USF Libraries

Stable hydrogen and carbon isotopic compositions of biogenic methanes

MISSING IMAGE

Material Information

Title:
Stable hydrogen and carbon isotopic compositions of biogenic methanes
Physical Description:
ix, 106 leaves : ill. ; 29 cm
Language:
English
Creator:
Burke, Roger Allen
Publisher:
University of South Florida
Place of Publication:
Tampa, Florida
Publication Date:

Subjects

Subjects / Keywords:
Marine sediments -- Gas content   ( lcsh )
Methane   ( lcsh )
Dissertations, Academic -- Marine science -- Doctoral   ( fts )

Notes

General Note:
Thesis (Ph. D.)--University of South Florida, 1985. Bibliography: leaves 92-106.

Record Information

Source Institution:
University of South Florida
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 030175771
oclc - 13894546
usfldc doi - F51-00166
usfldc handle - f51.166
System ID:
SFS0044210:00001


This item is only available as the following downloads:


Full Text

PAGE 1

STABLE HYDROGEN AND CARBON ISOTOPIC COMPOSITIONS OF BIOGENIC METHANE$ by Roger Allen Burke, Jr. A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Mari n e Science in the University of South Florida December, 1 985 Major P rofessor : Dr. William M Sackett

PAGE 2

Graduate Council University of South Florida Tampa, Florida CERTIFICATE O F APPROVAL Ph. D Dissertation This is to certify that the Ph.D. Dissertation of Roger Allen Burke Jr. with a major in Marine Sc i e nc e has been app r oved by the Examin i ng Committee o n August 1985 as satisfactory for t h e dissertation r eq u i r ement for t h e Ph.D. degree. Examining Comm i ttee: Member: Dr. Robert M Garrels Member: Dr. Edward S Van Vleet

PAGE 3

ACKNOWLEDGEMENTS I sincerely thank Dr. William M. Sackett for his advice, encouragement, and friendship during my graduate career. I also thank the other members of my committee: Dr. Robert Byrne, Dr. Robert Garrels, Dr. Harold Humm, and Dr. Edward Van Vleet for their time and input into this project. I thank Dr. James M. Brooks of Texas A&M University for providing most of the DSDP samples analyzed in this study. I thank Dr. George Claypool of the USGS, Denver for providing the DSDP Leg 84 samples. I thank Dr. Chris Martens of the University of North Carolina, Chapel Hill and the members of his research group "CH4Ao2sn for collecting and providing the Cape Lookout Bight samples analyzed in this study. I thank Suzy Wicks for the discussions we had about microbiology, for her interest and help with this project, and for her friendship. I especially thank my wife Jenny for her love and support throughout my stay at USF, and my little daughter Lisa, who for the past two and a half years has offered a different perspective on things. This work was financially supported by National Science Foundation grants OCE-8002455, OCE-83089115, and OCE-8417321 (Marine Chemistry Program). I also thank the Poynter Fund (Lake Fellowship) and the Graduate Council of USF for their generous support. ii

PAGE 4

TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ABSTRACT INTRODUCTION Natural Hydrogen Isotopic Compositions Natural Carbon Isotopic Compositions Origin of Natural Methanes Anaerobic Decomposition of Complex Organic Matter Methanogenic Pathways Methane Oxidation Natural Gas Classification Objectives iv v vii 1 3 7 8 9 1 4 16 19 22 EXPERIMENTAL METHODS 24 Preparation of Methane for Isotopic Analysis 24 Reduction of Water and oD Determination 26 + H Correction 27 sJmple Measurement Reproducibility 30 Determination of Ancillary Parameters 32 ISOTOPIC COMPOSITIONS OF METHANES FROM FRESHWATER SEDIMENTS 33 Sample Collection 34 Model of Biogenic Methane Formation 34 Results and Discussion 38 METHANE ISOTOPIC COMPOSITIONS IN CAPE LOOKOUT BIGHT SEDIMENTS 51 Sample Collection 54 Results and Discussion 54 METHANE ISOTOPIC COMPOSITIONS IN DEEP-SEA SEDIMENTS 67 Sample Collection 69 Results and Discussion 69 DSDP Leg 84 Middle America Trench Region 69 DSDP Leg 87 -Nankai Trough 73 DSDP Leg 96 Orca Basin 74 SUMMARY 85 LITERATURE CITED 92 iii

PAGE 5

Table 1. Table 2 Table 3. Table 4 Table 5. Table 6. Table 7. LIST OF TABLES Intercomparison of working standards and comparison of working standards with international reference samples (in /oo). oD-CH4 o13c-cH 4 and ancillary data for several freshwater environments and the Tampa Bay estuary. oD-CH4 o13c-cH 4 and oD-H2 0 from Cape Lookout Bight. Isotopic analyses of methane and water from DSDP Leg 84 (in /oo). Location of hydrates and% org. C data from Kvenvolden and McDonald (1985); Site locations, water depths, and sed. rates from von Huene, Aubouin et al. (1985). Isotopic anal6ses of methane and water from DSDP Site 583 (in /oo). Isotopic anal6ses of methane and water from DSDP Site 618 (in /oo). Comparison of isotopic equilibrium temperatures (T ) as calculated with the data of Lyon and (1984) with estimated environmental temperatures (T ). env iv 28 41 56 70 75 77 84

PAGE 6

Figure 1. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 1. LIST OF FIGURES Hydrogen isotope ranges of hydrogenaceous materials (data from: Friedman et al. 1964; Estep and Hoering 1980; Schiegl and Vogel 1970; Yapp and Epstein 1 982; Smith et al. 1983; Redding et al. 1980; Ziegler et al. 1976; Rigby et al. 1 981; Epstein et al. 1976; Smith and Epstein 1970; Yeh and Epstein 1981; Schoell 1980, 1983; Nakai et al. 1974; Woltemate et al. 1984; Ehhalt 1973; Ehhalt and Volz 1976; Begemann and Friedman 1968; L yon and Hulston 1984; Gunter and Musgrave 1971; Welhan 1981; and Cloud et al. 1958). The anaerobic deco mposition of complex organic matter to methane in sulfate-depleted environments. Percentages represent the flow of energy from the organic and inorganic compounds to methane (after Zehnder et al. 1982) Schematic of the system used for the preparation of methane for isotopic analyses. Model of methane formation. Lines represent the percentage mix of acetate dissimilation (AD) and carbon dioxide reduction (R) (after Woltemate et al. 1984). Plot of methane hydrogen isotopic composition versus environmental water hydrogen isotopic composition for several freshwater environments and the Tampa Bay estuary. The lines are the same as in Figure 4 Symbols: Crescent Lake, FL;_ Mirror Lake FL; A Lake Dias, FL; Kilmer Pond SC; Mississippi River Delta, LA; and $Tampa Bay estuary, FL. Plot of methane hydrogen isotopic composition versus methane carbon isotopic composition for several freshwater environments and the Tampa Bay estuary. Symbols the same as in Figure 5. Cape Lookout Bight, North Carolina, USA. The sampling site was Station A-1 (after Martens and Klump 1980). v 4 1 1 25 35 39 44 52

PAGE 7

Figure 8 Figure 9. Figure 10. variations of temperature, oD-CH4 and 6 C-CH4 in Cape Lookout Bight sediments. The temperature data are from Martens and Klump ( 1980 ) Martens et al. (1980), Crill and Martens (1983), and Martens (unpublished). Plot of methane hydrogen isotopic composition versus pore water hydrogen isotopic composition for samples from Cape Lookout Bight and the samples plotted in Figure 5. Symbols : 41 C a pe Lookout Bight; all others the same as in Figure 5 Plot of methane hydrogen isotopic composition versus methane carbon isotopic composition for the samples from Capl 3 Lookout Bight. The equation: 6D = -6.66 C 626 is a linear regression of all the data in this study. The equation: oD = -4.46 C 528 is a linear regression of data obtained for T 24C vi 55 59 64

PAGE 8

STABLE HYDROGEN AND CARBON ISOTOPIC COMPOSITIONS OF BIOGENIC METHANES by Roger Allen Burke, Jr. An Abstract A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Marine Science in the University of South Florida December 1985 Major Professor: Dr. William M. Sackett vii

PAGE 9

Stable hydrogen and carbon isotopic compositions of biogenic methanes collected from the sediments of several deep-sea, nearshore marine-estuarine, and freshwater environments were determined. The isotopic compositions of methane samples from eight different DSDP Sites (mean oD-CH4 = -185 /oo, std. dev. = 7/oo, n = 75; mean o13c-cH 4 0 0 = -71.3 /oo, std. dev. = 6.3 /oo, n = 44) are generally typical of m ethane formed via co 2 reduction in deep-sea sediments. Methane collected from several freshwater environments was 0 0 D-d epleted (mean oD-CH4 = -300 /oo, std. dev. = 26 /oo, n = 20) and 13c-enriched (mean o13c-cH 4 = -60.1/oo, std. dev. = 6.1/oo, n = 20) compared to the deep-sea methane. Normally, acetate dissimilation is thought to account for about 60 to 70% of the total methane production in freshwater sediments. Nearshore marine-estuarine methanes appear to be isotopically intermediate (mean oD-CH4 = -258/oo, std. dev. = 23/oo, n = 46; mean 0 0 -61.8 /oo, std. dev. = 3.1 /oo, n = 46) between deep-sea and freshwate r methanes. Variation in the relative importance of the two main methanogenic pathways, acetate dissimilation and co2 reduction, is prob ably the single most important factor responsible for the differences in methane isotopic compositions among these three different types of environm ents. Other factors that probably contribute to the methane isotopic differences are temperature, sedimentation rate, organic matter type and amount, concentration of alternate electron acceptors, rate of methane formation and possibly viii

PAGE 10

postgenerative isotopic equilibration. Shallow aquatic sediments are thought to be an important source of methan e to the atmosphere; the methane produced in these systems, including the ones sampled in this study, is generally substantially more 1 3 c-depleted than expected based on the o 13 c of atmospheric methane and the isotopic fractionation associated with the atmospheric sink process. T oo few oD data are available to allow evaluation of the role of shallow aquatic sediments in determining atmospheric oD-CH4 Abstract approved: Professor, Marine Science Date of\ approval ix

PAGE 11

INTRODUCTION The commerc ial production of natural gas, o f which m etha ne is usually the predominant co nstituent, accounted for about 27% of the total domestic energy production of the United States in 1975 (Natl. Acad Sci. 1 980) Natural gas is projected to account for about 20% of global e n e r gy consumption from 1990 to 2000 (Anonymous 198 3) The origi n of virtually all commercially important accumulations o f natura l gas is thought to be t h e degradation o f biologically derive d organic matter in aquatic sedim ents, m ediated eith e r by bacteria or b y the combined effects of time, temperature, an d catalysis after sediment burial. In this study, gas produced directly by bacteria is referred to as biogenic gas, and gas resulting from the effects of time, temperature and catalysis on organic matter is referred to as thermogenic gas. Rice and Claypool ( 1981) recently estima t e d that roughly 20% of the world's discovered gas reserves w ere formed by biogenic processes. In v i e w of the fact that biogenic methane occurs in geologically und erstoo d circu mstances, Rice and Claypool (1981) predict that even more than 20% of future gas discoveries will be of biogenic o rigin. Recent studies (Craig and Chou 1982; Khalil and Rasmussen 1982) of polar ice cores indicate an apparent doubling o f atmosp heric methane co ncen t ration s o v e r the past few hundred years. Recent high precision measurements (Khalil and Rasmussen 1983) indicate that the atmosp heric

