues in Pennsylvania (Leviton and Aldrich, 1992). Just below the newly made highway exposures is a substantial man-made exposure of Catskill Formation along the railroad bed between Blossburg and Covington. The lithology at the railroad cut is very similar to the red-brown, medium sandstone matrix of the original Sauripterus taylori specimen (AMNH 3341) (Fig. 5). The fossiliferous nature of the rocks along the railroad bed is confirmed by specimens from Hall at the American Museum of Natural History identified as cf. Eusthenodon and with labels indicating that they were collected “at the railroad cut between Blossburg and Covington.” The Late Devonian fossils from this section induced Charles Lyell to take a
Stop 3. Blossburg-Covington Section Coordinates: N41°42.872′ W77°04.732′ This section of Route 15 bypassing the town of Covington was built in 2003 and 2004. The entire Catskill Formation is relatively thin at this location and is undifferentiated with regard to member designations. The contact between the Catskill and the overlying Huntley Mountain Formation is gradational (Berg and Edmunds, 1979). Sandstone bodies become increasing common upsection and take on a gray color in this transition zone from the meandering river facies of the Catskill Formation to predominantly braided river facies. The Devonian-Carboniferous boundary is within the lower part of the Huntley Mountain Formation. We will examine rocks and fossil material that are found in the large debris slope off the western shoulder of the highway. A diverse fauna has been collected from this section of the highway that includes separate mass mortality layers of adult and juvenile Bothriolepis sp. A small area of the debris slope is the source of numerous samples that preserve the mass mortality of juvenile Bothriolepis sp. material. Criswell et al. (2007) reported head plus trunk shield lengths of 20 mm to 30 mm for most individuals from the mass mortality layers (Fig. 4). The morphology of these individuals matches the expectations for juvenile morphology of Bothriolepis canadensis as established by trends in allometric growth discussed by Werdelin and Long (1986). Of particular note is the relatively large head shield, the relative size of the orbital fenestra, and a large ventral opening in the trunk shield. The occurrence of numerous, similar-sized, articulated individuals in a closely packed taphocoenosis may provide clues about the nature of the earliest life history of Bothriolepis. Other material recovered from this section includes adult Bothriolepis sp. (Fig. 6D), a phyllolepid placoderm, a single specimen of a palaeoniscid actinopterygian, Holoptychius sp.,
Figure 4. Bedding plane with multiple articulated juvenile individuals (in ventral view) of Bothriolepis sp. from the Catskill Formation near Blossburg, Tioga County, Pennsylvania. Scale bar equals 2 cm.
Red Hill and other fossil sites in the Catskill Formation trip to Blossburg (along this rail line) to investigate the similarity between the fossil fauna from Blossburg and that of the Old Red Sandstone in the British Isles (Leviton and Aldrich 1992). Cumulative miles (km) 87.8 (141.3) 90.9 (146.3)
108.5 (174.6) 110.3 (177.5)
112.8 (181.5)
Directions and notes Continue on Route 15 southbound. Take exit for Blossburg. Go left at bottom of ramp, continue under overpass, and take immediate left onto ramp for Route 15 North. Catskill Formation grading downward into Lock Haven Formation Take exit for Route 287 / Tioga. Go left at bottom of ramp, continue under overpass, and take immediate left onto ramp for Route 15 South. Enter and park at Tioga Welcome Center Rest Stop.
Stop 4. Tioga Welcome Center Rest Stop Coordinates: N41°54.026′ W77°07.542′ This highway facility has restrooms and vending machines and offers nice views to the west. To the north of the welcome
7
center you can observe the gradational contact of the Catskill Formation (undifferentiated) with the underlying Lock Haven Formation. The contact zone between the gray sands and muds of the marine Lock Haven Formation and the red beds of the Catskill Formation is characterized by the presence of articulate brachiopods, Lingula and vertical burrows, and may represent a muddy tidal flat at the marine-terrestrial interface (Slane and Rygel, 2009). South of the welcome center the gradational contact between the Catskill Formation and the overlying Huntley Mountain Formation is mapped half way up the wooded hillside. The Catskill is thus relatively thin in this location (~100 m thick). The welcome center was built as part of a project to upgrade this section of the highway in 2001–2002. At that time, a section of the Catskill Formation was removed at the present site of the welcome center to create a flat building site. The excavations were examined numerous times during that phase of the work. Collected was fossil material of the placoderms Bothriolepis sp., Phyllolepis sp., a single element of a large dinichthyid, cf. Dunkleosteus sp., as well as the sarcopterygians Holoptychius sp., a tristichopterid, and the recently described lungfish Apatorhynchus opistheretmus (Friedman and Daeschler, 2006) (Fig. 6E). A palynomorph sample collected near road level across the highway from the welcome
Figure 5. Original illustration from Hall (1843) of Sauripterus taylori pectoral fin and scales found along a railroad cut near Blossburg, Tioga County, Pennsylvania.
8
Daeschler and Cressler
center suggests a Famennian 2b age for the lower part of this section (A. Traverse, 2004, personal commun.). Although fragmentary, the tristichopterid material from this site (also seen at Stop 3) is consistently associated with the same distinctive scale morphology (Fig. 6B). These distinct scales were named Holoptychius radiatus from specimens collected in the Blossburg area in the late nineteenth century (Newberry 1889). The type material of H. radiatus consists of distinctive scales with the characteristic tristichopterid feature of a raised, tear-drop shaped boss on the internal surface. With the suite of new specimens, this scale morphology can now be associated with lower jaw and skull material (e.g., Fig. 6C). A more complete description of this tristichopterid sarcopterygian (with a new generic attribution) is in order. Cumulative miles (km) 112.8 (181.5) 120.3 (193.6)
132.7 (213.6) 132.8 (213.7)
Directions and notes Continue southbound on Route 15. Take exit for Route 6, Mansfield/Wellsboro. Take a right at the bottom of the ramp following Route 6 West (towards Wellsboro). In Wellsboro, turn left at light (intersection of Route 6 and Route 660) onto Main Street. Turn right into parking lot for Penn Wells Hotel. End of Day 1.
DAY 2 Road Log Cumulative miles (km) 132.8 (213.7)
Directions and notes Turn right on Main Street and continue for one block. 132.9 (213.8) Left onto Route 287 South. 144.5 (232.5) Bear right to follow 287 South. 145.2 (233.8) Right on to Route 414 West. 175.3 (282.1) Right on to Route 44 North. 180.5 (290.5) Bear right to follow Route 44 North. 185.8 (299)Left on to Hyner Mountain Road (look for signs for Hyner Run State Park). 192.7 (310.1) Right on Route 120 West. 194.7 (313.3) Pull over on to wide shoulder at outcrop on right. Stop 5. Red Hill Coordinates: N41°20.645′ W77°40.800′ Introduction to Red Hill The Red Hill road cut is a kilometer-long exposure of slightly western-dipping alluvial sandstones, siltstones and paleosols of the Duncannon Member of the Catskill Formation. The western part of the exposure is dominated by channel sandstones. The face of the road cut is quite steep here and reaches a height of 30 m,
thus fossil prospecting has been restricted to examination of the frequent rock falls. A few fossil discoveries have been made in the finer-grained rocks that have fallen, but collecting has been minimal in this part of the outcrop due to taphonomic and logistical factors. The lack of collecting, however, is compensated for by the considerable sequence of fluvial strata that can be observed here, indicative of a dynamic Late Devonian alluvial plain. The eastern portion of the road cut is dominated by finergrained channel margin and inter-channel siltstones and paleosols. A vertically thin (3 m) but laterally broad (~200 m exposed) sequence of fossiliferous strata in this portion of the outcrop is the source of the abundant fossil remains from Red Hill. The fossiliferous lenses reflect several different styles of deposition from low energy standing water to higher energy lags. The slope of the outcrop is not cut as steeply in these fine-grained layers. The presence of talus and the creation of terraces during fossil prospecting allow more access to the strata for close observation and continued collecting. The talus slopes are good for collecting fragmentary specimens of fish teeth and scales, and plant compressions. Please USE CAUTION when climbing the rock face, and always be aware of traffic at this roadside location. Red Hill has produced one of the most diverse and abundant samples of life from a Late Devonian continental ecosystem. At least thirteen species of plants have been identified, representing an ecological patchwork of progymnosperm forests, lycopsid wetlands, zygopterid fern glades, and patches of early spermatophytes that occupied sites disturbed by fires. Two newly discovered terrestrial arthropods have been described from Red Hill, a trigonotarbid arachnid and a myriapod. The remains of at least fifteen species of vertebrate have been recovered, including placoderms, an acanthodian, chondrichthyans, palaeoniscid actinopterygians, finned sarcopterygians, and three early tetrapod species. A variety of inter-channel depositional settings formed a wide range of aquatic and terrestrial habitats as a result of the periodic avulsion of meandering rivers across the alluvial plain. These habitats formed a crucible of evolution for plants and animals, including innovations critical to the further development of life on land. Red Hill Sedimentology Traditionally, sedimentation in upper alluvial and coastal plain settings has been envisioned as being produced by singlethread meandering rivers. However, the pattern of sedimentological structures at Red Hill conforms to a model of periodic avulsion like that described in recent studies of modern fine-grained fluvial systems that show these systems cycling through two stages with a typical period on the order of 1000 years (Smith et al., 1989; Soong and Zhao, 1994; Slingerland and Smith, 2004). Stage I begins when a channel changes course by permanently breaching its levee. A sediment wedge progrades down-current from the avulsion site resulting in intense alluviation of the floodplain as the system changes from a single channelized flow into rapidly evolving distributary channels. These channels split and coalesce in a complex network while scouring the floodplain to create
Red Hill and other fossil sites in the Catskill Formation
Figure 6. Typical vertebrate fauna from the Catskill Formation along Route 15 (Interstate 99) in Lycoming and Tioga counties, Pennsylvania. (A) Scales of porolepiform, Holoptychius sp. (B) Scales of tristichopterid, Holoptychius radiatus. (C) Rostral portion of lower jaws of H. radiatus with submandibulars and gulars. (D) Dorsal view of adult antiarch placoderm Bothriolepis sp. from mass mortality zone. (E) Partial skull in dorsal view of lungfish, Apatorhynchus opistheretmus. (F) Pectoral fin of rhizodontid, Sauripterus taylori. Scale bars equal 2 cm.
9
10
Daeschler and Cressler
transient ponds and lakes as they encounter preexisting channel levees and floodplain vegetation. Commonly observed deposits of this stage are: (1) coarser-grained crevasse splays of a variety of lobate, elliptical, or elongate shapes (Smith, 1986; O’Brien and Wells, 1986; Bristow, 1999); and (2) finer-grained lake and distal splay deposits in which rapid burial has preserved organic debris from oxidation. The former are observed at Red Hill in the variety of sandstone bodies ranging in morphology from wedgeshaped lenticular, to flat-bottomed and convex-upward. The latter comprise the main fossiliferous strata at Red Hill, further described below as comprising four taphofacies. Stage II of the avulsion cycle begins when distributary channels once again follow the regional slope and start flowing subparallel to the parent channel where they are eventually abandoned as flow is captured into a new trunk channel similar in scale to the parent channel that initially avulsed (Smith et al., 1989). Sedimentation rates are low, allowing soil formation to resume on the floodplain. The new trunk channel incises into its earlier avulsion deposits, creating a new meander belt that is relatively narrow and only reworks a small fraction of the Stage I avulsion fill floodplain deposits into meander belt deposits. Evidence at Red Hill for Stage II of the avulsion cycle consists of the extensive paleosols easily observed at the eastern end of the outcrop and the massive channel sandstone bodies observable at the western end of the outcrop. The considerable lateral and vertical variation within the strata at Red Hill reflects the heterogeneity of depositional facies of the avulsion model. This variation in depositional facies, in turn, results in the uncommon circumstance of a range of taphonomic modes preserving both plant and animal remains at the same site. Four different taphofacies preserve fossil material: sorted microfossil horizons, basal lags, channel-margin and standing-water deposits. Well-sorted microfossil accumulations and basal lag deposits contain abundant, but fragmentary, vertebrate material that may be allochthonous and thus have poor time and ecological fidelity. The channel-margin taphofacies contains isolated and associated vertebrate material, often in discrete lenses. The character of the entombing sediments indicates that the fossils accumulated along the strandline of the aggrading margins of temporary channels in overbank areas after avulsion episodes. Deposits of this sort have the potential to accumulate relatively quickly, and the fact that the taphofacies shows little or no abrasion or pre-depositional weathering of accumulated material indicates that the associated taxa were living penecontemporaneously in the areas near the site of deposition. The standing-water taphofacies is represented by green-gray siltstones with abundant plant material and an occasional occurrence of arthropod and vertebrate remains. The vertebrate remains from this setting are black and “carbonized” suggesting different water chemistry (perhaps more acidic) and diagenetic conditions than other taphofacies at Red Hill. These deposits represent low energy, reducing environments, such as floodplain ponds and distal splay settings that can provide excellent temporal and ecological fidelity.
Red Hill Flora (Table 1) All the of the well-preserved plant fossils from Red Hill have come from a single gray-green siltstone pond deposit at the eastern end of the outcrop, with the exception of a few significant specimens found in a rock that had fallen from the western end. The floral characteristics of Red Hill are typical of a Late Devonian plant assemblage, specifically a subtropical Archaeopteris forest. Four Archaeopteris leaf morphospecies are dominated by A. macilenta and A. hibernica (Fig. 7C). This progymnosperm tree is an index fossil for the Late Devonian (Banks, 1980), as is the second most abundant set of plant remains at Red Hill, the zygopterid fern assigned to Rhacophyton ceratangium. The early diversification of arborescent lycopsids are represented by numerous decorticated stems (Fig. 7B), some identifiable as Lepidodendropsis. Also, well-preserved remains of cormose isoetalean bases and stems have been described as Otzinachsonia beerboweri (Cressler and Pfefferkorn, 2005). Two species of spermatophytes are present, including the newly described species Duodimidia pfefferkornii (Fig. 7A; Cressler et al., 2010b) characterized by fused symmetric cupule pairs, and Aglosperma quadrapartita (Cressler, 2006). The palynological age of the strata make it coeval with other sites preserving the earliest recorded spermatophytes in Belgium and West Virginia (FaironDemaret and Scheckler, 1987; Rothwell et al., 1989). Other minor floral elements include the stauripterid fern Gillespiea and a variety of barinophytes (Cressler, 2006). Major plant groups found at other Late Devonian sites, but not yet discovered at Red Hill, are the sphenopsids and cladoxylaleans. Red Hill Fauna (Table 1) The arthropod fauna discovered at Red Hill is likely only a very limited subset of the invertebrate community present in the floodplain ecosystem. A trigonotarbid arachnid (Fig. 7M; Shear, 2000) and archidesmid myriapod (Fig. 7L; Wilson et al., 2005) have been described from the standing-water taphofacies, but greater diversity is evidenced by enigmatic body impressions, burrow traces and walking traces. The vertebrate assemblage represents a diverse community that was living in aquatic habitats within the alluvial plain of the Catskill Delta Complex. These include bottom feeders, durophages, filter feeders and a wide range of predators. The placoderm assemblage is dominated by the small groenlandaspidid, Turrisaspis elektor (Fig. 7D), one of the most common taxa from the site (Daeschler et al., 2003). Lane and Cuffey (2005) recognize a new species of Phyllolepis, a cosmopolitan placoderm reaching the Euramerica landmass in the Famennian. Fin spines and pectoral girdle elements of the acanthodian Gyracanthus (cf. G. sherwoodi) (Fig. 7G) are common, as are small isolated teeth of the chondichthyan Ageleodus (Fig. 7F; Downs and Daeschler, 2001). Rare xenacanthiform chondrichthyan teeth have been recognized (Fig. 7E). Among the bony fish, the small palaeoniscid actinopterygian, Limnomis delaneyi (Fig. 7H; Daeschler 2000a), and the large tristichopterid sarcopterygian, Hyneria lindae (Fig. 7K; Thomson, 1968), are the dominant components. Other sarcopterygians
Red Hill and other fossil sites in the Catskill Formation
TABLE 1. RED HILL FAUNA AND FLORA RED HILL FAUNA
RED HILL FLORA
Chelicerata Arachnida Trigonotarbida Palaeocharinidae Gigantocharinus szatmaryi Shear Myriapoda Diplopoda Archidesmida Zanclodesmidae Orsadesmus rubecollus Wilson Vertebrata Placodermi Phyllolepida Phyllolepididae Phyllolepis rossimontina Lane and Cuffey Arthrodira Groenlandaspididae Groenlandaspis pennsylvanica Daeschler Turrisaspis elektor Daeschler Acanthodii Climatiiformes Gyracanthidae Gyracanthus cf. G. sherwoodi Newberry Chondrichthyes Ctenacanthiformes Ctenacanthidae Ctenacanthus sp. Xenacanthiformes Indet. Insertae Sedis Ageleodus pectinatus (Agassiz) Osteichthyes Actinopterygii Palaeonisciformes Limnomis delaneyi Daeschler Indet. Sarcopterygii Dipnoi Indet. Crossopterygii Rhizodontidae cf. Sauripterus sp. Indet. Megalichthyidae Indet. Tristichopteridae Hyneria lindae Thomson Amphibia Ichthyostegalia Hynerpeton bassetti Daeschler Densignathus rowei Daeschler cf. Whatcheeridae Indet.
Zosterophyllopsida Barinophytales cf. Protobarinophyton sp. Barinophyton obscurum (Dun) White Barinophyton sibericum Petrosian Lycopsida Isoetales Otzinachsonia beerboweri Cressler and Pfefferkorn cf. Lepidodendropsis Lutz Filicopsida Zygopteridales Rhacophyton ceratangium Andrews and Phillips Stauropteridales Gillespiea randolphensis Erwin and Rothwell Progymnospermopsida Archaeopteridales Archaeopteris macilenta (Lesq.) Carluccio et al. Archaeopteris hibernica (Forbes) Dawson Archaeopteris obtusa Lesquereaux Archaeopteris halliana (Göppert) Dawson Gymnospermopsida Pteridospermales cf. Aglosperma quadrapartita Hilton and Edwards Duodimidia pfefferkornii Cressler et al.
11
12
Daeschler and Cressler
Figure 7. Typical fauna and flora from the Catskill Formation at Red Hill, Clinton County, Pennsylvania. (A) Early spermatophyte, Duodimidia pfefferkornii. (B) Lycopod stem, cf. Lepidodendropsis. (C) Progymnosperm, Archaeopteris sp. (D) Head and truck shields in oblique lateral view of groenlandaspidid placoderm, Turrisaspis elektor. (E) Tooth of xenacanthiform chondrichthyan. (F) Tooth of chondrichthyan, Ageleodus pectinatus. (G) Pectoral fin spine of gyracanthid acanthodian, Gyracanthus cf. G. sherwoodi. (H) Body of small actinopterygian, Limnomis delaneyi. (I) Small lungfish tooth plate. (J) Skull roof of megalichthyid osteolepidid, gen. et sp. indet. (K) Isolated vomerine teeth of large tristichopterid, Hyneria lindae. (L) Millipede, Orsadesmus rubecollus. (M) Palaeocharinid arachnid, Gigantocharinus szatmaryi. Scale bars equal 2 cm (B, C, D, G, J, K), 5 mm (A, H, I, L, M), and 1 mm (E, F).
Red Hill and other fossil sites in the Catskill Formation include rhizodontid and megalichthyid (Fig. 7J) forms. Early tetrapod remains are rare and are represented by isolated skeletal elements, although recent analysis suggests that at least three penecontemporaneous taxa are present: Hynerpeton bassetti (Fig. 8A), Densignathus rowei (Fig. 8B), and a whatcheerid-like form (Fig. 8F, 8I) (Daeschler et al., 2009). Interestingly, taxa such as Bothriolepis and Holoptychius that are common at most other late Famennian sites in the Catskill Formation are absent at Red Hill, except for Holoptychius sp. scales from a rock fall from high in the Red Hill section. Red Hill Paleoecology A paleoecological analysis of the Red Hill plant community (Cressler, 2006) characterized the vegetation as a subtropical Archaeopteris floodplain forest interspersed with lycopsid wetlands and widespread stands of Rhacophyton on the floodplain
13
and along water margins (Figs. 9 and 10). Taphonomic and fossildistribution evidence was derived from systematic sampling of the floodplain pond deposit. The plant fossils had undergone little or no transport. The evidence provided in that study was interpreted to support a model of habitat partitioning of the landscape by the plants at a high phylogenetic level, a characteristic of midPaleozoic plant communities (DiMichele and Bateman, 1996). The pattern of plant distribution at Red Hill is similar to that seen in other Late Devonian paleoecological studies (Scheckler, 1986a, 1986b; Rothwell and Scheckler, 1988; Scheckler et al., 1999). Lycopsids dominated the wettest portions of the floodplain, whereas Rhacophyton dominated the poorly drained floodplain margins. Archaeopteris grew in the better-drained areas of the landscape and seed plants grew opportunistically. At Red Hill the seed plants apparently flourished following fires that cleared the Rhacophyton groundcover (Cressler, 2001). This is indicated
Figure 8. Isolated remains of early tetrapods from Red Hill, Clinton County, Pennsylvania. (A) Left shoulder girdle of Hynerpeton bassetti. (B) Left lower jaw of Densignathus rowei. (C) Partial right lower jaw. (D) Ventral and dorsal views of humerus (ANSP 21350). (E) Ventral view of femur. (F) Postorbital. (G) Jugal. (H) Jugal. (I) Lacrimal. Abbreviations: adb—adductor blade; adc—adductor crest; ang—angular; art—articular; cl—cleithrum; dent—dentary; ect—ectepicondyle; ent—entepicondyle; gf—glenoid fossa; it—internal trochanter; orb—orbit; raf—radial facet; sc—sensory canal; sgb—supraglenoid buttress; sur—surangular; tf—tibial facet; ulf—ulnar facet. Scale bars equal 1 cm except in F and I, where scale bars equal 0.5 cm.
14
Daeschler and Cressler
by a succession of Rhacophyton-to-charcoal-to-spermatophyte remains within the small-scale stratigraphic profile (Cressler, 2001; 2006). Fires presumably were associated with the wet and dry seasonal cycle that prevailed in the Late Devonian Catskill Delta system, as also indicated by the calcretes and vertic structures in the paleosols. The floodplain habitats at Red Hill provided a range of conditions for the cohabitation of plants and animals (Cressler et al., 2010a). Plant communities were partitioned on the floodplain across a range of environments from elevated and betterdrained levees to low, wetland habitats (Cressler, 2006). The aquatic settings include open river channels, shallow channel margins, anastomosing temporary channels, and floodplain ponds in interfluves that were subject to periodic flooding. This heterogeneity is expressed even on the local scale at the Red Hill site, as might be expected with the avulsion model of floodplain aggradation. Seasonal flooding and drying probably had a significant role in the annual cycles of plants and animals and their evolution.
Figure 9. The progymnosperm Archaeopteris and understory fern Rhacophyton growing on the edge of a Devonian floodplain pond. Illustration courtesy of Stephen Greb.
Cumulative miles (km) 194.7 (313.3) 195.7 (314.9) 196.0 (315.4)
Directions and notes Continue on Route 120 West. After crossing over bridge at Young Woman’s Creek, turn right onto School House Road. Turn left into rear parking area of the Chapman Township Municipal Building (white building with maintenance yard in rear). Park in the rear parking area.
Stop 6. Red Hill Field Station The people of Chapman Township have allowed us to renovate the second floor of their municipal building to use as our “Red Hill Field Station and Museum.” Mr. Doug Rowe is the driving force behind this facility. The Field Station serves as an important educational tool for visitors to the Red Hill site, including local school groups, fossil clubs, and casual visitors. It also provides a place for our research group to get out of the hot, cold,
Figure 10. Rhacophyton-dominated Devonian swamps, but arborescent lycopsids became increasingly common in the Carboniferous. Illustration courtesy of Stephen Greb.
Red Hill and other fossil sites in the Catskill Formation or wet weather while working at Red Hill. Please take some time to explore in the Field Station. Examples of most of the taxa from the Red Hill site are on display. Most are original fossil material. Almost 1,000 specimens from Red Hill are catalogued in the collections at the Academy of Natural Sciences in Philadelphia. Some of those specimens are presented as casts and photographs at the Field Station. Cumulative miles (km) 196.0 (315.4) 196.3 (315.9) 197.3 (317.5) 219.7 (353.8) 220.8 (355.3) 221.4 (356.3) 228.7 (368.1)
Directions and notes Turn right out of Municipal Building parking area. Turn left onto Route 120 East. Passing Red Hill exposure. Turn left at light to stay on Route 120 East in to Lock Haven. Turn right to stay on Route 120 East. Stay in right lane, and follow signs for Milesburg/Route 220 South Take exit for Route 80. End of Road Log.
ACKNOWLEDGMENTS There have been a large number of participants in our ongoing work in the Catskill Formation. We thank all for their dedicated work toward building a better understanding of the paleobiological and geological nature of the Catskill Formation in Pennsylvania. We would like to especially mention Douglas Rowe, Janice Pycha, Rudy Slingerland, Daniel Peterson, Stephen Greb, Frederick Mullison, Jason Downs, John Sime, and the Pennsylvania Department of Transportation for their invaluable contributions to this research and assistance in creating this field guide. REFERENCES CITED Ahlberg, P.E., Johanson, Z., and Daeschler, E.B., 2001, The Late Devonian lungfish Soederberghia (Sarcopterygii, Dipnoi) from Australia and North America, and its biogeographic implications: Journal of Vertebrate Paleontology, v. 21, no. 1, p. 1–12, doi:10.1671/0272-4634(2001)021[0001: TLDLSS]2.0.CO;2. Algeo, T.J., and Scheckler, S.E., 1998, Terrestrial-marine teleconnections in the Devonian: Links between the evolution of land plants, weathering processes, and marine anoxic events: Philosophical Transactions of the Royal Society of London, B, Biological Sciences, v. 353, p. 113–130, doi:10.1098/rstb.1998.0195. Arnold, C.A., 1936, Observations on fossil plants from the Devonian of eastern North America. II. Archaeopteris macilenta and A. sphenophyllifolia of Lesquereux: Contributions from the Museum of Paleontology, University of Michigan, v. 5, no. 3, p. 49–56. Arnold, C.A., 1939, Observations on fossil plants from the Devonian of eastern North America. IV. Plant remains from the Catskill Delta deposits of northern Pennsylvania and southern New York: Contributions from the Museum of Paleontology, University of Michigan, v. 5, no. 11, p. 271–314. Banks, H.P., 1980, Floral assemblages in the Siluro-Devonian, in Dilcher, D.L., and Taylor, T.N., eds., Biostratigraphy of Fossil Plants: Pennsylvania, Dowden, Hutchinson, and Ross, p. 1–24. Berg, T.M., and Edmunds, W.E., 1979, The Huntley Mountain Formation: Catskill-to-Burgoon Transition in North-Central Pennsylvania: Pennsylvania Geological Survey, Information Circular 83, p. 1–80.
15
Bristow, C.S., 1999, Crevasse splays from the rapidly aggrading sand-bed braided Niobrara River, Nebraska: effect of base-level rise: Sedimentology, v. 46, p. 1029–1047, doi:10.1046/j.1365-3091.1999.00263.x. Chamberlain, J.A., Jr., Friedman, G.M., and Chamberlain, R.B., 2004, Devonian Archanodont Unionoids from the Catskill Mountains of New York: Implications for the paleoecology and biogeography of the first freshwater bivalves: Northeastern Geology and Environmental Sciences, v. 26, no. 3, p. 211–229. Cope, E.D., 1892, On some new and little known Paleozoic vertebrates: Proceedings of the American Philosophical Society, v. 30, p. 221–229. Cressler, W.L., III, 2001, Evidence of earliest known wildfires: Palaios, v. 16, p. 171–174. Cressler, W.L., III, 2006, Plant paleoecology of the Late Devonian Red Hill locality, north-central Pennsylvania, an Archaeopteris-dominated wetland plant community and early tetrapod site, in DiMichele, W.A., and Greb, S., eds., Wetlands through Time: Geological Society of America Special Paper 399, p. 79–102, doi:10.1130/2006.2399(04). Cressler, W.L., III, and Pfefferkorn, H.W., 2005, A late Devonian isoetalean lycopsid, Otzinachsonia beerboweri, gen. et sp. nov., from north-central Pennsylvania, USA: American Journal of Botany, v. 92, p. 1131–1140, doi:10.3732/ajb.92.7.1131. Cressler, W.L., III, Daeschler, E.B., Slingerland, R., and Peterson, D.A., 2010a, Terrestrialization in the Late Devonian: a palaeoecological overview of the Red Hill site, Pennsylvania, USA, in Vecoli, M., Clement, G., and Meyer-Berthaud, B., eds., The Terrestrialization process: Modelling Complex Interactions at the Biosphere-Geosphere Interface: Geological Society of London Special Publication 399, p. 111–128. Cressler, W.L., III, Prestianni, C., and LePage, B.A., 2010b, Late Devonian spermatophyte diversity and the paleoecology at Red Hill, north-central Pennsylvania, USA: International Journal of Coal Geology, v. 83, p. 91–102, doi:10.1016/j.coal.2009.10.002. Criswell, K., Downs, J. and Daeschler, E., 2007, Mass mortality of juvenile placoderms (Bothriolepis sp.) from the Catskill Formation (Upper Devonian), Tioga County, Pennsylvania [abs.]: Journal of Vertebrate Paleontology, v. 27 (supplement to 3), p. 62A. Daeschler, E.B., 2000a, An early actinopterygian fish from the Catskill Formation (Late Devonian, Famennian) in Pennsylvania, U.S.A: Proceedings. Academy of Natural Sciences of Philadelphia, v. 150, p. 181–192. Daeschler, E.B., 2000b, Early tetrapod jaws from the Late Devonian of Pennsylvania, USA: Journal of Paleontology, v. 74, p. 301–308, doi:10.1666/0022 -3360(2000)074<0301:ETJFTL>2.0.CO;2. Daeschler, E.B., and Shubin, N.H., 1998, Fish with Fingers?: Nature, v. 391, p. 133, doi:10.1038/34317. Daeschler, E.B., Shubin, N.H., Thomson, K.S., and Amaral, W.W., 1994, A Devonian tetrapod from North America: Science, v. 265, p. 639–642, doi:10.1126/science.265.5172.639. Daeschler, E.B., Frumes, A., and Mullison, C.F., 2003, Groenlandaspidid placoderm fishes from Late Devonian of North America: Records of the Australian Museum, v. 55, no. 1, p. 45–60, doi:10.3853/j.0067 -1975.55.2003.1374. Daeschler, E.B., Clack, J.A., and Shubin, N.H., 2009, Late Devonian tetrapod remains from Red Hill, Pennsylvania, USA: How much diversity?: Acta Zoologica, v. 90, no. 1, p. 306–317, doi:10.1111/j.1463 -6395.2008.00361.x. Davis, M.C., Shubin, N.H., and Daeschler, E.B., 2004, A new specimen of Sauripterus taylori (Sarcopterygii, Osteichthyes) from the Famennian Catskill Formation of North America: Journal of Vertebrate Paleontology, v. 24, p. 26–40, doi:10.1671/1920-3. DiMichele, W.A., and Bateman, R.M., 1996, Plant paleoecology and evolutionary inference: two examples from the Paleozoic: Review of Palaeobotany and Palynology, v. 90, p. 223–247, doi:10.1016/0034-6667(95)00085-2. Downs, J.P., and Daeschler, E.B., 2001, Variation within a large sample of Ageleodus pectinatus teeth (Chondichthyes) from the Late Devonian of Pennsylvania, U.S.A: Journal of Vertebrate Paleontology, v. 21, no. 4, p. 811– 814, doi:10.1671/0272-4634(2001)021[0811:VWALSO]2.0.CO;2. Driese, S.G., and Mora, C.I., 1993, Physico-chemical environment of pedogenic carbonate formation in Devonian vertic paleosols, central Appalachians, U.S.A: Sedimentology, v. 40, p. 199–216, doi:10.1111/j.1365-3091.1993.tb01761.x. Eastman, C.R., 1907, Devonic fishes of the New York formations: New York State Museum Memoir, v. 10, 193 p. Faill, R.T., 1985, The Acadian Orogeny and the Catskill Delta, in Woodrow, D.L., and Sevon, W.D., eds., The Catskill Delta: Geological Society of America Special Paper 201, p. 39–50.
16
Daeschler and Cressler
Fairon-Demaret, M., and Scheckler, S.E., 1987, Typification and redescription of Moresnetia zalesskyi Stockmans, 1948, an early seed plant from the Upper Famennian of Belgium: Bulletin de L’Institut Royal les Sciences Naturalles de Belgique: Sciences del la Terre, v. 57, p. 183–199. Friedman, M., and Daeschler, E.B., 2006, Late Devonian (Famennian) lungfishes from the Catskill Formation of Pennsylvania, USA: Palaeontology, v. 49, no. 6, p. 1167–1183, doi:10.1111/j.1475-4983.2006.00594.x. Hall, J., 1843, Fourth annual report of the survey of the Fourth Geological District: Albany, New York, New York State Assembly Document 50, p. 393–394, 453. Harvey, A., 1998, A paleoenvironmental reconstruction in the Devonian Sherman Creek Member of the Catskill Formation in central Pennsylvania [M.S. thesis]: Philadelphia, Pennsylvania, Temple University, 90 p. Lane, J.A., and Cuffey, R.J., 2005, Phyllolepis rossimontina sp. nov. (Placodermi) from the uppermost Devonian at Red Hill, north-central Pennsylvania: Revista Brasileira de Paleontologia, v. 8, no. 2, p. 117–126, doi:10.4072/rbp.2005.2.04. Leidy, J., 1856, Description of two ichthyodorulites: Proceedings. Academy of Natural Sciences of Philadelphia, v. 8, p. 11–12. Lesquereux, L., 1884, Description of the Coal Flora of the Carboniferous Formation in Pennsylvania and throughout the United States. Second Geological Survey of Pennsylvania. Volume III. Report of Progress: State Printer of Pennsylvania, Harrisburg, p. 695–977. Leviton, A.E., and Aldrich, M.L., 1992, Fishes of the Old Red Sandstone and systemic boundaries, Blossburg, Pennsylvania 1830–1900: Earth Sciences History, v. 11, p. 21–29. Meyer-Berthaud, B., Scheckler, S.E., and Wendt, J., 1999, Archaeopteris is the earliest known modern tree: Nature, v. 398, p. 700–701, doi:10.1038/19516. Newberry, J.S., 1889, The Paleozoic fishes of North America: U.S. Geological Survey Monograph, v. 16, 340 p., 53 pls. O’Brien, P.E., and Wells, A.T., 1986, A small alluvial crevasse splay: Journal of Sedimentary Petrology, v. 56, p. 876–879. Pettitt, J.M., and Beck, C.B., 1968, Archaeosperma arnoldii—a cupulate seed from the Upper Devonian of North America: Contributions from the Museum of Paleontology, University of Michigan, v. 22, p. 139–154. Rackoff, J.S., 1980, The origin of the tetrapod limb and the ancestry of tetrapods, in Panchen, A.L., ed., The Terrestrial Environment and the Origin of Land Vertebrates: Academic Press, London. p. 255–292. Remington, K., Daeschler, E.B., and Rygel, M.C., 2008, Sedimentology of an Archanodon-bearing channel body in the Catskill Formation (Upper Devonian) near Steam Valley, Pennsylvania [abs.]: Geological Society of America Abstracts with Programs, v. 40, no. 2, p. 82. Rothwell, G.W., and Scheckler, S.E., 1988, Biology of ancestral gymnosperms, in Beck, C.B., ed., Origins and evolution of gymnosperms: New York, Columbia University Press, p. 85–134. Rothwell, G.W., Scheckler, S.E., and Gillespie, W.H., 1989, Elkinsia gen nov., a Late Devonian gymnosperm with cupulate ovules: Chicago, Illinois, Botanical Gazette, v. 150, p. 170–189, doi:10.1086/337763. Scheckler, S.E., 1986a, Floras of the Devonian-Mississippian transition, in Gastaldo, R.A., and Broadhead, T.W., eds., Land plants: notes for a short course: Knoxville, Tennessee, University of Tennessee, Studies in Geology, v. 15, p. 81–96. Scheckler, S.E., 1986b, Geology, floristics and paleoecology of Late Devonian coal swamps from Appalachian Laurentia (U.S.A.): Annales de la Société Géologique de Belgique, v. 109, p. 209–222. Scheckler, S.E., Cressler, W.L., Connery, T., Klavins, S., and Postnikoff, D., 1999, Devonian shrub and tree dominated landscapes (abstract): XVI International Botanical Congress, v. 16, p. 13. Sevon, W.D., 1985, Nonmarine facies of the Middle and Late Devonian Catskill coastal alluvial plain, in Woodrow, D.L., and Sevon, W.D., eds., The Catskill Delta: Geological Society of America Special Paper 201, p. 79–90.
Shear, W.A., 2000, Gigantocharinus szatmaryi, a new trigonotarbid arachnid from the Late Devonian of North America (Chelicerata, Arachnida, Trigonotarbida): Journal of Paleontology, v. 74, p. 25–31, doi:10.1666/0022 -3360(2000)074<0025:GSANTA>2.0.CO;2. Shubin, N.H., Daeschler, E.B., and Coates, M.I., 2004, The early evolution of the tetrapod humerus: Science, v. 304, no. 5667, p. 90–93, doi:10.1126/ science.1094295. Slane, D.C., and Rygel, M.C., 2009, Marginal-marine facies of the Catskill Formation (Upper Devonian), Tioga County, Pennsylvania [abs.]: Geological Society of America Abstracts with Programs, v. 41, no. 7, p. 145. Slingerland, R., and Smith, N.D., 2004, River avulsions and their deposits: Annual Review of Earth and Planetary Sciences, v. 32, p. 257–285, doi:10.1146/annurev.earth.32.101802.120201. Smith, D.G., 1986, Anastomosing river deposits, sedimentation rates and basin subsidence, Magdalena River, northwestern Colombia, South America: Sedimentary Geology, v. 46, p. 177–196, doi:10.1016/0037 -0738(86)90058-8. Smith, N.D., Cross, T.A., Dufficy, J.P., and Clough, S.R., 1989, Anatomy of an avulsion: Sedimentology, v. 36, p. 1–23, doi:10.1111/j.1365-3091.1989. tb00817.x. Soong, T.W.M., and Zhao, Y., 1994, The flood and sediment characteristics of the Lower Yellow River in China: Water International, v. 19, p. 129–137, doi:10.1080/02508069408686216. Stensiö, E.A., 1948, On the Placodermi of the Upper Devonian of East Greenland. II. Antiarchi: subfamily Bothriolepinae: Meddelelser om Grønland, v. 139, p. 1–622. Streel, M., Higgs, K., Loboziak, S., Riegel, W., and Steemans, P., 1987, Spore stratigraphy and correlation with faunas and floras in the type marine Devonian of the Ardenne-Rhenish regions: Journal of Palaeobotany and Palynology, v. 50, p. 211–229, doi:10.1016/0034-6667(87)90001-7. Thomson, K.S., 1968, A new Devonian fish (Crossopterygii: Rhipidistia) considered in relation to the origin of the Amphibia: Postilla, v. 124, p. 1–13. Thomson, K.S., 1972, New evidence on the evolution of paired fins in Rhipidistia and the origin of the tetrapod limb, with description of a new genus of Osteolepidae: Postilla, v. 157, p. 1–7. Thomson, K.S., and Thomas, B., 2001, On the status of species of Bothriolepis (Placodermi, Antiarchi) in North America: Journal of Vertebrate Paleontology, v. 21, no. 4, p. 679–686, doi:10.1671/0272-4634(2001) 021[0679:OTSOSO]2.0.CO;2. Traverse, A., 2003, Dating the earliest tetrapods: a Catskill palynological problem in Pennsylvania: Courier Forschungs-Institut Senckenberg, v. 241, p. 19–29. Weems, R.E., 2004, Bothriolepis virginiensis, a valid species of placoderm fish separable from Bothriolepis nitida: Journal of Vertebrate Paleontology, v. 24, no. 1, p. 245–250, doi:10.1671/20. Weems, R.E., Beem, K.A., and Miller, T.A., 1981, A new species of Bothriolepis (Placodermi: Bothriolepidae) from the Upper Devonian of Virginia (USA): Proceedings of the Biological Society of Washington, v. 94, p. 984–1004. Werdelin, L., and Long, J.A., 1986, Allometry in the placoderm Bothriolepis canadensis and its significance to antiarch evolution: Lethaia, v. 19, p. 161–169, doi:10.1111/j.1502-3931.1986.tb00727.x. Wilson, H.M., Daeschler, E.B., and Desbiens, S., 2005, New flat-backed archipolypodan millipedes from the Upper Devonian of North America: Journal of Paleontology, v. 79, p. 738–744, doi:10.1666/0022-3360 (2005)079[0738:NFAMFT]2.0.CO;2.
MANUSCRIPT ACCEPTED BY THE SOCIETY 20 DECEMBER 2010
Printed in the USA
The Geological Society of America Field Guide 20 2011
An introduction to structures and stratigraphy in the proximal portion of the Middle Devonian Marcellus and Burket/Geneseo black shales in the Central Appalachian Valley and Ridge Terry Engelder Rudy Slingerland Michael Arthur Department of Geosciences, The Pennsylvania State University, University Park, Pennsylvania 16802, USA Gary Lash Department of Geosciences, SUNY Fredonia, Fredonia, New York 14063, USA Daniel Kohl D.P. Gold Department of Geosciences, The Pennsylvania State University, University Park, Pennsylvania 16802, USA
ABSTRACT A Marcellus-Burket/Geneseo field trip in the Appalachian Valley and Ridge features both brittle and ductile structures. The degree to which these structures have developed depends on both lithology, which is a function of the stratigraphic architecture of the Devonian Appalachian Basin and position relative to the foreland during the Alleghanian Orogeny. Joints are best developed in the black shales and the units immediately above with the J2 joint set most prominent in the Brallier Formation just above the Burket/Geneseo Formation. Faults are seen in the form of cleavage duplexes and bedding-parallel slip accompanying flexural-slip folding. Cleavage duplexes are found in the Marcellus whereas bedding-parallel slip is more common in the overlying Mahantango Formation and further up the section in the Brallier Formation. Layer-parallel shortening decreases from greater than 50% to approximately 10% when crossing the Jacks Mountain–Berwick Anticline structural front from the hinterland portion to the foreland portion of the Valley and Ridge. Disjunctive cleavage, the primary mechanism for layer-parallel shortening, is best developed in carbonates whereas pencil cleavage is best developed in shales.
Engelder, T., Slingerland, R., Arthur, M., Lash, G., Kohl, D., and Gold, D.P., 2011, An introduction to structures and stratigraphy in the proximal portion of the Middle Devonian Marcellus and Burket/Geneseo black shales in the Central Appalachian Valley and Ridge, in Ruffolo, R.M., and Ciampaglio, C.N., eds., From the Shield to the Sea: Geological Field Trips from the 2011 Joint Meeting of the GSA Northeastern and North-Central Sections: Geological Society of America Field Guide 20, p. 17–44, doi: 10.1130/2011.0020(02). For permission to copy, contact
[email protected]. ©2011 The Geological Society of America. All rights reserved.
17
18
Engelder et al.
OVERVIEW During the first decade of the twenty-first century, the American energy portfolio was dramatically realigned after operators in Texas learned to use horizontal drilling and massive hydraulic (slickwater) fracture stimulations to recover natural gas from the Mississippian Barnett black shale, a source rock once considered of little economic value. In 2004, the Middle Devonian Marcellus black shale of the Appalachian Basin was first tested in a vertical well using slickwater fracturing. Late in 2007, Range Resources announced that their horizontal wells in the Marcellus were producing natural gas and associated liquids at economic rates. A Penn State press release on 17 January 2008 introduced the vast economic potential of natural gas production from the Marcellus to the public (Engelder and Lash, 2008). Production from the Marcellus through 30 June 2010 validated the earlier volumetric calculations verifying that the Marcellus of the Appalachian Basin was a super giant gas field and one of the largest in the world (Engelder, 2009). Because of the economic value of natural gas in the Marcellus, this formation has become a central focus for geological research in Appalachia. This field trip visits some of the key outcrops straddling a minimum of 1420 m of the Lower to Upper Devonian section in the Appalachian Valley and Ridge with a focus on the Marcellus black shale (Fig. 1). Of particular interest is the extent to which the structural fabric in the Middle Devonian section is dependent on the sedimentary geometry of the Appalachian Basin (Lash and Engelder, 2011). Sedimentary geometry is in turn controlled by a combination of global eustasy and a more local tectonically driven sea level change (Ettensohn, 1985). The fundamental observations for this trip were laid through a century of field work and herein we weave the threads of this work together in a context of a modern understanding of plate tectonics, sequence stratigraphy, and fracture mechanics. For example, the Marcellus Formation and associated strata, consist of at least four thirdorder depositional sequences, dividing the strata into lowstand (LST), transgressive (TST), and highstand systems tracts (HST) (Fig. 2). These sequences are defined by sequence boundaries associated with the Oriskany Sandstone (The Wallbridge Unconformity; Swezey, 2002), within the Selinsgrove (Senaca) Member of the Onondaga Limestone, the base of the Purcell/Cherry Valley Limestone, and the base of the Stafford Limestone (Lash and Engelder, 2011) (Fig. 2). The outcrops of the Middle Devonian section in the Appalachian Valley and Ridge reflect Laurentian plate tectonics from deposition during the early Acadian Orogeny (ca. 389 Ma) through the Alleghanian Orogeny (ca. 265 Ma). Before 390 Ma, the southeastern margin of Laurentia was loaded by the convergence of a microcontinent, Avalonia, which had an active volcanic arc (Ettensohn, 1987). This load created the accommodation space for the Marcellus transgressions (Ettensohn, 1994). At this time, Gondwana and Laurentia were separated by an ocean basin that was closing rapidly. The Middle Devonian Eifelian and Givetian Stages are characterized by two black shale sequences
(i.e., Marcellus Formation and the Geneseo/Burket Formation) both of which will be visited during this field trip (Fig. 1). Convergence culminated at approximately 315 Ma when Gondwana and Laurentia collided obliquely and slid past each other for at least 15 m.y. (Ferrill and Thomas, 1988). The oblique slip between Gondwana and Laurentia was dextral with Gondwana slipping west relative to Laurentia. This phase of the Alleghanian orogeny is characterized by major strike-slip faults
Figure 1. Stratigraphic column of the Devonian section in the vicinity of the Juniata Culmination of the Appalachian Valley and Ridge (adapted from the AAPG CSUNA Northern Appalachian Sheet). All thicknesses given in meters (Lindberg, 1985).
Middle Devonian Marcellus and Burket/Geneseo black shales including the Brevard fault zone and other faults in the Appalachian Piedmont (Gates et al., 1988) and a joint set that is found along the length of the Central and Southern Appalachians (the J1 set; Engelder and Whitaker, 2006). By 290 Ma, Laurentia and Gondwana ceased to slip in a dextral fashion and locked at the New York Promontory. The New York Promontory served as a fulcrum for a clockwise pivot of Gondwana relative to Laurentia (Hatcher, 2002). This clockwise pivot drove Gondwana directly into Laurentia, causing the fold-thrust belt of the Central and Southern Appalachians (Hatcher et al., 1989). The Northern and Maritime Appalachians were characterized by transform tectonics with no sign of a foreland arising from a direct continentcontinent collision. The timing of the head-on collision between Africa (Gondwana) and North America (Laurentia) during the
19
Alleghanian Orogeny of the Central and Southern Appalachians is poorly constrained but may have lasted as much as 35 m.y. (300 Ma to 265 Ma), depending on what geological indicator is used as a measure of this orogeny. Our field trip will cross the Appalachian Valley and Ridge by traveling from the Appalachian Plateau side of the Allegheny Front (Stops 1–2) to the middle coal fields of the Anthracite District (Stops 5–7) (Fig. 3). The trip starts along the Allegheny Front where beds are overturned, passes through the Nittany Anticlinorium where layer-parallel shortening strain is 15%, and then crosses the Jacks Mountain–Berwick Anticline structural front where layer-parallel shortening strain is greater than 50% on the hinterland side (Nickelsen, 1983). A layer-parallel shortening strain of 50% in the Marcellus is indicated by a strong
Figure 2. Gamma ray (GR; in American Petroleum Institute units) and bulk density (RHOB; in g/cm3) log typical of the Marcellus Shale and associated strata. This log is part of the suite of logs from the Appalachian Valley and Ridge that Penn State’s Appalachian Basin Black Shale Group has acquired with industrial support. LST— lowstand systems tract; TST—transgressive systems tract; HST—highstand systems tract; SB—sequence boundary; MFS— maximum flooding surface.
TERTIARY
MD.
INDIANA
12
BLAIR
20
11
30
1-2
MISSISSIPPIAN
50 MI
77°
UNION
PERRY
JUNIATA
DEVONIAN
ADAMS
CUMBERLAND
4
SULLIVAN
DAUPHIN
NO
D
MD.
ORDOVICIAN
YORK
76°
CHESTER
CAMBRIAN
LANCASTER
WAYNE
42°
MONTGOMERY
N.J.
LOWER PALEOZOIC
DEL.
41°
PIKE
75°
75°
75°
BUCKS
DELAWARE
LEHIGH
NORTHAMPTON
CARBON
MONROE
LACKAWANNA
BERKS
LUZERNE
WYOMING
SUSQUEHANNA
76°
SCHUYLKILL
LEBANON
BE UM H RT
N RLA
COLUMBIA
LYCOMING
SILURIAN
77°
N.Y. BRADFORD
SNYDER 5-7
3
TIOGA
80 KM
CLINTON
EXPLANATION
FRANKLIN
40 60
MIFFLIN 8
CENTRE
HUNTINGDON 9
78°
40
POTTER
20
10
78°
FULTON
PENNSYLVANIAN
BEDFORD
CAMBRIA
CAMERON
McKEAN
CLEARFIELD
ELK
PERMIAN
79°
SOMERSET
JURASSIC AND TRIASSIC
W.VA.
FAYETTE
WESTMORELAND
ARMSTRONG
JEFFERSON
FOREST
WARREN
79°
0
40°
PRECAMBRIAN
75°
Figure 3. The geology of Pennsylvania showing the location of field stops along the Allegheny Front (Stops 1 and 2) and into the Valley and Ridge. Dashed line indicates the approximate location of the Jacks Mountain–Berwick Anticline structural front. (Adapted from PA-DCNR Map 7; www.dcnr.state.pa/us/topogeo.)
QUATERNARY
GREENE
80°
ALLEGHENY
N.Y.
CLARION
VENANGO
BUTLER
80°
R
WASHINGTON
BEAVER
LAWRENCE
MERCER
CRAWFORD
ERIE
KE LA
IE ER
SCALE1:2,000,000
OU
40°
41°
42°
OHIO
10
NT A
0
ER IV
R
IL
MO
W.VA.
RE
WA
J.
LA
N.
DE
PH
Y.
PH IA
N.
DE L
20 Engelder et al.
Middle Devonian Marcellus and Burket/Geneseo black shales disjunctive cleavage, strain fibers about pyrite nodules, and cleavage duplexes. The second day of this trip (Stops 8–12) runs in the opposite direction across the Jacks Mountain–Berwick Anticline structural front. STOP 1. J2 (CROSS-FOLD OR DIP) JOINTS IN THE FORM OF NATURAL HYDRAULIC FRACTURES IN THE BRALLIER GRAY SHALE Brallier Formation: Chronostratigraphic Equivalent to the Genesee Group of New York Series (Eur. Stage): Middle Devonian (Givetian) Location: Howard, Pennsylvania (outcrop on north side of road along Route 150 between Milesburg and Howard) Coordinates: 40.992980° by −77.709799° (datum WGS84 spheroid) Background: Joint growth in the Middle and Upper Devonian section of the Appalachian Basin correlates with the presence of black shale (Lash et al., 2004; Lash and Engelder, 2005; Engelder, et al., 2009). Two major joint sets are found with the earlier set, J1, better developed within black shale and the later set, J2, better developed within the gray shale immediately above. In western areas of the basin where black shales are stacked, joints populate each black shale. In eastern portions of the basin where the Marcellus is the sole black shale, joint development decreases with distance above the black shale until the Carboniferous section is reached where jointing is very sparse. The correlation between jointing and black shale is largely a function of the driving mechanism for joint propagation which comes from internal pressure during maturation of kerogen to produce hydrocarbons, mainly natural gas (Lacazette and Engelder, 1992). These are natural hydraulic fractures (NHF) where tectonic stress controls the orientation of the joint set but tectonic stress is not the cause of joint propagation. Some of the best examples of black-shale– related NHF are found above the Geneseo/Burket black shale in the Finger Lakes District, New York (Fig. 4). Observations: Route 150 from Port Matilda to Bald Eagle State Park follows just south of a line of low ridges marking the position of the more resistive siltstones of the Brallier Formation. To the north of the road in several outcrops a well developed J2 joint set is seen in numerous of outcrops of the Brallier Formation (Fig. 5). The Burket Member, a black shale, is exposed just under the Brallier Formation in a small quarry near Bald Eagle State Park (Stop 2). The trip will stop at a Brallier outcrop Dbh-31-RU before stopping at Dbh-30-RU which is the small quarry of Burket black shale (Fig. 5). The general rule of thumb for joint development in Devonian black shales of the Appalachian Basin is that J1 is better developed in black shale and J2 is better developed in gray shale. Our first stop is an outcrop of Brallier sitting just above the black shale of the Burket (Geneseo) black shale, the oldest Upper Devonian black shale above the Hamilton Group. The basal portion of the Brallier tends to be finer grained with coarser turbidites appear-
21
ing farther up section (Fig. 6). We will visit the Brallier twice on the field trip with the second outcrop being higher in the section where turbidites vary in thickness up to nearly a meter (Stop 12). The difference in joint development between Stop 1 and Stop 12 is the vertical height to spacing with joints at Stop 1 having a height that far exceeds spacing. At Stop 12, the height to spacing ratio is roughly 1:1, a characteristic of well-developed mechanical beds and a characteristic of turbidite or carbonate beds interrupting shale beds. At Stop 1, the silt beds are thin (<5 cm), and the shale is laminated on such a fine scale that it acts as a single mechanical unit relative to joint growth (Fig. 6). When the spacing of joints is much closer than height, the standard explanation of jointing by bed-parallel extension (the stress-shadow theory) (Gross et al., 1995) does not apply. Another mechanism for closely spaced joints is joint-parallel compression; a model that does not work for joint propogation at depth under the rules of linear elastic fracture mechanics (Lorenz et al., 1991). Rather, the preferred model for driving joints in the Brallier at Stop 1 is a NHF mechanism for which the standard stress-shadow theory does not apply (Fischer et al., 1995). The NHF mechanism is appealing for joint development with a height to spacing ratio in excess of 10, a situation found only in the Devonian section of the Appalachian Basin when gray shale occurs just above black shale. The best examples of this behavior are found above the Geneseo (Burket) black shale in
Figure 4. J2 joints propagating as natural hydraulic fractures into the Penn Yan gray shale just above the Geneseo/Burket black shale at Taughannock Falls State Park, New York. Both units are members of the Genesee Group of western New York. The Penn Yan is chronologically equivalent to the Brallier Formation in Pennsylvania.
22
Engelder et al.
Figure 5. Present orientation of J2 joints in the vicinity of Bellefonte, Pennsylvania. Equal-area net projection. Dbh— Brallier and Harrel Formations undivided; Dh—Hamilton Group; dashed line—outcrop bedding; dotted line— local fold axis trend. Adapted from Engelder et al. (2009).
Fillmore Glen and Taughannock Falls State Parks in New York State (Fig. 6). At Stop 1 and in the previously stated locations in New York, it is the J2 joint that displays a height to spacing ratio greater than 10. The idea is that thermal maturation of the black shale reaches a peak at maximum burial during the Alleghanian orogeny when the tectonic stress field is in the cross-fold (i.e., J2) orientation (Engelder and Geiser, 1980). One characteristic of the entire Allegheny Front is that J2 is particularly well developed in the Brallier (Fig. 5). STOP 2. J2 (CROSS-FOLD OR DIP) JOINTS IN THE BURKET BLACK SHALE Burket Formation (sometimes called the Burket Member of the Harrell Formation): Chronostratigraphic equivalent to Geneseo black shale of the Genesee Group in New York Series (Eur. Stage): Middle Devonian (Givetian) Location: Howard, Pennsylvania (quarry along Route 150 between Milesburg and Howard) Coordinates: 40.998761° by −77.699038°
Observations: The Burket Formation, a black shale, is exposed just under the Brallier in a small quarry near Bald Eagle State Park. This black shale correlates with the Geneseo black shale of the Genesee Group, New York. Of note here is the absence of the J1 joint set which is present in many exposures of the Geneseo black shale in central New York. J2 joints are moderately developed, perhaps because of the low total organic carbon (TOC) in the parent rock. The thermal maturation of samples taken from the Burket black suggest that the Marcellus and shallower gas shales are prospective right up to the Allegheny Front (Table 1).
TABLE 1. SAMPLES SENT TO HUMBLE LABS FOR TOTAL ORGANIC CARBON AND ROCK-EVAL MEASUREMENTS TOC S1 S2 S3 Tmax Ro (calc) Fossil Plant 3.22 1.56 4.28 0.29 457 1.07 Matrix 0.90 0.37 0.62 0.06 453 0.99
Middle Devonian Marcellus and Burket/Geneseo black shales
Figure 6. An example of a J2 joint cutting through laminated bedding in the Harrell Formation at Stop 1.
STOP 3. OVERTURNED OATKA CREEK FORMATION WITH J1 AND J2 JOINTS PROPAGATING AROUND CONCRETIONS Oatka Creek Member of the Marcellus Formation of the Hamilton Group Series (Eur. Stage): Middle Devonian (Eifelian) Location: Antis Fort, Pennsylvania (Snook quarry along Old Fort Road off Route 44 west of Antis Fort) Coordinates: 41.191066° by −77.239754° Background: An ENE joint set was the first to propagate in many outcrops of Devonian, Mississippian, and Pennsylvanian rocks of the Central and Southern Appalachian Mountains (Kulander and Dean, 1993; Nickelsen, 1979; Nickelsen and Hough, 1967; Pashin and Hinkle, 1997). These early joint sets (here called J1) strike parallel to the orientation of the maximum horizontal stress, SH, in a stress field that was a prelude to the Alleghanian orogeny when Gondwana was sliding with dextral sense of slip relative to Laurentia between 315 Ma and 300 Ma. In total, these joint sets appear as one mega-set recording a rectilinear stress field extending for greater than 1500 km across
23
three promontories separated by oroclinal embayments of the Central and Southern Appalachians (Fig. 7). Given the arrangement of continental plates at the time of joint propagation, the plates were in the southern hemisphere and the maximum horizontal stress, SH, controlling the direction of propagation was oriented south of east relative to the earth’s axis between 315 Ma and 300 Ma. Observations: One of the most compelling cases for the pre- to early Alleghanian propagation of J1 joints is found at the Snook quarry in Antis Fort. Here, a relatively gray Oatka Creek Member of the Marcellus is overturned to dip 72° to the south. The view looking north in the Snook Quarry is the underside of bedding with J2 joints cutting vertically through the overturned bedding (Fig. 8). J1 joints are seen cutting from upper left to lower right when looking northward toward the underside of bedding. In map view, the acute angle between J2 and J1 appears clockwise from J1. Because this is the underside of bedding, the acute angle between J2 and J1 in map view is counterclockwise from J1. When observations were first made at Antis Fort in 2006, concretions of all sizes up to greater than 1 m could be seen in bedding (Fig. 8). J2 joints abut, but don’t cut the larger concretions as is expected for natural hydraulic fracturing. Some J2 joints are mineralized as was the case for the Eastern Gas Shales Project core recovered from the deeper portion of the Marcellus over 200 km to the west of Antis Fort (Evans, 1995). J1 joints have a shallow dip to the east (Fig. 9). When bedding is rotated to horizontal, joints of the J1 set are returned to a vertical position with a vector mean strike of 053°. J1 joints in overturned beds have approximately the orientation of J1 joints elsewhere in the Appalachian Basin, which supports the hypothesis that these are early and have survived 10%–15% layer-parallel shortening as measured nearby (Faill, 1977). However, two observations that temper the J1 hypothesis for joints in the Snook quarry are their weak cluster which looks like the clustering of ENE joints in the Hudson Valley fold-thrust belt and the fact that their strike is as much as 20° counterclockwise from the best developed J1 sets in black shale in the Finger Lakes District, New York. Rotation of the J1 joints to their position in horizontal bedding is accomplished about a rotation angle of 108°, assuming a plunge of 3° to the east for a fold axis at 074°. The rotation does not move the azimuthal mean to poles of joints to the horizontal (Fig. 9). Rather the rotation of bedding to horizontal leaves the joints dipping steeply to the south, on average. This phenomenon can be seen in Figure 9 where J1 joints appear to make an angle with bedding of ~85°. One interpretation is that bedding was subject to a layer-parallel shear but this shear is inconsistent with flexural slip folding which should be given the joints a steep dip to the north. At present the origin of this steep dip to the south is unknown. Flexural slip did take place as the Marcellus was overturned. This is indicated by slip fibers not only on bedding planes but also in the surface of concretions (Fig. 10). Differential slip between bedding and concretions is common other parts of the Valley and Ridge (Nickelsen, 1979).
24
Engelder et al.
A
B
Figure 7. (A) The distribution of ENE joint sets along the Appalachian Mountains. Reference to insets and their map locations (dashed rectangles) are found in Engelder and Whitaker (2006). (B) A time line for coal deposition and propagation of Appalachian cleats and joints. Ages are consistent with International Commission on Stratigraphy (www.stratigraphy. org) whereas the stage names are North American. AWSF—Appalachian-wide stress field.
Middle Devonian Marcellus and Burket/Geneseo black shales The most interesting aspect of the Rock-Eval work is that TOC in concretions relative to matrix (Table 2). During compaction TOC is preserved. Using a simple volumetric strain calculation, 39% of the volume of the initial rock (presumably sea floor mud) had to be removed to concentrate organic matter from 0.45% to 0.74%. This is consistent with compaction measurements made for concretions in Devonian shale elsewhere in the Appalachian Basin (Lash and Blood, 2007). The samples for TOC were taken toward the top of the Oatka Creek. Overturned Purcell and organic rich Oatka Creek are visible down section at the eastern end of the quarry (Fig. 11). STOP 4. UNION SPRINGS MEMBER OF THE MARCELLUS WITH J1 JOINTS PROPAGATING AROUND CONCRETIONS AND TILTED BY FOLDING Union Springs Member of the Marcellus Formation of the Hamilton Group Series (Eur. Stage): Middle Devonian (Eifelian) Location: Elimsport, Pennsylvania (Finck quarry along Pikes Peak Road off of Route 44 east northeast of Elimsport) Coordinates: 41.138081° by −76.992672°
25
TABLE 2. SAMPLES SENT TO HUMBLE LABS FOR TOTAL ORGANIC CARBON AND ROCK-EVAL MEASUREMENTS TOC S1 S2 S3 Tmax Ro (calc) Concretion 0.45 0.05 0.11 0.09 416 ? Matrix 0.74 0.08 0.2 0.05 469 1.28
TABLE 3. SAMPLES SENT TO HUMBLE LABS FOR TOTAL ORGANIC CARBON AND ROCK-EVAL MEASUREMENTS TOC S1 S2 S3 Tmax Ro (calc) Matrix 6.05 0.04 0.22 0.56 593 3.51
Background: Jumping over to the south flank of the Nittany Anticlinorium moves us into the transition between the Allegheny Front where gas shale is prospective to a region of the Valley and Ridge where gas shale is overmature (Table 3). In fact, industry dogma at the time of preparation of this field guide is that vigorous leasing of the Marcellus should remain north of an E-W line marked by PA Route 118 in Lucerne, Columbia, and Lycoming Counties. PA Route 118 is a virtual extension of the Allegheny Front east of the Susquehanna River. Observations: The Finck quarry at Elimsport exposes the Union Springs Member of the Marcellus somewhere above the top bentonite in the Marcellus (Fig. 12). The organic content of the shale (TOC > 6%) reveals that this portion of the Marcellus is in the hot bottom section as observed on gamma-ray logs (Fig. 2). J1 joints are well developed in this portion of the Marcellus and exhibit the characteristics of NHF by passing around some large concretions within the Union Springs (McConaughy and Engelder, 1999). In general, the surfaces are not as planar as seen in outcrops of black shale in the Finger Lakes District, New York. Like the J1 joints in the Snook quarry at Antis Fort, these joints form normal to bedding. When bedding is rotated to horizontal, the J1 joints return to vertical, again a sign of a prefolding origin. Also, like the Snook quarry, the J1 joints have a relatively weak cluster. In the Finck quarry the vector mean strike for the J1 set is 061°. While clustering is weak at both Stops 3 and 4, it is noteworthy that the orientation of J1 joints on both sides of the Nittany Anticlinorium are counterclockwise by approximately 10° from the strike of J1 joints in the Finger Lakes District of New York. This is consistent with the orientation of J1 joints along the entire Appalachian chain (Engelder and Whitaker, 2006). STOP 5. CLEAVED LIMEY SHALE AT THE BASE OF THE MAHANTANGO
Figure 8. Examples of joint development in the Oatka Creek Member of the Marcellus at the Ed Snook quarry along Old Fort Road off Route 44 west of Antis Fort. Bedding is overturned at 074°/72° (strike and dip by the right-hand rule). (A) Photo taken about March 2007. (B) Photo taken September 2008 after nearly a meter of bedding, including the layer of concretions, had been ripped away.
Mahantango Formation of the Hamilton Group Series (Eur. Stage): Middle Devonian (Eifelian) Location: Sunbury, Pennsylvania (road cut along Route 147 less than 2 miles south of the intersection with Route 61 in Sunbury, Pennsylvania) Coordinates: 40.839552° by −76.806962°
26
Engelder et al.
Figure 9. Examples of joint development in the Oatka Creek Member of the Marcellus at the Ed Snook quarry along Old Fort Road off Route 44 west of Antis Fort. Bedding is overturned at 082°/65°. View looking west and parallel to J1 joints. Joints plotted in present coordinates (left) and rotated to their position with horizontal bedding using a fold axis plunging 05° toward 082° with a rotation of 115° (right).
Middle Devonian Marcellus and Burket/Geneseo black shales
27
Figure 10. Left: Concretion with fibers indicating differential slip between the concretion and bedding of Marcellus. Right: In places the Marcellus is fossil rich. Psilophytes are seen at the Ed Snook quarry, for example. These are primitive vascular plants known as white ferns with both stems and branches looking like thick cylindrical pieces of grass. The picture, above, is the stem of a plant. Note the crinoids that have attached to the stem. (Swiss Army knife shown for scale.)
Background: Day 1 focuses on the Marcellus in the transition between the Allegheny Front and the Anthracite District of the Valley and Ridge. The initial two stops are along the Allegheny Front in the Brallier distal turbidites and Burket/ Geneseo black shale. The Marcellus is first encountered on the north side of Bald Eagle Ridge where rocks are vertical to overturned on the northern limb of the Nittany Anticlinorium (Stop 3). Next is a look at the Marcellus on the south limb of the Nittany Anticlinorium (Stop 4). The first four stops sit in a region where layer-parallel shortening (LPS) measures little more than that encountered within the Appalachian Plateau detachment sheet (<15%) (Engelder and Engelder, 1977). Stops 5–7 are south of the Jacks Mountain–Berwick Anticline structural front
where LPS can approach 50%. They are also on strike with and approximately15 km east of the famous Bear Valley Strip Mine (sample K of Nickelsen, 1979). In general, J2 joints in the region strike between 340° and 350° (Fig. 13). The larger LPS on the hinterland side of the Jacks Mountain–Berwick Anticline structural front appears to have no affect on the regional development of J2 joints. Observations: This outcrop is at the contact between the Marcellus and Mahantango Formations with the bottom portion of the Mahantango consisting of a limy shale. Disjunctive cleavage in the limy shale is strong which means that LPS was greater than 24% according to the scale by Alvarez et al. (1978). Nickelsen (1983) estimates that LPS was greater than 30% in the thick cleavage duplexes. Cleavage in the Mahantango is 260°/75°. Prominent cross fold joints have strike consistent with the regional pattern (Fig. 14). One interpretation for this cleavage duplex is a detachment thrust tipping out in the Mahantango just to the north of Stop 5 (Nickelsen, 1986). Small-scale folds are found on the north flank of the Selinsgrove Junction anticline, a second-order fold in the Appalachian Valley and Ridge. STOP 6. FORELAND TRANSPORT ON CLEAVAGE DUPLEX IN UNION SPRINGS FORMATION
Figure 11. Overturned Purcell exposed at Antis Fort quarry.
Union Springs Member of the Marcellus Formation of the Hamilton Group Series (Eur. Stage): Middle Devonian (Eifelian) Location: Selinsgrove Junction, Pennsylvania (quarry along Route 147 near the Selinsgrove Railroad Bridge over the Susquehanna River) Coordinates: 40.801418° by −76.838735°
28
Engelder et al.
Figure 12. Examples of joint development in the Union Springs Member of the Marcellus at the Delmar Finck quarry along Pikes Peak Road off of Route 44 east northeast of Elimsport, Pennsylvania. Bedding is 075°/10°. Joints plotted in present coordinates (left) and rotated to their position with horizontal bedding using a fold axis plunging 05° toward 075° with a rotation of 10° (right).
Middle Devonian Marcellus and Burket/Geneseo black shales
29
Figure 13. Present orientation of crossfold joints in the vicinity of the Selinsgrove Syncline. Equal-area net projection. Dbh—Brallier and Harrel Formations undivided; Dh—Hamilton Group; dashed line—outcrop bedding; dotted line—local fold axis trend. The Berwick Anticline structural front passes just north of the Selinsgrove Syncline. Adapted from Engelder et al. (2009).
Observations: Two shale pits in the Union Springs Member of the Marcellus Formation are located on the east side of State Route 147. Beds in these pits dip to the NNW. The top portion of the Selinsgrove Limestone (i.e., the Onondaga Limestone of the Appalachian Plateau) is exposed in a road cut just to the south of the southern pit (Fig. 15). Disjunctive cleavage in this limestone is consistent with LPS on the south side of the Jacks Mountain– Berwick Anticline structural front. The north shale pit exhibits a thin cleavage duplex dipping to the north. The duplex represents considerable shear strain in the form of a sigmoidal cleavage with a sense of vergence (i.e., thrusting) toward the Appalachian foreland to the
NNW (Figs. 16 and 17). The sigmoidal cleavage terminates abruptly against the floor and roof thrusts at strong strain discontinuities (Fig. 18). Such cleavage duplexes are common in both outcrops and core of the Marcellus within the Valley and Ridge. We will see another example at Stop 9. These would be detachments with far greater displacement than seen on bedding surfaces exhibiting fibrous growths of slickensides indicative of flexural-slip folding. No cleavage or other evidence of LPS is evident in overlying or underlying shale although it must be present as indicated by the extent to which the underlying limestone is cleaved without evidence of detachment from the shale.
Figure 14. The Selinsgrove Junction second order anticline, showing the stratigraphic setting in the thin and thick cleavage duplexes. Stop 5 will be equivalent to the region marked Figure 9, which is above the Purcell Limestone. Stop 6 is the area marked Figure 7 with a north-dipping thin cleavage duplex. Adapted from Nickelsen (1986).
30
Engelder et al.
TABLE 4. SAMPLES SENT TO HUMBLE LABS FOR TOTAL ORGANIC CARBON AND ROCK-EVAL MEASUREMENTS TOC S1 S2 S3 Tmax Ro (calc) Oatka Creek (Stop 7) 0.78 0.07 0.04 0.02 * ? Union Springs (Stop 6) 3.19 0.03 0.03 0.45 * ? Cleavage duplexes 1.59 0.01 0.02 1.42 * ? (Stop 6) *Tmax unreliable due to poor S2 peak.
Like the Marcellus at Stop 7, few joints have developed in this rock. The Marcellus outcrops near Stop 9 and at Stop 11 have a well developed J2 joint set. The reason why neither J1 nor J2 joints are better developed in the area of Selinsgrove Junction remains a mystery. Thermal maturity of organic-rich rocks in the anthracite district of Pennsylvania is so mature that estimates of Tmax and Ro are not possible (Table 4). STOP 7. OATKA CREEK MEMBER CONTAINING A HIGH ANGLE FAULT WITH A TRANSFER ZONE Oatka Creek Member of the Marcellus Formation of the Hamilton Group Series (Eur. Stage): Middle Devonian (Eifelian) Location: Sunbury, Pennsylvania (road cut at the intersection of Route 147 and State Highway 4018 SH) Coordinates: 40.834580° by −76.809977° Observations: Stop 7 is a thick section located in the Oatka Creek Member of the Marcellus black shale (Fig. 19). Here, neither J1 nor J2 joints are particularly well developed. Some of the silty layers in the black shale have joints exhibiting a spacing equivalent to bed thickness. These are parallel to the local
Figure 15. Disjunctive cleavage in a 20-cm-thick bed in the upper portion of the Onondaga Limestone (i.e., the Selinsgrove Limestone) at Selinsgrove Junction.
Figure 16. Correlated stratigraphic sections showing cleavage duplexes, thrust faults, and strain gradients. Section 3 is Stop 6. Section 4 is Stop 7. Section 5 is Stop 5. Adapted from Nickelsen (1986).
Middle Devonian Marcellus and Burket/Geneseo black shales
31
Figure 17. Cleavage duplex at Selinsgrove Junction. Looking to the east with tectonic transport toward the foreland (left). See Nickelsen (1986).
