The geochemistry of oils and gases from the Cumberland overthrust sheet in Virginia and Tennessee
<p>This study presents high-resolution gas chromatograms of oils and molecular and isotopic analyses of oil-associated gases from 17 wells producing in the…
Public-domain full text preserved in the Mountain Man Mining Library. Original source: pubs.usgs.gov.
The Geochemistry of Oils and Gases From the Cumberland Overthrust Sheet in Virginia and Tennessee By Kristin O. Dennen, Mark Deering, and Robert C. Burruss Chapter G.12 of Coal and Petroleum Resources in the Appalachian Basin: Distribution, Geologic Framework, and Geochemical Character Edited by Leslie F. Ruppert and Robert T. Ryder Professional Paper 1708 U.S. Department of the Interior U.S. Geological Survey
Suggested citation: Dennen, K.O., Deering, Mark, and Burruss, R.C., 2014, The geochemistry of oils and gases from the Cumberland overthrust sheet in Virginia and Tennessee, chap. G.12 of Ruppert, L.F., and Ryder, R.T., eds., Coal and petroleum resources in the Appalachian basin; Distribution, geologic framework, and geochemical character: U.S. Geological Survey Professional Paper 1708, 38 p., http://dx.doi.org/10.3133/pp1708G.12.
Contents Abstract 1 Introduction 1 Petroleum Geology Overview 1 Regional Geology 1 Structure of the Cumberland Overthrust Sheet 4 Reservoir Characteristics 4 Possible Source Rocks 4 The Gloeocapsomorpha Prisca Enigma 7 Methods 8 Sample Collection 8 Instrumental Analysis 8 Whole Crude Oil 8 Elemental, Isotopic, and API Gravity Analysis 8 Standard Biomarker Analysis 9 Natural Gas Analysis 9 Results 9 Oil Geochemistry 9 Elemental, Isotopic, and API Gravity Analysis 9 Biomarker Profiles 9 Terpanes 9 Steranes 11 Aromatic Fraction Biomarkers 11 Natural Gas Geochemistry 11 Molecular Composition 11 Isotopic Composition of Methane, Ethane, Deuterium, and Carbon Dioxide 11 Isotopic Composition of Gases With Higher Carbon Numbers 11 Discussion 16 Data Quality 16 Compositional Trends in Oil From the Cumberland Overthrust Sheet 16 Gas Chromatographic and Biomarker Profiles 16 Light Hydrocarbon Parameters 16 Isotopic Composition of Oil 18 Evidence for Variations in the Depositional Environment of the Source Rocks 18 Pristane-to-Phytane Ratio and Carbon Preference Index 18 Carbon Preference Index 18 Natural Gas Geochemistry 22 Methane 22 Ethane and Propane 22 Natural Gas Plots 22 Natural Gas From the Ben Hur and Rose Hill Fields 23 Natural Gas From the Swan Creek Field 23 Conclusions 24
Acknowledgments 26 References Cited 26 Appendix A.—Individual Gas Chromatographic Profiles of the Whole Crude Oils 29 Appendix B.—Representative Biomarker Profiles 37 Figures
1. Map showing the location of the study area 2
2. Generalized stratigraphic chart for the Cumberland overthrust sheet 3
3. Generalized cross sections of the study area 5
4. Bar charts showing percentage of each molecular component in the analyzed gas samples from the Cumberland overthrust sheet 15 5.-13. Graphs showing—
5. The n-heptane-to-methylcyclohexane ratio versus the toluene-to-n-heptane ratio of oil samples from the Cumberland overthrust sheet and from oil samples from other Paleozoic reservoirs in the United States and Canada 17
6. The iso-heptane values versus n-heptane values of oil samples from the Cumberland overthrust sheet and of oil samples from Ordovician to Tertiary reservoirs elsewhere in the United States and Canada 19
7. The isotopic compositon of carbon of the saturated fraction of the oil versus the isotopic composition of carbon of the aromatic fraction of the oil from the samples in this study, along with isotopic composition of carbon values from the saturated and aromatic fractions of oil samples from the Trenton Group in the Michigan basin for comparison 20
8. The carbon preference index versus the pristane-to-phytane ratio for the oil samples in this study and for oil samples from other geographic areas for comparison 21
9. The ratio of gas wetness versus the average depth below well head of a gas-producing formation for gas samples from the Cumberland overthrust sheet 22
10. The isotopic composition of carbon in methane relative to the Vienna Peedee belemnite standard versus the isotopic composition of hydrogen (deuterium) in methane relative to the Vienna standard mean ocean water standard for gas samples from the Cumberland overthrust sheet 23
11. The isotopic composition of carbon in propane versus that in ethane for the gas samples from the Cumberland overthrust sheet 23
12. The reciprocal of the component carbon number versus the isotopic composition of carbon in the hydrocarbon gas for four gas samples from Lee County, Va 24
13. The reciprocal of the component carbon number versus the isotopic composition of carbon in the hydrocarbon gas for four gas samples from the Swan Creek field, Hancock and Claiborne Counties, Tenn 25
Conversion Factors Multiply By To obtain Length foot (ft) meter (m) mile (mi) kilometer (km) micrometer (µm) inch (in.) meter (m) foot (ft) Volume barrel (bbl), (petroleum, 1 barrel=42 gal) cubic meter (m3) microliter (µL) cubic inch (in3) milliliter (mL) cubic inch (in3) Mass milligram (mg) 0.00003527 ounce avoirdupois (oz) Flow rate milliliters per minute (mL/min) cubic inches per minute (cm3/min) Velocity centimeters per second (cm/sec) inches per second (in./sec) Temperature in degrees Celsius (°C) may be converted to degrees Fahrenheit (°F) as follows: °F=(1.8×°C)+32 The isotopic composition of carbon (carbon 13, 13C) in methane is reported as the deviation (expressed as δ13C) in units of parts per thousand (per mil) relative to the Vienna Peedee belemnite (VPDB) standard. The isotopic composition of hydrogen (deuterium, 2H) in methane is reported as the deviation (expressed as δ2H) in per mil relative to the Vienna standard mean ocean water (VSMOW). Tables
1. Selected well information 6
2. Selected light hydrocarbon compositional ratios and reproducibility data 10
3. Molecular composition of gases 12
4. Isotopic composition of carbon and hydrogen in gas samples 13
Abstract This study presents high-resolution gas chromatograms of oils and molecular and isotopic analyses of oil-associated gases from 17 wells producing in the Upper Cambrian to Lower Ordovician Knox Group, the Middle and Upper Ordovician Stones River Group, and the Upper Ordovician Trenton Limestone in the Cumberland overthrust sheet. The wells are located in the Ben Hur and Rose Hill fields in Lee County, Va., and in the Swan Creek field in Hancock and Claiborne Counties, Tenn. They produce oils typical of those from source rocks that are rich in Gloeocapsomorpha prisca (G. prisca) (Assemblage A-type kerogen). The Rose Hill oils appear to come from a source that contains a higher proportion of Assemblage A-type kerogen than the Ben Hur and Swan Creek oils. Extrapolation of the δ13C compositions of oil-asso ciated gases to possible kerogen compositions gives estimates of -23 to -24 per mil within the range of isotopic compositions of known G. prisca source material. Gases produced from the Knox Group wells in the Swan Creek field are different from those in the Middle and Upper Ordovician reservoirs and come from a source with a broader range of isotopic values. Trends in isotopic and gasoline-range compositional para‑ meters of the oils and associated gas isotopic and molecular compositions are most likely influenced by changes in local source depositional facies. Introduction The study area is located in the Cumberland overthrust sheet, which is part of the central Appalachian basin. The Cumberland overthrust sheet is approximately 125 miles (mi) long and 25 mi wide and covers parts of southwestern Virginia, southeastern Kentucky, and northeastern Tennessee (fig. 1). It is bounded on the northwest side by the Pine Moun tain thrust fault, on the southeast side by the Clinchport thrust 1U.S. Geological Survey, Reston, Va. 2Virginia Department of Mines, Minerals and Energy, Division of Gas and Oil, Abingdon, Va. The Geochemistry of Oils and Gases From the Cumberland Overthrust Sheet in Virginia and Tennessee By Kristin O. Dennen,1 Mark Deering,2 and Robert C. Burruss1 fault, and on the northeast and southwest sides by the Russell Fork and Jacksboro faults, respectively (Miller and Brosge, 1954; Harris and Milici, 1977; Mitra, 1988) (fig. 1). The wells used in this study are located in the Rose Hill and Ben Hur fields in Lee County, Va., and in the Swan Creek field in Han cock and Claiborne Counties, Tenn. (fig. 1). This area had been explored for oil and gas for almost 100 years, but the first efforts at drilling were unsuccessful because there was little knowledge of the complex regional and local structure (Miller and Brosge, 1954). A produc ing well was completed in the Trenton Limestone in the Rose Hill field and commercial production started in 1946 (Bartlett, 1989). The Ben Hur field was discovered in 1963 and expanded in 1981 with the drilling of the James Dean No. 1 well (Bartlett, 1989). The first well in the Swan Creek field was completed in 1982 and commercial production began in 1988 (Hatcher and others, 2001). The Ben Hur and Rose Hill fields have produced almost 800,000 barrels of oil (Ryder, 1996; Virginia Center for Coal and Energy Research, 2008). Reserves in the Swan Creek field have been estimated to be over 35 billion cubic feet (bcf) of gas and over 300,000 barrels (42-gallon blue barrels, or bbl) of oil (Hatcher and others, 2001). Petroleum Geology Overview Regional Geology Harris and Milici (1977) divided the Paleozoic in the Cumberland overthrust sheet into three major depositional units: Cambrian through Lower Ordovician, Middle Ordovi cian through Lower Silurian, and Upper Devonian through Pennsylvanian. Stratigraphic units present in the Rose Hill, Ben Hur, and Swan Creek fields are shown in figure 2. Each unit is separated by a regional unconformity and each sequence records a marine transgression. The units also grade upward from clastic to dominantly carbonate regimes rep resented by the formation of carbonate banks, lagoons, and shallow-marine-shelf units. Late Proterozoic to early Paleo zoic Iapetan rifting was the first of four major Appalachian tectonic episodes recorded in the stratigraphy and structure of
2 Coal and Petroleum Resources in the Appalachian Basin Figure 1. Map showing the location of the study area. The Cumberland overthrust sheet, bordered on the west by the Pine Mountain thrust fault and on the east by the Clinchport thrust fault, is in the southeastern part of the Appalachian basin on the Tennessee-Virginia-Kentucky border. Colored ovals show the location of oil fields and circles show sample locations. Map is based on Mitra (1988). Figure 1 - Refer to Caption
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 3 Figure 2. Generalized stratigraphic chart for the Cumberland overthrust sheet showing previous stratigraphic nomenclature for the study area and some current terms used by the oil and gas industry, as used in this report. The chart also shows formations that are source rocks in other parts of the Appalachian basin. Figure 2 - Refer to Caption
