U.S. Geological Survey Open-File Report 03-041
Version 1.0
Robert C. Burruss, Paul G. Lillis, and Timothy S. Collett
U.S. Geological Survey
CONTENTS
Introduction
Regional Geologic Setting
Ellesmerian Sequence
Beaufortian Sequence
Brookian Sequence
Data Sources and Methods
Relationships to Depth, Stratigraphic and Structural Setting
Hydrocarbons
Carbon Dioxide
Nitrogen and Other Gases
Stable Isotope Relationships
Methane
Higher Hydrocarbons
Carbon Dioxide
Implications for Gases in NPRA
Acknowledgments
References
Table1 | 2A
| 2B
Figure Captions (These are separate
HTML files; download all figures with captions in a 622-KB
PDF file)
Download free Acrobat Reader software.
Visit the Adobe accessibility website.
The North Slope of Alaska contains a significant volume of natural gas. The known amount of gas within the oil and gas fields is large and the estimated amount of undiscovered resource is even larger (Bird, 2002a) . Increasing demand for gas as a relatively clean fossil fuel is driving plans to develop this resource either through a pipeline to the lower 48 states, LNG shipments, or gas-to-liquids technologies.
To increase confidence in U. S. Geological Survey (USGS) estimates of undiscovered gas resources there is a need to understand the role of natural gas in the petroleum systems of the North Slope. Of particular interest to resource estimates in the National Petroleum Reserve in Alaska (NPRA) is whether the petroleum systems of that part of the North Slope are gas-rich. Current exploration targets are oil reservoirs, but the historic drilling experience indicates that gas is abundant in the western part of the North Slope within NPRA.
Initial drilling by the U.S. Navy in 1944 to 1953 tested gas in a number of wells and discovered small accumulations at Barrow, Meade, Square Lake, and Wolf Creek and adjacent to NPRA at Gubik as shown on the North Slope location map in Figure 1. Of the three oil accumulations discovered by the Navy, the largest, Umiat, tested gas in a deeper reservoir horizon than the oil accumulation. Later drilling by the U.S. Navy and the USGS in 1973 to 1981 discovered gas accumulations on the Barrow Peninsula at Walakpa and Sikulik, and tested gas from a number of potential reservoirs at other locations, some at depths greater than 10,000 ft.
This paper addresses several natural gas resource questions based on a summary of the geochemistry of natural gas across the North Slope of Alaska. First, is there evidence of distinct gas-rich petroleum systems? Second, is there any evidence of unique gas systems in the foothills of the Brooks Range? Third, are there any potential problems with the quality of gas (significant quantities of non-hydrocarbon gases) in undiscovered gas accumulations?
To address these questions we present evidence for the source and maturity of the known natural occurrences. In particular we will examine the effects of microbial gas generation, microbial alteration of thermogenic gas, and the role of mixing of gases of various origins in controlling the present day composition of gas. Finally, we will examine the extent to which non-hydrocarbon gases, carbon dioxide and nitrogen, affect the quality of North Slope gas. Carbon dioxide is particularly important because a large fraction (5 to 18%) of the gas cap of the Prudhoe Bay field and surrounding satellite fields is CO2 and may amount to as much as 5 trillion cubic feet (TCF) of gas. Although this amount of CO2 is beneficial for current practices of miscible flooding for enhanced oil recovery in Prudhoe Bay, marketable natural gas must contain 4 % or less CO2. Therefore, gas from the Prudhoe Bay field gas cap will require processing or dilution with low CO2 gas before it is marketable. Costs associated with gas processing or dilution may impact the economics of natural development on the North Slope.
The petroleum geology of the North Slope is described in detail in (Bird, 2002b) . Figure 1 depicts the locations of a limited number of fields sampled by the USGS or for which data are available. More detailed summaries of the fields referred to in this study can be found in (Kumar, Bird, and others, 2002) . The stratigraphic setting of the oil and gas fields for which we have data is summarized in Figure 2. Essentially all stratigraphic intervals contain hydrocarbon reservoirs. These intervals range in age from the Mississippian Endicott Group, which overlies pre-Mississippian basement, to the early Tertiary age Sagavanirktok Formation.
Sediments of the Ellesmerian sequence were deposited on a passive continental margin with most units displaying both deep and shallow water facies (Hubbard, Edrich, and Rattey, 1987) . The reservoirs in the Paleozoic and lower Mesozoic Ellesmerian strata are known to be charged with oil, dominantly derived from the organic matter rich Shublik Formation. Ellesmerian traps charged with gas are present in the Kemik and Kavik gas fields in the eastern foothills of the Brooks Range. The Lisburne No. 1 well in the central foothills tested small amounts of gas in porosity associated with thrust faulted Lisburne Formation. The Ellesmerian reservoirs closest to the Brooks Range are affected by deep burial and imbricate thrust faults, which should result in high levels of thermal maturity.
The Beaufortian sequence contains sediments derived from uplifts related to initial rifting of the Ellesmerian margin. Ultimately rifting led to opening and spreading of the Arctic Ocean basin during the Cretaceous. These strata contain a number of sedimentary sequences related to changing sea level during rifting. Beaufortian reservoirs are present in the Alpine oil accumulation, and in small gas accumulations at Walakpa and the Barrow fields in the Avak impact structure. The Jurassic age Kingak Shale is an important source rock. The sequence is terminated by a regional unconformity known as the Lower Cretaceous Unconformity (LCU) that is visible in seismic sections and in outcrop. The origin of the LCU is not fully understood. Possible origins include erosion on a rift margin bulge or erosion on the Ellesmerian margin during collision with ancestral Brooks Range terrains.