PAGE 12

2 methane concentration has been increasing by 1-2% per year during the last several years. Because methane strongly absorbs infrared radiation within the atmospheric window (700 to 1400 em 1 ) that transmits most of the thermal radiation from the earth's surface to outer space (Wang et al. 1976), the atmospheric methane increase is a potential contributor to global warming. investigators (Wang et al . 1976; Lacis et al. 1981; Craig and Chou 1982) estimate the contribution of methane to increasing global temperatures to be about 20-40% of that attributed to the carbon dioxide increase. Microbial decomposition of complex organic matter in environments such as aquatic sediments and the intestines of ruminants is thought to be the most important source of methane to the atmosphere, accounting for an estimated 75 to 87% of the total input (Sheppard et al. 1 982; Hameed and Cess 1983). Improved understanding of the geochemistry of biogenic methane may be of practical significance to future energy exploration and climatic predictions. Two useful parameters characterizing natural gas are the stable hydrogen and carbon isotopic compositions of methane. The natural abundances of the stable hydrogen isotopes H and D are 99.9844% and 0.0156%, respectively (Hoefs 1980) The stable isotopes of carbon 12c and 13c have natural abundances of 98.89% and 1 .11%, respectively (Hoefs 1980). Stable isotopic compositions are usually reported as a parts-per-mil (0/oo) deviation (6) from a standard with a known isotopic ratio. The d efinition of the 6 value is: 6 = (R(sample)/R(standard) 1) X 1000 (1) where R is D/H and 13c ; 12c for hydrogen and carbon isotopic compositions, respectively. Values reported in this study are referred

PAGE 13

to the Vienna Standard Mean Ocean Water (V-SMOW; absolute 0/H = 155.76 -6 0.05 X 1 0 ; Hagemann et al. 1970; Gonfiantni 1978) for hydrogen and to the belemnite standard f rom the Cretaceous Peede e Formation of South Carolina (POB; absolute 13c;12c = 0 .01 1 2372; Cra i g 1953, 1957) for carbon. Natural Hydr oge n Isot opic Compositions Isotope effects are fundamentally the result of mass differences. Because hydrogen isotopes have such l a rge relative mass differences, large variations of 60 in nature are expected. Figure 1 shows approximate isotopic ranges for v arious types of hydrogen containing substances found in nature. The deuterium concentration of continental surface waters is determined primarily by the temperature at which local precipitation takes place; temperature i s in turn, influenced by latitude and altit u de (Friedman et al. 1 964) In arid regions the process of evaporati o n can lead to significant deuterium enrichment in standing water (Friedman e t al. 1 964) The 60 of the organically bound hydrogen of plants is largely determined by the 60 of moisture absorbed during g r o wth ( Schieg l and V oge l 1970) an d by the pathway o f photosynthetic carbon metabolism (c3 (Calvin-Benson), c4 (Hatch-Slack) or CAM (Crassulacean Acid Metabolism)) operative in t h e plan t (Ziegler et al. 197 6) The total o rganically b ound hydrogen is usually O-depleted relative to the water that the plants use for growth (Schiegl and Vogel 1970 ) with additional deuterium depletion d uring the formation of lipids (Smith and Epstein 1 970; Estep and Hoering 1980) When CAM plants are deprived of water their organically bound hydrogen can become O-enriched, apparently due to e nhanced evaporation of their tissue water (Ziegler et al. 1976) 3

PAGE 14

-1000 -800 -600 -400 -200 0 SURFACE WATERS HIGH LAT. MID. LAT. (20-50) LOW LAT. (0-20) LAND PLANTS -TOTAL CELLULOSE MARINE ORGANISMS TOTAL LIPIDS CELLULOSE FOSSIL FUELS COAL -CRUDE OIL -METHANE HYDROTHERMAL BIOGENIC ATMOSPHERIC THERMOGENIC HYDROGEN GAS HYDROTHERMAL BIOGENIC -1000 -800 -600 -400 -200 0 bD (%o) Figure 1 Hydrogen isotope ranges o f hydrogenaceous materials (data from : Friedman et al. 1964 ; Estep and Haering 1980; Schiegl and Vogel 1 970 ; Yapp_ and Epstein 1982; Smith et al. 1 983; Redding et al. 1980; Ziegler et al. 1976; Rigby et al. 1981; Epstein et al. 1976; Smith a n d Epstein 1970; Y e h and Epstein 1981; Schoell 1980 1 983; Nakai et al. 1974; Woltemate et al. 1984; Ehhalt 1973 ; Ehhal t and Volz 1976; Begemann and Friedman 1 968 ; Lyon and Hulston 1984; G unter and Musgrave 1971 ; Welhan 1981; and Cloud et al. 1958) 4

PAGE 15

5 Based on their early measurements of the oD of coal, Schiegl and Vogel (1970) suggested that there is some deuterium depletion in the conversion of plant material to coal and that some of the variability noted in coal oD might be attributable to climatic conditions at the time of plant growth Based on the results of a previous study by Redding et al. (1980) which indicated that coal oD is not greatly changed by the one-step formation of methane, and their own data, which indicated an inverse correlation between coal oD and maturity, Smith et al. (1983) concluded that coal oD is primarily source-controlled and that maturational effects are unimportant. Smith et al. (1983) assumed that there is essentially no isotope effect in the conversion of plant material to coal and that, following application of an average water-plant fractionation factor, their coal oD data could be used to estimate the paleolatitude of coal deposition. Based on analyses of over 100 crude oil samples and preliminary results from laboratory experiments, Yeh and Epstein (1981) concluded that the oD of oil is primarily source controlled and that maturity and chemical composition have little effect on the oD of normal crude oils. Yeh and Epstein (1981) based their conclusions on ( 1 ) the observation that there was no correlation between oD-oil and API gravity in their 0 sample set, (2) the range of oD-oil of their samples was only 10 0 loo compared to a oD range of more 250/oo for oil reservoir formation waters and, (3) the lack of detectable change in the oD of an oil sample after it had been in contact with water with a oD of about 0 0 2000 /oo at 1 80 C for two months. Laboratory and field studies by other authors, however, indicate that there is likely to be significant isotopic exchange between water and at least some of the fractions of

PAGE 16

oil (Koepp 1978; Schoell and Redding 1978), kerogen (Haering 1968; 1984), and coal (Rigby et al. 1981) during subsurface burial. 0 Field (Schiegl and Vogel 1970; Schoell 1980; 1 983) and laboratory pyrolysis (Chung 1976; Sackett 1978) studies indicate that methane formed from the thermal degradation of precursors (i. e., kerogen, coal, oil) is D-depleted relative to the precursor and that the magnitude of the depletion decreases with increasing precursor maturity. Field and laboratory studies (Nakai et al. 1 974; Schoell 1980; Woltemate et al. 1984) indicate that biogenic methane is D-depleted relative to the water in which the methanogenic bacteria grew and that oD-CH4 is affected by oD-H2o, oD of the precursor organic matter, and the pathway used in methane generation. Available measurements (Bainbridge et al. 1961; Begemann and Friedmann 1968; Ehhalt 1973) indicate that atmospheric methane is greatly D-enriched compared to its known sources. There appears, however, to be a large kinetic hydrogen isotope effect associated with the destruction of methane in the atmosphere by reaction with hydroxyl radicals (Senum and Gaffney 1985). Lyon and Hulston ( 1984) listed four possible origins for the methane in hydrothermal systems: (1) primordial or abiogenic methane from the mantle (i. e., Welhan and Craig 1 979; Gold and Soter 1980); (2) the high temperature decompositi on of organic matter (i.e., Welhan 1981; Burke et al. 1981); (3) chemical reaction between carbon dioxide and hydrogen (Fisher-Tropsch reaction, i.e., Hunt 1979; Giggenbach 1980) and (4 ) biological production (i. e., Burke et al. 1981; Baross et al. 1982; Baross and Deming 1983). Due to rapid isotopic equilibration between hydrogen-water and hydrogen-methane, any oD differences in

PAGE 17

methane and hydrogen formed by these processes are obscured, and oD-H 2 0 and oD-CH4 reflect the temperature (100-1200 C) of the water at the time of sampling and Musgrave 1971; Welhan 1981) Natural Carbon Isotopic Compositions The distribution of the stable carbon isotopes in natural 7 substances has been the subject of many reviews (i.e., Craig 1953; Fuex 1977; Deines 1980; and others) to which the interested reader is referred for more detail than will be presented here. Marine limestones typically have the greatest relative concentration s of 13c, with o 13 c values generally between 5 and +5 loo (Fuex 1977) Terrestrial plants that utilize the c 3 pathway are generally 13C-depleted ( -32 to -22/oo) relative to terrestrial plants fixing carbon dioxide via the c 4 pathway ( -1 8 to -9/oo; Deines 1980). Terrestrial plants that are known to use the CAM pathway exhibit o 13c that span the range of c 3 and c 4 plants (-32 to -11/oo) as a result of their unique ability to accomplish net carbon dioxide fixation via either the c 3 or c 4 pathway (Deines 1980). Most marine phytoplankton from tropical and temperate waters have a o 13 c of about 20 loo (Sackett in press). The o 1 3c of coals, crude oils, and kerogens are largely inherited from their precursors and are less apt to change as a result of maturational processes than are oD (Rigby et al. 1 981; Yeh and Epstein 1981). Crude oils (mode -21 to -30/oo) are generally 1 3c -depleted relative to coal (mode -23 to -26/oo) and kerogen, probably as a result of preferential retention o f isotopically light lipid components during oil formation (Deines 1980). Methane produced from the thermal decomposition of organic matter generally has a o 1 3c ranging from 60

PAGE 18

8 to -25/oo, whereas biogenic methane is generally more 13 c-depleted ( 90 to -50/oo; Fuex 1977) The o13c of the abiogenic methane emitted from the East Pacific Rise ranges from -17.6 to -15/oo (Welhan 1981) Origin of Natural Methanes Interest in primordial or abiogenic methane has increased recently as a result of observations that methane along with other mantle-derived gases ( 3He, co2 H 2), is injected into the deep sea from active spreading centers (Welhan and Craig 1979; Welhan 1981). These authors stated that it is uncertain whether the methane is an intrinsic mantle component that is extracted from the basalts by the circulating seawater or whether it represents a chemical equilibrium with co 2 and H 2 (Fisher-Tropsch reaction) under high pressure and decreasing temperatures (Welhan and Craig 1 979; Welhan 1981) Gold and Soter (1980) suggested that degradation of complex hydrocarbons, formed at the time of earth' s accretion under conditions of high temperature and pressure, is responsible for this deep-earth gas. Gold and Soter (1980) further suggested that the quantity of methane produced in thiS manner is enormous and that methane from this source is responsible for most of the gas and oil (through polymerization) in commercial accumulations. Most geochemists, on the other hand, think that thermal and biological decomposition of photosynthetically derived organic matter in marine or lacustrine sediments accounts for virtually all significant accumulations of natural gas. Early diagenetic (sed. T < 50C) reactions, primarily mediated by microorganisms, convert labile chemica l groups to gases (i.e., C02 CH4 ) and cause the remaining organic matter to be converted into a complex polymeric form termed

PAGE 19

kerogen. At temperatures between 50 and 200C (catagenesis) the combined effects of time, temperature, and catalysis lead to the formation of gas and oil or coal from the kerogen (Hunt 1979) Between temperatures of about 60 and 150C oil and some gas is formed by catalytic and thermal cracking of kerogen (Hunt 1979). Peak methane generation from kerogen, oil, and coal occurs between about 150 and 200C, with only a small amount of methane generation at higher (metamorphic) temperatures (Hunt 1979). In catalytic cracking, which is believed to be the dominant process in petroleum generation at 0 temperatures of up to about 125 C (Hunt 1979), positively charged, highly reactive, carbonium ions are generated by hydride removal from paraffin molecules a t Lewis acid centers of clay minerals (Sackett 1978). Carbonium ion reactions can then give rise to a large variety of products. In thermal cracking, which becomes increasingly important at temperatures above 125C (Hunt 1979), the reactive intermediate is a free radical that arises from the homolytic cleavage of a carbon to carbon bond in which each carbon retains one of the two shared electrons. As with carbonium ion reactions, a wide spectrum of products is possible when free radicals are created. Biogenic methane is produced by a specialized group of obligately anaerobic bacteria that are limited to a narrow range of low molecular weight substrates for growth and e nergy production. Known methanogenic substrates include carbon dioxide a nd hydrogen, acetate, methanol, and methylamines (Zehnder e t al. 1982). Methanogenic bacteria thus rely on other bacterial species to decompose complex organic compou nds to substrates suitable for their growth. Anaerobic Decomposition of Complex Organic Matter 9