4 Coal and Petroleum Resources in the Appalachian Basin the area. After this, upward-shallowing sequences of the Upper Cambrian and Lower Ordovician Knox Group were deposited on a widespread, stable passive-margin platform where subsid ence and depositional rates were generally equal (Pfiel and Read, 1980). In the Early to Middle Ordovician, the area was uplifted and eroded during a worldwide sea-level drop and a regional transition to an active foreland basin margin, resulting in the post-Knox unconformity (Hatcher and Repetski, 2007). The next three tectonic episodes (the Ordovician Taconian, the Devonian Acadian, and the Carboniferous Alleghanian orog enies) were compressional events. The hydrocarbon sources formed by each of the major depositional units may have generated hydrocarbons from Silurian to Permian time (Milici and others, 2006). In the study area, Middle and Upper Ordo vician sediments were deposited on a carbonate ramp between two depositional centers (one to the north and the other to the south, with highlands forming to the east) because of the Taconian orogeny (Read, 1980). The Knox Group carbonates are mainly dolomites, as are parts of the Ordovician carbonate sequences. Structure of the Cumberland Overthrust Sheet The complexity of the structure in the Cumberland overthrust sheet is illustrated by the fact that the producing horizon in an oil field can crop out within less than a mile of a well (Bartlett, 1989). The sheet consists of successions of Paleozoic marine strata that have been thrust westward on top of younger units (fig. 3). The Paleozoic sediments reached thermal maturity before folding and faulting; therefore, more thermally mature sequences were thrust on top of less mature strata (Harris and Milici, 1977; Repetski and others, this vol ume, chap. F.1). Westward displacement in the Cumberland overthrust sheet ranges from 2 mi on the northeast to more than 13 mi in the southwest with 10 mi of displacement in the Swan Creek area and about 8 mi of displacement in the vicinity of the Rose Hill and Ben Hur fields (Wentworth, 1921; Mitra, 1988). Sequences of Cambrian to Devonian strata are folded and duplicated by a series of imbricate thrust faults that climb along bedding planes in the Cambrian Conasauga Group and Rome Formation up through the Silurian Clinch Formation and Devonian Chattanooga Shale (Harris and Milici, 1977; Mitra, 1988) (fig. 3). Reservoir Characteristics Hydrocarbons are trapped in the thrust-faulted anticlines and fault slices above the faults (Ryder, 1996; Hatcher and others, 2001). The Lower Ordovician Mascot Dolomite in the upper part of the Knox Group produces most of the gas in the Swan Creek field and the Middle Ordovician Stones River and Nashville Groups produce most of the oil (table 1). Produc tion in the Swan Creek field currently comes from a fractured anticline between the Clinchport thrust fault (fig. 3) and the Wallen Valley thrust fault where older strata, starting with the Lower Cambrian Rome Formation, overlie a syncline that contains rocks as young as Mississippian (Hatcher and others and others, 2001). Production from the Ben Hur and Rose Hill wells comes from anticlinal structures in the Bales block, between the Pine Mountain thrust fault (which is exposed on the surface to the west in Kentucky) and the subsurface Bales thrust fault (Nolde, 1992). Most of the hydrocarbon production is from Middle to Upper Ordovician carbonate reservoirs along the axis of the Powell Valley anticline (fig. 3, table 1), with most of the gas production coming from what is called the "Stones River" (a phrase used by drillers in Lee County, Va.), which refers to all the formations between the base of the Trenton Limestone and the unconformity above the Knox Group (see fig. 2); for the purposes of this paper, these formations will be referred to as the "Stones River" formations. Two of the shal lowest wells in this study, PO-1 in the Rose Hill field and No. 108 in the Ben Hur field, were drilled in fensters where the Bales block anticlinal structure is exposed by erosion (fig. 3). Recognition and description of these fensters by Butts (1927) and mapping by Miller and Brosge (1954) greatly aided the successful drilling for hydrocarbons in the Ben Hur and Rose Hill fields (Bartlett, 1989). Reservoirs in the competent nonporous Ordovician limestones and dolomites above the Knox unconformity in the Cumberland overthrust sheet were formed mainly by second ary fracturing during tectonic movements and vug formation during dolomitization (Miller and Brosge, 1954; Nolde, 1992; Ryder, 1996). Fracturing in the Ordovician carbonates is con centrated at the top and bottom of the units (Gathright, 1981). The probability for successful drilling for hydrocarbons in all three fields is enhanced by finding intersecting sets of these fractures (Bartlett, 1989; Hatcher and others, 2001). Below the Knox unconformity, vuggy porosity (resulting from karst formation), subaerial exposure, and migration of deep basin fluids is a major factor in reservoir formation in the Knox Group carbonate reservoirs (Ryder, 1996; Hatcher and others and others, 2001). Possible Source Rocks Black shales from either the Ordovician Reedsville Shale (equivalent to the Utica Shale in other parts of the Appalachians) or the Devonian to Mississippian Chattanooga Shale are the most likely sources for the hydrocarbons in the Cumberland overthrust sheet. Middle and Upper Ordovician black shales and argillaceous limestones are source beds in the Appalachian basin from New York to Alabama, with thick nesses ranging from 200 to 400 feet (ft) and total organic carbon (TOC) values ranging from 0.5 to 3 weight percent (Ryder, 1996; Ryder and others, 1998). Ryder and others (1998) concluded that the Middle to Upper Ordovician Utica Shale (Reedsville Shale equivalent; fig. 2) is the hydrocarbon
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 5 Figure 3. Generalized cross sections of the study area. A, Section through the Ben Hur field, Lee County, Va., showing position of a well drilled into a fenster and a well drilled on the southeast flank of the Powell Valley anticline through the Pine Mountain thrust fault. Section is perpendicular to the Powell Valley anticline (see fig. 1). Modified from Bartlett (1989). B, Section through the Swan Creek field, Hancock County, Tenn., showing well drilled through the Clinchport thrust fault. Section is perpendicular to the Clinchport thrust fault (see fig. 1). Modified from Hatcher and others (2001). Fig ure - R efe r t o C apt ion
6 Coal and Petroleum Resources in the Appalachian Basin Table 1. Selected well information. [Elemental analyses are accurate to within 2 percent. Sulfur and nitrogen values are not reported in this table because they were below the instrumental detection limits in samples from the Ben Hur and Rose Hill oil fields and were not reported for samples from the Swan Creek oil field. Abbreviations are as follows: API, American Petroleum Institute; ft, feet; na, not analyzed; nr, not reported; δ13C, isotopic compo sition of carbon (carbon 13, 13C) reported as the deviation (expressed as δ13C) in units of parts per thousand (per mil) relative to the Vienna Peedee belemnite (VPDB) standard] Well name Type of sample API number Oil field Producing Age of producing formation Perforated interval (average depth, ft) Selected analyses for oil samples Carbon (weight percent) Hydrogen (weight percent) δ13C for aromatic fraction (per mil) δ13C for saturated fraction (per mil) API gravity (degrees) Appendix figure number No. 108 Oil and gas 4510500438 Ben Hur Trenton Limestone and "Stones River" formations Middle Ordovician 1,808 A1 James Dean No. 1 Oil 4510520452 Ben Hur Trenton Limestone Middle Ordovician 2,374 A2 Yeary No. 1 Oil and gas 4510520719 Ben Hur Trenton Limestone Middle Ordovician 2,553 A3 No. 8709 Oil and gas 4510501123 Ben Hur Trenton Limestone Middle Ordovician 3,806 nr A4 No. PO-1 Oil and gas 4510500371 Rose Hill Clinch Sandstone and Trenton Limestone Silurian and Ordovician 1,330 A5 Morgan Hensley No.1 Oil and gas 4510520672 Rose Hill Trenton Limestone (oil) and "Stones River" formations (gas) Middle Ordovician 2,711 A6 No. 8708 Oil and gas 4510501141 Rose Hill "Stones River" formations Middle Ordovician 2,910 A7 No. 1001 Oil and gas 4510505252 Rose Hill "Stones River" formations Middle Ordovician 3,083 A8 No. 1000 Oil and gas 4510504929 Rose Hill "Stones River" formations Middle Ordovician 3,096 A9 Sarah Hensley No. 1 Oil and gas 4510520572 Rose Hill Knox Group (upper part) Cambrian and Ordovician 3,994 A10 Paul Reed No. 5 Gas 4106720018 Swan Creek Stones River Group Middle Ordovician 2,155 na na nr nr nr No oil sample Paul Reed No. 2 Oil and gas 4106720013 Swan Creek Stones River Group Middle Ordovician 2,958 na na A11 Paul Reed No. 4 Gas 4106720017 Swan Creek Knox Group Cambrian and Ordovician 4,466 na na nr nr nr No oil sample Paul Reed No. 1 Oil and gas 4106720001 Swan Creek Knox Group Cambrian and Ordovician 4,519 na na A12 Gary Patton No. 1 Oil and gas 4106720006 Swan Creek Knox Group Cambrian and Ordovician 4,740 na na A13 Stephen Lawson No. 2 Oil and gas 4106720019 Swan Creek Nashville Group (oil) and Knox Group (gas) Cambrian and Ordovician 2,465 (oil) 4,839 (gas) na na A14 Sutton Heirs No. 1 Gas 4106720016 Swan Creek Knox Group Cambrian and Ordovician 4,902 na na nr nr nr No oil sample Woodrow Davis No. 1 Gas 4102520136 Swan Creek Knox Group Cambrian and Ordovician 4,908 na na nr nr nr No oil sample
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 7 source not only for Silurian reservoirs in Ohio and Indiana in the United States and Ontario, Canada, but also for the under lying Knox Group, Trenton Group, and Black River Group (Stones River Group equivalent) due to the updip migration of the produced oils. According to Nolde (1992), the Reedsville Shale and the Devonian to Mississippian Chattanooga Shale in Lee County, Va., are still in the zone of oil generation and may be local sources for the Cumberland overthrust sheet reser voirs because they underlie the producing formations in the Bales block (Ryder, 1996). The Gloeocapsomorpha Prisca Enigma In 1998, U.S. Geological Survey (USGS) analyzed oil samples from four wells in the Swan Creek field in Hancock and Claiborne Counties, Tenn. (U.S. Geological Survey, 1998, unpub. data). The analyses showed evidence that the oils were most likely from an Ordovician carbonate source (Hatcher and others, 2001) because the oil samples had unique characteristics. They had a strong odd-carbon preference in the n-C9 to n-C19 range, with a noticeable reduction in peak height after n-C19 (see appendix A). The samples were highly enriched in n-alkanes relative to other components and had very low relative con centrations of isoprenoids, including pristane and phytane. The American Petroleum Institute (API) gravity for oil from the Swan Creek field is around 52° (Hatcher and others, 2001). Martin and others (1963) first noted that this particular profile was unique to early Paleozoic oils. Reed and others (1986), Longman and Palmer (1987), Jacobson and others (1988), and Obermajer and others (2000) are examples of studies where other researchers documented