Figure 18. A major slip surface between the cleavage duplex, above and more weakly strained Union Springs shale, below.
fold axis and believed to have formed as a consequence of layerparallel stretching as the local fold developed near a high angle fault exhibiting a transfer zone (Fig. 20). Penn State’s Appalachian Basin Black Shale Group (ABBSG) sampled the Marcellus in two core holes within 3 km to the east of this outcrop. In core, the evidence of LPS is particularly strong in the Purcell Limestone, the middle member of the Marcellus Formation (Fig. 21). With these structures we have seen the manifestation of LPS in the limy portion of the basal Mahantango (Stop 5), the top of the Selinsgrove Limestone (Stop 6) and in the Purcell Limestone.
vegetation has been successful, particularly over the Marcellus Shale slopes, where rill wash has exposed bedrock. From the overpass bridge, anchored in limestones of the Keyser Formation, the following lithologies were encountered successively as one progresses northwestward by the Old Port Formation made up of the following members: Ceoymans (Limestone), New Scotland (Limestone), Mandata (black pyritic shale), Schriver (variegated light gray to white yellow chert with brown to purple interlayers and rusty hydration rinds) and Ridgley (Sandstone) and the Needmore Shale Member of the Onondaga Formation. As one progresses uphill to the west, faulting juxtaposes the Ridgley Sandstone and Needmore Shale units as well as the upper Onondaga strata of the Selinsgrove Limestone Member and the basal black shale beds of the Marcellus Formation. The latter two units were exposed in the steeper slope (2.25H:1V) above the road bed. Approximately 100 feet of the Marcellus black shales (Union Springs Member) is present in the road bank, bounded below (upright) and above (overturned) by the Selinsgrove Limestone in a tight overturned syncline (Fig. 22). The rest of the slope to the northwest is underlain by overturned Needmore Shale, Ridgley Sandstone, and Schriver Chert. Their distribution, of these units, photographed during the construction, is shown in Figure 23, with the axis of the overturned syncline in the center of the black band (Marcellus Shale). The Union Springs Member of the Marcellus Shale at Stop 8 is a highly fissile, dark gray to black shale with very fine grained framboidal pyrite, generally disseminated throughout, but more abundant near the base; tightly folded lenses, 5– 10 cm thick, of framboidal pyrite concentrations were observed near the center of the outcrop. Pyrite leaching is apparent in the orange stain (yellow boy) in the rip-rap at the bottom of the bank below the Marcellus Shale outcrop. The source is blooms
STOP 8. TIGHTLY FOLDED SYNCLINE IN THE UNION SPRINGS MEMBER OF THE MARCELLUS Union Springs Member of the Marcellus Formation of the Hamilton Group Series (Eur. Stage): Middle Devonian (Eifelian) Location: Lewistown, Pennsylvania (road cut along the 522 bypass west southwest of Lewistown) Coordinates: 40.581116° by −77.626651° Observations: This road cut is of interest for the repetition of the lower Devonian strata exposed and repeated in an overturned syncline and fault, as well as the relics of early mining activity (Fig. 23). Iron ore (secondary limonite and jarosite) was mined along the contact between the Selinsgrove Limestone and the Marcellus Shale. Sand has been extracted locally from the Ridgeley Member of the Old Port Formation. The construction of this interchange required engineering around existing underground mine workings, readjusting slope grade to stabilize the shaly rocks prone to slumping, and covering over the carbonrich black shales that produce acid drainage. Not all of the re-
32
Engelder et al.
of efflorescent minerals (copiapite and melanterite) that may be present (depending on weather conditions) on the surface of the Mandata Shale, and in the overhang areas on the Marcellus slopes. In the latter, pyrite-rich zones occur a meter or two above the Tioga-A ash bed and pyrite nodules up to 5 cm long were found in the Tioga-B ash bed. Seven ash beds (meta-bentonites) were exposed in the road cut (see Fig. 24), consistent with those
reported elsewhere by Way et al. (1986). The total carbon and organic hydrocarbon content of three samples of Marcellus Shales ranges from 6.78 to 11.5%. Trace elements with elevated concentration (V, Ni, Cu, Cr, Mn, Ag, Au, etc.,) are consistent with those of black shale deposits. The tight syncline is a third-order fold with numerous faults and fourth-order folds within it. The syncline is overturned and
Figure 19. Oatka Creek Member of the Marcellus along Route 147 south of Sunbury, Pennsylvania. Joints plotted in present coordinates (left) and rotated to their position with horizontal bedding using a fold axis plunging 00° toward 265° with a rotation of 32° (right). E-W joints are interpreted as fold-related, whereas J2 joints are well developed near the top of the Oatka Creek and up into the cleaved limy beds of the Mahantango Formation.
Middle Devonian Marcellus and Burket/Geneseo black shales
33
Figure 20. High-angle fault with a transfer zone.
faulted with beds dipping 60°–80°. The syncline plunges out approximately1.5 km to the west of Stop 8. The fault is interpreted as a back thrust (south vergent). Along the westbound exit ramp, a complete section of the footwall may be sampled, including the Needmore and Selinsgrove Members of the Onondaga Limestone, the Ridgley Sandstone, the Shriver Chert, and the black shales of the Mandata Formation. The organic content of this section is typical of the Union Springs and the thermal maturity is consistent with other samples taken on the south side of the Jacks Mountain structural front (Table 5). TABLE 5. SAMPLES SENT TO HUMBLE LABS FOR TOTAL ORGANIC CARBON AND ROCK-EVAL MEASUREMENTS TOC S1 S2 S3 Tmax Ro (calc) Union Springs 6.99 0.02 0.07 0.20 * ? *Tmax unreliable due to poor S2 peak.
Figure 21. Core from the Purcell Member of the Marcellus showing a vertical extension of pyrite nodules, cleavage about a carbonate concretion, and calcite veins filling a cleavage duplex. This core comes from the Handiboe well located less than 2 km to the east of Stop 7.
Figure 22. Structural cross section at Stop 7. View looking ENE. See Figure 1. Key to some formation names: Dop—Old Port; DSk—Keyser; Sto—Tonoloway.
34
Engelder et al.
Figure 23. Union Springs Member of the Marcellus Formation along the Route 522 bypass west southwest of Lewistown, Pennsylvania. (A) Marcellus in the core of an overturned syncline with axis plunging to the ENE and axial plane dipping NNW. Needmore and Selinsgrove Members of the Onondaga Limestone immediately above and below the Marcellus. (B) Upright kink fold near the axial plane of the overturned syncline with Marcellus in the core. (C) View of uncharted iron ore drift in Ridgley Sandstone on the southwest slope. These underground workings were discovered during construction, when a bulldozer was damaged by a cave-in. (Photograph by Tom McElroy.)
Middle Devonian Marcellus and Burket/Geneseo black shales STOP 9. FOLDED UNION SPRINGS MEMBER OF THE MARCELLUS WITH A TIOGA ASH BED, CLEAVAGE DUPLEX, AND SMALL-SCALE BUCKLE FOLDS Union Springs Member of the Marcellus Group of the Hamilton Group Series (Eur. Stage): Middle Devonian (Eifelian) Location: Newton-Hamilton, Pennsylvania (Forgy quarry along Ferguson Valley Road east of Newton-Hamilton) Coordinates: 40.397553° by −77.825627° Background: The Kistler railroad cut is the finest exposure of the Marcellus in the state of Pennsylvania and the only complete section of which we are aware (Fig. 25). The section is dip-
35
TABLE 6. SAMPLES SENT TO HUMBLE LABS FOR TOTAL ORGANIC CARBON AND ROCK-EVAL MEASUREMENTS TOC S1 S2 S3 Tmax Ro (calc) Union Springs 8.63 0.05 0.79 1.04 * ? *Tmax unreliable due to poor S2 peak.
ping gently to the WNW and just to the SE of the axis of a third order syncline that folds the Marcellus just to the SE of the Jacks Mountain structural front. The outcrop has a number of structural features including deformed strain markers, small-scale buckle folds, and pencil cleavage. The transition between the Onondaga Limestone and the Marcellus is particularly impressive (Fig. 26). Here, the Onondaga is an interlayered carbonate and shale, which is why it has a gamma ray signature that fluctuates with an API (American Petroleum Institute) reading well above that for a massive limestone like the Tully. Interbeds of scraggy limestone are seen at the top of the Union Springs Member. The Purcell is intensely cleaved. The Oatka Creek is a dark gray shale with local gray-black layers of higher organic content. The J2 joint set is well developed in the Union Springs Member. (See also Table 6.) Observations: The Forgy Quarry carries one bentonite which is a 4 cm layer of crystalline tuff with biotite (Fig. 27B). We believe this layer to be the seventh or top ash bed, the G-Layer of Way et al. (1986). The G bentonite serves as a superb marker bed in a 4 m section that Penn State graduate student Reed Bracht has mapped (Fig. 28). The notable layers in the Forgy quarry include a carbonate (unit 1) that we take to be the top of the transition zone from the Onondaga Limestone. At the Newton-Hamilton railroad cut the transition zone is several meters thick with shale layers thickening as the limestone layers thin. These layers are not scraggy carbonates like those at the top of the Union Springs. The 4-cm crystalline tuff is a half meter above the top limestone. Unit 6 is a layer that remains relatively intact and can be traced throughout the quarry. Unit 8 is an unusually low density material that is not even lithified in spots where it is similar to the gumbo-clay recovered from depths over 10,000 ft in the Tertiary section of the Texas Gulf Coast. Unit 10 is a cleavage duplex with sense of vergence toward the Appalachian foreland to the WNW (Fig. 27A). This is thickest of several cleavage duplexes in the Forgy quarry. The cleavage duplexes are a manifestation of a complexly faulted and folded Marcellus that is characteristic of its behavior to the SE of the Jacks Mountain structural front. STOP 10. WELL-DEVELOPED J2 JOINTS IN THE BRALLIER SILTSTONES
Figure 24. (A) View to northeast along the axis of the overturned syncline in center of black shale band. Beds of Selinsgrove Limestone were exposed at the road bed level below (upright) and in the slope above (overturned) the black band. (Photograph by Tom McElroy.) (B) View to northeast of three ash beds (meta-bentonites) in the Marcellus Shale at road-level during construction. (Photograph by D.P. Gold. Hugh Barnes for scale.)
Brallier Formation of the Genesee Group Series (Eur. Stage): Middle Devonian (Givetian) Location: Huntingdon, Pennsylvania (road cut along Penn Street off Route 22 in Huntingdon) Coordinates: 40.477211° by −77.997870°
36
Engelder et al.
Background: The major driving mechanism for joint propagation in the Middle Devonian section of the Appalachian Basin is natural hydraulic fracturing (Lacazette and Engelder, 1992; McConaughy and Engelder, 1999). Because the pressure for this drive develops during the maturation of hydrocarbons in source rocks, it is expected that source rocks should be most heavily fractured. The affinity between jointing and black shales is manifest by a decreasing joint density with thickness of section above the black shales (Fig. 29). Throughout the Appalachian Basin, the non-source rock that carries the most completely developed joint set is found right above black shales. This stop is an example of joint development above a black shale, the Burket/Geneseo. Observations: The Brallier Formation is a clastic unit with distal turbidites and shale interbedded immediately over the Burket black shale. This unit is often called the undivided BrallierHarrell where the Burket is the black shale member of the Harrell. Like portions of Mahantango above the Marcellus, this unit has a significant number of sheet sands that act as distinct mechanical units. With such mechanical units, the pattern of fracturing in the Brallier is distinct from other units visited during this field trip. The Brallier, like its counterpart in New York (the Ithaca Formation), gradually becomes more thickly bedded up section. At Stop 1 where the lower portion of the Brallier is exposed near the Burket, the siltstone interlayers were thinner and finer grained. J2 joints propagated through these thinner mechanical beds without stopping at bed boundaries. The same is true of this stop where
the earliest joints are mineralized J2 joints. In this outcrop, there is no evidence for J1 joints which favor black shales of the Appalachian Basin. At this stop, three episodes of joint propagation are evident starting with the mineralized J2 set often covered with euhedral crystals of quartz (Ruf et al., 1998). The second set consists of strike joints with either unmineralized surfaces or coated with a delicate pattern of microscopic crystals of unknown composition. Statistical analysis indicates that the third episode of jointing is a late-stage J2 joint set behaving like cross joints (Ruf et al., 1998). Certainly, the later J2 joints abut strike joints more commonly than the other way around (Fig. 30). It is, however, common to see these cross joints (late J2 orientation) cross cut the strike joints in the Brallier (Fig. 30). The strike joints are tilted slightly relative to bedding, a sign of fold-related joint growth (Engelder and Peacock, 2001). The development of surface morphology on the joints of the Brallier siltstones is magnificent. Two sets of systematic joints cutting the same bed may exhibit different rupture styles (Engelder, 2004). Joints oriented parallel to the strike of bedding formed prior to dip-parallel joints, as inferred from cross-cutting relationships. The strike joints typically have a surface morphology consistent with that of a short blade crack, whereas the dip joints exhibit a more complex morphology (Fig. 30). The earlier joints have surfaces with a typical plume-related topography (i.e., 1–3 mm within any cm2) that greatly exceeds the grain size
Figure 25. Google Earth™ image of the Kistler railroad cut. Penn State students are actively involved in a number of projects involving the Kistler section, including studying the effect of weathering on the Marcellus.
Middle Devonian Marcellus and Burket/Geneseo black shales
37
(<0.125 mm) of the host bed whereas the later joints have surfaces that are smooth to the touch and a topography on the order of the grain size of the host. The complex, irregular surface morphology on dip joints resembles a frosty window (Savalli and Engelder, 2005). Joint surfaces often contain one or more irregular primary plume axes with several small secondary detachment ruptures (as indicated by secondary plume axes) branching off of them. The detached ruptures behave as individual crack tips each propagating independently and each having a unique propagation velocity (vtl). One detached rupture may outrun an adjacent rupture. It is common for such detached ruptures to terminate against or cut off other ruptures. As a result, the bed-bounded joint surface is a composite of numerous secondary ruptures whose growth direction and vtl were impacted by nearby crack-tip stress concentrations. These are interpreted as subcritical joints with a much
slower propagation velocity. In this outcrop, the frosty-window morphology is interpreted as indicative of the slow growth of cross joints during exhumation. In Devonian clastic sections dominated by interlayered siltstones and shales, joint initiation usually starts in the siltstone layer (McConaughy and Engelder, 2001). During NHF propagation, least horizontal stress (Sh) is the governing parameter in dictating whether siltstones or shales should joint first and siltstones appear to carry the lower Sh (Engelder and Lacazette, 1990). This is largely because during consolidation, siltstones have a lower consolidation coefficient, which leads to the lower Sh during compaction (Karig and Hou, 1992). The difference in horizontal stress leads to later jointing in shales at a higher fluid pressure. If there is no rotation of the principal stresses, fluid-driven joints will propagate into the shale in plane with the earlier joints in siltstone. However, if the horizontal stress does rotate, then later,
Figure 26. Top and bottom portions of the Union Springs Member of the Marcellus along the Conrail railroad cut at Kistler, Pennsylvania. (A) The top portion showing scraggy lime layers that mark the transition between the Purcell Member (above) and Union Springs Member (below). (B) The bottom transition to the Onondaga Limestone.
Figure 27. Lower Union Springs Member of Marcellus in the Forgy quarry along Ferguson Valley Road east of Newton-Hamilton. (A) Cleavage duplex with vergence toward the foreland (top to the left). Michael Arthur for scale. (B) Tioga ash bed.
38
Engelder et al.
Figure 28. A 4-m stratigraphic column of the Union Springs Member of the Marcellus black shale at the Forgy quarry, Newton-Hamilton, Pennsylvania. Mapped by Penn State graduate student Reed Bracht. The rose diagrams given here are in the form of half-strike plots.
Middle Devonian Marcellus and Burket/Geneseo black shales higher fluid pressures will drive en echelon cracks (i.e., fringe cracks) into bounding shale beds (Pollard et al., 1982; Carter, et al., 2001; Younes and Engelder, 1999). Fluid-driven jointing in the Brallier at Huntingdon is witnessed by the trapping pressures of fluid inclusions in euhedral quartz along early J2 joints (Lacazette, 1991; Srivastava and Engelder, 1991). The Braillier also has the same NHF pattern as found in the Ithaca Formation with fringe cracks being driven from the interface of a parent joint (Fig. 31). STOP 11. DIPPING OATKA CREEK MEMBER OF THE MARCELLUS WITH J1 AND J2 MUTUALLY CROSS CUTTING Oatka Creek Member of the Marcellus Formation of the Hamilton Group Series (Eur. Stage): Middle Devonian (Eifelian) Location: Huntingdon, Pennsylvania (Cold Springs Road ~1 miles off of Route 26 to State College) Coordinates: 40.554403° by −77.963796° Background: One of the fundamental rules of joint propagation is that rupture is perpendicular to the plane of least principal stress (σ3). As σ3 occupies only one orientation, particularly in a static state, it is a physical impossibility for simultaneous crack growth in more than one orientation. Hence, when an outcrop contains multiple joint sets, there have been at least two propagation events with a rotation of stress between the events. During dynamic rupture, the stress field can be very complicated
Figure 29. Joint density of J2 joints in 2-D as measured in a series of outcrops along Route 15 between Williamsport, Pennsylvania and the New York–Pennsylvania border. Joint density was measured in cross section as cumulative joint length intersecting an area of 100 m2 as viewed along strike of joints.
39
on a local basis, but with the joints observed during this field trip, static fracture mechanics governs propagation. Abutting is a manifestation of the later joint set. However, in gas shales it is common to observe cross cutting joint sets (Fig. 32). The mechanism for cross cutting versus abutting is based on the projection of the crack-tip stress field in front of the later joints as they propagate. For deeply buried joints, the normal stress on early joints is large enough that a frictional contact between the walls of the early joint allows for an elastic distortion in the vicinity of the crack-tip cross over to the other side of the early joint. If the crack-tip distortion crosses an early joint, then the rupture of a later joint can cross the earlier joint. If the early joint is open and not in frictional contact, the elastic distortion of the crack tip from the propagation of the later joint is not transmitted across the early joint. In this case, the later joint will arrest at the earlier joint, thus abutting but not crossing the early joint. Observations: The Oatka Creek Member of the Marcellus at Hootenanny quarry, on the north flank of the Broadtop Syncline, has one of the nicest examples of cross cutting J1-J2 joint sets development found in the Valley and Ridge Marcellus (Fig. 33). Both joint sets are normal to bedding and rotate to vertical when bedding is restored to horizontal. The sharp corners of blocks defined by the cross-cutting joints are well developed. The outcrop also contains neotectonic joints with irregular planes. Aside from their irregular or curving planes and their non-systemic nature, there is very little else to allow a distinction between the J1-J2 sets and the curving neotectonic joints. The Hootenanny quarry is what the outcrop of Marcellus featured in Figure 32 might look like in cross section. The intersection of
Figure 30. Joints in interbedded siltstone of the Brallier Formation in a road cut along Penn Street off Route 22 in Huntingdon, Pennsylvania (Ruf et al., 1998). Late-stage J2 joints (chalk on two different joint surfaces with a hammer leaning against one of these) abutting a strike joint (joint propagating toward hammer).
40
Engelder et al.
joints in the Hootenanny quarry is at a smaller acute angle than joints at the Oatka Creek outcrop in Figure 32. As a general rule of thumb, J1 joints are better developed in black shales of the Appalachian Plateau. This is the case in the outcrop of the Oatka Creek (Fig. 32). Gray shale immediately over black shale has a better developed J2 joint set. This is true for the Brallier throughout the area covered by Day 2 of this field trip (Fig. 34). STOP 12. DIPPING UNION SPRINGS MEMBER OF THE MARCELLUS WITH INTERNAL LIMESTONE BEDS Union Springs Member of the Marcellus Group of the Hamilton Group Series (Eur. Stage): Middle Devonian (Eifelian)
Location: Frankstown, Pennsylvania (New Enterprise quarry off Locke Mountain Road in Frankstown) Coordinates: 40.435286° by −78.342129° Background: Up to 80 or more Lower and Middle Devonian (Lochkovian to Eifelian) K-bentonites in the Appalachian Basin are distributed through the succession as four major clusters of 8–15 closely spaced beds or as scattered multiple to single beds (ver Straeten, 2004). The most prominent cluster is known as the Tioga A through G ash beds/K-bentonites. Observations: The Union Springs Member of the Marcellus at the New Enterprise quarry in Frankstown is in the same structural position as the Finck quarry in Elimsport (Stop 4),
Figure 32. Cross cutting J1 and J2 joints. J1 joints strike left to right and J2 joints strike back to front.
Figure 31. (A) J2 joints in the Ithaca Formation at Taughannock Falls State Park where multiple en echelon cracks propagate down into shale from a siltstone-shale interface (Younes and Engelder, 1999). The parent joint cuts a siltstone. (B) J2 joints in the Brallier Formation in Huntingdon, Pennsylvania, with multiple en echelon cracks propagating upward into a siltstone. The parent joint cuts a silty shale.
Figure 33. The Marcellus at the Hootenanny quarry. Photo looking to the WNW along J2 joints. J1 joints cut parallel to the road and define the faces of blocks in this view. Duff Gold for scale.
Middle Devonian Marcellus and Burket/Geneseo black shales
Dbh-101-RU
Dbh-112-RU
Dbh-109-RU
Dbh-108-RU
Dbh-107-RU
Dbh-113-RU
Hootenanny Stop 11 Dbh-118-RU
Dbh-111-RU
N Huntingdon Stop 10
Dbh-110-RU
Dbh-117-RU
Frankstown Stop 12 Newton-Hamilton Stop 9
Dbh-106-RU
Dbh-116-RU
Dbh-105-RU
Dh-115-RU
Dbh-103-RU 0
Dbh-114-RU
10 km
Dbh-102-RU
Dbh-104-RU
Figure 34. Present orientation of cross-fold joints in the vicinity of the Broadtop Syncline. Equal-area net projection. Dbh—Brallier and Harrell Formations undivided; Dh—Hamilton Group; dashed line—outcrop bedding; dotted line—local fold axis trend. Joints at Hootenanny quarry plotted in present coordinates (left) and rotated to their position with horizontal bedding using a fold axis plunging 00° toward 050° with a rotation of 16° (right).
41
42
Engelder et al.
Figure 35. Examples of joint development in the Union Springs Member of the Marcellus at the New Enterprise quarry off Locke Mountain Road in Frankstown, Pennsylvania. Bedding is 336°/27°. Joints plotted in present coordinates (right) and rotated to their position with horizontal bedding.
Middle Devonian Marcellus and Burket/Geneseo black shales
43
Figure 36. Ash bed with low density, high total organic carbon shale at Frankstown, Pennsylvania.
Figure 37. Lingula found in the Union Springs Member of the Marcellus at Frankstown, Pennsylvania.
which is to say that this location is on the hinterland side of the Nittany Anticlinorium but to the foreland of the Jacks Mountain– Berwick Anticline structural front. The thermal maturation of the rock at Stops 4 and 12 is also similar and in both cases maturation has crossed into the non-prospective realm (compare Tables 3 and 7). Joints in this section of Union Springs are less well clustered than their counterpart along the Newton-Hamilton railroad cut. The strong N-S joint trend resembles that seen at NewtonHamilton railroad cut, but its significance is unknown (Fig. 35). Marcellus in the New Enterprise quarry carries three ash beds which are interpreted to be the Tioga E-, F- and G-ash beds of Way et al. (1986) (Fig. 36). An unusual bed of Lingula occurs toward the base of the Union Springs (Fig. 37). The top of the Onondaga Limestone is seen at the base of the exposure. Exposures of the Oriskany Sandstone are seen on the western wall of the quarry.
REFERENCES CITED
ACKNOWLEDGMENTS The Appalachian Basin Black Shale Group, an industrial affiliates group, supported research found in this guidebook.
TABLE 7. SAMPLES SENT TO HUMBLE LABS FOR TOTAL ORGANIC CARBON AND ROCK-EVAL MEASUREMENTS TOC S1 S2 S3 Tmax Ro (calc) Union Springs matrix 4.12 0.40 1.01 0.17 557* 2.87 Onondaga transition 0.10 0.03 0.07 0.16 * ? zone * Tmax unreliable due to poor S2 peak.
Alvarez, W., Engelder, T. and Geiser, P., 1978, Classification of solution cleavage in pelagic limestones: Geology, v. 6, p. 263–266, doi:10.1130/0091 -7613(1978)6<263:COSCIP>2.0.CO;2. Carter, B.J., Ingraffea, A.R., and Engelder, T., 2001, Natural hydraulic fracturing in bedded sediments: International Association for Computer Methods and Advances in Geomechanics: Annual Meeting, Tucson Arizona, p. 1–10. Engelder, T., 2004, Tectonic implications drawn from differences in the surface morphology on two joint sets in the Appalachian Valley and Ridge, Virginia: Geology, v. 32, p. 413–416, doi:10.1130/G20216.1. Engelder, T., 2009, Marcellus 2008: Report card on the breakout year for gas production in the Appalachian Basin: Fort Worth Basin Oil and Gas Magazine, August, p. 18–22. Engelder, T., and Engelder, R., 1977, Fossil distortion and decollement tectonics of the Appalachian Plateau: Geology, v. 5, p. 457–460, doi:10.1130/0091 -7613(1977)5<457:FDADTO>2.0.CO;2. Engelder, T., and Geiser, P.A., 1980, On the use of regional joint sets as trajectories of paleostress fields during the development of the Appalachian Plateau, New York: Journal of Geophysical Research, v. 85, p. 6319–6341, doi:10.1029/JB085iB11p06319. Engelder, T., and Lacazette, A., 1990, Natural hydraulic fracturing, in Barton, N., and Stephansson, O., eds., Rock Joints: Rotterdam, A.A. Balkema, p. 35–44. Engelder, T., and Lash, G.G., 2008, Marcellus shale’s vast resource potential creating stir in the Appalachia: American Oil and Gas Reporter, v. 51, p. 76–87. Engelder, T., and Peacock, D., 2001, Joint development normal to regional compression during flexural-slow folding: The Lilstock buttress anticline, Somerset, England: Journal of Structural Geology, v. 23, p. 259–277, doi:10.1016/S0191-8141(00)00095-X. Engelder, T., and Whitaker, A., 2006, Early jointing in coal and black shale; evidence for an Appalachian-wide stress field as a prelude to the Alleghanian Orogeny: Geology, v. 34, p. 581–584, doi:10.1130/G22367.1. Engelder, T., Lash, G.G., and Uzcategui, R., 2009, Joint sets that enhance production from Middle and Upper Devonian gas shales of the Appalachian Basin: American Association of Petroleum Geologists Bulletin, v. 93, p. 857–889. Ettensohn, F.R., 1985, The Catskill Delta complex and the Acadian Orogeny: A model, in Woodrow, D.L., and Sevon, W.D., eds., The Catskill Delta: Geological Society of America Special Paper 201, p. 39–49. Ettensohn, F.R., 1987, Rates of relative plate motion during the Acadian orogeny based on the spatial distribution of black shales: The Journal of Geology, v. 95, p. 572–582, doi:10.1086/629150.
44
Engelder et al.
Ettensohn, F.R., 1994, Tectonic controls on formation and cyclicity of major Appalachian unconformities and associated stratigraphic sequences: SEPM Concepts in Sedimentology and Paleontology #4, Tectonic and Eustatic Controls on Sedimentary Cycles, p. 217–242. Evans, M.A., 1995, Fluid inclusions in veins from the Middle Devonian shales; a record of deformation conditions and fluid evolution in the Appalachian Plateau: Geological Society of America Bulletin, v. 107, p. 327–339, doi:10.1130/0016-7606(1995)107<0327:FIIVFT>2.3.CO;2. Faill, R.T., 1977, Fossil distortion, Valley and Ridge Province, Pennsylvania: Geological Society of America Abstracts with Programs, v. 9, p. 262. Ferrill, B.A., and Thomas, W.A., 1988, Acadian dextral transpression and synorogenic sedimentary successions in the Appalachians: Geology, v. 16, p. 604–608, doi:10.1130/0091-7613(1988)016<0604:ADTASS> 2.3.CO;2. Fischer, M.P., Gross, M.R., Engelder, T., and Greenfield, R.J., 1995, Finiteelement analysis of the stress distribution around a pressurized crack in a layered elastic medium; implications for the spacing of fluid-driven joints in bedded sedimentary rock: Tectonophysics, v. 247, p. 49–64, doi:10.1016/0040-1951(94)00200-S. Gates, A.E., Speer, J.A., and Pratt, T.L., 1988, The Alleghanian southern Appalachian Piedmont: A transpressional model: Tectonics, v. 7, p. 1307–1324, doi:10.1029/TC007i006p01307. Gross, M.R., Fischer, M.P., Engelder, T., and Greenfield, R.J., 1995, Factors controlling joint spacing in interbedded sedimentary rocks: Integrating numerical models with field observations from the Monterey Formation, USA, in Ameen, M.S., ed., Fractography: Fracture Topography as a Tool in Fracture Mechanics and Stress Analysis: Geological Society of London Special Publication 92, p. 215–233. Hatcher, R.D., 2002, Alleghanian (Appalachian) orogeny, a product of zipper tectonics: Rotational transpressive continent-continent collision and closing of ancient oceans along irregular margins, in Martinez Catalán, J.R., Hatcher, R.D., Jr., Arenas, R., and Díaz García, F., eds., VariscanAppalachian Dynamics: The Building of the Late Paleozoic Basement: Geological Society of America Special Paper 364, p. 199–208. Hatcher, R.D., Jr., Thomas, W.A., Geiser, P.A., Snoke, A.W., Mosher, S., and Wiltschko, D.V., 1989, Alleghanian orogen in Hatcher, R.D., Jr., Thomas, W.A., and Viele, G.W., eds., The Appalachian-Ouachita Orogen in the United States: Boulder, Colorado, Geological Society of America, Geology of North America, v. F-2, p. 233–318. Karig, D. E., and Hou, G., 1992, High-stress consolidation experiments and their geological implications: Journal of Geophysical Research, v. 97, p. 289–300, doi:10.1029/91JB02247. Kulander, B.R., and Dean, S.L., 1993, Coal-cleat domains and domain boundaries in the Allegheny Plateau of West Virginia: American Association of Petroleum Geologists Bulletin, v. 77, p. 1374–1388. Lacazette, A, 1991, Natural hydraulic fracturing in the Bald Eagle sandstone in Central Pennsylvania and the Ithaca siltstone at Watkins Glen, New York [Ph.D. thesis]: University Park, Pennsylvania, The Pennsylvania State University, 224 p. Lacazette, A., and Engelder, T., 1992, Fluid-driven cyclic propagation of a joint in the Ithaca siltstone, Appalachian Basin, New York, in Evans, B. and Wong, T-F., eds., Fault Mechanics and Transport Properties of Rocks: London, Academic Press Ltd., p. 297–324. Lash, G.G., and Blood, D.R., 2007, Origin of early overpressure in the Upper Devonian Catskill Delta Complex, western New York state: Basin Research, v. 19, p. 51–66, doi:10.1111/j.1365-2117.2007.00318.x. Lash, G.G., and Engelder, T., 2005, An analysis of horizontal microcracking during catagenesis: Example from the Catskill Delta Complex: American Association of Petroleum Geologists Bulletin, v. 89, p. 1433–1449. Lash, G.G., and Engelder, T., 2011, Thickness trends and sequence stratigraphy of the Middle Devonian Marcellus Shale, Appalachian Basin: Implications for Acadian Foreland basin evolution: American Association of Petroleum Geologists Bulletin, v. 95, p. 61–103. Lash, G.G., Loewy, S., and Engelder, T., 2004, Preferential jointing of Upper Devonian black shale, Appalachian Plateau, USA: Evidence supporting
hydrocarbon generation as a joint-driving mechanism, in Cosgrove, J.W., and Engelder, T., eds., The Initiation, Propagation, and Arrest of Joints and Other Fractures: Geological Society of London Special Publication 231, p. 129–151. Lindberg, F.A., ed., 1985, Northern Appalachian Region: Tulsa, Oklahoma, American Association of Petroleum Geologists, Correlation of Stratigraphic Units of North America (COSUNA) Project. Lorenz, J.C., Teufel, L.W., and Warpinski, N.R., 1991, Regional Fractures I: A mechanism for the formation of regional fractures at depth in flat-lying reservoirs: American Association of Petroleum Geologists Bulletin, v. 75, p. 1714–1737. McConaughy, D.T., and Engelder, T., 1999, Joint interaction with embedded concretions: Joint loading configurations inferred from propagation paths: Journal of Structural Geology, v. 21, p. 1637–1652, doi:10.1016/S0191 -8141(99)00106-6. McConaughy, D.T., and Engelder, T., 2001, Joint initiation in bedded clastic rocks: Journal of Structural Geology, v. 23, p. 203–221, doi:10.1016/ S0191-8141(00)00091-2. Nickelsen, R.P., 1979, Sequence of structural stages of the Allegheny orogeny at the Bear Valley Strip Mine, Shamokin, Pennsylvania: American Journal of Science, v. 279, p. 225–271, doi:10.2475/ajs.279.3.225. Nickelsen, R.P., 1983, Aspects of Alleghanian Deformation, in Nickelsen, R.P., and Cotter, E., eds., Silurian depositional history and Alleghanian Deformation in the Pennsylvania Valley and Ridge: Guidebook for the 48th Annual Field Conference of Pennsylvania Geologists: Harrisburg, Pennsylvania, Pennsylvania Geological Survey, p. 29–39. Nickelsen, R.P., 1986, Cleavage duplexes in the Marcellus shale of the Appalachian foreland: Journal of Structural Geology, v. 8, p. 361–371, doi:10.1016/0191-8141(86)90055-6. Nickelsen, R.P., and Hough, V.N.D., 1967, Jointing in the Appalachian Plateau of Pennsylvania: Geological Society of America Bulletin, v. 78, p. 609– 629, doi:10.1130/0016-7606(1967)78[609:JITAPO]2.0.CO;2. Pashin, J.C., and Hinkle, F., 1997, Coalbed methane in Alabama, Circular: Geological Survey of Alabama, Report 192, p. 71. Pollard, D.D., Segall, P., and Delaney, P., 1982, Formation and interpretation of dilatant echelon cracks: Geological Society of America Bulletin, v. 93, p. 1291–1303, doi:10.1130/0016-7606(1982)93<1291:FAIODE> 2.0.CO;2. Ruf, J.C., Rust, K.A., and Engelder, T., 1998, Investigating the effect of mechanical discontinuities on joint spacing: Tectonophysics, v. 295, p. 245–257, doi:10.1016/S0040-1951(98)00123-1. Savalli, L., and Engelder, T., 2005, Mechanisms controlling rupture shape during subcritical growth of joints in layered rock: Geological Society of America Bulletin, v. 117, p. 436–449, doi:10.1130/B25368.1. Swezey, C.S., 2002, Regional stratigraphy and petroleum systems of the Appalachian Basin, North America: Denver, Colorado, U.S. Geological Survey Information Services, Geological Investigations Series Map I-2768. ver Straeten, C.L., 2004, K-bentonites, volcanic ash preservation, and implications for Early to Middle Devonian volcanism in the Acadian orogen, eastern North America: Geological Society of America Bulletin, v. 116, p. 474–489, doi:10.1130/B25244.1. Way, J.H., Smith, R.C., and Roden, M., 1986, Detailed correlations across 175 miles of the Valley and Ridge of Pennsylvania using 7 ash beds in the Tioga Zone, in Sevon, W.D., ed., Selected geology of Bedford and Huntington Counties: 51st Annual Field Conference of Pennsylvania Geologists, p. 55–72. Younes, A., and Engelder, T., 1999, Fringe cracks: Key structures for the interpretation of progressive Alleghanian deformation of the Appalachian Plateau: Geological Society of America Bulletin, v. 111, p. 219–239, doi:10.1130/0016-7606(1999)111<0219:FCKSFT>2.3.CO;2.
MANUSCRIPT ACCEPTED BY THE SOCIETY 11 JANUARY 2011
Printed in the USA
The Geological Society of America Field Guide 20 2011
Pennsylvanian climatic events and their congruent biotic responses in the central Appalachian Basin David K. Brezinski* Maryland Geological Survey, 2300 St. Paul Street, Baltimore, Maryland 21218, USA, and Section of Invertebrate Paleontology, Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213, USA Albert D. Kollar* Section of Invertebrate Paleontology, Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213, USA
ABSTRACT Pennsylvanian strata of western Pennsylvania exhibit evidence of a hierarchy of paleoclimatic changes. Long-term (107 years) climate trends reflect plate movement and tectonic events. These long-term trends are overprinted by changes of much shorter duration (100–400 k.y., and 10–20 k.y.). During deposition of the Pottsville and Allegheny formations (Bashkirian-Moscovian), the Appalachian climate exhibited perhumid to humid situations during periods of glacial advance, and humid to dry subhumid conditions during glacial retreats. Marine faunas and coal swamp floras during this interval of time exhibited a remarkably consistent taxonomic and ecological structure. Tetrapod amphibian faunas were highly aquatic. When the Conemaugh Group was deposited, the ancient Appalachian climate became progressively drier. Glacial stages were dry subhumid and during deglaciation semiarid to arid. This reduction in precipitation produced changes in coal-forming floras, as lycopsiddominated assemblages gave way to tree fern–dominated associations. Coincident with this climatic drying, tetrapod faunas became highly terrestrial in the basin. During the deposition of the Monongahela Group, the Appalachian climate returned to humid conditions during glacial periods. However, there is evidence of drier subhumid conditions during the intervening interglacial episodes as indicated by the pervasive presence of mudcracked nonmarine limestones. Nested lacustrine cycles within the Monongahela Group indicate short-term alternations between wet and dry periods that may have been driven by Earth’s precession. Coal-forming mires continued to be dominated by tree ferns, and vertebrate faunas tended to be found within fluvial lake environments. The latest Pennsylvanian and/or early Permian strata exhibit a return to Conemaugh-like deposition as evidenced by the pervasiveness of redbeds, dry climate floras, and highly terrestrial vertebrate faunas.
*
[email protected];
[email protected]. Brezinski, D.K., and Kollar, A.D., 2011, Appalachian Pennsylvanian climatic events and their congruent biotic responses, in Ruffolo, R.M., and Ciampaglio, C.N., eds., From the Shield to the Sea: Geological Field Trips from the 2011 Joint Meeting of the GSA Northeastern and North-Central Sections: Geological Society of America Field Guide 20, p. 45–60, doi:10.1130/2011.0020(03). For permission to copy, contact
[email protected]. ©2011 The Geological Society of America. All rights reserved.
45
46
Brezinski and Kollar
INTRODUCTION The Pennsylvanian Subsystem derives its name from rocks exposed in western Pennsylvania. This period of Earth’s history spans an interval of time of approximately twenty million years, from 318 to 299 million years ago (Gradstein et al., 2004). In ascending order, Pennsylvanian strata of western Pennsylvania consist of the Pottsville and Allegheny formations, and the Conemaugh and Monongahela groups (Edmunds et al., 1999). Current nomenclature parallels the subdivisions of Pennsylvanian rocks outlined by Rogers (1858). Rogers divided Pennsylvanian strata of western Pennsylvania into lower and upper coal-bearing intervals alternating with lower and upper barren or coal-poor intervals (Fig. 1). This early subdivision of the Late Carboniferous rocks of western Pennsylvania was based on the abundance, thickness, and areal extent of mineable coal beds. Cecil (1990) proposed that lithostratigraphic alternations of Pennsylvanian strata, known as cyclothems, are the manifestation of long-term oscillations of wet and dry conditions in the Appalachian climate. Consequently, lithostratigraphic subdivision of the Pennsylvanian appears related to long-term (107 year) climate changes. During these times, globally cool periods would cause all polar ice sheets to advance, which produced a sharp drop in sea level. While the high latitudes experienced increased levels of precipitation in the form of snow, the low latitudes experienced rainfall (Cecil et al., 1985). When the
Figure 1. Correlation diagram of Pennsylvanian stages, lithostratigraphic nomenclature, and early nomenclature according to Rogers (1858). Generalized climate curve modified from Frakes et al. (1992).
climate warmed, snowfall decreased and polar glaciers retreated, while lower latitudes experienced reduced levels of precipitation. The trend of long-term change in Appalachian Pennsylvanian climate is from ever-wet conditions during the Bashkirian, (lower Pennsylvanian) through the late Moscovian (middle Pennsylvanian) when evaporation periodically exceeded precipitation, to increasingly drier and drier conditions in the late Kasimovian and Gzhelian (upper Pennsylvanian) (Cecil et al., 1985). This progressive reduction in rainfall was followed by a temporary reversal in climatic trends to increased wetness during the latest Gzhelian. Decreased rainfall then continued into the Early Permian (Fig. 1). The long-term drying and warming trend can be subdivided into three stages. The climate shifted, first from tropical-seasonal in the early to middle Pennsylvanian (Bashkirian to early Moscovian), to wet-dry cycles in the late middle Pennsylvanian (late Moscovian), finally to dry-arid cycles in the Gzhelian (Cecil, 1990; Joeckel, 1999; Cecil et al., 2004). The climatic changes are evidenced by identifiable changes in lithology, particularly the presence, prominence, and character of the coals and paleosols (Fig. 2). Thus, Bashkirian to middle Moscovian coal beds are thick, regionally developed, and economically mineable. Thick early to middle Pennsylvanian coal beds indicate that tropical conditions extended throughout the Appalachian Basin, and into the Illinois Basin (Cecil, 1990). The interglacial periods of rainfall were, however, still sufficiently intense enough to produce well-leached, high-alumina paleosols, commonly known as flint clays. In terms of stratigraphic units applied to the central Appalachian Basin, this climate regime persisted throughout the deposition of the Pottsville and lower Allegheny formations, but ended by the deposition of the middle part of the Allegheny Formation. During the late Moscovian (upper Allegheny Formation), the wet climatic conditions of the earlier Pennsylvanian began to shift markedly toward a drier regime. This drying is indicated by the deposition of the lower and upper Freeport limestones. Deposition of the lower Conemaugh Group (Kasimovian) is characterized by localized, thin, uneconomical coal beds, red paleosols, and interbedded thin marine intervals. At the top of the Glenshaw Formation, thick, red, calcic-Vertisols (Fig. 2) (Joeckel, 1995a; Cecil et al., 1998, 2004), exemplified by the Pittsburgh Redbeds, are characteristic of the carbonate-rich paleosols produced during periods of low precipitation and high rates of evaporation (Joeckel, 1995a; Retallack, 2001). Near the middle of the Conemaugh Group, the Ames Limestone records extensive marine flooding of the upper Pennsylvanian in the Appalachian Basin (Brezinski, 1983). Its thickness and widespread extent suggest that sea level was near or at the maximum height for the Pennsylvanian. This widespread Appalachian inundation indicates that glacial melting in the southern polar hemisphere was extensive. The stratigraphic succession that characterizes the middle part of the Conemaugh Group, including the Ames Limestone, is coincident with a marked retreat of southern polar hemisphere glaciers and reduced deposition of high-latitude glacial tillites (Frakes et al., 1992; Birgenheier et al., 2009).
Appalachian Pennsylvanian climatic events In the Casselman Formation of the Conemaugh Group, coal beds become even patchier, while red paleosols become thicker and better developed. Flint clays, which were formed in wet, glacial periods of the early to middle Pennsylvanian, are replaced by thin, nonmarine limestones and paleosols (Cecil, 1990). The seasonally dry nonmarine limestones become progressively more prominent through the upper Conemaugh and Monongahela groups. The dry climate conditions marking the middle part of the Conemaugh Group were followed by a progressive reversal of the climate in the overlying Monongahela Group. A return to higher levels of precipitation in the uppermost Conemaugh and overlying Monongahela is indicated by a decreasing prominence of red paleosols and a renewed deposition of thick coals, such
47
as the Pittsburgh, Redstone, Sewickley, and Waynesburg. This return to high levels of precipitation is commensurate with the deposition of Rogers’ (1858) Upper Coal Measures. The climatic reversal to wetter conditions at the end of the Pennsylvanian is not restricted to the Appalachian Basin, as correlative strata in the U.S. midcontinent also reflect increased precipitation during the late Virgilian (Joeckel, 1995b). Although the increased precipitation and coal formation marked by deposition of the Monongahela Group reflect a reversal from the long-term drying of Appalachian Pennsylvanian climate, the short-term climate cycles took on a renewed dominance and cylothemic character. A wet Appalachian climate indicates a renewed dominance of glacial periods in polar regions as well. Thick coals suggestive of wet paleoclimate contrast with
Figure 2. Pennsylvanian stratigraphy of the central Appalachian Basin with interpreted precipitation curve (modified from Cecil et al., 1985; Cecil, 1990; and Cecil et al., 2004) with important paleoclimatic indicators and notable biotic influences.
48
Brezinski and Kollar
the intervening nonmarine limestone deposits that were presumably deposited during drier interglacial periods. Monongahela lacustrine deposits exhibit nested repetitions of shale and limestone that indicate alternating wet and dry conditions on an even shorter scale than that which produced the cyclothemic depositional patterns (Fig. 2). Above the Monongahela Group, the lower Dunkard Group exhibits depositional indicators that suggest a return to increased aridity (C.B. Cecil, 2010, oral commun.). While the lower part of the group contains several significant coals, these deposits become progressively less common upsection concurrent with the increased prominence of redbeds. This long-term trend in Appalachian climate is interpreted to be a response to long-lived geological events. Crowell (1978) and Frakes et al. (1992) proposed that such slow and protracted changes were caused by movement of plates through time (Table 1). The Pennsylvanian strata of North America have long been characterized as cyclic repetitions in lithology that are termed cyclothems (Weller, 1931). These cyclic packages lasting 105 to 405 years, are interpreted as depositional manifestations of glacioeustatic fluctuations caused by alternating glacial and deglaciation events in the southern high latitudes (Wanless and Shepard, 1936; Busch and Rollins, 1984; Heckel, 1990). Such short-term climatic oscillations appear to be consistent with rapid changes in atmospheric CO2 levels or perturbations in Earth’s orbit (Crowell, 1978; Horton and Poulsen, 2009). Appalachian precipitation models have shown a direct relationship between purported global climate models (Cecil, 1990). During glacial advances, global sea level dropped and precipitation rates increased in both low and high latitudes (Cecil, 1990; Heckel, 2008). Increased levels of precipitation created widespread swamplands in the equatorial lowlands of tropical Pangaea, which are manifested as thick coal beds and flint clays (Cecil, 1990; Greb et al., 2006). During periods of deglaciation, global sea level rose and tropical climates became more seasonal to markedly drier (Cecil et al., 2004). Glacial retreats also were responsible for the formation of paleosols and ephemeral carbonate lake deposits in topographic high and low areas, respectively (Cecil et al., 1985). Within cyclothems of the Monongahela Group alternating shale and limestone layers suggest that there is a scale of climatic cyclicity even shorter in duration than the cyclothems. Some individual cyclothems have as many as ten of these clasticcarbonate couplets. During this field trip we will examine the depositional signature and changes in character that the various scales of climate change record in Appalachian Pennsylvanian rocks.
DIRECTIONS TO STOP 1 Stop 1 is located along both sides of U.S. Route 422 in South New Castle, Lawrence County, Pennsylvania. The section proceeds eastward from the Moravia Street Interchange (Pennsylvania Route 168) to the top of the hill (40° 58′ 03″ N, 80° 21′ 54″ W). Stop 1. Upper Pottsville and Lower Allegheny Formations—The Early Middle Pennsylvanian Continuously Wet Interval The upper Pottsville and lower Allegheny formation rocks exposed at Stop 1 range in age from late Bashkirian (early Atokan) to middle Moscovian (early Desmoinesian) (Douglas, 1987; Brezinski et al., 1989). At the base of the exposure are upper Pottsville strata assignable to the Lowellville marine member though Upper Mercer Member (Fig. 3). The upper part of the roadcut exposes rocks of the lower Allegheny Formation that are capped by the early Desmoinesian Vanport Limestone (Douglas, 1987; Brezinski et al., 1989). Stop 1 exposes examples of Appalachian basin lower Pennsylvanian cyclothems. Each cycle consists of an underclay paleosol that is presumably unconformably overlain or replaced by thin coals or coaly shales (paleo-Histosols). The coal beds are commonly overlain by gray, argillaceous marine limestone or sideritic shale. The marine limestone or shale is overlain by greenish-gray silty shale, siltstone, or silty sandstone. These cyclothems contrast with the typical upper Pennsylvanian Midcontinent cyclothems in that they have very poorly developed marine limestones, but well-developed coal bed with underclay paleosols (Heckel, 1995). The cyclothems exhibited at this stop are interpreted to be equivalent in duration and mode of formation to those of the Missourian of the United States Midcontinent which Heckel (1995) proposed had a duration of between 100 k.y. and 300 k.y. Therefore, one can infer that the periodicity for these cycles is the same as that suggested for the slightly younger strata of the midcontinent. Throughout the northern part of the Appalachian Basin, numerous intervals of high-alumina fire clays, termed flintclays (Keller, 1968), are exposed. One of the best developed of these flint clays is the Mercer fire clay. In Clearfield and Cambria counties, Pennsylvania, this fire clay is a kaolinite-rich diaspore that has been interpreted as having formed by intense leaching on elevated areas marginal to the Mercer seaway (Williams et al., 1968; Williams and Bragonier, 1985; Bragonier, 1989; Cecil et al., 2004).
TABLE 1. CLASSIFICATION OF CLIMATIC EVENTS AND POSSIBLE CAUSATIVE FACTORS IN THE PENNSYLVANIAN OF THE APPALACHIANS Name Duration Possible cause 6 8 Long-term 10 –10 Plate movement/Ice buildup 5 5 Cyclothemic 10 –40 Milankovitch cycles 4 10 Earth’s precession Limestone/Shale alternations
Appalachian Pennsylvanian climatic events
49
At Stop 1, the best developed paleosol is not beneath either of the Mercer marine layers, but beneath the Brookville Coal bed at the base of the Allegheny Formation. This paleosol exhibits a highly leached upper layer coupled with an underlying layer enriched in iron similar to those found in Spodosols (Skema, 2005). At the top of the exposure is the Vanport Limestone. This marine limestone is the thickest and purest carbonate unit present in the Pennsylvanian strata of the central Appalachian Basin. It reaches a thickness in excess of 12 m (Bragonier, 2005), and it has been extensively mined as a flux for steel production.
Figure 3. Stratigraphic section of upper Pottsville Formation and lower Allegheny Formation rocks exposed along U.S. 422 at Moravia Street Interchange, New Castle, Pennsylvania (modified from Skema, 2005). This interval is interpreted to have been deposited during the early Pennsylvanian humid period. Climatic cycles (left column) are interpreted to be the Appalachian equivalent to Midcontinent cyclothems produced during repeated advances and retreats of high-latitude continental glaciers.
Climate Inferences The cyclothems displayed at this location are low-latitude manifestations of high-frequency climatic fluctuations that produced rapid glacial advances and retreats within Earth’s polar regions during the Late Paleozoic Ice Age (hereafter, the LPIA). These climate cycles are superimposed upon the long-term early Pennsylvanian climate trend toward cool conditions and high levels of rainfall (Cecil, 1990). The cyclothems preserved in this section are interpreted to result from alternating perhumid to humid paleoclimates. The underclays and coal beds were formed during perhumid to humid periods inferred to be contemporaneous with the advancement of polar glaciers (Fig. 3) (Cecil, 1990; Skema, 2005). These facies are replaced upsection by marine units that reflect regional flooding during sea level rise concurrent with deglaciation (Crowell, 1978; Skema, 2005). The greenish-gray shale and thin sandstone units that tend to separate individual cyclothemic packages are interpreted to represent somewhat drier interglacial periods (Fig. 3). The Mercer flint clay of Clearfield and Cambria counties, which was formed contemporaneously with the lower part of the section exposed at Stop 1, is an exceptionally well-developed, intensely leached, high-alumina paleo-Ultisol (Cecil et al., 2004). At this location, the Mercer fire clay would occupy the stratigraphic interval represented by the entire lower and upper Mercer cyclothems. The Mercer flint clay was formed on topographically elevated areas to the east during a prolonged period of intense rainfall and leaching. At the top of this exposure, the thick Vanport Limestone represents a period of extreme global warming that caused the flooding of the Appalachian Basin with deep and clear marine waters. Pennsylvanian cyclothems have been tied to sea level cycles with durations between 100 k.y. and 400 k.y. (Busch and Rollins, 1984; Heckel, 1990). The periodicity of such sea level cycles is consistent with Earth’s orbital incongruities termed Milankovitch cycles (Busch and Rollins, 1984; Heckel, 2008). Thus, the cyclothems exposed at Stop 1 are interpreted as the Appalachian depositional response to alternating intensities of rainfall caused by astronomically induced climate conditions during the early Pennsylvanian wet interval. Biotic Response The early to middle Pennsylvanian is characterized by long-term stasis and a lack of change within terrestrial plant
50
Brezinski and Kollar
communities (DiMichele et al., 2004). This lack of faunal turnover has long been recognized by biostratigraphers. Stanley and Powell (2003) hypothesized that the depressed rates of origination and extinction shown by marine animals during the LPIA was the normal macroevolutionary state during the rapidly changing climatic and sea level conditions of these glacial episodes. Stanley and Powell (2003) postulated that environmentally sensitive, mainly tropical, animals went extinct at the onset of the LPIA (Mid-Carboniferous extinction). This left behind only those forms that could exist in the highly variable climatic conditions and rapidly fluctuating sea level of the Pennsylvanian. A subdued evolutionary tempo for marine organisms during the LPIA has been documented for brachiopods (Powell, 2005; 2007), goniatites (Kullman, 1985), and trilobites (Brezinski, 1999). Thus, while Pennsylvanian sea level was rapidly changing concurrent with high-frequency climatic shifts, marine faunas underwent stasis. Spores recovered from coal beds reflect a flora dominated by lycopsid trees and tree ferns with lower abundances of small ferns, Calamites, and Cordaites (C. Eble, written commun., in Skema, 2005). These forms suggest a constantly wet substrate suggesting a humid to perhumid climate. Furthermore, the age of these floras suggests a large hiatus between the Lowellville interval at Stop 1, which may be as old as early Atokan, and the Lower Mercer, which may be as young as late Atokan (C. Eble, written commun., in Skema, 2005). The period of time represented by the deposition of this exposure is only slightly younger than that of the youngest Joggins strata of Nova Scotia (Calder et al., 2006), which has yielded a wealth of tetrapod remains as well as the earliest known reptile. Most of the tetrapods recovered from the Joggins strata are highly aquatic species that lived in mire-side lakes (Carroll, 1967). In the Appalachian Basin, the Five Points and Cannelton tetrapod sites from the Lower Kittanning coal interval represent similar habitats. These two locations have produced tetrapod faunas that are characterized by highly aquatic forms (Hook and Baird, 1986; Hook and Ferm, 1988; Berman et al., 2010).
ment that occurred prior to the formation of the overlying Bolivar fire clay. The brecciated intervals represent microkarsting related to subaerial exposure. The Bolivar fire clay has been mined locally for the manufacture of refractory bricks. This claystone is an intensely weathered and leached paleosol. Although the Bolivar fire clay is highly leached, it contains much higher levels of iron and silica than do the flint clays of the lower Allegheny and upper Pottsville formations (Williams et al., 1968). The Upper Freeport Coal can be subdivided into three, regionally recognizable benches (Ruppert et al., 2001). These benches can be traced throughout the basin, suggesting that the intervening partings represent more than just localized flooding events (Cecil et al., 2004). Kosanke and Cecil (1996) have shown that such partings exhibit enriched levels of spores from seed ferns, conifers, and fusain. In contrast, high levels of Lycospora are found within the coal benches. On a regional scale, the Upper Freeport coal bed exhibits localized areas of thickening that are pod-shaped (Fig. 5). These pod-like bodies led Pedlow (1977) to interpret them as depositional features known as peat islands. These local changes in coal bed thickness are suggestive of deposition in ombrogenous peat swamps. Such domal peat swamps form in areas of high precipitation where plant growth and peat accumulation can proceed above the water table. Although the Upper Freeport has been suggested as being deposited in a planar peat swamp, the pod-like character in local and regional isopach maps (Kertis, 1985; Ruppert et al., 2001) (Fig. 5) is inconsistent with the character of a planar or topogenous peat swamp.
DIRECTIONS TO STOP 2 Stop 2 is located along the William Flynn Highway (Pennsylvania Route 8) at Butler Plank Road, Shaler Township, Allegheny County, Pennsylvania (40° 32′ 21″ N, 79° 57′ 50″ W). Stop 2. Upper Freeport Coal Bed—Transition from the Early Pennsylvanian Continuously Wet Interval At Stop 2, we will examine the Late Moscovian (Westphalian D) Upper Freeport Coal. The Upper Freeport Coal is underlain by the Upper Freeport Limestone, and the Bolivar fire clay (Fig. 4). The Upper Freeport Limestone is a regionally discontinuous stratum of nonmarine limestone. At this location, the limestone bed is highly brecciated and appears to be pedogenically disrupted. The Upper Freeport Limestone represents an episode of lake develop-
Figure 4. Measured stratigraphic section of the Upper Freeport Limestone to the Upper Freeport Coal at Stop 2 with interpreted climate curve. The Upper Freeport coal has been interpreted as being deposited near the end of the wet interval that began in the Bashkirian (Cecil, 1990).
Appalachian Pennsylvanian climatic events Climate Inferences The Upper Freeport Coal was deposited during a phase of transition from the early Pennsylvanian (Bashkirian to Moscovian) wet interval to the late Pennsylvanian (Kasimovian-Gzhelian) dry period. In the Appalachian Basin, this period of transition is represented by the deposition of the upper Allegheny and lower Glenshaw formations. The climate ranged from perhumid to humid with high annual rainfall during glacial maxima to dry subhumid conditions during the intervening interglacial stages. This pattern appears to change toward the end of the deposition of the Allegheny Formation with the accumulation of the lower and upper Freeport Limestone beds. These strata are some of the earliest nonmarine limestone strata deposited in the Appalachian Basin. Their formation suggests that, during interglacial stages, evaporation sometimes exceeded precipitation, and the conditions may have become semiarid during deposition of these units (Cecil et al., 1985). The increased seasonal evaporation rates responsible for the deposition of the Upper Freeport Limestone were replaced by a return to continued year-round humid conditions and the resulting intense leaching during the formation of the Bolivar fire clay. This plastic, high-alumina fire clay suggests that the climate became increasingly wet during its formation, and the constant precipitation and intense leaching created this mineral paleosol. The Bolivar typically contains higher levels of iron and silica than the lower Allegheny fire clays, indicating that the severity of precipitation was somewhat reduced during the formation of earlier Allegheny Formation fire clays. The possibility of domed peat swamp formation during deposition of the Upper Freeport coal bed indicates that the
51
humid conditions and high rainfall that characterized the early Pennsylvanian may have intermittently returned during some glacial epochs. Such a return to very wet conditions may be responsible for the enigmatic thickness patterns displayed by the Upper Freeport Coal bed. However, this interpretation appears incongruous with the basin-wide partings that are enriched in conifer spores and fusain that suggests the Upper Freeport swamp infrequently experienced drier climatic conditions that may have temporarily terminated peat swamp growth. Alternately, the regional partings mean it was domed. Biotic Response Lowland, coal-forming plant communities show a remarkable persistence or stasis from the onset of the LPIA until the end of the Moscovian (DiMichele et al., 1996; Pfefferkorn et al., 2000). The end of the Moscovian marks a dramatic change in the character of these communities (DiMichele and Phillips, 1996; Kosanke and Cecil, 1996). The ecological stability in plant communities that persisted in peat-forming wetlands during the Bashkirian ended shortly after the deposition of the Upper Freeport Coal. Containing a diverse microflora (Kosanke and Cecil, 1996), the Upper Freeport represents one of the last coal beds that contain abundant Lycospora, the spore genera produced by lycopsid trees such as Lepidodendron (Cecil et al., 2004). Many of the well-known lycopsid trees became extinct within the lower Conemaugh Group (DiMichele and Phillips, 1996; Kosanke and Cecil, 1996). In the Appalachian Basin, three important vertebrate localities are known from the Allegheny Formation. These locations have yielded a diverse assemblage of vertebrates that are nearly exclusively highly aquatic. These three locations represent deposits of perennially submerged, sapropelic infillings of abandoned river channels (oxbow lakes) that produced black, organic-rich siltstones and shales (Hook and Ferm, 1985, 1988; Hook and Baird, 1986, 1993). These localities include: the Five Points locality of the Lower Kittanning Coal of Ohio; the Upper Kittanning interval at Cannelton, Pennsylvania; and the well-known Diamond Mine locality of the Upper Freeport Coal at Linton, Ohio. The latter site has yielded more than 6,000 specimens belonging to ~42 vertebrate taxa. Each of these localities suggests a climate that was perennially wet. Furthermore, the diverse tetrapod faunas obtained from each location indicate that these communities were distinctly highly aquatic in character. DIRECTIONS TO STOP 3
Figure 5. Isopach map of the Upper Freeport Coal in the central Appalachian Basin (modified from Ruppert et al., 2001). Note how coal bed thickness exhibits localized thickening. This is interpreted to have been produced by localized ombrogenous (domal) peat masses.
Stop 3 is located along Montour Run Road and adjacent FedEx Drive, north of Robinson Town Center, Allegheny County, Pennsylvania. The section begins within the Ames Limestone along Montour Run Road and continues through the overlying Grafton Sandstone at the intersection of Montour Run Road and FedEx Drive, then proceeds up the hill on FedEx Drive (40° 27′ 34″ N, 80° 10′ 20″ W).
52
Brezinski and Kollar
Stop 3. The Middle Conemaugh Group—The Late Kasimovian and Early Gzhelian Dry Period At Stop 3, we will examine Appalachian depositional responses to the late Pennsylvanian warming period. We will also discuss the turnover in the character of the marine and terrestrial vertebrate faunas. Stop 3 is within strata of the upper Glenshaw and lower Casselman formations of the Conemaugh Group. The Conemaugh Group is equivalent to the Lower Barren Measures of Rogers (1858). In western Pennsylvania, the Conemaugh Group is ~200 m thick and is subdivided into two formations. The lower formation, the Glenshaw, contains redbeds and numerous marine layers. The overlying Casselman Formation is characterized by redbeds and thin, freshwater limestones. A notable difference between the Conemaugh Group and the underlying Allegheny Formation is the paucity of mineable coals. The thin coals that are present are rich in fusain. In this part of the Appalachians, Pennsylvanian cylothems take on a different character than those of the Pottsville and Allegheny formations. Here, the Conemaugh Group has reddish paleosols, not gleyed underclays, beneath the patchy coal beds. The coal bed horizons are commonly replaced upsection by thin nonmarine limestones; these are in turn overlain by variably thick, channel-phase sandstone intervals and then reddish paleosols. At the base of the exposure at Stop 3 are the Pittsburgh Redbeds. These red strata represent a calcic-Vertisol to Aridisol (Joeckel, 1995a; Cecil et al., 1998, 2004). This red paleosol is overlain locally by a thin coal followed by the Ames marine horizon (Fig. 6). The Ames Limestone represents the final inundation of the Appalachian Basin by marine waters. The Ames is not only the final submergence of the basin, but also the most laterally extensive. These marine strata have been traced from southwestern Ohio to near Wilkes Barre, Pennsylvania (Brezinski, 1983). The areal extent of this marine unit indicates that the deepening episode that it represents was significant. In fact, the extensive distribution led Berman et al. (2010) to suggest that the Ames submergence of the basin reflected a significant retreat of polar glaciers and the concomitant loss of continental ice into marine settings. Above the Ames is ~10 m of interbedded sandstones and siltstones that represent overbank or river levee deposits. These strata are overlain by nearly 14 m of cross-bedded, mediumgrained, fining-upward channel-phase sandstone known as the Grafton Sandstone. The Grafton Sandstone is overlain by a 15-m interval of interbedded siltstone, shale, and mudstone that is capped by a gray limestone. This interval is correlative with the Birmingham Shale. Here the shale interval lacks the brackishwater fossils found at many other local sites; instead, it contains a highly rooted paleosol that displays abundant, in situ tree casts. More than 20 tree and root casts can be identified along this horizon. The tree zone is in turn capped by ~1 m of nonmarine lacustrine limestone. Many of the tree casts penetrate this limestone
interval and are truncated by the overlying Morgantown Sandstone. The carbonate within this interval forms the casts within the underlying tree zone. Near the western edge of the limestone outcrop, a thickening of the overlying dark gray, silty shale appears to represent an infilling of a topographic low (Berman et al., 2010). This paleotopographic change is interpreted as an abandoned stream channel or oxbow lake. It is here that the skull of the highly terrestrial tetrapod amphibian Fedexia striegeli was believed to have been recovered (Berman et al., 2010). Overlying the Birmingham Shale is a 45-m-thick, tanweathering, cross-bedded, coarse-grained, channel-phase sandstone with a pebbly, basal conglomerate and epsilon crossstratification. This unit is the Morgantown Sandstone, but is abnormally thick at this locality. Elsewhere in the central Appalachian Basin, this interval is represented by the Barton and Wellersburg coals and/or limestones and the upper Grafton and Barton sandstones (Swartz and Baker, 1920; Edmunds et al., 1999) (Fig. 6). At this site, the Morgantown Sandstone channel
Figure 6. Stratigraphic section of the Casselman Formation of the Conemaugh Group at Stop 3. This stop represents deposition during latter stages of the late Pennsylvanian warm and dry period.
Appalachian Pennsylvanian climatic events apparently eroded any evidence of these two intervals and may represent a stacking of several different episodes of river channel formation (Donaldson et al., 1985). The Morgantown Sandstone grades upward into a 12-m interval of interbedded grayish-brown siltstone and shale that represents deposition in a low-energy, overbank setting away from the main river channel. These clastics are overlain by the highest exposed strata of the roadcut. This 16-m interval of interbedded red and greenish-gray calcareous shale and nodular limestone is assignable to the Clarksburg Member of the Casselman Formation. Climatic Inferences The paucity of mineable coal beds, the high fusain content of the coals present, the occurrence of well-developed marine units, and the preeminence of calcic-Vertisols within the middle Conemaugh suggest that the climate during the time when these rocks were deposited was much drier than that in the underlying Allegheny Formation (Cecil et al., 1985; Cecil, 1990; Cecil et al., 1998, 2004; Berman et al., 2010). The red calcareous paleosol that underlies the Ames Limestone (Pittsburgh Redbeds) has been shown to be a good indication that the regional climate was semiarid to arid (Joeckel, 1995a). The Pittsburgh Redbeds are the most prominent of the numerous redbed horizons known from the middle and upper Conemaugh Group in this region. Other notable red paleosol intervals include the Meyersville Redbeds beneath the Pine Creek marine horizon, the Birmingham Redbeds below the Duquesne coal, and the Clarksburg Redbeds below the Clarksburg Limestone. While the intensity of pedogenesis and the prominence of calcic content of each of these intervals vary, all are suggestive of relatively dry, seasonal conditions (Retallack, 2001). The change in salient climatic indicators within the middle and upper Conemaugh Group reflects changes not only within the Appalachian Basin, but globally. Frakes et al. (1992) have shown that the frequency of high-latitude glacial events diminishes within the upper Pennsylvanian, and that a prolonged deglaciation occurred. This global warming trend is manifested in the Appalachian Basin by cyclothemic deposits that are very different from the earlier Pennsylvanian strata. The patchy, highfusain coal beds indicate that even during glacial maxima, the Appalachian climate was dry subhumid. During the drier interglacial periods the calcic-Vertisols indicate that the region’s climate was semiarid to arid. Biotic Response The effects of the long-term climate change of the Pennsylvanian Period appear more far-reaching than just influencing the presence or absence of coals and the character of the paleosols. During the early Missourian deposition of the Glenshaw Formation, there was a major extinction event of a significant number of coal-forming plants (Kosanke and Cecil, 1996). Arborescent lycopsid trees, several tree ferns, and sphenopsid tree-spore genera that dominated the coal-swamp flora throughout the early and
53
middle Pennsylvanian became extinct (Kosanke and Cecil, 1996; DiMichele and Phillips, 1996; Cecil et al., 2004; Pfefferkorn et al., 2008) between the deposition of the Mahoning Coal and the Brush Creek marine horizon. The main coal-forming, hygromorphic plants that inhabited perennially water-saturated substrates of the early and middle Pennsylvanian throughout Euramerica died out. This extinction is coincident with the WestphalianStephanian contact. These water-loving plants were supplanted in the late Pennsylvanian by forms adapted to drier conditions (DiMichele and Hook, 1992). This replacement was also accompanied by the changing sedimentological regimes resulting from increased aridity in the Kasimovian (late Missourian) (Kosanke and Cecil, 1996; DiMichele and Phillips, 1996). Thus, the drying of paludal environments during the early Missourian and Virgilian is thought to be directly related to the widespread diminishment of the coal-swamp plant clades (Kosanke and Cecil, 1996). Although the Late Paleozoic dry period had profound effects on terrestrial plant and animal communities, its impact on marine biota was less significant. Pennsylvanian marine faunas are renowned for their lack of biostratigraphic utility because of the absence of faunal turnover. Stanley and Powell (2003) have proposed that the rapidly fluctuating sea level of the Pennsylvanian constrained marine biota adapted to this constantly changing environment. Because cosmopolitan, ecological generalists were naturally selected for, little to no evolution occurred (Stanley and Powell, 2003; Powell, 2007). Brezinski (1999) has shown that the long-ranging, highly endemic trilobite faunas of the early Pennsylvanian are replaced within the Ames Limestone with highly pandemic forms. Brezinski et al. (1989) proposed that the high stand of global sea level, equivalent to the deposition of the Ames Limestone, eliminated marine interbasinal barriers to migration and spawned a period of pandemism. Berman et al. (2010) identified the new genus and species of tetrapod amphibian, Fedexia striegeli, from a well-preserved skull recovered from Stop 3. Fedexia is one of three early representatives of the tetrapod group known as the Dissorophoidea. All three of these early representatives are known from strata equivalent to the middle Conemaugh Group (Berman et al., 2010). The Dissorophoidea became a diverse and widespread vertebrate group in the Early Permian. This vertebrate group is commonly associated with rocks that exhibit evidence of being deposited in arid climate settings such as redbeds (Berman et al., 2010). The highly terrestrial dissorophid tetrapod amphibians from the older strata of the Conemaugh Group were recovered from a stratigraphic interval that is identified with a major deglaciation of Earth’s polar regions. The warming and drying of the climate during the late Kasimovian (Missourian and early Virgilian) appears to be a primary factor in a major ecological shift in tetrapod communities from predominately small, aquatic amphibians of the early and middle Pennsylvanian to medium-tolarge, terrestrially adapted vertebrates of the late Pennsylvanian to Early Permian. Throughout the early and middle Pennsylvanian, tetrapods are found in black shales and cannel coals associated with coal-forming and sapropelic environments (Hook and
54
Brezinski and Kollar
Ferm, 1988; Hook and Baird, 1986). The wet environmental conditions that existed throughout the deposition of the Allegheny Group have been interpreted as resulting from perennially high rainfall levels that were contemporaneous with periods of maximum development of southern hemisphere polar glaciers (Cecil, 1990; Cecil et al., 2004). Sediments that preserve the Appalachian Basin tetrapod assemblages of the late Pennsylvanian Conemaugh and Monongahela groups typically differ markedly from those of the earlier-formed Allegheny Formation. In this later stratigraphic interval vertebrates are largely confined to thin, freshwater limestones or their closely associated paleosols that are interpreted as most likely representing highly localized, seasonally dry ponds and lake deposits that formed during interglacial dry periods (Cecil, 1990). DIRECTIONS TO STOP 4 This stop is located at the Carnegie Interchange of Interstate 79. The section begins along Spring Street (Pennsylvania State Route 3038) and continues on the east and west sides of Interstate 79 immediately south of the interchange (40° 23′ 52″ N, 80° 06′ 20″ W). Stop 4. Monongahela Group—Gzhelian Humid Interval Stop 4 is within the uppermost Conemaugh and lower Monongahela groups. The Pittsburgh Coal is exposed near the base of the stratigraphic section that continues south along Interstate 79. Rogers (1858) named the stratigraphic interval that we now call the Monongahela Group the Upper Coal Measures, because of the numerous mineable coals present therein. At this location much of the Monongahela Group (below the top of the Benwood Limestone) is exposed along Interstate 79 (Fig. 7). The base of the Pittsburgh Coal marks the base of the Monongahela Group. This coal bed is perhaps the thickest and most extensive in the Appalachian Basin, covering more than 17,000 km2 (Eble et al., 2006). The Pittsburgh Coal exhibits a general regional thickening to the east, and does not show the localized pod-like thickness changes of the Upper Freeport Coal (Ruppert et al., 2001). The Pittsburgh has a thick, well-developed mineral paleosol and is divisible into two main benches, separated by a siliciclastic parting known as the “bearing-in bench.” The bearing-in bench is significant because it is present throughout the basin at or near the same stratigraphic level within the coal bed. This thick parting suggests a basin-wide change in deposition in the Pittsburgh peat swamp. At Stop 4, we will examine the Gzhelian strata that suggest that the Appalachian Basin returned to more humid conditions following deposition of the middle Conemaugh Group. The younger Monongahela Group is characterized by a number of mineable coals including the Pittsburgh, Redstone, Sewickley, and Waynesburg. Additionally, this stratigraphic interval also displays a return to clear cyclothemic deposition even though the
Figure 7. Stratigraphy of the Monongahela Group in western Pennsylvania. This interval is interpreted to have been deposited during a long-term humid period by short-term glacially driven climate cycles. Cyclothems are preserved as alternations of coal and limestone with coal beds interpreted as forming during humid periods contemporaneous with high latitude glacial advance and limestones created during drier intervals of deglaciation.