PAGE 20

10 The microbial degradation of complex organic matter to carbon dioxide and methane in anaerobic environments such as organic-rich sediments and sewage sludge digesters is a multi-step process requiring the participation of at least three different trophic groups of bacteria (Zehnder et al. 1982). In environments in which carbon dioxide and protons are the only available inorganic electron acceptors, this process is believed to proceed as illustrated in Figure 2. In the first step, biological polymers such as polysaccharides, proteins, and lipids are digested by extracellular enzymes to soluble materials that are then fermented to organic acids and alcohols by the acid-forming or chemoheterotrophic fermentative hydrogen-producing bacteria (Brock 1974; Mah 1982). In the next step organic acids larger than acetate and alcohols larger than methanol are converted to acetate, carbon dioxide, and hydrogen by the chemoheterotrophic non-fermentative obligate proton reducers (Mah 1982). The reducing equivalents (hydrogen) and acetic acid generated during the first two steps are disposed of in the final step by the methane-producing bacteria. The pathways for oxidation of organic compounds and ATP generation under anaerobi. c conditions can be divided into two major groups, fermentation and anaerobic respiration (Brock 1 974) In fermentation a compound at an intermediate oxidation level is disproportioned both to compounds that are more oxidized and compounds that are more reduced than the starting substrate in the abscence of an external electron acceptor. For example, the fermentation of glucose by Ruminococcus albus produces ethanol, acetate, hydrogen, and carbon dioxide (Wolin 1 979) Fermentation releases only a small part of the available

PAGE 21

BIOLOGICAL POLYMERS HYDROLYTIC AND FERMENTATIVE MICROORGANISMS ALCOHOLS, CARBOXYLIC ACIDS (except acetate) OBLIGATE PROTON REDUCERS Figure 2. The anaerobic decomposition of complex organic matter to methane in environments. Percentages represent the flow of energy from the organic and inorganic compounds to methane (after Zehnder et al. 1982). 11

PAGE 22

12 substrate energy; e.g., in the fermentation of glucose albus about 90% of the glucose energy remains in the end -products (Mah 1982). In anaerobic respiration, an external electron acceptor other than oxygen (i.e., nitrate, sulfate, protons, or carbon dioxide) is involved in the oxidation of organic compounds for energy production (Brock 1974). Examples of anaerobic respirations involved in the degradation of complex organic matter (Figure 2) are the production of acetate, hydrogen, and carbon dioxide by the obligate proton reducers, and methane production from carbon dioxide and hydrogen. Without the anaerobic respiration of the obligate proton reducers a significant portion of the complex organic matter would be only partially degraded to higher fatty acids and alcohols (Figure 2) Because degradatio n of these higher fatty acids and alcohols requires the use of enzymes that are strongly inhibited by hydrogen (Wolin and Miller 1982), the obligate proton reducers can operate only if methanogens, or some other hydrogenotroph, efficiently consume the hydrogen they produce. This process, referred to as interspecies hydrogen transfer (Iannott i et al. 1973; Wolin and Miller 1982), may also ensue when chemoheterotrophic fermentative hydrogen-producing bacteria are cultured with hydrogenotrophs (Wolin 1982) In this case interspecies hydrogen transfer is not obligatory because the fermentative organism has an alternate means of electron disposal. In monocul ture, Selenomonas rumina tum fermented lactate to propionate, acetate, and carbon dioxide (Chen and Wolin 1977). Addition of s. ruminatum to a pregrown culture of the methanogen Methanobrevibacter smithii (formerly Methanobacterium ruminatum) by a sequential transfer procedure caused a shift in lactate fermentation products to more

PAGE 23

1 3 acetate and less propionate with concomitant production of methane from carbon dioxide and hydrogen (Chen and Wolin 1977). Thus, interspecies hydrogen transfer is important to the anaerobic degradation process (Figure 2) not only because it creates an environment that is conducive to the growth of the obligate proton reducers, but also because it can cause a shift in the electron and carbon flow (Mah 1982 ) Electron flow is diverted away from the soluble reduced electron sink products (i.e., propionate) to proton reduction (hydrogen formation) with subsequent methane production, and carbon is channeled into the production of methane and the more oxidized soluble end products (i.e. acetate) at the expense of the more reduced soluble compounds such as propionate (Mah 1982). This is advantageous to the hydrogenotrophs ( esp.ecially methanogens) because they gain energy sources (hydrogen, acetate), a nd often to the hydrogenogens because they may obtain more ATP from their substrates due to the relatively more complete oxidation of the starting substrate (Wolin 1982). Other electron acceptors such as nitrate, iron and manganese oxides, and oxidized sulfur compounds are preferentially used (compared to protons and carbon dioxide) in anaerobic respirations (Zehnder et al. 1982), a nd when they are present in sufficient quantities carbon and electron flow (Figure 2) may be diverted away from methane production. Because of its high concentration in seawater, sulfate is the predominant inorganic electron acceptor in upper, anoxic marine sediments (Fenchel and Blackburn 1979; Reeburgh 1983) Consequently, many studies (e.g., Martens and Berner 1974; Claypool and Kaplan 1974 ; Ni kaido 1977) f o und that optimum rates of methane production and near-saturation methane concentrations are not encountered in marine

PAGE 24

sediments until sulfate is largel y (80-90%) depleted. This inhibition of methane production by high concentrations of dissolved sulfate is apparently due to the ability of sulfate-reducing bacteria to outcompete methanogens for substrates such as hydrogen and acetate (Winfrey and Zeikus 1977). Claypool and Kaplan (1974) proposed that sulfate reducers gain competitive advantage by virtue of the fact that sulfate reduction yields relatively more energy than methane production; however, McCarty (1972) argued that if two reactions are energetically feasible, then from a strictly thermodynamic viewpoint, one reaction can not inhibit the other simply because it is energetically more rewarding. Recent studies (Kristjansson et al. 1982; Schonheit et al. 1982) have shown that sulfate reducing bacteria have higher substrate affinities (lower K ) by factors of 5 and 15 for s hydrogen and acetate, respectively, than do methanogens. This difference in substrate affinities does not imply that methane production and sulfate reduction from these substrates are mutually exclusive (Kristjansson et al. 1982; Schonheit et al. 1982), but it does provide a satisfactory explanation for the vertical separation of the sulfate-reducing zone from the methane-producing zone that is generally observed in marine sediments (Claypool and Kvenvolden 1983) Methanogenic Pathways In sulfate-depleted marine sediments, as well as freshwater sediments, the dominant terminal process of anaerobic organic matter decomposition is methane production. Methanogenesis is generally 14 thought to occur mainly via two pathways, co2 reduction (2) and acetate dissimilation (3): + 8e + + ( 2)

PAGE 25

+ ( 3) The results of 4c labeling studies implied that acetate normally accounts for about 60 to 70% of the total methane production in sewage sludge digesters (Jeris and McCarty 1965; Smith and Mah 1966), paddy soils (Koyama 1963; Takai 1970), and some freshwater lake sediments (Cappenberg and Prins 1974; Winfrey and Zeikus 1979; Lovley and Klug 1982), with co2 reduction responsible for the remainder. In other freshwater lake sediments, however, the co2 reduction pathway is apparently predominant, accounting for about 90% of the total methane production in Russian Lake Kuznechikha (Belyaev et al. 1975) and about 75% of the methane generated in the sediments of Blelham Tarn in the English Lake District (Jones et al. 1982). Incubation of sediment samples from Cape Lookout Bight, North Carolina, a small, nearshore marine basin with rapidly accumulating, organic-rich sediments, with 15 ( 1 2-14c) sodium acetate indicated that more than 50% of the Cape Lookout Bight summertime methane flux could be accounted for by acetate dissimilation (Sansone and Martens 1981). If sediments receive a fairly specific type of organic matter input or if methanogens are competing with sulfate-reducing bacteria for H 2 and acetate, a significant fraction of the total methane production can result from other substrates such as methanol and trimethylamine (Oremland et al. 1982; King et al. 1983). All of the aforementioned studies used 14c labeled substrates as tracers to estimate the relative contributions of the methanogenic pathways to total m ethane production. An alternate method was proposed for estimating relative pathway contribution based on measurements of

PAGE 26

16 3 1 + when dissolved inorganic nitrogen (DIN = NH4 + + N03 -) concentration s are low bu t not when DIN concentrati o ns are above 1 5 to (Rudd et al. 1976; Sansone and Marte ns 1978). The aerobic oxidation process may require relatively high methane concentrati o n s For exa m p l e methane co ncentratio n s l imited the

PAGE 27

17 rate of methane -oxidizer population development in Lake Mendota Wisconsin (Harrits and Hanson 1980) and Sansone and Martens (1978) proposed that aerobic _methane oxidation in Cape Lookout Bight waters should be limited by other factors at methane concentrations Pure cultures of methanogenic bacteria (Zehnder and Brock 1979), freshwater lake sediments, and sewage sludge (Zehnder and Brock 1980) anaerobically converted 1 4cH 4 to 1 4co2 while producing methane; however, net methane production was not observed and the methane oxidation process was inhibited by the same conditions that inhibited methane production (i.e., presence of oxygen) implying that methanogenic bacteria were responsible for the methane conversion. Profiles of methane concentration in anoxic marine sediments frequently show extreme upward concavity in the zone of sulfate reduction (Barnes and Goldberg 1976; Reeburgh 1976; Martens and Berner 1977; Bernard 1978; Alperin and Reeburgh 1984 ) Barnes and Goldberg (1976) and Reeburgh (1976) suggested that net consumption of upward diffusing methane in the sulfate-reducing zone is required to explain the extreme upward concavity. Further, successful fits of diagenetic models to methane profiles from various anoxic marine sediments were < obtained only if the models included a methane consumption term (Martens and Berner 1977; Bernard 1978; Alperin and Reeburgh 1984 ) Reeburgh (1976, 1980) proposed that methane consumption in anoxic marine sediments primarily takes place in a relatively thin layer (estimated to be about 10 em in Cariaco Trench sediments) at the base of the sulfate-reducing zone where the profiles of methane, sulfate, and EC02 all typically exhibit slope changes. Several recent studies (Reeburgh 1980; Devol 1983; Alperin and Reeburgh 1984; Iversen and

PAGE 28

Jorgensen 1985) demonstrated the conversion of 14cH4 to 14co2 in relatively thin layers (about 15 em to 40 em thick) near the base of 18 the sulfate-reducing zone of various marine sediments. The location of the anaerobic methane consumption layer at depths where the concentrations of both sulfate and methane are appreciable, the demonstrated ability of the sulfate reducer Desulfovibrio desulfuricans to co-metabolize under anaerobic conditions (Davis and Yarbrough 1966), and the co-occurrence of peaks in methane oxidation rate and sulfate reduction rate (i.e., Iversen and Blackburn 1985), have been cited as evidence that anaerobic methane consumption is coupled to sulfate reduction, either directly (i.e., Reeburgh 1976) or indirectly via reactions coupled to other substrates (i.e., Devol 1983) The anaerobic conversion of 14cH4 to 14co2 in minimally disturbed marine sediments from the main anaerobic consumption zone of Skan Bay, Alaska was not inhibited by specific inhibitors of either methanogenesis (2-bromoethanesulfonic acid) or sulfate reduction (molybdate), implying that anaerobic methane consumption in marine sediments is not directly mediated by either methanogenic bacteria or sulfate-reducing bacteria (Alperin and Reeburgh 1985). Alperin and Reeburgh (1985) suggested that their results are consistent with two possibilities: anaerobic methane consumption in marine sediments is either mediated by an unknown organism, which possibly uses iron oxides, manganese oxides, or reduced sulfur compounds as an external electron acceptor, or methane consumption is accomplished by a bacterial consortium that includes an unknown organism capable of converting methane to carbon dioxide and hydrogen and sulfate-reducing bacteria that consume the hydrogen. Alperin and Reeburgh ( 1985)