that oils produced from Ordovician carbonate reservoirs in the United States and Canada also had this unique signature. Douglas and others (1991) showed in pyrolysis experiments that the Middle Ordo vician Guttenberg oil rock (a driller's term), which is enriched in the remains of the microfossil Gloeocapsomorpha prisca (G. prisca), produced this profile. The Ordovician Guttenberg Member of the Decorah Formation in Iowa has been well documented as being the purest example in the United States to date of a G. prisca-rich oil source rock (Peters and others, 2005). Oil source rocks that are equivalent to the Ordovician Trenton and Stones River Groups in midcontinental and eastcentral United States also are associated with G. prisca-rich source rocks (Reed and others, 1986; Hatch and others, 1987, 1990; Guthrie and Pratt, 1995). Kerogen is organic material that has a high molecular weight, occurs naturally in sedimentary rocks, and is capable of producing hydrocarbons. Early Paleozoic G. prisca-rich kerogens, found in many areas in the world, are similar to those found in Ordovician kukersites of Estonia; the kukersites are mudstones rich in marine carbonates, where G. prisca microfossils occur in the organic-rich laminae (Fowler and Douglas, 1984; Reed and others, 1986; Foster and others, 1990; Lille, 2003). G. prisca-enriched source rocks, which are primarily Ordovician, formed in marine areas in the tropical range restricted to about 5° on either side of the paleoequa tor (Tyson, 1995; Peters and others, 2005). These conditions existed during the time at which the Cambrian and Ordovician carbonate formations in the Cumberland overthrust sheet were forming (Pfiel and Read, 1980; Read, 1980). Although oils with G. prisca characteristics are classi fied as being from an algal source dominated by the lipid-rich membranes of microalgae (Peters and others, 2005), there is no identifiable living organism that produces the polymers that form those microfossils (Blokker and others, 2001). G. prisca microfossils appear as nearly structureless clumps and are difficult to identify consistently in thin sections (Jacobson and others, 1988). Oils from G. prisca-rich sources have very low relative amounts of isoprenoids, including pristane and phytane, as well as low relative amounts of steranes, terpanes, and other biomarkers (Reed and others, 1986; Longman and Palmer, 1987; Guthrie and Pratt, 1995; Obermajer and others, 1998, 2000). Both characteristics make it very difficult to bio logically classify G. prisca with certainty. Tyson (1995) and Peters and others (2005) summarized the controversy concern ing the biological classification of the G. prisca organism and the unique geochemical characteristics reported in oils derived from G. prisca-enriched source rocks. A study by Jacobson and others (1988) of core samples from the mostly carbonate Ordovician Decorah Formation in eastern Iowa, which is similar in age and type to Ordovi cian carbonate strata of the Appalachian basin, revealed kerogen that occurred as two different morphotypes. One was described as "platy" and designated "Assemblage A"; the other, described as "fluffy," was designated "Assemblage B." Assemblage A microfossils occur in carbonate-dominated environments and produce the characteristic G. prisca profile, which is a highly n-alkane-rich oil with very low amounts of isoprenoids, such as pristane and phytane. Assemblage B produces oil with more isoprenoids, including pristane and phytane, which is indicative of a marine kerogen deposited in a shale-rich environment. The G. prisca profile in oils is independent of the thermal maturity of the source rock. Reed and others (1986) showed that Ordovician oils with the G. prisca profile can come from sources ranging in thermal maturity from 0.65 percent to 1.25 percent calculated vitrinite reflectance (%Ro). G. prisca-rich source rocks and their oils exhibit a broad range of isotopic values. The δ13C value of G. prisca-enriched source rocks and oils can vary as much as 9.3 per mil within a single sample (Reed and others, 1986; Hatch and others, 1987; Ludvigson and others, 1996; Mastalerz and others, 2003). (δ13C represents the isotopic composition of carbon (carbon 13, 13C) in kerogen reported as the deviation in units (δ13C) of parts per thousand (per mil) relative to the Vienna Peedee belemnite (VPDB) standard.) Recent findings have shown that the G. prisca signa ture in oils can no longer be used to assign the age of the oil source rock to the Ordovician. Evidence of G. prisca material has been identified in Late Devonian source rocks in south ern Alberta, Canada (Fowler and others, 2004) and in the
8 Coal and Petroleum Resources in the Appalachian Basin Cambrian Rogersville Shale, a local source rock in the Rome trough in West Virginia and eastern Kentucky (Ryder and oth ers, 2005). To date, there are no reported analyses of any source rocks in the Cumberland overthrust sheet and no G. prisca-rich intervals have been reported in well cores from the Ben Hur, Rose Hill, or Swan Creek fields. In order to contribute to the understanding of the nature of the hydrocarbon sources in the Cumberland overthrust sheet, the four oils from the Swan Creek field, initially analyzed in 1998, were re-analyzed in 2006 for this study and complemented with samples collected in 2006 from 13 wells in the nearby Ben Hur and Rose Hill fields (fig. 1). The correlation of source rocks to petroleum accumulations is one of the most critical components of oil and gas assess ments based on petroleum systems (Magoon and Dow, 1994). This correlation is the basis for identifying oil migration paths and areas of potential undiscovered resources. Because no analyses of possible source rock candidates in the study area have been reported, analyses of the crude oils and natural gases were used to infer the characteristics of their sources. Although the results from the biomarker analyses of the Cumberland overthrust sheet oils are included in this study, the interpretations of oil source-rock characteristics in this report are based on compositional ratios calculated from gasoline-range hydrocarbons and isoprenoids. Oil-to-oil and oil-to-source-rock correlations using gasoline-range crude oil compositional ratios are usually complemented by analyses of biomarkers with a higher molecular weight. However, early Paleozoic oils have very low concentrations of biomarkers, making conventional biomarker analyses difficult to interpret (Reed and others, 1986; Guthrie and Pratt, 1995; Peters and others, 2005). For this reason, Obermajer and others (2000) used gasoline-range parameters to refine their biomarker studies of oils in the Williston basin, which included oils from Ordovician source rocks that contained G. prisca. Methods Oil and gas samples for this study were collected from the Rose Hill and Ben Hur fields in Lee County, southwestern Virginia, in 2006 by the authors. Richard J. Diecchio of George Mason University provided an oil sample from the James Dean No. 1 well. Oil and gas samples from the Swan Creek field in Hancock and Claiborne Counties, Tenn., were collected by Rob ert D. Hatcher of the University of Tennessee and Christopher Swezey and Robert C. Milici of the USGS in 1998. Sample Collection The oil produced in the Ben Hur and Rose Hill fields was pumped from the well head to stock tanks for periodic collection by a tanker truck. The well operator at each well site bailed crude oil samples from the stock tanks in the spring of 2006, with the exception of the sample from well PO-1, which was collected from a well-head drain line. Because the Swan Creek field is primarily a gas field, oil from wells in this field also were stored in well-head tanks before being trucked to the refinery (Hatcher and others, 2001); therefore, we assumed that the Swan Creek oils also were bailed from the stock tanks. In the field, the oil samples were collected in metal cans with screw-on lids. Upon receipt at USGS in Reston, Va., they were split into glass vials, which were capped by screw tops with nonstick liners. Gas samples were collected at the well head in evacuated stainless-steel gas sampling cylinders provided by Isotech Laboratories, Inc., of Champaign, Ill., and were shipped there for analysis. Instrumental Analysis Whole Crude Oil Whole crude-oil samples were analyzed at the USGS Gas Chromatography Laboratory in Reston, Va., using a Hewlett/ Packard Model 6890 gas chromatograph (GC) with a 60 meter (m) × 0.32 mm × 0.25 micrometer (µm) DB-1 fused silica capillary column. Samples (0.1 microliter (µL)) were injected manually into a split inlet system (split flow of 15 milliliters per minute (mL/min)) that split the samples 200 to 1. Helium carrier gas carried the samples at a constant velocity of 36 centimeters per second (cm/sec) to a flame ionization detector (FID). After an initial hold at 34°C for 6 min, the program mable oven increased temperature from 34°C to 320°C at 6°C/min and then held the temperature at 320°C for 20 min. A crude oil sample from Kentucky, which was analyzed under the same conditions and at the same time as the other oil samples, served as a check on instrument performance and reproducibility. Data from the Kentucky oil sample analyses were combined with other analyses of this oil sample con ducted over a period of two years in order to determine if the gasoline-range parameter ratios remained constant. The whole-oil chromatograms were integrated using Justice Innovations, Inc.'s ChromPerfect Spirit version 5.5 software. A baseline chromatogram (injectionless chromatogram), which was run on the same day under the same conditions as the oil samples that were being analyzed, was subtracted from each raw chro matogram. After subtraction of the baseline, the peak areas up to n-C9 were integrated from the horizontal baseline. After n-C9, the baseline was reset to integrate peak areas from peak valley to peak valley. Because they are more representative of concentra tion than peak heights, peak areas were used to calculate compo sitional ratios shown in the various diagnostic plots. Elemental, Isotopic, and API Gravity Analysis Splits of the oil samples were sent to the USGS Organic Geochemistry Laboratory in Denver, Colo., where a Carlo Erba Model 1110 Elemental Analyzer (EA) was used to determine the weight percent of carbon, hydrogen, nitrogen, and sulfur for the oil samples. The oil samples (approximately