Sediments shed from the rising Brooks Range in the south were deposited on the Beaufortian Sequence and filled the Colville basin of the North Slope. The shales deposited on the LCU are commonly rich in organic matter. Three stratigraphic units, the Hue Shale, the informally named pebble shale unit, and the informally named highly radioactive zone (HRZ, also known as the GRZ, for gamma ray zone) are important Lower Cretaceous source rocks. The organic matter in these rocks is dominantly of terrestrial origin (Type III) making them important sources for natural gas and petroleum with high gas-oil ratios (GOR).
Analytical measurements of the molecular and isotopic composition of natural gas come from several different sources. Compositions of gases tested during the U.S. Navy drilling program in 1944 to 1951 are reported in USGS Professional Paper 305 (Collins, 1958, 1959, 1961; Robinson, 1956, 1958a, 1958b, 1959a, 1959b, 1964; Robinson and Collins, 1959) . Composition and stable isotope measurements made during the U.S. Navy and USGS drilling program of 1973 to 1981 are reported in USGS Professional Paper 1399 (Claypool and Magoon, 1988; Kharaka and Carothers, 1988) and digital geochemical information is available as a USGS Digital Data Series (Threlkeld, Obuch, and Gunther, 2000) . A small number of analyses for Prudhoe Bay are reported by Chung (Chung, Gormly, and Squires, 1988) . A limited amount of industry data was released by British Petroleum and ExxonMobil to the U.S. Geological Survey (written communication to T. S. Collett, 1990). Recently, data from gas accumulations in the greater Prudhoe Bay area and solution gas from oil fields adjacent to the northeastern boundary of NPRA became available as a result of a Ph.D. dissertation (Masterson, 2001) .
During field work in 1998 we collected a gas sample from the Umiat No. 8 well (Umiat oil field) that was drilled in 1951 (Kumar, Bird, and others, 2002) . Following that opportunity we carefully reviewed all the records of testing of the wells drilled by the U.S. Navy in 1944 to 1951. We identified a number of wells where gas was tested and had the potential remained for new sampling at the wellhead. In 1999 we began a collaborative effort with the U.S. Bureau of Land Management to examine these wells and sample gas, where possible. These newly sampled wells all had some evidence that gas was leaking from wellhead fittings. This provided some assurance that the gas sampled was relatively fresh and not simply a static sample within the casing or production tubing. In 2001, with permission and assistance from the operating company, Upeagvik Arctic Slope, we sampled wells in the South Barrow and East Barrow gas fields. Finally, the last set of gas analyses that are relevant to this study are the result of a series of hydrous pyrolysis experiments on identified petroleum source rocks from the North Slope of Alaska. The original experiments focused on the laboratory simulation of oil generation and reported molecular gas compositions (Lillis, Lewan, and others, 1999) . The unpublished stable isotopic analyses of the gases are reported herein.
Two types of geochemical data are commonly used to interpret the origin and fate of natural gas. The most common is molecular composition, measured by gas chromatography or mass spectrometry and reported in the equivalent units of mole percent or volume percent. The prinicipal components measured are methane, ethane, propane, iso-butane, normal-butane, nitrogen, carbon dioxide, and hydrogen sulfide. The trace components are higher hydrocarbons, iso- and normal-pentane and “hexanes-plus” (hydrocarbons with 6 carbons and greater) and non-hydrocarbons such as oxygen (generally from air contamination during sampling), argon, helium, and hydrogen. In many cases, the trace components are below detection limits and results are not reported.
The second type of analyses are stable isotopic compositions of carbon and hydrogen. By convention, these results are reported in parts-per-thousand (per mil) relative to a standard. For carbon the standard isotopic composition is PDB (the PeeDee belemnite) and for hydrogen it is VSMOW (Vienna Standard Mean Ocean Water). The isotopic ratios are measured by ratio mass spectrometry and reported relative to the respective standards as follows:
 |
Where x = molecule of interest and Rx = (13C/12C)x or (D/H)x and Rstd is the ratio in the standard.
Carbon and hydrogen isotopic compositions can be measured on methane, but these analyses are only available on recently acquired samples. Carbon isotopic compositions of individual hydrocarbons through normal-pentane are available on a few samples, with many more samples having measurements on methane, ethane, and propane. The carbon isotopic composition of carbon dioxide is also measured, although it is not available on all samples.
The analytical results from natural gases used in this report are from samples of flowing gas, either as drill stem tests (DSTs), production tests, or samples from commercial production. Measurements on headspace gases of cuttings collected during drilling are not included. The available data consists of 113 samples from approximately 72 wells in 19 fields, 16 of which have some stable isotope measurements (Table 1). In addition, we have data from 44 samples in 15 wells outside of any known field, 2 of which have stable isotope measurements. Although this is a relatively large database of gas analyses that represents good stratigraphic coverage of gas occurrences (Figure 2), the data are spread over a very large area of the North Slope (Figure 1). Our sample coverage in areas of greatest interest, especially the Brooks Range foothills, is very limited.