PAGE 29

19 pointed out that anaerobic methane consumption would not be inhibited by molybdate if other hydrogen-consuming organisms replaced sulfate-reducers in the consortium. It is not clear, however, why this modified consortium does not function at greater depths where methane concentrations are e ven great er. The confinement of substa ntial methane consumption t o a thin layer implies that some component of the sulfate-reduction zone perhaps som e trace product of the sulfat e reduction process, is directly involved in anaerobic methane consumption. The impo rtance of iron and manganese oxides in anoxic sedi ments to the microbial decomposition of organic matter in general and methane in particular, is unknown because these metal oxides may be used as an e lectron acceptor directly by orga n isms o r by metabolic prod ucts (i.e. sulfide) of organisms ( Burdige and Nealson 1985), including a variety o f organic substances (Stumm and Morgan 1 981) Natural Gas Classificati on Isotopic and molecular compositional variations have been used frequently to develop models that describe the origin and subsurface history of natura l gas. All of these models assume that the observed isotopic and compositional variations of natural gases predominantly reflect processes (primarily kinetic isotope effects and source maturity) that occur d urin g gas formation (Schoell 1 983) althou gh this is not always explicitly stated. Galimov ( 1 968) presented data tha t indicated methan e becomes less 13c-depleted with increasing depth in the sediment, with most of the c h ange occurring during the initial stages of burial. To explain his data, Galimov (1968) develope d a model in which the sediment column was divided i nto three zones: (I) the biochemical zo n e (T < 25C) in which

PAGE 30

20 bacteria mediate the production of methane and its isotopic equilibration with carbon dioxide, (II) the catalytic zone (25C < T < 200C) in which methane is l3C-depleted relative to its precursor due to temperature-dependent kinetic isotope effects but is not in isotopic equilibrium with carbon dioxide because temperatures are too low for sufficient inorganic co 2 -cH4 isotopic exchange and bacterial activity ( ) ( C < T < 300C) 1 n is practically absent and III the thermal zone 200 which temperatures are sufficient for control of o13 c -cH4 by inorganic C02 -cH 4 isotopic equilibration. Stahl ( 1974) presented a model in which three separate stages of gas formation are apparent from a plot o f o13c-cH 4 versus c 1/[Cn (ratio of methane over all saturated gaseous hydrocarbons). In the initial stage the gas is formed predominantly by bacterial action and is characterized by a o13 c -cH4 ranging between about 80 and -68/oo and a c 1/[Cn near unity. The second stage represents thermocatalytic alterati on of marine source material as it is subjected to higher temperatures and time during deeper burial. Gas formed during the second stage maturational trend is characterized by progressive increases in o13c-CH4 from about -68 to 42/oo and progressive decreases in the c,;[cn ratio from a bout 1.0 to abou t 0.8. Upon further burial the source material and its hydrocarbo n products are subjected to increasing thermal stress; this results in gas that continually becomes more methane-and 13c-enriched. Gas formed in this third stage is typified by o13c-cH 4 between -42 and -35/oo and a C1/[Cn ratio of 0.8 to 1.0 (Stahl 1974). The results of Stahl (197 4 ) also indicate that o 13 c -CH 4 is dependent upon the type of source material from which it originates. Methane produced from predominantlY

PAGE 31

21 terrestrial source material during coal formation displays the same trends as noted above for marine source material, but with less 13c-depletion as high as -20/oo; Stahl 1974). The apparent decrease in c 1 /ECn with increasing thermal maturity during the second stage observed by Stahl (1974) is at odds with the laboratory pyrolysis results of Sackett ( 1978) that indicated the opposite trend, i.e., methane initially accounts for a small (<10%) fraction of volatile pyrolysis products and becomes increasingly more important with longer pyrolysis. Sackett's ( 1978) experiments were carried out at higher temperatures (400-500C) than would be found under natural conditions and, as pointed out by Bernard (1978), may not accurately simulate natural processes. Also, Hoering (1984) argued that because water is ubiquitous in sediments, dry pyrolysis experiments may be unrealistic in that some hydro gen transfer reactions in which water plays a role under natural conditions may be blocked. On the other hand, the carbon and hydrog e n isotopic compositions of several natural methanes associated with crude oils were qualitatively comparable to compositions that would be predicted from the results of Sackett (1978) implying that the dry pyrolysis process bears a close resemblance to gas formation in nature (Schoell 1980). Bernard et al. ( 1977) proposed that natura l gases could be efficiently characterized by a model in which c13 c-cH 4 was plotted against the ratio c 1/(C2 + c 3 ) (methane divided by the sum of ethane and propane) Bernard (1978) pointed out that molecular ratios expressed in this way vary by more than four orders of magnitude for natural samples, so this parameter should yield greater resolution than the C 1 /ECn ratio of Stahl ( 1974). Bernard et al. ( 1977) suggested

PAGE 32

22 that, in general, biogenic gas has a c1;(c2 + c3 ) ratio greater than 1000 and a o13c -cH4 more negative than -60/oo, whereas thermogenic gases generally have G1/(C2 + c3 ) ratios of less than 50 and o13c -cH4 0 greater than 50 /oo. The Bernard et al. (1977) model also takes into account the isotopic and compositional changes, with comcomitant loss of resolution, that may result from substrate depletion during biogenic methane formation, the mixing of biogenic and thermogenic gases, and subsurface migration of thermogenic gas. Schoell (1980, 1983) developed a more extensive gas characterization model in which o13c-cH4 is separately correlated wit h 13 the three parameters: 6D-CH4 6 C-C2 H 6 and c2+ (concentration of saturated gaseous hydrocarbons larger than methane). Use of the two 13 new parameters 6D-CH4 and 6 C C 2 H 6 made possible in part by recently improved analytical capabilities, allows a clearer distinction to be made between different genetic types of gas in some cases. For example Schoell (1980 1983) suggested that measurement o f 6D-CH4 might allow differentiation of biogenic gas produced under marine conditions from biogenic gas produced in a terrestrial system when o13c-CH alone is insufficient to make the distinction. 4 The isotopic and compositional variations that may result from secondary processes such as oxidation, mixing, and migration are also indicated in the Schoell ( 198 3) model. For example, migration may appreciably change the c2+ concentration but would generally not be expected to alter significantly the isotopic composition of methane (Schoell 1 983) Objectives My research had the following two objectives: (1) to add to the meager 6D-CH4 data base by measuring methane samples from a variety of

PAGE 33

23 aquatic environments such as deep-sea, nearshore marine-estuarine, and freshwater lake sediments. The resulting inhanced understanding of natural 6D-CH4 variations should then be of use in the improvement of natural gas characterization models and, in conjunction with concurrent o13c-cH4 measurements, eventually aid in gaining a clearer understanding of the relative contributions of the major methane sources to the atmosphere; (2) to use 6D-CH4 measurements to estimate the relative importance of acetate dissimilati on and co2 reduction to methane production in various aquatic sediment types. In the past, pathway importance was generally estimated from the results of 14c radiotracer studies. T o accomplish these objectives, it was necessary to construct a preparation system for the separation of methane from other gases and its conversion to products suitable for stable isotopic measurement The details of such a system h ave not been previously published and few investigators have made oD-CH4 measurements.

PAGE 34

24 EXPERIMENTAL METHODS Preparation of Methane for Isotopic Analysis Methane gas samples were prepared for isotopic analysis with the system illustrated in Figure 3 Preparation involves gas chromatographic (GC) separation of methane from other gases followed by combustion to co2 and water in a vacuum line. The GC used is a Hewlett-Packard 5710A equipped with a thermal conductivity detector and two 6mm OD stainless steel columns (3m grade 12 silica gel and 2m molecular sieve 5A) connected in series through a Valco ten port switching valve configured for column sequence reversal and backflush of the silica ge l column to the detector. This GC configuration allows baseline separation of H 2 o2 N 2 CH4 and co2 The helium carrier gas flow rate is 50ml/min and the oven temperature is held constant at 90C. Methane is combusted in a cupric oxide furnace (850C) to which 5-1 0 torr of oxygen is added before each run, as it is swept through the furnace by the GC carrier gas. Addition of oxygen to the system before each run is necessary to assure complete combustion of the methane. The isotopic composition of incompletely combusted methane was found to be significantly lighter than the actual composition; 0 3.5 /oo, respectively, were observed. The C02 and water resulting from the combustion of methane are condensed by liquid nitrogen ( 196C) in the trap immediately downstream from the combustion oven the excess oxygen

PAGE 35

S WITCHING VALVE tVACUUM VACUUM VACUUM \ GAUGE --H 2 0 COLLECTION C02 COLLECTION VENT-Figure 3. Schema tic o f the system used for the prep aratio n of methane for isotopfc a n a lyses. 02JN f\) V1

PAGE 36

26 and helium are pumped away, and the co2 and water are cryogenically (-89C) separated. The a13c of the co2 is then measured in a Finnigan nee Varian MAT 250 iso_tope ratio mass spectrometer (IRMS) by comparison With the NBS-21 standard (a13c = -28.16/oo; Schoell et al. 1983) Reduction of Water and aD Determination The water from the combustion of methane is then collected in a sample tube containing approximately 0.25g of treated zinc shot and reduced to H 2 gas by reaction with the zinc at 450C for 30 minutes (Coleman et al. 1982). The sample tube consisted of either a 15cm length of 9mm OD Pyrex tubing that was sealed at one end or a high vacuum teflon plug, 90 pattern stopcock (Model 8195-47 Ace Glass Inc., Vineland, NJ) with one arm sealed. The reducing agent used is AnalaR zinc shot (BDH Chemicals Ltd., Poole, England; presently obtainable from Gallard/Schlesinger Chemical Mfg. Corp., New York) that was sieved so that it passes through size 14 mesh but is retained by size 60 mesh. After sieving, the dull, gray oxide coating is removed from the zinc shot by reaction with dilute nitric acid until the zinc becomes quite shiny. This activated zinc is then washed several times With distilled water, dried and outgassed under vacuum overnight at about 300C. When not in use, the treated zinc is stored under vacuum The aD of H 2 gas is measured in the IRMS by comparison with a H 2 gas of known 6D. Through the assistance of Dr. Thomas Hoering of the Geophysical Laboratory, Carnegie Institution of Washington several high pressure cylinders of H 2 gas of differing 6D were prepared for use as routine working standards. Deuterium-depleted methane samples were always compared to Tank Standard-C (TS -C), which was assigned a oD of 220 .4/oo based upon comparison with the water reference sample

PAGE 37

Greenland Ice Sheet Precipitation (GISP) and Tank Standard-A (TS-A) which was assigned a oD of -31.9/oo from comparison with the NBS-1 water reference sampl e (Table 1) Comparison of these two tank standards with a third (TS-B) indicated that at least the isotopic difference between TS-A and TS-C is correct (Table 1) When isotopically heavier water samples wer e run, TS-A was used as the working standard. Each day that samples were run, the tank standard serving as the working standard was used in determining the oD of at least one of the other tank standards. The isotopic differences (TS-A vs. TS-C (242.4/oo) and TS-C vs. TS-A (-194.1/oo) were not quite internally consistent (Table 1) implying that H 2 gas admitted to the mass spectrometer through the sample-side of the inlet system is fractionated by about 0 .6/oo relative to gas flowing in through the standard-side capillary. Because the magnitude of this apparent fractionation is comparable to the uncertainty involved in determining the oD of H 2 gas with the IRMS and is small compared to the overall uncertainty involved with oD analysis of water (Table 1) and methane (see below) and would have no effect on any interpretations made here, n o correction for this effect has been applied to the data presented here. H + 3 Correction When mixtures of H 2 and HD gas a r e ionized in the mass 27 + + + spectrometer source, H 2 and HD 1ons are formed. Reaction of H 2 with + H 2 also results in the formation of H 3 ions that cannot be resolved from HD + ions by the mass spectrometer used in this study. Because d t + / + + e ermination of the ratio HD H 2 1s desired, the contribution of H 3 + to the mass-3 signal must be removed In this study the H 3 correction