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 9 2 milligrams) were weighed into tin capsules and introduced to the EA. After undergoing oxygen-enriched combustion, reduction, and oxidation, continuously flowing helium carrier gas transported the gas sample to a gas chromatograph for separation. A thermal conductivity detector then analyzed the gas sample compositions. The Organic Geochemistry Labora tory also analyzed the oil samples for API gravity and for the δ13C values for the aromatic and saturated fractions. Standard Biomarker Analysis The USGS Organic Geochemistry Laboratory in Denver, Colo., also analyzed the oil samples for biomarkers by gas chromatography with a mass-selective detector (GC-MS). The saturated and aromatic hydrocarbon fractions of the crude oil samples were analyzed simultaneously for terpanes, steranes, monoaromatic steroids, and triaromatic steroids by selected ion monitoring of mass-to-charge ratios (m/z) of 191.1800, 217.1956, 231.1174, and 253.1956, respectively. Quantities were calculated by normalizing the marker peak areas in each sample to the peak area of an internal standard, 5β-cholane, added to each sample. The aromatic fractions of the crude oil samples were analyzed by GC and FID methods described above. Natural Gas Analysis Natural gas samples were analyzed for molecular compo sition by gas chromatography and for stable isotopic composi tion by isotope-ratio mass spectrometry at Isotech Laborato ries, Inc. The carbon and hydrogen isotopic compositions were determined for methane in all samples. Four gas samples from wells in the Ben Hur and Rose Hill fields were submitted for isotopic analysis of δ13C and δ2H in propane through pentane, including iso-butane. δ13C represents the isotopic composition of carbon (carbon 13, 13C) in methane reported as the devia tion in units (δ13C) of parts per thousand (per mil) relative to the Vienna Peedee belemnite (VPDB) standard. δ2H repre sents the isotopic composition of hydrogen (deuterium, 2H) in methane reported as the deviation in units (δ2H) of parts per thousand (per mil) relative to the Vienna standard mean ocean water (VSMOW). The carbon isotopic composition of carbon dioxide was determined for the one sample having a carbondioxide concentration above the minimum detection limit for analysis. Natural-gas-geochemistry molecular compositions are based on standards accurate to within 2 percent. Carbon isotope precision is ±0.1 per mil and the precision is ±2 per mil for deuterium. Results Oil Geochemistry The resulting measurements (in weight percent) from the carbon, hydrogen, nitrogen, and sulfur analyses, the δ13C values for the aromatic and saturated oil fractions, and the API gravity values for the oil samples are shown in table 1. Selected light hydrocarbon compositional ratios are shown in table 2 along with reproducibility data from the in-house oil standard. Whole-oil gas chromatograms for all the oil samples are shown in appendix A (see appendix A, figures A1 to A14), along with a representative aromatic fraction chromatogram and representative biomarker fragmentograms (appendix B, fig. B1). The profiles of the crude oil samples are presented as integrated peak heights (relative amounts) in millivolts (verti cal axis) versus retention time in minutes (horizontal axis). The oil samples showed a strong odd-carbon preference in the n-C9 to n-C19 range, with a noticeable reduction in peak height after n-C19. They were highly enriched in n-alkanes relative to other components with a very low relative content of isopren oids, including pristane and phytane. Elemental, Isotopic, and API Gravity Analysis The oil samples from the three fields had high API gravities (43°-52°), with the samples from the Swan Creek field having slightly higher API gravities than samples from the Ben Hur and Rose Hill fields. (Sulfur and nitrogen were below the instrumental detection limit.) Carbon values for the Ben Hur and Rose Hill oil samples ranged from 61.3 to 82.4 weight percent, and hydrogen ranged from 10.6 to 14.1 weight percent. The carbon preference index (CPI) ranged from 1.2 to 1.6. δ13C values for the aromatic fractions of oil samples from the Ben Hur and Rose Hill fields ranged from -25.2 to -28.5 per mil; for the saturated fractions of the samples, the range was -22.2 to -28.8 per mil. Biomarker Profiles Biomarkers for terpanes, steranes, and aromatic com pounds were detected in most of the samples although the quantities were generally less than 10 parts per million, which is close to the detection limit for conventional biomarker analyses (appendix B, fig. B1). No steranes or terpanes were reported for the Paul Reed No. 1 well in the Swan Creek field. No triaromatic sterane markers (scanning range of m/z=231) or monoaromatic sterane markers (scanning range of m/z=253) were detected in the study oils. Terpanes Tricyclic terpanes from C19 to C29 were present, but the C22 tricyclic terpane marker was generally very small relative
10 Coal and Petroleum Resources in the Appalachian Basin Table 2. Selected light hydrocarbon compositional ratios and reproducibility data. [Compositional ratios (Thompson, 1983) are as follows: A benzene / n-hexane; B toluene / n-heptane; F n-heptane / methylcyclohexane; Heptane value (n-heptane / sum of peak volumes between and including cyclohexane and methylcyclohexane) × 100; iso-heptane value (2-methylcyclohexane + 2,3-dimethylpentane + 3-methylcyclohexanes / 1-cis-3-, 1-trans-3-, and 1-trans-2-dimethylcyclopentane. K1 (Mango, 1987) 2-methylhexane + 2,3-dimethylpentane / 3-methylhexane + 2,4-dimethylpentane. Carbon preference index [(15+17+19) / (14+16+18) + (15+17+19) / (16+18+20)] × 0.5 (numbers refer to number of carbons in n-alkanes)] Well name Oil-producing Oil field A B F Heptane value iso-heptane value K1 Carbon preference index Pristane / phytane Pristane / n-C17 Phytane / n-C18 Selected light hydrocarbon compositional ratios No. 108 Trenton Limestone and "Stones River" formations Ben Hur James Dean No. 1 Trenton Limestone Ben Hur Yeary No. 1 Trenton Limestone Ben Hur No. 8709 Trenton Limestone Ben Hur No. PO-1 Clinch Sandstone and Trenton Limestone Rose Hill Morgan Hensley No. 1 Trenton Limestone Rose Hill No. 8708 "Stones River" formations Rose Hill No. 1001 "Stones River" formations Rose Hill No. 1000 "Stones River" formations Rose Hill Sarah Hensley No. 1 "Stones River" formations Rose Hill Paul Reed No. 2 Stones River Group Swan Creek Paul Reed No. 1 Knox Group Swan Creek Gary Patton No. 1 Knox Group Swan Creek Stephen Lawson No. 2 Nashville Group Swan Creek Reproducibility of compositional ratios of 14 analyses of a control sample over a period of 2 years Mean standard deviation 95 percent confidence interval
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 11 to the other tricyclic terpanes. The C19 tricyclic terpane peak was the dominant peak of all of the tricyclics and was often the most abundant of the identified markers. C24 tetracyclic terpane was present in similar abundance to the tricyclic terpanes. There was little or no bisnorhopane and no nor moretane, except for the oil sample from the Stephen Lawson No. 2 well in the Swan Creek field, which had a measurable normoretane peak. The C27 18α-trisnorneohopane (Ts)-to-C27 17α-trisnorhopane (Tm) ratio was greater than 1 except for the Gary Patton No. 1 well in the Swan Creek field. The Ts peak was frequently one of the dominant peaks. Neonorho pane was more abundant than norhopane except for the oil sample from the Gary Patton No. 1 well. C30 hopane, although present, was usually not one of the predominant components. The four Swan Creek oil samples had more C30 hopane than oil samples from the Rose Hill and Ben Hur wells. The Swan Creek samples had relatively low amounts of the homohopane series, up to C33 homohopanes. The Ben Hur and Rose Hill samples had relatively low amounts of homohopanes, ranging from none to C35 homohopanes. Steranes In all of the oil samples, C27 steranes dominated the C29 steranes. The Swan Creek samples had more C27 20R ααα steranes than C27 20S βα diasteranes. The Ben Hur and Rose Hill samples had more C27 20S βα diasteranes than C27 20R ααα steranes and they also had more pregnanes. Aromatic Fraction Biomarkers Except for the oil sample from the Stephen Lawson No. 1 well in the Swan Creek field, which had only a very small naphthalene peak, 1- and 2-methylnaphthalene were the domi nant components in the aromatic fraction of the samples in this study. The oil samples were very low in sulfur-derived bio markers: there were no measurable dibenzothiophene peaks, phenanthrene was present (but in relatively low amounts), and 9-methylphenanthrene was the dominant methylphenanthrene peak. Natural Gas Geochemistry Molecular compositions and isotopic compositions of the natural gas samples are shown in tables 3 and 4, respectively. With the exception of two Swan Creek wells (Sutton Heirs No. 1 and Paul Reed No. 5), all of the wells produced gas as well as oil. Molecular Composition Molecular profiles of gas sample compositions are shown in figure 4. None of the gas samples had detectable levels of hydrogen sulfide. Gas samples from two of the wells, No. 8709 and Yeary No. 1, had elevated levels of nitrogen, oxygen, argon, and carbon dioxide and reduced values for the hydro carbon components, which indicated atmospheric contamina tion in the samples. These two wells are not shown on figure 4. The trace levels of oxygen were assumed to be the result of low levels of air contamination. The wells producing natural gas from the Ordovician formations in the Rose Hill and Ben Hur fields had approximately half the nitrogen content (1.71 to 2.9 weight percent) compared to the Swan Creek wells from the Knox Group with a range of 4.84 to 5.09 weight percent. Well No. 8708 in the Rose Hill field, which was redrilled down to the top of the upper part of the Knox Group in the spring of 2006, produced gas with little or no hydrogen sulfide (Mark Deering, Virginia Department of Mines, Minerals, and Energy, unpub. data, March, 2006). Isotopic Composition of Methane, Ethane, Deuterium, and Carbon Dioxide For all gas samples, the δ13C values for methane ranged from -44.1 per mil to -50.8 per mil and deuterium values for methane ranged from -164.6 to -215.1 per mil; the gas samples from the Knox Group wells in the Swan Creek field had the least negative values. The Knox Group gas samples from the Swan Creek field were almost identical: the δ13C values for methane ranged from -44.1 to -44.4 per mil, δ13C values for ethane ranged from -36.1 to -36.3 per mil, and deuterium values for methane ranged from -164.6 to -169.6 per mil. For the four deeper wells in Lee County, the δ13C values for methane in the Rose Hill field were -47.4 per mil and -47.6 per mil and the δ13C values for methane in the Ben Hur field were -45.1 per mil and -45.5 per mil. The shallowest wells (in the fensters) had the most negative δ13C values for methane: -49.4 per mil in the Ben Hur field and -50.8 per mil in the Rose Hill field. Only the No. 1001 well, producing from the "Stones River" formations in the Rose Hill field, had enough carbon dioxide for δ13C analysis. The δ13C composition of carbon dioxide from this well was -6.2 per mil. Isotopic Composition of Gases With Higher Carbon Numbers Of the four samples submitted for isotopic analyses of gases with a higher carbon number (table 4), two of the samples were from wells located in the fenster: (1) No. 108 in the Ben Hur field, producing from the Trenton Limestone and "Stones River" formations, and (2) PO-1 in the Rose Hill field, producing from the Clinch Formation and Trenton Limestone. The other two samples were from the No. 8708 and the No. 1001 wells located in the Rose Hill field, produc ing from the "Stones River" formations. Gases with higher carbon numbers from the Swan Creek field were not analyzed for isotopic composition. The δ13C values for ethane from the wells located in the fenster (-33.6 and -34.0 per mil) were more negative than