The analytical measurements were made with techniques that were standard at the time of the analysis. The oldest measurements are only of molecular composition and were made by mass spectrometry. Measurements made from the 1970s to present may include carbon stable isotope measurements on methane and CO2 by isotope ratio mass spectrometry. The most current measurements include hydrogen isotope measurements on methane and carbon isotope measurements on higher hydrocarbons up to iso-pentane. Isotopic measurements on some components may be missing because the concentration was too low to permit routine measurement of the stable isotope ratios. All measurements on samples collected by the USGS from 1998 to present were made by Isotech Laboratories, Inc. This set of modern data presented in Tables 2A and 2B.
The basic approach to interpretation of the analytical results is to present graphical displays of composition versus depth and stratigraphic sequence (color coded). These are examined for evidence of unique characteristics related to known petroleum systems. Additional bivariate graphics of gas properties that previous publications have demonstrated to be useful for identifying gas sources, mixing, and alteration are used to evaluate these aspects of North Slope gases.
The molecular composition of the hydrocarbons in natural gas varies as a function of source and thermal maturity, although the range of variations commonly overlaps between source and maturity effects. A compositional variable used routinely to evaluate gases is gas wetness or inversely, dryness. This refers to the fraction of ethane (C2H6, referred to herein as C2) and higher hydrocarbons in the gas (C2+). Greater amounts of C2+ hydrocarbons occur in gases that are associated with oils and condensates and are “wetter” in the sense of greater potential for liquid hydrocarbons. Conversely, as the fraction of methane increases the gas ultimately becomes “non-associated” with oil and may be either derived from microbial activity or very high levels of thermal maturation of source rocks or cracking of preexisting oil. In this discussion we will use the term gas dryness, defined as:
Dryness = C1/C2+
Many other mathematical definitions of gas “wetness” or “dryness” are in use, e.g. (Bernard, Brooks, and Sackett, 1976; Hunt, 1996) , but the ratio as we define it is useful because it is consistent with prior practice in the interpretation of stable isotope measurements (Bernard, Brooks, and Sackett, 1976) .
The dryness of gases as a function of reservoir depth and stratigraphic sequence is shown in Figures 3A, B, C. Values less than 100 indicate thermogenic gases that are distinct from microbial gases and values less than 30 indicate gases that are associated with condensate and oil accumulations. Gas tests within fields show general trends of decreasing dryness (increasing wetness) with increasing depth. Gas tests for individual wells outside known fields show much greater scatter with depth. In general, the individual well tests from zones within Brookian and Ellesmerian strata are very dry, non-associated gas.
Ploting all the data for gases from known accumulations in a single figure (Figure 4A), reveals patterns that are consistent with current knowledge of the processes of oil and gas generation, migration, and alteration. The dashed curve labeled “Qualitative oil and gas generation trend” shows the trend of gas compositions with depth in sediments during oil and gas generation (see for example (Hunt, 1996) , Figure 7-1, p. 187). Initially, shallow microbial gas generation produces a very dry, non-associated gas that becomes progressively wetter through mixing as more deeply buried sediments generate large volumes of higher molecular weigh hydrocarbons through thermal cracking. At intermediate depths of 7,000 to 10,000 feet (2.2 to 3.1 km) oil generation dominates and thermogenic, associated gases are produced. At much greater depths, super-mature gases are produced through cracking of residual kerogens or cracking of previously generated oil. These gases become progressively drier through the breakdown of higher molecular weight hydrocarbons to methane. Vertical arrows pointing toward the gas accumulations at Walakpa and Barrow and at the foothills accumulations at Kemik and Kavik are provided as indicators of possible upward migration of more deeply generated gases (Figure 4B).
The dryness of gases from individual wells plotted with the data from gas accumulations (Figure 4B) is consistent with the trends for the accumulations. The main difference is that many more gases are dry at intermediate depths, suggesting that these gases are migrated from more deeply buried sources. Four deep tests, Oumalik Test Well No. 1, Awuna No. 1, Lisburne No. 1, and Tulugak No. 1, in structures related to Brookian deformation are labeled. Of these, the most northerly, Oumalik, is the wettest and the most southerly, and deepest, Lisburne and Tulugak, are the driest. This is the only indication that there may be a unique, “foothills” gas type, that is very dry.
Carbon dioxide is the most abundant non-hydrocarbon gas that occurs in gas accumulations on the North Slope (Figure 5A). The highest concentrations (47.75 mole %) are reported in DST #1 in the Tern A-1 well of the Liberty Unit in the greater Prudhoe Bay area, although concentrations as high as 42.9 mole % are reported in the main Prudhoe Bay field (Kharaka and Carothers, 1988) . There is a clear trend of increasing CO2 concentration with depth, which, taken to a logical extension suggests that any gas accumulation deeper than about 12,000 feet should contain 100 % CO2 (Figure 5A). This extrapolation falls apart when gases from individual wells are added to the plot because the trend to greater depth does not continue (Figure 5B). The deepest gas sample in the Lisburne No. 1 well contains only a “trace” of CO2 (assumed to be less than 0.01 mole %) and the deepest gas sample in the foothills, Tulugak No. 1 well, contains only 0.4 mole %).