PAGE 38

2 8 Table 1. Intercomparison of working standards and comparison of working standards with international reference samples (in /oo). Sample Tank Standard A Tank Standard c 00TS-A 00SMOW 00TS-C 00SMOW TS-A -----------242 4 3 1.9 (aO o7, n) TS-B -83o 1 -112o4 138o8 -112.2 (a .. o.1, n=7) (aO o6, n) TS-C -194o1 -220oll ----------(a=Oo5, n"'42) V-SMOW 31.3 -1.6 -----------(a=1o 2 n=21) NBS-1 -15.7 -47 01 ------------( a-1 o 1 n=16) GISP ------------39.8 -189o 4 (a = 2 o3, n)

PAGE 39

29 was determined by the method of Nier (1947) using software supplied with the HP 9815A calculator that controls operation of the IRMS. + Because the H3 rate production is proportional to the square of the + hydrogen partial pressure, whereas the.rates of generation of the H 2 + and HD ions are directly proportional to the hydrogen partial pressure, measurement of the mass-3/mass-2 ratio (i3 +1i 2+) at different + pressures (proportional to i 2 ; Friedman 1953) yields a straight line (Nier 1947). + The product of the slope of this line and i 2 at the pressure at which an isotopic analysis is made is the contribution to + + + the ratio i 3 / i 2 resulting from the formation of H 3 and is simply + + subtracted out to yield the desired i (HD)/i (H2 ) ratio. Determination of this slope or H 3-factor requires a minimal investment of time (five minutes) and operator effort; this makes periodic redetermination practical if it is suspected that IRMS operating conditions have fluctuated. + The H 3 contribution is large (about 30%) at typical operating pressures employed here (mass-2 signal > 5V or sample size > 2cc(STP) H 2 ) but is fairly constant (within about 1 %) It can be calculated that a 1% fluctuation in the H 3-factor should cause an uncertainty of about 0.5/oo when a sample-standard pair differing in 0 cSD by about 200 loo is compared. This is close to the long term uncertainty that was observed for the comparison of TSA to TSC (Table + 1) further implying that the H 3 contribution remains constant to within 1% under typical operating conditions. Because the sample-standard pairs measured in this study were generally much closer in cSD than 200/oo, measurement uncertainty due to H 3-factor fluctuation should generally be less than 0. 5 /oo. At very low operating pressures the Nier ( 1947) relationship may deviate from

PAGE 40

30 linearity (Whelhan 1981) and may not be applicable when samples much smaller than the smallest measured here (about 2cc(STP) H 2 ) are analyzed. Sample Measurement Reproducibility The GC-combustion system (Figure 3) described here is a flow-through system in which methane is combusted as it is swept through the furnace by the GC carrier gas. There is no Toepler pump or bellows to cycle the gas repeatedly through the furnace. As a result, it is reasonable to expect that the time the methane spends in the oven which is determined by the GC carrier gas flow rate, could be an important variable. To test this, a methane working standard was analyzed several times using carrier gas flow rates of 25ml/min and 50ml/min. Isotopic values obtained at 50ml/min (mean OD-CH4 = 0 0 13 0 -164 /oo, std. dev. = 1.0 /oo, n = 12; mean o C-CH4 = -44. 2 /oo, std. 0 dev. = 0 .14 /oo, n = 12) were not significantly different from those obtained at a GC flow of 25ml/min (mean oD-CH4 = -164/oo, std. dev. = 1 .7/oo, n = 6; mean o 13 c-cH 4 = -44.1/oo, std. dev. = 0.17/oo, n = 10). Repeated analyses of a second methane working standard, analyzed daily during the course of this work, yielded standard deviations (mean oD-CH4 = -189.3/oo, std. dev. = 1.7/oo, n = 36; mean o 1 3c-cH 4 0 -39.6 /oo, std. dev. 0 21/oo, n = 32) that are similar to those obtained above for the GC flow rate test and are comparable to those reported by Schoell (1980). The combustion efficiency of the GC-combustion system (Figure 3) was tested by comparing the methane concentration of the gas atmosphere in the vacuum line resulting from the injection into the GC of 2 ml of methane under normal operating conditions (three separate trials) to

PAGE 41

31 the methane concentration of the vacuum line gas atmosphere after injection of 2 ml of methane into the GC with the combustion furnace turned off (three trials). The trap immediately downstream from the combustion oven was immersed in a Dewar flask containing an isopropyl alcohol slush (-89C) The resulting meth a ne concentrations, which were measured with a GC equipped with dual flame ionization detectors, indicated a combustion efficiency >99%. Water samples were prepared for oD determination by the zinc metal method (Coleman et al. 1982). In this case, water samples were simplY injected into the sample tubes with a syringe as described by Coleman et al. ( 1982). The long term accuracy and precision of oD-H2o determination are indicated by the analyses of water reference samples shown in Table 1. The oD-H2o value obtained here for NBS-1 by comparison with TS-A is essentially identical to that listed by Gonfiantini (1978). The GISP oD value of -189.4/oo determined by comparison with TS-C in this study is within analytical uncertainty of the value (-189.8/oo) reported for GISP by Gonfiantini (1981). The water reference sample V-SMOW was prepared from distilled Pacific Ocean water (Gonfiantini 1978) such that its oD would match the SMOW standard proposed by Craig (1961) as closely as possible. According to Hagema nn et al. (1970) V-SMOW is about -0.1/oo (relative to SMOW), whereas Coplen and Clayton (1973) obtained a oD of -1.2 0.4/oo (relative to SMOW) for V-SMOW and stated that the reason for the discrepancy is not known. Due to this apparent uncertainty in the oD of V-SMOW, the good agreement of TS-A and TS-C with NBS-1 and GISP, respectively, and the uncertainty (o = 1-2/oo) involved with determining oD-H2o, I have elected not to redefine the oD of TS-A such that the oD of V-SMOW

PAGE 42

32 0 0 measured here (-1 6 loo; Table 1 ) becomes 0 /oo. Instead, the 60 of water samples analyzed by comparison with TS-A and with a 60 near zero were corrected by adding 1 .6/oo to their measured 60. Possible errors associated with this method of correction are relatively s mall and have no significant effect on any interpretations made in this study. Determination of Ancillary Parameters Preparation of carbon dioxide gas for analysis by IRMS invol ved cryogenic separation of the co2 from the other gases. This was accomplished by injecting an aliquot of the sample gas through a rubber septum into a vacuum line with a trap cooled by liquid nitrogen (LN2 ) and then simply pumping away the LN2 non -condensable gases (02 N 2 CH4 ) Water vapor was then removed cryogenically (-89C) from the co2 Sedimentary organic matter was prepared for 6 13c analysis by IRMS in a Craig type combustion apparatus (Craig 1953). The methane and carbon dioxide concentrations (mol%) of the gas samp les were calculated from the GC-integration units (HP-3390A Integrator) following application of appropriate T. C. Weight Factors as given by McNair and Bonelli (1969).

PAGE 43

31 ISOTOPIC COMPOSITIONS OF METHANES FROM FRESHWATER SEDIMENTS Biogenic methane is a common constituent of anoxic sediments of freshwater environments such as lakes and swamps when large quantities of organic matter are vigorously decomposed (Figure 2) When methane production rates are high enough to supersaturate the pore waters of shallow sediments, bubbles may form and be released to the overlying water and then to the atmosphere. Few measurements of the isotopic compositi on of biogenic methane formed in freshwater sediments are available in the literature. Oana and Deevey (1960) determined the o1 3c of methane and carbon dioxide obtained in the gaseous state following agitation of the sediments of several Connecticut lakes. The methane was 13c-depleted (6 1 3c-cH 4 ranged from -80 to -57/oo) relative to both the gaseous carbon dioxide (o13c-co ranged from -25 to -4/oo) and the sedimentary organic matter 2 (6 1 3c -soM ranged from 33 to -27/oo; Oana and Deevey 1960). Stevens and Rust (1982) reported that methane collected from the bottom of a small c 3-plant dominat ed slough in Northern Illinois had a o1 3c of -57.8 to -54.7/oo, whereas that collected at the surface was somewhat less 13C-depleted (o13c -cH4 ranged from -51.3 to -49.4/oo) implying that there was partial oxidation of the methane during its transfer from the bottom to the surface. Fuex ( 1 977) measured 17 methane samples obtained from marsh sediments in Minnesota and reported o1 3c 0 ranging from -70.2 to 55 6 loo. Meth ane collected by agitating the

PAGE 44

sediments of Wurmsee, a shallow West German lake, was 13c-depleted (o 1 3c-cH 4 ranged from -64 to -52/oo) relative to the coexisting carbon dioxide gas (o 13c-co2 ranged from -21 to -3/oo) and was D-depleted (oD-CH4 ranged from -338 to -294/oo) relative to the pore water 0 (oD-H2 o ranged from -28 to -20 loo; Woltemate et al. 1984). In this section, measurements of oD-CH4 oD-H2o, o 13 c-CH 4 o13c-co2 and ancillary parameters from several freshwater environments and from a few locations within the Tampa Bay estuary are presented. Using these measurements and the model proposed by Woltemate et al. (1984), the relative importance of the two primary methanogenic pathways, acetate dissimilation and co 2 reduction, to methane production in these systems can be estimated. Sample Collection Gas samples were obtained by agitating the sediment with a rod and funneling the released gases into 125ml serum bottles (Wheaton Scientific, Millville, NJ) that were initially filled with lake water. The bottles were filled as completely as possible with gas and stoppered with a black rubber stopper (no. 2048-11800, Bellco Glass, Inc., Vineland, NJ) that was then secured with an aluminum crimp seal (Wheaton). The samples were stored on ice during transit back to the laboratory where they were kept in a freezer until analysis. Water samples for oD determination were collected in screw cap vials, the mouths of which were covered with parafilm to retard evaporation. Sediment samples were obtained by scraping surficial material into Whirlpak plastic bags and were frozen until analysis. Model of Biogenic Methane Formation Figure 4 is a model consisting of lines that describe predicted

PAGE 45

-0 ?fl. -J: () I 0 .0 -200 -100 Figure 4. Model of methane formation. Lines represent the percentage mix of acetate dissimilation (AD) and carbon dioxide reduction (R) (after Wolt e mate et al. 1 984) 35

PAGE 46

36 cSD-H20/ OD-CH4 isotopic pairs resulting from varying the relativ e contributions to methane production of the acetate dissimilation and co2 reduction pathways. This model was proposed by Woltemate et al. (1984) and used in that study to estimate that methyl group transfer (from acetate or other methyl group donors such as methanol) was responsible for about 76% of the total methane production in the sediments of Wurmsee, a shallow lake near Hannover, FRG. Lovley and Klug (1983) reported that methanol and methylamines were the precursors for les s than 5 and 1%, respectively, of total meth ane production in the sediments of eutrophic Lake Wintergreen, Michigan. The likely explanation for this is the low abundance of methanol and methylamine precursors relative to H 2 and acetate precursors in the organic matter input to the sediments (Lovley and Klug 1983) Low rates of methane productio n from methanol and methylamines in near-surface, sulfate-rich marine sediments have been reporte d recently (King et al. 1983; King 1984a) Although methane production from these "noncompeti ti ve11 s ub strates (King 198 4a) is of interest for several reasons, I would argue that it is likely to generate large quantities of methane (e.g., a significant fraction of that r equired to achieve saturation) only under r elatively unusual conditions. Many marine organisms use m ethylamines or methylamine precursors (King 1984b) in osmoregu lation (Yancey et al. 198 2). In response to salt stress, the halophyte Spartina alterniflora concentrated up to 30% of its total leaf nitrogen in glycine betaine ( Cavalieri and Huang 1981), which can b e metabolized t o acetate and trimethylamine in marine sediments (King 1984b). Oremland et al. ( 198 2 ) reported that metabolism of methanol and trimethylamine could account for the bulk of meth ane produce d in salt