12 Coal and Petroleum Resources in the Appalachian Basin Table 3. Molecular composition of gases. [Chemical analyses are based on standards accurate to within 2 percent. Hydrogen sulfide gas was not detected in any of the gas samples. na, not analyzed] Well name Oil field Producing
Average depth of perforated interval (ft) Molecular composition of gases, in percent Helium Hydrogen Argon Oxygen Carbon dioxide Nitrogen Methane Ethane Propane iso-butane n-butane iso-pentane n-pentane Hexane+ No. 108 Ben Hur Trenton Limestone and "Stones River" formations 1,808 Yeary No. 1 Ben Hur Trenton Limestone 2,553 No. 8709 Ben Hur Trenton Limestone 3,806 No. PO-1 Rose Hill Clinch Sandstone and Trenton Limestone 1,330 Morgan Hensley No. 1 Rose Hill "Stones River" formations 2,711 No. 8708 Rose Hill "Stones River" formations 2,910 No. 1001 Rose Hill "Stones River" formations 3,083 No. 1000 Rose Hill "Stones River" formations 3,096 Sarah Hensley No. 1 Rose Hill Knox Group (upper part) 3,994 Paul Reed No. 5 Swan Creek Stones River Group 2,155 Paul Reed No. 2 Swan Creek Stones River Group 2,958 Paul Reed No. 4 Swan Creek Knox Group 4,466 na na Paul Reed No. 1 Swan Creek Knox Group 4,519 na na Gary Patton No. 1 Swan Creek Knox Group 4,740 na na Stephen Lawson No. 2 Swan Creek Knox Group 4,839 na na Sutton Heirs No. 1 Swan Creek Knox Group 4,902 na na Woodrow Davis No. 1 Swan Creek Knox Group 4,908
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 13 Table 4. Isotopic composition of carbon and hydrogen in gas samples. [Isotopic composition of carbon (carbon 13, 13C) is reported as the deviation (expressed as δ13C) in units of parts per thousand (per mil) relative to the Vienna Peedee belemnite (VPDB) standard. Isotopic composition of hydrogen (deuterium, 2H) is reported as the deviation (expressed as δ2H) in units of parts per thousand (per mil) relative to the Vienna standard mean ocean water (VSMOW). Isotope precision is ±0.1 per mil for carbon 13 and ±2 per mil for deuterium. Abbreviations are as follows: ft, feet; na, not analyzed] Well name Oil field Perforated interval (average depth, ft) Isotopic composition (per mil) δ13C carbon dioxide δ13C methane δ2H methane δ13C ethane δ2H ethane δ13C propane δ2H propane δ13C iso-butane δ2H iso-butane δ13C n-butane δ2H n-butane δ13 C n-pentane δ2H n-pentane No. 108 Ben Hur 1,808 na Yeary No. 1 Ben Hur 2,553 na na na na na na na na na na na No. 8709 Ben Hur 3,806 na na na na na na na na na na na No. PO-1 Rose Hill 1,330 na Morgan Hensley No. 1 Rose Hill 2,711 na na na na na na na na na na na No. 8708 Rose Hill 2,910 na No. 1001 Rose Hill 3,083 No. 1000 Rose Hill 3,096 na na na na na na na na na na na Sarah Hensley No. 1 Rose Hill 3,994 na na na na na na na na na na na Paul Reed No. 5 Swan Creek 2,155 na na na na na na na na Paul Reed No. 2 Swan Creek 2,958 na na na na na na na na Paul Reed No. 4 Swan Creek 4,466 na na na na na na na na Paul Reed No. 1 Swan Creek 4,519 na na na na na na na na Gary Patton No. 1 Swan Creek 4,740 na na na na na na na na Stephen Lawson No. 2 Swan Creek 4,839 na na na na na na na na Sutton Heirs No. 1 Swan Creek 4,902 na na na na na na na na Woodrow Davis No. 1 Swan Creek 4,908 na na na na na na na na
14 Coal and Petroleum Resources in the Appalachian Basin A Figure 4A - Refer to Cap tion B Figure 4B - Refer to Cap tion
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 15 those from the deeper wells in the Rose Hill field (-31.6 and -31.9 per mil). This was also true for the δ2H values for ethane from the fenster wells (-148.8 and -155.0 per mil) compared with those from the deeper wells (-122.6 and -134.9 per mil). The range of δ13C values for propane was from -29.0 to -29.6 per mil, with the most negative value coming from one of the wells located in the fenster. The δ2H values for propane (-117.2 and -130.3 per mil) in the deeper Rose Hill wells were at the extreme ends of the range of values; the wells located in the fenster in the Rose Hill field had δ2H values for propane of -125.7 and -128.3 per mil. For all four gas samples, the δ13C values for iso-butane ranged from -29.3 to -31.0 per mil and the δ2H values for iso-butane ranged from -108.5 to -114.2 per mil; the deeper Rose Hill wells had both the most negative and the most posi tive δ2H values. Gas samples from three of the wells had δ13C values for n-butane of -27.1 per mil. The PO-1 well, which is located in the fenster in the Rose Hill field, had the most nega tive δ13C n-butane value of -27.9 per mil. The δ2H values for n-butane ranged from -114.0 to -123.8 per mil, with the most negative value coming from the PO-1 well. The δ13C values for pentane from the wells in the fenster (-26.1 and -26.5 per mil) were more negative than those in the deeper wells in the Rose Hill field (-24.8 and -25.1 per mil). The δ2H values for pentane in the four wells ranged from -109.9 to -118.9 per mil, with the deeper Rose Hill wells having the most extreme values. Figure 4 (facing page and this page). Bar charts showing percentage of each molecular component in the analyzed gas samples from the Cumberland overthrust sheet. A, Gas samples from the Trenton Limestone in Virginia and the Nashville Group in Tennessee. B, Gas samples from the "Stones River" formations in Virginia and the Stones River Group in Tennessee. C, Gas samples from the Knox Group. Figure 4C - Refer to Capt ion
16 Coal and Petroleum Resources in the Appalachian Basin Discussion Data Quality The total quantity of gasoline-range hydrocarbons can be affected by evaporative loss, which may be seen in rela tive amounts in the gas chromatographic signatures of the oil samples in this study. For example, the profile for the Yeary No. 1 well (appendix A, fig. A3) exhibited higher peak heights in the n-C5 to n-C7 range than in the n-C8+ range. This sample (and others with this type of profile) presumably represented relatively fresh samples because it had a greater abundance of the light hydrocarbons relative to the n-alkanes with higher carbon numbers and a flat baseline. The sample from the No. 8709 well (appendix A, fig. A4) exhibited reduced peak heights in the lighter n-alkane range compared to the sample from the Yeary No. 1 well, which indicates a relatively greater loss of light hydrocarbons in the sample from the No. 8709 well. This evaporative loss could have been caused by sample handling and storage. According to Thompson (1983), Ten Haven (1996), Mango (1997), Jarvie (2001), and references therein, com positional ratios in oil samples are very useful for assessing oil type, maturation, and other characteristics because the C4-to-C8 hydrocarbon ratios remain relatively constant. C4 to C8 hydrocarbons have similar boiling points and evaporative loss should not affect relative abundance. Over a period of two years, the gasoline-range parameters collected for the Ken tucky in-house standard oil sample (stored at room tempera ture in a capped glass vial) remained relatively constant with standard deviations ranging from 0.001 to 0.06 and 95 percent confidence intervals ranging from 0.0001 to 0.022 (table 2). Ten Haven (1996) found that ratios can remain constant even after oil samples are stored for 20 years at room temperature. Aerobic biodegradation can also affect data collected from oil samples by selectively removing n-alkanes, such as propane, and contributing compounds that cannot be identified using routine gas chromatography (Connan, 1984; Wenger and others, 2002). The Morgan Hensley No. 1 well profile (appendix A, fig. A6) exhibited an easily measurable pristane and phytane peak, a more extreme reduction in peak heights of the lighter n-alkanes, and an unresolved complex mixture "hump," which did not appear in chromatograms for the other oil samples. The loss of lighter n-alkanes in the Morgan Hens ley No. 1 well sample most likely was due to both evapora tive loss and slight aerobic biodegradation (using the criteria of Wenger and others, 2002) because the oil from which this sample was taken had been stored in the holding tank for over a year, according to the well operator. Isoprenoids in oil samples from Ordovician source rocks varied in abundance from almost undetectable to that typi cal of most oils generated from Phanerozoic marine sedi ments (Jacobson and others, 1988). The small amounts of the isoprenoids pristane and phytane in the oil samples in this study were confirmed using a gas chromatograph with a mass selective detector. Compositional Trends in Oil From the Cumberland Overthrust Sheet Gas Chromatographic and Biomarker Profiles The oil samples exhibited gas chromatographic profiles similar to oils associated with G. prisca-rich sources from Trenton Group and Stones River Group equivalent sources in the midcontinental and east-central United States (Reed and others, 1986; Hatch and others, 1987, 1990; Guthrie and Pratt, 1995). Biomarkers in the oil samples in this study (appendix B, fig. B1) also were similar to those reported in oils from G. prisca-rich Ordovician source rocks in basins in the midcon tinental United States and in Canada (Longman and Palmer, 1987; Obermajer and others, 1998, 2000). The abundance of biomarkers generally was very low, tricyclic terpanes domi nated the terpane profile, there were more diasteranes than steranes, and the C30 homohopanes were relatively scarce. Normoretane, bisnorhopane, and moretane were rare or not reported. The oil samples in this study appeared to have less C30 hopane and C30 homohopanes than was reported for Ordo vician oils from reservoirs in rocks equivalent to the Trenton and Stones River Groups in Canada and the midcontinental United States (Longman and Palmer, 1987; Obermajer and others, 1998, 2000). The very low amounts of organic sulfur compounds in the oil samples in this study were consistent with observations of Ordovician oils by Douglas and others (1991) and Hughes and others (1995). Light Hydrocarbon Parameters Oils from G. prisca-rich source rocks are typically low in branched and cyclic compounds and have a predominance of n-alkanes compared with oils from younger Paleozoic sources. This relation is shown in figure 5, where n-alkane enrichment in oils is shown by the n-heptane-to-methylcyclohexane ratio (paraffinicity) and enrichment in branched and cyclic com pounds is shown by the toluene-to-n-heptane ratio (aromaticity) (Thompson, 1983). The most extreme examples of n-alkane enrichment with low amounts of aromatic compounds are from oils from the Ordovician Red River Formation of the Williston basin. The sources for the Red River oils are known to be the G. prisca-rich kukersites in the Upper Ordovician Bighorn Group (Osadetz and others, 1992; Osadetz and Snowdon, 1995). Of all the oil samples in this study, the samples from the Rose Hill field were the closest in composition to the Red River oils in n-alkane enrichment, but they were slightly more aromatic. The samples from the Ben Hur and Swan Creek fields trend in composition toward the Devonian and Mississippian oils (Thompson, 1983).