In general, nitrogen contents of North Slope gases are low, with most values less than 4 mole % as shown in Figure 6. The only analyses from gas accumulations that are higher than this are a few points from the Barrow fields and Walakpa. The Walakpa measurements are known to be too high due to air contamination (Exxon, written communication to T. S. Collett. 1990). The obvious exceptions to the general range of values are the three tests in the South Simpson No. 1 well with more than 70% nitrogen. The origin of these high nitrogen contents is unknown, however, they appear to be a very localized phenomenon.
The nitrogen contents do not show any obvious trends with depth or stratigraphic sequence. This may be partially influenced by incomplete data sets. None of the analyses reported by Masterson (Masterson, 2001) include nitrogen, so we are missing points from Beaufortian reservoirs that are at intermediate depths. Also, the data released by industry for the Endicott field does not include nitrogen.
The only other gas of importance is hydrogen sulfide. The oil accumulations of the greater Prudhoe Bay and Kuparuk River fields are relatively high in sulfur, containing an average of 1.0 and 1.6 weight %, respectively (Masterson, 2001) . Also, there is evidence that the deep water facies of the Lisburne Formation, a thick carbonate section, generated a unique type of oil on the North Slope (Lillis and Magoon, 2002) , although it’s contribution to the accumulations at Prudhoe and Kuparuk appears to be minor (Masterson, 2001) . Worldwide, petroleum source rocks related to carbonate-rich sedimentary sequences produce high sulfur oils and gases containing H2S (Hunt, 1996) . Remarkably, no analyses of gases on the North Slope report H2S. The only possible significant H2S occurrence may be related to the occurrence of molten sulfur at total depth in the Inigok No. 1 well (Hunt, 1996) .
The carbon stable isotope composition of methane is shown as a function of depth (Figure 7). A trend of decreasing depletion in 13C with increasing depth is apparent. This trend is consistent with observations in many petroleum basins (Hunt, 1996) and reflects generation of gases at progressively higher levels of thermal maturity. The cluster of gases with isotopic compositions between –37 and –45 %o PDB at depths of about 2000 to 3500 feet may be due to migration of gases from depth as suggested by the vertical arrows in the figure. The points for the Alpine Field and the Liberty Unit of the Prudhoe Bay Field are isotopically lighter than expected for the depth of the reservoirs. The methane in the Alpine field appears to be microbially generated (Masterson, 2001) . The methane in the Liberty Unit may contain a significant fraction of microbial methane.
Comparison of variations in the dryness paramater, C1/C2+, with the variations in the stable isotopic composition of methane, provides evidence for the source and alteration of gases on the North Slope. There is a clear distinction between the microbially generated gases in the Simpson Field and thermogenic gases in almost all other accumulations on the North Slope (Figure 8). Three hypothetical mixing lines between different microbial endmembers and and thermogenic gas endmembers are shown on Figure 8. Each point on the curves represents a 20 % (mole basis) increment of mixing. If these mixing lines connect reasonable end-member gases, then the isotopically light methane in the Alpine, Tarn, West Sak, 2 Barrow field tests, and the Liberty Unit require mixing of 10 to 50% microbial gas with thermogenic gas to generate the observed compositions.
Gas samples from Walakpa, South Barrow, East Barrow, and Wolf Creek have relatively dry compositions compared to their thermogenic carbon stable isotopic compositions. This range of dryness in gases from shallow reservoirs is consistent with compositional fractionation due to gas hydrate formation. Gas hydrates preferentially remove heavier hydrocarbons from natural gas (Sloan, 1998) and the trend is observed in samples directly related to natural occurrences of gas hydrates in the MacKenzie Delta, Canada (Lorenson, Whiticar, and others, 1999) .
The hydrogen isotopic composition and carbon isotopic composition of methane provide an additional test of the thermal maturity and source of methane. These parameters are plotted in Figure 9 with fields of various gas types documented by (Schoell, 1983) . Many of the analyses do not include hydrogen isotopic composition, so our ability of compare variations among samples is limited. Gas from Alpine field plots with microbial gas from the Simpson field, although based on C1/C2+ (Figure 8), gas from Alpine should contain only 20 to 40 % microbial gas. Gas from the Tarn field appears to be a mixture of microbial and thermogenic gas, consistent with the dryness of the gas shown in Figure 8. All the other gases plot in the fields for gases associated with oil or condensate production. The Barrow and Walakpa gas fields produce small quantities of condensate (Holba, Ellis, and others, 2000; Magoon and Claypool, 1988) consistent with the hydrogen isotope evidence. With the exception of the Gubik and Wolf Creek gases, all the other samples are from the gas cap at Prudhoe Bay or gas in solution in oil in the reservoir. Oil staining was observed in the Gubik and Wolf Creek cuttings and core (Collins, 1959; Robinson, 1958a) and the hydrogen isotope data implies that these gases may have migrated from larger volumes of oil. Increasing levels of thermal maturity of the gas source cause both the carbon and hydrogen isotopic composition of methane to become heavier (less negative, or less depleted in 13C). On this basis the most thermally mature gases occur in the Walakpa field and the Lisburne Pool of the Prudhoe Bay field.