Appalachian Pennsylvanian climatic events cycles have a slightly different motif than that of the cyclothems in the Allegheny and Pottsville formations. Perhaps the most prominent nonmarine limestone in the Pennsylvanian of the Appalachian Basin is the Benwood Limestone of the Monongahela Group. This widespread unit is as much as 28 m thick and extends over much of southwestern Pennsylvania, eastern Ohio, and northern West Virginia (Petzold, 1990). In addition to the cyclothems that characterize the Monongahela Group, many limestone intervals in the Monongahela exhibit a nested succession of alternating carbonate and siliciclastic layers. These nested cycles are well-defined in the Benwood Limestone. The siliciclastic intervals in the Benwood consist of gray or greenish-gray to grayish-green, silty, locally sandy, shale to shaly siltstone that range from 0.3 to 1.5 m in thickness. Locally, these shales are dark gray. The shale layers alternate with light gray to tan, laminated to massive, argillaceous, limestone to dolomitic limestone (Fig. 8). The limestone layers commonly display brecciated intervals and pervasive mudcracks, especially near their top. Lake varves also have been noted (Marrs, 1981; Petzold, 1990). Coals and underclays within the Monongahela Group formed during high latitude glacial advances. However, the interglacial episodes produced deposition of extensive lacustrine deposits. This is reflected in the Monongahela cyclothems by alternations of coals and carbonates (Fig. 7). Just as with the cyclothems of the Pottsville and Allegheny formations, those of the Monongahela Group are interpreted to have been produced by Milankovitch-driven, glacial cycles (Heckel, 2008). Climate Inferences When compared to the middle Conemaugh Group, the sedimentological character of the Monongahela Group indicates a change to more humid conditions. The presence of mineable coal beds coupled with the lack of redbeds reflect an increased amount of rainfall (Cecil et al., 1985). Because of its geographi-
Figure 8. Idealized sedimentological manifestation of climate cycles with lake deposits in the Monongahela Group.
55
cally extensive character, the thick Pittsburgh Coal appears to be the product of copious levels of rainfall rather than local water table fluctuations. Furthermore, the mineral underclay beneath the Pittsburgh Coal infers a permanently waterlogged substrate that signifies a humid to wet subhumid paleoclimate (Cecil et al., 1998, 2004). While the presence of thick coal beds within the Monongahela cyclothems indicates a return to humid conditions during glacial periods, the intervening mudcracked lacustrine deposits suggest that during times of deglaciations, evaporation was greater than precipitation, and that the region’s climate may have been dry subhumid. The distinct coal-limestone cyclothems in this part of the stratigraphic succession also are suggestive of a return to icehouse conditions (Fig. 7). Within contemporaneous strata of the Midcontinent, Heckel (2008) identified a trend toward increased fluvial incision and cyclicity. This intimated to Heckel (2008) a greater level of continental glacier waxing and waning. Similar to the Pottsville and Allegheny formations, thick coals with well-developed underclays reflect humid conditions coincident with the advance of high-latitude glaciers. Drier, interglacial episodes within the Monongahela Group are represented by carbonate strata that intervene between the coal beds. In the Allegheny Formation interglacial episodes are suggested by sea level high stands that deposited marine to marginal marine limestones and sideritic shales (Skema, 2005). In contrast, the Monongahela interglacial intervals are indicated by carbonate intervals that are the result of intense evaporation and climate drying (Cecil et al., 2004). The couplets of shale and limestone within individual lacustrine units are interpreted to represent the sedimentologic manifestation of short-duration, climatically driven cycles (Fig. 9). The amount of time represented by each alternation has not been determined, but the vertical repetition of shale and limestone is identical to lacustrine cycles identified within the Late Triassic of the eastern United States (Van Houten, 1962; Olsen, 1986), and the Eocene of the western United States (Picard and High, 1981) (Fig. 8). Meter-scale shale and carbonate cyclicity is common to ancient lake deposits and has been attributed to changes in short-term depositional factors such as climate (Collinson, 1978; Picard and High, 1981; Dunagan and Turner, 2004). Each lacustrine cycle typically consists of a lower siliciclastic (shaly) unit produced by high turbidity during wet periods (Fig. 8). The upper chemically precipitated (limestone/dolomite) units are produced during dry periods when sediment supply and turbidity are low (Collinson,1978). Olsen (1986) hypothesized that the nested, meter-scale lake cycles of the Triassic rocks of the eastern United States were the result of alternating wet and dry periods produced by precessional factors of Earth’s orbit. A similar interpretation can be proposed for the Monongahela lacustrine deposits (Fig. 9). Biotic Response As mentioned at Stop 3, the shift from wet to dry climates that occurred during deposition of the lower Conemaugh precipitated
56
Brezinski and Kollar
a complete rearrangement of the structure of coal-forming plant communities (DiMichele and Phillips, 1996; Kosanke and Cecil, 1996). Many of the large lycopsid trees that characterized the early Pennsylvanian peat swamp are absent from floras in late Pennsylvanian coal swamps (Kosanke and Cecil, 1996). Instead, the main arborescent plant found in late Pennsylvanian swamps are tree ferns (DiMichele and Hook, 1992; DiMichele and Phillips, 1996; Eble et al., 2006). In a study of lower Monongahela coal beds, Eble et al. (2006) showed that the flora of the Pittsburgh, Redstone, and Sewickley coal beds was dominated by tree-ferns such as Psaronius and true fern species. This is quite different from the underlying Allegheny coal beds that have a more diverse Lycospora component, produced by large arborescent lycopsids such as Lepidodendron (Kosanke and Cecil, 1996; Eble, 2003). Eble et al. (2006) further suggested that the expansive geographic character of the Pittsburgh Coal indicates that it was formed in a topogenous or planar peat swamp in contrast to the more areally restricted, pod-like domal swamps of the early to early middle Pennsylvanian strata of the Morrowan and Atokan.
Tetrapod assemblages of the late Pennsylvanian Monongahela Group are preserved in sediments that differ markedly from those containing earlier Pennsylvanian forms. In this part of the stratigraphic interval, vertebrate remains are primarily confined to nonmarine limestone intervals or their closely associated paleosols. These strata represent shallow lakes that dried up during periods of decreased precipitation. The high-frequency lake cycles present in the Fishpot, Benwood, and Uniontown limestones demonstrate how ephemeral these environments were. Possible precessional cycles displayed by the alternation of wet climate shales and dry climate limestones reflect environments of deposition that were highly variable. Such variability would have selected for vertebrates that were able to move from one wet spot to another. Associated with this long-term shift in physical environments, seen during the Pennsylvanian, is the change from small aquatic amphibians to medium-to-large, terrestrially adapted vertebrates (Lund, 1975; Berman et al., 2010). The Appalachian stratigraphic record reflects this slow change in vertebrate faunas as it occurred within the Monongahela Group (Lund, 1975). The fossil record preserves some of the earliest larger predatory tetrapods including Diadectes, Deaphosaurus, and Eryops (Lund, 1975, fig. 1). DIRECTIONS TO STOP 5 Stop 5 is located in highwall excavations for The Foundry and Trinity Point shopping centers on both sides of U.S. Route 19 immediately north of Interstate 70, in Washington, Washington County, Pennsylvania (40° 11′ 02″ N, 80° 13′ 02″ W). Stop 5. Washington Formation, Dunkard Group—Late Gzhelian or Early Asselian Warming
Figure 9. Interpreted levels of short-term climate cycles preserved within the middle Monongahela Group at Stop 4. The high frequency shale/limestone couplets are interpreted to be similar to Van Houten’s (1962) lake cycles. These are presumably formed by orbital forces.
At Stop 5, we will look at strata in the type area of the Washington Formation of the Dunkard Group. Rogers (1858) referred to these strata as the “Upper Barren Measures.” The stratigraphic succession at this stop is distinctly similar to the rocks in the middle Conemaugh Group that we examined at Stop 3. The Washington Formation consists largely of interbedded nonmarine limestone, shale, sandstones, and thin discontinuous coals. A measured section (Fig. 10) spans ~40 m of the lower Washington Formation. The section is a composite of measurements taken from the two individual shopping centers along U.S. Route 19. The base of the exposed section is a massive tan-weathering limestone that is assigned to the Bristol Limestone. Approximately 6 m of platy, reddish-gray, micaceous, sandy siltstone separate this limestone from the lower bench of the Washington Coal. The Washington Coal is composed of two benches that are separated by ~2 m of platy siltstone. At this location, only a very poorly developed underclay occurs beneath the lower bench of the Washington Coal; there is no underclay evident under the upper coal bench. Above the Washington coal bed is the thinly bedded Lower Washington Limestone. This unit is best exposed behind Trinity Point Shopping Center. Above the Lower
Appalachian Pennsylvanian climatic events Washington Limestone is a variably thick, cross-bedded sandstone which is in turn overlain by a greenish-gray rooted claystone. The claystone exhibits abundant siderite nodules at the top. This rooted zone is overlain by ~3 m of interbedded limestone and shale assigned to the Middle Washington Limestone. Above the Middle Washington is ~2 m of cross-bedded sandstone which is in turn overlain by greenish-gray to reddish shale. At the top of the composite section is a carbonaceous shale correlated with the Washington A Coal. The exact age of the Dunkard Group has long been debated. It generally has been lumped into the Permian, but many consider it to be Pennsylvanian (Clendening, 1975; Lyons and Zodrow, 1995). Recovery of abundant remains of the fern Callipteris from the shales above the Washington Coal suggested to White (1891) that this part of the section was Permian. However, Clendening (1975) insisted that the strata were Pennsylvanian, based on spores. Romer (1952), Berman and Berman (1975), and Olson (1975) noted that the vertebrate faunas from the Washington and Greene formations were very similar to those found in the Early Permian of Texas, and therefore assigned those formations to the Permian.
Figure 10. Composite stratigraphic section and interpreted cyclothems of the lower Dunkard Group, Washington, Formation exposed at The Foundry and Trinity Point shopping centers, Washington, Pennsylvania.
57
Climate Inferences In contrast to the many widespread coal beds of the Monongahela Group, only the eponymous Washington coal is regionally correlatable in the Washington Formation. There is little question as to why Rogers (1858) termed this stratigraphic interval the Upper Barren Measures. Besides the Washington Coal, nearly all the other named coal beds are very thin and patchy in their distribution. Although the coals are spotty, the coal horizons are generally identifiable by well-developed paleosols at the stratigraphic level of many of the named coals. Examination of Dunkard Group coal beds indicates that they are very high in ash yield and sulfur content, similar to coal beds of the Conemaugh Group. This suggests that they were formed in planar peat swamps, with more clastic influx and possibly drier climatic conditions than were present during deposition of the underlying Monongahela Group (Milici, 2005). Coal beds of the Dunkard Group tend to lack well-developed underclays. Even though the Washington Coal is aerially widespread, it has a poorly developed underclay (White, 1891). Many of the Dunkard paleosols are characterized by iron-manganese or carbonate nodules and tend to be paleo-Vertisols (Stiles et al., 2001; Cecil et al., 2008). These paleo-Vertisols replace the patchy and discontinuous coal beds of the Washington and Greene formations. Furthermore, the paleosols become interstratified with redbeds farther upsection into the Greene Formation. The calcic-Vertisols present in the Washington Formation indicate long-term drying to a subhumid paleoclimate (Cecil et al., 2008). The presence of iron-manganese nodules within some of these Vertisols suggests a moisture deficit (Stiles et al., 2001). The long-term Appalachian paleoclimatic trend suggests a progressive drying of the basin from the deposition of the Monongahela Group through the Dunkard Group (Cecil et al., 2008). This trend toward increased seasonality then aridity is consistent with climatic trends interpreted for deposition of latest Pennsylvanian and Early Permian rocks of the western United States (DiMichele et al., 2006; Tabor et al., 2008). Based on the character of Early Permian paleosols from Texas and Oklahoma, DiMichele et al. (2006) proposed that these strata formed at a very low paleolatitude. Paleogeographic reconstructions have suggested that Texas and Oklahoma were at about the same paleolatitude as the Appalachian Basin during the latest Pennsylvanian and Early Permian. Consequently, similar climatic regimes between these two separate areas during this time period are expected. Cross (1950) showed that cyclothems of the Dunkard Group were very similar to the upper Pennsylvanian cyclothems of the Carbondale Formation in the Illinois Basin. Beerbower (1961) noted that the cyclothems of the Dunkard were distinct because they lacked any marine beds. The absence of marine influence within the Dunkard rocks indicates that the cyclothems displayed therein must have been controlled by other allocyclic factors. Beerbower (1961) reasoned that the Dunkard cyclothems were controlled by climate. Assuming that the periodicity of the Dunkard cyclothems is the same as equivalent Midcontinent cyclothems (100 k.y. to 400 k.y.), then it is likely that these
58
Brezinski and Kollar
climate cycles are of similar origin to those that produced the waxing and waning of Pennsylvanian and Permian ice sheets in Earth’s polar regions. Biotic Response Marine biota appear to show little change in composition, diversity, or paleogeography from the late Pennsylvanian to earliest Permian (Brezinski, 1999; Stanley and Powell, 2003; Powell, 2005; Groves and Yue, 2009). This suggests that, within marine settings, changes in environmental conditions were too minor to dramatically affect invertebrate faunas between the latest Pennsylvanian and earliest Permian. Floras of the earliest Permian exhibit a strong taxonomic similarity to those of the late Pennsylvanian (DiMichele and Hook, 1992). However, there is a long-term trend, within plant assemblages, from the water-loving lycopsids, that characterize the early Pennsylvanian, to tree fern, and conifer-dominated associations of the Early Permian. This change in faunal composition parallels the change in climate to increasingly drier conditions (DiMichele and Aronson, 1992; DiMichele et al., 2001). Some of these biotic changes have been interpreted as being coincident with the long-term climatic events that were produced by the creation of the supercontinent Pangaea (Crowell, 1978; DiMichele and Hook, 1992). The appearance of Callipteris near the end of deposition of the Washington Coal may be the Appalachian floristic heralding of this turnover to more characteristically Permian floras (White, 1891). Through the latest Pennsylvanian, Appalachian terrestrial vertebrates, like plants, continued their long-term trend toward adaptation to drier environments (Olson, 1975). By the Early Permian, reptiles had replaced amphibians as the dominant component of terrestrial tetrapod assemblages (DiMichele and Hook, 1992). The global trends toward increased terrestriality of tetrapods during the latest Pennsylvanian also is reflected in the composition of Appalachian tetrapod faunas (DiMichele and Hook, 1992). The mid-Conemaugh emergence of highly terrestrial tetrapod amphibians, such as Fedexia, was the earliest representative of highly terrestrial, dry climate vertebrates that would come to characterize Early Permian vertebrate faunas (Lund, 1975). Dunkard tetrapods such as Edaphosaurus, Trematops, and Dimetrodon indicate a distinct trend within the vertebrate faunas toward terrestriality and habitation of drier and more seasonal climates (Olson, 1975). These genera also are characteristic of Early Permian arid climate redbed faunas known from Texas and Oklahoma (Olson, 1975). These faunas inhabited environments suggestive of a much drier climate than those that characterized the early Pennsylvanian. ACKNOWLEDGMENTS We would like to thank Vik Skema of the Pennsylvania Geological Survey (retired) for background on Stop 1, and Nick Fedorko of Cove Geological Services, who provided insight
into Dunkard stratigraphic names and problems. Michael Fada provided permission to access the FedEx Drive locations, and Andy Boyd of THF Realty granted access to The Foundry properties for this study. Carla A. Kertis provided a careful review and helpful suggestions to early drafts of this field guide, and C. Blaine Cecil and Stephen Greb supplied helpful recommendations through their respective reviews. REFERENCES CITED Beerbower, J.S., 1961, Origin of cyclothems of the Dunkard Group, (upper Pennsylvanian–lower Permian) in Pennsylvania, West Virginia, and Ohio: Geological Society of America Bulletin, v. 72, p. 1029–1050, doi:10.1130/0016-7606(1961)72[1029:OOCOTD]2.0.CO;2. Berman, D.S., and Berman, S.L., 1975, Broiliellus kektotopos sp. Nov. (Temnospondyli: Amphibia) Washington Formation, Dunkard Group, Ohio, in Barlow, J.A., ed., Age of the Dunkard: Proceedings of the First I.C. White Symposium, West Virginia Geological and Economic Survey, p. 69–78. Berman, D.S., Henrici, A.C., Brezinski, D.K., and Kollar, A.D., 2010, A new trematopid amphibian (Temnospondyli: Dissorophoidea) from the Upper Pennsylvanian of western Pennsylvania: Earliest record of terrestrial vertebrates responding to a warmer, drier climate: Annals of the Carnegie Museum of Natural History, v. 78, p. 289–318, doi:10.2992/007.078.0401. Birgenheier, L.P., Fielding, C.R., Rygel, M.C., Frank, T.D., and Roberts, J., 2009, Evidence of dynamic climate change in sub-106-year scales from the Late Paleozoic glacial record, Tamworth Belt, New South Wales, Australia: Sedimentary Research, v. 79, p. 56–82, doi:10.2110/jsr.2009.013. Bragonier, W.A., 1989, Stratigraphy of flint clays of the Allegheny and Pottsville groups, western Pennsylvania, in Harper, J.A., ed., Geology in the Laurel Highlands of Southwestern Pennsylvania: 54th Annual Field Conference of Pennsylvania Geologists: Harrisburg, Pennsylvania, Pennsylvania Geological Survey, p. 69–89. Bragonier, W.A., 2005, Some findings relevant to the regional distribution of the Vanport Limestone, in Fleeger, G.M., ed., Type Sections and Stereotype Sections: Glacial and Bedrock Geology in Beaver, Lawrence, Mercer, and Crawford Counties: 70th Annual Field Conference of Pennsylvania Geologists: Harrisburg, Pennsylvania, Pennsylvania Geological Survey, p. 35–43. Brezinski, D.K., 1983, Developmental model for an Appalachian Pennsylvanian marine incursion: Northeastern Geology, v. 5, p. 92–95. Brezinski, D.K., 1999, The rise and fall of late Paleozoic trilobites of the United States: Journal of Paleontology, v. 73, p. 164–175. Brezinski, D.K., Sturgeon, M.T., and Hoare, R.D., 1989, Pennsylvanian Trilobites of Ohio: Ohio Geological Survey Report of Investigations 142, 18 p. Busch, R.M., and Rollins, H.B., 1984, Correlation of Carboniferous strata using a hierarchy of transgressive-regressive units: Geology, v. 12, p. 471–474, doi:10.1130/0091-7613(1984)12<471:COCSUA>2.0.CO;2. Calder, J.H., Gibling, M.R., Scott, A.C., Davies, S.J., and Hebert, B.L., 2006, A fossil lycopsid forest succession in the classic Joggins section of Nova Scotia: Paleoecology of a disturbance-prone Pennsylvanian wetland, in Greb, S.F., and DiMichele, W.A., eds., Wetlands through Time: Geological Society of America Special Paper 399, p. 169–195, doi:10.1130/2006.2399(09). Carroll, R.L., 1967, Labyrinthodonts from the Joggins Formation: Journal of Paleontology, v. 41, p. 111–142. Cecil, C.B., 1990, Paleoclimate controls on stratigraphic repetitions of chemical and siliciclastic rocks: Geology, v. 18, p. 533–536, doi:10.1130/0091 -7613(1990)018<0533:PCOSRO>2.3.CO;2. Cecil, C.B., Stanton, R.W., Neuzil, S.G., Dulong, F.T., Ruppert, L.F., and Pierce, B.S., 1985, Paleoclimate controls on late Paleozoic sedimentation and peat formation in the central Appalachian Basin (U.S.A.): International Journal of Coal Geology, v. 5, p. 195–230, doi:10.1016/01665162(85)90014-X. Cecil, C.B., Brezinski, D.K., and Dulong, F., 1998, Allocyclic controls on Paleozoic sedimentation in the central Appalachian Basin: U.S. Geological Survey Open-File Report 98-577, 75 p. Cecil, C.B., Brezinski, D.K., and Dulong, F., 2004, The Paleozoic record of changes in global climate and sea level: Central Appalachian Basin, in Southworth, C.S., and Burton, W., eds., U.S. Geological Survey Circular 1264, p. 77–135.
Appalachian Pennsylvanian climatic events Cecil, C.B., DiMichele, W.A., and Skema, V., 2008, The Permian of the central Appalachian Basin: Geological Society of America Abstracts with Programs, v. 40, no. 6, p. 535. Clendening, J.A., 1975, Palynological evidence for a Pennsylvanian age assignment for the Dunkard Group in the Appalachian Basin, Part I, in Barlow, J.A., ed., Age of the Dunkard: Proceedings of the First I.C. White Memorial Symposium, West Virginia Geological and Economic Survey, p. 195–216. Collinson, J.D., 1978, Lakes, in Reading, H.G., ed., Sedimentary Environments and Facies: Elsevier, New York, p. 61–79. Cross, A.T., 1950, Stratigraphy, sedimentation and nomenclature of the Upper Pennsylvanian and Lower Permian strata (Monongahela, Washington, and Greene Series) in the northern portion of the Dunkard Basin of Ohio, West Virginia and Pennsylvania: Field Guide to Special Field Conference, West Virginia Geological and Economic Survey and Ohio Division of the Geological Survey, 65 p. Crowell, J.C., 1978, Gondwana glaciation, cyclothems, continental positioning, and climate change: American Journal of Science, v. 278, p. 1345–1372, doi:10.2475/ajs.278.10.1345. DiMichele, W.A., and Aronson, R.B., 1992, The Pennsylvanian-Permian vegetation transition: A terrestrial analogue to the onshore-offshore hypothesis: Evolution: International Journal of Organic Evolution, v. 46, p. 807–824. DiMichele, W.A., and Hook, R.W., 1992, Paleozoic terrestrial ecosystems, in Behrensmeyer, A.K, Damuth, J.D., DiMichele, W.A., Potts, R., Sues, H.D., and Wing, S.L., eds., Terrestrial Ecosystems through Time: University of Chicago Press, Chicago, p. 205–325. DiMichele, W.A., and Phillips, T.L., 1996, Climate change, plant extinctions, and vegetation recovery during Middle-Late Pennsylvanian transition: the case of tropical peat-forming environments in North America, in Hart, M.B., ed., Biotic Recovery from Mass Extinction Events: Geological Society of London Special Publication 102, p. 201–221. DiMichele, W.A., Pfefferkorn, H.W., and Phillips, T.L., 1996, Persistence of Late Carboniferous tropical vegetation during glacially driven climatic and sea-level fluctuations: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 125, p. 105–128. DiMichele, W.A., Pfefferkorn, H.W., and Gastaldo, R.A., 2001, Response of Late Carboniferous and Early Permian plant communities to climate change: Annual Review of Earth and Planetary Sciences, v. 29, p. 461– 487, doi:10.1146/annurev.earth.29.1.461. DiMichele, W.A., Behrensmeyer, A.K., Olszewski, T.D., Labandeira, C.C., Pandolfi, J.M., Wing, S.L., and Bobe, R., 2004, Long-term stasis in ecological assemblages: Evidence from the fossil record: Annual Review of Ecology Evolution and Systematics, v. 35, p. 285–322, doi:10.1146/ annurev.ecolsys.35.120202.110110. DiMichele, W.A., Tabor, N.J., Chaney, D.S., and Nelson, W.J., 2006, From wetlands to wet spots: Environmental tracking and the fate of Carboniferous elements in Early Permian tropical floras, in Greb, S.F., and DiMichele, W.A., eds., Wetlands through Time: Geological Society of America Special Paper 399, p. 223–248, doi:10.1130/2006.2399(11). Donaldson, A.C., Renton, J.J., and Presley, M.W., 1985, Pennsylvanian deposystems and paleoclimates of the Appalachians: International Journal of Coal Geology, v. 5, p. 167–193, doi:10.1016/0166-5162(85)90013-8. Douglas, R.C., 1987, Fusulinid biostratigraphy and correlations between the Appalachian and Eastern Interior Basins: U.S. Geological Survey Professional Paper 1451, 95 p. Dunagan, S.P., and Turner, C.E., 2004, Regional paleohydrologic and paleoclimatic setting of wetland/lacustrine depositional systems in the Morrison Formation (Upper Jurassic), Western Interior, U.S: Sedimentary Geology, v. 167, p. 269–296, doi:10.1016/j.sedgeo.2004.01.007. Eble, C.F., Grady, W.C., and Pierce, B.S., 2006, Compositional characteristics and inferred origin of three Late Pennsylvanian coal beds from the northern Appalachian Basin, in Greb, S.F., and DiMichele, W.A., eds., Wetlands through Time: Geological Society of America Special Paper 399, p. 197–222, doi:10.1130/2006.2399(10). Eble, C.F., 2003, Palynological perspectives of Late Middle Pennsylvanian coal beds, in Cecil, C.B., and Edgar, T.N., eds., Climate controls on stratigraphy: Society of Sedimentary Geologists Special Publication 77, p. 123–135. Edmunds, W.E., Skema, V.W., and Flint, N.K., 1999, Pennsylvanian, in Shultz, C.H., ed., The Geology of Pennsylvania: Pennsylvania Geological Survey Special Publication 1, p. 149–169. Frakes, L.A., Francis, J.E., and Syktus, J.I., 1992, Climate modes of the Phanerozoic: Glasgow, Cambridge University Press, 274 p.
59
Gradstein, F., Ogg, J., and Smith, A., 2004, A geologic time scale 2004: Cambridge, Cambridge University Press, 589 p. Greb, S.F., DiMichele, W.A., and Gastaldo, R.A., 2006, Evolution and importance of wetlands in earth history, in Greb, S.F., and DiMichele, W.A., eds., Wetlands through Time: Geological Society of America Special Paper 399, p. 1–40, doi:10.1130/2006.2399(01). Groves, J.R., and Yue, W., 2009, Foraminiferal diversification during the late Paleozoic ice age: Paleobiology, v. 35, p. 367–392, doi:10.1666/0094-8373 -35.3.367. Heckel, P.H., 1990, Evidence for global (glacial-eustatic) controls over Upper Carboniferous (Pennsylvanian) cyclothems in midcontinent North America, in Hardman, R.F.P., and Brooks, J., eds., Tectonic Events Responsible for Britain’s Oil and Gas Reserves: London Geological Society Special Publication 55, p. 35–47. Heckel, P.H., 1995, Evaluation of evidence for glacial-eustatic control over marine Pennsylvanian cyclothems in North America and consideration of possible tectonic effects, in Dennison, J.M., and Ettensohn, F.R., eds., Tectonic and Eustatic Controls on Sedimentary Cycles: Society of Sedimentary Geology 4, p. 65–87. Heckel, P.H., 2008, Pennsylvanian cyclothems in Midcontinent North America as far-field effects of waxing and waning of Gondwana ice sheets, in Fielding, C.R., Frank, T.D., and Isbell, J.L., eds., Resolving the Late Paleozoic Ice Age in Time and Space: Geological Society of America Special Paper 441, p. 275–289, doi:10.1130/2008.2441(19). Hook, R.W., and Baird, D., 1986, The Diamond Coal Mine of Linton, Ohio, and its Pennsylvanian-age vertebrates: Journal of Vertebrate Paleontology, v. 6, p. 174–190, doi:10.1080/02724634.1986.10011609. Hook, R.W., and Baird, D., 1993, A new fish and tetrapod assemblage from the Allegheny Group (Late Westphalian, Upper Carboniferous) of eastern Ohio, U.S.A., in Heidtke, U., compiler, New Research on PermoCarboniferous faunas: Polichia-Buch 29, Bad Durkheim, p. 143–154. Hook, R.W., and Ferm, J.C., 1985, A depositional model for the Linton tetrapod assemblage (Westphalian D, Upper Carboniferous) and its palaeoenvironmental significance: Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, v. 311, p. 101–109, doi:10.1098/ rstb.1985.0142. Hook, R.W., and Ferm, J.C., 1988, Paleoenvironmental controls on vertebratebearing abandoned channels in the Upper Carboniferous: Palaeoclimatology, Palaeogeography: Palaeoecology, v. 63, p. 159–181, doi:10 .1016/0031-0182(88)90095-8. Horton, D.E., and Poulsen, C.J., 2009, Paradox of late Paleozoic glacioeustacy: Geology, v. 37, p. 715–718, doi:10.1130/G30016A.1. Joeckel, R.M., 1995a, Paleosol below the Ames marine unit (Upper Pennsylvanian, Conemaugh Group) in the Appalachian Basin, U.S.A.: Variability on an ancient depositional landscape: Journal of Sedimentary Research, v. 65, p. 393–407. Joeckel, R.M., 1995b, Tectonic and paleoclimate significance of prominent upper Pennsylvanian (Virgilian/Stephanian) weathering profile, Iowa and Nebraska, U.S.A: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 118, p. 159–179, doi:10.1016/0031-0182(95)00005-8. Joeckel, R.M., 1999, Paleosol in the Galesburg Formation (Kansas City Group, Upper Pennsylvanian), Northern Midcontinent, U.S.A.: Evidence for climate change and mechanisms of marine transgressions: Journal of Sedimentary Research, v. 69, p. 720–737. Keller, W.D., 1968, Flint clay and a flint-clay facies: Clays and Clay Minerals, v. 16, p. 113–128, doi:10.1346/CCMN.1968.0160202. Kertis, C.A., 1985, Reducing hazards in underground coal mines through the recognition and delination of coalbed discontinuities caused by ancient channel processes: U.S. Bureau of Mines Report of Investigations 8987, 23 p. Kosanke, R.M., and Cecil, C.B., 1996, Late Pennsylvanian climate changes and palynomorph extinction, in Wnuk, C., and Pfefferkorn, H.W., eds., Reviews of Palaeobotany and Palynology, v. 90, p. 113–140. Kullman, J., 1985, Drastic changes in Carboniferous ammonoid rates of evolution, in Bayer, U., and Seilacher, A., eds., Sedimentary and evolutionary cycles: New York, Springer-Verlag, p. 35–47. Lund, R., 1975, Vertebrate fossil zonation and correlation of the Dunkard Basin, in Barlow, J.A., ed., Age of the Dunkard: Proceedings of the First I.C. White Symposium, West Virginia Geological and Economic Survey, p. 171–182. Lyons, P.C., and Zodrow, E.L., 1995, Early to middle Twentieth Century floral zonation schemes of the Pennsylvanian (Late Carboniferous) of North America and correlation with the Late Carboniferous of Europe,
60
Brezinski and Kollar
in Lyons, P.C., Morey, E.D., and Wagner, R.H., eds., Historical Perspective of Twentieth Century Carboniferous Paleobotany in North America: Geological Society of America Memoir 185, p. 277–292. Marrs, T.O., 1981, Lithologic characteristics and depositional environments of the non-marine Benwood Limestone (Upper Pennsylvanian) in the Dunkard Basin, Ohio, Pennsylvania, and West Virginia [unpublished M.S. thesis]: Pittsburgh, Pennsylvania, University of Pittsburgh, 107 p. Milici, R.C., 2005, Appalachian coal systems: Defining the coal systems of the Appalachian Basin, in Warwick, P., ed., Coal System Analysis: Geological Society of America Special Paper 387, p. 9–30, doi:10.1130/0-8137 -2387-6.9. Olsen, P.E., 1986, A 40-million-year lake record of early Mesozoic orbital climatic forcing: Science, v. 234, p. 842–848, doi:10.1126/science .234.4778.842. Olson, E.C., 1975, Vertebrates and the biostratigraphic position of the Dunkard, in Barlow, J.A., ed., Age of the Dunkard: Proceedings of the First I.C. White Symposium, West Virginia Geological and Economic Survey, p. 155–165. Pedlow, G.W., 1977, A peat island hypothesis for the formation of thick coal [unpublished Ph.D. dissertation]: Columbia, South Carolina, University of South Carolina, 104 p. Petzold, D.D., 1990, Depositional environments of the lacustrine Benwood Limestone (Pennsylvanian) and paleogeography of the Benwood and associated lacustrine rocks of Late Paleozoic age, Ohio, West Virginia, and Pennsylvania [unpublished Ph.D. dissertation]: Bloomington, Indiana University, 211 p. Pfefferkorn, H.W., Gastaldo, R.A., and DiMichele, W.A., 2000, Ecological stability during the Late Paleozoic cold interval, in Gastaldo, R.A., and DiMichele, W.A., eds., Phanerozoic Terrestrial Ecosystems: Paleontological Society Special Papers 6, p. 63–78. Pfefferkorn, H.W., Gastaldo, R.A., DiMichele, W.A., and Phillips, T.L., 2008, Pennsylvanian tropical floras from the United States as a record of changing climate, in Fielding, C.R., Frank, T.D., and Isbell, J.L., eds., Resolving the Late Paleozoic Ice Age in Time and Space: Geological Society of America Special Paper 441, p. 305–316, doi:10.1130/2008.2441(21). Picard, M.D., and High, L.R., 1981, Physical stratigraphy of ancient lacustrine deposits, in Ethridge, F.G., and Flores R.M., eds., Recent and Ancient Nonmarine Depositional Environments: Models for Exploration: Society for Sedimentary Geology Special Publication 31, p. 233 – 259. Powell, M.G., 2005, Climatic basis for sluggish macroevolution during the late Paleozoic ice age: Geology, v. 33, p. 381–384, doi:10.1130/G21155.1. Powell, M.G., 2007, Geographic range and genus longevity of late Paleozoic brachiopods: Paleobiology, v. 33, p. 530–546, doi:10.1666/07011.1. Retallack, G.J., 2001, Soils of the Past: An introduction to paleopedology: Oxford, Blackwell Science, 600 p. Rogers, H.D., 1858, The Geology of Pennsylvania. v.2: Philadelphia, J.B. Lippincott and Co., 1045 p. Romer, A.S., 1952, Late Pennsylvanian and Early Permian vertebrates in the Pittsburgh–West Virginia region: Annals of Carnegie Museum, v. 33, p. 47–113.
Ruppert, L.F., Tewalt, S.J., Wallack, R.N., Bragg, L.J., Brezinski, D.K., Carlton, R.W., Butler, D.T., and Calef, F.J., III, 2001, A Digital Resource Model of the Middle Pennsylvanian Upper Freeport Coal Bed, Allegheny Group, Northern Appalachian Basin Coal Region: U.S. Geological Survey Professional Paper 1625-C, p. D1–D101. Skema, V., 2005, Stops 10 and 11-US 422 at Moravia Street Interchange, in Fleeger, G.M., ed., Type Sections and Stereotype Sections: Glacial and Bedrock Geology in Beaver, Lawrence, Mercer, and Crawford Counties: 70th Annual Field Conference of Pennsylvania Geologists: Harrisburg, Pennsylvania, Pennsylvania Geological Survey, p. 129–144. Stanley, S.M., and Powell, M.G., 2003, Depressed rates of origination and extinction during the late Paleozoic ice age: A new state for the global marine ecosystem: Geology, v. 31, p. 877–880, doi:10.1130/G19654R.1. Stiles, C.A., Mora, C.I., and Driese, S.G., 2001, Pedogenic iron-manganese nodules in Vertisols: A new proxy for paleoprecipitation: Geology, v. 29, p. 943–946, doi:10.1130/0091-7613(2001)029<0943:PIMNIV>2.0.CO;2. Swartz, C.K., and Baker, W.A., 1920, Second report on the coals of Maryland: Maryland Geological Survey, v. XI, no. Part 1, p. 29. Tabor, N.J., Montanez, I.P., Scotese, C.R., Poulsen, C.J., and Mack, G.H., 2008, Paleosol archives of environmental and climatic history of paleotropical western Pangea during the latest Pennsylvanian through Early Permian, in Fielding, C.R., Frank, T.D., and Isbell, J.L., eds., Resolving the Late Paleozoic Ice Age in Time and Space: Geological Society of America Special Paper 441, p. 291–303, doi:10.1130/2008.2441(20). Van Houten, F.B., 1962, Cyclic sedimentation and the origin of analcime-rich Upper Triassic Lockatong Formation, west-central New Jersey and adjacent Pennsylvania: American Journal of Science, v. 260, p. 561–576, doi:10.2475/ajs.260.8.561. Wanless, H.R., and Shepard, F.P., 1936, Sea level and climate change related to Late Paleozoic cycles: Geological Society of America Bulletin, v. 47, p. 1177–1206. Weller, J.M., 1931, The concept of cyclic sedimentation during the Pennsylvanian Period: Illinois State Geological Survey Bulletin 60, p. 153–177. White, I.C., 1891, Stratigraphy of the bituminous coal fields in Pennsylvania, Ohio, and West Virginia: Bulletin of the U.S. Geological Survey, v. 65, 212 p. Williams, E.G., Bergenback, R.E., Falla, W.S., and Udagawa, S., 1968, Origin of some Pennsylvanian underclays in western Pennsylvania: Journal of Sedimentary Research, v. 38, p. 1179–1193. Williams, E.G., and Bragonier, W., 1985, Origin of the Mercer high-alumina clay, in Gold, D.P., ed., Central Pennsylvania revisited: 50th Annual Field Conference of Pennsylvania Geologists: Harrisburg, Pennsylvania, Pennsylvania Geological Survey, p. 204–211.
MANUSCRIPT ACCEPTED BY THE SOCIETY 9 DECEMBER 2010
Printed in the USA
The Geological Society of America Field Guide 20 2011
Landslides in the vicinity of Pittsburgh, Pennsylvania Richard E. Gray DiGioia, Gray & Associates LLC, 570 Beatty Road, Monroeville, Pennsylvania 15146, USA James V. Hamel Hamel Geotechnical Consultants, 1992 Butler Drive, Monroeville, Pennsylvania 15146 USA William R. Adams Jr. Pennsylvania Department of Transportation, 45 Thoms Run Road, Bridgeville, Pennsylvania 15017, USA
ABSTRACT The Pittsburgh region has long been recognized as one of major landslide activity. This results from the geology and geomorphic processes shaping the region. The underlying bedrock of flat-lying interbedded strong and weak sedimentary strata has been acted upon by erosion, stress relief, and mass wasting, including creep and landsliding processes, to produce masses of marginally stable colluvial rock and soil on many of the steep hillsides common to the region. Landsliding often involves re-activation of such rock and soil masses. Recent landsliding is often triggered by heavy precipitation and by human activities, i.e., slope excavation, fill placement, and changes in long-established patterns of surface and subsurface drainage. This field trip has four stops, all within 20 mi of downtown Pittsburgh. Each stop is along a transportation corridor (railroad, local road, and two along an interstate highway). Each stop has various sized examples of the types of landslides common to the region. Most of these examples involve reactivation of unrecognized colluvial landslide masses.
INTRODUCTION By Richard E. Gray and James V. Hamel
unstable or marginally stable rock and soil masses (colluvium, i.e., old landslide debris) on many of the hillsides of the region. This field trip includes four stops (Fig. 1) which present examples of slope movements common to the Pittsburgh region: Stop 1—Grandview Avenue, Mount Washington, overlooking downtown Pittsburgh. • Regional geology and history. • Landsliding along Mount Washington Slope. Stop 2—Webster Road, Municipality of Plum Borough. • Typical colluvial hillside with common landslide features.
Pittsburgh is located in the Appalachian Plateau physiographic province (Fig. 1). With its steep hillsides, interbedded strong and weak sedimentary rocks, thick soil cover and precipitation of 890–1140 mm (35–45 in) per year, with the greatest amounts occurring in late winter and early spring, the Appalachian Plateau has long been recognized as an area of major landslide severity. Past landslides, both ancient and recent, have left
Gray, R.E., Hamel, J.V., and Adams, W.R., Jr., 2011, Landslides in the vicinity of Pittsburgh, Pennsylvania, in Ruffolo, R.M., and Ciampaglio, C.N., eds., From the Shield to the Sea: Geological Field Trips from the 2011 Joint Meeting of the GSA Northeastern and North-Central Sections: Geological Society of America Field Guide 20, p. 61–85, doi: 10.1130/2011.0020(04). For permission to copy, contact
[email protected]. ©2011 The Geological Society of America. All rights reserved.
61
62
Gray et al.
• Local road closed by small- to medium-sized landslides involving road fill over colluvium. Stop 3—Interstate Route 79 ~2.5 mi north of Ohio River. • Exposures of common local stratigraphy (including infamous Pittsburgh red beds) often involved in landslides, small masses of slumped bedrock, and a valley stress relief bedding plane shear zone. • Typical colluvial landslide features. Stop 4—Interstate Route 79 ~1 mi north of Ohio River. • Partially excavated colluvial soil and rock slide remnant. • Large sandstone blocks separated by open stress relief joints creeping and intermittently sliding downslope. • Huge amphitheater at rear of ancient rock slide.
• Rock slide scarp along stress relief joints in sandstone. Most of the information that follows on physiography, geology, slope formation, landsliding and geotechnical data has been derived from “Slope Stability in the Appalachian Plateau, Pennsylvania and West Virginia” by Gray et al. (1979). Most of this information is still applicable 32 years after original publication. There have, however, been some significant advances in our understanding of slope development and landslide processes over the past three decades (Ferguson and Hamel, 1981; Hamel, 1980; Hamel and Adams, 1981; Adams, 1986; Hamel, 1998; Hamel et al., 1998; Hamel, 2004). Some of these recent developments are inserted in the following text and others are presented in the descriptions of Stops 1–4.
Figure 1. Field trip stops, Appalachian Plateau, Western Pennsylvania. Modified from Gray et al. (1979).
Landslides in the vicinity of Pittsburgh, Pennsylvania PHYSIOGRAPHY AND GEOLOGY The Appalachian Plateau is a naturally dissected upland surface developed on gently folded but essentially flat-lying sedimentary rocks. In Pennsylvania and West Virginia, the Appalachian Plateaus Province trends northeast to southwest. It is bounded on the southeast by the Ridge and Valley Province and on the northwest by the Central Lowlands Province. Elevations range from 180 m (600 ft) along the Ohio River to ~1100 m (3700 ft) along the Allegheny Front, an escarpment forming the eastern boundary of the province. Erosion by streams and rivers has been intense and deep valleys and moderate to steep slopes form hilly to mountainous terrain. Along the major streams, local relief of 120–150 m (400–500 ft) is common. A small portion of the northwest section of the Appalachian Plateau was glaciated during the Pleistocene Epoch (Fig. 1). Glaciation subdued the previously existing topography, which was likely similar to the present topography in unglaciated sections, by infilling valleys and mantling upland surfaces with ice contact and other drift deposits. The Pittsburgh area has not been glaciated (Fig. 1). Although not directly affected by the ice sheets, unglaciated portions of the plateau were influenced in an important way by the periglacial climate. With regard to slope stability, the most significant periglacial effects were the greater rates of weathering, soil formation, and mass wasting (Denny, 1956; Philbrick, 1961; Rapp, 1967). Rock strata in the Appalachian Plateau are Devonian, Mississippian, Pennsylvanian, and Permian in age (Fig. 2). Rocks of Pennsylvanian age form the preponderance of surface strata. Pittsburgh is located within the portion of the Plateau underlain by Pennsylvanian rocks. Due to limitations on travel time for a one-day trip, all of our stops are in Pennsylvanian age rocks or soils derived from them. A thick band of Mississippian age rocks outcrops between the Pennsylvanian and Devonian rocks in the northerly portion of the area (Fig. 2) and Permian age rocks outcrop in the south-central portion. The structural trend of the region has a northeasterly direction and several anticlines and synclines extend for long distances. The dip angles associated with the fold structures are negligible to at most only a few degrees. On the east side of the plateau, adjacent to the Ridge and Valley Province, the dip of the strata increases to a maximum of 10° and Mississippian and Devonian strata are exposed on the higher ridges (Fig. 2). A detailed summary of the structure of the Appalachian Plateaus Province has been given by Rodgers (1970, p. 12–30). Mississippian and Devonian strata are predominantly shale, sandstone, and limestone (Fig. 3). The Permian and Pennsylvanian strata are characterized by thin cyclic sequences of sandstone, shale, claystone, coal, and limestone (Philbrick, 1953, 1959, 1960). SLOPE FORMATION Current slope development in the unglaciated portion of the Appalachian Plateau is consistent with flat-lying sedimentary
63
rocks in a temperate, humid climate. The occurrence of alternating weak and resistant rock strata is reflected topographically by breaks in slope and somewhat subdued to well-developed erosional benches. Hillside benches produced by landsliding exist at many locations; see Stops 3 and 4. Existing and past climatic conditions have resulted in substantial mechanical and chemical weathering which produced a residual or colluvial soil mantle over almost the entire rock surface. The sedimentary rock strata are normally not exposed. There is considerable evidence that rocks of this region remain highly stressed (Ferguson, 1967, 1974; Dahl and Parsons, 1971; Voight, 1974), except where stresses have been relieved near-surface in the valley walls and floors (Ferguson, 1967, 1974; Ferguson and Hamel, 1981). Stress relief fracturing is associated with many types of mass-wasting; see Stops 1−4. Joints caused by the local release of residual stress are closely spaced (2–3 m) in sandstone and limestone, whereas joints caused by tectonic stresses exhibit a spacing of many meters (Nickelsen and Hough, 1967). The finer-grained rocks have closely spaced joints. Nickelsen and Hough (1967) presented details of tectonic joint patterns, trends and spacing in the Appalachian Plateau of Pennsylvania. Except locally where sandstone may be abundant, the predominance of fine-grained rock (shale and claystone) within the geologic section results in soils typically being silty clay or clayey silt with rock fragments. Residual soils are characteristic of the flat upland surfaces and flat surfaces of larger erosional benches, with colluvial soils formed on slopes. Creep and sliding results in downslope movement of the soil and its accumulation on slopes and at the toes of slopes in colluvial masses. Colluvial soils tend to be 1.5–9 m (5–30 ft) thick on slopes and generally increase in thickness (to a maximum of ~30 m or 100 ft) near the toes of slopes unless there is active stream erosion. Colluvial soils are generally stiff to hard, and individual samples have relatively high shear strengths. However, creep or sliding processes (or both) during slope development have generally reduced the shear strength along movement surfaces to residual or near-residual levels. These low strength movement surfaces may occur at several levels within the colluvial mass but there is always a low strength movement surface at the soil-rock interface (Deere and Patton, 1971; Hamel, 1980). As the slope materials seek equilibrium between stress and strength, the soil mantle moves downslope and the mean slope angle decreases until a relatively flat slope angle, compatible with a state of marginal equilibrium, is achieved. This natural slope-flattening process accounts for the relatively thick soil cover on mature colluvial slopes, particularly at the base of slopes. Deere and Patton (1971) have suggested that there are no stable natural slopes in the Appalachian Plateau where the inclination exceeds 12°–14°. Terzaghi and Peck (1948, p. 357) reported movements on slopes as flat as 10°, whereas Gray and Donovan (1971) demonstrated that several mature colluvial slopes, with evidence of preexisting failure surfaces, had slope angles ranging from 7° to 10°. Hamel (1980) presents additional information on colluvial slope inclinations.
64
Gray et al.
We do not have any documented data on rates of creep of colluvial slopes in the area. Observations suggest, however, that colluvial slopes may creep at rates of a few centimeters per year. With the exception of creep, large colluvial masses appear stable unless disturbed by cutting, filling, drainage changes, or extreme precipitation events.
LANDSLIDING The Appalachian Plateaus Province is among the most severe for landsliding within the United States (Ladd, 1927, 1928; Sharpe and Dosch, 1942; Ackenheil, 1954; Eckel, 1958; Baker and Chieruzzi, 1959). Most landslides are in soil, the most
Figure 2. Regional geology. Modified from Gray et al. (1979).
Landslides in the vicinity of Pittsburgh, Pennsylvania common being slump-type slides or slow earth flows which range in size up to several million cubic meters. Rockfalls, the next most common type of slide, are typically much smaller with maximum volumes on the order of a few hundred cubic meters. Other types of slide movements do occur, however; see Stop 1.
65
Numerous small slump or slow earthflow slides occur during seasonal wet periods or due to local stream erosion, with catastrophic hydrological events being of major significance. For example, the great amount of precipitation associated with Hurricane Agnes in June 1972 caused a significant number of
Figure 3. Generalized geologic column for Allegheny County. Modified from Harper (1990); used with permission.
66
Gray et al.
such slides. However, most slides are a direct result of man’s disturbance of natural conditions. Frequent causes of sliding are (1) removal of toe support, (2) surcharging slopes by the placing of fill embankments, or (3) a change in surface and subsurface water flow. The largest slides usually result from disturbance of ancient landslide masses in soils and/or rock. These ancient landslides appear to have occurred in the main under periglacial conditions. Limited radiocarbon dating (Philbrick, 1961; D’Appolonia et al., 1967) suggests a Pleistocene age for some of these deposits. Peltier (1950) and Denny (1956) found fossil periglacial features close to the front of the maximum advance of the Wisconsinan glaciation in Pennsylvania and strongly supported the influence of Pleistocene periglacial processes on slopes. In a study of a portion of central Pennsylvania just east of the Appalachian Plateau, Rapp (1967), concurred with the above authors. In a review the effect of periglacial processes on slope profiles in areas currently experiencing humid temperate climates, Carson and Kirkby (1972) concluded the effect is not as great as indicated by the above studies. Where interbedded strong and weak rock strata are exposed, differential weathering and erosion result in the weaker rock being removed, leaving the more resistant rock as overhanging ledges. The result of this process is small but often dangerous rockfalls. Cuts containing hard sandstone or limestone beds underlain by relatively low-strength shale or claystone are common throughout the Pittsburgh area. Weathering causes relatively rapid decomposition and spalling of the softer rock, leaving unsupported ledges of limestone and sandstone. Rates of undercutting of ~60–180 mm (2–7 in) per year have been reported by Philbrick (1959), based on observations over a period of several years for a highway cut with claystone underlying massive Morgantown Sandstone. Average rates of undercutting of 13–38 mm (0.5–1.5 in) per year (over a period of two years) were measured by Bonk (1964) at two carefully prepared test sites in excavated slopes in Pittsburgh red beds claystone. Average rates of undercutting of 30–150 mm (1–6 in.) per year were reported by Bonk (1964) for Pittsburgh red beds exposed for periods of two to ten years in highway cuts. In relatively short periods of time, weathering can progress to the point where a resistant rock ledge can no longer sustain its cantilevered weight and the ledge falls. Vertical to subvertical stress relief joints often form the backs of rock fall blocks. Deep-seated rock slides are relatively rare under present climatic conditions, but many are believed to have occurred under the more severe climatic conditions of Pleistocene time; see Stops 1, 3, and 4. Recent deep-seated rock slides have typically involved excavated slopes in which large wedges of rock, separated from the valley walls by near-vertical stress relief joints, slide along or through weak claystone or shale beds. Water pressures in the slopes are usually significant contributing factors in such slides (see Case History 3 in Gray et al., 1979; Hamel, 1998; and introductory material for Stops 3 and 4).
GEOTECHNICAL DATA The engineering properties of intact rock materials of the northern Appalachian Plateau vary widely with lithology and degree of weathering. Almost no data on the permeability and deformability of these rocks have been published and relatively few strength data are available in the literature. Most of the rocks of the Appalachian Plateau are of very low to medium strength and low to medium modulus ratio according to the classification system of Deere (1968). As with most rocks, the engineering behavior of the sedimentary rocks of the Appalachian Plateau is controlled largely by geologic discontinuities; e.g., faults, joints, bedding contacts, weak beds or beds affected by fluid pressure conditions, and old sliding surfaces, rather than by properties of intact rock materials. Even the engineering behavior of colluvium derived from these rocks is governed mainly by old failure surfaces. Due to a history of previous shearing, these old failure surfaces generally have much lower shear strengths and much higher permeabilities and deformabilties than other portions of the colluvium or the rocks from which the colluvium was derived. Index properties and strength data on colluvium derived from claystones of the Appalachian Plateau have been given by D’Appolonia et al. (1967), Hooper (1969), Hamel (1969), Hamel and Flint (1969, 1972). This colluvium ranges from massive blocks of relatively intact claystone to silty or sandy clay soil with rock fragments. Dry unit weights typically range from ~1.9–2.2 Mg/m3 (118−137 pcf). A recent paper by Nakamura et al. (2010) identified three groups of soils. In the first group of soils, sliding appeared to be controlled by minerals such as quartz, feldspar, calcite, dolomite and layer silicate minerals other than smectite, vermiculite, chlorite, and mica and their residual strength was almost constant at ~32°. In the second group of soils, the controlling mineralogical factor for sliding shifted from non-preferred–orientation minerals to preferred-orientation minerals and residual strength decreased from 30° to 10°. In the third group of soils, sliding is well controlled by preferred-orientation layer silicate minerals and residual strength gradually decreased from 10° to 5°. They concluded that the total content of layer silicate minerals prone to preferred orientation (smectite, vermiculites, chlorite, and mica) in the sub–425 µm soil fraction is a suitable mineralogical parameter for predicting the magnitude of residual strength. The colluvium generally exhibits strain-softening behavior (Skempton, 1964) and its residual (large displacement) shear strength is generally less than half its peak (small displacement) strength at a given effective normal stress. For effective normal stresses of less than ~350 kN/m2, the peak strength of claystone colluvium is commonly characterized by cohesion intercepts of 7–35 kN/m2 and friction angles of 20–25° while the residual strength is usually characterized by negligible cohesion intercepts and friction angles of 8–20°. Measured residual friction angles for most claystone-derived colluvium are on the order of 11–16°. Experience in calculation of strength data from colluvial slide
Landslides in the vicinity of Pittsburgh, Pennsylvania masses (Hamel, 1969; Hamel and Flint, 1969, 1972; Gray and Donovan, 1971; Hamel, 1980, 2004) indicates that shear strengths characterized by residual level friction angles of 13–16°, with zero cohesion intercept, are commonly mobilized in place. STOP 1. MOUNT WASHINGTON By Richard E. Gray (Permission to use this amended text was granted by the Pennsylvania Geologic Survey, sponsor of the Annual Field Conference of Pennsylvanian Geologists, and by the author [Gardner, 1980].) Pittsburgh is located in a moderately dissected portion of the Appalachian Plateau where a relatively flat plateau surface is cut by streams producing steep sided valleys having a vertical relief of 425–560 ft (130–170 m). Here on Mount Washington, we are at approximately elevation 1140 ft. The normal pool level of the rivers below is elevation 710 ft. Pittsburgh is located at the confluence of the three largest rivers in the region, the Allegheny, Monongahela, and the Ohio. The Allegheny River, flows from the north originating in northern Pennsylvania and southern New York. The Monongahela River flows from the south, originating in east-central West Virginia. The Allegheny and Monongahela Rivers meet to form the Ohio. The Ohio River is a major artery of drainage that flows west and joins the Mississippi River at Cairo, Illinois, ~930 mi (1500 km) downstream from Pittsburgh. While Pittsburgh was not directly glaciated during the Pleistocene, the closest approach of Wisconsinan ice was ~30 mi (50 km) north of the city; the Allegheny and Ohio Rivers served as channels for glacial meltwaters. The heavy sediment load in the glacial outwash caused river aggradation that is now represented by coarse sand and gravel deposits and low lying terraces along the Allegheny and Ohio Rivers. Much older alluvial sediments, possibly of nonglacial origin, occur in high terraces and old alluvial channels that occur ~325 ft (100 m) above the present channels. The age of the high level deposits is thought to be Illinoian. The Monongahela River flows north through predominantly shales so that its alluvial deposits consist mainly of silt and clay. It was the strategic location at the confluence of the rivers that first attracted the attention of Europeans to the “Forks of the Ohio” at what is now Pittsburgh. The conflicts between the British and French in Europe in the early and mid-1700s were extended to North America as both nations struggled for domination of the continent. The French claimed the area west of the Allegheny Mountains as theirs, including the combined Ohio and Allegheny Rivers; the English did not recognize these claims. Ensuing clashes between the French and English trading in the area prompted Governor Dinwiddie to send a 21-year-old major of the Virginia Militia, George Washington, to deliver a protest to the French. En route, Major Washington traveled by the Forks and noted: …I spent some time viewing the rivers, and the land in the Fork; which I think extremely well situated for a fort, as it has absolute command
67
of both rivers…the Land at the point is 20 to 25 feet above the common surface of the water; and a considerable bottom of flat, well-timbered land all around it, very convenient for building…. (from Washington’s Chronicle, in Lorant, 1975)
The confrontations with the French prompted the Virginians to build a fort at the Forks as suggested by Washington. Construction of Fort Prince George was initiated in March 1754, and was the first recorded Euro-American construction on the land that is now Pittsburgh. The unfinished colonial fort was abandoned one month later when a superior force of French and Indians threatened attack. The French then erected their own fort, Fort Duquesne, at the Forks. The French controlled the forks for four years, repelling several English attempts to regain control. In November of 1758, the French burned and abandoned Fort Duquesne in the face of imminent attack by British forces led by General John Forbes and Colonel George Washington. The English erected their own fort (now partially reconstructed in Point State Park) on the ruins of Fort Duquesne, and Forbes named it Fort Pitt in honor of the English Prime Minister. Fort Pitt received no attacks from the French, although it suffered a siege by Indians in 1763 during “Pontiac’s War.” The end of the Indian uprising reduced the need for Fort Pitt, and it was gradually dismantled in the mid-1760s. The community that developed around Fort Pitt continued to grow as a center of trade for the ever increasing travel from east to west. When the community was incorporated as a city in 1816, it was the major center for commerce in the west since most travel from the east coast to the west went through Pittsburgh. Pittsburgh’s economy was primarily based on commerce in the late 1700s and early 1800s. As Pittsburgh grew, it required an ever-increasing supply of goods, most of which were manufactured in the east. However, transporting large quantities of goods was incredibly difficult and expensive because rugged mountains were a formidable barrier between Pittsburgh and the east. This led to Pittsburgh developing a manufacturing industry. By 1830, the commerce aspect of Pittsburgh’s economy was eclipsed by manufacturing. Thus, Pittsburgh was founded and began to flourish as a center of commerce and manufacturing because of its geography. But Pittsburgh was only born of its geography; it owes most of its growth and eventual status as a leading industrial center to its geology. Although less than two percent of the local rock stratigraphy is coal, it was coal that made Pittsburgh an industrial giant—the Iron and Steel Center of the World. Other industries developed in the region as a result of the exploitation of the local geologic resources. The Pittsburgh glass industry began about the same time as the iron industry, because the resources for making glass (sand and lime) were available, and it was difficult to import glass over the rugged mountains from the eastern manufacturing centers. Glass manufacturing first occurred in Pittsburgh in 1797. The first cheap aluminum ($2.00/pound) was produced in Pittsburgh on 25 November 1888, in a factory developed by Martin Hall, the inventor of the
68
Gray et al.
process. A large brick, pipe, and refractory manufacturing industry was fostered in the region by the occurrence of abundant clay (fireclay) and shale. One of the most significant resources, other than coal, to affect the development of the Pittsburgh area was oil and gas. In 1859, the Drake Well was drilled near Titusville, ~90 mi (145 km) north of Pittsburgh, and major oil production was born. By 1871, 60 oil refineries were operating in Pittsburgh producing 36,000 barrels of oil per day. Natural gas was developed about the same time, with the first gas well near Pittsburgh being drilled in 1878 in Murrysville, 12 mi (20 km) east of the city. In 1883, a gas pipeline was completed from Murrysville to Pittsburgh to feed, among other things, the gas street lights of the city. Pittsburgh’s strategic location as a “Gateway to the West” necessitated use of the rivers as the chief avenue of haulage. Railroads did not enter the area until the 1850s, and the rivers provided the quickest, easiest avenue for transporting large loads, as they still do today. Boatbuilding was an important industry of Pittsburgh from its earliest days. The first steamboat on western waters was the New Orleans, which was built in Pittsburgh in 1811. Although the Ohio River was a major pathway to the interior of the continent, traveling from Pittsburgh to the Mississippi in anything much larger than a canoe was usually restricted to the wetter seasons because the Ohio River was often too shallow for navigation during the summer and fall. In addition, transporting goods prior to the 1820s was usually unidirectional, downstream, because the current was too strong for upstream trips by larger boats. The advent of the steamboat on the Ohio River eliminated this problem, but the problem of seasonal navigation still remained. It wasn’t until 1929, when a series of 50 locks, dams, and canalization had been completed, that the Ohio River was totally free of seasonal restrictions to navigation. The Pittsburgh Coal is located ~100 ft below us. In 1759, British soldiers developed a coal mine on “Coal Hill,” now Mount Washington. Coal was mined on a small scale until industrialization created greater demand in the mid-1800s. The Pittsburgh coal is considered to be one of the richest economic deposits in the world. The U.S. Geological Survey estimated that the Pittsburgh coal alone yielded eight billion tons from the early 1900s to 1965, comprising 35 percent of all bituminous coal in the Appalachian Basin and 21 percent of the cumulative production for the entire United States. The Pittsburgh coal is essentially “worked-out” and no longer deep mined in Pittsburgh. The principal user of coal in the Pittsburgh region was the iron and steel industry. With this background of mining, mineral extraction, and heavy industry, along with its three rivers, topography, and geology, it is not surprising that Pittsburgh became a major center for practice in engineering geology and geotechnical engineering. A decade ago, Hamel and Adams (2000; see GSA Data Repository1) presented an update on engineering geology problems,
challenges, and practices in the Pittsburgh area. The material in that paper is still relevant today. Landsliding, of course, presents significant challenges at many locations in the Pittsburgh area, including the slope below, as described in the following guidebook section.
1
2 GSA Data Depository item 2011161, “Mt. Washington Slope—Duquesne Incline to Smithfield Street Bridge,” is available at www.geosociety.org/pubs/ ft2011.htm or on request from
[email protected].
GSA Data Repository item 2011160, “Update on Engineering Geology in the Pittsburgh Area” is available at www.geosociety.org/pubs/ft2011.htm or on request from
[email protected].
LANDSLIDING ALONG SLOPE BELOW MOUNT WASHINGTON By James V. Hamel Observation areas along Grandview Avenue on Mount Washington indeed provide a “grand view” of the Allegheny and Monongahela Rivers joining to form the Ohio River at the “Point” of downtown Pittsburgh. The slope below Mount Washington, along the downstream end of the Monongahela Valley and the upstream end of the Ohio Valley (Fig. 4), has a long history of mass-wasting, including various landslide and rockfall processes (Ackenheil, 1954, 1958; Hamel, 1998; Hamel et al., 1998; Pomeroy, 1974). The following information is summarized mainly from Hamel (2000; see GSA Data Repository2.) Permission to use this material was granted by the Pennsylvania Geologic Survey, sponsor of the Annual Field Conference of Pennsylvania Geologists. The three major rivers here now have nominal elevation (El.) 710 ft maintained as normal pool of the navigation dam at Emsworth, 6 mi down the Ohio River from the “Point.” The Mount Washington Slope extends from this river level up to approximately El. 1150 ft along Grandview Avenue (Figures. 4–7). Rocks here are flat-lying sedimentary strata of the Pennsylvanian age Conemaugh and Monongahela Groups. The Ames Limestone marking the top of the Lower Conemaugh Glenshaw Formation is ~40 ft above river level at nominal El. 750 ft. The Pittsburgh Coal marking the base of the Monongahela Group is ~100 ft below Grandview Avenue at nominal El. 1050 ft. The first known mining of the Pittsburgh Coal was done here on Mount Washington, then called “Coal Hill,” ca. 1760; see the Historical Marker on Grandview Avenue. This area is generally considered to have been eroded to Mount Washington ridgetop El. 1200 ft (south of Grandview Avenue) by the end of the Tertiary. During the Pleistocene, continental glaciers advanced to a line ~30 mi northwest of Pittsburgh on at least two occasions—at least once during the Illinoian period and at least once during the Wisconsinan period. Erosion of the three main river valleys down to the Parker Strath, nominal El. 900 ft at Pittsburgh, is generally thought to have been done by the Illinoian period when glacial outwash was deposited here up to nominal El. 1000 ft (Fig. 6). The three main rivers then eroded their channels deeper into bedrock, down to nominal El. 660 ft beneath the Pittsburgh “Point,” during late Illinoian time (Figs. 6 and 7). During the subsequent
Landslides in the vicinity of Pittsburgh, Pennsylvania
69
Figure 4. Map of Mount Washington Slope, Duquesne Incline to Smithfield Street Bridge.
Figure 5. Map of Pittsburgh with cross sections A–A′ and B–B′ (after Hamel, 1998)
70
Gray et al.
Figure 6. Cross section A–A′ (after Hamel, 1998).
Wisconsinan period, little, if any, additional downward erosion of the bedrock channels is thought to have occurred. Extensive Wisconsinan outwash was deposited and reworked in the river valleys, with remnants existing up to nominal El. 800 ft at and near Pittsburgh (Fig. 6). During the Illinoian and Wisconsinan periods, the southerly flowing Allegheny River carried huge quantities of meltwater and
outwash. Glacial outwash is thought to have dammed the Allegheny and Monongahela Rivers at Pittsburgh on several occasions, producing significant ponding events with extensive sediment deposition in slackwater. Ice jams from the Allegheny River may have caused additional ponding events. Valley stress relief (Ferguson, 1967; Ferguson and Hamel, 1981) during the relatively rapid (on a geologic time scale)
Figure 7. Cross section B–B′ (after Hamel, 1998).
Landslides in the vicinity of Pittsburgh, Pennsylvania Pleistocene erosion of bedrock valleys loosened rock masses in the valley walls. The severe periglacial climate and vigorous hydraulic activity during Pleistocene time caused rock sliding in the valley walls and accelerated weathering of rock slide debris. Over time, these produced the marginally stable colluvial rock and soil masses currently existing on many slopes in the Pittsburgh area, including the Mount Washington Slope. As Pittsburgh grew over the past 250 years, extensive development occurred in flatter areas along the top and bottom of the Mount Washington Slope. Circa 1850, the first railroad was constructed along the toe of this slope. As Pittsburgh industries grew, additional railroad tracks were constructed ca. 1900 on a bench locally excavated into the slope toe (Fig. 7). The slope itself, with a height of 300–400 ft and an overall inclination of 1.5H:1V (34°) with some steeper segments, remained essentially undeveloped. This resulted mainly from difficult access and extensive slope instability. Numerous landslides and rockfalls have come down onto the railroad over the past 160 years. In the early 1990s, a Busway to Pittsburgh International Airport was considered for the railroad shelf along the toe of the Mount Washington Slope (Hamel et al., 1998). Preliminary geotechnical investigations from 1991 to 1993 identified considerable rockfall potential. Detailed geotechnical investigations in 1994 and 1995 provided further information on slope geology, historic slope instability along the railroad, and landslide and rockfall hazards along the proposed Busway. Construction of various stabilization measures along the Mount Washington Slope began in late 1996 and was terminated in early 1997 prior to completion. This portion of the Busway was deleted at that time when the overall project was considerably reduced in scope and length due to cost and political considerations. Given present economic conditions, it seems unlikely that the Busway segment along the toe of the Mount Washington Slope will ever be constructed. Busway slope investigations in 1994 and 1995 included review of historical vertical aerial photographs as well as large scale low-level oblique aerial photographs taken in April 1994. Extremely difficult access, along with safety and environmental concerns, precluded drilling on the slope. Field work mainly involved detailed reconnaissance and mapping on large scale (1:360) oblique aerial mosaics (Hamel et al., 1998). This field work disclosed many of the slope failure types and processes of Varnes (1978). Rockfalls and rock topples are abundant here. Many, if not most, of these failures are related to lateral rock spreads, rock slides, and rock slumps associated with valley stress relief (Ferguson, 1967; Ferguson and Hamel, 1981; Hamel, 1998). Debris slumps, debris slides, and slump-earthflows, all typical colluvial slope failure processes of the region (Hamel and Ferguson, 1999) are common along the Mount Washington Slope. Debris slides, debris flows, and debris avalanches, all of which involve movement of colluvium and/or talus down ravines or chutes, are also common here (Hamel et al., 1998). The most important geologic finding along the Mount Washington Slope was the previously hypothesized (Ferguson and
71
Hamel, 1981; Hamel and Adams, 1981; Hamel and Ferguson, 1999) but seldom reported existence of numerous partially eroded and/or partially excavated remnants of deep-seated slumps and translational block slides in bedrock (Fig. 7). These rock slides are thought to have occurred during the Pleistocene (probably Illinoian) when the rivers were actively entrenching their valleys in bedrock and climatic and hydraulic conditions (involving both surface and subsurface waters) were much more severe than those during the Holocene (Hamel, 1998). Portions of these rock slide masses were eroded during the Late Pleistocene and some additional portions were excavated during the two major phases of railroad construction ca. 1850 and 1900. Discovery and documentation of numerous rock slide remnants along the Mount Washington Slope lead to review of previously observed but widely scattered rock slides in the region (Hamel, 1998). These rock slides have significant geologic implications regarding the Pleistocene history of the region, i.e., processes of valley formation and colluvial slope development. They also have significant engineering implications related to continuing rockfalls and rock slides along railroads and highways. STOP 2. WEBSTER ROAD By William R. Adams Jr. State Route 2090, Landslides Stop 2 provides several examples of the type of slope movements that typically impact the highways in southwestern Pennsylvania. There are numerous landslides visible along and causing distress to Webster Road, a roadway owned and maintained by the Commonwealth of Pennsylvania. Webster Road is designated by the Commonwealth as State Route (SR) 2090 and it is located in eastern Allegheny County ~16 mi east northeast of Downtown Pittsburgh (see Fig. 1) in the municipality of Plum Borough. Webster Road is ~10,800 ft in length with most of the eastern end of the roadway, ~3400 ft, barricaded by concrete Jersey Barriers prohibiting travel by the public as a result of the many landslides along the roadway. There are 6 larger slope movements identified along Webster Road and their locations can be seen in Figure 8. We will travel past Landslide No. 1 and then stop at the western barricade before walking to slope movements 2 through 4. Of the 6 slope movements, 5 fall within the section of the roadway that is closed. Landslide No. 1 was left outside the barricade to enable access by property owners to certain portions of their property. Webster Road trends along the lower portions of northern valley walls of a tributary stream to Pucketa Creek. Pucketa Creek then flows toward the northwest into the Allegheny River. The general topography of the area is shown in Figure 9. The relief of the valley walls in the vicinity of the slope movements is slightly more than 400 ft with the top of the valley walls reaching elevations of ~1250 ft to1350 ft above mean sea level (msl). The portion of the roadway most affected by the slope movements
72
Gray et al.
varies in surface elevations from ~900 ft to approximately elevation 1020 ft above msl. Stratigraphically the valley wall slopes are underlain by the Casselman Formation and upper part of the Glenshaw Formation (see Fig. 3). The bedrock is composed of cyclothems made up of primarily shales, sandstones, claystones, limestones, and coal. These cyclic deposits are well described by Philbrick (1959). The well-known mineable coal seam of our region, the Pittsburgh coal, is present in some of the hill tops in the area. Based on available published mapping, the Ames limestone (see Fig. 3) should be present above most of the roadway. Blocks of the Ames limestone have been observed ~10 ft above Webster Road in the vicinity of Slope Movements numbered 5 and 6 at approximately elevation 920. It is unknown if these blocks are in place or are float blocks in a colluvial mass; however, based
on regional geologic information the blocks are very near the expected elevation of the Ames limestone. This places the roadway on the slope at or just below most of the thicker red bed units in the local bedrock including the Pittsburgh red beds, a claystone well known for slope instability. The rapid weathering of the underlying red beds (claystones) have been well reported and documented (Adams, 1986; Hamel, 1969). Some investigators have, also, measured the rate of weathering (Bonk, 1964). These claystones weather to a very weak soil which is very susceptible to landsliding. Structurally, the regional dip of the underlying bedrock is in a south southeastern direction at a slope of approximate 1.2 ft per 100 ft. This is basically perpendicular to and out of the valley walls along which Webster Road trends. The regional dip continues into the southern or north-facing valley walls. This regional
Figure 8. Location of landslides. Base map produced of the New Kensington East 7-1/2 min Quadrangle by the U.S. Geological Survey, 2010.
Landslides in the vicinity of Pittsburgh, Pennsylvania dip would cause the groundwater to flow out of the south-facing slopes along which Webster Road is constructed. This increase in groundwater could be a contributing factor to the numerous failures along the roadway. While there are typically a number of contributing factors to the occurrence of a landslide, there is usually one triggering factor and it frequently is associated with increased groundwater levels and/or the introduction of water into the soils on the slopes (Adams, 1986). Figure 10 is a map of the active or recently active landslides; soil, and rock susceptible to landsliding; and ancient landslides in the vicinity of Webster Road. This mapping was conducted as part of a regional study in southwestern Pennsylvania by the U.S. Geological Survey in the mid-1970s. The ancient or prehistoric landslides frequently leave large amphitheaters or large arcuate shapes in areas on the slope that represent the upper reaches of these ancient landslides. Depending on the vegetation cover during our site visit, some of these amphitheaters that are present on the valley walls may or may not be readily visible. Very large examples of these amphitheaters are present and will be visible in and around Stops 3 and 4. From the mapping illustrated in Figure 10, it appears that there is more evidence of ancient and recent movement on the
73
northern valley wall which again may be reflective of the regional dip and probable increase in groundwater flow out of the south facing valley walls. The contribution of these ancient failure masses to the increased likelihood of historic landsliding is well documented (D’Appolonia et al., 1967; Hamel and Flint, 1969; Adams, 1986). Inability to recognize these failures has resulted in many recent/present day failures (Hamel and Flint, 1969; Hamel and Adams, 1981). The District Geotechnical Unit was first called out in the spring of 2007 to investigate the slope movements impacting the roadway. Because of the rapid development of the slope movements and the number of them, little actual geotechnical/geological investigation was performed at the slope failures before the roadway was closed. By late May 2007, the roadway was officially closed for a length of 3400 ft. I first became involved with the project in the fall of 2009 when I returned to the Geotechnical Unit. As a result of the anticipated direction for addressing these failures, that is, to abandon the roadway and, also, the likely cost of the remediations, it was decided not to perform a detailed subsurface investigation. Some borings were drilled at Landslide No. 1 and Landslide No. 2; however, due to a lack of funding and staff, the
Figure 9. Topographic setting. Base map from the New Kensington East 7.5-minute Quadrangle, 1998 NIMA 5056 II SWSERIES V831.