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 17 Figure 5. Graph showing the n-heptane-to-methylcyclohexane ratio versus the toluene-to-n-heptane ratio of oil samples from the Cumberland overthrust sheet and (for comparison) from oil samples from other Paleozoic reservoirs in the United States and Canada (Thompson, 1983). Data for other Paleozoic reservoirs is from Thompson (1983), Osadetz and others (1992), and Jarvie (2001). Fig ure - R efe r t o C apt ion
18 Coal and Petroleum Resources in the Appalachian Basin Another way to classify oils is to compare the ratio of n-heptane compounds (heptane value) to the ratio of iso-hep tane compounds (iso-heptane value) using the data provided in Thompson (1983) (fig. 6). The heptane and iso-heptane values for the oil samples in this study showed a generally positive correlation, but Ordovician oils from G. prisca-rich sources elsewhere usually do not follow this trend. Most of the oils from Devonian and younger reservoirs have heptane values of less than 30 and iso-heptane values of less than 1.5. The oils from the Ordovician Red River Formation in the Williston basin have heptane values of around 60, but the iso-heptane values have an average range of 0.5 to 1.5. There appears to be a compositional trend in the study oils similar to that seen in figure 5. The Rose Hill samples were most like the Red River Formation oils: they had the highest heptane values (between 55 and 60) and their isoheptane values of around 2.5 were well above the average compositional range. The compositions of the Ben Hur and Swan Creek oil samples trended toward those of the oils from younger reservoirs. Isotopic Composition of Oil When the δ13C values of the aromatic fraction of all the oil samples in this study are plotted against the δ13C values of the saturated fraction (fig. 7), the locations on the graph fall into the region indicating a marine kerogen source. The isotopic values for oils from the Ordovician Trenton and Black River Groups in the Michigan basin and for extracts of source rocks from the Guttenberg Member of the Decorah Formation (Hatch and oth ers, 1987) are included in figure 7 for comparison purposes. Oil samples from the Rose Hill field had a wider range of more positive isotopic values, similar to those from the Guttenberg Member. Oil samples from the wells in the fenster in the Rose Hill field and oil samples from the Ben Hur and Swan Creek fields had values which were more similar to those of the Trenton and Black River Group oils from the Michigan basin. The trend was consistent with the composi tional trends shown in the plot of aromaticity versus paraffi nicity (fig. 5) and in the plot of heptane values versus isoheptane values (fig. 6). The range in isotopic values probably reflects that kerogen δ13C values in G. prisca-enriched source rocks can vary as much as 9.3 per mil within a single sample (Hatch and others, 1987; Mastalerz and others, 2003). Evidence for Variations in the Depositional Environment of the Source Rocks Pristane-to-Phytane Ratio and Carbon Preference Index High percentages of G. prisca in kerogen are directly related to depositional environment and produce oils and source rock extracts with dominant odd n-alkanes and low amounts of isoprenoids (including pristane and phytane) (Jacobson and others, 1988). A plot of the pristane-to-phytane ratio versus the carbon preference index (CPI) for the oil samples in this study and elsewhere (fig. 8) may reflect (1) dif ferences in the amount of Assemblage A-type and Assemblage B-type kerogen in the source rock from which the study oils were derived and (2) variations in the depositional environ ment of the source rock. Although the ratio of pristane to phytane in oil can be influenced by other factors and should be used with caution (Hunt, 1996), it is nevertheless very useful for grouping oils and inferring the lithology and depositional environment of oil source rocks (Hughes and others, 1995; Peters and others, 2005). In general, oils derived from marine-carbonate-domi nated kerogen, deposited in water with a higher salinity content and reducing conditions that are far from terrestrial clastic input, have lower pristane-to-phytane ratios (less than or equal to 1) than do oils derived from kerogen that is dominated by conti nental sediments deposited in relatively lesser reducing condi tions (pristane-to-phytane ratios are greater than or equal to 1) (Peters and others, 2005, and references therein). Longman and Palmer (1987) and Jacobson and others (1988) suggested that Assemblage A-type, G. prisca-enriched kerogen was deposited in offshore shallow-shelf limestones far from terrigenous clastic input and that Assemblage B-type kerogen was deposited where more shale had accumulated, thus diluting the G. prisca sig nature; therefore, Assemblage A-type kerogen should produce oil with lower pristane-to-phytane ratios than oil derived from sources with more Assemblage B-type kerogen. Carbon Preference Index The CPI is calculated so that oil with an odd n-alkane predominance has a CPI of greater than 1; the resulting index number can be used to infer the amount of Assemblage A-type, G. prisca-rich kerogen in the oil source rock. Oils from the Mid dle Ordovician reservoirs in the Rose Hill, Ben Hur, and Swan Creek fields appear to have been derived from sediments depos ited in an environment that is similar to that of the Red River Formation oils (fig. 8). The Red River oils were derived from marine G. prisca-rich carbonate mudstones that were deposited so far from land that the amount of terrestrial sediment is very small (Osadetz and others, 1992). The higher pristane-to-phy tane ratios and the lower odd-carbon preference for the Ben Hur and Swan Creek oil samples may reflect more terrestrial input and more mixing with amorphous Assemblage B-type kerogen (Jacobson and others, 1988). The one oil sample in the Swan Creek field known to be from the Knox Group (from the Paul Reed No. 1 well) appeared to be closer in composition to the oils from the Knox Group and Trenton and Black River Group reservoirs in the northern Appalachian basin, which are derived from the Utica Shale (Ryder and others, 1998). Both the pristane-to-phytane ratios and the CPI values indicated that the oil source rocks for the Ben Hur and the Swan Creek oil samples received more clastic input and may reflect a
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 19 Figure 6. Graph showing iso-heptane values versus n-heptane values of oil samples from the Cumberland overthrust sheet and (for comparison) of oil samples from Ordovician to Tertiary reservoirs elsewhere in the United States and Canada. The orange arrow indicates that the composition of oil samples from the Cumberland overthrust sheet trends from Gloeocapsomorpha prisca-rich oils toward the average compositional trend of oil samples derived from Devonian to Tertiary source rocks. Fig ure Refe r to Cap tion
20 Coal and Petroleum Resources in the Appalachian Basin Figure 7. Graph showing the isotopic compositon of carbon (δ13C) of the saturated fraction of the oil versus the δ13C of the aromatic fraction of the oil from the samples in this study, along with δ13C values from the saturated and aromatic fractions of oil samples from the Trenton Group in the Michigan basin (Hatch, 1987) for comparison. Sofer (1984) used isotopic analyses of 339 oil samples from around the world, including the United States, to determine the equation of the line best separating oil derived from terrestrial kerogen sources from oil derived from marine kerogen sources. The oil samples plotted in this graph are below that line and therefore are inferred to have originated from a marine kerogen source. Abbreviations are as follows: ARO, aromatic; SAT, saturated; VPDB, Vienna Peedee belemnite standard. Fig ure - R efe r t o C apt ion
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 21 Figure 8. Graph showing the carbon preference index (CPI) versus the pristane-to-phytane ratio for the oil samples in this study and for oil samples from other geographic areas for comparison. The graph can be used to infer the depositional environment of the source rocks for the oil samples. The carbon preference index is calculated using the following formula: [(n-C14 + n-C16 + n-C18) / (n-C15 + n-C17 + n-C19) + (n-C16 + n-C18 + n-C20) / (n-C15 + n-C17 + n-C19)] × 0.5. Fig ure - R efe r t o C apt ion
22 Coal and Petroleum Resources in the Appalachian Basin different sources. Nevertheless, the variation in the isotopic composition of the gas samples in this study was consistent with the isotopic range found in G. prisca-enriched kerogen (Reed and others, 1986; Hatch and others, 1987; Ludvigson and others, 1996; Mastalerz and others, 2003). Natural Gas Plots Another way to relate the isotopic composition of gas to a source and process is to compare the δ13C composition of each component to the reciprocal of the number of carbons in the component, as shown in figures 12 and 13. This rela tion is based on theoretical concepts of the kinetic isotope effect that occurs during the thermal breakdown of high molecular weight hydrocarbons to low molecular weight gas (Chung and others, 1988; Rooney and others, 1995). An shift in depositional facies, as suggested by Jacobson and others (1988) and Mastalerz and others (2003). The shift could have been caused by wind or water currents affecting a large shallowmarine platform, thus dispersing the G. prisca remains to deeper offshore depositional areas (Foster and others, 1990). Deposits enriched by G. prisca also could have formed in place, such as in algal mats (Foster and others, 1990). The highest TOC values in the Red River Formation are found in biozones where G. prisca material is concentrated in burrows (Heinemann and others, 2005). The emergence of highlands to the east of the Cumber land overthrust sheet, which was caused by regional tectonic activity, may have caused changes in the amounts of clastic detritus in the source rocks (Harris and Milici, 1977; Read, 1980; Milici and others, 2006). The source rocks in the Rose Hill field may have been deposited at or near where they formed, thereby retaining their more pure Assemblage A-type, G. prisca-rich pro file compared with the Ben Hur and Swan Creek source rocks. Natural Gas Geochemistry Methane In this study, the methane contents in the gas samples increased with depth (fig. 9). The gas samples from the shal lowest wells (the wells in the fensters) had the least methane relative to other hydrocarbons. The samples from the deepest wells in the Swan Creek field, which produced gas from the Knox Group, had the highest methane content. This was also true for the deepest wells in the Ben Hur and Rose Hill fields. The stable isotopic composition of methane in the gas samples is shown in figure 10. The samples from the shallow est wells (in the fensters) had the most negative values of δ2H and δ13C. The values become more positive with depth in other wells. The samples from the deepest wells in the Swan Creek field had the most positive methane isotopic values. In sum mary, the deepest wells in the Cumberland overthrust sheet produce gas with the highest methane concentrations and with the most positive isotopic compositions. Ethane and Propane The trend in δ13C values for ethane and propane (fig. 11) is different from that for methane; therefore, depth was probably not a factor in the range of these values. All the gas samples plot in a range associated with early mature oil gener ation. The Rose Hill gas samples plotted in a group closest to the maturity trend line, and the Knox Group gas samples are in a group farthest away. The wells in the fensters and the wells producing gas from the Ordovician reservoirs in the Swan Creek and Ben Hur fields plotted in between these locations. In figure 11, the deviation away from the maturity (vitrinite reflectance, or %Ro) trend line suggests that the maturity of the kerogen source was not the determining factor in the composi tion of the gas with the higher hydrocarbon numbers. Whiti car (1994) infers this to be a result of mixing of gases from Figure 9. Graph showing the ratio of gas wetness (methane ÷ (methane + sum of heavier hydrocarbon gases), in percent) versus the average depth below well head of a gas-producing formation for gas samples from the Cumberland overthrust sheet. The graph shows that methane concentrations increase with depth. Fig ur e Refer to C aptio n