The carbon isotopic composition of the higher hydrocarbons, ethane through pentane, is related to the isotopic composition of the source organic matter, the thermal maturity of the source rock, and post-generation alteration of the gas (Chung, Gormly, and Squires, 1988; Prinzhofer and Huc, 1995; Whiticar, 1994) . One method of displaying these relationships is the “natural gas plot” developed by Chung (Chung, Gormly, and Squires, 1988) . Examples using North Slope gas analyses are shown in Figure 10. In these plots, the carbon isotopic composition of the individual hydrocarbons is plotted as a function of the number of carbon atoms (Cn) in the hydrocarbon. As discussed by Chung (Chung, Gormly, and Squires, 1988) the kinetics of cracking alkyl side chains from high molecular weight source organic matter favors breaking 12C – 12C bonds, resulting in enrichment of the product gas molecules in 12C. Therefore, C1, methane, is most depleted (lower d13C values) in 13C, and the progressively higher carbon number alkanes are less and less depleted because of the increasing probability of including a 13C atom in the cracking product. The theoretical arguments presented by Chung (Chung, Gormly, and Squires, 1988) , indicate that gases generated by cracking of source kerogens should plot on a straight line when the carbon number is plotted as its inverse. Furthermore, extrapolation of the line to high carbon number (1/ Cn approaches 0 as Cn gets large), should approximate the carbon isotopic number of the parent kerogen. This is shown in Figure 10A with gas analyses for the Walakpa, Wolf Creek, and Umiat fields compared to the isotopic compositions of North Slope kerogens (Burwood, Cole, and others, 1985; Schoell, Wehner, and Coleman, 1985) . Extrapolation (as indicated by the dashed green arrow) of most of the gas analyses intersect with isotopic compositions of organic matter in the upper Jurassic Kingak Shale, and the Lower Cretaceous Torok Formation or the pebble shale unit of the Hue Shale.
Isotopic compositions of higher hydrocarbons in gases from the South and East Barrow gas fields show a wide range of patterns. Gases from two wells, S. Barrow No. 13 and E. Barrow No. 19, plot as straight lines (Figure 10B) that extrapolate to isotopically lighter values than Walakpa and Umiat and appear to be consistent with generation from organic matter in the Shublik Formation (see location of Shublik on Figure 10A). The most striking feature of the Barrow data is the fact that most analyses do not fall on straight lines. Either ethane or propane is isotopically heavier than expected from the straight line plots. This type of variation occurs in gases that are affected by microbial alteration (oxidation) of the gas (James and Burns, 1984; Pallaser, 2000) .
A plot of gases from the Prudhoe Bay area in Figure 10C shows evidence of microbial alteration of ethane and propane as originally noted by Chung (Chung, Gormly, and Squires, 1988) . The least altered gases appear to be those from the western part of the field area (West End Test #1, and Prudhoe, Eileen) and the offshore exploration well, Mukluk No.1. Extrapolation of the isotopic compositions of these gases gives an isotopic composition of source kerogen between the values for the Shublik Formation and the Beaufortian and Brookian sources shown in Figure 10A. This suggests mixing of gases from several sources, a process that is well documented for the oils in the Prudhoe Bay Field (Masterson, Dzou, and others, 2001) and should be expected for gases, as well. The greatest amount of microbial alteration appears to have affected gases from the Mississippian Kekiktuk Formation reservoir of the Endicott field (Sag Delta #8 and #10) and the Liberty Unit in the eastern part of the greater Prudhoe area.
Similar alteration patterns are shown by gases from the Prudhoe Bay gas cap and other gases reported by (Masterson, 2001) , and displayed in Figure 10D. The higher hydrocarbons that show the least evidence of alteration occur in the western-most fields in the Beaufortian reservoirs at Alpine and Kalubik and in the similar but stratigraphically higher reservoir at Tarn. The least altered gas related to the Prudhoe Bay area is the solution gas recovered from the Lisburne pool. This gas is relatively isotopically heavy, possibly due to mixing with gases from Beaufortian or Brookian sources, but more likely due to higher levels of thermal maturity, consistent with the hydrogen and carbon isotopic composition of methane (Figure 9).
Gases generated by laboratory simulation of thermal maturation of source rocks by hydrous pyrolysis (Lewan, 1997) can be compared with field data to look for evidence of variations due to organic matter type and thermal maturity of source rocks. Isotopic analyses of gases from hydrous pyrolysis experiments on one sample of Shublik Formation (Shublik 31) and 4 samples of Hue Shale (Hue 34, 35, 38, and 442) are shown in Figure 10E. Experiments on potential source rocks from Kingak Formation did not yield enough gas for stable isotopic measurements (Lillis, Lewan, and others, 1999) . As inferred from the comparisons shown in Figures 10A and 10B, the gases generated by hydrous pyrolysis from Shublik source rocks are isotopically lighter than those generated from Lower Cretaceous sources in the Brookian sequence. It is clear, however, that extrapolation of the experimental results yields estimates of the isotopic composition of the source that are lighter than published measurements. Carbon isotope compositions of kerogens used in the hydrous pyrolysis experiments were not measured so we cannot directly compare these results with previously published kerogen and gas analyses.
Like the hydrocarbons, the carbon isotopic composition of CO2 is controlled by the source of CO2 (thermogenic cracking of organic matter, thermal oxidation of organic matter, microbial processes, or thermal breakdown of carbonate minerals) and mixing between sources. Thermally driven reactions of organic matter tend to yield isotopically light CO2, similar in isotopic composition to the source. Thermal metamorphism of carbonates yields isotopically heavy carbon dioxide. Microbial processes can yield either heavy or light CO2, depending on the dominant microbial activity such as anaerobic reduction of CO2 to methane or aerobic oxidation of hydrocarbons to CO2 (Whiticar, 1999) .