74
Gray et al.
borings were not logged by a geologist or geotechnical engineer. Again because the investigation was terminated and no funds were available, the locations and surface elevations of the borings were not surveyed, so it was difficult to correlate them with the underlying stratigraphy. The closure of the roadway and the drilling of these borings all occurred prior to my returning to the Geotechnical Unit. The slope movements are in various stages of development with Landslides No. 1 (see Fig. 11) and 4 (see Fig. 14) most severely impacting the roadway. Landslide No. 2 (see Fig. 12) and Landslide No.3 (see Fig. 13) while slightly smaller than Landslides No.1 and 4, they are still large enough to restrict travel
to 1 lane or less. In total, there are 6 relatively larger slope movements. Besides these 6 there are numerous locations along the roadway where smaller slope movements are starting to develop. About 4 locations can be seen just east of Landslide No. 1, where smaller slope movements are starting to impact the roadway including an area where the guiderail has been completely undermined and is suspended in air over a portion of its length. While no detailed investigation has been conducted on these slope movements, they have the general appearance of one of the most common types of failures impacting roadways in this region, that is, complex type landslides. These typically have a slump type movement near the heads or upper portions of the
Figure 10. Slope Movement Features. Modified from Landslide Susceptibility Map of the New Kensington East 7-1/2′ Quadrangle, Allegheny County and Vicinity, Pennsylvania, Open File Map 74-283 by William E. Davies (Davies, 1974).
Landslides in the vicinity of Pittsburgh, Pennsylvania
Figure 11. View of Landslide No. 1 looking west.
landslides with the lower or toe areas turning into flow type movements (Eckel, 1958; Varnes, 1978). Landslides 5 and 6 are shown in Figures 15 and 16. They appear to be in the early stages of development at least at the roadway level. They also appear to be coalescing into one large landslide. The Pennsylvania Department of Transportation (PennDOT) is presently evaluating the permanent closing and possible abandonment of a portion of Webster Road. Preliminary cul-de-sac designs are under way to determine the most effective locations to permit turning around where the roadway is closed by barriers. Additionally, consideration must be given to possible relocation of property owners and land locking of property which may have permanent residents on it.
Figure 13. View of Landslide No. 3 looking east.
75
Figure 12. View of Landslide No. 2 looking east.
STOPS 3 AND 4. LANDSLIDES ALONG INTERSTATE ROUTE 79 By James V. Hamel Introduction Stops 3 and 4 on this field trip are along Interstate Route 79 (I-79) about 9 mi northwest of Pittsburgh (Fig. 17). Along this section of I-79, deep-seated ancient landslide masses high on the valley walls were not recognized during investigation and design of the highway in the early to mid-1960s. Sidehill excavations for highway construction in 1968 and 1969 removed the toes of marginally stable landslide masses and initiated progressive failures which propagated upslope. By mid-1969, reactivated ancient
Figure 14. View of Landslide No. 4 looking east.
76
Gray et al.
Figure 15. View of Landslide 5 looking east.
Figure 16. View of Landslide 6 looking east.
Figure 17. Location map for Stops 3 and 4 along Interstate 79 (after Flint and Hamel, 1971).
Landslides in the vicinity of Pittsburgh, Pennsylvania landslide masses extended discontinuously along the east valley wall from approximate highway Station (Sta.) 900–955. Stop 3, near the north end of this landslide zone, is at approximate Sta. 975–985 where partially excavated remnants of some of the smaller landslides remain. Stop 4 is near the south end of the landslide zone at approximate Sta. 900–910, where partially excavated remnants of the largest landslide mass remain. Before describing Stops 3 and 4, it is appropriate to present some background on the landslide situation along this highway segment. Initial Work, 1968–1969 Dr. Norman K. Flint (now deceased) and I had a rock slope research project under way with the Pennsylvania Department of Highways (predecessor of the Pennsylvania Department of Transportation) at the University of Pittsburgh in the fall of 1968. At that time, extensive landsliding occurred along this segment of I-79 which was then under construction. Because of their magnitude, proximity, and importance to the Department of Highways, the I-79 landslides became the focus of our research project. These landslides were located high on the east valley wall and they had basal failure surfaces at or near the base of a thick zone of weak claystone (Pittsburgh red beds) and/or colluvium derived from this claystone and overlying rock units (Fig. 18). Slope excavation and related landsliding proceeded rapidly from the summer of 1968 through the summer of 1969. By December 1968, there was a mile-long active landslide laboratory, mainly along the east side of Kilbuck Run Valley. I spent about three days there each week from December 1968 through May 1969 trying to keep up with the continuing development and progression of landslides along this active construction corridor. Field observations, notes, sketches, and photographs were all I could manage initially. In early 1969, I began taking samples of failure surface clay seams for laboratory testing. By May 1969, Bill Adams (one of the leaders of the present field trip) and I were obtaining high-quality block samples from fresh failure surfaces. Most of the laboratory shear strength tests on these samples were done by Bill Adams. Results of our work on the I-79 landslides were presented in several reports and papers from 1969 to 1972 (Hamel, 1969; Hamel and Flint, 1969, 1972; Flint and Hamel, 1971). Portions of the report by Hamel and Flint (1969) and the entire field trip guidebook section by Flint and Hamel (1971) are available from the GSA Data Repository.3,4 Note that the Interstate Route numbers were later changed from those in these early publications. I-279 in these early publications is now I-79 and I-79 in these early publications is now I-279. 3 GSA Data Depository item 2011162, “A Slope Stability Study on Interstate Routes 279 and 79 near Pittsburgh, Pennsylvania” is available at www.geosociety .org/pubs/ft2011.htm or on request from
[email protected].
4 GSA Data Depository item 2011163, “Engineering Geology at Two Sites on Interstate 279 and Interstate Route 79 Northwest of Pittsburgh, Pennsylvania” is available at www.geosociety.org/pubs/ft2011.htm or on request from
[email protected].
77
Subsequent Work, 1972–2010 I was away from the Pittsburgh area from 1969 to 1972. Since returning in 1972, I have continued to intermittently observe, consider, and evaluate the I-79 slopes. Most of the interpretations, conclusions, and recommendations presented from 1969 to 1972 are still considered valid. A few items, however, require clarification and/or amplification. Some of these items were mentioned, but not fully treated, in two later papers (Ferguson and Hamel, 1981; Hamel and Adams, 1981). Other items were addressed in a recent discussion (Hamel, 2004). At the time the work on I-79 was done in 1969, my focus was on residual strength (Skempton, 1964), progressive failure (Bjerrum, 1967), back-calculation of shear strength (Hamel, 1969), and the only published case history of a near-by colluvial slope in Weirton, West Virginia (D’Appolonia et al., 1967). I recognized that the I-79 colluvial slopes were similar in some ways to the Weirton slope, but that the rear (upslope) portions were more like old rock slide masses. Subsequent observations along I-79 (Hamel and Adams, 1981), along with experience elsewhere, indicated that this was indeed the case. Briefly, the connecting links between rock slides and colluvial slopes in the Upper Ohio River drainage basin (and probably elsewhere), which I had not yet recognized ca. 1970, are (1) the theory of valley stress relief in flat-lying sedimentary rocks (Ferguson 1967; Ferguson and Hamel, 1981) and (2) the Pleistocene history of the region (Leverett, 1902, 1934; Jacobson et al., 1988; Harper, 1997, 2002). These concepts are discussed further by Hamel (1998). The original rock slides, extending up to ridge-top level along the east wall of Kilbuck Run valley, probably occurred in the Early to Middle Pleistocene (Pre-Illinoian or Illinoian) on the order of 500,000–1,000,000 years ago (Harper, 1997, 2002). Lateral stress relief accompanying valley down-cutting, probably in conjunction with high pore and joint water pressures in valley wall rock due to periglacial precipitation (glacial ice front some 20 mi to northwest) and perhaps rapid drawdown of glacially ponded water, caused rock sliding along the valley wall. Subsequent sliding, creep, and weathering broke down the outer portion of the original rock slide masses into more typical colluvium. Along the I-79 corridor, bedrock strata dip westerly (out of the east valley wall) at ~2% (Fig. 19). Groundwater flow down these strata and down through stress relief fractured rock, rock slide masses, and colluvium contributed to both ancient and recent sliding of these materials, especially where groundwater discharge zones were blocked (covered or confined) by ice formations and/or clayey colluvium. X-ray diffraction analyses showed concentrations of expandable lattice clay minerals (vermiculite, smectite) along the failure surface clays of the I-79 slides. These secondary minerals were hypothesized to result from groundwater flow and weathering or alteration effects along the failure surfaces and to be at least partially responsible for the low measured residual shear strengths of the failure surface clays (Hamel, 1969; Hamel and Flint, 1969).
78
Gray et al.
Subsequent mineralogical and geochemical investigations of the claystone-derived shear zone at the base of a similar ancient rock slide located along present I-279 some 3 mi east of the I-79 slides showed an accumulation of randomly interstratified clay minerals (illite, chlorite, vermiculite, kaolinite) precipitated from colloidal solution as a result of groundwater interactions in overlying fissured claystone (Elnaggar and Flint, 1976). Chigira (1989) noted oxidation of pyrite in mudstones transforming chlorite to smectite as a potential cause of landslides in Japan. Anson
and Hawkins (1999, 2002) also noted implications of geochemical processes in residual strength reductions along landslide failure surfaces in England. This is a potentially fruitful area for further research (Hamel, 2004; Nakamura et al., 2010). Hamel (1969) and Hamel and Flint (1969) plotted residual friction angles from multiple reversal direct shear tests on I-79 failure surface clays versus clay size fraction in the manner of Skempton (1964). The I-79 data (with clay size fractions of 10%–29%) fell well below Skempton’s band. They attributed this
Figure 18. Stratigraphic section for Stops 3 and 4 (after Flint and Hamel, 1971).
Landslides in the vicinity of Pittsburgh, Pennsylvania
Figure 19. Structure contour map on top of Ames limestone (by N.K. Flint) for Stops 3 and 4 (from Hamel and Flint, 1969).
79
80
Gray et al.
to platy shaped aggregates of clay minerals (claystone and shale fragments of fine sand and silt size) becoming oriented parallel to the failure surface so that the measured residual friction angle was close to that of their mineral constituents. Hamel (2004) showed that residual friction angles measured and back-calculated for claystone and claystone-derived col-
luvium of Western Pennsylvania plotted versus plasticity index also fell below empirical strength correlation bands given by Mesri and Shahien (2003). He recommended that these strength correlation bands be used with caution, particularly for low to moderate plasticity materials like Western Pennsylvania claystone and claystone-derived colluvium.
Figure 20. Plan, Stop 3 (on portion of Allegheny County Topographic Map, 1992).
Landslides in the vicinity of Pittsburgh, Pennsylvania Reflection on these concepts over the past 40 years, along with reconsideration of failure surface clays from I-79 and elsewhere observed in the field as well as the laboratory, suggests that the low (relative to index properties) residual friction angles measured and back-calculated for these clays may also result in part from thin films of (generally pale gray and slippery) clay coating platy shaped sand and gravel size (and presumably smaller) rock fragments along the failure surfaces. This inference is supported
81
by the work of Elnaggar and Flint (1976) regarding clay mineral accumulation in a landslide shear zone and by the more recent work of Nakamura et al. (2010). STOP 3 Stop 3 includes a length of ~1500 ft along the east side of I-79 from approximate Sta. 973–988 (Fig. 20), Figure 21 is a
Figure 21. Stratigraphic column, Stop 3.
82
Gray et al.
stratigraphic column measured up the erosion gully at Sta. 976 and Figure 22 is a slope cross section at Sta. 974. This cross section was developed from pre-construction topography and the logs of borings drilled in 1963. For Stop 3, we leave the vehicle(s) at the Mount Nebo Road Park and Ride lot near Sta. 988 (Fig. 20). We walk south along the east side of the northbound exit ramp and the east side of I-79 to approximate Sta. 973. Along the way, we observe landslide features including colluvial soil masses, bent trees, and spring discharge zones in the slope above and discuss the stratigraphic column in Figure 21. At or near Sta. 974, we see slumped rock masses in the slope above and discuss the pre-construction cross section in Figure 22. Then, we walk north to the erosion gully at Sta. 976 (Fig. 20). Just south of this gully, we climb the slope on a zigzag path. Along the way, we observe rocks exposed in the gully (Fig. 21) and, south of the gully, a slumped mass of carbonaceous shale of the Birmingham Shale Unit (Figs. 21 and 22). On a bench at approximately elevation (El.) 1040, we see an apparent valley stress relief bedding plane shear zone at the base of the Morgantown Sandstone Unit (Fig. 21). We continue upslope to a relatively level area at or near the top of the Morgantown Sandstone at approximately El. 1100 (Fig. 21). From there, we follow a highway construction access road northerly and downslope to the northbound exit ramp at or near Sta. 985 (Fig. 20). Along the way, we see typical colluvial landslide features including hummocky topography, bent trees, spring discharge zones, and closed depressions with ponded water.
We then walk back to the Park and Ride Lot, board the vehicle(s), cross over I-79 to the southbound entrance ramp (Fig. 20), and drive ~1.5 mi south on I-79 to the next exit where we cross under I-79 to a northbound entrance ramp and Stop 4. STOP 4 Stop 4 includes a length of ~1000 ft along the east side of I-79 from approximate Sta. 900 to Sta. 910 (Fig. 23). This is the largest and most spectacular landslide area along this section of I-79 and it was the main one investigated by Hamel (1969) and Hamel and Flint (1969). They investigated two separate landslides which were reactivated by highway slope excavation: Slide A from approximate Sta. 906–910 and Slide B from approximate Sta. 899–904. Plans of Slides B and A are shown in figures 5-2 and 5-3, respectively, of Hamel and Flint (1969). Cross sections through Slide B at Sta. 899 + 00, 901 + 50, and 903 + 50 are shown in figures 5-6, 5-7, and 5-8, respectively, of Hamel and Flint (1969). A cross section through Slide A at Sta. 908 + 00 is shown in figure 5-9 of Hamel and Flint (1969). All of these figures show conditions as they existed in the spring of 1969. Later in 1969 and 1970, Slides A and B coalesced and the northern part of the coalesced landslide propagated several hundred feet easterly to form the huge landslide amphitheater extending from approximate Sta. 904-910 on Figure 23. In 1969, the Department of Highways acquired additional right-of-way to contain the enlarging landslide area. The magnitude of this landslide problem occurring during a large
Figure 22. Cross section, I-79, Sta. 974+00, Stop 3.
Landslides in the vicinity of Pittsburgh, Pennsylvania construction project, along with the low residual strengths of clay seams along basal failure surfaces (Hamel and Flint, 1969), indicated that the most economical solution here involved removal of a substantial portion of the landslide mass. A bench was excavated at or below the base of the landslide along the outer part of the slope to catch landslide debris and the landslide debris further
83
upslope was excavated to a nominal inclination of 5H:1V, producing the topography shown in Figure 23. Slope excavation was completed in 1970. Since then, movements have propagated further upslope, opening preexisting stress relief joints in Morgantown Sandstone. Some of these joints at the rear of the slide area were open to depths on the order
Figure 23. Plan, Stop 4 (on portion of Allegheny County Topographic Map, 1992).
84
Gray et al.
of 100 ft by 1981. Large sandstone blocks have been detaching from the valley wall and creeping and/or intermittently sliding downslope on underlying claystone. These movements pose no threat to the highway. For Stop 4, we leave the vehicle(s) on the road shoulder at approximate Sta. 908 and walk up the slope at approximate Sta. 910 (Fig. 23). We walk south along the excavated bench to approximate Sta. 905, then northerly along the partially excavated landslide mass to approximate Sta. 909. On this partially excavated surface, we see numerous sandstone stress relief joints opened by the continuing slide movements. We then walk south along the partially excavated surface to approximate Sta. 904 where we turn northeast and walk into the amphitheater at the rear of the landslide area (Fig. 23). There we see the slide scarp along stress relief joints in Morgantown Sandstone, a graben-like canyon where rock moved away from the scarp, and huge sandstone blocks separated by various fissures and openings. Some of the latter are several feet wide with depths of 100 ft or more. All of this block movement and separation at the rear of the landslide mass has occurred since 1969. From the rear of the amphitheater, we walk north along the inside of the right-of-way fence to approximate Sta. 910, then proceed westerly and downslope along the south side of the valley at approximate Sta. 911. Along the way we see various landslide features, e.g., rock debris and spring discharge areas. One of the springs has an abandoned Appalachian livestock watering trough—spring water formerly carried by a pipe to an overflowing cast iron bathtub. We return to the vehicle(s) on the road shoulder at approximate Sta. 908 and then leave for Pittsburgh. REFERENCES CITED Ackenheil, A.C., 1954, A soil mechanics and engineering geology analysis of landslides in the area of Pittsburgh, Pennsylvania [Ph.D. dissertation]: Pittsburgh, Pennsylvania, University of Pittsburgh. Ackenheil, A.C., 1958, Report on landslides, Fort Pitt Tunnel, in Stability Investigation, North Portal, Fort Pitt Tunnel: Michael Baker, Jr., Inc. Adams, W.R., Jr., 1986, Landsliding in Allegheny County, Pennsylvania—Characteristics, causes, and cures [Ph.D. dissertation]: University of Pittsburgh. Anson, R.W.W., and Hawkins, A.B., 1999, Analysis of a sample containing a shear surface from a recent landslip, south Cotswolds, UK: Géotechnique, v. 49, p. 33–41, doi:10.1680/geot.1999.49.1.33. Anson, R.W.W., and Hawkins, A.B., 2002, Movement of the Soper’s Wood Landslide on the Jurassic Fuller’s Earth, Bath, England: Bulletin of Engineering Geology and the Environment, no. 61, p. 325–345, doi:10.1007/ s10064-002-0151-8. Baker, F.F., and Chieruzzi, R., 1959, Regional concept of landslide occurrence: Highway Research Board Bulletin, v. 216, p. 1–16. Bjerrum, L., 1967, Progressive failure of slopes in overconsolidated plastic clay and clay shales: Journal of the Soil Mechanics and Foundations Division, v. 93, p. 1–49. Bonk, J.G., 1964, The weathering of Pittsburgh redbeds [M.S. Thesis]: Pittsburgh, Pennsylvania, University of Pittsburgh. Carson, M.A., and Kirkby, M.J., 1972, Hillslope Form and Process: London, Cambridge University Press, 475 p. Chigira, M., 1989, Chemical weathering of mudstone in mountainous areas: Abstract in Abstracts and Program, Association of Engineering Geologists Annual Meeting, Vail, Colorado, p. 56.
Dahl, H.D., and Parsons, R.C., 1971, Ground control studies in the Humphrey No. 7 mine, Christopher Coal Division, Consolidation Coal Company: American Institute of Mining and Metallurgical Engineering, Preprint. D’Appolonia, E., Alperstein, R., and D’Appolonia, D.J., 1967, Behavior of a colluvial slope: Journal of the Soil Mechanics and Foundations Division, v. 93, p. 447–473. Davies, W.E., 1974, Landslide Susceptibility Map of the New Kensington East 7-1/2′ Quadrangle, Allegheny County & Vicinity, PA: U.S. Geological Survey Open File Map 74-283, scale: 1:24,000. Deere, D.U., 1968, Geological considerations, in Stagg, K.G., and Zienkiewicz, O.C., eds., Rock Mechanics in Engineering Practice: New York, New York, Wiley, p. 1–20. Deere, D.U., and Patton, F.D., 1971, Slope stability in residual soils: Proceedings, 4th Panamanian Conference on Soil Mechanics and Foundation Engineering: American Society of Civil Engineers, New York, New York, v. 1, p. 87–170. Denny, C.S., 1956, Surficial geology and geomorphology of Potter County, Pennsylvania: U.S. Geological Survey Professional Paper 288, p. 72. Eckel, E.B., ed., 1958, Landslides and Engineering Practice: Highway Research Board, Special Report no. 29, 232 p. Elnaggar, H.A., and Flint, N.K., 1976, Analysis and Design of Highway Cuts in Rock: Report by University of Pittsburgh to Pennsylvania Department of Transportation, 62 p. Ferguson, H.F., 1967, Valley stress release in the Allegheny Plateau: Bulletin of the Association of Engineering Geologists, v. 4, p. 63–71. Ferguson, H.F., 1974, Geologic observations and geotechnical effects of valley stress relief in the Allegheny Plateau: Paper presented at American Society of Civil Engineers, Water Resource Engineering Meeting, Los Angeles, California, January 1974, p. 31. Ferguson, H.F., and Hamel, J.V., 1981, Valley stress relief in flat-lying sedimentary rocks, in Akai, K., et al., eds., Weak Rock: Rotterdam, Balkema, v. 2, p. 1235–1240. Flint, N.K., and Hamel, J.V., 1971, Engineering geology at two sites on Interstate 279 and Interstate 79 northwest of Pittsburgh, Pennsylvania, in Thompson, R.D., ed., Environmental Geology in the Pittsburgh Area: Geological Society of America, Annual Meeting., November 1971, Guidebook for Field Trip No. 6, p. 36–45. Gardner, G.D., 1980, An introduction to the geology of Pittsburgh and its impact on the activities of man, in 45th Annual Field Conference of Pennsylvania Geologists Guidebook, Pittsburgh, Pennsylvania. Gray, R.E., and Donovan, T.D., 1971, Discussion of slope stability in residual soils, by D.U. Deere and F.D. Patton: Proceedings, 4th Panamanian Conference on Soil Mechanics and Foundation Engineering: American Society of Civil Engineers, New York, New York, p. 127–130. Gray, R.E., Ferguson, H.F., and Hamel, J.V., 1979, Slope stability in the Appalachian Plateau of Pennsylvania and West Virginia, in Voight, B., ed., Developments in Geotechnical Engineering, vol. 14B, “Rockslides and Avalanches”: New York: American Elsevier Publishing Company. Hamel, J.V., 1969, Stability of slopes in soft, altered rocks [Ph.D. thesis]: University of Pittsburgh, Pittsburgh, Pennsylvania, 232 p. (No. 70-23, University Microfilms, Ann Arbor, Michigan). Hamel, J.V., 1980, Geology and slope stability in western Pennsylvania: Bulletin of the Association of Engineering Geologists, v. 17, p. 1–26. Hamel, J.V., 1998, Mechanism of Pleistocene rock slides near Pittsburgh, Pennsylvania: International Journal of Rock Mechanics and Mining Science, v. 35, n. 4-5, paper no. 32. Hamel, J.V., 2000, Mount Washington Slope—Duquesne Incline to Smithfield Street Bridge: Guidebook for the Annual Field Conference of Pennsylvania Geologists, v. 65, p. 98–105. Hamel, J.V., 2004, Discussion of residual shear strength mobilized in first-time slope failures: Journal of Geotechnical and Geoenvironmental Engineering, v. 130, p. 544–546, doi:10.1061/(ASCE)1090-0241(2004)130:5(544). Hamel, J.V., and Adams, W.R., Jr., 1981, Claystone slides, Interstate Route 79, Pittsburgh, Pennsylvania, USA, in Akai, K., et al., eds., Weak Rock: Rotterdam, Balkema, v.1, p. 549–553. Hamel, J.V., and Ferguson, H.F., 1999, Landsliding, Chapter 48, in Shultz, C.H., ed., The Geology of Pennsylvania, Harrisburg: Pennsylvania Geological Survey and Pittsburgh: Pittsburgh Geological Society, p. 704–711. Hamel, J.V., and Flint, N.K., 1969, A slope stability study on interstate routes 279 and 79 near Pittsburgh, Pennsylvania: Report by Departments of Civil Engineering and Earth and Planetary Sciences, University of Pittsburgh to
Landslides in the vicinity of Pittsburgh, Pennsylvania Pennsylvania Department of Highways and U.S. Department of Transportation, Bureau of Public Roads, 130 p. Hamel, J.V., and Flint, N.K., 1972, Failure of colluvial slope: Proceedings of the American Society of Civil Engineers: Journal of the Soil Mechanics and Foundations Division, v. 98, SM2, p. 167–180. Hamel, J.V., Lasko, J.D., and Ruppen, C.A., 1998, Rock slope evaluation for Pittsburgh Airport Busway in Moore, D.P., and Hungr, O., eds., Engineering Geology—A Global View from the Pacific Rim: Rotterdam, Balkema, v. 5, p. 3121–3128. Harper, J.S., 1990, Fossil collecting in the Pittsburgh area: Pittsburgh Geological Society, Field Trip Guidebook, 50 p. Harper, J.A., 1997, Of ice and waters flowing: The formation of Pittsburgh’s Three Rivers: Pennsylvania Geology, v. 28, no. 3/4, p. 2–8. Harper, J.A., 2002, Lake Monongahela: Anatomy of an immense ice age pond: Pennsylvania Geology, v. 32, no. 1, p. 2–12. Hooper, J.R., 1969. Slope movements of residual clays in Southeastern Ohio: Proceedings, Symposium on Landslides: Athens, Ohio, Ohio University, February 1969, p. 33–57. Jacobson, R.B., Elston, D.P., and Heaton, J.W., 1988, Stratigraphy and magnetic polarity of the high terrace remnants in the upper Ohio and Monongahela Rivers in West Virginia, Pennsylvania, and Ohio: Quaternary Research, v. 29, p. 216–232, doi:10.1016/0033-5894(88)90031-2. Ladd, G.E., 1927–1928, Landslides and their relation to highways. A report of observations made in West Virginia and Ohio to determine the cause of slides and devise means of control: Public Roads, Part 1, 8 (2), p. 21–35; Part 2, 9 (8), p. 153–163. Leverett, F., 1902, Glacial Formations and Drainage Features of the Erie and Ohio Basins: U.S. Geological Survey Monograph XLI, 781 p. Leverett, F., 1934, Glacial deposits outside the Wisconsin Terminal Moraine in Pennsylvania: Pennsylvania Geological Survey Bulletin G-7, 123 p. Lorant, S., 1975, Pittsburgh, The Story of an American City: R.R. Donnelley and Sons Company, 3rd ed., 608 p. Mesri, G., and Shahien, M., 2003, Residual strength mobilized in first-time slope failures: Journal of Geotechnical and Geoenvironmental Engineering, v. 129, p. 12–31, doi:10.1061/(ASCE)1090-0241(2003)129:1(12). Nakamura, S., Gibo, S., Egashira, K., and Kimura, S., 2010, Platy layer silicate minerals for controlling residual shear strength in landslide soils of different origins and geology: Geology, v. 38, p. 743–746, doi:10.1130/G30908.1. Nickelsen, R.P., and Hough, V.N.D., 1967, Jointing in the Appalachian Plateau of Pennslyvania: Geological Society of America Bulletin, v. 78, p. 609– 630, doi:10.1130/0016-7606(1967)78[609:JITAPO]2.0.CO;2.
85
Peltier, L.C., 1950, The geographic cycle in periglacial regions as it is related to climatic geomorphology: Annals of the Association of American Geographers. Association of American Geographers, v. 40, p. 214–236. Philbrick, S.S., 1953, Design of deep rock cuts in the Conemaugh Formation: State Road Commission of West Virginia, Proceedings, 4th Symposium on Geology as Applied to Highway Engineering, p. 79–88. Philbrick, S.S., 1959, Engineering geology of the Pittsburgh area: Geological Society of America, Pittsburgh Meeting, Guidebook for Field Trips, p. 191–203. (A copy of this guide is available as GSA Data Depository item 2011164 at www.geosociety.org/pubs/ft2011.htm or on request from
[email protected].) Philbrick, S.S., 1960, Cyclic sediments and engineering geology: Proceedings, 21st International Geological Congress, pt. 20, p. 49–63. Philbrick, S.S., 1961, Old landslides in the Upper Ohio Valley [abs.]: Geological Society of America, 1961 Annual Meeting, Cincinnati, Ohio, 2–4 November 1961, Program, Abstracts of Papers, p. 121A. Pomeroy, J.S., 1974, Landslide Susceptibility Map of the Pittsburgh West 7-1/2′ Quadrangle, Allegheny County, Pennsylvania: U.S. Geological Survey Open-File Map I-1035, scale: 1:24,000. Rapp, A., 1967, Pleistocene activity and Holocene stability of hillslopes, with examples from Scandinavia and Pennsylvania, in L’Evolution des Versants: International Symposium on Geomorphology, Liege, June 1966: Liege, University of Liege, p. 230–242. Rodgers, J., 1970, The Tectonics of the Appalachians: New York, New York, Wiley-Interscience, 271 p. Sharpe, C.F.S., and Dosch, E.F., 1942, Relation of soil-creep to earthflow in the Appalachian plateaus: Journal of Geomorphology, v. 5, p. 312–324. Skempton, A.W., 1964, Long-term stability of clay slopes: Géotechnique, v. 14, p. 77–101, doi:10.1680/geot.1964.14.2.77. Terzaghi, K., and Peck, R.B., 1948, Soil Mechanics in Engineering Practice: New York, New York, Wiley, 566 p. Voight, B., 1974, A mechanism for “locking-in” orogenic stress: American Journal of Science, v. 274, p. 662–665, doi:10.2475/ajs.274.6.662. Varnes, D.J., 1978, Slope movement types and processes, in Schuster, R.L., and Krizek, R.J., eds., Landslides Analysis and Control: Washington, D.C., Transportation Research Board, p. 11–33.
MANUSCRIPT ACCEPTED BY THE SOCIETY 2 FEBRUARY 2011
Printed in the USA
The Geological Society of America Field Guide 20 2011
Quaternary geology of northwestern Pennsylvania Gary M. Fleeger Pennsylvania Geological Survey, 3240 Schoolhouse Road, Middletown, Pennsylvania 17057-3534, USA Todd Grote Department of Geography and Geology, 205 Strong Hall, Eastern Michigan University, Ypsilanti, Michigan 48197, USA Eric Straffin Department of Geosciences, Edinboro University of Pennsylvania, Edinboro, Pennsylvania 16444, USA John P. Szabo Department of Geology & Environmental Science, University of Akron, Akron, Ohio 44325, USA
ABSTRACT Northwestern Pennsylvania was glaciated by the Grand River sublobe of the Erie Lobe. Glacial advances occurred at least three times during the pre-Illinoian (Slippery Rock, Mapledale, and Keefus), once during the Illinoian (Titusville), and four Late Wisconsinan (Kent, Lavery, Hiram, and Ashtabula) tills have been identified. While the area was studied for over 50 years by George White and associates, there are numerous details that remain unknown. The Titusville Till, which comprises the bulk of the glacial sediment, contains up to five separate sheets separated by sand and gravel. The origin of the numerous sheets is still not clear. The Kent glaciation resulted in extensive deposition in proglacial lakes and caused numerous local drainage diversions. Interpretations of the surficial geology around Conneaut Lake have changed a couple of times. Originally interpreted to be formed by a lobe of Hiram ice, it was later determined to be part of a widespread area of Lavery ice. Recent work supports the original Hiram interpretation. Since glaciation, streams in northwestern Pennsylvania have incised into the glacial sediments and have developed fine-grained floodplains within glacially scoured valleys. Lake sediments and alluvial stratigraphy suggests that general climate amelioration during the Holocene epoch was interrupted by episodes of landscape instability. Deforestation by European settlers is the most recent event appearing in the stratigraphic record and resulted in deposition of as much as 2 m of post-settlement alluvium.
Fleeger, G.M., Grote, T., Straffin, E., and Szabo, J.P., 2011, Quaternary geology of northwestern Pennsylvania, in Ruffolo, R.M., and Ciampaglio, C.N., eds., From the Shield to the Sea: Geological Field Trips from the 2011 Joint Meeting of the GSA Northeastern and North-Central Sections: Geological Society of America Field Guide 20, p. 87–109, doi:10.1130/2011.0020(05). For permission to copy, contact
[email protected]. ©2011 The Geological Society of America. All rights reserved.
87
88
Fleeger et al.
INTRODUCTION Glacial Lobes Northwestern Pennsylvania was glaciated by the Erie lobe of the Laurentide ice sheet and one of its sublobes, the Grand River lobe. The Grand River lobe advanced and spread out from the lowland of the Grand River valley in Ohio. The entire lobe advanced within the Appalachian Plateau, most of it in Ohio. The Kent, Lavery, and Hiram Till borders show the control of the Grand River lobe on its deposition. The Kent and especially the Defiance Moraines outline the lobe very well. The Titusville border does not show control by the Grand River lobe in Ohio, and the distribution of the Slippery Rock Till is inadequately known to determine lobe control. The Ashtabula and maybe the Keefus glaciers were restricted to the Erie lobe, not having advanced far enough to enter the Grand River valley. In northwestern Pennsylvania, the glacial margin of the eastern quarter of the Grand River lobe extends from the Ohio state line into Beaver County, and to the Venango–Crawford County border. From there, the glacial border turns more northeasterly, more closely paralleling the Lake Erie shore, and is considered as part of the Erie lobe. Distribution of Glacial Sediments The distribution of tills on the Appalachian Plateau (Fig. 1) is compressed compared to the till plains of the Midwest. The outcrop areas are narrower. In northwestern Pennsylvania and northeastern Ohio, the glacial border is within 120 km of Lake Erie, whereas in the till plains of central Ohio and west, the glacial border is over 320 km from the Great Lake shores. The end moraines on the Plateau are less prominent and closer together, because the topography has greater relief and is more bedrock controlled than in the till plains. Ground moraine areas generally contain thin till. Most valleys contain extensive kame terraces along the valley walls. They usually are a series of knolls on the lower valley wall and were deposited between an ice-remnant in the valley and the valley wall. The deposits are usually rather chaotic, and consist of coarse, poorly sorted sands and gravels, till, and well-sorted, sands, gravels, silts, and clays. Sediment was deposited by streams flowing between the ice and valley wall, slumped from the ice surface, and/or in lakes formed in depressions. Many of the valleys parallel to the general direction of ice flow contain outwash. Some valleys are completely buried. Most are partially buried with maximum depths to bedrock exceeding 100 m. Lithostratigraphy The tills of northwestern Pennsylvania have been assigned formal lithostratigraphic names of formational rank (Shepps et
al., 1959; White et al., 1969; Fig. 2). No non-till units have been given such designations. Most tills, except maybe for the Titusville Till, are usually thin, generally not more than 6 m thick. The median thickness of individual Wisconsinan tills in northeastern Ohio is 1.5 m (White, 1982). The uppermost till in most places is so thin that it is often weathered completely through into the next lower till, creating a single weathering sequence through both tills. Identification of tills also can sometimes be difficult because of local facies changes, and sufficient age control of the various units is lacking. Until, 1959, all of the tills in the Grand River lobe were designated by their age (Illinoian till, early Cary till, etc.). Shepps et al. (1959) were the first to apply formal lithostratigraphic nomenclature. Slippery Rock Till. The oldest known till is the Slippery Rock Till, named from the borough of Slippery Rock in northern Butler County, Pennsylvania. No unoxidized or unleached Slippery Rock Till has ever been found. The Slippery Rock Till has not been found at the surface, but is everywhere buried beneath younger glacial deposits. At places in northwestern Pennsylvania, erratic boulders have been found beyond the mapped limit of tills. These boulders may or may not be related to the Slippery Rock Till. Mapledale Till. The Mapledale Till is named for the village of Mapledale, near Franklin, in Venango County, Pennsylvania. It can be identified in the field by its coarse matrix, pebbly and cobbly nature, predominance of sandstone clasts, yellow-brown oxidized color, and carbonate content (White, 1968b). The latter, especially, serves to distinguish unleached samples from the Titusville and Kent Tills, because the carbonate content of only the Mapledale Till is usually too low to react visibly with hydrochloric acid. The Mapledale Till (originally called “Outer Illinoian” by Shepps et al., 1959) is the surface till in a fringe beyond the Kent margin in some places (Fig. 1). Keefus Till. The Keefus Till is named for Keefus Road in Conneaut Township, Ashtabula County, Ohio (White, 1982). It is a very hard, compact, coarse-grained, dusky red till. It is higher in matrix carbonate than any other pre-Wisconsin episode till in the Grand River lobe, averaging ~9% at its type section (Bruno, 1988). The red color and high carbonate content, which make this till very distinctive, come from the Queenston and Grimsby shales in the Lake Ontario basin (Szabo and Totten, 1995). The Keefus has only been found in water wells and outcrops within 32 km of Lake Erie and does not extend nearly as far onto the Plateau as other Grand River lobe tills (Fig. 2). It is usually found in the subsurface and has not been identified in situ beneath the Titusville Till anywhere in Pennsylvania. Before the discovery of the Keefus, such a till was predicted, because masses of a pink, high-carbonate till were sometimes found within Titusville Till (White et al., 1969). It was usually described as a purple-pink or maroon till, and as being exceedingly calcareous (G.W. White field notes). However, one section illustrated in White et al. (1969; their fig. 17) shows 0.6 m of a
Quaternary geology of northwestern Pennsylvania
Figure 1. Distribution of the tills of northwestern Pennsylvania. Note the lobe of Hiram Till beyond the Defiance Moraine at Pymatuning Reservoir (Stop 4) and around Conneaut Lake. Modified from White et al. (1969) to add the Hiram lobe around Conneaut Lake (Stop 6). See Stop 6 for discussion. The “fringe” is the outcrop area of the Titusville and Mapledale Tills. Map names are 15′ quadrangles.
89
90
Fleeger et al.
Keefus-like, extremely calcareous, purple-pink till appearing to lie in place on top of the Titusville Till. Titusville Till. The Titusville Till is named for Titusville, Crawford County. Droste and Tharin (1958) first described the Titusville type section in one of their pioneering clay mineralogy studies of tills. It is a very hard sandy, cobbly till. It is olive-gray and oxidizes to olive-brown. The color, texture, and hard compact nature of the till aid in its field identification. Titusville sediments make up the bulk of the drift in much of northwestern Pennsylvania (White et al., 1969). The thickness sometimes exceeds 30 m. White et al. (1969) believe that the bulk of the Kent Moraine is composed of pre-Kent sediments, mostly Titusville, and that the Kent is simply draped over the moraine, making it a palimpsest moraine. Straffin and Grote (2010) found no evidence to support a thin Kent Till in the Sugar Lake area in eastern Crawford and northwestern Venango County. White et al. (1969) have also determined that the Titusville Till is often divided into up to five separate till sheets, separated by sand and gravel layers of varying thickness. They speculate that the Kent Moraine is composed of these multiple till sheets. They interpreted the separate sheets as resulting from minor retreats and readvances of a fluctuating ice margin and that each
Figure 2. Time-distance diagram showing the stratigraphic classification of the glacial deposits of northwestern Pennsylvania. The Keefus Till is shown, even though it has never been seen in place in Pennsylvania. Modified from White et al. (1969).
readvance extended less far than the previous one. Stop 3 will look at a rare exposure of what White et al (1969) would have interpreted as all five till sheets (see Fig. 9), well behind the Kent Moraine, and well back from the Titusville margin. The Titusville Till (called “Inner Illinoian” in Shepps et al., 1959) is the surface material in part of the fringe beyond the Kent Margin (Fig. 3). Totten and Szabo (1987) thermoluminescence dated the loess overlying the Titusville-correlative Millbrook Till in Ohio at ~140,000 years. Kent Till. The Kent Till is named for Kent, Ohio. Shepps et al. (1959) were the first to apply the name Kent to these deposits. Prior to that, it was known as “early Cary till” (Shepps, 1955). The Kent Till is a friable (relative to the Titusville Till), sandy, pebbly gray till that oxidizes to a yellow-brown. It can be distinguished from the Titusville Till by its color and friable nature, and from the younger tills by its sandy, pebbly matrix. Younger tills have a much finer-grained matrix and are more sparingly pebbly. White et al. (1969) determined that the Kent Till is generally very thin. In most places, weathering has extended completely through the Kent Till into the underlying material (older drift or bedrock). Because it is so thin, unoxidized Kent Till is rare. Although the Kent Till is interpreted to be very thin, White et al. (1969) correlated the extensive kame terrace deposits throughout the valleys in northwestern Pennsylvania with the Kent glaciation. The Kent Till had been related to the Kent Moraine for many years (White et al., 1969). However, the Kent Till has been mapped in part of the fringe beyond the Kent Moraine, so in many places, the Kent Moraine does not mark the boundary of the Kent Till. Based on the radiocarbon dating of wood in what was interpreted to be pro-Kent lacustrine sediments near Cleveland, Ohio, the Kent Till is thought to be ~23,000 radiocarbon years old (White, 1968a). Lavery Till. The Lavery Till was named by Shepps (Shepps et al., 1959) for Lavery, near Edinboro, in Erie County. Prior to the formal lithostratigraphic name, the Lavery Till was referred to as the “middle Cary till” (Shepps, 1955). The Lavery Till is a light gray, compact, silty, pebbly till that oxidizes to a yellow brown color. It has a few cobbles and boulders. It can easily be distinguished in the field from older tills by its fine-grained matrix and fewer pebbles and cobbles. It can be difficult to distinguish from the younger Hiram Till in the field. The Lavery Moraine may be a palimpsest moraine and does not mark the limit of the Lavery Till, as shown on the map by Shepps et al. (1959). White et al. (1969) extended the Lavery limit to cover the area shown on Figure 1. This map is in error, however, because White’s field maps show additional outcrops of Lavery Till farther south, and the glacial map of northeastern Ohio (White, 1982) shows the Lavery border more closely matching the outcrops on White’s Pennsylvania field maps. The Lavery border actually extends, at the state line, to southern Lawrence County, rather than southern Mercer County, as shown on Figure 1. Recent work by Weinreich (2006) and Hartley (2009) suggest that the Lavery may not extend beyond the Lavery Moraine, so it is uncertain in Pennsylvania whether it is a palimpsest moraine.
Quaternary geology of northwestern Pennsylvania Weinreich (2006) found a more calcareous phase of the Lavery Till in its type area that probably represents superglacial meltout of the englacial ice. This phase is more typical of Ohio. A radiocarbon date of marl preserved below peat in a bog at Corry, Erie County (Droste et al., 1960), is within the extended Lavery border, and provides a minimum age of 14,000 radiocarbon years for the Lavery Till. White (1982) indicates that it might be ~19,000 years old. Hiram Till. The Hiram Till is named for Hiram, Ohio (Shepps et al., 1959). It was previously referred to as the “late Cary till” (Shepps, 1955). The Hiram Till appears very similar to the Lavery Till, and distinguishing them in the field can be difficult. White (1982) reports that the Hiram is the most clayrich till in northeastern Ohio. It is a bluish-gray, clay to siltyclay, sparingly pebbly till (White et al., 1969). Its oxidized color is described as drab brown (Shepps, 1955). Pebbles can be rare enough that it sometimes appears to be a lacustrine deposit (White, 1982). The oxidized color difference between the Hiram and Lavery is subtle (White, 1982). The limit of the Hiram advance is generally marked by the Defiance Moraine. The Defiance Moraine, named for Defiance, in northwestern Ohio, is mapped (Shepps et al, 1959) as emerging from beneath the younger Ashtabula Morainic System on the lake plain south of Erie, and rising 70 m up onto the plateau to the southwest (Figs. 1 and 19). The Defiance Moraine extends westward across Ohio, extending southward into a number of other sublobes in central and western Ohio, wraps around the end of Lake Erie, and trends north to its terminus near Pontiac, Michigan. The Defiance Moraine reappears from
91
beneath the Ashtabula Morainic System at the New York State line (Shepps et al., 1959). In the Grand River lobe, the Defiance Moraine, like other moraines, may be a palimpsest moraine created by earlier glaciations. Hiram Till was mapped by Shepps et al. (1959) with a couple of lobes of Hiram Till extending beyond the Defiance Moraine at Conneaut Lake and the western arm of Pymatuning Reservoir. Later, White et al. (1969) showed two maps, one of which showed a Hiram lobe only at Conneaut Lake, and another that showed a Hiram lobe only at Pymatuning Reservoir. White (1982) estimates the age of the Hiram Till to be ~17,000 years. The only date associated with it is of wood preserved in peat in a kettle hole in Medina County, Ohio (Totten, 1976). Its date of 14,050 radiocarbon years provides only a minimum date for the withdrawal of the Hiram glacier, when peat could begin to accumulate in the kettle. Ashtabula Till. The youngest till in northwestern Pennsylvania is named for its type area near Ashtabula, Ohio. Its type section is a road cut three miles east-southeast of Ashtabula (White, 1960). It was previously referred to as the “latest Cary” till (Shepps, 1955). Ashtabula Till can also be difficult to distinguish from the Hiram and Lavery Tills, but it appears to be somewhat sandier than the older tills, and has more pebbles. It also usually has a greater depth of leaching than the two older tills (White, 1982). It is a pebbly, bluish-gray, silt till. Its oxidized color is also similar to the Hiram and Lavery Tills. It crops out in extensive bluffs along the shore of Lake Erie, where it is usually overlain by lacustrine sands and silts deposited in early, higher levels of Lake Erie.
Figure 3. Slopeshade image of the Jacksville Esker–Delta complex based on the lidar digital elevation model data. The light areas are flat, and the dark areas are steep. The flat top and steep sides of the esker, and the flatness of the top of the delta and lake plain are evident in this image. The seven substops are located on this image. Modern Hogue Run in both the preglacial Black Run and Hogue Run valleys is labeled. Direction of modern drainage: Direction of ice flow:
92
Fleeger et al.
The Ashtabula Till is present at the surface only within the Eastern Lake Section. The distribution of the Ashtabula Till indicates that it is an Erie Lobe deposit and does not extend into the Grand River lowland. The Ashtabula advance was stopped by the plateau escarpment. A series of moraines, known as the Lake Escarpment Morainic System (Leverett, 1902) or Ashtabula Morainic System (Shepps et al., 1959), was deposited against the escarpment. The morainic system is composed of a series of individually named moraines (Leverett, 1902). Farther east, in New York, the Defiance and Lavery Moraines merge with the Ashtabula moraines and become part of the Lake Escarpment Morainic System. We are not aware of any work having been done to evaluate whether the moraines are palimpsest moraines or were formed by the Ashtabula advance. There are no radiocarbon dates associated with the Ashtabula Till. White (1982) estimates its age at ~15,000 years. The Fringe Beyond the Kent Moraine is a “fringe” of glacial sediments. The fringe was first recognized (I.C. White, 1880) and named (Lewis, 1884) by geologists from the Second Geological Survey of Pennsylvania. The age of the fringe has been debated for years. I.C. White (1880) believed it to be the same age as the rest of the deposits. Leverett (1934) believed that the fringe to the south of Slippery Rock was Wisconsinan in age based on weathering characteristics (depth of oxidation and pebble freshness), as did G.F. Wright (1892). Others, including G.W. White (1942), believed that the fringe was Illinoian, based on the weathered, thin, discontinuous nature of the deposits that contrasted with the thick, continuous character of the deposits within and behind the Kent Moraine. Ireland (1940) mapped the fringe in Ohio as Wisconsinan. Wilber Stout in Ohio considered it to be pre-Illinoian in age based on the topography and weathering characteristics (Preston, 1949). More recent mapping (Shepps et al., 1959) showed the fringe to be mostly Illinoian and older in age. D’Urso (2000) remapped the margins in the uncertain area of the Slippery Rock Creek basin, using the thickness of weathering rinds on granitic clasts to determine age. He concluded that the fringe south of Slippery Rock was Wisconsinan, as Leverett (1934) had mapped. During Shepps mapping, the model was that the bulk of the drift in northwestern Pennsylvania was Kent Till, and that the older tills were thin and discontinuous. When White et al. (1969) restudied the area, with the benefit of new interstate highway construction and more extensive strip mining, they determined that the bulk of the drift was Titusville. The Kent Till was apparently discontinuous, and thin enough that in most places where it is the surface material, that the weathered horizon extended completely through the Kent Till and into the underlying material. This change in model may have bearing on the interpretation of the fringe. An important characteristic of the drift in the fringe is that it is discontinuous, thin, and weathered completely. With the Shepps model, that is the description of the Titusville and Mapledale Tills where they are at the surface. The Kent Till
matches that description with the later White model. White et al. (1969) never considered the effect of changing the model on the interpretation of the fringe. Their model of Kent Till weathered through, and into the underlying material may also be valid in the fringe. The only difference is that in the fringe, the underlying material is bedrock, rather than older drift. If the fringe is determined to be composed of Kent drift, then the glacial borders need to be revised. At Stop 2, we’ll see how this affects the determination of the age of a drainage diversion. STOP 1. JACKSVILLE ESKER, DELTA, LAKE PLAIN, AND DRAINAGE DIVERSION COMPLEX The Jacksville Esker complex (Fig. 3) is a remarkably complete complex of related glacial landforms. The 15-km-long Jacksville Esker terminates at a large (1.6 km × 0.8 km × 12 m high) kame delta that prograded into a lake created by the glacier through which the subglacial stream creating the esker flowed. A flat lacustrine plain extends beyond the distal slope of the delta. The delta extends completely across the Black Run valley. Lake drainage channels are eroded through the east side of the Black Run valley wall, and through the delta sediments where they impinge upon the valley walls. The Jacksville Esker The Jacksville Esker (Stop 1A) is probably the best-preserved esker in Pennsylvania. It was deposited during the Kent glaciation (Shepps et al., 1959) ~23,000 radiocarbon years ago (White et al., 1969). It occurs in multiple segments over a distance of 15 km between Schollard, Lawrence County (Durco, 2008), and West Liberty, Butler County, separated by post-glacial erosion by Slippery Rock Creek and various smaller streams. There are several short eskers that appear to form a tributary pattern with the main esker. The 5-km-long southeastern segment ends at a kame delta that was deposited in a proglacial lake. Over its entire length, the esker passes over topographic highs and lows, but the southeastern segment increases in elevation toward the glacial border and follows the axis of buried pre-glacial valleys of Black Run and Slippery Rock Creek (Poth, 1963). There is also no evidence of deformation of the entire esker (Geyer and Bolles, 1979) that would indicate letdown from an englacial or supraglacial position, suggesting that the tunnel in which it formed was a full-flowing subglacial tunnel. Higher mounds on the esker at fairly regular intervals indicate that there was continued flow through the tunnel and deposition at its mouth during the retreat of the glacier (Fleeger, 1986). The Kame Delta Geomorphology The kame delta is a very flat-topped feature at an elevation of ~1295 ft (395 m; Stop 1C) that was built into a lake dammed by the Kent glacier (Preston, 1977). The topography of the proximal,
Quaternary geology of northwestern Pennsylvania ice-contact side of the delta, the north side, is hummocky, probably due to collapse after the supporting glacier retreated (Stop 1B). The distal delta-front slope is smooth, merging with the lacustrine plain (Stop 1D). The Black Run valley splits upstream into two tributary valleys, Black Run E and Black Run W. The center of the delta extends to the divide between these two tributary valleys (Stop 1E). Internal Delta Composition—Glacial Sand and Gravel Co. Pit (Stop 1G) A sand and gravel pit, recently opened by the appropriately named Glacial Sand and Gravel Company, has exposed the internal composition and structure of part of the delta. Their exploratory borings indicate that the sediment is generally at least 12 m thick and ranges up to 37 m thick in the vicinity of the current pit, varying with the bedrock topography. The sediment within the delta ranges from fine sand to cobbles. Much of the sediment is well sorted, but the coarsest sediment, in the topset beds, is more poorly sorted, ranging from coarse sand to cobbles. Most of the sedimentary structures are cross bedding and ripples in the sand beds. Contacts between beds are usually sharp, sometimes channeled, and frequently between beds of very different grain size, indicating fluctuating flow. Frequently, there are large pebbles within the sand foresets, with no indication of any deformation of the laminae. There are few faults in the sediments. The most prominent feature is the foresets, up to 5 m high (Fig. 4). Two sets of foresets/topsets have been exposed vertically as of this writing. The foresets are mostly medium sand, but there are some fine sand or silt beds within them. The dip of the foresets varies. At least one of the foresets is overlain by a thin, dirty sand and gravel topset. These foresets were probably deposited on the frontal slope of what was at the time of deposition, the outer margin of the delta. The foresets that are/were exposed here indicate transport from the west or northwest, which suggests
93
that the source of sediment was not from the esker tunnel, which joins the delta to the north-northeast of the pit. There were likely multiple sediment input locations to the delta. The multiple foreset/topset beds and flow direction indicators suggest that the delta is not a single, simple, large delta, but is a composite feature, built up over a period of time from multiple episodes of deposition, probably though multiple small delta lobes. Lacustrine Plain The lacustrine plain is a flat lowland extending beyond the delta in Black Run E and W. The only exposure was a temporary pond excavation just beyond the delta front in the Black Run W valley. The sediments in that excavation were gray and black clay, silt, and some till-like material. The latter was seen in the excavated sediments, and not in place. Exploratory borings for the sand and gravel pit indicate a clay thickness beneath the sand, silt, and gravel of 0–6 m. The thickness here on the lacustrine plain is unknown. Black Run–Hogue Run Drainage Diversions Because the delta impinged upon the bedrock hills surrounding the valley on three sides, small lakes remained in the upper Black Run valley after the melting of the glacier dam. One diversion that drained much of the remaining lake eroded the valley wall on the east side of the valley. It forms a small bedrock gorge through the divide into the adjacent Hogue Run valley (Fig. 3). Two other smaller diversion channels were eroded between bedrock and the delta where it impinged upon the bedrock. One is on the south side of the delta. The Black Run E and W were isolated where the delta encroached upon the hills dividing them (Stop 1E; Fig. 3). Drainage from the Black Run W lake into the Black Run E valley eroded this channel. The other diversion channel was eroded where a lake in a small tributary to Black Run W was isolated by delta encroachment upon the surrounding hills (Stop 1F; Fig. 3). Once the lake in Black Run W began to drain into Black Run E at Stop 1E, this channel was eroded by the draining of the small lake into Black Run W. STOP 2. SLIPPERY ROCK GORGE
Figure 4. Delta foresets overlain by the darker, thin topset bed.
The Slippery Rock Gorge is ~19 km long, extending from Kennedy Mill to Wurtemburg, where it enters Connoquenessing Creek (Fig. 5). The maximum depth of the gorge is 120 m at Cleland Rock. Slippery Rock Creek at Kennedy Mill abruptly shifts its course southward into the gorge from its broad southwest-northeast valley. The southern end of the gorge is not so abrupt. From Cleland Rock to Wurtemburg, the valley gradually becomes less deep, but remains narrow to its confluence with Connoquenessing Creek. The gorge is a Pleistocene drainage diversion. There is a suggestion (D’Urso, 2000) that the formation of the gorge did
94
Fleeger et al.
Figure 5. Lidar digital elevation model of the Slippery Rock Gorge area. Dashed line indicates the pre-glacial divide at Cleland Rock. The current course of Slippery Rock Creek marks the course of McConnells Run flowing north past McConnells Mill, and Wurtemburg Run flowing south past Wurtemburg from the divide. The course of the preglacial streams can be discerned on the image.
not require glacial diversion, but, presumably, was a consequence of pre-glacial stream piracy. D’Urso’s basis for this is that he interpreted his data to suggest that discharge through this gorge was less during the Pleistocene lake drainage than numerous stream discharges measured during historic times, eliminating the possibility of a catastrophic glacial lake outburst, as have been previous (Preston, 1977) and subsequent (Fleeger et al, 2003) interpretations. Progressive Nature of Drainage Reversal The almost total reversal of the drainage in northwestern Pennsylvania was not caused by a single ice advance. Progressive changes probably occurred with each advance, with changes
occurring during later advances being minor relative to those resulting from early advances. One or more early advances (Jacobson, et al., 1988) formed the Allegheny-Ohio River by unifying four or more northwest flowing preglacial rivers (Upper Allegheny, Middle Allegheny, Pittsburgh, Teays). This reversed the overall drainage for much of the area from the St. Lawrence River to the Mississippi River. Later advances caused more minor changes. They, in effect, caused the divide between the St. Lawrence and Ohio Rivers to migrate further north to its present location in Erie and Crawford Counties. Migration of a divide probably frequently took place by headward erosion of a stream draining a proglacial lake. Drainage causes the col to be eroded downward and lakeward until the
Quaternary geology of northwestern Pennsylvania lake is completely drained, either because headward erosion has progressed far enough to cause the drainage to be away from, instead of toward, the ice front, or because the glacier dam has been removed by retreat. Slippery Rock Gorge Frank Preston (1977) spent many years studying the glacial lakes and drainage history of this area. Prior to glaciation, the Slippery Rock Gorge did not exist, but two smaller streams, McConnells Run to the north to Slippery Rock Creek, and Wurtemburg Run to the south to Connoquenessing Creek, flowed away from a divide at Cleland Rock (Fig. 5). Preston interpreted two large glacier-dammed lakes during the maximum advance of the Kent glaciation—Lake Edmund in the Slippery Rock–Wolf Creek basin, and Lake Watts in the Muddy Creek basin (Fig. 6). Modern Lake Arthur in Moraine State Park is a smaller recreation of Lake Watts. At the maximum glacial advance, Lake Edmund overflow drained into Lake Watts at a col near West Sunbury, and Lake Watts overflow drained through a col at Queen Junction and into the Connoquenessing Creek and Beaver River (Fig. 6.) At the maximum advance of the Kent glacier, Preston also identified a small lake created when the ice advanced far enough to isolate and dam McConnells Run, which he named Lake Prouty (Figs. 6 and 7). The overflow from Lake Prouty was through a col at Cleland Rock and into Wurtemburg Run. Preston interpreted that this overflow created the incipient Slippery Rock Gorge. Part
95
of the erosion of the Cleland Rock col may have been completed by Lake Watts drainage before the isolation of Lake Prouty. When the glacier retreated, it opened a series of lower cols, first draining Lake Watts and later Lake Edmund, all through the Slippery Rock Gorge. Each col opening partially drained one of the lakes. Here is where D’Urso (2000) disagrees with Preston’s interpretations. Preston interpreted each opening, which he named Alpha, Beta, Gamma (Lake Watts), Delta, and Epsilon (Lake Edmund) Passes, to have resulted in multiple, shortterm, catastrophic drainages of the lake to the levels of those cols. Because D’Urso found no clasts in the deposits downstream that would require a flow greater than 380 m3/s, he interpreted the lake drainages to not be catastrophic. The maximum historical flow on Slippery Rock Creek at the U.S. Geological Survey gauge at Wurtemburg is 538 m3/s in January 1937. During a field trip in October 2004 (D’Urso et al., 2004), we found a boulder larger than any identified by D’Urso (2000) that had apparently been moved during a 334 m3/s discharge on 9 September 2004 during the remnants of Hurricane Frances. Preston’s interpretation states that the lake drainage through the various passes eroded the incipient Slippery Rock Gorge to its present depth and eroded the divide headward past McConnells Mill. That and the filling of the old Slippery Rock channel permanently diverted the stream through the gorge. The rapid erosion left a number of hanging valleys, such as at Alpha Falls. Dating the Formation of the Slippery Rock Gorge Dating the time of eroding the gorge has proven to be difficult. Popular publications describing the origin of the gorge state
Figure 6. Distribution of pro-glacial, ice-dammed lakes in northern Butler and Lawrence Counties. Modified from Preston (1977).
Figure 7. Detail of glacier and Lake Prouty at Cleland Rock col. Modified from Preston (1977).
96
Fleeger et al.
that it was formed either 10,000 or 20,000 years ago (e.g., Bushnell, 1975; Palmer, 1980). Several attempts were made to obtain wood samples from deltaic deposits and lacustrine sediments of glacial Lakes Prouty and Watts, for radiocarbon dating, but none were successful (Preston, 1977). Preston determined that the reversal occurred as a result of the last ice advance to reach the area (Woodfordian Kent glaciation; 23,000 yr B.P.; White et al., 1969), based on the youthful appearance of the Slippery Rock Gorge (F.W. Preston, 1978, personal commun.). There are several lines of evidence that suggest that the diversion is associated with the glacial border mapped as “Inner Illinoian” (Titusville) by Sitler (1957), part of the “fringe” beyond the Kent Moraine. However, more recent work (D’Urso, 2000) casts some doubt about whether that border is the Titusville or Kent border. The change from a thick Kent, thin, weathered Titusville model (Shepps, et al., 1959) to a thick Titusville, thin, weathered Kent model (White, et al., 1969) supports D’Urso’s interpretation that the border could be the Kent. It is likely that the erosion of the gorge occurred during several glacial advances and associated proglacial lake system drainages. Cleland Rock Cleland Rock is a ledge of Kittanning Sandstone which forms a scenic overlook 120 m above the deepest part of the 19-km-long Slippery Rock Gorge. It is also the highest part of the rim of the gorge, along the pre-diversion divide between the
south-flowing Wurtemburg Run and the north-flowing McConnells Run (Figs. 5 and 7). Lake Prouty lay in the valley to the right (north) of this divide. Overflow from Lake Prouty, assisted by meltwater and sediment released from the nearby glacier, spilled over the divide at Cleland Rock from right to left (north to south) into Wurtemburg Run. This overflow eventually eroded a channel through the divide at Cleland Rock deep enough to join McConnells Run and Wurtemburg Run to form the ancestral south-flowing Slippery Rock Creek, and drain Lake Prouty. Preston (1977) believed Lake Prouty to be the key to the creation of the gorge. While D’Urso (2000) disagrees with Preston’s interpretation, evidence in the Cheeseman Run valley supports it (Figs. 5 and 8). East of Breakneck Bridge, Cheeseman Run forms a gorge through sand and gravel deposited in a kame delta in Lake Prouty and into the underlying bedrock (Fig. 8). Cheeseman Run flows northwest from Portersville. When it reaches the border of the Titusville Till, as mapped by Sitler (1957), it turns to the southwest into the gorge through the kame delta, and forms an arc, concave to the north, to its mouth at Slippery Rock Creek. It enters the Slippery Rock Gorge, 45 m above Slippery Rock Creek, from a hanging valley, flowing over Breakneck Falls. Cheeseman has three characteristics that are different from other similar sized streams flowing into the Slippery Rock Gorge. First, most streams have a series of small waterfalls
Figure 8. Hillshade image of the Cheeseman Run area derived from lidar digital elevation model data. Proglacial Lake Prouty was in the Cheeseman Run valley. Kame deltas were deposited in Lake Prouty.
Quaternary geology of northwestern Pennsylvania where they cross resistant beds and many boulders in the bottom of the valleys. Cheeseman Run gorge does not have any waterfalls until it reaches the top of the Homewood Sandstone several hundred meters above Breakneck Falls. Resistant beds do exist along the course of Cheeseman Run because a tributary from the south (did not drain ice front) does have several small waterfalls up to 1.5 m high at these resistant beds. The bottom of the Cheeseman Run gorge is also free of large boulders and choked with a cobble bedload. Secondly, while most streams essentially fill the valley bottom, lower Cheeseman Run flows in a valley bottom about five times the stream width. Finally, other streams flowing into the Slippery Rock Gorge enter the gorge over waterfalls at the top of the Homewood Sandstone (e.g., Gardiner Falls, Kildoo Falls). Cheeseman Run has eroded 10+ m of the Homewood Sandstone before entering the Slippery Rock Gorge at Breakneck Falls at an elevation of 1070 ft (326 m). These characteristics suggest that Cheeseman Run at one time carried a considerably greater flow than the other streams or carried a large flow for a longer period of time. Sand and gravel foresets in a kame delta north of Cheeseman Run (Fig. 8) are overlain by a topset bed, with cobbles up to 0.3 m in diameter. Long abandoned gravel pits in the part of the kame delta south of Cheeseman Run (Fig. 8) appear to be sand and gravel that is finer grained than in the northern pit, and is the material at the surface of the kame delta. This, and the characteristics of Cheeseman Run gorge suggest that the glacier did not reach this position, but stopped somewhere between the pit north of Cheeseman Run and the pits south of Cheeseman Run. Cheeseman Run gorge may well have been eroded as an icemarginal channel. Sitler (1957) mapped the border of the Titusville Till between the northern and southern gravel pits. The Wurtemburg–McConnells Run divide was not removed prior to glaciation because the kame delta at elevation 1260 ft (384 m) indicates that the divide was at least that high during the advance. After Lake Prouty drained, the new Slippery Rock Gorge was probably no more than half of its present depth (currently elevation 920 ft [280 m]). From this beginning, it has been eroded to greater than its present depth by the draining of glacial Lakes Watts and Edmund, further upstream. STOP 3. BOOTH RUN SECTION At Stop 3, we will see a thick section containing multiple tills. White et al. (1969) and White (1982) described the Titusville Till in the subsurface as being composed of up to five separate till sheets, separated by sand and gravel layers. Here we have five tills beneath the Kent, separated by four sand beds. White et al. (1969) suggested that multiple Titusville sheets “stacked up” to form the bulk of the Kent Moraine. However, we are many miles behind the Kent Moraine here. The section also illustrates complex weathering patterns associated with jointing in the till and the effects of the sand beds on the adjacent till.
97
Stratigraphy Up to 7 tills have been identified here based on field evidence and laboratory data (Fig. 9). Whether each of the 7 tills represents a separate glacial advance is open to speculation. The till immediately above the stone concentration at 6 m is most likely Kent Till. Below the stone line, there are multiple possible interpretations of the stratigraphy, based on our current knowledge of the glacial stratigraphy of northwestern Pennsylvania and northeastern Ohio. The mineralogy, showing the changes resulting from weathering, suggests that there may be fewer than five tills, and that the sands do not all represent breaks between glacial events. Using till mineralogy and weathering evidence, there are probably only three tills beneath the Kent Till (Fig. 9). However, that also has problems in that weathering can extend through one till into underlying tills, complicating the interpretation of the stratigraphy, especially if the tills are thin, as they are here. White would probably have interpreted them as multiple Titusville Till sheets. If they are not multiple Titusville sheets, then they must be older tills. The older (Illinoian and older) tills identified in the Grand River Lobe are the Keefus, Mapledale, and Slippery Rock (White et al., 1969; White, 1982). The Keefus Till has been identified, mostly in water well logs, only within 32 km of Lake Erie (White, 1982). It has a distinctive reddish color and is high in matrix carbonate content (9.1% at its type section; Bruno, 1988), relative to the other tills in the lobe. Its existence was predicted earlier by White et al. (1969) because it was found as inclusions in the Titusville Till. White (1982) reports that the Titusville Till averages ~3% carbonate. The till of samples BR-7 to BR-11 has the highest matrix carbonate content in the section, including 2 samples (BR-8 and BR-9) with carbonate contents greater than 6% (Fig. 9), and there is no known nearby source of carbonate material. It is in the stratigraphic position of the Keefus Till at its type section (White, 1982). Maybe the Keefus Till is just one of White’s Titusville sheets with local red coloration near Lake Erie. The red color is thought to be from eroded Grimsby Shale in the Niagara region. Flowing over the escarpment may have caused increased erosion of local bedrock, resulting in compositional and color changes. Szabo (1987) discussed compositional changes resulting from glacial flow over the Allegheny Escarpment in Ohio. Gross and Moran (1971) determined that, on the Plateau in northwestern Pennsylvania, 50% of the Titusville Till is derived from within 32 km of the site of deposition. Once the Keefus glacier advanced over the gray bedrock of the plateau, perhaps the till lost its red color and high carbonate content, and that may be why it has not been identified further south. The till exposed at the base of the section (sample BR-12) also has a carbonate content of 4%. The Mapledale Till generally does not react visibly to dilute HCl in the field (White et al., 1969). White et al. (1969) indicate that there is a second Mapledale sheet in places that has a greater carbonate value. This lowest till may be a lower Mapledale sheet. There is a
98
Fleeger et al.
3+ m covered section between the sand and gravel underlying the Keefus (?) Till and the lowest till, which may contain the upper, low-carbonate Mapledale sheet. A Slippery Rock Till possibility is more difficult to determine. To our knowledge, no unweathered Slippery Rock Till has ever been recognized, so its original mineralogy is unknown. Weathering This section displays some weathering phenomena that complicate interpretations. Patterns of oxidized sediments overlying
unoxidized sediments have traditionally been interpreted as a single sequence of subaerial weathering. However, in this section there are two other factors controlling the patterns of oxidation. One is a weathering “shadow” produced by increased weathering adjacent to sand beds due to groundwater flow through the sand beds. The other is the uneven base of the oxidized zone because of oxidation to greater depths along joints. There are two sand beds in this section that appear to have oxidation shadows in the overlying and/or underlying till. One is adjacent to the sand bed between Titusville 2 and 3, and the other between Titusville 3 and 4. The upper of the two sand beds
Figure 9. Two possible interpretations of the pre-Kent stratigraphy in the Booth Run section based on the mineralogy of the till samples. The Titusville Till 1, 2, 3, 4, and 5 designations, indicate multiple Titusville Till sheets, as described by White et al. (1969), separated by multiple sand beds. The Titusville Till, Keefus Till?, and Mapledale Till? designations suggest fewer till sheets, based on weathering sequences, and suggest possible correlations with other Grand River Lobe tills.
Quaternary geology of northwestern Pennsylvania appears to have a shadow above it. Whether a shadow exists in the subjacent till is difficult to determine. The underlying till has a 1 m thick oxidized zone. Oxidation shadows generally extend considerably less than 0.3 m into the adjacent till. This till may also have undergone subaerial weathering. The overlying 18-cm shadow is more typical. The lower sand bed has only a very thin shadow in the underlying till. The other weathering complication is caused by weathering agents penetrating deeper into tills along joints. The joints in the till are usually discreet linear fractures, which serve as preferred pathways for weathering agents. Oxidation proceeds downward along the joints and also outward from the joints, faster than between joints. This section displays many examples of oxidized till adjacent to joints within generally unoxidized till. The oxidation extends outward from some joints only a centimeter or less, but 10 or more centimeters at other joints. Weathering along joints produces the same initial mineralogical changes as subaerial weathering. For example, samples BR-5 and BR-6 were taken beside each other (Fig. 9). BR-6 is partially leached of carbonates and has had its clay minerals altered from the composition of BR-5. This section has sequences that progress downward from oxidized till, to oxidized till containing unoxidized till pods (masses of gray, unoxidized till completely surrounded by oxidized till), to unoxidized till with oxidation only along joints, to completely unoxidized till (Fig. 10). The unoxidized pods appear to be remnants of till that have not yet been affected by weathering moving downward along joints or laterally along other preferred paths. The downward sequences from oxidized to unoxidized till overlap with the oxidation shadows, making it difficult to determine the cause of oxidation in some places, and where stratigraphic breaks may be.
99
STOP 4. LUNCH STOP 5. HARMONSBURG RECESSIONAL WASHBOARD MORAINE As we travel on this road, we observe rolling topography that has been recently opened up for the exploitation of underlying sand and gravel deposits that may have been deposited as ice melted back to the Harmonsburg recessional moraine. More importantly is the occurrence of washboard moraine (Figs. 11 and 14) originally described by Shepps (19591, 1962). The occurrence of a washboard moraine in Pennsylvanian is rare because this landform is more common in places such as central Iowa in the area glaciated by the Des Moines lobe or in central Canada. In the literature there are several modes of origin and many other names for washboard moraine (Benn and Evans, 1998). The Pleistocene geology of the area would appear to rule out the glacio-lacustrine origin, which attributes the ridges to grounding lines of ice terminating in a large lake. Sudgen and John (1984) suggest that these ridges may form at the base of thrust planes near a compressed ice terminus. If an advancing tongue of Hiram ice formed these, that lobe should have been undergoing extension at this location. Another mode of origin includes seasonal or annual moraines formed during a general retreat of ice. This may involve some pushing of sediments in the winter advance followed by mass wasting of sediments onto the top of the pushed sediment during the summer retreat (Benn and Evans, 1998). The internal structure of the individual ridges might provide a clue as to their origin. After retreat of the Hiram lobe to the north, this recessional moraine blocked the south end of the Conneaut Creek valley (Fig. 12). The Hiram glacier blocked the northern outlet of the valley, forming a lake. Lacustrine sediments are present throughout the valley north of the moraine. The valley-blocking moraine eventually was breached, allowing the lake to partially drain to the south into the Ohio River basin. When the Hiram glacier retreated far enough to open an outlet to the north, the lake drained into the St. Lawrence basin. The present divide between the Ohio and St. Lawrence basins is on this moraine. Presumably, this moraine formed when the block of ice that created Conneaut Lake (Fig. 12) broke off from the front of the Hiram lobe. Outwash issuing from this ice front buried the ice block, which when melted formed the Conneaut Lake kettle. Some of the drainageways leading from the area of the moraine are visible on Figure 12. STOP 6. CONNEAUT LAKE Conneaut Lake (Figs. 1 and 14) is the largest kettle lake wholly within Pennsylvania and formed during late Wisconsinan
Figure 10. Sketch illustrating an idealized weathering sequence from oxidized to fresh till, and oxidation shadows adjacent to sand layers within a till mass. It also shows the complexities created when two sources of oxidation (from surface and sand beds) intersect.
1
V.C. Shepps’ 1959 field guide, “Glacial geology of northwestern Pennsylvania,” is available as GSA Data Repository Item 2011102, at www.geosociety .org/pubs/ft2011.htm, or on request from
[email protected].
100
Fleeger et al.
Figure 11. View of the front (south) slope of the washboard moraine from Harmonsburg Road.
deglaciation. It is situated just to the south of the Harmonsburg recessional moraine and within a major outwash valley that drained southward from the margin of the Grand River sublobe into the Upper Ohio River watershed. Portions of the original kettle form amphitheater-like topography adjacent to the northern margin of the modern lake (Figs. 12 and 18). This area was the main focus of an M.S. thesis produced by Kelley Hartley (2009) who investigated the limits of late Wisconsinan glacial advances near Conneaut Lake. The extents of different till sheets around the lake have been mapped differently throughout the years. Two major reports about the glacial deposits within the region, published within ten years of each other,
show conflicting interpretations of till sheet limits within Crawford County. The study completed by Shepps et al. (1959) shows Kent Till covering much of Crawford County and tongues of Hiram Till deposited by small tongues of ice that advanced down valleys associated with Conneaut Lake and Pymatuning Reservoir to the west (Fig. 1). White et al. (1969) eliminated the Conneaut Lake tongue, replaced it with a discontinuous blanket of Lavery Till in this part of Crawford County, and placed the Hiram boundary farther north in southern Erie County (Fig. 1). Till of a discontinuous sheet would be expected to be preserved on the highest parts of the topography in the area mapped as Kent by Shepps et al. (1959). Therefore Hartley (2009) not only sampled
Figure 12. Hillshade image of the Stop 5 area derived from lidar digital elevation model data. North is to the top of the image.