PAGE 47

37 marsh sediments containing 8ml of methane per liter of wet sediment. experiments that yielded these results were conducted in flasks containing 60ml of San Francisco Bay water, 40ml of sediment, and 10ml of homogenized Spartina foliosa materials (Oremland et al. 1982). Although known (macroalgae; Blunden et al. ( 1982)) and potential (mangroves, seagrasses) contributors of methylamine precursors (unknown concentrations) account for some organic input, sediments as rich in organic matter as those used above (Oremland et al. 1982) are rarely encountered (Sackett et al. in press) in the Tampa Bay estuary. From all of the preceding, I conclude that methane production from methanol and methylamines is quantitatively unimportant in the samples analyzed during this study; therefore, in applying the model (Woltemate et al. 1984) to the data presented in this section, I consider acetate to be by far the major source of methyl groups to the methanogens active in these sediments. In constructing this model, Woltemate et al. (1984) assumed that for the co2 reduction pathway all four methane hydrogens are supplied by the environmental water (Schoell 1980; Daniels et al. 1980), and that for the acetate dissimilation pathway only one hydrogen comes from water with the remaining three coming from the methyl group of acetate (Pine and Barker 1 956) As a result, the slopes o f the 0:100 (acetate dissimilation: co2 reduction) and the 100:0 lines are 1 and 0 .25, respectively. The slopes of the intermediate lines are obtained by multiplying the relative amount of each pathway by the appropriate slope (1 or 0.25) and summing. For example, the slope of the 80:20 line would be : 0 .80(0.25) + 0.20(1) = 0.4. They-intercept of the 80:20 line was obtained from the sewage sludge incubation experiment of

PAGE 48

38 Schoell (1980). In that experiment, aliquots of sewage sludge were incubated in plastic.bottles spiked with different amounts of 0 20. The results indicated a linear correlation between 6D-H20 and 6D-CH4 that was described (Schoell 1980) by the equation: OD-CH4 0 323 /oo (4). The y -intercept of the 0:100 line resulted from the observation that OD-H2 0/6D-CH 4 pairs obtained from the measurement of natural samples, in which methane was presumably formed via co2 reduction (Schoell 1980) fit the relationship (Schoell 1980; Nakai et al. 1974): 6D-CH4 (5). The equations of the remaining lines (Woltemate et al. 1984) are obtaine d by assuming that all of the lines pass through the point of intersection of equations (4) and (5). Equations (4) and (5) are based o n analyses of meth ane formed under widely different env ironmental conditions. Variations in the magnitude of isotopic fractionation, due to variation in param eters such as rate of reaction and time (degree of isotopic equilibration), may introduce error into predictions of r e lative methanog enic pathway selection obtained with the model It appears, however that the model yields predictions that are generally valid (e.g., Woltemate e t al. 1984), and that application of the model to data presented in this study should yield useful interpretation. Results and Discussion Measu r e ments of oD-H2 0/oD-C H 4 pairs from various freshwater and estuarine sediments presented in Figure 5 imply that about 5 0 to 80% of the total methane production in these sediments can be attributed to acetate dissimilation. This agrees with some earlier studies (Cappenbe r g and Prins 1974; Winfr ey and Zeikus 1979; Lovley and Klug

PAGE 49

-0 ?fl. -.::t I () I 0 .0 -250 .& .& -270 -290 ... .& -* - - -330 -40 -20 0 20 Figure 5. Plot of methane hydrogen isotopic composition versus environmental water hydrogen isotopic composition for several freshwater environments and the Tampa Bay estuary. The lines are the same as in Figure 4. Symbols: Lake, FL.; Mirror Lake, FL; Dias, FL; Kilmer Pond, SC; Mississippi River Delta, LA: and Tampa Bay estuary, FL. 39

PAGE 50

1982; Woltemate et al. 1984), but is in conflict with others that .indicated co2 reduction dominates methane production in other freshwater lake sediments (Belyaev et al. 1975; Jones et al. 1982) 40 The variation in pathway importance between individual lakes and even between different locations within the same lake (Figure 5) may be attributable to natural qualitative and quantitative variations in the organic matter and bacteria occurring in these sediments (i.e., Belyaev et al. 1975). The predominance of the acetate dissimilation pathway under nearshore marine-estuarine (S = 6-34/oo; Table 2) conditions indicated in Figure 5 was also found in other nearshore-marine circumstances (Sansone and Martens 1981; Mountfort et al. 1980). In contrast, the biogenic methane found in deep-sea sediments is thought to originate almost exclusively from co2 reduction (Schoell 1 980; Claypool and Kaplan 1974) and is isotopically distinguishable (oD = -170 to -190/oo; Schoell 1983) from the marine methane measured in this study (Figure 5). The 6D-CH4 values reported here support the idea that with respect to methane production, highly productive brackish and marine sediments can bear a greater resemblance to freshwater than deep-sea sediments if sulfate depleted near the sediment-water interface due to high rates of respiration (Rudd and Taylor 1980; King 1983). As discussed previously, partial oxidation of a quantity of methane, mediated by aerobic bacteria, can leave the remaining methane markedly less 13c-and D-depleted (Coleman et al. 1981). If the percentage of methane oxidation and the degree of isotopic fractionation varied depending upon location within a given freshwater lake, for example, isotopic variation (Figure 5) could occur. The

PAGE 51

1 3 and ancillary data for several freshwater environments and Table 2. oD-CH4 o C-cH4 the Tampa Bay estuary. LOCATION/ oD (0/oo) o13c (0 /oo) r.c Sed. Cone. (mol % ) DATE u CH4 H 2 0 CH4 C02 SOMa SOM T(C) aCb aDC CH4 C02 A 08/3l/84 1 297:!:1. 0 -7:t0. 9 -61. 3:!:0.1 -7.2 31 1.058 1.413 59.8 6.5 It 3 304.5 -58.8:!:0.1 -6. 9 1.061 1.427 55.4 2 6 10/02/84 1 -316:!:0.6 '-58.8.1 -11.6 26 1.050 45.9 2.2 11/17/84 1 -298 -62.0 -8.1 22 1.057 50.5 2.9 12/21/84 1 -296 -11:!:0. 7 -62.9 -8. 8 23 1.058 1.405 46.2 2.5 ... 2 -267:!:0.4 -75.0:!:0.3 -7.2 1.073 1.349 50.9 7.5 01/23/85 1 -10:!:0.5 -27.7 3.3 16 2 -263 -71. 4 -5.0 -27.9 3.5 1.072 1.343 59.1 4 3 B 12/2l/84 1 -285 3:!:0.4 55 1 -2.2 21 1.056 1.394 85.3 7.2 II 2 -314 -52.2 -6.8 1.048 1.453 85.8 2 9 01/23/85 1 -292 -2:!:0.2 -56.4 -10.8 -22. 5 0.8 14 1.048 1.410 65.0 .1.4 2 -286 -55.8 -9.0 1.050 1.398 77.0 3.5 c 12/25/84 1 -292 +5:!:0.2 -60.3 -3.7 -27.5 11.1 21 1 060 1.419 60.8 8.4 II 2 -256 -66.7 -3.7 -28.0 33.6 1.068 1.351 49.0 13.1 It 3 -265:!:0.5 -64.8 -2. 4 -26.9 0.5 1.067 1.367 74.4 8.0 D 04/23/82 2 -335 -48:!:0.2 -59.9:!:0.1 1.432 63.9 1.2 5 346 -60.0 1.456 61.0 0.6 II 6 -342:!:0 5 -60.4. 1 -12.2 1.051 1.447 69.4 3.6 E 10/14/84 1 -315 -53.6 -10.9 24 1.045 41.0 2.8 II 2 -319 -51.5 -12.0 1.042 34. 0 1.9 12/12/84 1 -311 -16:!:0.6 -55.3 9 1.428 33.6 0.7

PAGE 52

Table 2 (Cont'd). LOCATION/ cSD (0/oo) 613c (0 /oo) DATE u CH4 H 2 0 CH4 C02 sm1 a F 09/16/84 1 -323:!:1. 2 +11 70 .8:!:0.1 10/11/84 2 -296 +11 -63.3 12/03/84 3 288 +7:!:0.4 -64.7 23 2 12/03/84 4 -291 0:!:0.2 -63.8 TABLE LEGEND Lake, 12ha, fresh, St. Petersburg (SP), FL Mirror Lake, 7ha, fresh, SP, FL f -Lake Dias, 300ha, fresh, Volusia County, FL -Mississippi River Delta, fresh, LA Pond, fresh, near Columbia, S.C. -Tampa Bay estuary, FL 0 1-Ft. DeSoto Pk., S = /oo 2 Bishop Harbor, S = 320/oo 3 Lake Seminole, S = 20 /oo 4 -Feather Sound S = 6/oo % C SOM 1.6 Sed. Cone (mol%) T("C) aCb aDC CH4 29 1.493 54.9 28 1.436 79.3 24 1.414 70.7 21 1.4 10 74.1 a SOM = Sedimentary organic matter b aC = c -aD = 13 (C02 /cH 4 ) a 6 C02 + 1000 13 6 C-CH4 +1000 (H20/CH4) = 6 D-H20 + 1000 6 D -CH4 + 1000 co2 2.0 13.5 7.5' 6.9 +:"" [\)

PAGE 53

43 generally negative correlation between o 1 3c-cH 4 and oD-CH4 indicated in Flgure 6 implies that the isotopic variation is not solely due to methane consumption. The negative correlation does not mean that these samples were totally unaffected by in situ methane oxidation, but it does imply that some additional process (i.e., variation in path way) contributed to the isotopic variation. Methane concentration profiles in the upper 1 0 to 1 5 em of anoxic freshwater sedi ments are generally linear, implying that production and transport are of greater importance than anaerobic methane consumption in controlling the methane distributions (Reeburgh and Heggie 1977) Methanogenic bacteria are strict anaerobes and produce methane only under highly reducing conditions (Mah 1982). Methane produced under these conditions could only be oxidized by the aerobic process if it either migrated (as a bubble or in solution) to a less reducing location or if the depth of oxygen penetration increased. A bubble rising from the depth at which it was formed would likely exit the sediments entirely, because the distance between the methane production zone and the sediment-water interface is short (50 em or less), and would not be sampled by the methods employed here. Likewise, methane that dissolved would not be sampled, and because the process of partial dissolution itself would not be expected to induce much isotopic fractionation (Bernard et al. 1976), any methane remaining in bubble form should maintain its isotopic signature. Deeper penetration of oxygen into the sediments because of a temperature-induced decrease in oxygen consumption might allow partial oxidation of trapped bubbles. This could be responsible for the slight increase (Table 2) in the isotopic compositions of methane sampled from Crescent Lake (#2) on 1/23/85

PAGE 55

45 compared to that sampled at the same location on 12/21/84. This slight difference has very little effect on any interpretations, and when considered along with the other discussion advanced above 1 argues against extensive isotopic fractionation of samples reported here (Figures 5 ,6) by methane consumption All of the isotopic and ancillary data considered in this section are listed in Table 2 Concentrations of methane and carbon dioxide ranged from 33 to 86% and from 4 to 13%1 respectively, in the samples that were analyzed. The carbon isotopic composition of sedimentary organic matter (SOM) ranged from -26.9 to -28.0/oo for all but one of the freshwater sediment samples (Table 2); this implies that higher land plants (c3 pathway) and low 13c;12 c aquatic plants (Smith and Epstein 1971) dominate organic matter input to these sediment s For instance, a n unidentified emergent aquatic macrophyte from Lake Dias (#3) yielded a o13c of -28.1/oo. A o13c of -22.5/oo was determined for the SOM of Mirror Lake (Table 2); this is substantially heavier than the other freshwater values. Possible explanations for this include a relatively greater contribution of isotopically heavy c 4 pathway plant material (i.e., domestic Bermuda grass; Smith and Epstein 1971) and utilization of isotopically heavier source carbon by the submerged aquatic plants (Osmond et al. 1981). There is no obvious relationship between either o13c-SOM or %C-SOM and o13 c-cH 4 for the samples reported here (Table 2). For example, although the o13 c-cH 4 of Crescent Lake #2 is 8-12/oo lighter than the o1 3 c-cH 4 of Crescent Lake #1 I there is no substantial difference between the two sites with respect to o1 3c-SOM or %C-SOM. Factors other than amount and o1 3c of the source organic matter, such as pathway selection (Figure 5), must