Comparison of the CO2 content of North Slope gases to their carbon isotopic compositions shows several distinct groups of gases (Figure 11). The CO2 in the solution gas from West Sak field is considered by Masterson (2001), to originate from anaerobic reduction of CO2 to methane, yielding isotopically light methane (Figures 4 and 7) and a residual, heavy CO2. The similarity of the compositions of CO2 in 3 South Barrow gases suggests that this mechanism may control CO2 in this field also. The Walakpa gases with very low CO2 content are difficult to interpret because of uncertainty in the reliability of these isotopic compositions. The main group of analyses in an elongate cluster suggests a possible mixing trends. Based on a smaller number of analyses, Masterson, 2001, suggested that this range of compositions is due to mixing between a thermogenic gas with a possible origin in the Shublik Formation and a CO2-rich endmember derived from low grade metamorphic reactions in Lisburne Group carbonates. To test this idea we show a mixing line drawn for end-members containing 50% CO2 with 0 per mil d13C and 0.1% CO2 and –24 per mil d13C (Figure 11). Although this line spans the data, the end-members are arbitrary and there is no reason to define the metamorphic endmember as any other concentration other than 100% CO2. If we do this, it will be very difficult to define a mixing model that approximates the observations. Furthermore, mixing does not explain the approximate 10 to 20 per mil range of isotopic composition for any CO2 concentration shown (Figure 11). This range of isotopic compositions suggests that other processes or sources of CO2 are affecting the samples. Furthermore, the fact that gas tested in the Lisburne formation in the Lisburne No. 1 well has a very low (“trace”) CO2 content (Figure 5B), implyies that deeply buried Lisburne Formation is not universally a source of high-CO2 gas.
The evidence that the isotopic composition of ethane and propane in most gases was affected by microbial oxidation (Figure 10) suggests that there should be some correlation between the extent of hydrocarbon alteration and CO2 content. In Figure 12, the extent of alteration is shown as difference in carbon isotopic composition between ethane and propane which is plotted against percent CO2. Where ethane has been most extensively altered and isotopically heaviest, for example, in the Prudhoe Bay gas cap, the difference between ethane and propane is positive. Where propane is preferentially altered, as in the Barrow field, the difference is large and negative. For normal, unaltered gases, the difference is negative with a magnitude of 1 or 2 per mil. Both trends can be seen in Figure 12. This suggests that some fraction of the CO2 content of the gases of the greater Prudhoe Bay area are derived by microbial processes and not solely from an external, metamorphic source.
The interpretations of the gas analyses discussed in this report are preliminary. Not all samples have a complete molecular analysis nor do many have complete isotopic analyses. In general, however, the data are consistent with current empirical and theoretical concepts of natural gas generation, migration, and alteration. Therefore we believe we can draw reasonably consistent conclusions from the results presented in the previous section.
The variation in molecular composition as a function of depth (Figures 3 and 4) suggests that there is the potential for a Foothills gas type. This gas is very dry, non-associated, and based on the deepest tests in the Lisburne and Tulugak wells (Figure 4B), may have low CO2 concentrations. However, this conclusion is based on very little data and has no isotopic measurements to constrain the source or maturity of the gas. Furthermore, evidence for composition of gases generated at great depth during thrusting in the eastern Brooks Range (Parris and Burruss, 2001; Parris, Burruss, and O'Sullivan, 2002, in review) indicates that deep reservoirs could have as much as 20% CO2. These limited observations from gases within the Foothills of the Brooks Range do not support a distinct gas-rich petroleum system in NPRA.
Gases from Beaufortian reservoirs adjacent to NPRA at Walakpa on the west and Alpine on the east offer some insight to the possible gas charge to Beaufortian plays within NPRA. These plays are important to current exploration strategies within NPRA (Houseknecht, 2002) . Gases in Walakpa are the most thermally mature gases west of the greater Prudhoe Bay field area. They appear to be generated from sources within Beaufortian or Brookian sedimentary sequences. This contrasts with the solution gas in the Alpine reservoir that contains some fraction of biogenic methane. The isotopic composition of the heavier hydrocarbons at Alpine suggests that they are derived from organic matter within the Shublik Formation (Masterson, 2001) and Figure 10D. This contrast in source and thermal maturity of gases within essentially the same stratigraphic horizon suggests that mixed oil and gas accumulations may be discovered as exploration proceeds west along the trend of the Beaufortian shelf edge. The possible effects of microbial alteration of hydrocarbons may be limited. The alteration observed in the stratigraphically deeper Barrow fields is undoubtably due to the present day shallow burial depth of these reservoirs. These reservoirs contain gas from Shublik Formation source rocks, making the charge to these reservoirs much more similar to the charge to the reservoirs at Prudhoe Bay than that to the Beaufortian reservoirs in the Walakpa field.
One of the most surprising facts about the proven gas reserves on the North Slope, the gases of the greater Prudhoe Bay and Kuparuk River fields, is that essentially all the known gas is affected by microbial alteration. The main exceptions appear to be gases in the west end of the Prudhoe Bay complex, in the area of the Eileen fault and the thermally mature solution gases in the Lisburne pool. The extent of microbial alteration was described by (Burruss and Collett, 2000) and newly available data (Masterson, 2001) confirms this fact.