Quaternary geology of northwestern Pennsylvania
Figure 13. Stratigraphy and correlations of units found in boreholes near Conneaut Lake; subunits are defined in Table 1 (modified from Hartley, 2009).
borings in the Hiram lobe area but also in the adjacent Kent area to the west of Conneaut Lake. Samples from 15 auger borings (Fig. 13), a measured section at Pymatuning reservoir, and diamict samples from large gravel pit were described in the field, and textural, mineralogic, and lithologic properties were analyzed in the laboratory. Matrix textures (% < 2 mm) were determined using methods of Folk (1974). The gasometric method of Dreimanis (1962) was employed to determine percentages of fine carbonate in the <0.074-mm component of diamicts. Diffraction intensity ratios (DI of Willman et al., 1966) of the 1.0 nm illite peak area divided by the 0.7 nm chlorite and kaolinite peak area were calculated. Lithology of the 1–2 mm sand fractions was examined as being representative of pebble content (Anderson, 1957). Data were obtained from eight borings located in the area mapped as Kent Till by Shepps et al. (1959), seven boreholes located within the tongue of Hiram Till (Fig. 13), the section at Pymatuning Reservoir, and the large gravel pit on the east side of the lake. Boreholes in the Kent area penetrated yellowishbrown, sandy diamicts, averaged ~3.5 m in depth, and terminated in bedrock, whereas those in the Hiram area contained brown, fine-grained diamicts, averaged 8 m in depth and ended in a gray, sandy, uniform diamicts easily recognized among boreholes. Using descriptions and laboratory data, four units were designated from A, old, to D, young (Table 1). Unit A is associated with the Kent advance (Fig. 13), and its diamict is typically
Unit D C B1 B2 A1 A2 A3
101
loamy. Although diamicts B2 and C were initially separated by clay contents, they have similar silt contents and similar amounts of calcite, dolomite, and total carbonate (Hartley, 2009). Unit B was limited to two boreholes and may represent slight textural variations of Unit C; both are correlated with the Hiram advance (Fig. 13). Unit D is problematic and was found to overlie a complex sequence of ablation sediments in the large pit east of Conneaut Lake in March of 2008. This unit had been stripped off and pushed into spoil piles on the east side of the property by June of the same year. Unit D is a dark brown, fine-grained diamict having an average fine-carbonate content approaching 18%, whereas other diamicts in the area average less than 5% (Hartley, 2009). Diamict D overlies Unit C, which displays a wide range of textures and facies where it overlies thick gravels of ice-contact deposits. Unit D is typical of the Hiram Till as found near its limit in northeastern Ohio (Szabo, 2006). Both the Hiram and Lavery end moraines lay several kilometers northwest of Conneaut Lake, and there is no evidence that the main front of their advances extended farther south, especially on the uplands. There is sufficient evidence that the Kent advance extended southward far past Conneaut Lake and blanketed the region. The Lavery advance did not extend much beyond its moraine running diagonally across Erie County (Shepps et al., 1959; O’Brien, 2004). There is no fine-grained unit on the uplands that matches descriptions of the Lavery Till as described by Weinreich (2006). One hypothesis for explaining the occurrence of fine-grained diamicts beyond their traditional limits is that tongues of Hiram ice advanced over proglacial lacustrine deposits in outwash valleys of Conneaut Lake and Pymatuning Reservoir. This resulted in the deposition of fine-grained diamicts of units B and C in the lowlands. As Hiram ice melted back, ice blocks detached from these ice tongues and were surrounded by outwash. The origin of the brown diamict of unit D and its restricted areal extent is problematic. Although diamicts of units C and D have similar clay contents (Hartley, 2009), diamicts of unit D have four times as much fine carbonate as those of unit C. One hypothesis is that Unit D may represent the calcareous debris carried englacially and released by superglacial meltout of the ice block responsible for Conneaut Lake. An alternative hypothesis is that englacial ice may have sheared off of stalled basal ice (Szabo and Totten, 1992) and flowed over ice marginal sand and gravel deposits and also deformed Unit C on the east side of the Conneaut Lake.
TABLE 1. UNITS AND SUBUNITS DEFINED BY HARTLEY (2009) Distinguishing characteristics Very calcareous, silty, dark brown diamict Weakly calcareous, yellow-brown diamict having a clay content near 35% Weakly calcareous interbedded silt, clay, and sparsely pebbly diamict Weakly calcareous diamict having a clay content between 25% and 30% and a small pebble content Weakly calcareous, interbedded silt, clay, and diamict Weakly calcareous sand and gravel Weakly to moderately calcareous diamict having sand content near 34%
102
Fleeger et al.
Unfortunately, little has been done in terms of understanding the post-glacial history of the lake, but a limited glimpse of what is known of its Holocene history is presented here. During the late glacial and presumably into the early Holocene, the lake was deeper and encompassed more geographic area extending to the steep amphitheater-like topography. The area’s rich archaeological history suggests a landscape attractive to Native American populations. Like many lakes in the region (e.g., Yu et al., 1997; Finkelstein and Davis, 2006; Li et al., 2007), Holocene levels of Conneaut Lake likely fluctuated within a few meters of its modern level during much if not all of the Holocene. Conneaut Lake was near its current lake level by ~6350 calendar years before the present based on a radiocarbon date (BETA 238191) from peat that unconformably underlies silty floodplain alluvium to the north of the modern lake inlet. Lake sediment underlies the peat in a few cut banks, but more commonly underlies alluvium. The alluvium has been largely deposited during the late Holocene, predominantly over the past several hundred years. Euro-American settlement within the watershed largely converted the native forested land cover to an agriculturally dominated landscape (Grote et al., 2010). Historic land-cover changes within the watershed likely are responsible for the deposition of significant volumes of sediment within the flood plain and in the lake itself. Gravel Pit in Kames, East Side of Conneaut Lake Conneaut Lake Sand and Gravel, Inc., currently is mining the large gravel pits through which a county road passes on the high ridge east of Conneaut Lake (Figs. 14 and 15). This pit is on the edge of the Hiram boundary as mapped by Shepps et al. (1959). The original topography consisted of a north-south line of kames veneered with very calcareous, dark brown diamict of Unit D (Hartley, 2009) previously mentioned. Hartley (2009) examined the spoils from the stripping of the unit along the eastern property line of the gravel pit. She also observed greenish-blue gleyed diamict suggesting that there were several closed water-logged depressions in the landscape. Ten samples of Unit D averaged 3% sand, 61% silt, and, 36% clay and contained an average fine-carbonate content consisting of 8.8% calcite and 9.1% dolomite. Their clay mineralogy was dominated by illite, and their diffraction intensity ratios averaged 1.9, typical of lower Paleozoic rocks of the Niagara peninsula (Szabo, 2006). Granules and small pebbles were scarce in the diamicts, but larger 2–8-cm-long pebbles were common. Gleyed deposits contained more sand (10%–23%) and less fine carbonate (3.8%) suggesting transport of weathered diamict from topographic highs into the depressions. The exposed sediments beneath Unit D consist of variable deposits of interbedded gravel, sand, silt, and diamicts (Fig. 16). The ice-contact environment of deposition of Unit C is illustrated in Figure 17, where glacio-fluvial sediments are truncated by a near vertical wall of diamict. Photographs taken over the years by trip leaders show dipping silt and sand lenses
accentuated by weathering in addition to large gravel bodies, some of which, represent meltwater channels. Layers of diamict seem to appear randomly and truncate or veneer other sediments. Ten samples of gray, diamict of Unit C averaged 11% sand, 52% silt, and 37% clay; they contained 1.1% calcite and 2.7% dolomite. Borehole 12 (Fig. 13) located at the site of an old waterslide park east of State Route 18 and west of the pit contained nearly 4 m of fill overlying 1 m of weathered Unit C that had an average texture of 10% sand, 60% silt, and 30% clay. This unit also had a fine-carbonate content averaging 1.3% fine carbonate. The remainder of the boring penetrated less than a meter of sand and gravel overlying diamict of Unit A. The stratigraphy of the gravel pit suggests a stagnant mass of Hiram ice wedged against the upland east of the lake. As this ice mass melted, several topographic inversions may have occurred accounting for the dipping silt and sand bodies and the occurrences of diamict produced by sediment flows off higher parts of the decaying ice. Kames and thick outwash deposits, not only of the most recent advance but also of older advances, are more common along the eastern margin of the former Hiram ice tongue than along its southern or western margins. The southern and western margins of the former ice tongue are dominated by rolling topography of end moraine that merges into the bedrock-supported uplands veneered with thin Kent deposits (Hartley, 2009). The Hiram deposits may be at least 8.5 m thick in some areas of the moraine (Hole 2; Fig. 13), but are only a few meters thick in other areas mirroring the underlying Kent deposits. Kames also occur in the large flat area west of the lake as ground moraine is replaced by deposits of an outwash plain northwest of the lake. The asymmetry of the topography may have influenced meltwater flow as the ice margin retreated. The topography along the east side of the lake is relatively more rugged and steep when compared to flatter landscape adjacent to the west side of the lake. This may suggest that meltwater flowed southward away from the ice margin along the western side of the detached ice block that would become Conneaut Lake. STOP 7. GENEVA MARSH: LATE WISCONSINAN DEGLACIATION AND POST-GLACIAL VALLEYS SYSTEMS At this stop we will discuss our current understanding of the post-glacial (<15 ka) valley system evolution in the Upper Ohio–Allegheny River headwaters region. Geneva Marsh is located within the Conneaut outlet valley, which drains Conneaut Lake, our previous stop (Fig. 18). The Conneaut outlet valley and other major northwest-southeast–trending valleys throughout the region are predominantly the result of subglacial erosion. These valleys have been partly buried by a variety of glacigenic, lacustrine, paludal, and fluvial deposits, in some cases hundreds of feet thick. The hill tops surrounding the valley are mapped as having a cover of Kent ground moraine, while high-elevation, hummocky
Quaternary geology of northwestern Pennsylvania
103
Figure 14. Topographic map of the area of Stop 6. From the Harmonsburg (north) and Conneaut Lake (south) 7½-minute quadrangles.
Figure 15. View of the exposure in the Conneaut Sand and Gravel pit in July 2010.
104
Fleeger et al.
Figure 16. Interbedded glacio-fluvial deposits and diamicts (at the base of the section).
Figure 17. Abrupt vertical transition from glacio-fluvial deposits to diamicts.
landforms flanking the valley are mapped as kame terraces and outwash deposits associated with retreating late Wisconsinan ice (Shepps et al., 1959). The flat to gently sloping, low-lying surface is the Holocene floodplain. Fluvial terraces of known Holocene age are conspicuously lacking, while underfit streams, swamps, and lakes are features typical of post-glacial drainages in northwestern Pennsylvania.
Grand River Sub-lobe of the Laurentide Ice Sheet into northwestern Pennsylvania (Fig. 19). The end moraine complex can be subdivided into three landform-sediment assemblages: (1) end moraine, (2) ice-contact terrain, and (3) proximal outwash terrain. The end moraine itself is characterized by conspicuous high-relief hummocky topography that is underlain by a thick sequence of interbedded till and outwash deposits. The core of the end moraine may contain remnants of pre-Wisconsinan till sheets. Ice-contact terrains are characterized by low-relief hummocky topography, primarily along valley walls, that is underlain by a chaotic sequence of englacial and ablation/flow tills interbedded with poorly sorted, stratified to massive glaciofluvial
Deglaciation of Northwestern Pennsylvania The Kent end moraine complex marks the earliest and most extensive Late Wisconsinan ice advance associated with the
Figure 18. Lidar hillshade model of the region around field trip Stops 5, 6, and 7. Lidar data from Pennsylvania Imagery Navigator.
Quaternary geology of northwestern Pennsylvania sediments. Proximal outwash terrain is characterized by moderately flat to undulating topography, sometimes exhibiting distinct braid bar morphologies, that is dominantly underlain by stratified glaciofluvial sand and gravel deposited in braided stream environments. Occasionally, proglacial lake sediments are interbedded with glaciofluvial sediments. When taken as a whole, the landform-sediment assemblages associated with the Kent end moraine complex indicate that initial Late Wisconsinan deglaciation in northwestern Pennsylvania occurred as stagnation zone retreat along a dead ice margin (Straffin and Grote, 2010). Post-Glacial Landscape Evolution The over-deepened glacial valleys of the region contain an aggradational sedimentary record from lakes, floodplains, swamps, and alluvial fans that provide evidence of changing Holocene landscape stability related to prevailing environmental conditions and land use/land cover history. In particular, several natural kettle lakes (e.g., Conneaut and Edinboro Lakes), the moraine dammed Sugar Lake, and peat/marsh sediments from Conneaut Lake valley all contain long sedimentary
Figure 19. Section of Glacial Map of Shepps et al. (1959) with moraines and glacial lakes.
105
archives that have been examined by coring and radiocarbon dating. Natural and artificial exposures within the French Creek watershed, which drains much of northwestern Pennsylvania, have also been examined in order to develop the regional floodplain stratigraphy. Sugar Lake (Fig. 19), a moraine-dammed lake within the outer margin of the Kent end moraine complex which marks the maximum extent of Wisconsinan glaciation in northwestern Pennsylvania, contains a continuous record of sedimentation spanning the last 14 thousand years. Analyses of lithology and magnetic susceptibility from a 12-m-long sediment core permit interpretations regarding environmental change within the watershed during the late and post-glacial period. The oldest sediments record an early phase of predominantly clastic, silty sedimentation when glacial ice still existed within the valley. Annual summer/winter cycles (varves) are evident as alternating light gray silt and black organic laminations, respectively. As glaciers retreated northward, varves were gradually replaced by increasingly organic sediments. That trend was terminated by a rapid change to predominantly homogeneous organic sedimentation, which likely marks the transition to an ice-free valley and climatic amelioration at the beginning of the Holocene Epoch. Decreases in lake sediment magnetic susceptibility values suggest that clastic sedimentation generally declined as organic content increased through the early and middle Holocene, although several distinct pulses interrupt that trend, suggesting episodes of landscape instability and/or hillslope soil erosion during the past ~3 thousand years. The French Creek fluvial system also appears to have been active during the same period of time. The French Creek floodplain is characterized by middle Holocene lateral accretion facies that are overlain by fine-grained (silty) vertical accretion facies that accumulated during the past ~3.5 thousand years (Fig. 20). A similar story is documented by alluvial fan sediments, which indicate several discrete depositional events separated by weakly developed soils over the past ~2 thousand years. The timing of fan deposition is modestly correlative with phases of floodplain aggradation. Independent paleoenvironmental proxies for the area suggest the late Holocene prior to Euro-American settlement was a time of cool and wet climate under which a pine/hickory/beech forest predominated (Grote et al., 2010). This paleoenvironmental setting would have been conducive to frequent overbank flood events in the French Creek watershed. Deforestation and the transition to an agriculturally dominated landscape beginning in the late 1700s significantly altered geomorphic and hydrologic processes within the watershed. The Sugar Lake core shows gyttja dominates the late Holocene record, with a final phase of increased magnetic susceptibility likely recording human-induced hillslope erosion over the last 100+ years. Lastly, many portions of the French Creek floodplain, tributary floodplains, and adjacent watersheds all contain post-settlement alluvium within the floodplain stratigraphy, some as much as ~2 m thick, that definitively document the profound impact.
106
Fleeger et al.
Figure 20. Regional climate, vegetation, and lake level status derived from independent proxy records of environmental change, compared with valley responses of northwestern Pennsylvania. NE regional climate periods inferred from pollen (Shuman et al., 2004), local vegetation inferred from pollen (Walker and Hartman, 1960), vegetation transitions after Gajewski et al. (2006), and northeastern lake levels after Shuman and Donnelly (2006).
Quaternary geology of northwestern Pennsylvania ROAD LOG Mileage Int Cum 0.0 0.0
0.2 0.1
0.2 0.3
14.9
15.2
13.8
29.0
8.8 3.4
37.8 41.2
5.7
46.9
0.9 0.3
47.8 48.1
1.4
49.5
1.3 0.3
50.8 51.1
0.8
51.9
0.3 0.6
52.2 52.8
1.7
54.5
0.8
55.3
0.8 0.6
56.1 56.7
Description From William Penn Hotel, start on William Penn Ave., heading south (one way). The first few turns happen quickly, so pay attention! Make 4 turns, each after one block: Turn left onto Oliver Ave. Turn left onto Grant St. Turn right onto 6th Ave. Turn left onto Bigelow Blvd. Keep left on Bigelow to Veterans Bridge. Left on I-579 north and Veterans Bridge, crossing the Allegheny River. Keep left and continue to I-279 north to I-79 north. Merge with I-79 north, continue ~26 miles, then watch out for zombies! Evans City was the filming location for Night of the Living Dead. Cross the Connoquenessing Creek, site of pre-Illinoian pro-glacial Lake Connoquenessing. Pass the exit to Portersville. Cross US 422, which was the site of the Late Wisconsinan border and an ice dam creating glacial Lake Watts (till on the west side of the interchange, bedrock on the east side). Exit right to PA 108 (Slippery Rock exit) to stop sign, then turn left on PA 108 toward Slippery Rock. Turn right onto West Park Rd. Cross Slippery Rock Creek. This stream was ice-dammed to form glacial Lake Edmund. Turn left onto Dickey Rd. in the village of Jacksville. Road travels along the Jacksville Esker. Turn left onto West Liberty Rd. Stop 1A. View of Jacksville Esker to the north. Turn right onto Moore Rd. and Stop 1B. Hummocky backside of delta (to south). Stop 1C. Flat top of delta. Turn right at stop sign onto Rohrer Rd. (lake floor to left, south, side of road). Rohrer Rd. passes along the base of the front (distal) slope of the delta. Stop, turn right onto Mt. Union Rd. Cross lake bottom, then up delta front (bedrock underlying hill to left, delta to right of road). Stop 1D. Glacial Sand and Gravel pit in kame delta. Stop, turn left onto West Liberty Rd. Stop, turn right onto Dickey Rd.
1.3 1.7 0.8 8.9 0.1 0.6 0.2 1.2 0.2 1.0 0.1
58.0 59.7 60.5 69.4 69.5 70.1 70.3 71.5 71.7 72.7 72.8
0.1 1.0 0.2 1.2
72.9 73.9 74.1 75.3
3.1 0.1 7.1 6.7 2.1 12.1
78.4 78.5 85.6 92.3 94.4 106.5
0.5 3.1 6.8
107.0 110.1 116.9
2.7 1.6 0.6 1.4 1.0 1.0 0.4 1.0 5.6 3.4 1.7
119.6 121.2 121.8 123.2 124.2 125.2 125.6 126.6 132.2 135.6 137.3
0.7
138.0
0.4 0.7 1.2 0.8 1.9 3.2 5.6 0.9
138.4 139.1 140.3 141.1 143.0 146.2 151.8 152.7
107
Stop, turn right onto West Park Rd. Stop, turn left onto PA 108. Exit right to I-79 south. Exit right at Exit 96 to PA 488. Turn left onto PA 488 west. Turn right onto US 19 north/PA 488 west. Turn left onto PA 488 west. Turn right onto Pfeifer Rd. Turn left onto Magee Rd. Turn right onto Breakneck Bridge Road. Turn left to Cleland Rock Scenic Vista. Stop 2. Slippery Rock Gorge. Return to Breakneck Bridge Rd. Turn right onto Breakneck Bridge Rd. Turn left onto Magee Rd. Turn right onto Pfeifer Rd. Turn left onto PA 488 east. Turn left onto US 19 north. We will cross two of the temporary glacial Lake Watts drainage passes in the next 3 miles. Turn right onto ramp to US 422. Bear right onto the ramp to US 422 west. Bear right to stay on US 422 west. Merge with I-376 north. Pass US 422 exit and Mahoning River. Pass over I-80. Route number changes to PA 760. Exit to PA 18 north through Hermitage. Intersection of PA 18 and US 62. Exit right and turn left in 0.1 miles. Cross PA 18 headed west onto Rutledge Rd. Stop. Continue straight. Turn right (north) onto Summit Rd. Stop. Continue straight on Summit Rd. Turn left (west) on Woods Rd. Turn right (north) on S. Barry Rd. Stop 3. Booth Run Section. Stop, turn right (east) onto PA 358. Turn left (north) onto Summit Rd. Stop. Turn right (east) onto PA 58. Turn left (west) onto US 322.. Turn right onto West Lake Rd. into Pymatuning State Park. Turn right toward the Jamestown Marina. Lunch and Stop 4 in Environmental Classroom. Turn left onto West Lake Rd. Left onto the dam road. Left onto East Lake Rd. Right onto East State Rd. Turn left onto US 322. Stop. Continue straight on US 322. Conneaut Lake. Turn left onto US 6 (west). Turn right (north) onto PA 618.
108
Fleeger et al.
3.4 0.8 0.9 0.6
156.1 156.9 157.8 158.4
0.4 0.9 0.9 5.2
158.8 159.7 160.6 165.8
0.6 0.2
166.4 166.6
0.4 1.4
167.0 168.4
0.2
168.6
5
173.6
0.5
174.1
0.5
174.6
0.2 1.2 0.1 69.4 11.4
174.8 176.0 176.1 245.5 256.9
0.9 0.6 0.1
257.8 258.4 258.5
Turn left onto PA 18 (north). Turn left onto Harmonsburg Rd. Turn right onto Hindman Rd. Stop 5. Harmonsburg recessional washboard moraine. Turn right onto Pachuk Rd. Turn right on PA18 (south). Traffic light. Keep straight on PA 18 south. Turn left onto Lake Side Dr. There are 3 stop signs; continue east, straight ahead, at each. Stop, then turn left onto Ellion Rd. Stop 6. Conneaut Lake Sand and Gravel pit in large kame. Leave kame gravel pit heading south on Ellion Rd. Turn right (west) onto 322. Conneaut Lake to north, the largest natural lake in Pennsylvania, is a large kettle lake. Turn left onto 285 at second light in Conneaut Lake Turn left (east) onto Marsh Rd. (unmarked Fish Commission road). Drive to the end of Marsh Rd. Stop 7. Geneva Marsh, with Holocene history. Turn around, drive back to entrance to Marsh Rd. Turn left (south) onto PA 285. Continue to I-79. View of glacial valley and swamp to the left. Stop and continue straight across US 19. Exit right to enter onramp to I-79 south. Slight left at I-279 S (signs for Pittsburgh). Take exit 8A to merge onto I-579 S toward Veterans Bridge. Exit onto 7th Ave. Turn left onto William Penn Pl. Go 2 blocks to the William Penn Hotel.
ACKNOWLEDGMENTS We thank Duane Braun, professor emeritus from Bloomsburg University of Pennsylvania and Mike Angle, Mapping Group supervisor at the Ohio Department of Natural Resources, Division of Geological Survey for many useful comments. REFERENCES CITED Anderson, R.C., 1957, Pebbles and sand lithology of the major Wisconsin glacial lobes of the central lowlands: Geological Society of America Bulletin, v. 68, p. 1415–1450, doi:10.1130/0016-7606(1957)68[1415:PASLOT] 2.0.CO;2. Benn, D.I., and Evans, D.J.A., 1998, Glaciers and glaciation: London, Arnold Publishing Co., 734 p. Bruno, P.W., 1988, Lithofacies and depositional environments of the Ashtabula Till, Lake and Ashtabula Counties, Ohio [unpublished M.S. thesis]: Akron, Ohio, University of Akron, 207 p. Bushnell, K., 1975, Slippery Rock Gorge: Pennsylvania Geological Survey, 4th series, Park Guide 9.
D’Urso, G., 2000, Revised glacial margins and Wisconsin meltwater paleoflood hydrology in Slippery Rock Creek basin, central western Pennsylvania [unpublished Ph.D. dissertation]: Morgantown, West Virginia, West Virginia University, 174 p. D’Urso, G., Burkhart, P., Livingston, J., and Iksic, C., 2004, An examination of field methods for glacial margin mapping and paleoflood reconstruction in Slippery Rock Creek basin, western Pennsylvania: Pittsburgh Geological Society Field Guide, 28 p. (http://www.pittsburghgeologicalsociety .org/publications.htm). Dreimanis, A., 1962, Quantitative gasometric determination of calcite and dolomite by using a Chittick apparatus: Journal of Sedimentary Petrology, v. 32, p. 520–529. Droste, J.B., and Tharin, J.C., 1958, Alteration of clay minerals in Illinoian till by weathering: Geological Society of America Bulletin, v. 69, p. 61–68, doi:10.1130/0016-7606(1958)69[61:AOCMII]2.0.CO;2. Droste, J.B., Rubin, M., and White, G.W., 1960, Age of marginal Wisconsin drift at Corry, northwestern Pennsylvania: Science, v. 130, p. 1760. Durco, N., 2009, Mapping eskers with lidar in Butler and Lawrence Counties, Pennsylvania: Geological Society of America Abstracts with Programs, v. 41, no. 3, p. 66. Finkelstein, S.A., and Davis, A.M., 2006, Paleoenvironmental records of water level and climate changes from the middle to late Holocene at a Lake Erie coastal wetland, Ontario, Canada: Quaternary Research, v. 65, p. 33–43, doi:10.1016/j.yqres.2005.08.021. Fleeger, G.M., 1986, The West Liberty Esker: Pacific Geology, v. 17, no. 1, p. 7–9. Fleeger, G.M., Bushnell, K.O., and Watson, D.W., 2003, Moraine and McConnells Mill State Parks, Butler and Lawrence Counties—glacial lakes and drainage changes: Pennsylvania Geological Survey, 4th series, Park Guide 4, 13 p. Folk, R.L., 1974, Petrology of sedimentary rocks: Austin, Texas, Hemphill Publishing Co., 182 p. Gajewski, K., Viau, A., Sawada, M., Atkinson, D., and Fines, P., 2006, Synchronicity in Climate and Vegetation Transitions Between Europe and North America During the Holocene: Climatic Change, v. 78, p. 341–361, doi:10.1007/s10584-006-9048-z. Geyer, A.R., and Bolles, W.H., 1979, Outstanding scenic geologic features in Pennsylvania, Pennsylvania Geological Survey, 4th series, Environmental Geology Report 7, 508 p. Gross, D.L., and Moran, S.R., 1971, Grain-size and mineralogical gradations within tills of the Allegheny Plateau, in Goldthwait, R.P., ed., Till, a symposium: Columbus, Ohio, Ohio State University Press, p. 251–274. Grote, T., Straffin, E., Kerschner, A., Malzone, J., and Jones, K., 2010, Records of late Holocene landscape evolution in the French Creek watershed, Northwestern Pennsylvania: Geological Society of America Abstracts with Programs, v. 42, no. 1, p. 152. Hartley, K.A., 2009, Stratigraphic analysis of areal discontinuities of Late Wisconsinan till sheets near Conneaut Lake, northwestern Pennsylvania [unpublished M.S. thesis]: Akron, Ohio, University of Akron, 124 p. Ireland, H.A., 1940, New evidence for an Illinoian glacial boundary in northeastern Ohio: Geological Society of America Bulletin, v. 51, p. 1337–1358. Jacobson, R.B., Elston, D.P., and Heaton, J.W., 1988, Stratigraphy and magnetic polarity of the high-terrace remnants in the upper Ohio and Monongahela Rivers in West Virginia, Pennsylvania, and Ohio: Quaternary Research, v. 29, p. 216–232, doi:10.1016/0033-5894(88)90031-2. Leverett, F., 1902 Glacial formations and drainage features of the Erie and Ohio basins: U.S. Geological Survey, Monograph XLI, 802 p. Leverett, F., 1934, Glacial deposits outside the terminal moraine in Pennsylvania: Pennsylvania Geological Survey, 4th series, General Geology Report 7, 123 p. Lewis, H.C., 1884, Report on the terminal moraine in Pennsylvania and western New York: Pennsylvania Geological Survey, 2nd series, Report Z, 299 p. Li, Y.X., Yu, Z.C., and Kodama, K.P., 2007, Sensitive moisture response to Holocene millennial-scale climate variations in the Mid-Atlantic region, USA: The Holocene, v. 17, p. 3–8, doi:10.1177/0959683606069386. O’Brien, A.M., 2004, Surficial map, subsurface analysis, and interpretation, Pleistocene-age glacial deposits in Cambridge Springs and Blooming Valley 7.5′ quadrangles, Crawford and Erie Counties, northwestern, Pennsylvania [unpublished B.S. thesis]: Meadville, Pennsylvania, Allegheny College, 70 p. Palmer, T., 1980, Rivers of Pennsylvania: Keystone Books, 229 p.
Quaternary geology of northwestern Pennsylvania Poth, C.W., 1963, Geology and hydrology of the Mercer quadrangle: Pennsylvania Geological Survey, 4th series, Water Resource Report 16, 149 p. Preston, F.W., 1949, More on the status of the Illinoian in western Pennsylvania: Preston Collection, AGR International, Inc., unpublished Report 49-069, 20 April 1949. Preston, F.W., 1977, Drainage changes in the late Pleistocene in central western Pennsylvania: Pittsburgh, Carnegie Museum of Natural History, 56 p. Shepps, V.C., 1955, The glacial geology of a part of northwestern Pennsylvania [unpublished Ph.D. thesis]: University of Illinois, 117 p. Shepps, V.C., 1959, Field Trip 5: Glacial geology of northwestern Pennsylvania, in Guidebook for field trips, Pittsburgh meeting, 1959, National GSA Guidebook. Shepps, V.C., 1962, Pennsylvania and the Ice Age: Pennsylvania Geological Survey, 4th series, Educational Series 6, first edition, 33 p. Shepps, V.C., White, G.W., Droste, J.B., and Sitler, R.F., 1959, Glacial geology of northwestern Pennsylvania: Pennsylvania Geological Survey General Geology Report G 32, 59 p. Shuman, B., and Donnelly, J.P., 2006, The Influence of Seasonal Precipitation and Temperature Regimes on Lake Levels in the Northeastern United States during the Holocene: Quaternary Research, v. 65, p. 44–56, doi:10.1016/j.yqres.2005.09.001. Shuman, B., Newby, P., Huang, Y., and Webb, T., 2004, Evidence for the close climatic control of New England vegetation history: Ecology, v. 85, p. 1297–1310, doi:10.1890/02-0286. Sitler, R.F., 1957, Glacial geology of a part of western Pennsylvania [unpublished Ph.D. dissertation]: University of Illinois at UrbanaChampaign, 119 p. Straffin, E.C., and Grote, T., 2010, Surficial geology of the Sugar Lake 7.5-minute quadrangle, Crawford and Venango Counties, Pennsylvania: Pennsylvania Geological Survey, 4th ser., Open-File Report OFSM 10-05.0, 23 p., Portable Document Format (PDF). Sudgen, D.E., and John, B.S., 1984, Glaciers and landscape: London, Edward Arnold Publishing Co., 326 p. Szabo, J.P., 1987, Textural and mineralogical composition of pre-Woodfordian tills, north-central Ohio, in Totten, S.M., and Szabo, J.P., leaders, PreWoodfordian stratigraphy of north-central Ohio: Guidebook for the 34th Midwest Friends of the Pleistocene meeting, p. 26–45. Szabo, J.P., 2006, Quaternary geology of the interlobate area between the Cuyahoga and Grand River lobes, northeastern Ohio: Ohio Division of Geological Survey Guidebook 20, 52 p. Szabo, J.P. and Totten, S.M., 1992, Glacial dispersal rejuvenation on the Allegheny Plateau, north-central Ohio, based on till carbonate patterns: Journal of Sedimentary Petrology, v. 62, p. 1044–1053. Szabo, J.P., and Totten, S.M., 1995, Multiple pre-Wisconsinan glaciations along the northwestern edge of the Allegheny Plateau in Ohio and Pennsylvania: Canadian Journal of Earth Sciences, v. 32, p. 2081–2089.
109
Totten, S.M., 1976, The “up in the air” late Pleistocene beaver pond, Lodi, Medina County, northern Ohio [abs.]: Geological Society of America Abstracts with Programs, v. 8, no. 4, p. 514. Totten, S.M., and Szabo, J.P., 1987, Stop 5: Mount Gilead Fairgrounds Section, in Totten, S.M., and Szabo, J.P., leaders, Pre-Woodfordian stratigraphy of north-central Ohio: Guidebook for the 34th Midwest Friends of the Pleistocene meeting, p. 63–67. Walker, P.C., and Hartman, R.T., 1960, The forest sequence of the Hartstown Bog area in western Pennsylvania: Ecology, v. 41, p. 461–474, doi:10.2307/1933321. Weinreich, M., 2006, A detailed sedimentological and geomorphic investigation of the Wisconsinan tills near the Lavery type section, northwestern Pennsylvania [unpublished M.S. thesis]: Akron, Ohio, University of Akron, 114 p. White, G.W., 1942, Illinoian and Wisconsin drift of the Grand River Lobe in eastern Ohio: Geological Society of America Bulletin, v. 53, p. 1813. White, G.W., 1960, Classification of Wisconsin glacial deposits in northeastern Ohio: U.S. Geological Survey, Bulletin 1121-A, 12 p. White, G.W., 1968a, Age and correlation of deposits at Garfield Heights (Cleveland), Ohio: Geological Society of America Bulletin, v. 79, p. 749–756, doi:10.1130/0016-7606(1968)79[749:AACOPD]2.0.CO;2. White, G.W., 1968b, Pleistocene deposits of the north-western Allegheny Plateau, USA: Quarterly Journal of the Geological Society of London, v. 124, p. 131–151, doi:10.1144/gsjgs.124.1.0131. White, G.W., 1982, Glacial geology of northeastern Ohio: Ohio Department of Natural Resources Division of Geological Survey, Bulletin 68, 75 p. White, G.W., Totten, S.M., and Gross, D.L., 1969, Pleistocene stratigraphy of northwestern Pennsylvania: Pennsylvania Geological Survey, 4th series, General Geology Report 55, 88 p. White, I.C., 1880, The geology of Mercer County: Pennsylvania Geological Survey, 2nd series, Vol. QQQ, 233 p. Willman, H.B., Glass, H.D., and Frye, J.C., 1966, Mineralogy of glacial till and their weathering profiles in Illinois, part II. Weathering profiles: Illinois State Geological Survey Bulletin 94, 204 p. Wright, G.F., 1892, The extra-morainic drift of the Susquehanna valley: American Geologist, v. X, p. 219. Yu, Z.C., McAndrews, J., and Eicher, U., 1997, Middle Holocene dry climate caused by change in atmospheric circulation patterns: Evidence from lake levels and stable isotopes: Geology, v. 25, p. 251–254, doi:10.1130/0091 -7613(1997)025<0251:MHDCCB>2.3.CO;2.
MANUSCRIPT ACCEPTED BY THE SOCIETY 6 DECEMBER 2010
Printed in the USA
The Geological Society of America Field Guide 20 2011
The history and geology of the Allegheny Portage Railroad, Blair and Cambria Counties, Pennsylvania John A. Harper Pennsylvania Geological Survey, Pittsburgh, Pennsylvania 15222-4745, USA
ABSTRACT The Allegheny Portage Railroad, just one leg of the Pennsylvania Mainline Canal system, was the first railroad over the Allegheny Mountains, an imposing physiographic barrier to westward migration in the early 1800s. Construction of the canal system began in 1826 and continued until ca. 1840 without interruption. The Allegheny Portage Railroad began construction in 1831 and opened for business in 1834. This astonishing engineering feat took less than four years for completion, despite the necessity of 10 inclined planes and the use of the new-fangled railroad locomotives. Construction made use of many of the natural resources occurring along and adjacent to the right-of-way, especially the Pennsylvanian-aged sandstones used for the “sleepers” that held the rails in place. Travel occurred in sectional canal boats, boats that were built in two or three pieces that could be easily loaded onto rail cars. Passengers and goods were loaded onto the boat sections in Philadelphia, which were then hauled by horse or locomotive to the Susquehanna River west of Lancaster. The boats traveled north on the Susquehanna River canal to the mouth of the Juniata River north of Harrisburg, then along the Juniata River canal to Hollidaysburg near the foot of Allegheny Mountain. There, the boats were taken from the water, loaded onto rail cars, and hauled over the mountain on the Allegheny Portage Railroad to Johnstown where they were unloaded into the Conemaugh River canal for the journey to Pittsburgh. A New Allegheny Portage Railroad was built in the 1850s to bypass the inclined planes. It was no sooner built, however, when the state sold the entire canal system to the Pennsylvania Railroad for less than half the cost of construction. The Pennsylvania Railroad promptly dismantled the Allegheny Portage Railroad and filled in the canals. Today, the Allegheny Portage Railroad National Historic Site oversees and administers the preservation of the few remaining aspects of the old railroad. INTRODUCTION The Allegheny Portage Railroad (APR) was the first railroad over Allegheny Mountain (the Allegheny Front). Considered a technological wonder in its day (1834–1854), the railroad played a critical role in opening the interior of the United States
to further settlement and additional trade. It allowed settlers and traders to travel from the east coast to the center of the North American continent without major interruption by forming a link between the Juniata River portion of the Pennsylvania Mainline Canal System in central Pennsylvania and the Ohio River drainage (via the Conemaugh and Allegheny rivers). David Stevenson,
Harper, J.A., 2011, The history and geology of the Allegheny Portage Railroad, Blair and Cambria Counties, Pennsylvania, in Ruffolo, R.M., and Ciampaglio, C.N., eds., From the Shield to the Sea: Geological Field Trips from the 2011 Joint Meeting of the GSA Northeastern and North-Central Sections: Geological Society of America Field Guide 20, p. 111–141, doi: 10.1130/2011.0020(06). For permission to copy, contact
[email protected]. ©2011 The Geological Society of America. All rights reserved.
111
112
Harper
a distinguished English civil engineer, spoke of the APR as a mountain railway that had no equal. He compared it to the passes of the Simplon, and Mont Cenis, in Sardinia, but felt that even they paled in comparison (Stevenson, 1838). During its brief existence, thousands of people made the journey on the APR, including the English novelist Charles Dickens, who traveled to America in 1841 and 1842 and chronicled his trip in “American Notes for General Circulation” (Dickens, 1842). Dickens wrote of giddy precipices, valleys full of light and softness, scattered cabins, and people and animals going about their daily routines without a hint of curiosity. The APR went through two iterations, then was bought and dismantled by the Pennsylvania Railroad (PRR) in 1854. Today, only parts of the APR remain. The Allegheny Portage Railroad National Historic Site, a unit of the National Park Service, covers 6 km2 of the site where the APR arrived at the top of Allegheny Mountain (see Stop 3 for details). The park also oversees the Staple Bend Tunnel, located ~6 km east of Johnstown in Cambria County. The inclined planes and some of the railroad right-ofway west of Allegheny Mountain remain, much of it being used as PA Route 53 between the towns of Summit and Portage. The route has been marked by the Allegheny Ridge Heritage Corridor, a private not-for-profit organization operating with state support. HISTORICAL PERSPECTIVE The Pennsylvania Mainline Canal By the beginning of the 1800s, the main U.S. population centers were found only along the Atlantic seaboard. Traveling from Europe to the United States had been relatively easy—you got on a boat and sailed across the Atlantic. Traveling westward from the coast, however, presented a host of difficulties for the average person. As General Forbes and his army found out in the previous century (Briggs, 1998), the Allegheny Mountains presented a formidable barrier to westward travel. A few brave souls made the trek—pioneers, soldiers, and eventually those who saw a better life away from the main population centers. As tales of vast stretches of fertile farmland, great rivers, abundant forests and game, and troves of mineral resources found their way back east, others began following the pioneers westward. At that time nothing moved faster than the speed of a horse and except on a racetrack, no horse moved very fast (Ambrose, 1996). There weren’t many roads in the United States at the beginning of the nineteenth century, and the condition of the roads that existed ranged from bad to abominable. Travel on the best highway in the country, which ran from Boston to New York, took three days to make the 280-km journey. It required Thomas Jefferson 10 days to travel the 362 km from his home at Monticello to Philadelphia (Ambrose, 1996). Turnpikes began to be built early in the 1800s. The Huntingdon, Cambria and Indiana Turnpike (now old U.S. Route 22), completed in 1819, was just a dirt track that crossed the Allegheny Mountains within the deeply etched valley called Blair Run
Gap between Hollidaysburg and Cresson (Jacobs, 1945) (Fig. 1). Roads such as these often had little to recommend them. Rivers offered better, faster, and safer means of transportation, but they were controlled by both the landscape and the vagaries of nature. Drought meant no river transportation, whereas too much rain created potentially disastrous floods within the river valleys. Canals helped solve this problem, as well as a myriad of others. Pennsylvania had considered creating a canal system late in the 1700s, but it wasn’t until the 1820s that any progress on the concept began to occur. On 31 March 1824 the Pennsylvania Legislative Assembly appointed a Board of Canal Commissioners to investigate possible canal routes between Harrisburg (“civilization”) and Pittsburgh (“the boondocks”) (Wilson, 1897). The Board considered two possibilities: (1) the Juniata and Conemaugh rivers (the latter flowed into the Allegheny and, eventually, into the Ohio); and (2) the West Branch of the Susquehanna River and Sinnemahoning Creek with a crosscountry link to the Allegheny River in McKean County. A year later a second commission began making surveys and estimates for a system of canals that would link: (1) Philadelphia to Pittsburgh; (2) Allegheny (now Pittsburgh’s North Side) to Erie; and (3) a section that would ultimately connect with the existing canals in New York. A canal convention convened in Harrisburg in August 1825 generated a concept that won much public support. This resulted in petitions, circulated throughout the state, being presented to the legislature. Construction of the Pennsylvania canal system began in 1826 and continued until ca. 1840 without interruption. The system eventually became a hybrid of public and private canals and railroads (Fig. 2), but for travelers trying to get to Pittsburgh from Philadelphia, the entire journey was made in canal boats. The boats were loaded in Philadelphia and unloaded in Pittsburgh and
Figure 1. Allegheny Mountain as seen from Chimney Rocks Park above Hollidaysburg, showing the imposing nature of the Allegheny structural and topographic front. Blair Run Gap is on the far left, and Sugar Run Gap is on the right. Photograph by John A. Harper.
Allegheny Portage Railroad the great western frontier. The Pennsylvania Mainline Canal, as it was called, opened in 1834 and operated for 20 years. The story of how the APR eventually came to be a part of the Pennsylvania Mainline Canal system is one of false starts, varied perspectives and opinions, numerous trials and tribulations, and the inability of the legislature, the Board of Canal Commissioners, and several governors to make decisions. A Canal Tunnel through Allegheny Mountain The original concept of the Pennsylvania canal system called for a continuous waterway from Philadelphia to Pittsburgh (Wilson, 1897; Jacobs, 1945), including constructing a canal tunnel through Allegheny Mountain. It would have been 7.2 km long and filled with water to provide continual passage of canal boats. Two of the canal commissioners sent a report to the governor in February 1825 that stated, “Even good men, who love to see the improvement of their country, have been startled at the idea of burrowing in the ground for a few miles, to let large boats pass through the bowels of the Allegheny.” (Wilson, 1897, p. 37) They estimated the cost of such a tunnel was $480,000. The third commissioner suggested a few weeks later that the tunnel was completely impractical. The whole region of the Allegh-
113
eny Mountains, especially where the proposed tunnel was to be constructed, is heavily forested and surrounded by high mountains with steep slopes, separated by narrow ravines. The governor, faced with such contrary information, decided to not decide on the matter. In 1826, the commissioners sent another report to the governor describing the results of ongoing engineering surveys. In this report, they recommended the Juniata-Conemaugh route as preferable to the Susquehanna-Allegheny route, but they also recognized that the canal tunnel concept was impractical. There seemed to be no practical way to keep such a tunnel filled with water other than by tunneling at a much lower elevation than originally proposed. This would have increased its length, and was, therefore, considered an “insuperable objection.” By the time the 1826–1827 legislative session commenced, the tunnel idea was all but dead. The alternative, the Board of Commissioners decided, was a portage across Allegheny Mountain to Johnstown. Canvass White, the engineer in charge of the 1826 survey, suggested that canal boats could be constructed in three or four pieces, divided transversely, so they could be transported over the portage without changing the cargo. This was the first official suggestion that sectional boats should be built. These boats
Figure 2. Map of the Pennsylvania canal system and associated railroads (dotted) (modified from Shank in American Canal Society, 2004). The Juniata Canal, a division of the Pennsylvania Mainline Canal, ran 204 km between Duncan’s Island in the Susquehanna River and Hollidaysburg, and had 86 locks. It remained in operation from 1832 until 1888.
114
Harper
played a very important role in the history of the Pennsylvania Mainline Canal. The Advantage of Railroads In 1802, Thomas Jefferson predicted that steam would change human civilization if it were applied to a carriage on wheels (Ambrose, 1996). Jefferson was a man of great vision, but he never lived long enough to see a train, let alone the automobile that he envisioned. While the first Board of Canal Commissioners was convening, a series of reports on the new-fangled railroad began to attract a lot of attention, particularly as to the possibility of using railroads rather than canals to cross the state (Wilson, 1897). As the number of railroad supporters increased, the demands on the legislative assembly also increased for a study of the efficacy of railroads to serve the transportation needs of the commonwealth. In February 1825, the Pennsylvania Senate appointed a committee to study the practicality of constructing a railroad from Philadelphia to Pittsburgh. The proposition was ahead of its time—Pennsylvania just wasn’t ready for crossstate railroads in 1825—but it focused attention on the fact that the best way to cross the state was along the route eventually chosen for the Pennsylvania Mainline Canal, and the “unrealistic” railroad concept was seen as an acceptable means of portage over the mountains. An alternative was travel by highway. The mud trail of the Huntingdon, Cambria and Indiana Turnpike could have been improved into a well-maintained highway westward from Hollidaysburg. By the end of 1826, opinion for the portage was fairly balanced between the concepts of turnpike and railroad. In January 1827, George T. Olmstead, assistant engineer on the canal survey, reported that the horizontal distance from Hollidaysburg to Johnstown over the mountain was ~66 km and the vertical distance was 895.93 m. The railroad proposal eventually began to gain ground. It took another year before the Board of Canal Commissioners could accomplish anything substantive. The engineer in charge of the 1828 surveys favored a railroad side by side with a paved (macadam) turnpike road sharing the same grade of one degree or less. Since he couldn’t convince the canal board to make a clear decision, he resigned and was replaced by Moncure Robinson. Robinson had thought long and hard about the topic of portage before actually ascending to his important role. Early in 1829, Robinson visited the Allegheny Mountains already convinced that a railroad would be a far superior method of portage than either a canal tunnel or turnpike roads. He also believed that stationary steam engines and locomotives would be far more effective and economical sources of power than horses. He saw two determining factors for siting the portage: a deep gap in the mountain and the shortest distance between Hollidaysburg and Johnstown. He reported to the canal commissioners in November 1829, proposing the best way to cross the mountain was by means of a system of straight, inclined planes operated
by stationary engines, combined with railroad lines on the intervening flat lands. The proposed railroad wouldn’t exceed 61 km in length, in contrast with any possible macadam turnpike that would have to be at least 80 km long to be as effective. In addition, he recommended shortening the distance over the mountain summit by constructing a 1.6-km-long tunnel ~1.5 km north of the existing turnpike. The tunnel could be constructed at an elevation of 385 m above the canal at Hollidaysburg, fully 54 m lower than the summit. He estimated the cost of this railroad would be $936,004.87 (Wilson, 1897). Robinson’s recommendations were far too bold and straightforward for the governor, the legislature, or the canal board to make any decisions until they could receive confirmation by other civil engineers. In 1830, the canal board appointed a Board of Engineers, consisting of Moncure Robinson, Lieutenant Colonel S.H. Long, and Major John Wilson, to re-survey (once again!) the Allegheny Mountains. The engineers reported to the board that fall. They concluded that Robinson’s original proposal for a railroad was best and that it should cross the mountain at Blair Gap. They also suggested reducing the length of the railroad by building a viaduct across the Little Conemaugh River at one meander bend and digging a 1000-foot-long tunnel through another. Long and White disagreed with Robinson on the proposed summit tunnel. They favored a route that included eleven inclined planes, six on the east side of the mountain and five on the west, connected by a cut through the summit that would be 457 m long and no more than 5.5 m deep. Robinson vigorously defended his tunnel proposal and predicted that within five years it would become obvious that he was correct. This, generally, was the only part of Robinson’s original proposal that was not accepted. On 21 March 1831, the governor finally signed a legislative act authorizing the building, without delay, of the APR (Wilson, 1897). Nine days later, Sylvester Welch took control of the project as principal with Robinson as consulting engineer, and Samuel Jones as superintendent. Thus, the building of the APR began. Construction (Finally!!!) The surveys from Johnstown to the summit of Allegheny Mountain commenced early in April 1831 and were completed by 20 May, and the construction work was contracted out at Ebensburg on 25 May. The surveys from the summit to Hollidaysburg were completed and the work was contracted out on 29 July (Wilson, 1897). The contractors built ten inclined planes, the Conemaugh Viaduct, the Staple Bend Tunnel, and the railroad right-of-way. The right-of-way was 37 m wide to accommodate any additional track that might be necessary in the future and to ensure that any fallen trees would not block the tracks. Because most of the chosen route was heavily forested, the contractors had to cut much timber before they could begin grading the railroad. But fresh timber does not burn readily and the logs typically were too big to haul out unless cut into smaller segments. The work went slowly. In addition, the contractor originally hired to build the
Allegheny Portage Railroad Conemaugh Viaduct backed out of the contract without having accomplished much, and that portion of the construction had to be re-contracted the next year. At 8 m wide, the railroad bed was wide enough for two sets of tracks. One set of tracks was laid between the inclined planes and two sets were laid on the inclines so cars could be raised and lowered simultaneously, thus counterbalancing each other (see below). The bed was constructed so that the steepest grade was no more than 10.25 percent. The tracks consisted of rolled iron rails in sections 5.5 m long and 237 pounds forged by Harfords, Davis and Co. of Wales, and set in cast-iron “chairs” secured by iron wedges. The whole assembly was then attached to a set of cut stone blocks called “sleepers” (Fig. 3) chiseled from local sandstone by stonemasons. Where such rock wasn’t available, or would have cost too much, wooden timbers (ties) replaced the “sleepers.” “Sleepers” were spaced approximately every 0.9 m along the road. The blocks tended to shift with weather and moisture variations, so the rails often separated, making it impossible for trains to move safely along the tracks. Eventually wooden cross ties, such as those seen on modern railroads, replaced most of the “sleepers.” The ties were not as prone to movement and were much easier to repair or replace. The early ropes used to raise and lower cars on the inclined planes consisted of a combination of Italian and Russian hemp laid in shrouds of from 360 to 450 yarns. These varied in length from 1.1 to 2 km, with a composite length of 17.77 km, and had a total cost of $20,531.05 (Welch, as quoted in Shank, 1975). Seven of the ropes were ~5.7 cm in diameter. Four of them were each made in one piece, whereas the others were each made of several pieces spliced together. Unfortunately, these ropes, as strong as they were, broke frequently. John Roebling, the architect and engineer who built the Brooklyn Bridge, suggested using “wire rope” such as he was developing in his shop in Saxonburg, Butler County, and by 1849 all of the APR’s rope had been replaced by metal cables. The track segments between Hollidaysburg and the foot of Allegheny Mountain, and those between the inclined planes, had low grades that allowed horses to haul the railroad cars. In 1835, the engine “Boston” became the first locomotive to run on the APR. It did the work of 18 horses and proved so successful that over the next few years the railroad acquired 16 more locomotives. Eventually the use of horses was phased out completely. The inclined planes used high-pressure stationary steam engines to raise and lower railroad cars. There were two engines at each incline, each fed by three boilers. Most were 35-horsepower (HP; = 26,100 kilowatt [kW]) engines with 35.6-cm cylinders and a 1.5-m piston stroke that allowed ascending cars to move at ~6 km/h. These had boilers 0.8 m in diameter and 6 m long. Engines at inclined planes 2, 5, 9, and 10 were somewhat smaller 30-HP (= 22,370 kW) engines having 33-cm cylinders and a 1.5-m stroke, and had boilers 0.8 m in diameter and 5.5 m long. By increasing the amount of steam, the engines could produce up to 60 HP (= 44,740 kW), thus enabling the systems
115
to haul heavier loads, or to move the cars up the inclines at a faster pace. The machinery for raising and lowering canal cars originally was to be operated by single-cylinder engines with flywheels. Because of safety factors, however, the canal commissioners decided instead on two-cylinder engines with no flywheels, and the accompanying machinery had to be adapted to that power source. Sylvester Welch felt that flywheels were the major cause of accidents on other inclined planes where stationary engines were used, and encouraged the use of rope alone to stop the engine without danger of being broken (Shank, 1975). The engines, boilers, and machinery for operating the ropes resided below track level in an engine house at the top of each incline. Two cast-iron wheels, 2.4 m in diameter and with 15-cm grooves for the ropes, were placed vertically in the center of each set of tracks 30.5 m from the head of the incline (Figs. 4 and 5). Cogwheels on the assembly that were 1.2 m in diameter made them revolve in opposite directions, and a set of clutches allowed the wheels to be disengaged while the engines were operating. A third cast-iron wheel, 292 cm in diameter (the distance between
Figure 3. Stone “sleepers” along the Allegheny Portage Railroad. (A) Rails attached to “sleeper” along the right-of-way at the Allegheny Portage Railroad National Historical Site. Photograph by John A. Harper. (B) Details of a rail and “sleeper” (based on Shank, 1975).
Figure 4. Schematic plan view diagram of a typical engine house used for the stationary engines on the inclined planes by Fred R. Connacher (from Shank, 1975). See Figure 5 for a cross-sectional schematic diagram.
116 Harper
Figure 5. Cross-sectional schematic diagram of a typical engine house used for the stationary engines on the inclined planes by Fred R. Connacher (from Shank, 1975). See Figure 4 for a schematic plan view.
Allegheny Portage Railroad
117
the centers of the two sets of tracks on the incline), revolved horizontally on a movable carriage between the vertical wheels and the head of the inclined plane. This latter wheel generally was anchored in one spot by a weight suspended in a well, but it could be moved ~4.6 m when necessary. At the bottom of the incline, a similar horizontal wheel on a carriage was anchored in a similar fashion 12 m from the foot of the incline. A double pulley block, rope, and windlass allowed it to be moved ~15 m. The rope that hauled the canal cars ran a complicated pattern. First it passed over one of the two vertical wheels, then through a hole in the pit wall, around the horizontal wheel, back through another hole in the pit wall, under and over the second vertical wheel, down the incline, around the horizontal wheel at the bottom, and finally back up the incline. Forty-six cm wheels spaced 7 m apart between the tracks supported the rope on the incline. Since the rope was moving down one track and up the other simultaneously, one car could be hauled up the incline while another was going down. If the weight of cars descending the incline exceeded that of cars ascending, or if there were no ascending cars at all, the engines were disengaged from the hoisting mechanism and gravity took over. A water brake (see Fig. 5) regulated the velocity of the descending cars, keeping them from running out of control. This brake consisted of a water-filled cylinder, 35.6 cm in diameter and 1.8 m long, bolted to the pit wall separating one of the engines from the horizontal wheel. The cylinder had air chambers at either end and a side pipe with a sliding valve that allowed the water to pass into and out of the cylinder at the stroke of the piston that was attached by gears to the drive shaft. The sliding valve was used to regulate the water flow, which in turn regulated the velocity of the descending cars. A clutch allowed the brake to be disengaged when the engines worked the hoisting mechanism. “Safety cars” (Fig. 6)—special trucks attached to the rope on the downhill side of the cars— provided additional braking ability. Completion of the Allegheny Portage Railroad The first track was completed and open for traffic 18 March 1834, less than three years after the final surveys had been done. The second track was finished in late spring of 1835. The APR was 59 km long and had a total rise and fall of 783 m between Hollidaysburg and Johnstown. The inclination of the planes varied from ~0.07 percent to a little over 10 percent. The viaduct over the Little Conemaugh River 13 km east of Johnstown comprised a single semi-circular, 24-m arch 8.5 m wide and standing 21 m above the surface of the water. The tunnel at Staple Bend on the Little Conemaugh River, 6 km east of Johnstown, was the first tunnel built in America. It was 274.6 m long, 6 m wide, and 5.8 m high at the top of the arched ceiling. At the opening of the APR on 18 March 1834, there were 25 cars ready for use. Another 25 became available by 1 April and 30 more by 18 April. These could be hauled by either horse or locomotive. Although designed specifically to haul canal boats, the APR also hauled other types of cars (Fig. 7). The canal boats
118
Harper
were built in sections (Fig. 8) to make their movement overland practicable. Passengers and goods boarded half a canal boat that was mounted on wheeled trolleys in Philadelphia. Horses or mules hauled the sections through the streets to the railway terminus where the boat halves were transferred to special railroad trucks. At the Susquehanna River, the boat sections were reassembled and drawn by horses to the Juniata River canal, then along the Juniata to Hollidaysburg. There they were transferred once again to railroad trucks and hauled over the mountain by locomotives and stationary engines. Finally, at Johnstown at the western end of the Allegheny Mountain portage, the boats were reassembled to complete the journey to Pittsburgh on the Conemaugh River canal. The whole journey from Philadelphia to Pittsburgh, a distance of ~644 km that used to require months of hardship, could be covered in relative safety in less than a week. Canal boats apparently were not meant as luxury accommodations, however. Charles Dickens (1842, p. 154) described them quite vividly as “a barge with a little house in it, viewed from the outside; and a caravan at a fair, viewed from within.” He described the sleeping arrangements as “three long tiers of hanging book-shelves, designed apparently for volumes of the small octavo size. Looking with greater attention at these contrivances (wondering to find such literary preparations in such a place), I descried on each shelf a sort of microscopic sheet and blanket; then I began dimly to comprehend that the passengers were the library, and that they were to be arranged, edge-wise, on these shelves, till morning.” The APR was completed in less than four years, no mean feat for the early nineteenth century, especially considering the length of time required for much smaller projects today. For the next 20 years, the railroad would serve well those willing to travel west to make their fortunes.
doubled the cost of operation as compared with the level or very low-grade parts of the railroad. On the last day of the 1835–1836 legislative session, the state legislature charged the canal commissioners with finding a better way of taking travelers over Allegheny Mountain. In October 1836, chief engineer Charles DeHaas looked to keep as much of the old road as possible, including the Staple Bend Tunnel and the Conemaugh Viaduct. He recommended increasing the overall distance to provide a lower grade of not more than 253 m/km to allow the APR to enter the Staple Bend Tunnel without use of an inclined plane or deepening the floor of the tunnel. His new road would have increased the distance
The New Allegheny Portage Railroad Shortly after the APR opened people began complaining about the lack of convenience and the safety of the inclined planes. Generally regarded as “nuisances,” the inclines just about
Figure 6. “Safety car” used on inclined planes to prevent rail cars from crashing out of control down the inclines if the cable broke (from Shank, 1975).
Figure 7. Early rail cars on the Allegheny Portage Railroad. (A) Box car (from Davis, 1931). (B) Passenger car (from Thompson, 2010).
Allegheny Portage Railroad
119
to 93 km but it would have bypassed every inclined plane but the first. In 1840, S.M. Fox, principal assistant engineer, determined that a railroad line could cross the mountain, avoiding the inclined planes entirely, and increasing the overall distance only slightly. From Summit to Johnstown no grade would exceed 238 m/km, and the route would make use of 13 km of the old grade with an increased distance of only 1.5 km. Fox determined that the best place to put a tunnel was at the summit of the Sugar Run Gap. From Hollidaysburg to the Sugar Run summit was only 6 km longer than the existing line to Blair Gap summit, and it would avoid Inclined Planes 6 through 10 while keeping the grade to less than 238 m/km. The Board of Canal Commissioners, however, delayed any decisions until construction of the PRR began in 1847. The old APR was constantly being repaired or adjusted because of frost heave, landslides, foundation problems in the engine houses, embankment failures, and rotting wood in road and bridge superstructures. Repairs were done piecemeal on a daily basis, and the APR was never actually in good working order. About the only important improvement made before 1850 was the replacement of hemp ropes with John Roebling’s metal cables. With the beginning of construction of the PRR across the state in 1847, the legislature began to understand that the “old” APR had outlived its usefulness. This became especially apparent when the PRR hoped to use the APR as part of its line until it could complete its own line over the mountains. Such economic interests had their impact. It cost at least $10,000 per year to keep each inclined plane in repair. The governor suggested spending
$500,000 to bypass four of the five inclined planes between Summit and Johnstown, and only repairing the five on the east side of Allegheny Mountain. He felt the PRR would be able to make good use of the APR for many years if these changes were made. Work on the New APR began in June 1851 after the legislature authorized reconstruction that would avoid the planes on the western slope. The new road bypassed Inclined Planes 1 through 3 by the beginning of 1853. From there the road continued along the western slope of the mountains, paralleling the PRR line to a small branch of Clearfield Creek where the two lines diverged. The PRR line went through a 1088-m-long tunnel through the summit, down the northern face of Sugar Run Gap and around the eastern face of Allegheny Mountain to Altoona via Horseshoe Curve. The New APR went through the summit via a tunnel only 549 m in length and ran down the southern face of Sugar Run Gap. It curved around the eastern slope of Allegheny Mountain to Blair Run Gap where it crossed the Old APR at the foot of Inclined Plane 8 on the Muleshoe Curve viaduct (Fig. 9). From there it paralleled, and occasionally crossed, the Old APR line to the top of Inclined Plane 10, then down the slope of the foothills to Newry and north to Duncansville where it used the old railroad line. A 10-km-long branch railroad from Duncansville to Altoona provided access between the PRR and APR. By February 1854 the PRR was ready to go over the mountain on its own road, and by July 1855 the New APR, although incomplete, began operations. The New APR was 72 km long—29 km from Hollidaysburg to the summit, and 43 km from the summit to Johnstown. The summit was 46 m lower than on the old road, reducing the total ascent and descent by 91 m. The maximum grade for the new APR road was 349 m/km on the western slope and 396 m/km on the eastern slope of the mountain. The minimum curvature radius was 213 m. The summit tunnel was dug at the narrowest and
Figure 8. Illustration of a sectional canal boat and safety car ascending Inclined Plane No. 6 at the top of Allegheny Mountain (from Davis, 1931).
Figure 9. Muleshoe Curve Bridge, formerly used by the New Allegheny Portage Railroad and the Pennsylvania Railroad. Photograph by John A. Harper.
120
Harper
lowest point at Sugar Run Gap, 41 m below the summit. The PRR tunnel, by comparison, was as much as 61 m below the summit, even though it was only a few hundred yards farther north. The Board of Canal Commissioners pronounced the New APR superior to other railroads (Wilson, 1897). Thus, the Old APR, once considered one of the “wonders of the age,” had done its work, lived its history, and was decommissioned. The Demise of the Allegheny Portage Railroad But no sooner had the New APR gone into operation than the state decided to sell the entire canal system. The New APR had cost over $2 million dollars to build, and there was no way that revenues from its use were ever going to equal or exceed its operating costs. The entire canal system was in bad shape financially as well as physically, and the advent of the cross-state PRR had put a serious crimp in any revenues the system could have generated. The cost of the state canal system from Philadelphia to Pittsburgh had been $16,504,655.84 (Jacobs, 1945). When the system went on the blocks, the PRR was the only bidder, making payment with bonds valued at $7,500,000—less than half of the total cost of construction. On 1 August 1857 the PRR took possession of the Eastern, Juniata, and Western divisions of the Pennsylvania Mainline Canal, including the APR, and began dismantling them. By 1858 the APR was gone, and by 1864 the PRR had abandoned the entire Western Division of the Pennsylvania Mainline Canal. In 1866 they sold the remaining 286 km of canal to the Pennsylvania Canal Company, a private enterprise, for $2,750,000. The PRR used most of the rails from the APR to extend the Pittsburgh, Fort Wayne and Chicago Railroad 132 km from Plymouth to Chicago, Illinois. Many of the stone “sleepers” were sent to Altoona for use in masonry for the railroad shops. A small side track allowed use of the New APR tunnel as a subsidiary route, and portions of the tracks at Hollidaysburg and Lilly were used as sidings. Long neglected by all but local historians, many of the inclined planes disappeared beneath a cover of forests, and the railroad right-of-way in places was used, and eventually paved with asphalt, for road traffic.
vicinity of Hollidaysburg (Fig. 10). North of the city it is called Brush Mountain; to the east, where it forms a great angular loop, it is known as Lock Mountain; Loop Mountain is the segment to the southeast of town; and Short Mountain and Dunning Mountain can be found south of Hollidaysburg. Subsidiary ridges, such as Catfish Ridge to the south of town, are also common. These ridges typically are the remnant flanks of breached anticlines and synclines. Ridge and Valley Province ridges have fairly steeply dipping slopes, as much as 20 degrees in places. The valleys within the province generally have floors of easily eroded shales and/or carbonate rocks. Some fairly resistant rock layers produce much reduced ridges and knobs within the valleys. The long continuous valley west of Hollidaysburg generally lies at an elevation between 366 and 427 m but rises to between 488 and 533 m in the foothills of Allegheny Mountain. This valley is called Bald Eagle Valley north of Altoona; between Altoona and Hollidaysburg it is called Logan Valley; and west and south of Hollidaysburg it goes by the name of Frankstown Valley.
REGIONAL GEOLOGY Physiography and Drainage The APR crossed three physiographic sections representing two of the major provinces of the Appalachians, which can be recognized by specific topographic and geological characteristics. Each presented different sets of problems during construction of the APR. The Appalachian Mountain Section of the Ridge and Valley Province consists of long, narrow ridges and broad to narrow valleys exhibiting moderate to very high relief (Sevon, 2000). Within the Hollidaysburg area the most prominent ridge trends generally northeast-southwest, but has prominent twists and turns in the
Figure 10. A portion of the shaded-relief map of Pennsylvania (Commonwealth of Pennsylvania, 1999) showing the contrast between the Ridge and Valley (right) and the Appalachian Plateau (left). County boundaries are in white. Locations of towns include: A—Altoona; B— Bedford; H—Hollidaysburg; and J—Johnstown. Arrows indicating cross-strike structural fault systems include: TF—Transylvania fault zone; and TM—Tyrone–Mount Union lineament. See text for details.
Allegheny Portage Railroad The Allegheny Front Section of the Appalachian Plateaus Province consists of a series of rounded to linear hills, cut by deep, narrow valleys, rising by steps to the escarpment of the Allegheny Front (Fig. 1), then tapering away to the west in a series of undulating hills. Local relief is moderate to high (Sevon, 2000). The section, which is commonly called Allegheny Mountain, is an enormous wall of Upper Devonian through Pennsylvanian-age rock with ridge tops between 610 and 732 m above sea level. It is the eastern scarp of the Appalachian Plateau. The section spans ~1126 km along the boundary of the Ridge and Valley and Appalachian Plateaus provinces (Hunt, 1974). It forms the major drainage divide, with streams on the eastern side flowing to the Atlantic Ocean and streams on the western side flowing to the Gulf of Mexico. Wide ridges decreasing in elevation toward the north and separated by broad valleys characterize the Allegheny Mountain Section of the Appalachian Plateaus Province. Like the Allegheny Front Section, local relief is moderate to high (Sevon, 2000). Structures consist of broad, open folds, parallel to the tighter ridges of the Ridge and Valley Province, but becoming more subdued toward the west. The underlying Pennsylvanian and Mississippian rocks typically lie in horizontal strata except where the folds occur. All of the ridges, escarpments, and folds have been dissected at intervals by deep valleys or gaps. The most obvious in the vicinity of Hollidaysburg are the 244-m-deep McKee Gap, where Hatter Creek separates Short Mountain from Dunning Mountain ~10 km to the south, and the 274-m-deep Point View Gap, where the Frankstown Branch of the Juniata River flows through Lock Mountain ~15 km to the northeast. Along Allegheny Mountain, the most prominent in the vicinity of the field trip are Burgoon Run, Sugar Run, and Blair Run gaps. The Juniata River is the major waterway of the Appalachian Mountain Section, draining ~8800 km2 of central Pennsylvania. The main channel of this river forms at the confluence of the southern Frankstown Branch (whose tributary, the Beaverdam Branch, flows through Hollidaysburg), and the northern Little Juniata River. The Little Juniata drains 886 km2 whereas the Frankstown Branch drains 1026 km2 (Juniata Clean Water Partnership, 2000). Both of these rivers originate as numerous tributary creeks on the east slope of Allegheny Mountain. Other tributaries originate on the west slopes of the first set of ridges and within the shale-floored valley between Allegheny Mountain and the ridges. The major rivers of the Allegheny Mountain Section include the Conemaugh River, which drains the northern half of the section, and the Youghiogheny River, which drains the southern half. The Conemaugh forms at the confluence of the Little Conemaugh River and Stoneycreek River in Johnstown. Near Saltsburg, in Indiana County to the west, Loyalhanna Creek joins the Conemaugh and, at this point, the name changes to Kiskiminetas River despite the Conemaugh being a major river in the area. The combined Kiskiminetas-Conemaugh watershed drains ~4890 km2. This is the largest single sub-basin of the Allegheny
121
River, draining 16 percent of the Allegheny’s total drainage (GAI Consultants, 1999). The combined Stoneycreek-ConemaughKiskiminetas River flows 196 river km through western Pennsylvania from northern Somerset County to the Allegheny River. Because of its steep topography, the river valley is one of the most flood-prone areas of the state. Stratigraphy Although the rock strata within the general vicinity of the field trip range from Middle Cambrian to Late Pennsylvanian, the field trip route itself crosses strata ranging in age only from Early Silurian (Clinton Group) to Late Pennsylvanian (Conemaugh Group) (Fig. 11). Middle Cambrian through Middle Ordovician bedrock occurs only east of the crest of the Brush-Dunning Mountain ridge complex, whereas the western flanks of the ridge complex expose Upper Ordovician through Lower Devonian strata. Middle and Upper Devonian shales and siltstones form the floor of the Logan-Frankstown Valley between Hollidaysburg and Allegheny Mountain. Upper Devonian and Lower Mississippian shales and sandstones underlie the eastern escarpment of Allegheny Mountain, and Mississippian sandstones form the summit. From there westward the bedrock consists of Middle Mississippian through Middle Pennsylvania terrigenous rocks. Lower Silurian Tuscarora sandstones, derived from erosion of the Taconic highlands to the east, gave way to mudrocks and carbonates later in the Silurian. During this transition, the amount of clastic material decreased upward until, by the end of the Silurian, the rocks comprise fossiliferous marine limestones. The Tuscarora acts as the major ridge former in the Ridge and Valley Province; it is very noticeable as the nearly ubiquitous boulder fields and talus deposits of white sandstone crowning the upper slopes of the higher ridges. Lower Devonian rocks in central Pennsylvania range from bioclastic shelf carbonates to very coarse-grained sandstones, the result of fluctuating shallow marine depths and increasing clastic input from the east. Sea level continued to fluctuate through the Middle Devonian with carbonates replacing clastics replacing carbonates. By this time, the Acadian orogeny was taking place on the eastern margin of Laurentia. Faill (1985, 1999) found no evidence of Acadian deformational structures in Pennsylvania west of the Piedmont, but noted that sedimentological changes provide a good record of the event. Numerous K-bentonites within the upper Onondaga and lower Marcellus formations (Tioga ash falls) demonstrate that the Acadian orogeny was under way by that time. The encroachment of the Catskill deltaic complex, which began in eastern Pennsylvania in the Middle Devonian, dominated the Late Devonian and Early Mississippian in central and western Pennsylvania. As a result of the Catskill progradation, Upper Devonian rocks consist almost entirely of sandstones, siltstones, and shales, with a few minor limestone beds punctuating the section. The lower part of the section is dominated by marine shelf mudrocks with a general increase in grain size upward through the section. The rocks also reflect a generally
122
Harper upward-shallowing sequence, from prodeltaic through distributary to continental alluvial deposition. Pennsylvanian rocks in the area consist of a highly variable sequence of fluvial and deltaic sandstones, shales, siltstones, claystones, coals, and both marine and nonmarine limestones. With the exception of a few marine limestones, the individual beds generally have limited areal extent, changing facies rapidly over short distances. Surficial deposits of Quaternary age generally consist of unconsolidated material lying on bedrock. These deposits can be separated into distinct types of material based on how and where they were formed. Faill et al. (1989) recognized regolith, colluvium, alluvium, and artificial fill or “works of man” in the Cambria/Blair County area. The composition and arrangement of materials varies greatly depending on the location of the material. Structural Geology The Ridge and Valley structural province is the classic example of a folded and faulted foreland mountain system; the structures formed during the Alleghanian orogeny (Faill and Nickelsen, 1999). The majority of the fold belt extends 1125 km along the eastern interior of North America, from Pennsylvania to Alabama. The Appalachian Plateau is characterized by relatively flatlying rocks, interrupted at intervals by low, broad anticlines lying parallel to the Ridge and Valley. The folds are gentle features with wavelengths between 8 and 32 km. Wavelengths and amplitudes decrease to the west. They are asymmetric, having dips generally of three or four degrees on the northwestern limbs and five or six degrees on the southeastern limbs. Surface faults in the plateau generally are normal faults associated with ancient landslides (slumps) adjacent to Paleozoic river banks. The few thrust faults occur mostly within the major folds in the eastern part of the plateau. Numerous faults of different types have been mapped in the subsurface during oil and gas exploration. Ridge and Valley folds were once thought to be long, continuous folds broken here and there by subsidiary structures resulting from Alleghanian compression alone. As Faill and Nickelsen (1999) observed, however, the folds represent only one stage of deformation that extended over a long period when the principal stress directions changed orientation. Deformation stages included: (1) pretectonic hydraulic jointing; (2) tectonic cross-fold extension jointing and layer-parallel shortening; (3) flexural-slip folding overprinting the previous structures; (4) layer-parallel extension on the steep limbs of folds; and (5) late strike-slip faults and out-of-sequence high-angle reverse faults (Faill and Nickelsen, 1999). Although Appalachian Plateau folds generally are limited in size and scope, folds in the Ridge and Valley come in all sizes, from anticlinoria
Figure 11. Generalized stratigraphic column of the rocks of western Pennsylvania (modified from Harper and Laughrey, 1987).