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 23 deflection in the δ13C values of methane in each reservoir is interpreted to be the result of mixing with biogenically gener ated methane. Extrapolation of the isotopic compositions of the gas samples from the Ben Hur and Rose Hill fields to possible kerogen compositions yields estimates of about -23 to -24 per mil, which is at the more positive end of the range of kerogen values for G. prisca-enriched rock samples reported by Hatch and others (1987) and Mastalerz and others (2003). Natural Gas From the Swan Creek Field The natural gas plots for the gas samples from the Stones River Group in the Swan Creek field in Hancock and Clai borne Counties, Tenn., have profiles that are almost straight, which indicates that simple thermal cracking of higher hydro carbons is the major factor in the gas composition (fig. 13). The exception is methane, where the slight downward deflec tion in δ13C suggests some mixing with biogenically generated methane. Extrapolation to possible kerogen source composi tions for these gas samples yields compositions of about -23 to -24 per mil, which is similar to samples from from the Ben Hur and Rose Hill fields. The natural gas plots for the eight gas samples from the Knox Group rocks in the Swan Creek field are almost important consequence of this relation is that the isotopic compositions of gas from a single source should plot on a straight line between the isotopic compositions of methane and the source. Deviations from a straight thermal cracking line suggest that something other than thermal cracking of higher hydrocarbons is the determining factor in the isotopic composition of the gas. Natural Gas From the Ben Hur and Rose Hill Fields The gas samples from the reservoirs in the "Stones River" formations and Trenton Limestone reservoirs in the Ben Hur and Rose Hill fields in Lee County, Va., have differ ent profiles, which suggest that the gas had different sources in each reservoir (fig. 12). The gas samples from the reservoirs in the "Stones River" formations exhibit a deviation from a straight thermal cracking line, with both ethane and methane isotopically more positive than the corresponding gas samples from reservoirs in the Trenton Limestone. The gas samples from the reservoirs in the Trenton Limestone also do not fall on a straight line; instead, they form a concave-downward pat tern, which suggests that the isotopic compositions of ethane and propane are more positive than if they had been generated and trapped without mixing or alteration. The slight downward Figure 10. Graph showing the isotopic composition of carbon (δ13C) in methane relative to the Vienna Peedee belemnite (VPDB) standard versus the isotopic composition of hydrogen (deuterium) (δ2H) in methane relative to the Vienna standard mean ocean water (VSMOW) for gas samples from the Cumberland overthrust sheet. The graph shows that values increase (become less negative) with depth. Figur e 10 - Ref er to Capt ion Figure 11. Graph showing the isotopic composition of carbon (δ13C) in propane versus that in ethane for the gas samples from the Cumberland overthrust sheet. The vitrinite reflectance (maturity) values of the source rocks were calculated from the difference between isotopic compositions of carbon in propane and ethane in their produced gases by Faber and Stahl (1987). The gas samples in this study are inferred to be from source rocks of similar maturity. Deviations from the maturity trend line were inferred by Whiticar (1994) to indicate mixing of gas sources. Fig ure Ref er to Cap tio n
24 Coal and Petroleum Resources in the Appalachian Basin compositions of the gas samples from all three fields show evidence that a simple maturation trend has been altered, with the Knox Group samples exhibiting the strongest evidence of mixing from more than one isotopic source. Conclusions The oil from the Rose Hill and Ben Hur fields in Lee County, Va., and the Swan Creek field in Hancock and Claiborne Counties, Tenn., are from a marine-algal source identical and appear as a single line (fig. 13). There is a downward deflection toward more negative values for ethane relative to propane. The carbon isotope values for methane in the Knox Group samples are more positive than the values for the Stones River Group samples, which may have been mixed with a small amount of biogenic methane. The range of possible thermal cracking lines for the Knox Group samples (indicated by the shaded area) leads to a much wider range of extrapolated carbon isotope values for kerogen sources, but the range (-22.6 to -33.2 per mil) is still consistent with that found for G. prisca-enriched source rocks by Hatch and others (1987) and Mastalerz and others (2003). The isotopic Figure 12. Graph showing the reciprocal of the component carbon number versus the isotopic composition of carbon (δ13C) in the hydrocarbon gas for four gas samples from Lee County, Va. (Chung and others, 1988). The gas samples from the "Stones River" formations and the Trenton Limestone wells are slightly different and may be mixed with some biogenic methane. The dashed line represents the hypothetical line of isotope values generated by thermal cracking (thermogenic gas). The shaded area shows the range of possible straight cracking lines, given the isotopic range of the inferred source rock. The range of δ13C values found in Ordovician Gloeocapsomorpha prisca-bearing kerogen sources is -22.6 per mil to -33.2 per mil (Hatch and others, 1987; Mastalerz and others, 2003). VPBD, Vienna Peedee belemnite standard. Fig ure Ref er to Cap tio n
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 25 enriched in Assemblage A-type, G. prisca-enriched kerogen. The differences in the oil compositions are probably due to localized differences in the specific marine depositional facies that was enriched in G. prisca. The environments range from a more carbonate-rich environment in the Rose Hill field to an environment with more shale and more mixing of Assemblage A-type kerogens with Assemblage B-type kerogen in the Ben Hur and Swan Creek fields. The associated gases from all three fields have geochemi cal characteristics consistent with those of the oils. The gases are from sources with variable isotopic signatures within the same range found in G. prisca-enriched source rocks elsewhere. Gases from the Cambrian to Lower Ordovician Knox Group in the Swan Creek field are different from gases in the Middle and Upper Ordovician reservoirs in all three fields, which is evidence of a source with a broader isoto pic range. The gas profiles from each reservoir within and between the fields are different, which suggests that the source for the gases may be contained within each reservoir. There is evidence of some biogenic methane in the Ben Hur and Rose Hill gases and in the gases from the Stones River Group rocks in the Swan Creek field. Figure 13. Graph showing the reciprocal of the component carbon number versus the isotopic composition of carbon (δ13C) in the hydrocarbon gas for four gas samples from the Swan Creek field, Hancock and Claiborne Counties, Tenn. (Chung and others, 1988). The graph shows that two samples from the Stones River Group reservoirs form nearly straight lines that are almost parallel to a theoretical thermal cracking line. The six gas samples from Knox Group reservoirs appear as a single line on the graph. The range of δ13C values found in Ordovician Gloeocapsomorpha prisca-bearing kerogen sources is -22.6 per mil to -33.2 per mil (Hatch and others, 1987; Mastalerz and others, 2003). VPDB, Vienna Peedee belemnite standard. Fig ure Ref er to Cap tio n
26 Coal and Petroleum Resources in the Appalachian Basin Acknowledgments This work would not have been possible without the help of the staff at the Abingdon, Va., office of the Division of Gas and Oil of the Virginia Department of Mines, Miner als, and Energy, who provided well data, arranged for access to the wells, and assisted with sampling. Daugherty Petro leum Company kindly allowed us to sample the wells and provided their oil-field technician from Harlan County, Ky., who enthusiastically collected samples from stock tanks and provided invaluable assistance and information. Mike Pribil and Augusta Warden of the U.S. Geological Survey's Organic Geochemistry Laboratory in Denver, Colo., provided analy ses and information. Comments and suggestions from James Coleman, William Dennen, Joseph Hatch, Peter Warwick, and Nancy Stamm greatly improved the manuscript. References Cited Bartlett, C.S., 1989, Trenton Limestone fracture reservoirs in Lee County, southwestern Virginia, in Keith, B.D., ed., The Trenton Group (Upper Ordovician Series) of Eastern North America: American Association of Petroleum Geologists Studies in Geology, v. 29, p. 27-35. Blokker, Peter, van Bergen, Pim, Pancost, Rich, Collinson, M.E., de Leeuw, J.W., and Sinninghe Damste, J.S., 2001, The chemical structure of Gloeocapsomorpha prisca microfossils; implications for their origin: Geochemica et Cosmochimica Acta, v. 65, no. 6, p. 885-900. Butts, Charles, 1927, Fensters in the Cumberland overthrust block in southwestern Virginia: Virginia Geological Survey Bulletin, v. 28, 12 p. Chung, H.M., Gormly, J.R., and Squires, R.M., 1988, Ori gin of gaseous hydrocarbons in subsurface environments; theoretical considerations of carbon isotope distribution: Chemical Geology, v. 71, p. 97-103. Connan, Jacques, 1984, Biodegradation of crude oils in reser voirs, in Brooks, Jim, and Welte, Dietrich, eds., Advances in petroleum geochemistry: New York, N.Y., Academic Press, p. 299-335. Douglas, A.G., Sinninghe Damste, J.S., Fowler, M.G., Eglinton, T.I., and de Leeuw, J.W., 1991, Unique distribu tions of hydrocarbons and sulphur compounds released by flash pyrolysis from the fossilized alga Gloeocapsomor pha prisca, a major constituent in one of four Ordovician kerogens: Geochimica et Cosmochimica Acta, v. 55, no. 1, p. 275-291. Faber, E., and Stahl, W., 1984, Geochemical surface explora tion for hydrocarbons in North Sea: American Association of Petroleum Geologists Bulletin, v. 68, no. 3, p. 363-386. Foster, C.B., Wicander, R., and Reed, J.D., 1990, Gloeocapso morpha prisca Zalessky, 1917, a new study; part II—Origin of kukersite, a new interpretation: Geobios, v. 23, no. 2, p. 133-140. Fowler, M.G., and Douglas, A.G., 1984, Distribution and structure of hydrocarbons in four organic-rich Ordovician rocks: Organic Geochemistry, v. 6, p. 105-114. Fowler, M.G., Stasiuk, L.D., Hearn, Mark, and Obermajer, Mark, 2004, Evidence for Gloeocapsomorpha prisca in Late Devonian source rocks from Southern Alberta, Canada: Organic Geochemistry, v. 35, no. 4, p. 425-441. Gathright, T.M., II, 1981, Lineament and fracture trace analy sis and its application to oil exploration in Lee County: Vir ginia Division of Mineral Resources Publication 28, 40 p. Guthrie, J.M., and Pratt, L.M., 1995, Geochemical character and origin of oils in Ordovician reservoir rock, Illinois and Indiana, USA: American Association of Petroleum Geolo gists Bulletin, v. 79, no. 11, p. 1631-1649. Harris, L.D., and Milici, R.C., 1977, Characteristics of thinskinned style of deformation in the Southern Appalachians, and potential hydrocarbon traps: U.S. Geological Survey Professional Paper 1018, 40 p. Hatch, J.R., Jacobson, S.R., Witzke, B.J., Risatti, J.B., Anders, D.E., Watney, W.L., Newell, K.D., and Vuletich, A.K., 1987, Possible late middle Ordovician organic carbon iso tope excursion; evidence from Ordovician oils and hydro carbon source rocks, mid-continent and east-central United States: American Association of Petroleum Geologists Bulletin, v. 71, no. 11, p. 1342-1354. Hatch, J.R., Risatti, J.N., and King, J.D., 1990, Geochemis try of Illinois Basin oils and hydrocarbon source rocks, in Leighton, M.W., Kolata, D.R., Oltz, D.F., and Eidel, J.J., eds., Interior cratonic basins: American Association of Petroleum Geologists Memoir 51, p. 403-423. Hatcher, R.D., Jr., and Repetski, J.E., 2007, The post-Knox unconformity—Product of global, not regional processes: Geological Society of America Abstracts with Programs, v. 39, no. 2, p. 8. Hatcher R.D., Jr., Williams, R.T., and McCown, M.W., 2001, Swan Creek field—Isolated success or tip of the iceberg?: Oil and Gas Journal, v. 99, no. 40, p. 38-44. Heinemann, K.A., Qing, H., and Bend, S., 2005, Preliminary results of organic matter analyses, Yeoman Formation (Red River), Saskatchewan—Implications for biozones and dolo mitization patterns, in Summary of investigations, 2005, volume 1: Saskatchewan Geological Survey Miscellaneous Report 2005-4.1, p. 1-9. Hughes, W.B., Holba, A.G., and Dzou, L.I.P., 1995, The ratios of dibenzothiophene to phenanthrene and pristane to phy