The gases that are the most extensively altered by microbial processes are also the gases that have the highest CO2 contents. Masterson (Masterson, 2001) proposed that the gases with high concentrations of CO2 are mixtures of a normal, thermogenic natural gas with a CO2-rich endmember derived from thermal breakdown of carbonates in the Lisburne Group. This hypothesis is questionable because the only gases tested by drilling that have high CO2 contents are the gases of the greater Prudhoe Bay area, no other gases approach these values. Whatever the source of CO2, high CO2 gases appear to be restricted to this area. Thus, there is no compelling evidence to suggest that gases with high CO2 contents will be found in NPRA.
We want to acknowledge the field support and enthusiasm of Don Meares, U.S. Bureau of Land Management, Northern Field Office, Fairbanks, Alaska, for visits to the historic NPRA well sites and the opportunity to obtain modern samples of gas from wells drilled by the US Navy in the 1940’s and 1950’s. George Finlayson and Rich Evans, Ukpeagvik Arctic Slope, Barrow, Alaska, provided the opportunity and assistance to sample wells in the South and East Barrow gas fields. Kristin Dennen, USGS, Reston, VA, provided creative ideas and assistance with displaying data. Finally we must acknowledge the enthusiasm and support of the USGS project chief for North Alaska petroleum studies, Kenneth Bird. Ken made this work possible. Constructive reviews by Robert Ryder and Peter Warwick improved the presentation. Finally, any mention of commercial product names, trade names, or contract analytical facilities does not constitute endorsement by the U.S. Geological Survey, the Department of Interior, or the Federal Government.
Bernard, B.B., Brooks, J.M., and Sackett, W.M., 1976, Natural gas seepage in the Gulf of Mexico: Earth and Planetary Science Letters, v. 31, p. 48-54.
Bird, K.J., 2002a, Estimates of undiscovered natural gas volumes, onshore North Slope: A retrospective (abs). in AAPG-SPE Joint Technical Conference, Anchorage, Alaska, AAPG, p. 62.
2002b, Geologic Setting of NPRA, in Bird, K.J., and Nelson, P.E., eds., Petroleum Resource Assessment of the National Petroleum Reserve in Alaska, U.S. Geological Survey, p. GS1- GS41.
Burruss, R.C., and Collett, T.S., 2000, Stable isotope geochemistry of natural gas, North Slope, Alaska: Evidence for multiple sources, mixing, and alteration: Bulletin American Association of Petroleum Geologists, v. 83, no. 13, p. 91.
Burwood, R., Cole, G.A., Drozd, R.J., Halpern, H.I., Penfield, I.E., and Sedivy, R.A., 1985, Carbon isotopic characterization of some North Alaska petroleums and potential source rock kerogen assemblages, in Claypool, G.E., ed., Alaska North Slope Oil/Rock Correlation Study: Tulsa, American Association of Petroleum Geologists.
Chung, H.M., Gormly, J.R., and Squires, R.M., 1988, Origin of gaseous hydrocarbons in subsurface environments: Theoretical considerations of carbon isotope distribution: Chemical Geology, v. 71, p. 97-103.
Claypool, G.E., and Magoon, L.B., 1988, Oil and gas source rocks in the National Petroleum Reserve in Alaska, in Gryc, G., ed., Geology and Exploration of the National Petroleum Reserve in Alaska, 1974 to 1982: Washington, D.C., United States Government Printing Office, p. 451-482.
Collins, F.R., 1958, Test wells, Umiat area, Alaska: Geological Survey Professional Paper, v. 305-B, p. 71-206.
1959, Test wells, Square Lake and Wolf Creek areas, Alaska: Geological Survey Professional Paper, v. 305-H, p. 423-484.
1961, Core tests and Test wells, Barrow area, Alaska: Geological Survey Professional Paper, v. 305-K, p. 569-644.
Holba, A.G., Ellis, L., Decker, P., Wilson, G., Watts, A., and Masterson, D., 2000, Organic geochemistry of gas fields flanking the Avak impact structure, western NPRA, North Slope, Alaska (abs.), in Alaska Geological Society & Geophysical Society of Alaska 2000 Science and Technology Conference, Anchorage, Alaska, p. unpaginated.
Houseknecht, D., 2002, Beaufortian Stratigraphic Plays, in Bird, K.J., and Nelson, P.E., eds., Petroleum Resource Assessment of the National Petroleum Reserve in Alaska, U. S. Geological Survey, p. BF1-BF33.
Hubbard, R.J., Edrich, S.P., and Rattey, R.P., 1987, Geologic evolution and hydrocarbon habitat of the 'Arctic Alaska Microplate, in Tailleur, I., and Weimer, P., eds., Alaska North Slope Geology: Bakersfield, The Pacific Section, Society of Economic Paleontologists and Mineralogists, p. 797-830.
Hunt, J.M., 1996, Petroleum Geochemistry and Geology (2nd ed.): New York, W. H. Freeman and Company, 743 p.
James, A.T., and Burns, B.J., 1984, Microbial alteration of subsurface natural gas accumulations: American Association of Petroleum Geologists Bulletin, v. 68, no. 8, p. 957-960.
Kharaka, Y.K., and Carothers, W.W., 1988, Geochemistry of oil-field water from the North Slope, in Gryc, G., ed., Geology and Exploration in the National Petroleum Reserve in Alaska, 1974 to 1982: Washington, D.C., United States Government Printing Office, p. 551-562.