Allegheny Portage Railroad (wavelength >16 km) to specimens you can hold in your hand (Nickelsen, 1963). Many of the anticlinoria contain subsidiary en echelon folds that are controlled by local thrust faults occurring within different stratigraphic intervals along the length of the major folds. These smaller (second- and third-order) folds generally stand upright, many with vertical or overturned limbs. Some are recumbent (Epstein et al., 1974). They consist of small, straight segments that change orientation in increments, from one or two to as many as 20 degrees, and generally have lengths less than 113 km (Faill and Nickelsen, 1999). Some of these folds are elongate domes, whereas others are long and linear. Most of the surface faults present at the surface in the Ridge and Valley are thrust faults; however, wrench faults do occur in major zones that cross regional strike. Normal faults typically are rare and small, restricted to vertical and overturned beds in the northwest limbs of the anticlines. The most important tectonic element in central and western Pennsylvania actually is a system of southeast-dipping thrust faults that rise through the Paleozoic rock section (Faill and Nickelsen, 1999). These faults generally act as décollement surfaces within certain ductile rock units, before ramping upward through more brittle ones. Many of these faults are blind. Seismic survey data suggest ramping typically occurs above basement normal faults associated with the breakup of the supercontinent Rodinia in the Late Precambrian and Early Cambrian. During the Alleghanian orogeny, the Upper Ordovician through Permian sedimentary rock section was transported northwestward along a basal Cambrian décollement and deformed in a series of imbricate thrust sheets. These strata are generally deformed in approximate imitation of the basal deformation, but contain folds and faults of their own as well. The Cambrian décollement ramps upward at the Allegheny Structural Front to the level of the Silurian Salina Group, which forms the principal basal detachment horizon beneath the Appalachian Plateau from West Virginia to New York (Gwinn, 1964; Frey, 1973). Thrust faults that splay off the basal décollement to form the cores of anticlines tend to be moderately steep and parallel to bedding in the southeastern anticlinal limbs. In contrast, the faults in northwestern anticlinal limbs have low dips and crosscut bedding (Gwinn, 1970). Transverse fracture zones also occur, some with documented strike-slip components. The Transylvania fault zone of Root and Hoskins (1977) (TF in Fig. 10) is the longest in central Pennsylvania. Most large transverse zones have been described as lineaments or cross-strike structural discontinuities (CSDs) (Kowalik and Gold, 1976; Rodgers and Anderson, 1984; Harper, 1989; Gold, 1999). The most prominent of these is the Tyrone–Mount Union lineament, which appears to be the longest in the state. It stands out in the Ridge and Valley as the parallel courses of the Little Juniata and Juniata Rivers (TM in Fig. 10). In addition, Faill (1987) described numerous small structures exposed along its trace in the Little Juniata River valley near Birmingham, including slickensided transverse faults, mesoscopic disharmonic folds, and high fracture density. It terminates many second- and
123
third-order folds, both in the Ridge and Valley and the Appalachian Plateau. It apparently has a right-lateral strike-slip component of ~60 km in the basement (Lavin et al., 1982, based on evidence from gravity data), but a down-to-the-southwest normal fault component in the Paleozoic cover (Rodgers and Anderson, 1984). According to Canich and Gold (1985), this zone is still seismically active. EDWARD MILLER’S GEOLOGY OF THE ALLEGHENY PORTAGE RAILROAD The geology of the area around the APR was first explored during the initial surveys for the right-of-way. Early (i.e., pre– Revolutionary War) explorations for mineral resources were limited and very much confined to lead, zinc, iron ore, and coal deposits. Some early settlers made use of coal and iron ore throughout the Johnstown and Hollidaysburg area, but we don’t actually know when that first occurred. We do know that coal had been found at least as early as 1769, and probably used at that time. And from the journal of Joshua Gilpin, a businessman from Philadelphia who owned property in western Pennsylvania, we learned that iron mills existed in Johnstown by 1809, and that limestone and aluminum-rich claystones cropped out in the hillsides around the town (Brice, 1989). The numerous surveys for the Pennsylvania Mainline Canal made prior to 1831 were more concerned with topography than with geology. In 1831, early in the design phase of the APR, a young engineer named Edward Miller traveled to England to get the latest information on railroads. In 1832, because of the knowledge he had gained, he was named principal assistant engineer on the APR and was placed in charge of the inclined plane machinery, most of which he designed (Roberts, 1878). Miller made quite a name for himself, both with the APR and later in other business. He served as chief engineer of the Pennsylvania Mainline Canal in 1838–1839, became an associate engineer of the PRR under J. Edgar Thompson in 1847, and eventually established his own company in 1862. Wilson (1899) called Edward Miller and Co. a business with integrity, ability, and good resources. Director of the Second Geological Survey of Pennsylvania J. Peter Lesley called Miller “a Civil Engineer of great ability” (Lesley, 1876, p. 48). Besides being an engineer of great ability, Miller also affirmed his place in history by making the first geological report of the Blair and Cambria County area at the 1835 meeting of the Geological Society of Pennsylvania in Philadelphia (Miller, 1835). Although he was quite busy with his duties in the main office with Principal Engineer Sylvester Welch, in his “leisure” time, Miller examined rock outcrops, gathered specimens, took instrument readings, and speculated on what he found. Miller’s 1835 contributions included: (1) an outline map of ~520 km2 of what are now Cambria, Blair, Bedford, and Huntingdon counties at a scale of 1:63,360; (2) a cross section along the railroad from the west side of Hollidaysburg to the area around Inclined Plane No. 3; and (3) specimens presented to the Geological Society of
124
Harper
Pennsylvania that were studied by prominent geologists and paleontologists of the day. The map showed, among other things, the crest line of Allegheny Mountain, the courses of all the streams, the APR right-of-way, and the dip and strike of the strata. The cross section is by far the most interesting and informative contribution (Fig. 12). Miller’s original was 25.4 by 14 cm and had a horizontal scale of 1 cm = 360 m and a vertical scale of 1 cm = 36 m (Fig. 12 is a redrafted version, so the scales aren’t exact). He apologized for the distortion but stated that it was necessary because of the great difference between the scales. He divided the cross section into areas numbered 1–4 and included detailed information on the types of strata, dip of beds, bed thicknesses, and total thicknesses of each. J.P. Lesley, ever the unsatisfied critic, had only slight praise for the man who later became one of the most distinguished civil engineers in the country. “Had he not…exaggerated the vertical scale to eight times that of the horizontal scale, so as to distort all the dips, this section would be not only of the highest interest as a classic in the science, but would stand us in capital stead in our annual report this year (1874-’5)…. Mr. Platt will give Mr. Miller’s text in his report on Cambria and Somerset counties, (Report of Progress in 1875,); but the section as Mr. Miller published it is worthless….” (Lesley, 1876, p. 49). Miller took dips and strikes along the whole railroad and found that the dip direction did not vary greatly from “W.N.W.” He also found that the dips between Inclined Plane No. 3 and Inclined Plane No. 6 did not vary much from 3.5 degrees, but that eastward from Inclined Plane No. 6 the dip gradually increased to 23 degrees at Hollidaysburg. Miller’s cross section contained letter designations used to reference the various strata and the specimens collected in them (Fig. 12). Section No. 1 included what we now map as Rose Hill (Lower Silurian) to the top of the Onondaga Limestone (Fig. 11). He noted that one limestone bed (Onondaga at b in Fig. 12) was ~15 m thick but wasn’t very fossiliferous except where it had been weathered. The other limestones (probably Wills Creek and Tonoloway) contained beds as thin as 2.5 cm and were fossiliferous. Section No. 2 included the Hamilton Group through the lower half of the Catskill Formation (Fig. 11), measured at 1740 m thick. Although predominantly shale, the number and thickness of sandstones increased westward (Brallier to Scherr to Foreknobs), and many of the beds contained marine fossils. The western edge of section No. 2 contained the red shales of the lower Catskill. Section No. 3 included the upper Catskill Formation and at least a portion of the Rockwell Formation (Fig. 11), with alternating shale and predominant micaceous sandstone measured at 1027 m. The color of the rocks changed gradually from red to green traveling westward. Miller noted that the sandstones could be easily quarried in thin slabs of large size. Miller called section No. 4 “The coal measures.” He determined that the starting point for this sequence of rocks was at point e (on Fig. 12) where he found a decided change in the char-
acter of the rocks—sandstone containing plant fossils. In fact, he started the “coal measures” too low in the section. The plant fossils he found probably occurred in the upper Rockwell Formation or, perhaps, the Burgoon Sandstone. Swartz (1965), Inners (1987) and Faill et al. (1989) all described plant remains in this portion of the section. Section No. 4 also contained a quartzose limestone bed ~9 m thick (Loyalhanna Formation at f in Fig. 12). The first true coal appeared at the top of Inclined Plane No. 7 (g on Fig. 12), probably Pennsylvanian in age or, less likely, a welldeveloped seam in the Upper Mississippian. It was only a few cm thick, and since Incline Plane No. 7 passes through the Burgoon and Loyalhanna, and the overlying Mauch Chunk Formation (Berg and Dodge, 1981), it probably was found upslope from the plane, rather on it. Miller also found iron ore (siderite nodules) a little above the coal. Siderite is common above the Mercer coal in western Pennsylvania, so it is likely that this is the section he was describing. He found the coal seams numerous, ranging from 2.5 cm to 1.8 m in thickness, and of varying quality, often within a single seam. Miller found only three limestone beds cropping out in section No. 4: (1) the Loyalhanna, mentioned above; (2) a light blue limestone in a bed 0.9 m thick along Bens Creek near Cassandra (Inclined Plane 3), which is probably the Upper Freeport limestone in the upper Allegheny Formation; and (3) a limestone in the bed of Limestone Run, a small creek between Lilly and Cassandra. He assumed it was the same limestone as the one along Bens Creek. Sandstones varied both in appearance and quality, with outcrops showing a lot of jointing. The best quarries occurred within stream valleys where large blocks lay strewn upon the ground. Considering the amount of sandstone needed for “sleepers” and construction of foundations, bridges, and culverts, it was extremely important to make note of this. Miller stated that rare minerals would not be found along the APR, and that his specimens were valuable only to illustrate the geology of the Allegheny Mountains. Miller’s specimens included: (1) plant fossils from the surface of a coal bed at the top of Allegheny Mountain, and described by Richard Harlan (1835), a well-known physician and paleobotanist of the early 1800s; (2) fossil marine shells collected at the top of Inclined Plane No. 3 that were described and illustrated by Timothy Abbott Conrad (1835), a distinguished paleontologist; and (3) specimens of coal and siderite that were chemically analyzed and discussed by Thomas G. Clemson (1835), who had been with the Royal School of Mines in Paris, France, and later became superintendent of the Flemington Mines in New Jersey (Lesley, 1876). Unfortunately, the suite of rock, mineral, and fossil specimens he sent to the museum of the Geological Society of Pennsylvania in 1833 were lost or misplaced (Lesley, 1876). Some or all of these specimens probably reside in a variety of collections around the state, but no one knows where—this valuable historical collection appears to have been lost for all time. But at least we have published reports with illustrations that can help us deduce what Miller provided to science.
Figure 12. Cross section of the Allegheny Mountains, showing prominent strata and the locations of the inclined planes (redrawn from Miller, 1835). See text for explanation.
Allegheny Portage Railroad 125
126
Harper
Figure 13. Chimney Rocks Park, showing the old limestone quarry walls of Keyser and Tonoloway limestone. Photograph by John A. Harper.
STOP 1. CHIMNEY ROCKS PARK: GEOLOGY AND FIELD TRIP OVERVIEW This stop affords a spectacular view of Hollidaysburg, the Juniata River, and Allegheny Mountain. PLEASE TAKE NOTE: this is a municipal park. Hammers are forbidden, and collecting specimens is discouraged. Photos, however, are encouraged. “Take nothing but pictures; leave nothing but footprints.” Introduction The Chimney Rocks perch on the northwest edge of Catfish Ridge, a low ridge supported by durable carbonates in the Upper Silurian and (possibly) Lower Devonian. The old Chimney Rocks quarry (Fig. 13), and Chimney Rocks themselves (Figs. 14A and 14B), provide dramatic panoramic views of Hollidaysburg and the surrounding area, including the Juniata River, the remains of the Hollidaysburg canal basin, and Allegheny Mountain. Chimney Rocks (Butts, 1945, called them “Tower Rocks”), which include a series of tall, rugged pillars or “chimneys” of faulted and eroded Keyser Formation limestone standing in relief, can be seen from as far away as U.S. Route 22 on Allegheny Mountain. According to legend, the Oneida Indians, who lived in this area, used Chimney Rocks as a vantage for signal fires, alerting villages in the Juniata River valley of impending events. The Chief’s Seat in particular, a rock pad atop the pillars, is supposed to be the site where the chief of the tribe would sit in this area and view his world below.
Figure 14. Keyser limestone at Chimney Rocks. (A) One of the limestone pillars partially obscured by trees. (B) The limestone is very fossiliferous, containing remnants of stromatoporoids and crinoid columnals. Photographs by John A. Harper.
Allegheny Portage Railroad The Chimney Rock Limestone Company operated the quarry, mining the high-grade “calico” rock at the base of the Keyser Formation (see below), which ranges in thickness from 6 to 8 m (Miller, 1934). The limestone was burned for lime in the kiln seen just below the parking lots. Other limestone layers provided crushed stone used for aggregate and road metal. O’Neill (1964) indicated the quarry, which was still active in 1962, had the potential to provide riprap and railroad ballast as well. The Blair County Historical Society obtained this land and deeded ~0.4 km2 to the municipality of Hollidaysburg late in 1994. With the assistance of state funding, the borough created the park, which includes the old quarry and Chimney Rocks, as part of the Allegheny Ridge State Heritage Park in 1998. The main quarry area has been cleaned up and sewn with grass. Be aware of the signs restricting access to the quarry walls. If you would like to view the rock close up, try the unrestricted outcrop near the upper parking lot. Although much can be seen from the lower viewing area at the quarry, the best viewing is from a platform built on the Chief’s Seat at the top of the ridge. A relatively short, though somewhat strenuous, hike up the hillside is necessary to get to Chimney Rocks and the Chief’s Seat. Geology The highwall of the old quarry (Fig. 13) exposes the highest beds of the Tonoloway Formation, all of the basal (“Chimney Rocks”) member of the Keyser and the basal beds of the middle (Jersey Shore) member of the Keyser. Chimney Rocks, at the top of the ridge, is a series of prominent pillars of faulted and eroded Tonoloway and Keyser carbonates. Ulrich (1911) named the Keyser Formation for a series of gray, fossiliferous, crystalline to nodular limestone beds exposed in quarries on the eastern outskirts of the town of Keyser, Mineral County, West Virginia. The Keyser ranges from 27 to 62 m in thickness in central Pennsylvania (O’Neill, 1964), averaging ~30.5 m in Blair County (Faill et al., 1989). Throughout central Pennsylvania, the Keyser consists of three members. In the vicinity of Hollidaysburg, the lower “Chimney Rocks” member (Taylor et al., 2002) lies conformably on the uppermost 1.8 m of the finely laminated Upper Silurian Tonoloway Formation. This member contains several different types of limestone. The basal layer of the member, which Butts (1945) called the “calico rock,” comprises ~9 m of medium-gray, massive, fossiliferous, lime mudstone to bioclastic wackestone/floatstone containing vugs of white, pink, or yellow calcite. These beds often contain crinoids, corals, brachiopods, stromatoporoids, and columnar and mound-like algal stromatolites (Fig. 14B). The highest 1.5– 3 m of the member contain the most abundant fossils. The fossils can be seen in weathered and etched exposures where they stand out against the matrix. Above the basal beds is a sequence, ~2 m thick, of shaly, nodular, lime mudstone with minor interbeds of very fine grainstone. The nodular limestone is fossiliferous, medium-gray, and contains isolated dense nodules up to baseball size. Fossils include corals, brachiopods, bryozoans, and
127
crinoid debris. The upper beds of the member consist of ~2.7 m of echinoderm grainstones, with the lower 0.6 m being thickly bedded and fine grained and the upper 2 m being thinly bedded and coarse grained. Above the “Chimney Rocks” member is the Jersey Shore Member of the Keyser Formation, which can be distinguished from the underlying member by the appearance of a more diverse marine invertebrate fauna, including horn corals, halysitid corals, brachiopods, bryozoans, and stromatolites, and by an increase in yellowish shaly material (Taylor et al., 2002). The contact between the two members can be sharp or gradational. The uppermost member of the Keyser Formation, the LaVale Member, is not exposed at Stop 1 and is not discussed further. The trail up to Chimney Rocks at the top of the park has two side trails in addition to the main trail. The first side trail to the left leads to the top of the old quarry highwall, and the second side trail leads to the base of the more southerly of two isolated limestone towers, or chimneys. This trail allows attendees to view the contact between the Tonoloway and Keyser Formations. The main trail leads to the overlook called Chief’s Seat (which has a metal safety rail and a plaque describing its archaeological significance) and the opportunity to examine the upper parts of some chimneys. Chimney Rocks consist of 2.4–3 m of Tonoloway Formation surmounted by 4.6–6 m of “calico” rock that have been displaced upward with respect to the main body of rock in Catfish Ridge by a vertical fault with 4.6–6 m of throw (Taylor et al., 2002). As a result, the more resistant “calico” rock on the northwest side of the fault is juxtaposed with the more easily eroded nodular limestones and overlying grainstones exposed on the southeast side. In addition, there is a well-developed set of N15°W-striking joints that promoted dissolution (Taylor et al., 2002), allowing the more resistant chimneys to stand out in relief from Catfish Ridge. If you take a few moments to examine the chimneys, you will see numerous fossils, particularly stromatoporoids, on the weathered surfaces, and, depending on the amount of time available, you might even be able to detect the “Chimney Rocks fault” separating the cliff from one or more chimneys. Juniata River Canal and Hollidaysburg Canal Basin Hollidaysburg had been established in 1796 as a tavern stop on the Huntingdon, Cambria and Indiana Turnpike, too far out in the frontier area of Pennsylvania to be considered of any significance. That all changed as the result of one reluctant farmer. Frankstown, which lies 3 km northeast of Hollidaysburg along the Juniata River, and was a far more important town than Hollidaysburg in 1830, was originally selected for the western terminus of the Juniata River Canal because it lies at the confluence of the Little Juniata and Frankstown branches of the Juniata River. It was, therefore, a natural location for a canal basin. The canal engineers decided the best place for the basin was on the farm of Jacob Wertz. Wertz apparently disagreed, however,
128
Harper
because he refused their offer of $10,000 for the land. Then John Blair, who lived at Blair’s Gap (present-day Foot of Ten) at the foot of Allegheny Mountain, used his influence as a member of the Pennsylvania Legislature to have the basin relocated to Hollidaysburg. Many people argued that there wasn’t enough water in the Beaverdam Branch at Hollidaysburg to operate the canal. (Jacobs, 1945, quoted one legislative wag as remarking that the boatmen would be blinded by the dust rising from the bottom of the basin!) Despite the criticism, the Canal Commission in 1831 chose Hollidaysburg for the canal basin. Within the space of five years the town grew from a tiny frontier village with a population of ~76 to an inland seaport with a population of 3,000. And all because a farmer wouldn’t sell his land! With the opening of the Pennsylvania Mainline Canal, Hollidaysburg developed into an important inland shipping terminal and transfer point. The town became a major commercial gateway between the Atlantic seaboard and the western frontier, and the county seat of Blair County in 1846. Altoona soon became more important commercially, thanks to the PRR establishing a major base of operations there in the 1850s. Hollidaysburg, however, continued to play an important role through most of the twentieth century as a commercial, industrial, and governmental center for the region. Many of the canal-era buildings still survive in Hollidaysburg, providing interesting studies for historical architecture buffs. The Hollidaysburg Canal Basin consisted of three interconnected basins (Jacobs, 1945). The upper basin, situated along Bedford Street where the Canal Basin Park is now located (Fig. 15), was ~335 m long and 37 m wide. The main basin was 517 m long, 37 m wide, and 1.8 m deep, extending along South Juniata Street from Montgomery Street to Jones Street (Fig. 15), approximately where PA Route 36 crosses the Conrail yards today. A 183-m canal segment connected these two basins with a lock to a smaller basin located just below what is now Juniata Street. The third pool, located near the intersection of U.S. Route 22 and Allegheny Street, acted as a feeder reservoir for the other two. Canal boats proceeded into the upper basin along the Juniata River. At the lower basin, the boats were unloaded either onto the APR or the various wharves that were located along the basin margin (Fig. 15). A railroad track actually ran down into the canal basin in order to facilitate the easy transfer of boats to the railroad (Jacobs, 1945). The south side of the canal basin, about where the cloverleaf on PA Route 36 is now, was occupied by three boatyards where boats were built and repaired. The canal weigh lock was situated at the foot of Jones Street, and the weighmaster lived in the brick house on the northwest corner of Juniata and Jones Streets (Jacobs, 1945) (Fig. 15). A boat was weighed by running it into a lock at either end and fastening the lock gates to make them as water-tight as possible. Then the water was drawn off through a channel leading to a waste weir. At that point, the boat would be resting on large scales where it could be weighed as accurately as it could be on land. The section adjacent to the canal was a typical seaport, with businesses servicing the canal workers and travelers.
The port of Hollidaysburg was a very busy place, with one boat arriving every 20 minutes on average. The number of passengers and the amount of freight coming into Hollidaysburg required the building of boat slips and docks, large warehouses and freight forwarding houses, and hotels and taverns that catered to boatmen and travelers alike. At the height of the canal’s popularity there were five large forwarding warehouses on the north bank of the basin. In the early days of the canal, the boats were unloaded at the docks and the freight and passengers transferred to railroad cars for the trip across the mountains. The invention of sectional canal boats (Fig. 8) improved this process immeasurably by allowing the boat sections to be hauled out of the water, disassembled, and placed on trucks (railroad undercarriages). After the trip to Johnstown the boat sections were reassembled and continued down the canal to Pittsburgh. Freight shipped separately typically was loaded onto boxcars (similar to the passenger car shown in Fig. 7A) by brute strength. Railroad workers often worked all night loading cars. And then, in the morning, they traveled by rail the 58 km over the Allegheny Mountains to Johnstown where they then had to unload the cars. At times they must have worked 72 hours with their only rest, if you can call it that, being the ride to Johnstown. One canal boat loaded with freight could fill as many as 12 boxcars, making two trains of five or six cars each (Jacobs, 1945). During the apex of its operation, the canal basin and railroad handled between 195 million and 230 million tonnes (t) of freight annually, bringing about $115,000 per year to the state coffers (Jacobs, 1945). Tens of thousands of people passed through Hollidaysburg during the canal era. By the late 1850s, however, the canal system was obsolete. Although the PRR bought the Pennsylvania Mainline Canal system in 1857 and dismantled the APR, the Hollidaysburg canal basin continued in limited operation, serving the needs of area farmers and manufacturers. Eventually, the last canal boat, the William A. Fluke, left Hollidaysburg on 22 April 1872 (Jacobs, 1945). What was left of the canal was abandoned that year. Necessary repairs had been largely ignored, and frequent flooding caused extensive damage to the infrastructure. The feeder reservoir for the canal basin, for example, had been a beautiful, 2 km2 body of water that became a favorite resort for fishermen. The farmers in the river valley below Hollidaysburg feared that the dam would eventually break and drown them (Johnson, 1889). When the canal was abandoned in 1872, the PRR cut the breast of the dam to prevent the water from overflowing and flooding the adjoining land. On 10 February 1882, the dam finally gave way and the water flowed into the Juniata River. Wagonloads of stranded fish were gathered and sold in Hollidaysburg (Jacobs, 1945). The dam was never repaired and, since that time, the land has been used for other purposes. The canal bed over much of its distance became the right-of-way for a branch of the PRR between Hollidaysburg and Huntingdon. In Hollidaysburg the canal basin was filled with cinders; it is now part of the rail yard south and east of the Borough. East
Figure 15. Map of the Hollidaysburg canal basin.
Allegheny Portage Railroad 129
130
Harper
of Huntingdon the canal was so greatly damaged by a flood in 1889 that it also was abandoned. Traces of the once busy waterway can be seen still at several places in the Juniata River valley, east of Hollidaysburg. In the late 1980s, plans for restoration of the canal and canal basin in Hollidaysburg soon began to take shape. After several years of research and fund-raising, the Borough began to study creating a linear park, connecting the Legion Park on North Juniata Street to the original site of the Canal Basin. Eventually, the plan encompassed Chimney Rocks. The newly built Canal Basin Park lies along the Beaverdam Branch of the Juniata River in the Gaysport section of Hollidaysburg. For anyone interested in studying the canal basin in further detail, Canal Basin Park has a museum residing in the restored lock-keepers house. It is a Victorian building housing artifacts and historical documents, an interactive exhibit, and a beautiful mural depicting the history of the canal boats and the region. STOP 2. INCLINED PLANE NO. 10 The area we’ve been traveling through is called the village of Foot of Ten. Inclined Plane No. 10 is in the woods to the south of the village. After we park the bus at the entrance to the access road, we will walk the 30 m or so back through the woods to view the plane.
Inclined Plane No. 10 stretched 700 m up the mountain to overcome 55 m of elevation (Wilson, 1897). The portion of the inclined plane we will view is fill material rather than excavated hillside. As you will notice, the lower end of the inclined plane is missing. It apparently fell victim to the need for building space along the floodplain of Blair Run in Foot of Ten (or, perhaps, to the ravages of Blair Run flash floods—the storm of 31 May 1889 caused considerable damage throughout central Pennsylvania). Our vantage allows us the opportunity to view a stone culvert built over a small tributary of Blair Run (Fig. 16A). This sandstone block construction gives us our first clue to the quality workmanship of early nineteenth century stonemasons. The National Park Service has stabilized and added erosion control systems to this and other structures (Garcia and Smith, 1999). The culvert at Inclined Plane No. 10 is constructed of cut coursed ashlar stone (Fig. 16B) in a radiating arch. Reconstruction required retention of all the original stone work, installation of new culvert liner pipes and flared end sections, a mat slope protection system, and the re-establishment of the top stones to their original orientation. Since railroad construction crews used local material for the railroad, it is safe to say that the fill material is probably Brallier Formation shales and siltstone (the bedrock in this area). The stone for the culvert might be from the Foreknobs or Scherr Formation, which form bedrock 1.5– 3 km to the west. It is also possible they quarried the Mahoning
Figure 16. (A) Cut stone culvert beneath Inclined Plane No. 10 near Foot of Ten. Photograph by John A. Harper. (B) Differences in ashlar stonework: left—random ashlar, made from cut stones that fit together but are not in rows; right—coursed ashlar, made from cut stones that fit together and are placed in rows. (Course refers to a row of stones.)
Allegheny Portage Railroad sandstone (Conemaugh Group, Glenshaw Formation) at the top of Allegheny Mountain and hauled it the down the Huntingdon, Cambria and Indiana Turnpike. Bedrock Geology As mentioned above, the bedrock at Inclined Plane No. 10 is Late Devonian Brallier Formation (Fig. 11). This formation consists of interbedded shale, silty shale, and siltstone of variable thickness, which are relatively easy to dig and make a good source of road material and random fill (Geyer and Wilshusen, 1982). Faill et al. (1989), using measured outcrops, calculated a thickness for the Brallier between 732 and 853 m in the Altoona area to the north. Gas wells just west of Hollidaysburg, however, indicate the thickness is actually between 366 and 549 m. The Brallier consists mostly of medium- to dark-gray, generally homogeneous, variably silty and only sparsely fossiliferous shales (constituting 50–80 percent of the formation according to Faill et al., 1989), and light- to medium-gray siltstones. Lundegard et al. (1980) characterized the Brallier as a series of turbidites having sharp planar bases and undulatory upper contacts deposited in submarine fans. The rock sequences are better explained as submarine ramp turbidites, however (Harper, 1999). The Brallier grades upward into the Scherr Formation which is distinguished by the presence of scattered beds of sandstone. An outcrop of the Brallier can be seen on the east side of Mill Road just before the entrance to the Incline Plane No. 10 access. STOP 3. ALLEGHENY PORTAGE RAILROAD NATIONAL HISTORIC SITE Feel free to roam through the Visitors Center, visit the “sleeper” quarries, Inclined Plane No. 6 (which overcame 81 m of elevation over its 827-m length; Wilson, 1897), Engine House No. 6, the Lemon House, the Skew Arch Bridge, and the Allegheny Portage Railroad Memorial. Hiking trails and the grasscovered inclined plane allow easy access to all of these features. PLEASE BE EXTRA CAREFUL crossing Old U.S. Route 22 when visiting the Skew Arch Bridge and railroad memorial. Traffic coming uphill (from the left) is often very fast, so be sure to look that way before crossing.
131
tives used on the railroad, stands in one corner. The Visitors Center also has a 60-seat auditorium where they show a 20-minute orientation film about the APR. Other amenities include a bookstore, restrooms, pay phones, and a water fountain. Access to the outside historic exhibits is easily obtained on a wheelchairaccessible boardwalk. Engine House Six Interpretive shelter includes the excavated original engine house foundation, as well as full-scale models of stationary steam engines and boilers used in raising and lowering cars on the inclined plane. Informative and interactive exhibits help explain the workings. If park staff is present, they might give you a demonstration. The Lemon House (Fig. 17) is an historic hotel and tavern on the railroad. It was a common rest and dining stop for railroad passengers. The National Park Service has restored this historic tavern’s first floor to how it would have looked in 1840, using both reproduction and period furnishings. Docents can explain about the social and economic aspects of the railroad in the furnished tavern’s bar room, dining room, parlor, and exhibit area. Bedrock and Surficial Geology The stonemasons hired to cut stones and construct “sleepers,” culverts, engine house foundations, and the Skew Arch Bridge did much of their work within an easy walk of the Visitors Center. In fact, a leisurely stroll down the boardwalk will provide a good view of the many small pits where the stone was extracted and cut. The bedrock at the park consists of the upper part of the Allegheny Formation and the Glenshaw Formation, the lower part of the Conemaugh Group (Figs. 11 and 18). The Allegheny and Glenshaw formations both comprise variable sequences of rock types consisting predominantly of medium-gray silt shale and dark-gray clay shale with some light-gray sandstone, and red
Introduction The National Park Service established the Allegheny Portage Railroad National Historical Site in 1964 to commemorate the first railroad crossing of the Allegheny Mountains. The park protects many of the remnants of the railroad, including Inclined Plane No. 10, Inclined Plane No. 8, the features associated with Inclined Plane No. 6, and the Staple Bend Tunnel ~6 km east of Johnstown. The Visitors Center has informative and educational exhibits, including artifacts and models that help tell the story of the railroad. A full-scale model of the Lafayette, one of the locomo-
Figure 17. Restored Lemon House at the Allegheny Portage Railroad National Historic Site. Photograph by John A. Harper.
132
Harper
and gray claystone. The Allegheny Formation contains numerous, thick, mineable coal seams such as the Upper Freeport and Lower Kittanning, whereas only thin, generally unmineable coals occur sparsely throughout the Glenshaw Formation. Nonmarine limestones are common in the Allegheny but rare in the Glenshaw. The Glenshaw has four or five marine zones (Fig. 18), typically consisting of a central argillaceous limestone surrounded by fossiliferous shales, whereas Allegheny marine zones typically
are more brackish and sparely fossiliferous (the major exception being the Vanport Limestone). Inclined Plane No. 6 crosses the Upper Freeport coal (the formation boundary) ~31 m below the engine house. Glover (1990) showed that the Upper Freeport has been mined extensively throughout the area, including beneath the park. A hike of only 0.8 km northeast or southwest from Old U.S. Route 22 will bring you to Upper Freeport strip mines. An old coal mine shaft
Figure 18. Generalized stratigraphic section of the upper part of the Allegheny Formation and the Glenshaw Formation (lower half of the Conemaugh Group). Modified from Harper and Laughrey (1987).
Allegheny Portage Railroad supposedly exists across the railroad from the Lemon House. Although a shaft is no longer visible, there are shallow depressions in the low bank on the northeast side of the railroad, nestled among the evergreen trees, that might represent remnant shaft or mine-collapse features. Most of the area downhill from the Visitors Center is covered with a surficial deposit of regolith and colluvium consisting of sandstone boulders in an unsorted mixture of unconsolidated material. The boulders are Mahoning sandstone (lowermost Glenshaw) (Fig. 18), which is well developed in this area (Geyer and Wilshusen, 1982 indicated that it is well developed in the subsurface at Cresson, just west of the park). It is also conspicuous in the Patton area of north-central Cambria County where it forms a flat surface of considerable extent (Stone, 1932). It is a heavy, coarse, and, in places, conglomeratic sandstone that typically occurs in two sequences ~11 or 12 m thick each, separated by shales, a minor coal, and/or sometimes a thin nonmarine limestone. In the subsurface of western Pennsylvania it often has no recognizable parting at all, forming one mass of sandstone up to 30 m thick. Where exposed at the surface, it tends to break down into abundant boulders that seem to be strewn across the landscape in helter-skelter fashion. In Cambria County, such accumulations occur in the Patton area, at Cassandra (near Inclined Plane No. 3), and here at the park. Stone (1932) indicated that some blocks are 3 m across and provided an excellent source of local building stone. It would have been relatively easy, even in the 1830s, to sift through the boulder field here at the edge of Allegheny Mountain to find suitable stone to cut. Feel free to examine the boulder quarries, both along the boardwalk and down the slope from the engine house. One of the exhibits along the boardwalk illustrates how the stonemasons shaped the boulders into cut stone. This exhibit stands beside a boulder that still shows the marks of the early nineteenth century stonecutter’s trade.
133
that reads, “This tablet erected in 1928 by the Blair County Historical Society of Pennsylvania to perpetuate this Skew Arch built in 1832-33 to carry the Huntingdon-Blairsville section of the modern turnpike over Inclined Plane No. 6 of the Allegheny Portage Railroad.” The site also contains a monument erected in 1934 to commemorate the centennial of the completion of the APR (Fig. 19B). The monument originally was 3 m high, erected from “sleepers” from the APR. It was unveiled on 1 October 1929 in the presence of a large assemblage of PRR and Pennsylvania State officials, historians, and other interested persons. The site for the monument occupies ~2020 m2, purchased by the Blair County Historical Society from James Glass and wife in 1928. The society presented the monument to the Commonwealth of Pennsylvania at the unveiling, but maintained the custodianship.
Skew Arch Bridge and Portage Railroad Memorial Down the inclined plane and across the westbound lanes of Old U.S. Route 22 are the Skew Arch Bridge and the lower half of the inclined plane (Fig. 19A). When the APR was being built, it became obvious that the Huntingdon, Cambria and Indiana Turnpike would cross the railroad about midway along Inclined Plane No. 6. The Skew Arch Bridge was built in 1832–1833 to carry the turnpike traffic over the railroad (Hoenstine, 1952). They built the bridge of coursed ashlar (Fig. 16B) laid diagonally across the inclined plane without mortar. The stonework, which is undoubtedly Mahoning sandstone from the quarries upslope, is held firmly in place by keystones. The bridge was in continual for over 100 years until a new concrete road, now the eastbound lanes of Old U.S. Route 22, was built. Amazingly, the bridge is still in a near-perfect state of preservation. In 1928, the Blair County Historical Society erected a tablet on the bridge bearing an inscription by the Honorable Plymouth W. Snyder
Figure 19. Scenes from the Allegheny Portage Railroad National Historic Site. Photographs by John A. Harper. (A) Skew Arch Bridge. (B) Portage Railroad Memorial.
134
Harper
STOP 4. JOHNSTOWN FLOOD NATIONAL MEMORIAL The Johnstown Flood National Memorial is part of the National Park System. It is open year-round except for New Year’s Day, Martin Luther King Day, Washington’s Birthday, Veterans Day, Thanksgiving, and Christmas. It has a Visitors Center with restrooms, programs, and a book sales area. A film entitled Black Friday, shown daily at the Visitors Center, helps recreate the Johnstown Flood of 1889. The Park Service offers a variety of talks, tours, and other programs during the summer months (you can obtain a complete schedule of daily activities at the Visitors Center). The Visitors Center also features multimedia exhibits, including a fiber-optic map, which describes the path of the flood. Other exhibits tell the story of the fabled South Fork Fishing and Hunting Club. The park has walking trails to the north and south abutments of the South Fork Dam. There is a picnic area located near the south abutment available for public use (we will pass the access road to the picnic area on the way to South Fork). The Great Johnstown Flood The Commonwealth of Pennsylvania began building the South Fork Dam (Fig. 20A) across South Fork Creek in 1839 to supplement the Pennsylvania Mainline Canal during periods of drought in the Conemaugh River and its tributaries. The dam was 22 m high and more than 274 m long (Francis et al., 1891)— at the time, it was the world’s largest earthfill dam, holding back the world’s largest manmade lake, 2.7 million cubic ft of water (WGBH Educational Foundation, 2001). It had a spillway that, at its narrowest, was ~21 m wide, and five cast-iron pipes that discharged through an arched stone culvert at the base of the dam (Francis et al., 1891). In the original design, these sluice
Figure 20. Illustration of South Fork Dam and Lake Conemaugh before 1889 (from Johnstown Area Heritage Association, 2005).
(feeder) pipes were to be used to supplement water level in the canal during low flow periods. They could also be used to control the lake level. The dam cost $166,647 and was finished in 1852, just in time for the demise of the Pennsylvania Mainline Canal system. When the PRR took over the western division of the canal system, they had no use for the dam and lake. The cast-iron pipes were removed and sold as scrap, while the dam was allowed to deteriorate without any thought as to possible future problems. Benjamin Ruff bought the property from the PRR in 1879, hoping to interest the wealthy families of western Pennsylvania into investing in a private summer resort where they could escape the hectic city life and relax in the mountains (Wichterman, 1998). South Fork Creek was considered one of the best trout streams in the state, and the lake would have provided a wealth of recreational opportunities for those with the cash to afford it. Ruff sold shares in the resort to people such as Andrew Carnegie, Henry Clay Frick, Andrew Mellon, and many other Pittsburgh entrepreneurs. The group built a 47-room clubhouse and elaborate Victorian palaces, which they referred to as “cottages,” stocked the lake with bass, trout, and other game fish, and called themselves the South Fork Fishing and Hunting Club. The lake they dubbed Lake Conemaugh. When the club bought the old reservoir, the dam was in need of major repairs. A section ~49 m long had collapsed (Johnson, 1889), creating a sag in the top of the dam, and this was rebuilt for $17,000. At one point, the club members discovered that the overflow pipe was not large enough to handle the frequent storms in the region, so they cut 1.5 m of rock away in order to increase the mouth of the lake. John Fulton, a geologist and engineer who was also the general manager of Cambria Iron Works in Johnstown, evaluated the South Fork Dam in November of 1880 and found it to be in very bad shape (Richardson, 2002). Johnson (1889) reported that a local mine owner, whose mine was adjacent to the lake, often inspected the dam with his engineer in tow. They reported that the dam was constructed of shale and clay, and that only straw was used to stop water leaks during reconstruction work. Not surprisingly, the engineer hired by the club members pronounced the dam perfectly safe. With all their great wealth, the members of the club adamantly refused to pay to have the dam and lake professionally repaired. No one in a position of authority seriously considered warnings that the dam was not strong enough for the water it retained. It was, after all, 82 m wide at the base as originally designed, and the top, 21 m above, was ~6 m thick (Francis et al., 1891). Thus, only minor repairs were made, and other changes to improve the attractiveness of the vacation spot actually worsened the situation: (1) the remains of the culvert used to control the lake level was blocked, and the water was allowed to rise to the level of the spillway; (2) a road was constructed across the dam wide enough for carriages to pass over two abreast, but in order to facilitate the road, the dam actually had to be lowered; and (3) the spillway stood only ~1.5 m below the lowest section of the dam where the sag occurred. The South
Allegheny Portage Railroad Fork Dam and Lake Conemaugh became a major disaster just waiting to happen. And happen it did. A moderate rain that hit the Ohio Valley on 30 and 31 May 1889 dropped only 5–10 cm of rain on the Pittsburgh area, but orographic effects due to the mountains in the Allegheny Mountain Section intensified the storm in the Johnstown area and points east (Kaktins and Fry, 1989). According to the local weather observer, Johnstown had already gotten 5.8 cm of rain by 11 a.m. of 31 May (Blodget, 1890). There was already some flooding in Johnstown, but as this was not unusual, no one thought much of it. No one knows exactly how much total precipitation occurred that night because the Johnstown station observer was one of the great Johnstown Flood’s casualties. The Franklin Institute estimated that the City of Johnstown received 13–15 cm of rain during the 26-hour storm duration (Kaktins and Fry, 1989). Interestingly, the storm hit the area east of Allegheny Mountain even harder than Johnstown, dumping 20.8 cm on Harrisburg, 22.84 cm on McConnellsburg, and causing record flooding in the Juniata River valley. Kaktins and Fry (1989) reported that floods also played havoc in the James and Potomac River basins in Maryland and Virginia, and in the upper Allegheny River and Genesee River in New York. Yet, across an area greater than 31,000 km2, the Johnstown Flood remains in our collective memories. All of that rain hitting the Johnstown area raised the lake level and caused debris to clog the spillway, thus reducing its effectiveness even more. At just before 3:00 p.m., the water spilling over the center sag of the dam caused the dam to fail, sending 18,000,000 t of water sluicing down the narrow Little Conemaugh River valley at a peak discharge rate of ~9000 m3/sec (Coleman et al., 2009). The valley walls are steep, so the floodwaters had nowhere to go but downstream. The leading edge of the flood swelled with enormous chunks of debris and churned toward Johnstown at 64 km/h. To get an idea of the force of the flood, imagine if Niagara Falls were suddenly turned into the Little Conemaugh River valley in the very short period of time it must have taken for the lake to empty. Thousands of people were swept up in a slurry of dirty water, mud, and tonnes of grinding, crushing debris that has been described as a mountain of rubbish moving at the speed of a locomotive. Nothing stood in its way; even the Conemaugh Viaduct, constructed to carry the APR over the Little Conemaugh River at the entrenched meander loop west of South Fork, was swept away. When the floodwall hit Johnstown, everything in its path was destroyed. The wall only stopped moving forward when it hit the side of Laurel Hill, causing a wave of water to back up into the Stoneycreek River, destroying many of the buildings in the floodplain south of Johnstown as well. Amazingly, the flood lasted only ~10 minutes in Johnstown; but the flood was just the beginning of the disaster. The Old Stone Bridge of the PRR just downstream of the convergence of the Little Conemaugh and Stoneycreek rivers stood its ground. Much of the debris, which included dead horses, trees, railroad cars and locomotives, houses, and other things the flood had
135
destroyed, got caught up against the arches of the bridge and created a huge dam ~9 m high (Uldis Kaktins, 2010, written commun.). The floodwaters spread out over 0.12 km2, trapping numerous victims of the flood. Before anyone could be rescued, the debris caught fire and those who had escaped the ravages of the flood, only to be caught in the debris, were soon burned alive. Many of the bodies recovered from the devastation were never identified, and many hundreds of people went missing and were never found. In all, more than 2,200 people died, thousands were left homeless, and the once prospering city of Johnstown became a debris-clogged wasteland. The cleanup took years, with virtually no assistance from the wealthy people who were directly responsible for the disaster. Johnstown eventually rebuilt, but it took five years for the city to recover from the devastation. With the reservoir drained (Fig. 20B), the South Fork Fishing and Hunting Club disbanded and the wealthy families returned to their main residences with little thought of the suffering and death they caused by their negligence. Andrew Carnegie was the only one who contributed to Johnstown’s rebuilding, providing the money to build the town’s library. Of the “cottages” and the clubhouse, most can still be seen along Main Street, just uphill from Locust Street (PA Route 896) in St. Michael, the small town that now stands on the southwest side of the former lake. As we exit the Johnstown Flood National Memorial, the bus will stop briefly at the bottom of the hill to allow attendees to view and photograph the remains of the breached dam (Fig. 21). For the complete story of the Johnstown Flood, see McCullough (1968) and Law (1997). I also recommend Johnson (1889), who provided an early, florid, account of the flood. Just be prepared for some good old-fashioned, late Victorian, passionate and very opinionated purple prose.
Figure 21. The breached South Fork Dam as it appears today. The bridge in the background carries U.S. Route 219 over the South Fork of the Little Conemaugh River. Photograph by John A. Harper.
136
Harper
STOP 5. THE CONEMAUGH VIADUCT From this vantage point we can look west to the great entrenched meander of the Little Conemaugh River and the place where the Conemaugh Viaduct used to be. The Conemaugh Viaduct, built for the APR in 1832–1833, was one of the earliest railway bridges in the country. When the railroad reached Horseshoe Bend on the Little Conemaugh River, an entrenched meander loop (Fig. 22), the railroad crossed the stream and cut across the narrow neck of the loop, thereby saving more than 2.4 km of distance (the length of the loop). This required a rather high bridge, which turned out to be an imposing structure (Fig. 23). Solomon W. Roberts designed and supervised its erection by a stonemason named John Durno (Roberts, 1878). Originally, the bridge was to have two arches spanning 15 m each, but the plans were altered and the final structure was 24 m high with a single semicircular arch 12 m high and spanning 24 m. The arch was 1.1 m thick at the springing line, and 0.9 m at the crown. The embankment at the end of the viaduct was 20 m high. The foundation was built on timber on one end and on rock on the other (Wilson, 1897). From the foundation to the springing line of the
arch the abutment walls were 8.8 m high. The viaduct was 8.5 m wide at the top of the parapets and 12 m wide at the foundation. The arch stones were cut from large erratic blocks of lightcolored sandstone that were found lying in the woods nearby. These were probably Pottsville Formation sandstone blocks—the Pottsville crops out on the hillside where the river cuts through the Ebensburg anticline (Fig. 22), and appears to have been quarried near the viaduct. The backing stone for the bridge was cut from Loyalhanna Formation sandy limestone, which also crops out near the site. The joints of the face stones were pointed using mortar made from the Loyalhanna without the need for adding sand to the mix (Roberts, 1878). The viaduct was completed in the early spring of 1833 at a total cost of $54,562.54 (Wilson, 1897). It stood solid, in constant use for 55 years by the APR and the PRR until it was destroyed on 31 May 1889 by the Johnstown Flood. A previous flood, in October 1847, caused the bed of the Little Conemaugh River to wash out beneath the foundations of the viaduct. The railroad attempted to repair the damage by building a dam downstream and filling the washed out areas with brush and stone. Wilson (in Baumgardner and Hoenstine, 1952)
Figure 22. Topography and geologic structure of a portion of the Geistown and Nanty Glo 7½-minute topographic quadrangles, showing the locations of the entrenched meander loop of the Little Conemaugh River and the locations of the bygone Conemaugh Viaduct and Staple Bend Tunnel.
Allegheny Portage Railroad speculated that these actions may have ultimately assisted in destroying the viaduct. When the huge wave of debris-clogged water, later called the Johnstown Flood, came roaring down the Little Conemaugh River valley, the first obstacle it struck was the Conemaugh Viaduct. Wilson (1897) reported that the weight of the water was estimated to be 18,000,000 t, moving at a velocity of 24 km/h, in a narrow gorge with a fall of 280 m/km (a 1 percent grade). When it hit the viaduct, the water dammed up against it to the depth of 24 m (Uldis Kaktins, 2010, written commun.). The viaduct temporarily stood solid, blocking the flood, which backed up in the river valley. When the viaduct collapsed and was swept away, the backed-up flood surged forward at a peak discharge on the order of 12,000 m3/sec (Coleman et al., 2009), washing out the PRR for several km. It also swept away the village of Mineral Point. It is hard to believe the flood didn’t cut the neck of the meander loop and create a new channel. The PRR quickly built a temporary wooden trestle and eventually replaced it with the present bridge over the Little Conemaugh River at the location of the Conemaugh Viaduct, but this one has stood for over 110 years. STOP 6. JOHNSTOWN AND THE JOHNSTOWN CANAL BASIN The City Conemaugh, the original name for Johnstown, is a corruption of the Delaware Indian word(s) Connu-macht, meaning “Otter Creek.” The Delaware Indians occupied a town situated at
Figure 23. Illustration of the Conemaugh Viaduct (from Johnstown Area Heritage Association, 2005). Built to carry the Allegheny Portage Railroad over the Little Conemaugh River, this historic structure was one of the victims of the great Johnstown Flood of 1889.
137
the forks of the Conemaugh River long before European settlers entered this area in the late 1700s. As the Indian name suggests, the river once had a pristine quality that was lost at the end of the 1800s when large-scale coal mining began, allowing the degradation of water by abandoned mine drainage. Settlers of European descent first came to this area immediately before the French and Indian War in the mid-1700s. The Allegheny Front had long been the treaty boundary between the Native American tribes and the settlers, but, as was typical of the early settlers, the natives were driven out and the area west of Allegheny Mountain was opened for settlement. Joseph Johns laid out the town he called Conemaugh, located at the confluence of the Little Conemaugh and Stoneycreek rivers where they form the Conemaugh River, in 1800. The town was incorporated in 1831, and in 1834 it was renamed Johnstown in honor of its founder. The city had a major growth spurt during the mid-1880s as a result of iron and steel manufacturing. This was the home of Andrew Carnegie’s Cambria Iron Works, which later became Bethlehem Steel. The surrounding hills had a seemingly endless supply of coal, and iron ore and limestone could also be found. But its greatest resource was the water of its rivers—water for manufacturing, water for drinking, water for transportation. Despite its manufacturing history, Johnstown is probably most famous today for its lack of foresight and the disastrous consequences that resulted. Johnstown is defined by its topography—the city lies within a 3-km-wide floodplain formed by three rivers draining 88 km2 of upland, with one of the deepest river gaps in the eastern United States (Reese, 2008) as their only egress from the region. In the early settlement years, the floodplains (bottomlands) were used for agriculture. The settlers drained the wetlands to make use of the organic-rich soils and establish numerous fertile fields, which removed the natural sumps for storing storm water runoff. In addition, dams built to hold large volumes of water for commercial uses increased evaporation and reduced the flow of clean water into the system. The subsequent discovery of coal, iron ore, and limestone in the surrounding hills, along with the construction of the APR and PRR, helped establish Johnstown as a thriving manufacturing and transportation center. People flocked to Johnstown where they built houses, stores, churches, and other buildings on the former wetlands-turned-farmland. The already overused soil became compacted and covered with structures, macadam, concrete, paving stone, and a variety of other impervious materials that increased runoff even more. Although flooding was an annual event before the land was modified, the Native Americans learned to live with it and use it. The paving of Johnstown resulted in increased flooding in the valley, much of it flash flooding during storms. People came to expect getting their feet wet. But even the devastation caused by the anomalous floods of 1889, 1936, and 1977 (Kaktins and Fry, 1989) couldn’t deter people from building in the floodplains. It is likely that Johnstown will continue to have floods, possibly even larger, more devastating ones as a result of global warming. There are, after all, still several large dammed reservoirs in
138
Harper
the hills around Johnstown. A modern engineer would tell you that these dams are safe, and that failures such as occurred at South Fork and Laurel Run in 1889 and 1977, respectively, are highly unlikely to occur. But in these turbulent times, anything is possible. And even the concrete flood walls on the Conemaugh River can provide only a measure of protection. The only solution to avoiding the devastation of future large-magnitude floods is to retreat to the hillsides. The Canal Basin The APR reached its final destination (on westward runs) at the canal basin in Johnstown. This basin was situated along the Little Conemaugh River between the intersection of Clinton and Railroad streets on the west and Church and Railroad streets on the east. The canal basin (Fig. 24) was essentially enclosed by the city to the south and to the north by an island, officially known as Long Island, but commonly called simply “The Island” (Storey, 1907). The Island was ~90 m wide at the widest point. A smaller island called Goose Island, ~30 m wide, lay to the west of The Island. Goose Island acted as the north wall of the Johnstown canal (A in Fig. 24). The two islands were separated by a waste weir (B), which acted as a sluice for the basin. A feeder channel from the upstream side of the Little Conemaugh River entered
the basin through a sluice at the northeast end of The Island (C). A second feeder stream, from Suppes’ Dam on the Stoneycreek River south of the city, came into the basin at the present-day location of Feeder Street (D). It separated the city into Johnstown proper on the west and Conemaugh on the east. The APR ran between Railroad Street and the basin on the mainland side and between Portage Street and the basin on The Island. The area of Conemaugh at the northeast end of The Island was known as Five Points because Portage, Railroad, Church, and Depot Streets, and the APR, converged at that location. This was where railroad, road, and canal met to continue the Pennsylvania Mainline Canal (Storey, 1907). The canal basin was shaped like a semicircle. On the south, it started at the packet slip where Canal Street (now called Washington Street) and Clinton Street intersected and followed the curve of Railroad Street to Depot Street at the northeast end of The Island. On the north, it paralleled Portage Street in a straight line to the waste weir. It was 549 m long and 183 m wide at the widest point (at Singer Street) (Fig. 24). Both sides of the basin were lined with warehouses and docks, or slips. Each warehouse had a slip, generally 4.6 m wide by 24 m long, long enough for two boats. One could be loaded while the other was being unloaded. Warehouses extended into the basin from either Railroad Street on the south or Portage Street on the north. Each occupied a strip of land ~30 m long and 23 m wide.
Figure 24. Map of the Johnstown Canal Basin, showing the locations of streets, railroads, and other features (based on Johnstown Area Heritage Association, 2005).
Allegheny Portage Railroad One or two sidings extended from the railroad to the rear end of each dock. At the southwest end of The Island, the Overhead Bridge (E, Fig. 24) extended from Canal Street, across the canal and the south end of the waste weir, to The Island. It provided access to the freight houses and piers on The Island. It was 91 m long and wide enough for wagons to pass. It sat on a pier on the Canal Street side, and gradually descended to the level of The Island at Portage Street (Storey, 1907). Prior to construction of the bridge in 1835, the only ways to reach The Island and Goose Island from the mainland were by driving a team across the bed of the Little Conemaugh River or by driving all the way up to Five Points. Unfortunately, the former was difficult at best and at worst, during high water, impossible. The Johnstown canal (A, Fig. 24) was ~18 m wide and contained at least 1.2 m of water. At the basin end of the canal on the north side, beneath the Overhead Bridge, there was a weigh lock (F) with weigh scales. The purpose of the waste weir (B) was to control the necessary amount of water in the basin to float the canal boats, and to make it easier to repair them. If repairs were necessary, water could be let out of the basin. At the northwest end of the canal, an aqueduct carried the canal boats across the Little Conemaugh River to a canal that crossed the floodplain between the Little Conemaugh and Conemaugh rivers. Up until 1835, all canal boats were weighed in Pittsburgh. In 1835, Johnstown got its own weigh lock (F, Fig. 24; Fig. 25). This allowed the Commonwealth to determine if there were any sneaky business going on between Hollidaysburg and Johnstown, and between Johnstown and Pittsburgh. After a boat had been weighed and its freight accounted for, the captain paid the toll at the collector’s office and received his “clearance papers,” which allowed the boat to continue its trip on the canal. At various places between Johnstown and Pittsburgh the captain had to show his “clearance papers” to state officials. This helped prevent the smuggling of freight onto the boats between weigh
139
locks. The procedure for weighing a boat was the same here as it was at Hollidaysburg. The era of the Johnstown canal basin came to a close around 1851 when the PRR started hauling freight between Johnstown and Pittsburgh. The railroad was faster and less expensive than the canal, and would run even during times of low water. When the Pennsylvania Mainline Canal was finally abandoned, Principal Engineer Sylvester Welch and Samuel Jones, who owned The Island, donated Portage Street to the city. There’s not much left of the canal basin in Johnstown today. The basin was filled in and Bethlehem Steel built a foundry at the site that encompasses The Island and Goose Island. Much of what was once Portage Street is now occupied by PA Route 271. Canal Street is now called Washington Street, and several other streets shown in Figure 24 have been given different names as well. The aqueduct over the Little Conemaugh River was replaced by a steel railroad trestle and the canal by railroad tracks. Truth or Consequences Shortly after the canal was built, the canal engineers realized that there would be times during the dry summers when there would be too little water in the Little Conemaugh River to keep the basin in operation. The Commonwealth began building a dam on the South Fork in 1835. However, the finances ran out and the project was abandoned for a few years. It was finally completed in 1845 and dammed up South Fork as far as 4.8 km. The South Fork Dam broke in 1847, causing considerable damage to the canal and basin in Johnstown. Two smaller breaks occurred in 1862, but they didn’t cause any serious damage. By this time, the PRR had dismantled the canal system and abandoned the dam and reservoir. They were purchased and the South Fork Fishing Club rebuilt the dam (more or less). Then came 31 May 1998! (Refer to Stop 4 for more details.) The Johnstown Flood Museum sits at the corner of Washington (formerly Canal) Street and Walnut Street. It is an impressive building, built in 1891 as a library. Andrew Carnegie gave Johnstown the money to build the library, supposedly to assuage his guilty conscience for having been a member of the South Fork Fishing and Hunting Club, whose failure to maintain the South Fork Dam essentially wiped Johnstown off the face of the earth. The museum has exhibits, displays, photographs, and extensive archives (which can be researched by appointment only). The museum also tells the stories of the 1936 and 1977 floods. ACKNOWLEDGMENTS
Figure 25. Illustration of the weigh lock at the Johnstown Canal Basin (from Johnstown Area Heritage Association, 2005).
This field guide benefitted immeasurably from reviews by Dr. Uldis Kaktins and Dr. William R. Brice, University of Pittsburgh at Johnstown (emeritus), and by Ms. Kristin M. Carter, Pennsylvania Geological Survey. It is based largely on a field trip organized for the Pittsburgh Geological Society in 2002. I thank the Society members who attended that field trip for their feedback, which assisted with the compilation of this field guide.
140
Harper
REFERENCES CITED Ambrose, S.E., 1996, Undaunted Courage: Meriwether Lewis, Thomas Jefferson, and the Opening of the American West: New York, Touchstone Books, 521 p. American Canal Society, 2004, The American canal guide: A bicentennial inventory of America’s historic canal resources, http://www.americancanals .org/index.html. Baumgardner, M.J., and Hoenstine, F.G., 1952, The Allegheny Old Portage Railroad, 1834–1854: Building, operation and travel between Hollidaysburg and Johnstown, Pennsylvania: Blair County Chapter, S.A.R., and Cambria County Chapter, S.A.R., 90 p. Berg, T.M., and Dodge, C.M., compilers and editors, 1981, Atlas of preliminary geologic quadrangle maps of Pennsylvania: Pennsylvania Geological Survey, 4th ser., General Geology Report 61, 636 p. Blodget, L., 1890, The floods in Pennsylvania, May 31 and June 1, 1889: Commonwealth of Pennsylvania, Annual Report of the Secretary of Internal Affairs, Part I, p. A143–A149. Brice, W.R., 1989, Early geological explorations of the Johnstown area, in Harper, J.A., ed., Geology in the Laurel Highlands of southwestern Pennsylvania: Guidebook, 54th Annual Field Conference of Pennsylvania Geologists, Johnstown, Pennsylvania, p. 121–138. Briggs, R.P., 1998, Conquest of the Allegheny Mountains in Pennsylvania: The engineering geology of Forbes Road, 1758–1764: Environmental & Engineering Geoscience, v. IV, no. 3, p. 397–414. Butts, C., 1945, Description of the Hollidaysburg and Huntingdon quadrangles: U.S. Geological Survey, Folio 227, 20 p. Canich, M.R., and Gold, D.P., 1985, Structural features in the Tyrone–Mount Union lineament, across the Nittany anticlinorium in central Pennsylvania, in Gold, D.P., ed., Central Pennsylvania geology revisited: Guidebook, 50th Annual Field Conference of Pennsylvania Geologists, State College, Pennsylvania, p. 120–137. Clemson, T.G., 1835, Analysis of the minerals accompanying Mr. E. Miller’s donation: Transactions of the Geological Society of Pennsylvania, v. 1, p. 271–274. Coleman, N.M., Davis Todd, C., Myers, R.A., and Kaktins, U., 2009, Johnstown flood of 1889—Destruction and rebirth: Geological Society of America Abstracts with Programs, v. 41, no. 7, p. 216. Conrad, T.A., 1835, Description of five new species of fossil shells in the collection presented by Mr. Edward Miller to the Geological Society: Transactions of the Geological Society of Pennsylvania, v. 1, p. 267–270. Davis, T.S., 1931, A history of Blair County, Pennsylvania: Harrisburg, National Historical Association, Inc., 322 p. Dickens, C., 1842, American Notes for General Circulation, reprinted as p. 14–254 in Schwarzbach, F.S., ed., 1997, American Notes and Pictures of Italy: Vermont, Everyman, Charles E. Tuttle Co., Inc., 511 p. Epstein, J.B., Sevon, W.D., and Glaeser, J.D., 1974, Geology and mineral resources of the Lehighton and Palmerton quadrangles, Carbon and Northampton Counties, Pennsylvania: Pennsylvania Geological Survey, 4th ser., Atlas 195cd, 460 p. Faill, R.T., 1985, The Acadian orogeny and the Catskill delta, in Woodrow, D.L., and Sevon, W.S., eds., The Catskill Delta: Geological Society of America Special Paper 201, p. 15–17. Faill, R.T., 1987, The Birmingham window; Alleghanian décollement tectonics in the Cambrian-Ordovician succession of the Appalachian Valley and Ridge Province, Birmingham, Pennsylvania: Geological Society of America Centennial Field Guide, Northeastern Section: Field Trip 10, p. 37–42. Faill, R.T., 1999, Chapter 33, Paleozoic, in Shultz, C.H., ed., The Geology of Pennsylvania, Part VI: Geologic History: Pennsylvania Geological Survey, 4th ser., Special Publication 1, p. 418–433. Faill, R.T., and Nickelsen, R.P., 1999, Chapter 19, Appalachian Mountain Section of the Ridge and Valley Province, in Shultz, C.H., ed., The Geology of Pennsylvania, Part III, Structural Geology and Tectonics: Pennsylvania Geological Survey, 4th ser., Special Publication 1, p. 269–285. Faill, R.T., Glover, A.D., and Way, J.H., 1989, Geology and mineral resources of the Blandburg, Tipton, Altoona, and Bellwood quadrangles, Blair, Cambria, Clearfield, and Centre Counties, Pennsylvania: Pennsylvania Geological Survey, 4th ser., Atlas 86, 209 p. Francis, J.B., Worthen, W.E., Becker, M.J., and Fteley, A., 1891, Report of the Committee on the cause of the failure of the South Fork Dam: Transactions of the American Society of Civil Engineers, v. 24, p. 431–469.
Frey, M.G., 1973, Influence of Salina salt on structure in New York– Pennsylvania part of Appalachian Plateau: The American Association of Petroleum Geologists Bulletin, v. 57, p. 1027–1037. G.A.I. Consultants, 1999, Final Kiski-Conemaugh River Basin conservation plan: Windber, Pennsylvania, Kiski-Conemaugh River Basin Alliance and Environmental Information Services, variously paginated. Garcia, D.M., and Smith, N.L., 1999, Allegheny Portage Railroad: New support for old arches: National Park Service, Cultural Resource Management, v. 22, no. 10, p. 44–46, http://crm.cr.nps.gov/archive/22-10/22-10-17.pdf. Geyer, A.R., and Wilshusen, J.P., 1982, Engineering characteristics of the rocks of Pennsylvania, 2nd ed.: Pennsylvania Geological Survey, 4th ser., Environmental Geology Report 1, 300 p. Glover, A.D., 1990, Coal resources of Cambria and Blair Counties, Pennsylvania. Part 1. Coal crop lines, mined-out areas, and structure contours: Pennsylvania Geological Survey, 4th ser., Mineral Resource Report 96, 129 p. Gold, D.P., 1999, Chapter 22, Lineaments and their interregional relationships, in Shultz, C.H., ed., The Geology of Pennsylvania, Part III: Structural Geology and Tectonics: Pennsylvania Geological Survey, 4th ser., Special Publication 1, p. 418–433. Gwinn, V.E., 1964, Thin-skinned tectonics in the Plateau and northwestern Valley and Ridge Provinces of the Central Appalachians: Geological Society of America Bulletin, v. 75, p. 863–900, doi:10.1130/0016 -7606(1964)75[863:TTITPA]2.0.CO;2. Gwinn, V.E., 1970, Kinematic patterns and estimates of lateral shortening, Valley and Ridge and Great Valley Provinces, Central Appalachians, southcentral Pennsylvania, in Fisher, G.W., Pettijohn, F.J., Reed, J.C., and Weaver, K.N., eds., Studies of Appalachian Geology: Central and Southern: New York, Interscience Publishers, p. 127–146. Harlan, R., 1835, Notice of fossil vegetable remains from the bituminous Coal Measures of Pennsylvania, being a portion of the illustrative specimens accompanying Mr. Miller’s essay or geological section of the Alleghany Mountain, near the Portage Railway: Transactions of the Geological Society of Pennsylvania, v. 1, p. 256–259. Harper, J.A., 1989, Effects of recurrent tectonic patterns on the occurrence and development of oil and gas resources in western Pennsylvania: Northeastern Geology, v. 11, p. 225–245. Harper, J.A., 1999, Chapter 7: Devonian, in Shultz, C.H., ed., The Geology of Pennsylvania, Part II, Stratigraphy: Pennsylvania Geological Survey, 4th ser., Special Publication 1, p. 108–127. Harper, J.A., and Laughrey, C.D., 1987, Geology of the oil and gas fields of southwestern Pennsylvania: Pennsylvania Geological Survey, 4th ser., Mineral Resource Report 87, 166 p. Hoenstine, F.G., 1952, The Skew Arch Bridge and Old Portage Railroad Monument, in Baumgardner, M.J., and Hoenstine, F.G., The Allegheny Old Portage Railroad, 1834–1854: Building, operation and travel between Hollidaysburg and Johnstown, Pennsylvania: Blair County Chapter, S.A.R., and Cambria County Chapter, S.A.R., p. 84–88. Hunt, C.B., 1974, Natural Regions of the United States and Canada: San Francisco, W.H. Freeman and Company, 725 p. Inners, J.D., 1987, Upper Paleozoic stratigraphy along the Allegheny topographic front at the Horseshoe Curve, west-central Pennsylvania: Geological Society of America Centennial Field Guide, Northeastern Section, Field Trips 8 and 9, p. 29–36. Jacobs, H.A., 1945, The old Juniata Canal and Portage Railroad, in Wolf, G.A., ed., Blair County’s First Hundred Years, 1846–1946: Altoona, Pennsylvania, Mirror Press, p. 159–171. Johnson, W.F., 1889, History of the Johnstown Flood, Including All the Fearful Record; The Breaking of the South Fork Dam; The Sweeping out of the Conemaugh Valley; The Over-Throw of Johnstown; The Massing of the Wreck at the Railroad Bridge; Escapes, Rescues, Searches for Survivors and the Dead; Relief Organizations, Stupendous Charities, etc., etc. with Full Accounts also of the Destruction on the Susquehanna and Juniata Rivers, and the Bald Eagle Creek: Edgewood Publishing Co., also found at http://prr.railfan.net/documents/JohnstownFlood/. Johnstown Area Heritage Association, 2005, Johnstown Flood Museum: The Great Johnstown Flood of 1889: http://www.jaha.org/edu/flood/index.html. Juniata Clean Water Partnership, 2000, Juniata watershed management plan: http:// www.jcwp.org/final%20plan/JCWP%20Final%20Plan/jcwpplan.pdf. Kaktins, U., and Fry, H.C., 1989, The floods of Johnstown, in Harper, J.A., ed., Geology in the Laurel Highlands of southwestern Pennsylvania: Guidebook, 54th Annual Field Conference of Pennsylvania Geologists, Johnstown, Pennsylvania, p. 139–149.
Allegheny Portage Railroad Kowalik, W.S., and Gold, D.P., 1976, The use of Landsat-1 imagery in mapping lineaments in Pennsylvania, in Hodgson, R.A., et al., eds., Proceedings of the First International Conference on the New Basement Tectonics: Utah Geological Association, Publication 5, p. 236–249. Lavin, P.M., Chaffin, D.L., and Davis, W.F., 1982, Major lineaments and the Lake Erie-Maryland crustal block: Tectonics, v. 1, p. 431–440, doi:10.1029/TC001i005p00431. Law, A.S., 1997, The great flood, Johnstown, Pennsylvania, 1889: Johnstown Area Heritage Association, 106 p. Lesley, J.P., 1876, Historical sketch of geological explorations in Pennsylvania and other states: Second Geological Survey of Pennsylvania, v. A, 200 p. Lundegard, P.D., Samuels, N.D., and Pryor, W.A., 1980, Sedimentology, petrology, and gas potential of the Brallier Formation—Upper Devonian turbidite facies of the central and southern Appalachians: U.S. Department of Energy, USDOE/METC/5201-5, Morgantown, West Virginia, 220 p. McCullough, D.G., 1968, The Johnstown Flood: Simon & Schuster, New York, 302 p. Miller, B.L., 1934, Limestones of Pennsylvania: Pennsylvania Geological Survey, 4th ser., Mineral Resource Report 20, 729 p. Miller, E., 1835, Geological description of a portion of the Alleghany Mountain, illustrated by drawings and specimens: Transactions of the Geological Society of Pennsylvania, v. 1, p. 251–255. Nickelsen, R.P., 1963, Fold patterns and continuous deformation mechanisms of the central Pennsylvania folded Appalachians, in Tectonics and Cambrian-Ordovician stratigraphy, central Appalachians of Pennsylvania: Pittsburgh Geological Society and Appalachian Geological Society, Guidebook, p. 13–29. O’Neill, B.J., 1964, Atlas of Pennsylvania’s mineral resources: Part 1. Limestones and dolomites of Pennsylvania: Pennsylvania Geological Survey, 4th ser., Mineral Resource Report 50, 40 p. Reese, S.O., 2008, The geologist who went into a narrows but came out through a gorge: Pacific Geology, v. 38, no. 2/3, p. 2–12. Richardson, D., 2002, The Johnstown Flood of 1889: National Park Service, http:// www.nps.gov/archive/jofl/histiography.htm (accessed November 2010). Roberts, S.W., 1878, Reminiscences of the first railroad over the Allegheny Mountain: The Pennsylvania Magazine of History and Biography, v. 2, no. 4, p. 370–393. Rodgers, M.R., and Anderson, T.H., 1984, Tyrone–Mount Union cross-strike lineament of Pennsylvania: A major Paleozoic basement fracture and uplift boundary: The American Association of Petroleum Geologists Bulletin, v. 68, p. 92–105. Root, S.I., and Hoskins, D.M., 1977, Lat 40°N fault zone, Pennsylvania: A new interpretation: Geology, v. 5, p. 719–723, doi:10.1130/0091 -7613(1977)5<719:LNFZPA>2.0.CO;2.
141
Sevon, W.D., 2000, Physiographic provinces of Pennsylvania: Pennsylvania Geological Survey, 4th ser., Map 13, Scale 1:2,000,000. Shank, W.H., 1975, Sylvester Welch’s report on the Allegheny Portage Railroad, 1833: York, Pennsylvania, American Canal and Transportation Center, not paginated. Stevenson, D., 1838, Sketch of the Civil Engineering of North America; Comprising Remarks on the Harbours, River and Lake Navigation, Lighthouse, Steam Navigation, Waterworks, Canals, Roads, Railways, Bridges, and Other Works in that Country: London, J. Weale, 320 p. Stone, R.W., 1932, Building stones of Pennsylvania: Pennsylvania Geological Survey, 4th ser., Mineral Resource Report 15, 316 p. Storey, H.W., 1907, History of Cambria County, Pennsylvania, v. 1: New York, Lewis Publishing Co., 590 p. Swartz, F.M., 1965, Guide to the Horse Shoe Curve section between Altoona and Gallitzin, central Pennsylvania: Pennsylvania Geological Survey, 4th ser., General Geology Report 50, 56 p. Taylor, J.F., Bragonier, W.A., and Hall, F.W., 2002, Stop 3 and lunch, Chimney Rocks Park, in Way, J.H., Doden, A.G., Gold, D.P., and Fleeger, G.M., eds., Geology on the edge: Selected geology of Bedford, Blair, Cambria, and Somerset Counties: Guidebook, 68th Annual Field Conference of Pennsylvania Geologists, Altoona, Pennsylvania, p. 104–109. Thompson, J., 2010, Cable car lines in Pennsylvania: Allegheny Portage Railroad: http://www.cable-car-guy.com/images/all_port_car.jpg. Ulrich, E.O., 1911, Revision of the Paleozoic systems: Geological Society of America Bulletin, v. 22, p. 281–680. WGBH Educational Foundation, 2001, Wonders of the world databank—South Fork Dam: http://www.pbs.org/wgbh/buildingbig/wonder/structure/ south_fork.html (November 2010). Wichterman, L., 1998, The Johnstown Flood; National disaster: http://www. geocities.com/Heartland/4547/johnstown.html. Wilson, W.B., 1897, The evolution, decadence and abandonment of the Allegheny Portage Railroad, p. 36–83 in Baumgardner, M.J., and Hoenstine, F.G., 1952, The Allegheny Old Portage Railroad, 1834–1854: Building, operation and travel between Hollidaysburg and Johnstown, Pennsylvania: Blair County Chapter, S.A.R., and Cambria County Chapter, S.A.R., 90 p. [Reprinted from The Pennsylvania Railroad Men’s News, v. 9, September, 1897, p. 289–305; October, 1897, p. 317–323.] Wilson, W.B., 1899, History of the Pennsylvania Railroad Company, With Plan of Organization, Portraits of Officials and Biographical Sketches: Philadelphia, Henry T. Coates & Co., v. 1, 418 p.
MANUSCRIPT ACCEPTED BY THE SOCIETY 29 NOVEMBER 2010
Printed in the USA
The Geological Society of America Field Guide 20 2011
Early industrial geology of western Pennsylvania and eastern Ohio: Early gristmills and iron furnaces west of the Alleghenies and their geologic contexts Joseph T. Hannibal* Cleveland Museum of Natural History, 1 Wade Oval Drive, Cleveland, Ohio 44106, USA Tammie L. Gerke* Glenn A. Black Laboratory of Archaeology, Indiana University, 423 N. Fess Avenue, Bloomington, Indiana 47408, USA Mary K. McGuire* Department of Geology and Planetary Science, University of Pittsburgh, 4107 O’Hara Street, SRCC, Room 200, Pittsburgh, Pennsylvania 15260, USA Harry M. Edenborn* Geosciences Division, National Energy Technology Laboratory, U.S. Department of Energy, Pittsburgh, Pennsylvania 15236, USA Ann L. Holstein* Kelvin Smith Library, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, Ohio 44106, USA David Parker* Mercyhurst Archaeological Institute, Mercyhurst College, 501 East 38th Street, Erie, Pennsylvania 16546, USA
ABSTRACT Even before 1800, geological resources such as chert, iron, limestone, and coal were being utilized from the Pennsylvanian rocks of eastern Ohio and western Pennsylvania. These materials were of great interest to the early geologists of the region. This field trip discusses these products in the context of early grain milling, iron furnaces, and allied industries of Ohio and Pennsylvania in the late eighteenth and early nineteenth century, with a focus on two publicly accessible sites: McConnells Mill Park in western Pennsylvania, and Mill Creek Park in eastern Ohio. These parks contain publicly accessible gristmills and iron furnaces, and outcrops. We also provide new observations on cultural materials related to these industries, especially ironfurnace slag and millstones.