PAGE 56

46 also be involved in determining o 13c-CH 4 Fractionation factors ac(a(CO/CH4)) and aD(a(H 20/CH4)) of the freshwater samples, ca,lculated from measured isotopic compositions, ranged from 1 042 to 1 073 and from 1 343 to 1 456 respectively (Table 2). These values are similar to those determined for German lake Wurmsee (Woltemate et al. 1 984) although the ranges reported here are greater. The greater ranges (higher aC, lower aD) probably reflect the larger relative contribution of the co2 reduction pathway in some of these samples (Figure 5) compared to the Wurmsee samples (Woltemate et al. 1984) The fractionation factors calculated from typical deep -sea isotopic compositions (Rice and Claypool 1981; Claypool and Kaplan 1974; Schoell 1980) are significantly larger (aC-1 08) and smaller (aD1.25) than the factors listed in Table 2; this implies that both isotopes are affected by methanogenic pathway and indicates the potential applicability of these parameters (particularly together) to the genetic characterization of biogenic methane deposits. As discussed earlier, recent studies (Craig and Chou 1982; Khalil and Rasmussen 1982) indicated a doubling of atmospheric methane concentrations during the past few hundred years. Furthermore, during the last 3 to 4 years, atmospheric methane concentrations have increased at rates of 1 to 1 .9% per year (Khalil and Rasmussen 1983) Decomposition of organic matter in water-covered soils, intestinal fermentation in ruminants, biomass burning, and direct anthropogenic input (i.e. leakage of fossil fuels) are thought to be the major sources of methane to the atmosphere contributing approximately 46, 22, 5, and 7% of the total flux, respectively (Khalil and Rasmussen 1983) The major sink of atmospheric methane is thought to be reaction with OH

PAGE 57

47 radicals in the troposphere (Khalil and Ras mussen 1983). A recent study by Harriss et al. ( 1982) indicated that periodically inundated soils may also be a site of atmospheric methane removal during dry periods. Recent measurements indicate that the o1 3c of atmospheric methane is about -47. 0 0 .3/oo (Stevens and Rust 1982). The average isotopic 0 fractionation associated with the sink process i s -2.5 1 5 /oo, and there is a +0.3/oo isotope effect resulting from the nonsteady state increasing methane concentrations (Stevens and Rust 1982). This implies that the average o13c -CH4 for all sources is -49. 2 1 .5/oo; this is significantly less 13 c -depleted than most known biogenic methane sources (Stevens and Rust 1982), and is about 1 o0 /oo less negative than the average o13c-cH 4 of all sources (-61 0 3 /oo) calculated by Senum and Gaffney ( 1985). Biomass burning and fossil fuels are the on l y known significant methane sources mor e 13c-enriched than -49/oo (Stevens and Rust 1982), but they apparently account for a relatively small fraction of the total input. Bubble ebullition is the dominant mode of methane input to the atmosphere from some shallow freshwater (Robertson 1979; Cicerone and Shetter 1981) and marine (Martens and Klump 1980) environments. All of the gas samp les analyzed for this section were collected as bubbles from sediments covered by very shallow (1m or less) water. Three of the freshwater lakes (Crescent Lake, Mirror Lake and Kilmer Pond) hav e been observed to release gas as bubbles from their sediments and it is very likely that active bubbling also occurs in the other environments. There is a good deal of variation ( -24 /oo) in the o 13 c -cH4 data presented in Table 2; however, most of the samples are substantially

PAGE 58

48 lighter than -49/oo, in agreement with other data from shallow aquatic environments (Stevens and Rust 1982). Though limited in number, these data imply that ebullition from these sediments is unlikely to provide an atmospheric input of methane as 13c-enriched as the source average. Based on the results of a study by Rust (1981) which demonstrated that the o 13 c of methane produced by ruminants depends on the of the plants in their diet, Stevens and Rust ( 1982 ) proposed that the anaerobic decomposition of isotopically heavy c 4 plant debris in large wetlands, such as the Sudd marshes and the Florida Everglades, might be a significant source of isotopically heavy methane to the atmosphere. As mentioned before, no such simple relationship between the o 1 3c-cH 4 and of aquatic sediments is indicated by the data presented in Table 2. According to Wolfe (1979), four well-defined groups of bacteria are involved in the conversion of complex organic matter to methane in sediments, whereas in the rumen only two of the groups are involved and there is no significant production of methane from acetate in the rumen. Also, substrate input, product output, and environmental conditions are more closely controlled in the rumen than in sediments (Wolin 1982). Thus, in addition to other factors such as pathway selection and possibly kinetic effects are probably important in setting the o 13 c-cH 4 of aquatic sediments. Escape of dissolved methane across the air-water interface, rather than ebullition, may be the more likely means of tranferring isotopically heavy methane from shallow aquatic environments to the atmosphere. Dissolved methane should be more readily available for aerobic methane oxidation (Rudd and Taylor 1980) and isotopic fractionation (Coleman et al. 1981); however, assessment of the potential contributions of ebullition and

PAGE 59

dissolved methane to atmospheric 613c-cH4 is hindered by a lack of relevant data. Using data obtained in 1960 and 1970 (Bainbridge et al. 1961; Begemann and Friedman 1968; Ehhalt 1973), Senum and Gaffney (1985) noted an apparent decrease in atmospheric methane 60 with time that they extrapolated to the present to yield an estimate of -104 4/oo for the 60 of present-day atmospheric methane. The extrapolated 60 value was corrected for kinetic isotopic fractionation resulting from the reaction of methane with hydroxyl radicals (the magnitude of the fractionation was attributed to Gorden and Mulac 1975) in the troposphere to yield an estimate of -322/oo as the present-day average 60 of all methane sources to the atmosphere (Senum and Gaffney 1985) Analyses reported in Table 2 and by Woltemate et al. (1984) indicate that methane as O-depleted as -322/ oo is produced in some freshwater sedi ments Organic matter decomposition in sewage sludge may also yield methane as O -depleted as 322 / oo (Schoell 1980 ) According to the genetic characterization model of Schoell (1983), which is based on approximately 500 natural gas analyses, thermogenic methane is generally more 0-enriched than -300/oo. To my knowledge, there are no estimates of the 60 of methane produced from either the intestinal fermentation in ruminants or biomass burning. Although an important source of atmospheric methane with a 60 near the estimated source average (Senum and Gaffney 1985 ) has been identified (freshwater sediments) there are presently too few measurements of the 60 of atmospheric methane and its sources available to determine whether or not this apparent agreement has any real significance in regard to evaluating the role of shallow aquatic sediments in the atmospheric

PAGE 60

methane budget and in regard to determining the 6D of atmospheric methane. 50

PAGE 61

5 1 METHANE ISOTOPIC COMPOSITIONS IN CAPE LOOKOUT BIGHT SEDIMENTS Cape Lookout Bight is a small, partially enclosed marine basin located on the Outer Banks of North Carolina (Figure 7). The b ight interior is somewhat protected from high energy winds a nd waves, and serves as an efficient trap of fine-grained sediments and organic debris that exit the biologically productive barrier island lagoonal system to the north through Barden's Inlet in response to tidal flushing and winter storms (Martens and Klump 1980, 1984) Organic matter in bight sediments comes f rom plankton, salt marsh and seagrass detritus eroding shoreline peat deposits (Martens and Klump 1 984) and terrigenous materials washed into t h e barrier isl and lagoonal system by freshwater runoff. Rapid remineralization of organic carbon in the rapidly accu mulating (8.4 to 11.8 em/yr ; Chanton et al. 1983), organic-rich (3.3 t o 5% organic carbon) sediments of the b ight interior (Station A-1; Figure 7) creates anoxic conditions at or near (0.5 to 1.0 em s ubbottom) the sediment-water interface (Mar tens and Klump 1 984) The a noxic conditions prevent inhabitation of the sediments by macrobenthic infauna (Menzies et al. 1968), which pump sulfate-rich seawater into their burrows (Martens 1976) except during winter months (Bartlett 1981 as cited by Chanton et al. 1983; Martens and Klump 1984). As a result, dissolved sulfate is quickly depleted (about 1 mM) and meth a ne is produced rapidly enough to supersaturate the interstitial waters

PAGE 62

CAPE 38' N \ 1000 M ATLANTIC OCEAN Figure 7 Cape Lookou t Bight, North Carolina USA. The sam p ling s ite was Stati o n A 1 (aft e r Martens an d Klump 1980). 52

PAGE 63

53 near (about 8-15 em subbottom) the sediment-water interface during warm (16-29C) summer months (late May to mid November; Martens and Klump 1980; Sansone and Martens 1982; Kipphut and Martens 1982) Reduction of hydrostatic pressure associated with ebb tide (tidal range about 1 m) triggers expansion and upward migration of trapped gas bubbles (82 to 90% methane, mean 86%) through sedimentary "bubble tubes" (Martens 1976) to the water co lumn and finally, to the troposphere (Martens and Klump 1980). An estimated 70 to 80% of the yearly flux of methane ejected to the atmosphere from the Cape Lookout Bight system comes from direct bubble transport, with the remaining 20 to 30% contributed by escape of dissolved methane from the waters across the air-sea interface (Martens and Klump 1980). Blair et al. (1984) reported that methane bubbles produced in bight sediments display a seasonal o 1 3c variation (o13 c-cH 4 ranging between -64 and -58/oo), with ebullition of the more 13c-enriched methane o ccurring during the summer when bubbling rates are maxima l In this section, measurements of the oD and o13c of methane gas samples from Cape Lookout Bight sediments collected on fifteen different dates from 1 6 June 1983 to 17 November 1984, and oD measurements of three pore water samples from various times and depths are presented. All of the samples described here were collected by the research group "CH4Ao2sn under the direction of Dr. Christopher S. Martens of the University of North Carolina, Chapel Hill. These measurements provide an opportunity to co mpare the estimates of relative methanogenic pathway importance obtained with the Woltemate et al. (1984) model to those derived from 14c labeling studies reported by Sansone and Martens ( 1981) and Crill and Martens (in press).

PAGE 64

54 Further, the large range of temperatures (7 to 29C) at which these samples were collected allows assessment of seasonal influences on methanogenic pathway selection. Sample Collection Trapped gas bubbles were displaced by physical disturbance from different depths within the upper 20 to 40 em of the sediments into a large inverted cone placed about 1 m above the bottom (water depth about 8 m), and transferred at depth from the cone to 125 ml (except for the 19 September 1984 samples (30 ml)) Wheaton serum bottles that were initially filled with water. The serum bottles were filled as completely as possible with gas, stoppered, and c rim ped with an aluminum seal. During sample collection, gas was continuously released from the sediment to the cone as gas was withdrawn from the cone. As a result, replicate samples, particularly during summer whe n up to five 125 ml samples were taken, did not all come from exactly the same aliquot of sedimentary gas. The samples were kept in a freezer until analysis, except during transit between North Carolina and Florida. Results and Discussion Methane isotopic compositions over the course of a year and sediment temperatures from several years (Martens and Klump 1980; Martens et al. 1980; Crill and Martens 1983; Martens unpublished data) are plotted versus sampling date in Figure 8. The data (Figure 8, Table 3) indicate great isotopic variability among replicate sample bottles during the summer, and indicate that oD-CH4 and o13c-cH 4 vary in an opposite manner in response to change in season, with methane more D-depleted and more 13c-enriched (also Blair et al. 1984) during 0 summer months (July to September; mean oD-CH4 = 269 /oo, std. dev. =

PAGE 65

30 ,..... (.) 0 1-w 20 za: w :::> :::E 1-<( 8 ffi10 (/) (L :::E w 1-cfi -...; :-250 (.) I 0 .0 -270 ,..... 0 cfi -...; ' .. l res> (2) . J A 0 D F BOTTLE +: 2 3 4 + 5 ... t... .... J. ......... '\ ...... ... ...... '\. I 1 (2) (2) (2) A 1984 J (4)J: & y I A ...... -\4 ... ... 1 (2) 1(3) . 0 D Figure 8. Seasonal variations o f temperature, oD-CH4 and o13c-cH 4 in Cape Lookout Bight sediments. The temperature data are from Martens and Klump (1980), Martens et al. (1980), Crill and Martens (1983) and Martens (unpub lished). V1 V1