Kumar, N., Bird, K.J., Nelson, P.H., Grow, J.A., and Evans, K.R., 2002, A digital atlas of hydrocarbon accumulations within and adjacent to the National Petroleum Reserve-Alaska (NPRA): U.S. Geological Survey Open-file Report 02-71, p. 70 pp.
Lewan, M.D., 1997, Experiments on the role of water in petroleum formation: Geochimica et Cosmochimica Acta, v. 61, p. 3671-3723.
Lillis, P.G., Lewan, M.D., Warden, A., Monk, S.M., and King, J.D., 1999, Identification and characterization of oil types and their source rocks, in Team, A.A., ed., The Oil and Gas Resource Potential of th Arctic National Wildlife Refuge, U.S. Geological Survey, p. OA-1 to OA-101.
Lillis, P.G., and Magoon, L.B., 2002, Geochemistry of Petroleum Systems of NPRA, in Bird, K.J., and Nelson, P.E., eds., Petroleum Resource Assessment of the National Petroleum Reserve in Alaska, U. S. Geological Survey, p. GC1-GC53.
Lorenson, T.D., Whiticar, M.J., Waseda, A., Dallimore, S.R., and Collett, T.S., 1999, Gas composition and isotopic geochemistry of cuttings, core, and gas hydrate from the JAPEX/JNOC/GSC Mallik 2L-38 gas hydrate research well, in Collett, T.S., ed., Scientific Results from JAPEX/JNOC/GSC Mallik 2L-38 Gas Hydrate Research Well: Ottawa, Geological Survey of Canada, p. 143-163.
Magoon, L.B., and Claypool, G.E., 1988, Geochemistry of oil occurrences, National Petroleum Reserve in Alaska, in Gryc, G., ed., Geology and Exploration in the National Petroleum Reserve in Alaska, 1974 to 1984: Washington, D.C., United States Government Printing Office, p. 519-550.
Masterson, W.D., 2001, Petroleum filling history of central Alaskan North Slope fields: Dallas, Texas, University of Texas at Dallas, Ph.D., 222 p.
Masterson, W.D., Dzou, L.I.P., Holba, A.G., Fincannon, A.L., and Ellis, L., 2001, Evidence for biodegradation and evaporative fractionation in the West Sak, Kuparuk and Prudhoe Bay field areas, North Slope, Alaska: Organic Geochemistry, v. 32, p. 411-441.
Pallaser, R.J., 2000, Recognizing biodegradation in gas/oil accumulations through the d13C compositions of gas components: Organic Geochemistry, v. 31, p. 1363-1373.
Parris, T.M., and Burruss, R.C., 2001, Tectonism, fracturing, and gas generation, northeast Brooks Range, Alaska.: AAPG Program with Abstracts.
Parris, T.M., Burruss, R.C., and O'Sullivan, P.O., 2002, in review, Deformation and the timing of gas generation and migration in the eastern Brooks Range foothills, Arctic National Wildlife Refuge, Alaska: American Association of Petroleum Geologists Bulletin, p. 41p., 12 figures, 2 tables.
Prinzhofer, A., and Huc, A.Y., 1995, Genetic and post-genetic molecular and isotopic fractionations in natural gases: Chemical Geology, v. 126, p. 281-290.
Robinson, F.M., 1956, Core Tests and Test Wells Oumalik Area, Alaska: Geological Survey Professional Paper, v. 305-A, p. 1-70.
------1958a, Test wells, Gubik area, Alaska: Geological Survey Professional Paper, v. 305-C, p. 207-264.
------1958b, Test well, Grandstand area, Alaska: Geological Survey Professional Paper, v. 305-E, p. 317-339.
------1959a, Test wells, Titaluk and Knifeblade areas, Alaska: Geological Survey Professional Paper, v. 305-G, p. 377-421.
------1959b, Test wells, Simpson area, Alaska: Geological Survey Professional Paper, v. 305-J, p. 523-568.
------1964, Core tests, Simpson area, Alaska: Geological Survey Professional Paper, v. 305-L, p. 645-730.
Robinson, F.M., and Collins, F.R., 1959, Core test, Sentinel Hill area and Test well, Fish Creek area, Alaska: Geological Survey Professional Paper, v. 305-I, p. 485-521.
Schoell, M., 1983, Genetic characterization of natural gases: American Association of Petroleum Geologists Bulletin, v. 67, p. 2225-2238.
Schoell, M., Wehner, H., and Coleman, M.L., 1985, North Slope Oil-Rock Correlation Study, in Claypool, G.E., ed., Alaska North Slope Oil/Rock Correlation Study: Tulsa, American Association of Petroleum Geologist, p. 443-458.
Sloan, E.D., Jr., 1998, Clathrate hydrates of natural gas: New York, Marcel Dekker, 705 p.
Threlkeld, C.N., Obuch, R.C., and Gunther, G.L., 2000, Organic Geochemistry Data of Alaska (1.0 ed.), U.S. Geological Survey.
Whiticar, M.J., 1994, Correlation of Natural Gases with Their Sources, in Dow, W.G., ed., The petroleum system - from source to trap: Tulsa, American Association of Petroleum Geologists, p. 261-283.
------1999, Carbon and hydrogen isotope systematics of bacterial formation and oxidation of methane: Chemical Geology, v. 161, p. 291-314.
[an error occurred while processing this directive]