*Hannibal—
[email protected]; Gerke—
[email protected]; McGuire—
[email protected]; Edenborn—
[email protected]; Holstein—
[email protected]; Parker—
[email protected]. Hannibal, J.T., Gerke, T.L., McGuire, M.K., Edenborn, H.M., Holstein, A.L., and Parker, D., 2011, Early industrial geology of western Pennsylvania and eastern Ohio: Early gristmills and iron furnaces west of the Alleghenies and their geologic contexts, in Ruffolo, R.M., and Ciampaglio, C.N., eds., From the Shield to the Sea: Geological Field Trips from the 2011 Joint Meeting of the GSA Northeastern and North-Central Sections: Geological Society of America Field Guide 20, p. 143–167, doi: 10.1130/2011.0020(07). For permission to copy, contact
[email protected]. ©2011 The Geological Society of America. All rights reserved.
143
144
Hannibal et al.
INTRODUCTION American colonists began to move westward beyond the Allegheny Mountains at the end of the French and Indian War in 1763. The first industries were typically sawmills and grain mills, with products of the sawmills used to build the mills for grinding grain. Local stone was utilized for the manufacture of millstones, key components of the grain mills. Iron furnaces followed soon after, where there was a suitable source of iron ore, limestone for flux, and trees suitable for making charcoal. By the early 1800s, coal was beginning to be used to stoke these early blast furnaces. In western Pennsylvania and eastern Ohio, the same, or adjacent, Pennsylvanian rock units from the Pottsville and Allegheny Groups (Fig. 1) could contain buhrstone for manufacturing millstones, iron ore suitable for blast furnaces, clay and sandstone for use in bricks and as building stone for construction of furnaces, lime for flux, and coal for fueling the furnaces (Rogers, 1836; Hildreth, 1838; Mather, 1838). The geologists of the first Pennsylvania and Ohio geological surveys paid special attention to buhrstone for millstone, iron ore for blast furnaces, and other resources.
Mills and iron furnaces were key components—and sources of wealth—along the frontier (Harper, 1991, p. 50–53). In this guide, we provide an overview of earlier and ongoing geological and archaeological investigations of some of these early industries in western Pennsylvania and eastern Ohio west of the Allegheny Mountains, with a concentration on mills, millstones, and iron furnaces, placing recent archaeological work in its geological context. We define the Allegheny Mountains as being bordered on the east by the southwest side of the Appalachian Mountains and on the west by the Allegheny Plateau, that is the general high-elevation area delineated as the “Allegheny Mountain Section” by Way (1999) which is similar (at least on its western border) to the “Allegheny ‘Mountains’” section of Murphy and Murphy (1937, fig. 4). Western Pennsylvania and eastern Ohio have much in common. Both are on the Appalachian Plateau, both have Pennsylvanian bedrock, and portions of these areas between 41 and 42 degrees north latitude were once claimed by the state of Connecticut. The two major parks discussed in this guidebook chapter are similar in many respects, but have complementary geological and cultural features that recommend a visit to both. MILLS, MILLSTONES, AND BUHRSTONE
Figure 1. Chart showing relative stratigraphic position of selected rock units noted in the text. The base of the Pottsville Group is dated to ~318 million years ago. Rock-unit usage is in conformance with that of the Ohio Division of Geological Survey (e.g., see Ruppert et al., 2010) which considers all units below the level of group in the Pennsylvanian to be informal, except for the Sharon Formation which is used as in Foos (2003).
Mills were established early on as Euro-Americans moved across the Alleghenies, and so, along with sawmills, were the first industries in western Pennsylvania and eastern Ohio. Mills were constructed where there was a drop in water along a stream, typically at a waterfall or where such a drop could be engineered. In eastern Ohio and western Pennsylvania, waterfalls, typically formed by resistant beds of sandstone, provided the necessary drop to turn a waterwheel, especially when aided by a dam. The waterwheel provided the power to turn pairs of millstones that actually ground the grain. Millstones in the region actually predated water-powered mills, however, as hand mills (querns) utilizing local stone were used from an early date along with treestump mortars and wooden pestles. Stone millstones used in the eighteenth and nineteenth centuries produced nutritious flour that contained some bran and midlings (Fletcher, 1950, p. 326). Newer, more efficient types of mills invented in the mid- and late-1800s, along with finer bolting cloths, resulted in whiter flour minus bran, wheat germ, and other unwanted material. This resulted in less nutritious flour for people and some nutritious midlings for cattle and pigs. Stoneground flour has continued production, however, to this day, and has come into greater prominence with the renewed interest in healthful breads in the later decades of the twentieth century. In addition to processing grain, millstones were also used for other purposes, including crushing calcined lime at lime kilns. A variety of stones have been used for millstones in Ohio and Pennsylvania. The best overall references to date on these and other millstones of the United States are the compilation of papers on millstones by Ball and Hockensmith (2007) and the recent book on the millstone industry by Hockensmith (2009a).
Early industrial geology of western Pennsylvania and eastern Ohio In this paper, we provide additional information on millstones in western Pennsylvania and eastern Ohio. Granite Millstones The bedrock of western Pennsylvania and eastern Ohio is composed of sedimentary rock, but some early millstones were fashioned from granitic glacial boulders (Saja and Hannibal, 2009). There are a number of mentions of such usage in the historical literature. Fletcher (1950, p. 325) noted that early millstones in Pennsylvania “were made of native granite and were three to seven feet in diameter.” The millstones seen at the ruins of McConnells Mill (Stop 1) confirm this statement. Work in northeastern Ohio (Saja and Hannibal, 2009) shows that the situation was similar there. The oldest millstones used for Youngstown’s 1845 Lanterman’s Mill (Stop 4) are said to have been granite (Melnick, 1976, p. 235). Early millstones in Shenango Township, Pennsylvania, have been described as being composed of “country stone” (Durant and Durant, 1877, p. 112), which may or may not have been granite. Elsewhere in Pennsylvania, the term “country stone” has been used in contrast with “burrs” (Hazen, 1908, p. 299). Leung (1981, p. 38), in a study of mills in Ontario, noted country stone as being granite or conglomerate. Conglomerate Millstones Conglomerate, along with sandstone, has been widely used for millstones in Europe and the United States (Safford, 1880; Tucker, 1984; Hockensmith, 2009a). This use gave rise to the old European and American rock-strata name “Millstone grit.” The nineteenth-century term “millstone grit” (lower case) referred to a coarse sandstone or a pebbly sandstone, presumably intermediate in grain size between a conglomerate and sandstone (Dana, 1884, p. 426). As a rock-unit name, Millstone grit was used synonymously with the Pottsville conglomerate (Chamberlin and Salisbury, 1909, p. 620, 641). Apparently influenced by European models, many early millstones in the United States were also made from conglomerates and conglomeritic sandstones. There is a reference to conglomerate and millstones in one of the early Ohio Geological Survey reports (Whittlesey, 1838, p. 58), but that reference notes that the local conglomerate (now known as the Sharon Formation) was not good for millstones. This is an unusual statement that seems to imply that someone was using, or attempting to use, this conglomerate for millstones. Indeed they were (Hannibal and Saja, 2009); a millstone found along Mill Creek (Stop 4) at the Lanterman’s Mill site in Youngstown is an example of such a conglomerate millstone. Berg (1986) also documented a millstone quarry in the Olean Conglomerate in Tioga County, north-central Pennsylvania. In addition, Hockensmith (2009b) documented, in detail, six conglomerate millstone quarries in Powell County, Kentucky. The advantages of conglomerate millstones appear to be their heterogeneity and, perhaps, by the presence of hollows created by plucked-out pebbles. These hollows retained, or appeared
145
to retain, sharp edges as the stone was ground down through use. In this way conglomerate was like French buhr, the most desirable of millstones, which was made of chert. Conglomerate millstones tended to glaze after repeated use (C.D. Hockensmith, December 2000, personal commun.), however, so this comparison of conglomerates with French buhr may not hold true. Chert Millstones (Buhrstone Millstones), Especially French Buhr The premier material for manufacture of millstones in the eighteenth and nineteenth centuries was the rock traditionally known as buhrstone (also spelled burrstone, or burstone, and sometimes known simply as buhr). This term was used typically for siliceous rock that is suitable for manufacture of millstones (Arkell and Tomkeieff, 1953, p. 16). The term has been used in geological literature (e.g., Stout, 1927, p. 256) and by those who study millstones (Hockensmith, 2009a, p. 215). Buhrstone is typically a light-colored chert. The name buhrstone continues to be used for this rock today, but the term is also known as part of the name of “buhrstone ore,” an iron ore associated with buhrstone. The term buhrstone has also been used for millstones made of buhrstone (e.g., Lepper et al., 2001, p. 55). The best known buhrstone by far is French buhr, that is, buhrstone from France. Classic descriptions of this Cenozoic French stone quarried in the Paris Basin go back to Cuvier (e.g., Cuvier, 1815, p. 308–311) and the stone was well known in North America being prominently noted in publications such as Hughes’ (1851) classic book, The American Miller and Millwright’s Assistant, which went through many editions. French buhr was long quarried in La Ferté-sous-Jouarre and vicinity in France (Ward, 1993) and exported to Britain, the British colonies, and the United States. In many of the larger cities in the United States, manufacturers imported blocks of the French buhrstone, which were then assembled into complete millstones at their workshops. City directories show that French buhr was being used by millstone manufacturers in Cleveland (Mac Cabe, 1837; Fig. 2) and Pittsburgh (Harris, 1837; Hockensmith, 2009a, p. 97–98; Fig. 3) in the 1830s. Newspaper advertisements indicate that by April 1825, composite French-buhr millstones were being produced in Cleveland (Fig. 2). Pittsburgh City Directories in the middle decades of the 1800s show that William W. Wallace was also one of the Pittsburgh manufacturers who sold Chesnut [Chestnut] Ridge and Laurel Hill millstones as well as Frenchbuhr millstones in Pittsburgh. In southern Ohio, the earliest millstones were French buhrs and millstones from Redstone and Laurel Hill, Pennsylvania (Garber, 1970, p. 77–78). The Laurel Hill, Pennsylvania, stone has been mistakenly referred to as granite (Garber, 1970, p. 11; this is not the only case of misidentification of chert millstones as being composed of granite!) and as “Laurel sandstone” (Melnick, 1976, p. 248), but was a cryptocrystalline chert quarried from the Pennsylvanian rocks near Brownsville in Fayette County, western Pennsylvania. By 1790, Laurel Hill stones were sent by
146
Hannibal et al.
Figure 2. Woodcut (from Mac Cabe, 1837) showing three men fashioning millstones out of pieces of French buhr at a millstone manufacturer near the Cleveland lakefront in the 1830s. A gristmill is shown in the left background and a side-wheel steamer and sailing ships (presumably used to transport millstones) are shown in right background.
flatboat to Marietta, Ohio (which was founded only in 1788) via Pittsburgh (Garber, 1970, p. 11). The Laurel Hill material was used early on in the Pittsburgh area as Pittsburgh was downriver from Brownsville on the Monongahela River. The stone continued to be available for sale in Pittsburgh well into the nineteenth century (James M’Kinney ad in Harris, 1837; Fig. 3). Laurel Hill millstones were also shipped down the Ohio River to Kentucky, and were in use there by at least 1802 (Hockensmith, 2008). Historically, buhrstone was produced in Vinton (Raccoon buhr), Muskingum, and Licking counties, Ohio, from greyish or yellowish white micro/cryptocrystalline-quartz rocks of Pennsylvanian Age. Raccoon buhr, a variety of chert quarried from the Vanport limestone near McArthur, Vinton County, Ohio, was especially well esteemed, although second to French buhr in reputation (Foster, 1838, p. 90–91; Safford, 1880, p. 176). Contemporary advertisements (Figs. 2 and 3) make this ranking clear. Stout (1927, p. 259) described the best stone used for buhrstone in Elk Township of Vinton County as “somewhat cellular but firmly bonded” flint (“cellular” refers to stone that has rounded to subrounded hollows). The Vanport was also quarried at other locations in Vinton County (Stout and Schoenlaub, 1945, p. 75–78) as well as at Flint Ridge (Garber, 1970, p. 80–82; Carlson, 1991, p. 14–16, 65–67; Hockensmith, 2007), Ohio for millstones. Carlson (1991, p. 15) described the Vanport flint used as millstones as coming from the “impure, porous phases.”
Figure 3. Ad (from Harris, 1837) for a millstone manufacturer in Pittsburgh. The illustration shows one of the typical groove-and-furrow patterns used at the time. It also shows the sources of millstones at the time, giving French buhr (burr), the most desirable stone, the most prominence in the ad. Image courtesy of the Special Collections and Archives of the Kent State University Libraries.
Early industrial geology of western Pennsylvania and eastern Ohio The Pennsylvanian cherts from Pennsylvania and Ohio quarried for millstones can presumably be distinguished from the French by their fossil content. The Vanport and other Pennsylvanian units contain Paleozoic fossils, including fusulinids (Smyth, 1957; Carlson, 1991). The French material is Cenozoic in age. However, a direct comparison of material at La Ferté-sous-Jouarre with the material from Pennsylvania and Ohio still needs to be made. IRON, IRON FURNACES, AND IRON ORE IN WESTERN PENNSYLVANIA AND EASTERN OHIO Early Iron Industry in Ohio and Pennsylvania Iron was (and is) produced by blast furnaces (Fig. 4) operated by combining three elements (fuel, flux, and ore) to create two (iron and slag). The fuel used in the late eighteenth and early nineteenth century was typically hardwood charcoal, which was readily available by cutting swaths of the virgin hardwood forests of Pennsylvania and Ohio. As these forests were cleared, the fuel switched around the 1850s to coal, which was and still is abun-
147
dant in Pennsylvania and Ohio. Around 1875, these would give way to coke (White, 1979, p. 4). The fuel provided the necessary heat and reducing power needed to melt the ore inside the furnace. Flux, often limestones such as the Vanport limestone, was added to the furnace to bond with molten iron-ore impurities in the blast furnace. This bonding would create a glassy material, called slag, which would float upon the molten iron at the bottom of the furnace. When the impurities separated from the iron, the slag was removed from the furnace, the molten iron metal would be tapped from the furnace, and the process started again. The name “blast furnace” comes from the air, or blast, which is forced into the furnace. With the addition of this air, the furnace is able to reach much hotter temperatures and is essential to melting the iron ore (iron melts at around 1200 °C). In the nineteenth century, this blast could be produced by following one of three general methods: using a trompe, bellows, or blowing tubs. A trompe was a crude blast device that used falling water to push air into the furnace. Because the air came directly from the area where the water was stored, it was cold and the pressure was not as great as under other styles. The Hopewell Furnace (Stop 6) originally had a trompe device; however, it eventually was switched to bellows to produce a more efficient blast. The bellows are another way to force a blast into a furnace. In this method, a waterwheel is attached to a shaft with a cam upon it, and this cam operates a bellows or set of bellows that force air into the tuyère opening (see illustration on p. 78 of Bining, 1938). The final method is the use of blowing tubs. In this method, a waterwheel is attached to a set of piston arms which operate two blowing tubs alternatively. The tubs supply their oxygen into a central chamber, known as a plenum, which is connected to a blow pipe that forces air into the tuyère. Later innovations would heat the blast and channel it into the tuyère. Because the hot blast was already several hundred degrees Celsius, it allowed the furnace to operate more efficiently and reach hotter temperatures with less fuel. In contrast to that innovation, furnaces that utilize a trompe, bellows, or blowing tub method are known as “cold blast.” The early furnaces created a type of iron known as cast iron. Cast iron is an iron which has a high carbon content (3–4.5%). This high carbon content makes it brittle after casting, and it is unable to be worked by a smith after casting. This cast iron was either cast directly into goods or into ingots for transport. Because of the inflexibility of cast iron, it was often cast into ingots, known as pigs, and transported to a bloomery, where it would be rendered into a more workable form, wrought iron. Until its replacement by mild steel in the early twentieth century, wrought iron was the metal of choice for smithing (Light, 2000). Early Iron Manufacture in Pennsylvania
Figure 4. Generalized cross section of an early iron furnace. Reprinted with slight modifications from The Ohio Journal of Science (White, 1980c, fig. 1) with the permission of the Ohio Academy of Science.
Pennsylvania’s history of blast furnaces predates the American Revolution. The year 1692 saw the first blast furnace producing iron in the colony and by 1720 there were four blast furnaces there. By the time of the revolution, there would be nearly 60. In 1841, there were well over 200 (Moldenke, 1920, p. 15–16).
148
Hannibal et al.
As settlement moved westward, blast furnaces too moved westward to supply the iron stove pieces, nails, andirons, pigs, and utilitarian products necessary to settling the frontier. Pittsburgh, a city long associated with ferrous metallurgy, saw its first furnace erected in 1792, but it was not successful. In the first half of the nineteenth century, Pittsburgh would be known not for its pig iron production, but for the foundries which converted the cast iron pigs into wrought iron (Moldenke, 1920, p. 18). Two types of furnace systems existed in Pennsylvania: the plantation and the entrepreneur furnaces. The plantation systems frequently emerged in areas near established populations (White, 1979, p. 5). These plantations were nearly self-sufficient communities, creating even their own food (Schallenberg and Ault, 1977, p. 436). An example of this is the Hopewell Furnace National Historic Site in southeast Pennsylvania. Entrepreneur furnaces were those furnaces that were willing to supply a local market with readily available goods for a finite amount of time. These took advantage of the abundant timber and readily available low-grade iron ore found in western Pennsylvania. When the distances from raw materials or markets became too great, these furnaces would close, and a new furnace would be built that could more profitably produce iron. It would be these entrepreneurial furnaces that would sprout west of the Alleghenies. In Mercer County, Pennsylvania, alone, no fewer than fifteen entrepreneurial furnaces operated (Sharp and Thomas, 1966). Early Iron Manufacture in Ohio Like its eastern neighbor, Ohio has a long history of manufacture of ferrous metal. The first blast furnace in Ohio, circa 1802, was the Hopewell (Eaton) Furnace. Its name is now synonymous with metallurgy (White, 1978, p. 391). Before 1820, blast furnaces would be erected in Akron and Tallmadge as the frontier moved ever westward (Moldenke, 1920, p. 27). The furnaces constructed in Ohio in the first half of the nineteenth century would be almost exclusively charcoal furnaces. As the hardwood forests began to disappear, many of these furnaces closed. Some furnaces, such as the Mill Creek (Trumbull), were able to make the switch to fossil fuels (coal) that fueled the furnaces after 1840–1850. By 1884 (Wright, 1884, p. 129), Ohio ranked second highest in iron manufacture behind Pennsylvania. In that span, iron production in Ohio transformed from many small production furnaces (around two to five tons a day) to massive mill complexes each capable of many hundreds of tons per day. Ohio was fortunate enough to be located near early iron rich ores and hardwood forests essential for early nineteenth century iron production. Following the discovery of iron-rich deposits around Lake Superior in the mid to late nineteenth century and the rise in use of coal due to the depletion of Ohio’s forests, Ohio still had a role as the nexus of where those materials could be inexpensively transported for manufacture. By the mid nineteenth century, the Mahoning Valley, with Youngstown as its center, was leading the entire state in iron manufacturing (Wright, 1884, p. 131).
One of the most important influences on the rise of iron, and later steel manufacturing, was the availability of transportation. Ohio’s industries profited early on (though it would nearly bankrupt the state) by the creation of a number of canals. These permitted goods to find readily available markets both farther from the source of manufacture and also at a fraction of the time and cost previously employed. Lasting for only a generation, these vital waters would be supplanted by railroads, which permitted coals from West Virginia, Pennsylvania, and Ohio to be combined with Lake Superior ores, shipped inexpensively to ports such as Cleveland and Ashtabula on Lake Erie (Wright, 1884, p. 133). Iron Minerals Used for Iron Ore The common forms of iron minerals used in the early furnaces were iron carbonates (siderites) or iron oxides (limonite and hematite) that were available near the furnace. These minerals were found near the surface and could be extracted from pits or trenches. Early furnaces were sited near sources of iron ore, limestone (used for flux), fuel (charcoal or coke made from coal) and running water for powering a blast machine. An iron carbonate (siderite) was one of the earliest types of ore used in eastern Ohio and western Pennsylvania. In this area, this mineral is found as layers, concretions, or nodules. The mineral siderite is a precipitate of iron and carbonate ions which form with other minerals within a soft mud sediment close to the sediment-water interface (Fisher et al., 1998). Siderite forms a dark to medium gray, dense rock which weathers to a reddish brown color. Density of siderite varies from 3.00 to 3.80 depending on the purity of the siderite (Stout, 1944a). All carbonate ores were roasted prior to use in the furnace to drive off water and carbon dioxide (Stout, 1944a). Continuous siderite layers can sometimes be found above a limestone bed as a layer intermixed with varying amounts of calcium and magnesium carbonate and silica. An example of this type of ore is the Buhrstone iron ore which is found above the marine Vanport limestone in western Pennsylvania (Coyle, 2003). The Vanport limestone is an important marker bed for the identification of adjacent rock layers because of its unusual thickness which is greater than 20 ft in Lawrence, Butler and Armstrong counties, Pennsylvania (Berkheiser, 1999). Below the Vanport limestone are typically three other marine limestones, the Upper Mercer, the Lower Mercer, and the Lowellville, of which the Upper Mercer and Lower Mercer are exposed along U.S. Route 422 south of New Castle, Pennsylvania, near the Moravia Street exit. Layers of siderite can occur above the Vanport limestone, the Upper Mercer Limestone and the Lower Mercer limestone which were used for iron ore. The ore above the Upper Mercer limestone was also known as the Big Red Block ore (Stout, 1944a, p. 116). The ore over the Lower Mercer limestone has been called the Little Red Block ore (Stout, 1944a, p. 64). Siderite can also be found incorporated into concretions or nodules found in shale beds (Newberry, 1878, section between
Early industrial geology of western Pennsylvania and eastern Ohio p. 804 and 805; Willis, 1886), including dark shales above limestone or coal beds. These concretions were referred to as “kidney,” or reniform, ore because they were oval or kidney shaped (Willis, 1886, p. 235; Stout, 1944a, p. 7). This kidney ore, however, is not as heavy and not as iron rich as some of the classic kidney ores of England and the eastern United States. The current Glossary of Geology (Neuendorf et al., 2005, p. 352) defines kidney ore as a variety of hematite, but also as a concretionary ironstone. Thus the definition can include iron carbonate (e.g., siderite) as well as minerals such as hematite. Newberry (1870a, p. 41) described the Ohio kidney ore as an “earthy carbonate of iron” which “generally forms balls or concretions, lying in the shales of the coal formation.” By the time Newberry wrote this, however, its use had been supplanted by other types of ore. Fossils or shell fragments have been found within concretions or nodules but are not found in all concretions (Pye et al., 1990, p. 325). Concretions can form in shallow sediment, close to the sediment-water interface (less than 10 m) (Fisher et al., 1998, p. 1). Carbonate cement forms in the pore space of the sediment and incorporates clay particles as it grows, eventually forming a spherical or ovoid shape. The original laminations of the clay can sometimes still be seen inside a concretion when broken open. Size of the concretions or nodules vary and can be up to 4+ inches in diameter and irregularly shaped. The concretions can be aligned in layers or occur randomly in a shale. Concretions can have a rind of limonite or hematite which is a yellowish to brownish red iron oxide. Concretions with rinds of iron oxides can be found embedded in shale along the streambed of Yellow Creek, in Struthers, Ohio (Stop 6), where the first iron furnace in Ohio was built. Such concretions have also been collected by archaeologists at furnace sites. Mined buhrstone and siderite nodules can vary in iron content from 25 to 45 percent (Harper and Ward, 1999, p. 29). Siderite can also be found as black, hard layers within or on top of coal seams where it was called blackband ore. Blackband ore was found just above the Upper Freeport coal seam in Tuscarawas, Carroll, Perry, Stark, Guernsey, and Gallia counties, Ohio (Stout 1944a, p. 181). This blackband ore was a black, bituminous shale impregnated with iron in the form of an iron carbonate (Stout, 1944a, p. 182). The quantity of metallic iron in blackband ore varies from 25 to 40 percent (Stout, 1944a, p. 182). After weathering, the blackband ore breaks down into thin rusty flakes (Stout, 1944a, p. 183). This ore was sometimes overlain by a calcareous layer with nodules of siderite (Camp, 2006, p. 213). Blackband ore is not noticeably denser (its specific gravity ranges from 2.3 to 2.5) and looks like a black shale (Stout, 1944a, p. 189). It was not discovered until 1854 by an English miner, John Lewis, who was familiar with the blackband ore in the Victoria mines in England (Stout, 1944a, p. 29). The Sharon blackband ore played an important part in the development of iron industry in Youngstown. The Sharon blackband ore is found within the Sharon coal or is found at the bottom of the Sharon coal and is limited to Trumbull and Mahoning counties. Within the Sharon coal it occurs as a layer of iron
149
carbonate in the form of a coal parting. It is banded in brown and black layers and displays a varvelike structure (Stout, 1944a, p. 28). After 1854, the blackband ore began to be used in furnaces in Mahoning and Trumbull counties (Stout, 1944a, p. 29). Slucher and Rice (1994, fig. 2) located a number of siderite beds in their column of the Pottsville Group in Ohio. Bog-iron ore formed relatively recently in the Quaternary age and has been mined along the beach ridges from Cleveland, Ohio, to the Pennsylvania-Ohio state line (Stout, 1944a, p. 6). It is generally yellow-brown in color and variable in thickness from a few inches to several feet (Stout, 1944a, p. 6). Bog ore is open and spongy in texture and contaminated with impurities such as clay (Stout, 1944a, p. 6). It precipitates in shallow waters such as springs or swamps as a yellow or orange sediment that consolidates into an iron ore (Harper and Ward, 1999, p. 29). Bog iron is a limonite precipitated as nodules or sheets over several acres (Stout, 1944a, p. 6). Bog iron ore was used at the Van Buren Furnace in Cranberry Township, Venango County, Pennsylvania (Harper and Ward, 1999, p. 29). Iron ores were of special interest to the early geologists of Ohio and Pennsylvania including Ohio’s W.W. Mather (Mather, 1838, p. 7–9) and Pennsylvania’s H.D. Rogers, who even had his own iron furnace (Gerstner, 1994, p. 62,132–133). Slag and Slag Analysis Slag from old iron furnaces can be found in many places in western Pennsylvania and eastern Ohio, including places where there have been iron furnaces and places where the slag has been transported by streams, or more often, dumped or reused. Such slag can take on a variety of external forms, ranging from irregular material resembling aa lava to vitrified material that bears a resemblance to obsidian or bottle-glass (Fig. 5). Because it is eyecatching, people pick up pieces of slag and bring it to museums (at least in the United States and Great Britain) for identification. Sometimes those bringing the slag to museums are doing so with the hope of confirming their find of a “meteorite.”
Figure 5. Slag excavated by archaeologists at Trumbull Furnace, Mill Creek. These samples show some of the color variants and the glassy nature of some of the slag at this location.
150
Hannibal et al.
Slag is both the most prevalent object at blast furnaces and also the item that best tells the story of blast furnaces (White, 1979, p. 7). Two of us (H.E. and T.G.) have analyzed a suite of slag samples collected over a period of two years from 36 charcoal furnace sites in Venango, Clarion, Forest, Mercer, and Lawrence counties in western Pennsylvania and Lebanon and Berks counties in eastern Pennsylvania (Fig. 6; Edenborn et al., 2009). Where discernible, representative slag samples, as well as any slag that seemed unusual in terms of color or texture, were collected from each site. In addition, ore, limestone flux, and iron metal samples were collected, if observed. In the laboratory, the samples were broken open and examined. Pieces with weathered surfaces were discarded and representative pieces with fresh surfaces were crushed in a mortar and pestle. The magnetic susceptibility of crushed slag samples was determined from the ratio of inductance obtained with and without the sample inside of a 2.8 cm inner diameter measuring coil (SI-2 Magnetic Susceptibility and Anisotropy Instrument, Sapphire Instruments, Ruthven, Ontario, Canada). Specific gravity of samples was estimated by fully suspending solid samples of known weight in deionized water (Mursky and Thompson, 1958), and slag color was estimated using the Munsell Rock Color Chart (1991). Powdered (<75 µm) slag samples were analyzed for major and minor elements using a molten salt fusion analysis and inductively coupled plasma–atomic emission spectrometry (modified ASTM Method D6349). A subset of the slag sample set was analyzed by X-ray fluorescence.
Analysis of Slag Samples Preliminary analyses of slag samples from cold-blast charcoal iron furnaces in northwest Pennsylvania suggest the following. (1) Slags from a given furnace site are generally physically similar in appearance and contain similar trace elements. (2) Low refractory indices (RI) suggest that furnace charge materials (ore, flux) were of a composition that permitted relatively low furnace operating temperatures. (3) The desulfurizing capacity (DI) of tested slags was low, but this ability was unneeded where low-sulfur charcoal fuel and ores were routinely used. (4) The principal component analysis (Fig. 7) indicates the presence of two distinct groups: When the control of Axis 1 is plotted against the control of Axis 2, it is clear that the slag samples are chemically distinct from the ore samples. This difference may provide insight to the types of fluxes utilized during smelting. Pioneering research on the composition of early Ohio and Pennsylvania iron blast furnace slags was conducted by John White (1980), who compared slags from twelve early blast furnaces in Ohio (including the Hopewell [Eaton] and Trumbull), Pennsylvania (Wilroy), and Europe in terms of their metallurgical features and physical attributes. White (1980) was able to
Axis 2 (controlled by V)
Description of Slag Samples Slags from these sites were frequently dark green–colored and glassy, reflective of high silica and residual iron content. Slags from specific furnace sites tended to have similar suites of minor trace elements that may be traceable to given ore or flux sources. Magnetic susceptibility tests were able to screen for slags containing small iron prills, likely indicative of inadequately heated furnaces. Short-wave fluorescence was intense in
some samples and likely only occurs when correct ratios of activator and quenching elements are present. Lower specific gravity was generally indicative of greater amounts of entrained air or gas in slag, also lightening the slag color. Two general metallurgical indices can be calculated based on chemical analysis of slags. The refractory index (RI) reflects the amount of alumina in slag relative to lime and silica, high values indicating a more refractory slag that requires a higher furnace temperature to melt. The desulfurization index (DI) is calculated as the ratio of calcium and magnesium oxides to silica and alumina, a high index indicating a greater sulfurremoving capacity.
Group 1
Group 2
Axis 1 (controlled by K2O) Figure 6. Slag-analysis sampling sites at charcoal furnace sites in western (Venango, Clarion, Forest, Mercer, and Lawrence counties) and eastern Pennsylvania (Lebanon and Berks counties).
Figure 7. Preliminary principal component analysis (PCA) of chemical and physical slag and ore variables. Two major groupings can be distinguished based on this plot of the first and second principal components, controlled by K2O and vanadium (V).
Early industrial geology of western Pennsylvania and eastern Ohio show that properties of slag could be used to indicate the likely operating conditions and relative efficiency of the furnaces at the time. An “optimal” slag in an active furnace would demonstrate the following two important metallurgical characteristics: it would be fusible, or easily melted at high temperatures; and be fluid, with a low viscosity, at those temperatures. Additionally, as mentioned previously, the chemical composition of the slag was important in the scavenging of stray sulfur, seldom a problem in charcoal iron furnaces, but a more common problem when coal was used. Many of White’s observations on the characteristics of slag are consistent with recent studies (Edenborn et al., 2009) of a much larger number of additional furnaces in Pennsylvania. Not surprisingly, the metallurgical characteristics of studied slags suggest that blast conditions at small charcoal iron furnaces were seldom optimal, probably reflecting the consistent use of lowgrade iron ores, which could be composed of a wide range of ore
151
types and qualities, and technical difficulties maintaining proper blast temperatures. Interestingly, White was able to determine that blast conditions at the Hopewell (Eaton) furnace in Ohio, one of the first to attempt to use both charcoal and coal with a higher sulfur content, probably resulted in the demise of the furnace, due to its inability to fully remove the additional sulfur, which would have resulted in an inferior iron product. FIELD TRIP STOPS Stops cover parts of western Pennsylvania and eastern Ohio (Fig. 8). These stops were chosen so that sites in both states could be visited during a one-day field trip. The stops may, of course, be visited in a different order. Information on the stop locations is given at the beginning of each stop. Web sites for the two major parks systems visited on this trip contain detailed maps.
Figure 8. Map of field trip stops: 1—McConnells Mill, McConnells Mill State Park, Lawrence County, Pennsylvania; 2—Hells Hollow, McConnells Mill State Park, Lawrence County, Pennsylvania; 3—Route 422, New Castle Pennsylvania; 4—Lanterman’s Mill, Mill Creek Park, Youngstown, Ohio; 5—Mill Creek Furnace, Mill Creek Park, Youngstown, Ohio; 6—Hopewell (Eaton) Furnace, Yellow Creek Park, Struthers, Ohio; 7—Struthers Historical Society, Struthers, Ohio.
152
Hannibal et al.
Stop 1. McConnells Mill, McConnells Mill State Park, Slippery Rock Gorge, Lawrence County, Pennsylvania McConnells Mill State Park is located west of Portersville, in southeastern Lawrence County, Pennsylvania. With the exception of the inside of the mill, it is open year-round. This park includes sites related to several early industries, including grain milling, iron production, and the burning of agricultural lime. Visitors to the park should obtain a copy of the Pennsylvania Trail of Geology Moraine and McConnells Mill State Parks guide (Fleeger et al., 2003). Maps of the park are available on the Pennsylvania State Park Web site. The eastern end of McConnells Mill State Park contains the site of the McConnells Mill (Figs. 9 and 10) itself as well as the ruins (mostly foundation material) of an earlier gristmill. Access
Figure 9. McConnells Mill, McConnells Mill State Park, and adjacent dam.
Figure 10. Geologic map of the McConnells Mill area, McConnells Mill State Park. Allegheny Group rocks are shown in yellow and Pottsville Group rocks are indicated by brown. Two lakes are shown in the bottom right. The numeral 1 indicates the site of McConnells Mill. Park boundaries are indicated by the intermittent dots and dashes. Trails are indicated by dotted lines. Geologic data for this map is from Berg et al. (1980), and Pennsylvania Bureau of Topographic and Geologic Survey (2001).
Early industrial geology of western Pennsylvania and eastern Ohio to McConnells Mill is from a set of stairs leading from a trail from a parking lot along McConnells Mill Road or from a park road leading downhill to a small parking area by the mill itself. The trail from the upper parking area is recommended as it is scenic and provides good views of a thick sequence of the Homewood sandstone (Fig. 11). Here the Homewood is composed of coarse-grained sandstone and conglomerate. Joints, crossbeds, and honeycomb weathering can be seen in the unit. (A discussion of the Homewood and a stratigraphic column of McConnells Mill State Park in Skema [2005b], however, shows that identifying a unit as the Homewood is not without its problems.) McConnells Mill (Forest Mills) McConnells Mill (Figs. 9 and 10) is located in the eastern part of McConnells Mill State Park, near the entrance to the park off of State Route 422. The mill is open seasonally, closing at the end of October and re-opening in the spring. This mill, also known in the past as the Forest Mills, was one of a number of gristmills constructed along Slippery Rock Creek in this area,
Figure 11. Homewood sandstone at McConnells Mill showing typical cross-bedding and jointing. Staff is 1.5 m high.
153
the first of which was constructed in or just after 1825. McConnells Mill is the best known and only remaining intact mill here. It was run by Captain Thomas McConnell (1822–1905) and his son James (Durant and Durant, 1877). McConnells Mill was constructed in 1870 on the foundation of an even earlier mill which burned in 1868. McConnells Mill has its foundation built into the Homewood sandstone, and it appears to have been built out of blocks quarried from the Homewood. The Homewood has been quarried in other areas of Lawrence County (Stone, 1932, p. 191–192) and across the state border in Ohio as well (Stout, 1944b). Red sandstone in part of the present mill foundation is said to have been reddened by the fire that burned down the predecessor to the current mill. Fire can redden sandstones rich in iron, but we have not verified if this red color was caused by fire or if the stone was originally red. This mill was once noted for its early use of rolling mills, which utilized steel rollers to crush grain. Despite this, a number of traditional millstones remain in and around the mill. Two complete French-buhr millstones are preserved here, one inside the mill (Fig. 12) and one outside (Fig. 13). The stone inside the mill is a composite millstone composed of polygonal blocks (known as “panes”) of light-colored chert (yellowish gray 5Y 8/1), 106 cm in diameter. This is a runner stone (the top stone in a working pair of millstones), still with its plaster top (see Hannibal and Evans, 2010, fig. 30, for a view of another runner stone whose plaster is deteriorating because the stone has been left outdoors). Such plaster, typically incorporating stone rubble, was added to finish and balance the runner. (Balance boxes were also used for weights to balance the runner.) The millstone contains a number of rounded to suboval cavities, several of which exceed 3 cm in maximum diameter. Another composite French-buhr millstone outside of the mill has the same coloration and diameter. Microfossils and other particles can be seen in the stone. Cavities
Figure 12. French-buhr millstone set on edge inside of McConnells Mill. The pieces of chert used for this millstone are cemented together with plaster and held in place with an iron band. This is the runner (top) stone of a pair. Staff is marked in 1-dm increments.
154
Hannibal et al.
range up to 4.5 cm in maximum diameter. One block of this millstone is cut in an unusual manner, showing bedding structures that indicate that its cutting surface was cut to be perpendicular to bedding. Most millstones made of sedimentary rock are cut parallel to bedding. Bedding surfaces with abundant cavities (cells) were typically chosen as the cutting face of millstones (Hildreth, 1838, p. 343). There are other millstones and related materials in the mill. These include a level used in finishing millstones. A 33 by 48 cm piece of a French-buhr millstone can also be seen inside the mill. Only the cutting surface of this millstone is finely finished. The sides are very rough. There is also a sandstone millstone inside the mill. Mill Ruins The ruins of an earlier mill are located along the Kildoo Trail which follows Slippery Rock Creek to the south from McConnells Mill. (From the mill, the trail to the ruins runs along the east side of the stream south past the covered bridge.) The ruins are ~100 ft past the wooden footbridge over the falls. Some sandstone building blocks used for the mill are in place while others are strewn about. Shallow, rectangular depressions used for anchoring wooden beams can be seen in the 5-m-high stone block at stream level. Three millstones can be seen at the ruins just below the level of the trail. Two of the stones are entire; a third is broken. Two of the stones are monolithic feldspathic stones. The entire stone that is readily measurable is 115 cm across. It has large crystals that delineate some foliation or preferred orientation along which some cracks are developed. The third, broken millstone is a light-colored granitic stone and was presumably monolithic. It is ~105 cm in diameter. Roughly one-third of the stone has broken off at some time, with the break partly along an old 3-cmdiameter, 9-cm-deep, drill hole.
Figure 13. French-buhr millstone set into sidewalk outside of McConnells Mill. The iron band which held all of the pieces (panes) of this composite millstone together can be seen. The central cavity has been filled with concrete. Staff is 1.5 m in length.
A fourth monolithic millstone (Fig. 14) is leaning against a tree downslope from the trail. This granitic stone, ~109 cm in diameter, is very high in quartz and contains biotite. It has a maximum thickness of 20 cm and has a very irregular bottom. Stop 2. Hells Hollow, McConnells Mill State Park, Lawrence County, Pennsylvania Hells Hollow (also known as Big Hollow in the past) is located in the western part of McConnells Mill State Park (Fig. 15), with access from a parking lot along Shaffer Road. A path leads from the parking area downstream along Hell Run, a tributary of Slippery Rock Creek. The trail bifurcates just before the first footbridge over Hells Hollow Run; the trail along the southern side of the river leads to an old quarry area and an old quarry. There are excellent exposures along the trails of the Vanport limestone as well as associated karst features indicated by very evident changes in stream flow over relatively short distances (see Fleeger et al., 2003). Hells Hollow was once known for its disappearing streams and “darksome dells” (Durant and Durant, 1877, p. 115). This site is not the only “Hells Hollow” in western Pennsylvania. The Hells Hollow in McConnells Mill State Park should not be confused with the (equally interesting as far as Pennsylvanian rocks and nineteenth-century industrial geology) Hells Hollow in Mercer County. The Vanport has been widely quarried in this and other areas of Lawrence County because of the access to exposures and the uniformity of the rock (Miller, 1934, p. 482). The old quarry adjacent to the trail is shallow. If missed, it can be found by backtracking from the lime kiln which is more evident. According to Fleeger et al. (2003, p. 9) this quarry was the source of flux (limestone) and iron ore. The lime kiln (Figs. 16 and 17) preserved at this stop is unusual as it is completely, rather than partly, built into the limestone and shale bedrock of the hillside. The rock here was carved
Figure 14. Granitic millstone leaning against a tree downslope from the Killdoo Trail, McConnells Mill Park. Staff is marked in 1-dm increments.
Early industrial geology of western Pennsylvania and eastern Ohio
155
Figure 15. Geologic map of the Hells Hollow area, McConnells Mill State Park showing distribution of Pennsylvanian Pottsville (brown) and Allegheny (yellow) group rocks. Park boundaries are indicated by the intermittent dots and dashes. Trails are indicated by dotted lines. P indicates parking area; number 2 indicates site of lime kiln. Basic map adapted from Fleeger et al. (2003); geologic data from Berg et al. (1980), and Pennsylvania Bureau of Topographic and Geologic Survey (2001).
out and lined with fire brick, with the bricks aligned with their ends facing in. The bottommost part of the kiln is in shale; the top in limestone. This is a vertical (also known as continuous) kiln; the top opening was for raw material, while the bottom opening allowed for removal of the calcined lime. This kiln was used to produce agricultural lime. Kilns were used to make agricultural lime in the nineteenth century as this method reduced the limestone to smaller pieces that could be readily powdered. Early lime kilns and iron furnaces in the United States had a similar morphology (Hahn and Kemp, 1994,
p. 9–11). Lime kilns in western Pennsylvania and eastern Ohio during the seventeenth and eighteenth centuries were typically built to produce agricultural lime and natural cement (a hydraulic cement that could set under water). The calcining process helped to reduce the size of lime particles for agricultural lime. Limestone was further crushed here by a horse (or similar animal) mill (Natalie Simon, October 2010, personal commun.). Lawrence Furnace, located on private property (“X” on Fig. 15) adjacent to the park property, was built in 1865 or 1866. White (1986, table 1) reported that this furnace was built into
Figure 16. Side entrance to the lime kiln at Hells Hollow, McConnells Mill State Park. Staff is 1.5 m high. Pictured: Kathleen Farago.
Figure 17. Looking down into the lime kiln at Hells Hollow, McConnells Mill State Park.
156
Hannibal et al.
limestone bedrock and that slag and a tipple retaining wall are present at the site but his description of the site (his table 1) as being well-preserved indicates that he probably confused the furnace site with what is now interpreted as the lime kiln. This is understandable as there are a number of similarities between early blast furnaces and lime kilns. Both, for instance, are tall, have openings, and are lined with fire-resistant materials. The first ore used for the Lawrence Furnace was presumably quarried nearby, but already by 1870 on this furnace utilized “red ore” from the iron ore banks of southern Shenango Township (Durant and Durant, 1877, p. 117; the ore banks are also shown on a map between p. 5 and 6). The limestone used for flux has been described as being local, thin, brittle, bluish-gray in color (Durant and Durant, 1877, p. 117), a description which fits the Vanport limestone at Hells Hollow. The iron produced at Lawrence Furnace in the 1870s was sent mainly to Youngstown, Ohio. Stop 3. Route 422, New Castle Pennsylvania: Allegheny/ Pottsville Outcrops A spectacular outcrop (Fig. 18) of Pennsylvanian rocks at New Castle, Pennsylvania, along U.S. Route 422 south of New Castle, on the New Castle South topographic quadrangle at latitude N 40° 58′ 6.42″ and longitude W −80° 21′ 47.88″, is conveniently located between McConnells Mill Park in western Pennsylvanian and Mill Creek Park in eastern Ohio. This stop is situated along a busy highway and should only be viewed with extreme caution. The rocks are exposed on the north and south side of the road over a length of 3000 ft. (~914 m). The outcrop extends from
the Martha Street Overpass west to the Moravia Street ramps. Vertical relief is ~200 ft from the upper Pottsville Formation to lower Allegheny Formation. This stop has an abundance of siderite (iron carbonate, FeCO3) deposits that were used as iron ore in many early Ohio and western Pennsylvania iron furnaces. The following description of the stratigraphy at this stop is adapted from Skema (2005a). Stratigraphy Rock units at this exposure (Fig. 19) range from the Brookville and Clarion coals of the Allegheny Formation down to the Lowellville limestone horizon of the Pottsville Formation. The north side of the roadcut exposes the Clarion coal at the top to the Flint Ridge coal at the base. On the south side of U.S. Route 422, rocks below the Flint Ridge coal crop out as low as the Lowellville limestone. This outcrop shows: (1) the repetitive nature of coal, limestone, shale, and sandstone deposition during the Pennsylvanian, (2) siderite in various forms, and (3) a channel, incised into preexisting sediment layers, that locally cuts out important marker beds such as the Lower Mercer limestone. Note that the Lower Mercer limestone is missing on the north side of the roadcut whereas it is present on the south side of the road. The Upper Mercer limestone crops out on both sides of the road. Iron carbonate (siderite) is present as nodules or layers above the marine limestone beds and as concretions or “kidney stones” in the shale above the limestone beds. Siderite is reddish on weathered surfaces and light gray, bluish gray, or dark gray on fresh, unweathered surfaces. If tested with mild acid, it will fizz. The iron carbonate may be siliceous, however, and contain Mg-calcite and/or pyrite, each of which lessens its tendency to
Figure 18. Section of upper Pottsville and lower Allegheny Group rock exposed at roadcut on U.S. Route 422 at Moravia Street Exit Ramp, New Castle, Pennsylvania. Vehicle for scale.
Early industrial geology of western Pennsylvania and eastern Ohio fizz under acid. The siderite is finely crystalline and breaks with a conchoidal fracture (Inners, 1999, p. 563). It has a higher specific gravity than the surrounding shale. A strip mine for the Vanport limestone was located in the hillside above this outcrop. An important iron ore horizon called the Buhrstone ore was mined above the Vanport. It was named for the light bluish-gray chert, or buhrstone, that occurs between the ore bed and the limestone. Rogers (1840, p. 189–190) compared this buhrstone to the classic French buhr. Unweathered buhrstone ore is medium gray, calcitic siderite, locally siliceous, weathers brick red. The buhrstone ore (also known simply as buhrstone,
157
but that term is best used only for the siliceous rock for which the buhrstone ore was named) was mined in the 1800s to supply iron ore to the iron furnaces (Coyle, 2003). The ore in the old mines was generally 6–12 inches thick (Chance, 1880, as stated in Inners, 1999, p. 563). Weathering of the siderite resulted in thick pockets of secondary limonite (Inners, 1999, p. 563). In the roadcut exposure, more erosion-resistant sandstone and limestone layers stand out from the “softer” claystone or shale. The most prominent layer in the middle of the outcrop is the Upper Mercer limestone (see Fig. 19) (Skema, 2005a, p. 132, fig. 10–3). Just above this limestone is a thin layer of siderite. At
Figure 19. Stratigraphic Section at the U.S. Route 422 Moravia Street Exit Ramp. (Adapted from Skema, 2005a, p. 131.)
158
Hannibal et al.
the far western end of the outcrop, a layer of siderite is exposed in the bottom of the drainage ditch along the road. This layer of siderite is interesting because faint ripple marks on the bedding surface indicate that the siderite was forming a layer on top of the mud while under water and not formed as a hardpan at the bottom of a soil (Skema, 2005a, p. 132). Concretions are found in the shale above the limestone layers. The photo of the outcrop (Fig. 18) has some of the limestone and siderite layers identified. Loose nodules or concretions that have fallen from the shale may be present on the ground. In this area, nodules with a “lumpy” surface may have barite, a barium sulfate mineral, precipitated within internal fractures. Siderite concretions in the dark shale above marine limestone or coal may have septarian fracturing in the center of the nodules. These fractures may contain calcite, barite, and zinc minerals, such as sphalerite and wurtzite, and clay (Skema, 2005a, p. 130). Partial fossils may be preserved
within nodules (Skema, 2005b, p. 151) indicating that the fossil may have served as a nucleus for the growth of the concretion. However, there are concretions with no apparent nuclei (Pye et al., 1990, p. 325). Concretions which occur in the shale are disc-like and conform to the surface of the underlying and overlying shale layers. Concretions probably grew within the soft sediment as seen in Recent marsh sediments of the Mississippi River deltaic plain (Moore et al., 1992, p. 357). Siderite nodules (<2 cm in diameter) and clayey tabular siderite accumulations (<5 cm thick) that parallel bedding are common in the lacustrine and back-swamp muds of the southern lower Mississippi Valley floodplain (Aslan and Autin, 1999, p. 803). A channel cuts into preexisting sediments at the western end of the roadcut. There is a layer of siderite nodules above, but parallel to, the base of the channel. These nodules may have been transported by the eroding stream and then deposited as part of
Figure 20. Geologic map of part of the Youngstown, Ohio, region showing locations of Stops 4–7: 4—Lanterman’s Mill, Mill Creek Park, Youngstown, Ohio; 5—Trumbull (Mill Creek) Furnace, Mill Creek Park, Youngstown, Ohio; 6—Hopewell (Eaton) Furnace, Yellow Creek Park, Struthers, Ohio; 7—Struthers Historical Society, Struthers, Ohio. Mississippian Cuyahoga Formation and the Pennsylvanian Allegheny and Pottsville Groups from Slucher (2002a, 2002b) and top of the Lower Mercer limestone from Stephenson (1933, pl. 4).
Early industrial geology of western Pennsylvania and eastern Ohio lag gravel along with the clay and sand. Another theory is that this is an example of the siderite precipitating on plants, logs and other carbonaceous lag debris along the edge of the stream channel. The original organic matter would have been completely replaced (Skema, 2005a, p. 132). Stop 4. Lanterman’s Mill, Mill Creek Park, Youngstown, Ohio Lanterman’s Mill (Stop 4 on Fig. 20) is a restored gristmill in the valley of Mill Creek, the early industrial heartland of Youngstown, Ohio. The mill (Fig. 21) is located at 980 Canfield Road (Route 62), on the east side of Mill Creek, immediately south of the Canfield Road bridge over the creek, in Mill Creek Park. This park is one of the parks in the Mill Creek MetroParks system. There is a parking area located just to the north of the bridge. The best easily accessible view of the mill (and the best site for photographing the mill) is from the Canfield Road bridge over Mill Creek. The park and area around the exterior of the mill, including an elevated boardwalk along the cliffside of the valley of Mill Creek that extends in a downstream direction, is open all year long except when closed due to extreme weather conditions. The mill itself is open seasonally, typically between May and October. The mill is located alongside a scenic waterfall formed by a resistant layer of the Pennsylvanian Massillon sandstone (Stephenson, 1933, p. 69–71). (As with the Homewood as noted above, the correlation of the Massillon can be problematical; e.g., see Ruppert et al., 2010, fig. 1; Szmuc, 1957, p. 136) A path from the mill leads to the elevated boardwalk allowing a close look at the rocks along the stream. The Massillon here (Fig. 22) is a coarse-grained, crossbedded, cliff-forming quartz sandstone. Typical Pennsylvanian plant fossils (Sigillaria, Lepidodendron)
Figure 21. Lanterman’s Mill, Mill Creek Park, Youngstown, Ohio. The mill is built into an outcrop of the Massillon sandstone, part of which forms the lip of the falls.
159
can be seen in places in the outcrop (Stephenson, 1933, p. 70–71) and as float along the stream. Despite these fossils, the identification and correlation of the Massillon (also known as the Connoquenessing sandstone; Rau, 1970, p. 79) in this area of Ohio remains problematical (Slucher and Rice, 1994, p. 37). The earliest mill along Mill Creek was erected by Abraham Powers and his son ca. 1799. According to the Mahoning Valley Historical Society (1876, p. 167), the pair of millstones for this mill were split from a rock ~3 ft in diameter. The same source notes that this rock was found “in the vicinity of where Lincoln Avenue will cross Holmes Street” in Youngstown. This location would have been on the north side of the Mahoning River, just to the northeast of where Mill Creek flows into the Mahoning River. Butler (1921, p. 658) called the material “a native boulder,” and Melnick (1976, p. 235) called the source material “local granite boulders.” This is likely, based on the location of the site, its description in the literature, and other cases of early use of glacial
Figure 22. Exposure of Massillon sandstone along path along Mill Creek, just north of Lanterman’s Mill. Pictured: Kathleen Farago.
160
Hannibal et al.
granite boulders in northeastern Ohio for millstones (Saja and Hannibal, 2009). The present Lanterman’s Mill, the third to stand at this site, was constructed by German Lanterman and Samuel Kimberly between 1845 and 1846 and restored between 1982 and 1984. The mill is built into an outcrop of the Massillon sandstone where more resistant layers of the rock unit form a waterfall. The mill is also built of the Massillon sandstone, which was historically quarried along the Mill Creek gorge at various places. Quarry walls can be seen, for instance, in the Bears’ Den area to the northwest of the mill in Mill Creek Park. The Massillon is a coarse-grained quartz sandstone at the mill. Sedimentary features seen in the stone used for the mill include Liesegang rings. The mill was built at this location to utilize the drop to turn a waterwheel. The present, still functional, waterwheel is located inside the mill structure. The interior of the mill also provides a good look at a bright reddish brown iron precipitate forming at a seep in the natural sandstone forming part of the mill foundation. There are a number of such iron-rich seeps in the Massillon sandstone in this area. Millstones at Lanterman’s Mill A number of millstones and millstone fragments can be seen outside and inside of Lanterman’s Mill. Some of the millstones are monolithic, and others are composed of multiple pieces. Millstones at this site are made of conglomerate, granite, and, presumably, French buhr. A granite millstone (Fig. 23) with a bronze plaque is preserved near the entrance of Lanterman’s Mill. While it is likely that this millstone was fashioned from a glacial boulder, we have not been able to determine its history. A conglomerate millstone (Fig. 24) is the best documented, as it has long been preserved in the bed of Mill Creek, 155 m (500 ft) south of the downstream edge of the mill and 170 m (~548 ft) from the lip of the falls. By 2009, the millstone had become almost completely covered by typical stream gravel and
Figure 23. Granite millstone near entrance to Lanterman’s Mill, Mill Creek Park.
sediment. The upward-facing surface of the millstone, however, was uncovered during the summer of 2009 and subsequently swept off in 2010. (One year of normal stream deposition had partially covered it again.) The millstone can be seen from the boardwalk, where a sign points it out on the opposite side of the stream. Published sources indicate that this millstone is from the Baldwin Mill, the second (1823) mill at this site. The composition of the millstone is consistent with the identification of Melnick (1976, p. 244), who noted that the millstones here at the mill were made of “‘Pudding stone’ or perhaps Sharon conglomerate.” The millstone is a monolith made of a conglomerate with rounded to angular, white quartz pebbles. The quartz pebbles, as well as the rounded to subrounded cavities (Fig. 24, where clay clasts had broken away) present, are consistent with it having been fashioned from rock from the Sharon Formation. Interestingly, the Sharon (then known as the Carboniferous Conglomerate) was once correlated with the Millstone Grit in Europe (Newberry, 1870b). The Sharon Formation is exposed along the lower reaches of Mill Creek as well as elsewhere in the region. Conglomerates were once a preferred stone for manufacture of millstones in both England and America, but fell out of favor as the popularity of imported French-buhr millstones increased as transportation networks (especially canals) allowed for easy transport to places in western Pennsylvania and eastern Ohio. To complicate things, however, Galaida (1941, p. 10) reported that conglomerate buhrstones from Lisbon, Ohio, were brought to the old Woolen Mill (Pioneer Pavilion) area of Mill Creek Park for use in crushing flax. This type of buhrstone would have been an edge runner, that is, a buhrstone whose edge was utilized as a grinding surface. It is possible, but unlikely, that the stone in the stream by the mill is one of those other buhrstones. Three multiple-piece chert millstones (Fig. 25) are prominently displayed outside of the mill. They are presumably made of French buhr. Most large millstones sold in the United States that were composed of French buhr are composite. But,
Figure 24. Conglomerate millstone along the streambed of Mill Creek, Mill Creek Park (2009 photograph). Staff is marked in 1-dmlong increments.
Early industrial geology of western Pennsylvania and eastern Ohio
161
as Hockensmith (2009a, p. 71) noted, domestic Ohio chert millstones were also produced in segments by the 1820s. The presence of French buhrstones at this site in the past is indicated by the inclusion of “3 damaged French bur [sic] mill stones” in an 1842 inventory of the personal property of Eli Baldwin (MSS 2097, Container 1, Folder 6, Estate of Eli and Mary Baldwin, 1840–1881, Eli Baldwin papers, Archives of the Western Reserve Historical Society). There are several individual segments of French-buhr millstones (e.g., see Fig. 26) preserved inside the mill. These clearly show that only the grinding surface of the stone pieces was finely finished, with the other sides of the stones finished to various degrees. Also, it is not uncommon for Frenchbuhr millstones to be constructed from blocks of varying thicknesses (C.D. Hockensmith, 2010, personal commun.). At quite an early date, Hildreth (1838, p. 33–34) explained the basic geologic difference between French buhr and the Ohio millstones: French buhr was Tertiary and contained fresh-water shells; Ohio stone was from the Coal Measures and contained marine forms. Thus, easily identifiable fossils such as fusulinids (all Paleozoic) and horn corals are potential index fossils to the Ohio cherts used for millstones. Some of the chert millstones at this mill contain molds of trace fossils and low-spired snails, and French buhr was noted as having small fossil shells (Safford, 1880, p. 177; Hockensmith, 2009a, p. 61). A detailed geological and paleontological comparison of the various cherts used for millstones, however, remains to be made to definitively identify the sources of this stone.
ion. The furnace is built into the side of an outcrop of an unnamed Pennsylvanian shale subjacent to the Massillon sandstone. Originally built as a charcoal-fueled furnace, this furnace later utilized bituminous coal. Butler (1921, p. 177) stated that the furnace began production in either 1826 (p. 177) or 1832 (p. 663) as a strictly charcoal furnace, and that after twenty years it was rebuilt to accommodate both charcoal and bituminous coal. However, competition from other more efficient furnaces drove it out of business some time after that in the 1840s or 1850s. Also, the furnace was located roughly three miles from the Pennsylvania and Ohio Canal (completed in 1848), so both raw and finished
Stop 5. Trumbull (Mill Creek) Furnace, Mill Creek Park, Youngstown, Ohio
Figure 26. A single piece of a French-buhr millstone preserved inside of Lanterman’s Mill. Note cellular nature of the stone indicated by concave, light-colored depressions.
Trumbull Furnace (Fig. 27; Stop 5 on Fig. 20), also known as the Mill Creek Furnace, is located along Old Furnace Road just to the east of its intersection with Cohasset Drive. This is just to the north of Lake Cohasset. The mill is next to Pioneer Pavil-
Figure 25. One of three French-buhr millstones outside of Lanterman’s Mill. This millstone was constructed of several pieces; one of the segments missing. A part of an adjacent millstone is also seen to the left. The staff is 1.5 m long.
Figure 27. Trumbull (Mill Creek) Furnace. Doorway to rear of furnace is in lower right. Woman with 1.5-m-high staff is standing upon refractory sand from inner furnace with the bosh and crucible visible on the left. In the five years or so since the excavation some of the facing blocks have tumbled and the once-vertical sidewalls which permitted easy entrance to the doorway have collapsed.
162
Hannibal et al.
materials had to be moved that distance by other means (Williams, 1882, p. 371). The furnace was excavated by archaeologist John R. White in the summers of 2003–2005. A number of innovations over the original Hopewell Furnace were discovered, including stone-lined recesses under the casting floor and an access door to either side of the furnace. The recesses served a dual purpose of removing water from both the casting floor and also the molten iron and of cooling the casting floor from below. The access doorway to the left of the crucible (defined below) extends from the casting floor to the back of the furnace from which it curves around the back of the furnace before emerging on the right of the crucible. This doorway would provide access for the workers to the other side of the casting floor and would be especially useful during those times when the casting floor was covered in molten iron. The use of the site has changed over time. The furnace was built against the base of a rise, used, and abandoned. Williams (1882, p. 371) revealed that the machinery was stripped from the furnace after it was closed. Some of the blocks from the furnace were scavenged. Also, as the road above the furnace (the aptly named Old Furnace Road) was constructed, the earth from the top was leveled and pushed over the sides and took much of the front of the furnace with it, causing many of the large blocks from the outer shell of the furnace to become jumbled above the casting floor. Additionally, the iron salamander from the furnace came to extend several meters out from the furnace. Sometime following all this, a gas line was laid through the casting floor to reach Pioneer Pavilion. Following 150 years of deposition, only the most superior edges of the top two topmost blocks were visible prior to excavation in 2003. Ironically, because of the burial of the furnace, it was able to endure much longer than otherwise would be the case. Already in the few years since excavation, dangerous cracks have emerged, which, if left untended, threaten to destroy some of the most salient features of the site. The dominating feature of the furnace is the central inner furnace. The inner furnace can best be thought of as two truncated cones with their widest sections placed back to back, like an inverted hourglass. The point where the inner furnace is widest is known as the bosh and is visible at the Mill Creek Furnace. Just below the bosh (Fig. 4) is the hole where the tuyère was located. (The tuyère is where air, known as blast, was forced into the furnace; see the section “Early Iron Industry in Ohio and Pennsylvania” above for additional discussion.) Below the tuyère opening is a narrower area, now solidified. This is the crucible and would have represented an area where the molten iron and slag congregated. Below that is the point at which the iron would have emerged onto the casting floor, which would have been covered with a dam stone or clay plug. Just outward from the inner furnace is a red layer of brick. This is the inwall and would have consisted of refractory brick used to insulate the inner furnace, some of which is still visible. Another insulating layer composed of loose sand lined the exterior of the brick lining. Finally, just outward from that are the sandstone blocks used as a shell of the furnace. The area in front of the furnace was known as the casting floor and would have been where the iron castings and pigs were
made. The casting floor extends beyond the excavation area. Just in front of the furnace is a large conglomeration of iron and slag, called a salamander or bear. This represented the last batch of iron and slag from the furnace and still remains at the site. White (1980c, table 3) described the slag at this site as green and stony, but also gave a more detailed analysis (1980a) and a greater color range. Samples (now in the University of Youngstown Department of Sociology and Anthropology collections) from this site that have been collected by archaeologists vary from gray to blue-green to green-brown to brownish black in color and range from glassy to vesicular in appearance. White (1980a, p. 59, table 3) noted that the slag found at Trumbull Furnace was high in sulfur in comparison to other early furnaces, hypothesizing that this may have been due to the use of coal. Pioneer Pavilion, located next to the furnace, was originally constructed in 1822 as a wool carding and fulling mill and operated until 1830 (Butler, 1921, p. 663). It was once a simpler building (Blue et al., 1995, p. 19); additions have been made over time. It is made of locally quarried Massillon sandstone. The stone is coarse-grained and contains nested Liesegang rings. It is likely that the mill race that provided the water power for whatever system the furnace used to force air into the furnace is located between the pavilion and the hillside, possibly under the more recently added restrooms on the southwest side of the pavilion. During construction of additions on the pavilion, pieces of a likely waterwheel were recovered; however, whether that wheel belonged to the carding mill or the blast furnace remains unknown. Both the carding mill and furnace likely used the same mill race to supply their hydropower needs. Source of Iron for Trumbull Furnace It is likely that the source of iron for this furnace was local. Charles Whittlesey showed (1838, cross-section between p. 56 and 57) iron ore and iron strata below the Conglomerate ( = the Sharon Formation) as well as beds of iron in the rocks between the Conglomerate and the Blue Limestone. That would indicate iron sources at and below the level of the furnace. Orton (1884, p. 383) referred the kidney ore which had been mined in the Mahoning Valley to the Sharon shale, an informal shale unit above the sandstones and conglomerate of the Sharon Formation as used here. Indeed, the gray shale adjacent to the furnace, which is below the Massillon sandstone, contains ironstone concretions. Galaida (1941, p. 13) indicated that the original ore used was kidney ore obtained from local outcrops, and that the supply of this ore was short-lived. Belfast, in an unpublished 1979 class report (“A geologic field guide of Mill Creek Park”) noted the presence of an iron ore mine on the east side of Lake Cohasset, below the dam on the downstream side of the dam. Stop 6. Hopewell (Eaton) Furnace, Yellow Creek Park, Struthers, Ohio The Hopewell Furnace, also known as the Eaton (also spelled Heaton) Furnace, is located in Yellow Creek Park in Struthers,
Early industrial geology of western Pennsylvania and eastern Ohio
163
Ohio (Stop 6 on Fig. 20). (The name Hopewell is the most used name for this furnace and is the name used on the historic marker in this park, but the name should not be confused with the older, even better-known Hopewell Furnace located in Berks County, Pennsylvania [see Walker, 1966, for a description of that furnace]). This Mill Creek MetroPark stretches from where Lowellville Road crosses Yellow Creek south to the northeast side of the dam which impounds Lake Hamilton. The furnace is built into the hillside, into an outcrop of the Massillon sandstone, just to the east of Lake Hamilton, in the southernmost section of Yellow Creek Park in Struthers, Ohio. Hopewell Furnace (Fig. 28) is accessible via the park trail that extends southward from a parking area along Wetmore Avenue, an east-west–trending street which roughly bisects this north-south–trending park along Yellow Creek. There is a historic marker for the furnace by the parking area. When curves of the winding trail are taken into account, the distance between the parking area and the furnace approaches 1 mi (1.6 km). The trail is easiest to traverse in dry weather when the stream is low, as it is necessary to cross Yellow Creek at least two times along the path to the furnace. The trail to the furnace site is scenic, however, and of special interest to geologists. Outcrops of medium- to dark-gray shale containing iron-rich beds and concretions can be seen along the stream en route to the furnace. The gray shale, rich in ironstone nodules, is an unnamed shale subjacent to the Massillon sandstone (Stephenson, 1933). The iron-rocks derived from these layers can also be seen as float. These iron-rich rocks can readily be identified by oxidized exterior layers which are red in color. These can be interpreted as kidney or something similar to kidney ore. These iron-rich layers and concretions, however, do not readily fizz in acid, presumably because they are leached of siderite sure to deep weath-
ering. In his archaeological reports, White (1978, p. 392; 1996, p. 234) described the ore used here as being kidney or reniform ore “generally composed of concentric layers or shells made distinct by weathering” and exfoliation, and having a “deep redbrown color.” He also noted (White, 1980c, p. 57) that the ore was found in “pockets or layers and as float material in Yellow Creek” or “from beds of shale” (White, 1996, p. 234). Butler (1921, p174) stated that the “ore found along Yellow Creek” was used as the raw material. So it is reasonable to infer that the iron-rich layers and concretions in the gray shales exposed here are the source of the iron. These layers, however, are not among those shown in typical columns (e.g., fig. 2 in Slucher and Rice, 1994) of the Pottsville. White did some analysis of ore at this furnace site, describing (White, 1982, p. 24) the kidney ore used here as having 39.8% iron. In other papers, he also described the locally collected kidney ore at Hopewell as having an “average iron content of 51.3%” (White, 1980b, p. 513; White, 1979, p. 8). The contact between the shale and the overlying Massillon along Yellow Creek is interesting for a number of reasons. The shale contains sideritic concretions that presumably were the source of iron ore. Also, the rock just below and above the contact with the Massillon appears to preserve some soft-sediment deformation and evidence of penecontemporaneous faulting and slumping (Fig. 29). Alternatively, this faulting might be interpreted as being due to fracture-relief faulting as identified elsewhere by Ferguson (1967). The Massillon sandstone contains anastomosing layers of coal to at least 6 cm in thickness; at least some of these layers are coalified Pennsylvanian trees. The Massillon also contains prominent sets of cross beds and channelform structures. The furnace is built into the hillside above the creek, into an outcrop of the Massillon sandstone, just beyond and uphill of a stone arched-bridge carrying a large pipe. This hillside
Figure 28. Hopewell (Eaton) Furnace, Yellow Creek Park, Struthers, Ohio. View shows tuyère arch (2010 photo). Trail leading to furnace is seen to the left. Staff is marked in 1-dm-high increments.
Figure 29. Contact of Massillon sandstone with subjacent dark-gray shale. Note deformed beds in the top few decimeters of the shale. The white staff is 0.9 dm high.
164
Hannibal et al.
construction (Figs. 30 and 31), a common construction element in blast furnaces in both Ohio and Pennsylvania, allowed for the delivery of ore, limestone, charcoal, and coal from above. This hillside feature is so ubiquitous in furnaces built before the invention of the skip hoist that they are often referred to as “bank” furnaces (White, 1979, p. 6). Additionally, this placement was often found in conjunction with elevation changes in water, which were essential to powering the mechanisms used before steam power to force the air blast into the furnace. An overshot waterwheel here was powered by the flow of Yellow Creek, aided by a high dam (Reese, 1929). The upper layer of material at the base of the furnace is composed of a large amount of slag. Most of the slag seen here, as well as that found during archaeological investigation of the site by White (1980a, p. 57), is a dense, glassy black-colored slag. This furnace is frequently cited in Ohio geological, historical, and cultural literature (the furnace and the ore along Yellow Creek are mentioned in the introductory stanzas of the Bruce Springsteen song, “Youngstown”). It was probably constructed by 1802–1803 (White, 1978, 1996) or at least by 1804 (Butler, 1921, p. 174; Stout, 1944a, p. 28), although Upton (1910, p. 602) cites an 1807 construction date. The Hopewell Furnace has been accepted by some authors as the earliest blast furnace west of the Alleghenies (Upton, 1910, figure caption on p. 593, but also see p. 602; Melnick, 1976, p. 170; White, 1996 and other references; Deblasio, 2010, p. 75). This claim, however, conflicts with claims for earlier furnaces west of the Alleghenies (Swank, 1878, p. 49; Mathews, 1885, p. 177) in western Pennsylvania, or “over the Alleghenies” in western Pennsylvania (Bining, 1938, p. 61–64). White, the principal modern proponent of this claim, was familiar with iron furnaces in western Pennsylvania (see White, 1986), so these claims may rest on one’s interpretation of the boundaries of the Alleghenies. The first trans-Appalachian
Figure 30. Edge of Hopewell (Eaton) Furnace (left of photo) showing where it is in contact with Massillon sandstone outcrop (right of photo). Staff is marked in 1-dm-high increments.
iron furnace in western Pennsylvania, the Alliance Furnace (blown-in in 1789) is located near the western boundary of the Allegheny Mountain Section as defined by the Pennsylvania Geological Survey physiographic map (reproduced in Shultz, 1999, p. 342; Briggs, 1999, p. 364; and elsewhere). If the Allegheny Mountain Section is taken as a benchmark, then the Alliance Furnace is indeed just west of the western boundary of that section in Fayette County. Based on this criterion, the Alliance is indeed the first west of the Alleghenies. However, the confusion over this boundary and the exact location of the Alliance Furnace in relation to the boundary contribute to this ongoing dilemma. It should be noted, however, that the historic marker for Hopewell Furnace in the park claims that the furnace was “one of the first west of the Allegheny Mountains,” not the first. The Hopewell Furnace has also been noted as the first industry in Ohio, but that claim ignores the Ohio gristmills and sawmills established before 1802. Gristmills and sawmills were the first industries (manufacturers) in eastern Ohio as well as in western Pennsylvania (Hazen, 1908, p. 114). The Hopewell Furnace was well studied by Youngstown State University archaeologist John R. White (1937–2009), who published a series of articles describing aspects of the furnace, (e.g., White, 1977, 1978, 1980b, 1980c, 1982, 1996). It is important for its very early construction date and for its early use of coal as well as charcoal as fuel and as the first iron furnace in Ohio. White provided a series of maps and diagrams of the site (e.g., in White, 1978, and especially White, 1996, which contain photos of the site as recently excavated). The Hopewell Furnace originally used the abundant forests surrounding the furnace to produce charcoal, which the furnace initially used exclusively as fuel (Butler, 1921, p. 174). Charcoal is a great fuel and with the availability of the virgin forests of
Figure 31. Historic photo of Hopewell Furnace along Yellow Creek during the winter of 1900–1901 (photo courtesy of the Struthers Historical Society). A version of this photo was published in Upton (1910, p. 593). Compare this figure to Fig. 28 to see how structure has changed since 1901 and to fig. 5 in White (1996) to see how it has deteriorated, in part due to vandalism, since the 1970s.