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 27 tane as indicators of depositional environment and lithology of petroleum source rocks: Geochimica et Cosmochimica Acta, v. 59, no. 17, p. 3581-3598. Hunt, J.M., 1996, Petroleum geochemistry and geology (sec ond edition): New York, N.Y., W.H. Freeman and Company, 743 p. Jacobson, S.R., Hatch, J.R., Teerman, S.C., and Askin, R.A., 1988, Middle Ordovician organic matter assemblages and their effect on Ordovician-derived oils: American Asso ciation of Petroleum Geologists Bulletin, v. 72, no. 9, p. 1090-1100. Jarvie, D.M., 2001, Williston Basin petroleum systems—Infer ences from oil geochemistry and geology: The Mountain Geologist, v. 38, no. 1, p. 19-41. Le Van, D.C., and Rader, E.K., 1983, Relationship of stra tigraphy to occurrences of oil and gas in western Virginia: Virginia Division of Mineral Resources Publication, v. 43, 1 sheet. Lille, U., 2003, Current knowledge on the origin and struc ture of Estonian kukersite kerogen: Oil Shale, v. 20, no. 3, p. 253-263. Longman, M.W., and Palmer, S.E., 1987, Organic geochem istry of mid-continent Middle and Late Ordovician oils: American Association of Petroleum Geologists Bulletin, v. 71, no. 8, p. 938-950. Ludvigson, G.A., Jacobson, S.R., Witzke, B.J., and Gonzalez, L.A., 1996, Carbonate component chemostratigraphy and depositional history of the Ordovician Decorah Formation, Upper Mississippi Valley, in Witzke, B.J., Ludvigson, G.A., and Day, Jed, eds., Paleozoic sequence stratigraphy—Views from the North American craton: Geological Society of America Special Paper 306, p. 67-86. Magoon, L.B., and Dow, W.G., eds., 1994, The petroleum system: American Association of Petroleum Geologists Memoir 60, p. 3-24. Mango, F.D., 1997, The light hydrocarbons in petroleum— A critical review: Organic Geochemistry, v. 26, no. 7-8, p. 417-440. Martin, R.L., Winter, J.C., and Williams, J.A., 1963, Distribu tions of n-paraffins in crude oils and their implications to the origin of petroleum: Nature, v. 199, p. 100-113. Mastalerz, M., Schimmelmann, A., Hower, J.C., Lis, G., Hatch, J., and Jacobson, S.R., 2003, Chemical and isotopic properties of kukersites from Iowa and Estonia: Organic Geochemistry, v. 34, no. 10, p. 1419-1427. Milici, R.C., Ryder, R.T., and Repetski, J.E., 2006, Explora tion for hydrocarbons in the southern Appalachian basin— An overview: Geological Society of America Abstracts with Programs, v. 38, no. 3, p. 10-11. Miller, R.L., and Brosge, W.P., 1954, Geology and oil resources of the Jonesville District, Lee County, Virginia: U.S. Geological Survey Bulletin 990, 240 p. Mitra, Shankar, 1988, Three-dimensional geometry and kine matic evolution of the Pine Mountain thrust system, south ern Appalachians: Geological Society of America Bulletin, v. 100, no. 1, p. 72-95. Nolde, J.E., 1992, Oil and gas well data and geology, Lee County, Virginia: Virginia Division of Mineral Resources Publication 113, 13 p. Obermajer, M., Fowler, M.G., and Snowdon, L.R., 1998, A geochemical characterization and a biomarker re-appraisal of the oil families from southwestern Ontario, Canada: Bulletin of Canadian Petroleum Geology, v. 46, no. 3, p. 350-378. Obermajer, Mark, Osadetz, K.G., Fowler, M.G., and Snowdon, L.R., 2000, Light hydrocarbon (gasoline range) parameter refinement of biomarker-based oil-oil correlation studies— An example from Williston Basin: Organic Geochemistry, v. 31, no. 10, p. 959-976. Osadetz, K.G., Brooks, P.W., and Snowdon, L.R., 1992, Oil families and their sources in Canadian Williston Basin (southeastern Saskatchewan and southwestern Manitoba): Bulletin of Canadian Petroleum Geology, v. 40, no. 3, p. 254-273. Osadetz, K.G., and Snowdon, L.R., 1995, Significant Paleo zoic petroleum source rocks in the Canadian Williston Basin—Their distribution, richness and thermal maturity (southeastern Saskatchewan and southwestern Manitoba): Geological Survey of Canada Bulletin 487, 60 p. Peters, K.E., Walters, C.C., and Moldowan, J.M., 2005, The biomarker guide, volumes 1 and 2 [second edition]: Cambridge, United Kingdom, Cambridge University Press, 1135 p. Pfeil, R.W., and Read, J.F., 1980, Cambrian carbonate plat form margin facies, Shady Dolomite, southwestern Virginia: Journal of Sedimentary Petrology, v. 50, no. 1, p. 91-116. Read, J.F., 1980, Carbonate ramp-to-basin transitions and foreland basin evolution, Middle Ordovician, Virginia Appalachians: American Association of Petroleum Geolo gists Bulletin, v. 64, no. 10, p. 1575-1612. Reed, J.D., Illich, H.A., and Horsfield, B., 1986, Biochemi cal evolutionary significance of Ordovician oils and their sources: Organic Geochemistry, v. 10, no. 1-3, p. 347-358. Rooney, M.A., Claypool, G.E., and Chung, H.M., 1995, Modeling thermogenic gas generation using carbon isotope
28 Coal and Petroleum Resources in the Appalachian Basin ratios of natural gas hydrocarbons: Chemical Geology, v. 126, no. 3-4, p. 219-232. Ryder, R.T., 1996, Appalachian Basin Province (067), in Gautier, D.L., Dolton, G.L., Takahashi, K.I., and Varnes, K.L., eds., 1995, National Assessment of United States Oil and Gas Resources—Results, methodology and supporting data: U.S. Geological Survey Digital Data Series DDS-30, release 2, 1 CD ROM. Ryder, R.T., Repetski, J.E., and Harris, A.G., 1997, Strati graphic framework of Cambrian and Ordovician rocks in the central Appalachian Basin from Campbell County, Ken tucky, to Tazewell County, Virginia: U.S. Geological Survey Miscellaneous Investigation Series Map I-2530, 1 sheet. Ryder, R.T., Burruss, R.C., and Hatch, J.R., 1998, Black shale source rocks and oil generation in the Cambrian and Ordo vician of the Central Appalachian Basin, USA: American Association of Petroleum Geologists Bulletin, v. 82, no. 3, p. 412-441. Ryder, R.T., Harris, D.C., Gerome, Paul, Hainsworth, T.J., Burruss, R.C., Lillis, P.G., Jarvie, D.M., and Pawlewicz, M.J., 2005, Evidence for Cambrian petroleum source rocks in the Rome trough of West Virginia and Kentucky, Appalachian Basin: U.S. Geological Survey OpenFile Report 2005-1443, 79 p., available only online at http://pubs.usgs.gov/of/2005/1443/. (Accessed September 23, 2008.) Sofer, Zvi, 1984, Stable carbon isotope compositions of crude oils—Application to source depositional environments and petroleum alteration: American Association of Petroleum Geologists Bulletin, v. 68, no. 1, p. 31-49. Ten Haven, H.L., 1996, Applications and limitations of Mango's light hydrocarbon parameters in petroleum cor relation studies: Organic Geochemistry, v. 24, no. 10-11, p. 957-976. Thompson, K.F.M., 1983, Classification and thermal history of petroleum based on light hydrocarbons: Geochimica et Cosmochimica Acta, v. 47, no. 2, p. 303-316. Tyson, R.V., 1995, Sedimentary organic matter—Organic facies and palynofacies: London, United Kingdom, Chap man and Hall, 615 p. Virginia Center for Coal and Energy Research, 2008, Histori cal crude oil production (1951-2004): Blacksburg, Va., Virginia Polytechnic Institute and State University, available only online at http://www.energy.vt.edu/vept/petroleum/ crude_historical.asp. (Accessed July 29, 2008; to access, click on Home Petroleum Historical Production.) Wenger, L.M., Davis, C.L., and Isaksen, G.H., 2002, Multiple controls on petroleum biodegradation and impact on oil quality: Society of Petroleum Engineers Reservoir Evalua tion and Engineering, v. 5, no. 5, p. 375-383. Wentworth, C.K., 1921, Russell Fork fault of southwest Virginia, in Giles, A.W., The geology and coal resources of Dickenson County, Virginia: Virginia Geological Survey Bulletin 21, p. 53-67. Whiticar, M.J., 1994, Correlation of natural gases with their sources, in Magoon, L.B., and Dow, W.G., eds., The petro leum system—From source to trap: American Association of Petroleum Geologists Bulletin Memoir 60, p. 261-283.
Appendix A.—Individual Gas Chromatographic Profiles of the Whole Crude Oils Individual gas chromatographic profiles of the whole crude oils are shown in figures A1 through A14. Representative biomarker fragmentograms and an aromatic fraction chromatogram are shown in appendix B, figure B1.
30 Coal and Petroleum Resources in the Appalachian Basin Figure A1. Gas chromatogram for whole crude oil from the No. 108 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A2. Gas chromatogram for whole crude oil from the James Dean No. 1 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A1 - Refer
to aption Figure A2 - Refer
to aption
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 31 Figure A3. Gas chromatogram for whole crude oil from the Yeary No. 1 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A4. Gas chromatogram for whole crude oil from the No. 8709 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A3 - Refer
to aption Figure A4 - Refer
to aption
32 Coal and Petroleum Resources in the Appalachian Basin Figure A5. Gas chromatogram for whole crude oil from the PO-1 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A6. Gas chromatogram for whole crude oil from the Morgan Hensley No. 1 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A5 - Refer
to aption Figure A6 - Refer
to aption
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 33 Figure A7. Gas chromatogram for whole crude oil from the No. 8708 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A8. Gas chromatogram for whole crude oil from the No. 1001 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A7 - Refer
to aption Figure A8 - Refer
to aption
34 Coal and Petroleum Resources in the Appalachian Basin Figure A9. Gas chromatogram for whole crude oil from the No. 1000 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A10. Gas chromatogram for whole crude oil from the Sarah Hensley No. 1 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A9 - Refer
to aption Figure A 10 - Re fer to Caption
Chapter G.12 Geochemistry of Oils and Gases from the Cumberland Overthrust Sheet in Va. and Tenn. 35 Figure A11. Gas chromatogram for whole crude oil from the Paul Reed No. 2 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A12. Gas chromatogram for whole crude oil from the Paul Reed No. 1 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A 11 - Re fer to Caption Figure A 12 - Re fer to Caption
36 Coal and Petroleum Resources in the Appalachian Basin Figure A13. Gas chromatogram for whole crude oil from the Gary Patton No. 1 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A14. Gas chromatogram for whole crude oil from the Stephen Lawson No. 2 well, where n-Cx indicates a normal alkane with x number of carbons. Figure A 13 - Re fer to Caption Figure A 14 - Re fer to Caption
Appendix B.—Representative Biomarker Profiles Representative biomarker fragmentograms and an aromatic fraction chromatogram are shown in figure B1.
38 Coal and Petroleum Resources in the Appalachian Basin Figure B1. Representative biomarker profiles for oil samples from the James Dean No. 1 well in the Ben Hur field, Lee County, Va. Producing formation is Trenton Limestone. Average depth of producing formation is 2,374 ft. A, Terpane mass fragmentogram, scanning range mass-to-charge ratio (m/z) is 191. B, Sterane mass fragmentogram, scanning range m/z is 217. C, Aromatic fraction chromatogram. D, Triaromatic sterane mass fragmentogram, scanning range m/z is 231. E, Monoaromatic sterane mass fragmentogram, scanning range m/z is 253. Figure B1 - Refer to Capti o n