Early industrial geology of western Pennsylvania and eastern Ohio Pennsylvania/Ohio, charcoal was originally readily available. White (1996, p. 240) estimated that this charcoal production would have used ~240–250 acres of hardwood timber per year for the Hopewell. As the timber around the furnace was rapidly being depleted, the owners attempted to supplement the charcoal fuel with locally available bituminous coal. According to White (1980a) the high sulfur content that the coal was adding to the iron, and the inability of the flux to effectively remove it, caused the furnace to be abandoned after it underwent a structural failure around 1808. The furnace utilized water power directed by a headrace (White, fig. 1) originating at the site of the present dam. This present high dam, impounding Lake Hamilton, was built much later (1907) than the furnace. Stop 7. Struthers Historical Society, Struthers, Ohio The Struthers Historical Society Museum (Stop 7 on Fig. 20) is located in a historic (1884) house at 50 Terrace Street, Struthers, Ohio, within a few blocks of the northern side of Yellow Creek Park. The museum is open by appointment. There is a mining car in the yard next to the museum building. The museum itself includes artifacts related to mining and iron manufacture in the area, including artifacts such as iron, ceramics, and bone excavated at the Hopewell Furnace under the supervision of John White. The museum also includes a bound set of the local newspaper and various reprints, photographs, and other items related to the furnace and the town of Struthers. ACKNOWLEDGMENTS Ray Novotny, Gary Meiter, Robert Orr, Julie Pantelas, and other staff at Mahoning County MetroParks provided help, permission, or aid in the study materials in the park. Theresa Kalka, Hiram College, and Veronica Fusco, Oberlin College, helped to uncover the millstone in the stream at Mill Creek Park in 2009 and 2010. Kathleen Farago, Cleveland Heights/ University Heights Public Library, and Cleveland and Maple Heights High students Terryn Mathis and Marcus Jackson also helped in the field. Matt O’Mansky, Department of Sociology and Anthropology, Youngstown State University, made archaeological materials from the Mill Creek Furnace available, as well as files of John White; Marian Kutlesa, Struthers Historical Society, provided images and information on iron furnaces. Natalie Simon, McConnells Mill State Park, Tom Anderson, University of Pittsburgh, John Harper, Pennsylvania Geological Survey, Ernie Slucher, U.S. Geological Survey, Ann G. Harris, Youngstown State University, and David Saja, Doug Dunn, Wendy Wasman, and Evan Scott, Cleveland Museum of Natural History, provided additional help, references, and other aid. Kathleen Farago and Lars Benthien, Case Western Reserve University, proofread versions of the text. The manuscript was further improved by the formal reviews of Charles D. Hockensmith, Frankfort, Kentucky, and Ann G. Harris.
165
REFERENCES CITED Arkell, W.J., and Tomkeieff, S.I., 1953, English rock terms, chiefly as used by miners and quarrymen: London, Oxford University Press, 139 p. Aslan, A., and Autin, W.J., 1999, Evolution of the Holocene Mississippi River floodplain, Ferriday, Louisiana: Insights of the origin of fine-grained floodplains: Journal of Sedimentary Research, v. 69, no. 4, p. 800–815. Ball, D.B., and Hockensmith, C.D., 2007, Millstone studies: papers on their manufacture, evolution, and maintenance: Murray, Kentucky, and East Meredith, New York, Symposium on Ohio Valley Urban and Historic Archaeology and the Society for the Preservation of Old Mills, 223 p. Belfast, M.A., 1979, A geologic field guide of Mill Creek Park: Unpublished Youngstown State University report, 62 p. Berg, T.M., 1986, A sesquicentennial story: Early millstone quarry in Tioga County: Pennsylvania Geology, v. 17, no. 1, p. 3–6. Berg, T.M., Edmunds, W.E., Geyer, A.R., et al., compilers, 1980, Geologic map of Pennsylvania: Pennsylvania Geological Survey, 4th ser., Map 1, 2nd ed., 3 sheets, scale 1:250,000. Berkheiser, S.W., Jr., 1999, Nonmetals—Limestone-dolostone: Specialty uses, in Shultz, C.H. ed., The Geology of Pennsylvania: Pennsylvania Geological Survey and Pittsburgh Geological Society, Special Publication 1, p. 628–637. Bining, A.C., 1938, Pennsylvania iron manufacture in the eighteenth century: Harrisburg, Pennsylvania Historical Commission, v. 4, 227 p. Blue, F.J., Jenkins, W.D., Lawson, H.W., and Reedy, J.M., 1995, Mahoning memories: A history of Youngstown and Mahoning County: Youngstown, Donning, 192 p. Briggs, R.P., 1999, Appalachian Plateaus Province and the eastern lake section of the Central Lowland Province, in Shultz, C.H., ed., The Geology of Pennsylvania: Harrisburg and Pittsburgh, Pennsylvania Geological Survey and Pittsburgh Geological Society, Special Publication 1, p. 362–377. Butler, J.G., 1921, History of Youngstown and the Mahoning County, Ohio: Chicago, American Historical Society, v. 1, p. 173–670. Camp, M.J., 2006, Roadside geology of Ohio: Missoula, Montana, Mountain Press, 411 p. Carlson, E.H., 1991, Minerals of Ohio: Ohio Division of Geological Survey Bulletin 69, 155 p. Chamberlin, T.C., and Salisbury, R.D., 1909, A college text-book of geology: New York, Henry Holt and Company, 978 p. Chance, H.M., 1880, The geology of Clarion County: Pennsylvania Geological Survey, 2nd ser., Report VV, 232 p. Coyle, P.R., 2003, Structural and lithologic controls on the distribution of the buhrstone siliceous iron ore, Allegheny Plateau, west central Pennsylvania [master’s thesis]: Pittsburgh, Pennsylvania, University of Pittsburgh, 151 p. Cuvier, G., 1815, Essay on the theory of the Earth, translated by Robert Kerr: Edinburgh, W. Blackwood, 332 p. Dana, J.D., 1884, Manual of mineralogy and lithology: New York, John Wiley & Sons, 474 p. Deblasio, D.M., 2010, Youngstown’s Idora Park: Creating a fantasyland in an industrial landscape: Ohio History, v. 117, p. 74–92, doi:10.1353/ ohh.2010.0011. Durant, S.W., and Durant, P.A., 1877, History of Lawrence County, Pennsylvania; with illustrations descriptive of its scenery, palatial residences, public buildings, fine blocks, and important manufactories: Philadelphia, L.H. Everts Co., 228 p. Edenborn, H.M., Gerke, T.L., and Thompson, R.P., 2009, Preliminary analysis of historic charcoal blast furnace slags from northwestern Pennsylvania: Geological Society of America Abstracts with Programs, v. 41, no. 4, p. 60. Ferguson, H.F., 1967, Valley stress release in the Allegheny Plateau: Bulletin of the Association of Engineering Geologists, v. 4, no. 1, p. 1–17. Fisher, Q.J., Raiswell, R., and Marshall, J.D., 1998, Siderite concretions from nonmarine shales (Westphalian A) of the Pennines, England: controls on their growth and composition: Journal of Sedimentary Research, v. 68, no. 5, p. 1034–1045. Fleeger, G.M., Bushnell, K.O., and Watson, D.W., 2003, Moraine and McConnells Mill State Parks, Butler and Lawrence Counties—Glacial lakes and drainage changes: Pennsylvania Geological Survey, 4th ser., Park Guide 4, 12 p. Fletcher, S.W., 1950, Pennsylvania agriculture and country life: 1640–1840: Harrisburg, Pennsylvania Historical and Museum Commission, 605 p.
166
Hannibal et al.
Foos, A.M., ed., 2003, Pennsylvanian Sharon Formation, past and present: Sedimentology, hydrogeology, and historical and environmental significance: A field guide to Gorge Metro Park, Virginia Kendall Ledges in the Cuyahoga Valley National Park, and other sites in Northeast Ohio: Ohio Division of Geological Survey Guidebook 18, 67 p. Foster, [J.W.], 1838, Report of Mr. Foster: Ohio Geological Survey Second Annual Report, p. 73–107. Galaida, E., 1941, Mill Creek Park: Cleveland, [Steffan Printing Co.], 104 p. Garber, D.W., 1970, Waterwheels and millstones: A history of Ohio gristmills and milling: Ohio Historical Society, 139 p. Gerstner, P., 1994, Henry Darwin Rogers, 1808–1866, American geologist: Tuscaloosa, University of Alabama Press, 311 p. Hahn, T.F., and Kemp, E.L., 1994, Cement mills along the Potomac River: Institute for the History of Technology & Industrial Archeology, Monograph Series, v. 2, no 1, 90 p. Hannibal, J.T., and Evans, K.R., 2010, Civil War and cultural geology of southwestern Missouri, part 1: The geology of Wilson’s Creek Battlefield and the history of stone quarrying and stone use, in Evans, K.R., and Aber, J.S., eds., From Precambrian Rift Volcanoes to the Mississippian Shelf Margin: Geological Field Excursions in the Ozark Mountains: Geological Society of America Field Guide 17, p. 39–68, doi:10.1130/2010.0017(04). Hannibal, J.T., and Saja, D.B., 2009, Millstones along the Cuyahoga and other streams of the Western Reserve: rock type, provenance, and trends in usage: Geological Society of America Abstracts with Programs, v. 41, no. 4, p. 66. Harper, J.A., and Ward, A.N., Jr., 1999, Rocks, oil, gravel, iron: The surficial, bedrock, and economic geology of Venango County, Pennsylvania, Guidebook for the Pittsburgh Geological Society Field Trip 1 May 1999: Pittsburgh, Pennsylvania, Pittsburgh Geological Society, 61 p. Harper, R.E., 1991, The transformation of western Pennsylvania, 1770–1800: Pittsburgh, University of Pittsburgh Press, 273 p. Harris, I., 1837, Harris’ Pittsburgh Business Directory for the year 1837 including the names of all the merchants, manufacturers, mechanics, professional, & men of business of Pittsburgh and its vicinity: Pittsburgh, Isaac Harris, 340 p. Hazen, A.L., editor and compiler, 1908, 20th century history of New Castle and Lawrence County, Pennsylvania and representative citizens: Chicago, Richmond-Arnold Publishing Co., 1015 p. Hildreth, S.P., 1838, Report of Dr. S.P. Hildreth: First annual report on the Geological Survey of the State of Ohio: Columbus, Samuel Medary, Printer to the State, p. 25–63. Hockensmith, C.D., 2007, Ohio buhr millstones: the Flint Ridge and Raccoon Creek Quarries, p. 134–143, in Ball, D.B., and Hockensmith, C.D., Millstone Studies: Papers on their Manufacture, Evolution, and Maintenance: Special Studies No. 1, Murray, Kentucky, Symposium on Ohio Valley Urban and Historic Archaeology, and East Meredith, New York Society for the Preservation of Old Mills. Hockensmith, C.D., 2008, Millstones from Ohio and Pennsylvania imported into Kentucky: Raccoon buhrs and Laurel Hill stones, in Hockensmith, C.D., ed., Foreign and domestic millstones used in Kentucky: Papers examining archival records: Clay City, Kentucky, Kentucky Old Mill Association, p. 39–54. Hockensmith, C.D., 2009a, The millstone industry: A summary of research on quarries and producers in the United States, Europe, and elsewhere: Jefferson, North Carolina, McFarland & Co., 269 p. Hockensmith, C.D., 2009b, The millstone quarries of Powell County, Kentucky: Contributions to Southern Appalachian Studies 24: Jefferson, North Carolina and London, McFarland Publishing, 202 pages. Hughes, W.C., 1851, The American miller and millwright’s assistant, revised edition: Philadelphia, Henry Cary Baird, 223 p. Inners, J.D., 1999, Metallic mineral deposits—sedimentary and metasedimentary iron deposits, Chapter 40a, in Shultz, C.H., ed., Geology of Pennsylvania: Pennsylvania Geological Survey and Pittsburgh Geological Society, Special Publication 1, p. 556–565. Lepper, B.T., Yerkes, R.W., and Pickard, W.H., 2001, Prehistoric flint procurement strategies at Flint Ridge, Licking County, Ohio: Midcontinental Journal of Archaeology, v. 26, p. 53–78. Leung, F.L., 1981, Grist and flour mills in Ontario: from millstones to rollers, 1780s–1880s: Ottawa, National Historic Parks and Sites Branch, Environment Canada, History and Archaeology 53, 293 p. Light, J.D., 2000, A field guide to the identification of metal, in Karklins, K., ed., Studies in Material Culture Research: The Society for Historic Archaeology, California, Pennsylvania, p. 3–19.
Mac Cabe, J.P.B., 1837, Directory, Cleveland and Ohio City, for the years 1837–38: Cleveland, Sanford & Lott, 144 p. Mahoning Valley Historical Society, 1876, Historical collections of the Mahoning Valley: Youngstown, Mahoning Valley Historical Society, 524 p. Mather, W.W., 1838, First Annual Report of the Ohio Geological Survey: Columbus, Samuel Medary Printer, 134 p. Mathews, A., 1885, Pittsburgh, II, an outline of the city’s industrial and commercial development: Magazine of Western History, p. 175–191. Melnick, J.C., 1976, The green cathedral: History of Mill Creek Park, Youngstown, Ohio: Youngstown, Youngstown Lithographing, 446 p. Miller, B.L., 1934, Limestones of Pennsylvania: Pennsylvania Geological Survey, Fourth Series, Bulletin M 20, 729 p. Moldenke, R., 1920, Charcoal iron: Lime Rock, Connecticut, Salisbury Iron Corporation, 64 p. Moore, S.E., Ferrell, R.E., Jr., and Aharon, P., 1992, Diagenetic siderite and other ferroan carbonates in a modern subsiding marsh sequence: Journal of Sedimentary Petrology, v. 62, no. 3, p. 357–366. Murphy, R.E., and Murphy, M., 1937, Pennsylvania: a regional geography: Harrisburg, Pennsylvania, Pennsylvania Book Service, 591 p. Mursky, G.A., and Thompson, R.M., 1958, A specific gravity index for minerals: Canadian Mineralogist, v. 6, p. 273–287. Neuendorf, K.K.E., Mehl, J.P., and Jackson, J.A., 2005, Glossary of geology, Fifth Edition: Alexandria, Virginia, American Geological Institute, 779 p. Newberry, J.S., 1870a, Report on the progress of the Geological Survey of Ohio in 1869, Part 1: Geological Survey of Ohio, p. 3–53. Newberry, J.S., 1870b, Chart of geological history [chart bound in at the end of] Report on the progress of the Geological Survey of Ohio in 1869 Part 1: Geological Survey of Ohio, p. 3–53. Newberry, J.S., 1878, Report on the geology of Mahoning County. Ohio: Ohio Geological Report, v. 3, no. part 1, p. 781–814. Orton, E., 1884, The iron ores of Ohio: Report of the Geological Survey of Ohio, vol. 5: Economic Geology, p. 371–435. Pennsylvania Bureau of Topographic and Geologic Survey, Department of Conservation and Natural Resources, 2001, Bedrock Geology of Pennsylvania, edition 1.0, digital map. Retrieved from Internet 30 September 2004; http://www.dcnr.state.pa.us/topogeo/map1/bedmap.aspx, DL Data: pageoexp.zip. Pye, K., Dickson, J.A.D., Shiavon, N., Coleman, M.L., and Cox, M., 1990, Formation of siderite-Mg-calcite-iron sulfide concretions in intertidal marsh and sandflat sediments, north Norfolk, England: Sedimentology, v. 37, p. 325–343, doi:10.1111/j.1365-3091.1990.tb00962.x. Rau, J.L., 1970, Pennsylvanian System of northeast Ohio, in Banks. P.O., and Feldmann, R.M., eds., Guide to the geology of northeastern Ohio: Cleveland, Northern Ohio Geological Society, p. 69–124. Reese, V.S., 1929, Iron history of district reads like fiction: Struthers Journal, December 18, p. 1. Rogers, H.D., 1836, Annual Report of the State Geologist: Harrisburg, Samuel Patterson, 22 p. Rogers, H.D., 1840, Fourth annual report on the geological survey of the state of Pennsylvania: Harrisburg, Holbrook, Henlock, and Bratton, 215 p. Ruppert, L.F., Trippi, M.H., and Slucher, E.R., 2010, Correlation chart of Pennsylvanian rocks in Alabama, Tennessee, Kentucky, Virginia, West Virginia, Ohio, Maryland, and Pennsylvania showing approximate position of coal beds, coal zones, and key stratigraphic units: U.S. Geological Survey Scientific Investigations Report 2010-5152, 9 p., 3 plates. Safford, J.M., 1880, Millstones: U.S. Centennial Commission Report and Awards, v. 3, p. 176–182. Saja, D.B., and Hannibal, J.T., 2009, Late 18th and early 19th century granite millstone production in northeastern Ohio: Geological Society of America Abstracts with Programs, v. 41, no. 4, p. 66. Schallenberg, R.H., and Ault, D.A., 1977, Raw materials supply and technological change in the American charcoal iron industry: Technology and Culture, v. 18, no. 3, p. 436, doi:10.2307/3103901. Sharp, M.B., and Thomas, W.H., 1966, A guide to old stone blast furnaces in western Pennsylvania: Pittsburgh, Historical Society of Western Pennsylvania, 90 p. Shultz, C.H., ed., 1999, Geology of Pennsylvania: Pennsylvania Geological Survey and Pittsburgh Geological Society, Special Publication 1, 888 p. Skema, V., 2005a, Stops 10 and 11—US 422 at Moravia Street interchange, in Fleeger, G.M. and Harper, J.A., eds., Type Sections and Stereotype Sections: Glacial and Bedrock Geology in Beaver, Lawrence, Mercer
Early industrial geology of western Pennsylvania and eastern Ohio and Crawford Counties, 70th Annual Field Conference of Pennsylvania Geologists, Pennsylvania Geological Survey, p. 129–144. Skema, V., 2005b, Stop 12, US 422 at Toll 60 interchange, in Fleeger, G.M. and Harper, J.A., eds., Type Sections and Stereotype Sections: Glacial and Bedrock Geology in Beaver, Lawrence, Mercer and Crawford Counties, Guidebook for the 70th Annual Field Conference of Pennsylvania Geologists, p. 145–154. Slucher, E.R., 2002a, Reconnaissance bedrock geology of the Campbell, OhioPennsylvania, quadrangle [revised edition]: Ohio Division of Geological Survey Digital Map Series BG-2 [1-p. map]. Slucher, E.R., 2002b, Reconnaissance bedrock geology of the Youngstown, Ohio, quadrangle [revised edition]: Ohio Division of Geological Survey Digital Map Series BG-2 [1-p. map]. Slucher, E.R., and Rice, C.L., 1994, Key rock units and distribution of marine and brackish water strata in the Pottsville Group, northeastern Ohio, in Rice, C.L., ed., Elements of Pennsylvanian stratigraphy, Central Appalachian Basin: Geological Society of America Special Paper 294, p. 27–40. Smyth, P., 1957, Fusulinids from the Pennsylvanian rocks of Ohio: Ohio Journal of Science, v. 57, p. 257–283. Stephenson, E.L., 1933, The Geology of the Youngstown region [master’s thesis]: Columbus, Ohio State University, 129 p. Stone, R.W., 1932, Building stones of Pennsylvania: Pennsylvania Geological Survey, Bulletin M 15, 316 p. Stout, W., 1927, Geology of Vinton County: Geological Survey of Ohio, Bulletin 31, 402 p. Stout, W., 1944a, The iron ore bearing formations of Ohio: Geological Survey of Ohio, Fourth Series, Bulletin 45, 230 p. Stout, W., 1944b, Sandstones and conglomerates in Ohio: Ohio Journal of Science, v. 44, p. 75–88. Stout, W., and Schoenlaub, R.A., 1945, The occurrence of flint in Ohio: Ohio Division of Geological Survey Bulletin 46, 110 p. Swank, J.M., 1878, Introduction to a history of ironmaking and coal mining in Pennsylvania: Philadelphia, J.M. Swank, 125 p. Szmuc, E.J., 1957, Stratigraphy and paleontology of the Cuyahoga Formation of northern Ohio [Ph.D. dissertation]: Columbus, Ohio State University, v. 1, 230 p. Tucker, D.G., 1984, Millstone making in Scotland: Proceedings of the Society of Antiquaries of Scotland, v. 114, p. 539–556. Upton, H.T., 1910, History of the Western Reserve, v. 1: Chicago, Lewis Publishing Co., 709 p. Walker, J.E., 1966, Hopewell Village: The dynamics of a nineteenth century iron-making community: Philadelphia, University of Pennsylvania Press, 526 p.
167
Ward, O., 1993, French millstones: Notes on the millstone industry at La Fertésous-Jouarre: Reading, England, International Molinological Society, 75 p. Way, J.H., 1999, Appalachian Mountain section of the Ridge and Valley province, in Shultz, C.H., ed., Geology of Pennsylvania: Pennsylvania Geological Survey and Pittsburgh Geological Society, Special Publication 1, p. 352–361. White, J.R., 1977, X-ray fluorescent analysis of an early Ohio blast furnace slag: Ohio Journal of Science, v. 77, p. 186–188. White, J.R., 1978, Archaeological and chemical evidence for the earliest American use of raw coal as a fuel in ironmaking: Journal of Archaeological Science, v. 5, p. 391–393, doi:10.1016/0305-4403(78)90059-6. White, J.R., 1979, Nineteenth century blast furnaces of Mercer County: A postscript: Mercer County History, v. 9, p. 3–20. White, J.R., 1980a, Historic blast furnace slags: Archaeological and metallurgical analysis: Journal of the Historical Metallurgy Society, v. 14, p. 55–64. White, J.R., 1980b, The Eaton blast furnace: Current Anthropology, v. 21, no. 4, p. 513–514, doi:10.1086/202504. White, J.R., 1980c, Preliminary archaeological examination of Ohio’s first blast furnace: The Eaton (Hopewell): Ohio Journal of Science, v. 80, p. 52–58. White, J.R., 1982, Analysis and evaluation of the raw materials used in the Eaton (Hopewell) furnace: Ohio Journal of Science, v. 82, p. 23–27. White, J.R., 1986, Survey and analysis of pre-1850 blast furnace sites in western Pennsylvania: Proceedings of the Symposium on Ohio Valley Urban and Historic Archaeology, v. 4, p. 74–85. White, J.R., 1996, The rebirth and demise of Ohio’s earliest blast furnace: An archaeological postmortem: Midcontinental Journal of Archaeology, v. 21, p. 217–245. Whittlesey, C., 1838, Mr. Whittlesey’s report: Ohio Division of Geological Survey, Second Annual Report, p. 41–71. Williams, H.Z., 1882, History of Trumbull and Mahoning County, with illustrations and biographical sketches: Cleveland, H.Z. Williams & Brothers, 153 p. Willis, B., 1886, Notes on the samples of iron ore collected in Ohio: U.S. 10th Census, v. 15, p. 235–243. Wright, N., 1884, Iron manufacture of Ohio: Ohio Mining Journal, v. 2, no. 3, p. 129–135.
MANUSCRIPT ACCEPTED BY THE SOCIETY 21 JANUARY 2011
Printed in the USA
The Geological Society of America Field Guide 20 2011
The old, the crude, and the muddy: Oil history in western Pennsylvania Kristin M. Carter Pennsylvania Geological Survey, Pittsburgh, Pennsylvania 15222, USA Kathy J. Flaherty ABARTA Oil & Gas Company, Pittsburgh, Pennsylvania 15238, USA
ABSTRACT Western Pennsylvania is rich in oil history, and many of the fledgling petroleum industry’s “firsts” happened right here between the late nineteenth and early twentieth centuries. First and foremost, it is home to the Drake Well, the first-ever economic well intentionally drilled to produce oil. In addition, western Pennsylvania has bragging rights to various industry advancements, from the way geologic samples were collected and interpreted, to how wells were drilled and stimulated for production, to early oil refining techniques, marketing, and production. In this field guide, we discuss the petroleum geology and history of western Pennsylvania in general and three important oil-producing sites in particular: (1) the McClintock #1 Well, Rouseville; (2) the Drake Well, Titusville; and (3) Muddy Creek oil field, Prospect. Each of these sites produced oil from Upper Devonian reservoirs and spurred additional petroleum exploration and development in the central Appalachian basin.
INTRODUCTION From Pittsburgh to Titusville (Fig. 1), we will be driving through old oil country, from where, at one time in history, the world’s supply of oil was produced (Flaherty, 2003). Few traces of this early oil activity remain, but in its heyday, the fledgling petroleum industry had a profound effect on the establishment, growth, and development of much of western Pennsylvania. Settlements along the creeks and Allegheny River became important, thriving shipping ports. People and goods moved in one direction, and oil moved in the other. Fortunes were earned and spent, investments were made, and structures were built where lives and industries dependent upon petroleum evolved. The remainder of this introductory section is largely excerpted from Flaherty (2003).
Traveling north from Pittsburgh, we will also be moving back in time relative to the progress of the modern oil industry. Starting with Drake’s well at Titusville in 1859, oil exploration and development grew outward, sometimes taking long leaps, and sometimes making short hops, through Pennsylvania, the Appalachian basin, and beyond. As an example, oil was not produced in the fields nearest Pittsburgh until the 1880s. Those who developed the oil fields in Allegheny County in the 1880s were many of the same explorers and innovators of the early oil patches in Venango County in the preceding decades. A fundamental reason for the lack of successful oil well drilling in Allegheny County prior to the 1880s was that early drilling equipment was not capable of penetrating the oil sands here. The drilling equipment and methods used in the 1860s in
Carter, K.M., and Flaherty, K.J., 2011, The old, the crude, and the muddy: Oil history in western Pennsylvania, in Ruffolo, R.M., and Ciampaglio, C.N., eds., From the Shield to the Sea: Geological Field Trips from the 2011 Joint Meeting of the GSA Northeastern and North-Central Sections: Geological Society of America Field Guide 20, p. 169–185, doi:10.1130/2011.0020(08). For permission to copy, contact
[email protected]. ©2011 The Geological Society of America. All rights reserved.
169
170
Carter and Flaherty
Figure 1. Location of field trip area in western Pennsylvania.
Venango County and vicinity were relatively simple; many wells were “kicked down” using a spring pole and simple tools suspended on a rope. This approach was adequate for the Titusville area, where Venango Group oil sands typically occur at depths of 122 m (400 ft) (Fig. 2) or less. Here, the only special requirements to kick down a well were time, determination, and brawn. Time, determination, and brawn, however, were no match for the depths of oil-bearing sands in the greater Pittsburgh area, which generally exceed 457 m (1500 ft). The desire to drill deeper led to the use of new materials and modifications in drilling equipment design; many took the lead of “Colonel” Edwin Laurentine Drake, using steam engines for power and wooden derrick towers for drawing tools up and down the hole. As time and experience increased, tools were adapted to the special needs of oil well operators, and so by the early 1890s, steel derricks were popular in the oil fields around Pittsburgh. They were stronger than wood and virtually indestructible, perfect for supporting the tools needed for deeper drilling. Early Oil Experiments in Western Pennsylvania Samuel Martin Kier, a partner in the Mechanic Canal Transportation Line, operated canal boats between Pittsburgh and Philadelphia, but his interests took him far beyond the waterways.
Together with his partner B.F. Jones, Kier purchased an iron furnace and forge in the Allegheny Mountains in 1847. The iron works were relocated to Pittsburgh and, with the addition of two partners, became the American Iron Works, precursor to Jones and Laughlin Steel (Frasure, 1952). Kier’s entrepreneurial spirit also led him to petroleum, and his father’s salt well. Salt, an important food preservative, was an essential commodity. Throughout the Appalachian region, salt was obtained by drilling for salt water and heating it until the salt crystallized. Salt wells operated by Mr. Thomas Kier, Samuel’s father, near Tarentum, Pennsylvania (32 km, or 20 mi up the Allegheny River from Pittsburgh) were occasionally tainted with crude oil. In 1846, the senior Mr. Kier drilled a salt well with more than a mere taint of oil—he quite literally had a greasy nuisance on his hands. Unaware of any value for this unwelcome substance, Kier’s father let the oil discharge onto the ground, where it often found its way to the nearby Pennsylvania Canal. A few years later (ca. 1849), Samuel M. Kier’s wife became ill, and her doctor recommended “American Oil” to ease her consumptive symptoms. The substance looked and smelled so much like the waste his father was disposing of onto the ground in Tarentum that Kier concluded they were the same substance. Thus, Samuel M. Kier’s career as a hawker of medicinal oil began, and he marketed the product through local apothecary shops.
Oil history in western Pennsylvania
171
Figure 2. Generalized stratigraphic cross-section illustrating the relative depths of oil-bearing sandstones from the northern portion of Allegheny County north to Venango County (modified from Flaherty, 2003).
172
Carter and Flaherty
Oil produced from the Tarentum salt well far outpaced Kier’s market for sales, so he began a study of how it could be burned without its characteristic unpleasant smoke and odor. Kier made some contacts and learned from a Philadelphia chemist that distilling the liquid would transform the crude into a quality illuminant. With that advice, Kier set up a distillery on Seventh Avenue near Grant Street in downtown Pittsburgh and was the first to distill crude oil in Pennsylvania in 1852 (Carter and Sager, 2010). Distillation did not render the crude, now called “carbon oil,” to be either smokeless or odorless when burned in common oil lamps. Through trial and error, though, Kier successfully treated the oil using acids and a “Virna burner,” a lamp he invented, so that the distilled crude could be burned smoke-free and provide a strong, radiant light (Henry, 1873). Alas, the odor emitted from Kier’s operations was so pungent that city authorities invited Kier to relocate his business outside of Pittsburgh due to fears of fire and explosion. Kier moved his oil works a short distance up the Allegheny River to Lawrenceville, a suburb of Pittsburgh. Natural Gas in the Pittsburgh Area History shows us that the benefits of using the “other” form of petroleum—natural gas—in manufacturing were recognized by the mid 1870s. Fuel economy and improved product quality were just two such advantages to using gas over coal. The following examples illustrate this evolving appreciation for natural gas in western Pennsylvania. John B. Pearse wrote of the first application of natural gas to iron manufacture in 1875 (Giddens, 1938). He described the operation of the Siberian Mill in Leechburg, Armstrong County, Pennsylvania, established in 1872 by Rogers and Birchfield. The mill had six single puddling furnaces, six heating furnaces, six trains of rolls, two steam hammers, one refinery, and two “knobbling fires” (forges). Natural gas fueled the annual production of 2540 tonnes (t) (2800 tons) of tin plate, sheet iron, and other products at this facility. A three-inch pipeline connected the mill with a substantial gas well in nearby Ladnorsville, which Rogers and Birchfield had purchased from the oil seekers who abandoned it as a failure. Rogers and Birchfield found that, compared to coal furnaces, the heat produced in a gas furnace was more intense, could be adjusted readily, and resulted in more homogeneous iron. They estimated that the savings in fuel amounted to 54 t (60 tons) of coal per day, and even found that a cheaper grade of iron could be used successfully (Pearse, 1876). Noting the efficiency of natural gas in the manufacture of iron, both the Spang, Chalfont and Company works at Sharpsburg and Graff, Bennett and Company works at Allegheny City made arrangements with J.J. Vandergrift’s Natural Gas Company, Ltd., to pipe the gas from the Ladnorsville well to their respective plants. Fear of a natural gas pipeline explosion by the farmers between Ladnorsville and Sharpsburg resulted in the inclusions of many bends in the pipeline to avoid crossing the farms. When the six-inch pipeline was completed, gas burned at Sharpsburg
20 minutes after opening the well at Ladnorsville. The 27-km (17-mi) line fed gas at a rate of 53 thousand cubic feet (Mcf) per hour, enough to run the 45 furnace and boiler connections at the Sharpsburg facility and 25 furnaces at Graff, Bennett and Company in Allegheny City (Pearse, 1876). Some manufacturers, such as J. Painter and Sons, went to great lengths to secure steady supplies of natural gas. In 1884, they drilled for gas on their manufacturing site, located on the south side of Pittsburgh just a short distance from the point where the Three Rivers meet, but found only salt water. Consequently, they procured gas from the 35-km (22-mi) long, six-inch diameter pipeline leading from the McGuigan #1 well in Washington County directly to the J. Painter and Sons Iron Works (Carll, 1886). The McGuigan #1 was completed in 1882 and was the first very productive gas well in Washington County (Carll, 1886). In addition to serving the Painter Iron Works, gas from the McGuigan well also provided fuel for the Taylor Salt Works located near the iron manufacturer, and to 50 families along the pipeline route (Carll, 1886). George Westinghouse, fascinated by the natural gas industry, drilled successful wells at several of his manufacturing sites in the greater Pittsburgh area. The first well that Westinghouse contracted was on the grounds of his home, “Solitude,” in Pittsburgh. The well was drilled between late December 1883 and at the end of February 1884. The well yielded enough gas for Solitude and the homes of Westinghouse’s neighbors, but Westinghouse wanted to continue drilling. The driller, Mr. Gillespie, described the results of this deepening by recalling, “We struck such a volume of gas that it blew the tools out and ripped off the casing head with such a roar and racket that nobody could hear his own ears, within a block.” The once-peaceful neighborhood was lit with a 35-m (100-ft) high roaring torch for weeks (Prout, 1922). Westinghouse applied for many patents related to the drilling and production of natural gas, the most significant of which was the automatic cut-off regulator, an important safety device in the young gas-distribution industry (Prout, 1922). He established the Philadelphia Company and drilled hundreds of wells in southwestern Pennsylvania before leaving the business in the late 1890s. FIELD TRIP OBJECTIVES The objectives of this field trip are threefold: (1) to discuss the subsurface geology of western Pennsylvania as it relates to shallow oil and gas production; (2) to experience historical sites representative of early oil ingenuity; and (3) to participate in a dedication ceremony at the Drake Well Museum, where original rock cuttings collected at well sites in the 1870s by members of the Second Geological Survey of Pennsylvania will be officially presented to museum staff for display and archival. We intend to fully immerse attendees in both science and serendipity—both of which have played significant roles in the success of petroleum production here in the heart of the Appalachian basin, as well as the world.
Oil history in western Pennsylvania
173
STOP 1: MCCLINTOCK # 1 WELL Hamilton McClintock was one of the first settlers on Oil Creek. With much effort, he farmed the unimpressive, rocky land along the floodplain. The McClintock farm had an oil spring that discharged into the creek. McClintock built a dam around this spring, and the family would retrieve the oil by soaking flannel rags in the oil pool and wringing them into a bucket. The annual yield of the oil spring was ~1 barrel (Bbl), which they sold for $0.75–1.00 per gallon to the bottlers of “Seneca Oil.” Operators began drilling the McClintock Farm in 1859. The farm and surroundings became the town of McClintockville, center of the oil patch. This small town was described as everything from a crowded shantytown to an attractive town with a two-story hotel and a bridge over Oil Creek. One visitor noted that McClintockville:
Figure 3. The famous Coal Oil Johnny Farm, Oil Creek, Pennsylvania, 1861 (The Rotograph Co., New York City, Printed in Germany, no date).
… has the appearance of a California ranch or settlement. A hundred or more rough board shanties have been erected, and on every side you will see carpenters busy with barns, sheds, and houses of every description the idea predominant, however, to do nothing permanent. There must be a lack of faith somewhere in this oil business, or people would build more enduring structures, and those better adapted for the winter blasts. All are making haste to be rich, and the whole affair is so novel, and startling, and bewildering, that we cannot blame the oilseekers much for their hurry to get into the ground, or to strike a vein. We should think, if the supply of oil continues, there must be a flourishing town here, and that mechanics of all kind would find employment. What is said to be wanted is a good store—there is none now there. Laborers receive from twenty-six to thirty dollars per month. From McClintock’s to Franklin one is never out of sight of the peculiar institution of this region, viz., the derrick, and in the latter town every man almost has one in his garden. (Giddens, 1947)
The McClintock #1 well, originally known as the Colby well, was “kicked down” using a spring-pole drill by J.D. Angier to a depth of 189 m (620 ft) for Brewer, Watson and Company. The proceeds from this and other wells on the McClintock farm were well enjoyed by John Washington Steele. Better known as “Coal Oil Johnny,” John had been unofficially “adopted” by Culbertson McClintock, one of Hamilton’s three sons, at the age of 18 months. Coal Oil Johnny became sole heir to the McClintock farm in 1864 when Culbertson’s widow perished from an oil burn in the kitchen (Flaherty, 2003; Steele, 1902). Figure 3 is a view of the McClintock/Coal Oil Johnny Farm from across Oil Creek. Figure 4 illustrates the location of the McClintock #1 on a topographic base, along with more recently obtained light detection and ranging (lidar) imagery. The McClintock well site (Fig. 5) is now managed and nurtured by Pennsylvania Historical and Museum Commission and Friends of Drake Well, Inc. Lithology and Reservoir Characteristics The McClintock #1 produced from the Second and Third sands of the Upper Devonian Venango Group (Fig. 6), which are the primary petroleum-producing reservoirs throughout the
Figure 4. Location of the McClintock #1 Well and immediate vicinity, using (A) modern (1972) topographic and (B) recent (2003) digital lidar maps (Oil City 7.5 min quadrangle map, prepared by the U.S. Geological Survey, photorevised 1972; Pennsylvania Geological Survey, PAMAP lidar data, 2003).
174
Carter and Flaherty
Figure 5. McClintock #1, the oldest continuously producing well in the world, drilled in 1861. Photo: K.J. Flaherty, November 2002.
Oil City–Rouseville field (north-central Venango County). Lytle (1950) described the Venango Second sand as light gray, finegrained, and hard, with occasional pebbly zones 1–1.5 m (3– 5 ft) thick near the top of the sand. The average thickness of the Venango Second sand is 6.1–10.7 m (20–35 ft) (Harper and Holtz, 2009), and occurs at depths of ~180–275 m (600–900 ft), depending on the elevation of the wells relative to the Oil Creek floodplain (Fig. 7). Porosities in the Venango Second sand average 17 percent, and permeabilities range from 8 to 200 millidarcies (Lytle, 1950). The upper portion of the Venango Third sand is coarse, loose, and pebbly. Porosity ranges between 10 and 16 percent, and permeability varies from several darcies in the pebbly zones to less than a millidarcy in the finer-grained portions of this sand unit (Harper and Holtz, 2009; Lytle, 1950). Averaging 244 m (800 ft) deep, the Venango Third is ~4.6 m (15 ft) thick (Lytle, 1950). The water-to-oil ratio for wells in this part of the Oil City–Rouseville field has always been high (1:1), according to Lytle (1950), and many were subjected to secondary recovery methods. Air drive was used to increase oil flow from the Venango Second sand, and suction was applied to the Venango Third in the 1920s (Lytle, 1950). Production In August 1861, the McClintock #1 was completed in the Venango Third sand and initially produced 50 Bbl of oil per day, together with substantial amounts of brine. Subsequent operators of the well included John and Joe Bowers, Robinson Oil Company, Brundred Oil Corporation, and Quaker State (Flaherty, 2007). The McClintock #1 now produces ~30 Bbl of oil per year. Although the total amount of oil pumped is unknown, this historic well holds the distinction of being the oldest continuously producing oil well in the world. STOP 2: THE DRAKE WELL
Figure 6. Subsurface rock correlation diagram for the field trip area. Oil-bearing sandstones associated with each of the stops in this field guide are highlighted with a black stippled pattern (modified from Carter, 2007).
The story of the Drake Well is steeped in serendipity and influenced by many, from Samuel M. Kier to Dr. Francis Brewer, Mr. George H. Bissell, Professor Benjamin Silliman Jr., and Mr. James M. Townsend (the “Fantastic Four”), to Edwin L. Drake himself. Several works have addressed this historically significant story over the years, but perhaps the most comprehensive and well-researched history of Drake was published by Dr. William Brice (2009). Other recent renditions of Drake’s story have been published by Carter (2009), Flaherty (2003), Harper (1998), and Harper et al. (2009). Additionally, oil history texts such as Giddens (1947) and Henry (1873) provide excellent discussions of the time, the technology, and the series of events that gave rise to Drake’s success in Titusville and the advent of the modern petroleum industry. Figure 8 illustrates the location of the Drake Well Museum grounds and surrounding area over time using historical farmline and topographic maps, as well as recently obtained lidar imagery.
Oil history in western Pennsylvania
175
Situated on the floodplain of Oil Creek, the site has relatively little topographic relief. These lands were originally donated for preservation as an historic site in 1908; cleared, excavated, and drained to prevent flooding in the early 1930s; and eventually taken over by the Commonwealth of Pennsylvania in 1934, which has since cared for the site (Harper et al., 2009). The small feature circled in Figure 8C is an area where pits were dug in the earth to gather and extract oil from the shallow subsurface (Fig. 9 is a photograph taken at this particular site). Such pits have been noted in journals and on maps since the earliest explorers observed them in western Pennsylvania and southwestern New York, and several of these occur adjacent to Oil Creek, along the southern end of the Drake Well Museum grounds. In most cases, the pits are ~2–2.5 m (7–8 ft) long, 1.2–2.5 m (4–8 ft) wide, and variable in depth. Measured depths range from ~1.8 to 3.6 m (6 to 12 ft), with some as deep as 6 m (20 ft) in particular localities (Pees, 2004). Believed to have been dug by early Native Americans, the wood cribbing and ladders found in some of the pits bear evidence that the wood was prepared by use of stone axes (Pees, 2002). Artifacts were dated to the 1400 A.D. time period, prior to Christopher Columbus’ expeditions (Pees, 2004). At the Drake Well Museum site, the pits were known to be among the natural oil seeps that attracted the attention of Brewer, Watson, and eventually Drake. It is known from the diaries of early Europeans exploring the region that the Native Americans skimmed oil from the surface of the water that collected in the pits and used it for medication. This “Seneca Oil” was believed to be a curative for assorted ailments afflicting humans as well as horses and cattle. Initial attempts to collect commercial quantities of oil by Brewer, Watson, J.D. Angier, and others mimicked the Native American pits until Drake’s success proved that larger quantities of oil could be efficiently produced by drilling (Pees, 1993). At present, the Drake Well Museum is maintained by the Pennsylvania Historical and Museum Commission and includes 219 acres (Harper et al., 2009). The working board-for-board replica of the Drake Well (Fig. 10) pumps oil actually produced from the McClintock # 1 (Stop 1 of this field trip). The highlight of our visit to the museum grounds will be our participation in a formal dedication ceremony with Pennsylvania Historical and Museum Commission personnel, at which time we will dedicate Second Pennsylvania Geological Survey rock cutting samples originally gathered from wells in Clarion and Warren Counties to the museum for archival and display.
Figure 7. “Type” geophysical log for the Rouseville area. Obtained from a deeper, Medina Group–producing gas well (Permit #12143911) located on a bluff to the northwest of the McClintock #1, this log illustrates the gamma-ray signature of the productive Venango Second and Third sands (shaded gray using a 50% sand cutoff) in this area of Venango County.
176
Carter and Flaherty
Figure 9. Abandoned oil pits near the Drake Well Museum grounds, Titusville, Pennsylvania (the location of these oil pits is circled in Fig. 8C). Photo: K.J. Flaherty, November 2010.
Lithology and Reservoir Characteristics The Drake Well produced oil from the “Drake sand,” an unnamed sandy lens in the Upper Devonian Riceville Formation (Fig. 6). This was truly a long shot, considering the fact that the Riceville is composed mostly of shale, not the most promising of reservoir rocks. As described by Dickey et al. (1943), the Riceville Formation consists of 26–29 m (85–95 ft) of purplish-red, dark brown, gray, and greenish-gray shales and sandy shales, with occasional thin beds of fine-grained white sandstone. Although the Drake Well struck oil in a sandy lens of the Riceville, subsequent drilling and exploration in northwestern Pennsylvania typically produced oil from one of the Devonian Venango Group sands underlying this formation (Fig. 11). Even the historical farmline map shown in Figure 8A indicates an oil pool producing from the Venango Third (V3) in the immediate vicinity of Drake Well.
Figure 8. Location of Drake Well Museum and immediate vicinity, using (A) historical (1946) farmline, (B) modern (1973) topographic, and (C) recent (2003) digital lidar maps (after Dickey and Heeren, 1946; Titusville South 7.5 min quadrangle map, prepared by the U.S. Geological Survey, photorevised 1973; Pennsylvania Geological Survey, PAMAP lidar data, 2003).
Figure 10. Replica of the Drake Well on the grounds at the Drake Well Museum, Titusville, Pennsylvania. Photo: K.J. Flaherty, May 2003.
Oil history in western Pennsylvania
177
Production
The “Drake Well Formation”
“Colonel” Drake retained the services of William A. Smith, a blacksmith familiar with salt well drilling, to drill his well (Giddens, 1947). Known as “Uncle Billy” to many, Smith completed Drake’s well to a total depth of 21 m (69.5 ft) by the evening of Saturday, 27 August 1859. When he returned the next morning to check the well, he found it to be full of oil. Drake’s well required pumping and had an initial production of 40 barrels per day (Bpd), but only for a short while. By the end of December 1859, well production decreased to 16 Bpd, and in August 1860, it produced 18 Bpd. In October 1865, the well was still reported to produce, although at a much lower rate—5 Bpd (Boyle, 1898).
Our visit to the Drake Well Museum will also give us the opportunity to see the informally named “Drake Well formation” in outcrop. This unit, situated between the Tidioute Shale and Riceville Formation (Fig. 6), has been studied by many, beginning with Caster in 1934 and followed by some confusion in correlations since then. Such problems were identified by Dodge (1992) in reviewing local subsurface log data, and as part of the 63rd Annual Field Conference of Pennsylvania Geologists, Harper (1998) revisited this stratigraphy, basing his correlations on subsurface geophysical log information. Harper (1998) rediscovered a marine interval equivalent to the lower Knapp- and Bedford-equivalent section—so-called the Kushequa Member by Caster (1934)—in rocks exposed at the railroad cut adjacent to the Drake Well Museum, which contains fossils from several phyla, including Porifera, Bryozoa, Brachiopoda, Cephalopoda, and Chordata, among others (Harper et al., 2009). Harper further proposed that this outcrop be identified as the temporary “type section” for this new, informally named unit. The Drake Well formation type section (Fig. 12) includes a maximum rock exposure of 7.3 m (24 ft). Based on the work of Harper (1998), this outcrop represents only about the lower third of the entire 18-m (60-ft) Drake Well formation thickness. The lowermost 3.5 m (10 ft) of the exposure is partially covered today and includes shales and thin, tabular siltstone beds. The upper part of the outcrop is dominated by weather-resistant sandstones and siltstones teeming with fossils (Harper et al., 2009). Recent work in the greater Titusville area by Baird corroborates the work of Harper, and suggests that the base of the Drake Well formation at this type section occurs at or slightly below the elevation of Oil Creek at the latitude of the railroad cut (Harper et al., 2009). On a larger scale, the Drake Well formation has several stratigraphic equivalents—the Berea, Corry, Cussewago, Knapp, and Murrysville formations—all of which have been found to produce oil, gas, and/or brine throughout western Pennsylvania (Carter, 2007). The approximate extent of these correlative units is illustrated in Figure 13. In fact, the Berea Sandstone is the producing reservoir associated with Muddy Creek oil field, Stop 3 of this field trip. STOP 3: MUDDY CREEK OIL FIELD At the turn of the twentieth century, Butler County was one of the hubs of the oil business in southwestern Pennsylvania.
Figure 11. “Type” geophysical log for the Titusville area. Obtained from a deeper, Medina Group–producing gas well (Permit #12143954) on a bluff situated to the northeast of the Drake Well, this log illustrates the gamma-ray signature of various oil-producing sands of the Venango Group (shaded gray using a 50% sand cutoff) in this portion of northwestern Pennsylvania. Note that the “Drake sand” is not clearly visible in the Riceville Formation portion of this gamma ray log; this is attributed to the limited extent of the “Drake sand” in the immediate vicinity of the Drake Well Museum.
178
Carter and Flaherty
Muddy Creek oil field played an important part in this regard, providing much of the oil processed by refineries in East Butler (Stokes, 2001; Williams, 2008). Oil was first discovered in Muddy Creek oil field two miles north of Prospect in 1891 on the Daniel Shanor farm. Figure 14 illustrates the location of Muddy Creek oil field on an historical farmline base map, as
well as using recently obtained lidar imagery. The most significant development seen in this “time-lapse” presentation is the appearance of Lake Arthur (Fig. 14B). This surface water body is a manmade recreation of glacial Lake Watts (Fleeger et al., 2003) and was constructed in the 1960s during development of Moraine State Park. So-named for local glacial ground and end moraine deposits, the park is situated just to the east of the southeast limit of glaciation 140,000 years ago (Fleeger et al., 2003). The Pennsylvania Department of Forest and Waters (predecessor of today’s Pennsylvania Department of Conservation and Natural Resources, which happens to be the parent agency of the Pennsylvania Geological Survey) purchased the lands that would become Moraine State Park in 1963. In order to create a recreational lake at this location, Muddy Creek and its tributaries would need to be flooded, so most oil wells in this field were plugged, and associated pipelines, tanks, and pumping equipment were also removed. The state maintained operation of 46 Muddy Creek oil wells until Lake Arthur and the park were completed (Stokes, 2001). Plugging certificates for several of these
Figure 12. “Drake Well formation” outcrop exposed at the Drake Well Museum grounds, Titusville, Pennsylvania (after Harper, 1998).
Figure 13. Approximate extent of the “Drake Well formation” in the shallow subsurface of Crawford, Forest, Venango, and Warren Counties (modified from Carter, 2007).
Oil history in western Pennsylvania wells, submitted in the mid to late 1960s, remain on file at the Pennsylvania Geological Survey in Pittsburgh, Pennsylvania. Lithology and Reservoir Characteristics The producing reservoir in Muddy Creek oil field was the Upper Devonian Berea Sandstone, ~360 million years in age (Fig. 6). Newberry (1870) named the Berea Sandstone for its type locality in Berea, Ohio, and called the sandstone the “Berea grit” based on outcrop observations he made at this location (Tomastik, 1996). The Berea Sandstone was described by Lytle (1950) as ranging “from a dark gray, hard, fine grained sandstone, to a coarse-grained sandstone.” Lytle (1950) indicated that the pay zone was found in the upper part of the unit, and that if coarse sand was encountered at the bottom of this formation, fresh water was also usually encountered, too. Stratigraphic trapping is responsible for oil production in Muddy Creek field. The Cuyahoga Group siltstones and shales
179
that overlie the Berea Sandstone are laterally extensive throughout the area and more than 30 m (100 ft) thick (Fig. 15). In addition, facies changes within the Berea Sandstone itself impact the permeability of this reservoir from place to place; such permeability barriers are associated with the fluvial-deltaic trend of the Berea play (Tomastik, 1996), to which Muddy Creek field belongs. There could also be a minor structural component to hydrocarbon trapping in this area, as Muddy Creek field is located along the southeastern flank of the Homewood Anticline. This low-amplitude structure is oriented approximately N40°E, and is illustrated in Figure 16 by structure contours on the top of the Pennsylvanian Vanport Limestone. Natural gas was first discovered in the Berea Sandstone in East Liverpool, Ohio, ca. 1860 (Tomastik, 1996). Like so many other natural gas discoveries in the Appalachian basin, the one in East Liverpool occurred by happenstance because drillers were really looking to strike oil or brine. Natural gas from this and subsequent wells was used to heat and light the town of East
Figure 14. Location of Muddy Creek oil field, using (A) historical (1911) farmline and (B) recent (2003) digital lidar maps (unpublished Zelionople NE 15 min quadrangle farmline map; Pennsylvania Geological Survey, PAMAP lidar data, 2003).
180
Carter and Flaherty
Figure 15. “Type” geophysical log for the Prospect area (Permit# 01920690). This gamma-ray log spans subsurface formations from the Pennsylvanian Conemaugh Group through the Devonian Chadakoin Formation. The Upper Devonian Berea Sandstone and Venango Group sands (see gray shading using a 50% sand cutoff) are the typical oil and gas-producing reservoirs in this area.
Oil history in western Pennsylvania
181
Figure 16. Geologic structure on top of the Vanport Limestone (modified from Richardson, 1936).
Liverpool. Spurred by this discovery, shallow wildcat wells were drilled throughout Ohio and Pennsylvania in the early 1860s (Tomastik, 1996). The earliest production from the Berea play in Pennsylvania is associated with the discovery of oil in Slippery Rock field, Lawrence County, in 1864 (Carter, 2005). In this locality, Slippery Rock Creek was so named because of a rock that was slippery due to the discharge of an oil seep. Curiously enough, the observation of this oil seep and “slippery rock” gave rise to the drilling activity that discovered the oil field of the same name (Lytle and Lytle, 1974). Production The Muddy Creek discovery well initially produced 4 Bpd (Lytle, 1950; Flaherty, 2003), and by 17 July 1891, Muddy Creek oil field had 23 wells producing around 250 Bpd (Boyle, 1898). Bessemer gas engines, fueled by natural gas, operated the wells here. One gas engine could operate as many as 17 wells at a time (Stokes, 2001). Approximately 360 oil wells were completed in Muddy Creek oil field, based on the assessment prepared by Lytle (1950); the authors of this field guide have been able to document locations for 250 of these (Fig. 14B). These wells did not produce large amounts, but gave a steady supply of two or three Bpd over a period of several years (Richardson, 1936; Lytle, 1950; Flaherty, 2003). The oil was piped to a central collection
tank, located on the old Whippoorwill Hill Road (now Christley Road) by Big Run on the Leisey Farm (Stokes, 2001), and then sent to refineries. The Marshall-Barr Site Well No. 19 on the old Marshall-Barr tract of land was drilled in 1932 to a depth of 297 m (974 ft) and produced oil for more than 30 years. When the Commonwealth of Pennsylvania purchased the lands now known as Moraine State Park, the Marshall-Barr tract was selected as the site for a pumping oil well exhibit. Lack of funding prevented its refurbishment in the 1960s, but with assistance from volunteers at the park as well as from the Steam Engine and Old Equipment Association (Portersville, Pennsylvania) in the 1990s, the Marshall-Barr site is now a fitting homage to the once-thriving oil industry in Muddy Creek valley. The site contains an old Bessemer gas engine, an engine house, and the necessary equipment, such as hangers and pumping jacks (Stokes, 2001) (Figs. 17 and 18). ROAD LOG Our day-long field trip will involve a drive of almost 340 km (~211 mi) from the junction of Interstates I-79 and I-279 in Allegheny County to stops in Butler and Venango Counties, and back again (Fig. 1). Our first stop will be the McClintock Well in Rouseville, where we’ll spend about an hour touring and
182
Carter and Flaherty
discussing the geology and history associated with the oldest, continuously producing oil well in the world. From there, we will travel north to Titusville to tour the Drake Well Museum and grounds. Here, we will have lunch, take a tour of the grounds and antique equipment, and participate in the dedication of original rock cutting samples collected by geologists of the Second Geologic Survey of Pennsylvania; our visit will be approximately two and a half hours long. The final stop is Muddy Creek oil field at Moraine State Park in Butler County, located roughly 130 km (80 mi) south of Titusville. We will spend an hour at the Muddy Creek site, where volunteers will lead an historical tour of the exhibit and grounds. We plan to arrive back at the Omni William Penn Hotel in Pittsburgh by 5:30 p.m.
55.2 (88.8) 82.5 (132.7) 82.6 (132.8) 82.6 (133.0) 95.0 (152.9) 96.5 (155.3) 98.0 (157.7)
Cumulative mi (km)
directions
133.9 (215.4)
0.0 (0.0) 45.4 (73.0)
Junction of I-79 and I-279 northbound. Follow I-79 north; exit onto I-80 eastbound.
144.1 (231.8) 161.8 (260.4) 168.2 (270.6) 168.3 (270.9) 172.2 (277.1) 176.1 (283.4) 176.2 (283.6) 182.6 (293.7) 210.6 (338.8)
Figure 17. The Bessemer gas engine used to operate the oil pump at the Marshall-Barr site was manufactured in Grove City, Pennsylvania. Photo: K. Carter, October 2007.
Follow I-80 eastbound; exit at PA Route 8 northbound (Exit 29). Follow Route 8 north to Rouseville; turn left onto Waitz Road. Follow Waitz Road to parking area for McClintock Well. Follow Waitz Road east; left turn onto Route 8 north. Follow Route 8 north to Titusville; turn right onto East Bloss Street. Follow East Bloss Street to Drake Well Museum parking area. Depart Drake Well Museum parking area. Follow East Bloss Street; turn left onto Route 8 southbound. Follow Route 8 south; merge onto I-80 westbound. Exit I-80 to I-79 southbound (Exit 19). Follow I-79 south; exit at PA Route 422 east (Exit 99). Follow Route 422 east to Route 528 north. Turn left onto Park Road. Follow Park Road to parking area for Moraine State Park/Muddy Creek oil field. Exit Moraine State Park via Park Road to Route 528 south. Turn right onto ramp for Route 422 westbound. Follow Route 422 west; turn left onto ramp for I-79 south. Follow I-79 south to junction with I-279.
Figure 18. Restored Marshall-Barr Site near the shores of Lake Arthur. Photo: K. Carter, October 2007.
Oil history in western Pennsylvania APPENDIX A. SECOND GEOLOGICAL SURVEY OF PENNSYLVANIA ROCK CUTTINGS FOR DEDICATION TO THE DRAKE WELL MUSEUM John Franklin Carll, geologist with the Second Geological Survey of Pennsylvania, was truly the world’s first petroleum geologist. He was a stickler for detail and very carefully sought to obtain the most complete and accurate data regarding oil drilling in western Pennsylvania. To Carll’s credit are the first published geologic structure maps in North America (1875), the earliest known geologic strip log (ca. 1877), and invention of a wooden rack for correlating subsurface formation data using rock cuttings bottles (1880). The reader is referred to John Harper’s historical piece on this amazing man for more details (Harper, 2002). Rock cuttings described in Tables A1 and A2 are associated with two of Carll’s well-respected reports: his 1880 report on the geology of the oil regions of Warren, Venango, Clarion, and Butler Counties (Report III); and his 1883 report on the geology of Warren County and neighboring oil regions (Report I4). During our stop at the Drake Well Museum in Titusville, we will be dedicating rock cuttings from the Beatty #1 Well, completed in March 1875 in Glade Township, Warren County; and the Columbia Oil Company #19 Well, completed in July 1877 near Edenburg, Clarion County. Many thanks are given to Ms. Megan Achille, summer extern at the Pennsylvania Geological Survey, who carefully gathered, cleaned, and organized these rock cuttings.
183
and Gas Company, Inc., who continually support our respective endeavors to research oil history. In addition, many thanks go to Gary Fleeger and John Harper of the Pennsylvania Geological Survey and Samuel F. Pratt Jr. of Waverly Production, L.P., who provided thorough and thoughtful technical reviews of this field guide. Last but not least, the authors appreciate the work of Megan Achille, undergraduate student at the University of Pittsburgh, who organized and catalogued the Second Survey sample cuttings dedicated to the Drake Well Museum as part of this field trip.
ACKNOWLEDGMENTS The authors wish to acknowledge several individuals in association with this work. First and foremost, we wish to recognize the Pennsylvania Geological Survey and ABARTA Oil
Megan Achille at the Pennsylvania Geological Survey, summer 2010.
TABLE A1. BEATTY WELL NO. 1, MARCH 1875* Thickness Depth Elevation (ft) (ft below ground surface) (ft mean sea level) Well mouth above ocean, 1217 N.A. 0 1217 Drive pipe 90 90 1 127 Shales 170 260 957 Chocolate shale, slaty (specimen 1) 8 268 949 Dark slaty shale, micaceous, hard (2) 52 320 897 Chocolate shale-mud (3) 8 328 889 Shales, blue and brown, some red layers 57 385 832 SS and shale, gray, fossiliferous, hard (4) 10 395 822 SS and shale, gray, softer, less fossiliferous (5) 45 440 777 35 475 742 Mud rock with sand shells at 465′ and 475′ (6a, 6b, 7) ? (Shales?) 34 509 708 SS, with shale, gray (8) 18 527 690 Shale and slate (spe cimens lo st) (9) 33 560 657 Muddy shale, with hard streaks (10) 32 592 625 Sandy slate, micaceous, bluish (11) 12 604 613 Very fine, flaky SS, some fossils and slate (12) say 2 606 611 ? Probably same as above 4 610 607 Same as No. 12, less slate and grayer (13) 5 615 602 † 597 5 620 Very fine-grained and very fossiliferous SS, hard (14) Gray SS, soft and friable (15) 5 625 5 92 Gray SS, soft and friable, with some pebbles and slate (oil here) (16) 4 629 588 Same as No. 16 (17) 3 632 5 85 Note: Near Mr. Beatty’s residence at East Warren, Glade township. Record compiled from specimens of sand-pumpings preserved by Mr. F.A. Randall. *Carll, J.F., 1883, Geological report of Warren County and the neighboring oil regions, with additional oil well records: Second Geological Survey of Pennsylvania: 1880 to 1883, Report I4, p. 1–2. † Oil and gas from 620′, increasing to 629′. Natural production probably about 5 barrels. Description
184
Carter and Flaherty TABLE A2. COLUMBIA OIL COMPANY WELL NO. 19, 7 JULY 1877* Driller’s Thickness Depth Elevation Sand Name (ft) (ft below ground (ft above mean surface) sea level) N.A. Well mouth above ocean in feet N.A. 0 1443 N.A. Conductor 12 12 1431 1 Slate, dove colored and very muddy 18 30 1413 Slate, dark and dove colored, very muddy 47 77 1366 2 3 Slate, black with iron pyrites 45 122 1321 N.A. Slate, with gray sand shells 16 138 1305 Coal 1 139 1304 N.A. 4 Slate, shelly 7 146 1297 5, 6 SS, gray, fine, gritty 44 190 1253 Slate, dark with an occasional thin shell 33 223 1220 7, 8 9 to 12 SS, very dark, fine, micaceous 54 277 1166 13, 14 SS, white, medium fine, fossils? 33 310 1133 SS, white, fine 5 315 1128 15 N.A. SS, gray, fine Mountain Sand 20 335 1108 16 SS, dark-gray, fine, micaceous, soft 19 354 1089 SS, dark-gray, very fine, tough, “rubber rock” 15 369 1074 17 18 Slate, with dark shells, hard drilling 11 380 1063 19 SS, dark gray, fine, hard, flaky, muddy 15 395 1048 Slate, dark blue 31 426 1017 20, 21 22, 23 SS, dark gray, very fine, tough 30 456 987 24, 25, 26 SS, dark gray, coarser, gritty 29 485 958 27 SS, or sandy micaceous slate 12 497 946 28, 29 SS, like No. 1 30 527 916 40′ Rock 30, 31, 32 SS, gray, very fine, fossils? 21 548 895 33 to 37 SS, gray and white, medium, slaty, micaceous, fossils? 22 570 873 38 to 42 Slate, dark, tough, granulating like sand 48 618 825 43 to 46 Slate, with an occasional very fine sand shell 37 655 788 47 Slate, with frequent shells, hard, blue, micaceous 40 695 748 48, 49 Slate, purplish 7 702 741 50 Slate, dark and fawn color with gray shells 12 714 729 51, 52 SS, dark gray, fine, micaceous, soft 7 721 722 53 to 61 Slate, with very fine shells, green and gray 118 839 604 62 SS, white, with white and yellow pebbles 2 841 602 63 SS, pure white, medium, fossils? 4 845 598 64, 65 SS, pure white, fine 10 855 588 Slate, soft, “soapstone” 24 879 564 66 N.A. Sand shells, gray 2 881 562 1 882 561 N.A. Red rock 67 Slate, with gray sand shells 5 887 556 68 SS, white, very fine, hard 1 888 555 69 SS, white, medium, fossils? 5 893 550 70 SS, gray, very fine, micaceous 15 908 535 Slate, with gray shells 4 912 531 71 72 Slate, with green shells 8 920 523 Slate, with gray shells frequent 18 938 505 73 74 Red, fine gray sand and red clay 3 941 502 75, 76 Slate, dark, with fine hard gray shells 40 981 462 77, 78 Sand shells, gray and olive, fossils? 4 985 458 32 1017 426 79, 80, 81 Slate, dark, with gray shells 82, 83 Red rock, red and green sandy shale 9 1026 417 84 Red rock, brownish red, sandy, micaceous slate Big Red 30 1056 387 Slate 3 1059 384 N.A. 85 SS, gray, gritty, micaceous, flaky 9 1068 375 86 Slate, sandy 13 1081 362 N.A. Red rock 2 1083 360 87 Slate, purplish and green, sandy 21 1104 339 88 Shells, gray, hard, with yellow pebbles 2 1106 337 Slate, with gray, fine, hard shells 11 1117 326 89, 90 91 SS, with white and yellow pebbles 5 1122 321 92 SS, gray, very fine 1 1123 320 † 93, 94,95 SS, cream color, fine Third Sand 4 1127 316 96, 97 SS, fine-pebble sand, fossils? 4 1131 312 98 to 101 SS, cream color, fine, fossils? 12 1143 300 Note: Owned by Columbia Oil Company and situated on their J.H. Kiser farm, ¾ of a mile S20°E from Edenburg, 1 mile N85°E from Brundred Well No. 4, and half a mile west from McGrew Bros. Well No. 4. *Carll, J.F., 1880, The geology of the oil regions of Warren, Venango, Clarion, and Butler Counties: Second Geological Survey of Pennsylvania: 1875 to 1879, Report III, p. 215–217. † Not fully through the 3d sand. Second sand only represented by thin shells. Drilled dry. Cased at 567′. No water found below casing. No gas above 3d SS, and a little in that. Production from 5 to 6 barrels per day. This well was torpedoed before the tubing was inserted and made one flow over the top of the derrick. Specimen numbers
Description
Oil history in western Pennsylvania REFERENCES CITED Boyle, P.C., 1898, The Derrick’s Handbook of Petroleum: A Complete Chronological and Statistical Review of Petroleum Developments from 1859 to 1898: Oil City, Pennsylvania, Derrick Publishing Company, 1062 p. Brice, W.R., 2009, Myth, Legend, Reality: Edwin Laurentine Drake and the Early Oil Industry: Oil Region Alliance, Oil City, Pennsylvania, 661 p. Carll, J.F., 1886, Annual report, Geological Survey of Pennsylvania, Part II, Report on the oil and gas regions: Second Geological Survey of Pennsylvania, Harrisburg, Pennsylvania, 918 p. Carter, K.M., 2005, Oil and gas geology of Lawrence and Mercer Counties, Pennsylvania, in Fleeger, G.M., and Harper, J.A., eds., Type sections and stereotype sections, glacial and bedrock geology in Beaver, Lawrence, Mercer, and Crawford Counties: Guidebook, 70th Annual Field Conference of Pennsylvania Geologists, Sharon, Pennsylvania, p. 44–58. Carter, K.M., 2007, Subsurface rock correlation diagram, oil and gas producing regions of Pennsylvania: Pennsylvania Geological Survey, 4th ser., OpenFile Report OFOG 07-01.1, web version (24 September 2010). Carter, K.M., 2009, 1859: A year that changed the world: Oilfield Journal, v. 8, 2008–2009, p. 5–16. Carter, K.M., and Sager, K., 2010, Addendum to Park Guide 4—Muddy Creek oil field: Pennsylvania Geological Survey, 4th ser., Park Guide 4A, 7 p., Portable Document Format (PDF). Caster, K.E., 1934, The stratigraphy and paleontology of northwestern Pennsylvania, part 1: stratigraphy: Bulletins of American Paleontology, v. 21, no. 71, 185 p. Dickey, P.A., and Heeren, L., 1946, Oil and gas field property line maps of the Titusville quadrangle, Pennsylvania: Pennsylvania Geological Survey, 4th ser., Special Bulletin No. 4, 9 p. Dickey, P.A., Sherrill, R.E., and Matteson, L.S., 1943, Oil and gas geology of the Oil City quadrangle, Pennsylvania: Pennsylvania Geological Survey, 4th ser., Mineral Resources Report M25, 201 p. Dodge, C.H., 1992, Bedrock lithostratigraphy of Warren County, Pennsylvania, in Sevon, W.D., ed., Geology of the Upper Allegheny River region in Warren County, northwestern Pennsylvania: Guidebook, 57th Field Conference of Pennsylvania Geologists, Warren, Pennsylvania, p. 1–20. Flaherty, K.J., 2003, Hills, dales & oil trails, a guide to some historic oil fields between Pittsburgh and Titusville, Pennsylvania: Field trip guidebook, American Association of Petroleum Geologists–Society of Petroleum Engineers 2003 Joint Eastern Meeting, Pittsburgh, Pennsylvania, 7 September 2003, 91 p. Flaherty, K.J., 2007, Drill bits—A town grows in oildom: Oilfield Journal, v. 6, 2006–2007, p. 82. Fleeger, G.M., Bushnell, K.O., and Watson, D.W., 2003, Moraine and McConnells Mill State Parks, Butler and Lawrence Counties—Glacial lakes and drainage changes: Pennsylvania Geological Survey, 4th ser., Park Guide 4, 12 p. Frasure, W.W., 1952, Longevity of manufacturing concerns in Allegheny County: Pennsylvania, University of Pittsburgh, 226 p. Giddens, P.H., 1938, The Birth of the Oil Industry: New York, New York, MacMillan & Company, 216 p. Giddens, P.H., compiler and ed., 1947, Pennsylvania Petroleum: 1750–1872, a Documentary History: Pennsylvania Historical and Museum Commission, Titusville, Pennsylvania, 420 p. Harper, J.A., 1998, Stop 6 and Lunch: Drake Well Memorial Park, in Harper, J.A., ed., Geotectonic Environment of the Lake Erie crustal block: Guide-
185
book, 63rd Annual Field Conference of Pennsylvania Geologists, Erie, Pennsylvania, p. 61–74. Harper, J.A., 2002, The incredible John F. Carll: the world’s first petroleum geologist and engineer: Oilfield Journal, v. 2, 2001–2002, p. 2–14. Harper, J.A., and Holtz, A., 2009, Stop 2: “McClintockville and the McClintock #1 Well,” in Harper, J.A., ed., History and geology of the oil regions of northwestern Pennsylvania: Guidebook, 74th Annual Field Conference of Pennsylvania Geologists, Titusville, Pennsylvania, p. 89–93. Harper, J.A., McKenzie, S.C., Baird, G.C., and Sullivan, J.S., 2009, Stop 6: Drake Well Museum and “type” “Drake Well Formation,” in Harper, J.A., ed., History and geology of the oil regions of northwestern Pennsylvania: Guidebook, 74th Annual Field Conference of Pennsylvania Geologists, Titusville, Pennsylvania, p. 120–131. Henry, J.T., 1873, The Early and Later History of Petroleum, Vol. I and II: Oil City, Pennsylvania, Oil Region Alliance, second reprint (2008), 607 p. Lytle, W.S., 1950, Crude oil reserves of Pennsylvania: Pennsylvania Geological Survey, 4th ser., Mineral Resources Report M32, p. 132. Lytle, V., and Lytle, W., 1974, But is there really a Slippery Rock?: Pacific Geology, v. 5, no. 1, p. 4–8. Newberry, J.S., 1870, Report on the progress of the Geological Survey of Ohio in 1869: Ohio Division of Geological Survey, 176 p. Pearse, J.B., 1876, Natural gas in iron working, in Special report on the coke manufacture of the Youghiogheny River Valley in Fayette and Westmoreland Counties, v. L: Second Geological Survey of Pennsylvania, Harrisburg, Pennsylvania, 252 p. Pees, S.T., 1993, The Search for North America’s First Oilmen: Oilfield Barker, v. 1, no. 1, p. 1, 3. Pees, S.T., 2002, Petroleum Mining in NW Pennsylvania: Oil-Industry History, v. 3, no. 1, p. 34–55. Pees, S.T., 2004, Pre-Columbian Mining, http://www.petroleumhistory.org/ OilHistory/pages/Gathering/precolumbian.html, (18 November 2010). Prout, H.J., 1922, A Life of George Westinghouse: New York, New York, Charles Scribner’s Sons, 375 p. Richardson, G.B., 1936, Geology and mineral resources of the Butler and Zelienople quadrangles, Pennsylvania: Washington, D.C., U.S. Geological Survey Bulletin 873, p. 61–65 and plate 4. Shank, W.H., 1981, The Amazing Pennsylvania Canals: Historical Society of York County, York, Pennsylvania., 128 p. Steele, J.W., 1902, Coal Oil Johnny—His Book: New York, New York, Press of Hill Publishing Company, 211 p. Stokes, N.H., 2001, History of the Muddy Creek oil field and the MarshallBarr-site: informational brochure prepared for the Pennsylvania Department of Conservation and Natural Resources, 33 p. Tomastik, T.E., 1996, Play MDe: Lower Mississippian–Upper Devonian Berea and equivalent sandstones, in Roen, J.B., and Walker, B.J., eds., The atlas of major Appalachian gas plays: West Virginia Geological and Economic Survey, Publication V-25, p. 56–62. Williams, C.E., 2008, Western Pennsylvania’s oil heritage: Arcadia Publishing, Charleston, South Carolina, p. 60.
MANUSCRIPT ACCEPTED BY THE SOCIETY 6 DECEMBER 2010
Printed in the USA
Contents Preface 1. Late Devonian paleontology and paleoenvironments
at Red Hill and other fossil sites in the Catskill Formation of north-central Pennsylvania Edward B Daeschler and Walter L Cressler III 2. An introduction to structures and stratigraphy in the
proximal portion of the Middle Devonian Marcellus and Burket/Geneseo black shales in the Central Appalachian Valley and Ridge Terry Engelder, Rudy Slingerland, Michael Arthur, Gary Lash, Daniel Kohl, and D .P. Gold
3. Pennsylvanian climatic events and their congruent biotic responses in the central Appalachian Basin David K Brezinski and Albert D Kollar
4. Landslides in the vicinity of Pittsburgh, Pennsylvania Richard E. Gray, James V. Hamel, and William R Adams Ir
5. Quaternary geology of northwestern Pennsylvania Gary M . Fleeger, Todd Grote, Eric Straffin, and John P Szabo
6. The history and geology of the Allegheny Portage Railroad, Blair and Cambria Counties, Pennsylvania John A. Harper 7. Early industrial geology of western Pennsylvania and
eastern Ohio: Early gristmills and iron furnaces west of the Alleghenies and their geologic contexts Joseph T. Hannibal, Tammie L. Gerke, Mary K. McGuire, Harry M . Edenborn, Ann L. Holstein. and David Parker
8. The old, the crude, and the muddy: Oil history in western Pennsylvania Kristin M . Carter and Kathy J. Flaherty
. . THE GEOLOGICAL SOCIETY ,....OF AMERICA®
ISBN 978-0 -8137-0020-5
3300 Penrose Place • P.O. Box 9140 Boulder, CO 80301-9140, USA
9 7808 13 7002 0 5