PAGE 66

56 Table 3 oDC H 4 o 1 3c-cH4 a nd o DH 2 0 from Cape Look o u t Bight. goc H 4 o 1 3 C C H Date Bottle IF ( /oo) (o/oo) 4 T(C) 6/16/83 1 22 7 63.5 23 2 -225 63 7 3 -227 63.8 4 -222 63 9 5 2 2 6 63 8 8/19/83 1 2 5 9 -62.0 27 2 264 -60. 5 3 2 84 -56. 8 4 2 77 -58 9 5 258 62 2 9/14/83 1 26 1 61.4 27.8 2 263 6 1 9 10/16/83 1 265 60 0 22 8 2 2 65 60.3 2/2/84 -238 64 3 7.3 2 23 1 -64.3 4/7/84 1 24 1 63 6 14 2 239 63 0 5/6/84 1 -231 -64. 0 1 8 2 233 63 9 5/31 I 8 4 23 1 -68.4 23 2 228 -68. 3 6/14/84 234 65 1 23 2 233 64. 1 7/2/84 -270 59 2 24 2 -271 59 3 7/18/84 25 1 -62.1 26 2 263 61.3 8/10/84 1 277 57 0 25 2 276 56 0 3 275 57 9 4 275 58 3 8/31/84 -278 -58.3 26 2 258 61.2 3 -281 -57.4 4 260 60 7

PAGE 67

57 Table 3 (Cont 'd). 8D-CH4 o 13c-CH Date Bottle II ( loo) (oloo) 4 T(C) 9119184 276 -58.9 26 2 -273 -59.0 1 1 I 17 I 84 250 -60. 4 1 6 2 257 -60. 5 3 -260 -61.0 4 -262 60.4 Pore wat e r
PAGE 68

o .r 13C-CH o o 9 /oo, n = 21; mean v 4 = -59.5 /oo, std. dev. = 1.9 /oo, n= 21) when bubbling rates are maximal (Martens and Klump 1980), as compared to winter months (December to April; mean oD-CH4 = -236/oo, std. dev. 0 13 0 0 = 4.4 /oo, n = 6; mean o C-CH4 = -63.9 /oo, std. dev. = 0.5 /oo, n 6) when Cape Lookout Bight sediments are not observed to bubble (Martens and Klump 1980, 1984). The indicated seasonal fluctuation of oD-CH4 (Figure 8) probably results from seasonal variation in methanogenic pathway selection, with acetate dissimilation accounting for a greater proportion of the total methane production during summer than during winter. Evaluation of these data with the Woltemate et al. (1984) model (Figure 9) implies that acetate dissimilation accounts for about 45 to 60% of the total methane production during summer months, a n d about 25 to 40% of the total during winter months Sansone and Martens (1981) measured the 1 4 production rates of carbon dioxide and methane from C-labeled acetate in the 5 em depth interval immediately below the 1 mM sulfate isopleth of bight sediments at various times of the year, and estimated that over 50% of the summertim e methane flux could be accounted for by acetate dissimilation. In contrast, measurements made from January to May indicated that the rate of methane production from acetate in this layer was near zero during the winter (Sansone and Martens 1981). Crill and Martens (in press) measured methane production rates from 14 C-labeled bicarbonate a n d acetate at various depths in Cape Lookout Bight sediments during the summer of 1983, and found that the relative importance of the acetate dissimilation pathway to total methane production, integrated over the upper 30 e m of the sediments, increased over the course of the summer from 23% on 21 July to 30% on 30 August.

PAGE 69

0 '#. :r: () I Cl .0 I -230 I -250 I I !.& --270 -290 A -. -330 -20 0 20 bD-H20 ( %o ) Figure 9 Plot of methane hydrogen isotopic composition versus pore water hydrogen isotopic composition for samples from Cape L ookout Bight and the samples plotted in Figure 5. Symbols: ecape Lookout Bight; all other the same as in Figure 5 59

PAGE 70

In the sulfate-reducing zone of bight sediments the acetate methyl is oxidized only to carbon dioxide (Sansone and Martens 1982). Although appreciable rates of methane production from carbon dioxide reduction were measured in Cape Lookout Bight sediments at all depths except the top 2 em (including some in which high rates of sulfate reduction were measured), acetate was a significant precursor of methane only in sulfate-depleted (< 1 mM) layers; further, acetate dissimilation accounted for an increasing percentage of the total methane production with increasing depth to a maximum of 48% in the 20 to 22 em layer on 7 August 1983 (Crill and Martens in press). Net 60 methane production rates as determined by the tube incubation technique of Crill and Martens (1983) were nearly equivalent to the sum of the methane production rates from 14c-labeled bicarbonate and 'acetate during the summer of 1983, implying that nearly all of the methane production in Cape Lookout Bight sediments during the summer results from carbon dioxide reduction and acetate dissimilation (Crill and Martens in press). Measurements of in situ concentrations of alternate methanogenic substrates, such as methanol and methylamines, and 14c radiotracer studies will be required for a more accurate assessment of the quantitative signficance of alternate methanogenic pathways in bight sediments. The interpretation presented here (Figure 9) that acetate dissimilation is relatively less important during the winter in Cape Lookout Bight sediments qualitatively agrees with the aforementioned results of Sansone and Martens (1981). Further, if significant methane production from acetate occurs only when sulfate concentrations are < mM (Crill and Martens in press), then acetate dissimilation would be

PAGE 71

61 expected to be of less importance during the winter because penetration of the mM sulfate isopleth is 20 to 22 em in winter and 7 to 10 em in summer (Sansone and Martens 1982). Because more than 98% of the calculated total remineralization of metabolizable organic matter takes place within three years in Cape Lookout Bight sediments (Martens and Klump 1984 ) which corresponds to about 30 em depth (Chanton et al. 1983), the microbes living at depths where acetate dissimilation to methane should be possible during winter would have to subsist on the more refractory organic matter fractions. Lowered rates of methane production from acetate during winter would be expected due to the combination of lowered temperatures (Zeikus and Winfrey 1976) and poorer nutrition. The logical prediction that acetate dissimilation is relatively less important during the winter and the general agreem ent of the estimates of summer Cape Lookout Bight methanogenic pathway selection presented in Figure 9 with those from 14c radiotracer studies (Sansone and Martens 1981; Crill and Martens in press) imply that interpretations made using the model (Woltemate et al. 1984) employed in this study are reasonable. Variation of the relative importance of acetate dissimilation with depth in'the sediment (Crill and Martens in press) is probably the factor most responsible for the great isotopic variability observed among replicate samples (Figure 8 Table 3) taken during the summer (19 August 198 3 and 31 August 1984) when bubbling rates are maximal (Martens and Klump 1980, 1984) Because the method of sample collection employed in this study did not assure that all "replicates" sampled exactly the same aliquot of sedimentary gas, any isotopic inhomogeneity (i.e., with depth) of gas bubbles trapped within the

PAGE 72

62 sediment could express itself in the replicates analyzed here. Kipphut and Martens (1982) observed 222Rn profiles in Cape Lookout Bight 222 sediments during early and late summer that exhibited a sharp Rn maximum at 6 to 8 em depth and a 222Rn deficit below. The unusual shape of the 222Rn profiles was attributed to stripping of 222Rn by methane bubbles at depth followed by partial dissolution of the bubbles at 6 to 8 em depth, particularly during times of the year when bubbling rates are relatively low and the residence time of the bubbles in the sediment should be longer (Kipphut and Martens 198 2) The results of Bernard et al. (1976) indicate that very little isotopic fractionation occurs when methane gas is partially dissolved in water; therefore, it is unlikely that variation in the extent of bubble dissolution with depth in the sediment or season would induce much isotopic variability. If all of the methane produced in Cape Lookout Bight sediments had the same isotopic composition and was then subjected to varying degrees of bacterial oxidation, depending upon season and/or depth, differential isotopic fractionation (Coleman et al. 1981) and isotopic variability could result. The zone of oxic conditions is very thin (0.5 to 1.0 em; Martens and Klump 1984) in bight sediments during the winter and essentially absent during the summer, greatly decreasing the possibility of extensive oxidation of sedimentary methane bubbles by the assimilatory, microaerophilic process (Martens and Klump 1980) The potential significance of dissimilatory anaerobic methane oxidation in Cape Lookout Bight sediments is difficult to assess as there are no direct radiotracer measurements of the process (i.e. Alperin and Reeburgh 1984) available for this system. Because appreciable anaerobic methane oxidati on in marine sediments is apparently confined

PAGE 73

63 to a thin layer at the base of the sulfate-reducing zone (Reeburgh 1980; Devol 1 983; Alper in and Reeburgh 1984; Iversen and Jorgensen 1985) methane bubbles shou l d not be subject to extensive oxidation during entrapment below the sulfate-reducing zone The linear to convex-up pore water methane concentration profiles that are observed during summer and winter (Martens and Klump 1980; Kipphut and Martens 1982) can be interpreted to mean that the rate of anaerobic methane oxidation is significantly less than the rate of methane transport out of the sediment (Martens and' Klump 1984) The negative linear correlation between Cape Lookout Bight oD-CH4 and o 13 c-cH 4 (all data): oD-CH4 = -6.1 o13C -CH4 626/oo (r = 0 .90, n = 42) (6) indicated in Figure 10 is good evidence that the isotopic variability is not simply the result of differential oxidation of a singl e isotopic end-member, as partial bacterial consumption would likely enrich the residual methane with respect to both isotopes (i. e Coleman et al. 1981). Thus it is unlikely that any secondary process such as partial dissolution or bacterial oxidation could account for all of the seasonal and replicate sample isotopic variability indicated in Figure 8 Linear regression of oD-cH4 and o13c-cH 4 data obtained during the main Cape Lookout Bight bubbling period (1 July to 1 October, sediment T 24C, Martens and Klump 1984) yields the equation: oD-CH4 528/oo (r = 0.92, n = 21) ( 7) With the exception of the 3 1 May 1 984 samples, the methane isotopic compositions of winter samples taken from February to mid -June (Table 3) deviate significantly from the line (7) derived from the summer data. Partial bacterial oxidation of trapped methane bubbles resulting

PAGE 74

0 ?fl. -'It J: 0 I 0 .0 -240 -260 -280 : bD=-4.4 b 13 C-5{8 T>24C, n=21, r=0.92 bD=-6.1 b 13 C -626 n=42, r=0.90 CAPE LOOKOUT BIGHT, N.C -68 -66 -64 -62 -60 -58 -56 Figure 10. Plo t of methane hydrogen isotopic composition versus meth a ne carbon isotopic composition for from Cape Lookout Bight. The equation: oD = -6.16 C626 is a linear regression of all the1gata considered in this study. The equation : oD = -4.4 C 528 is a linear regressi on of the data obtained for T 24C. 64

PAGE 75

65 from deeper penetration of sulfate during the winter (Sansone and Martens 1982) would be expected to shift the methane isotopic compositions in the direction required to explain the deviation. Blair et al. ( 1984) suggested that near-sur.face methane oxidation could account for the variation in pore water methane c 1 3c with depth (-61 .0 0.6 /oo at 2-7 em depth to -65 0.1/oo at 35-47 em depth) that they observed during February of 1984. The c13c-CH of the winter 4 samples discussed here range from -65.1 to -63.0/oo, implying that they originated in deeper layers where methane oxidation is probably of less significance. The disagreement of the winter isotopic compositions with equation ( 7) probably results from the greater_ relative importanc e of the carbon dioxide reduction pathway to methane production during the winter as discussed above. The biogenic methane found in deep-sea sediments, which is thought to originate almost exclusively from carbon dioxide reduction (Claypool and Kaplan 1974; Schoell 1980), is generally significantly less D-depleted (cD = -190 to -170/oo, Schoell 1983) than the methane produced during the summer in 0 Cape Lookout Bight sediments (mean cD = -269 /oo). It is reasonable to expect that methane produced by carbon dioxide reduction in Cape 0 Lookout Bight sediments would be less D-depleted than 269 /oo, although it might be more D-depleted than the methane produced in deep-sea sediments as a result of environmental differences (i.e., rate of organic matter input and remineralization, sedimentation rate, methane production rate, extent of isotopic exchange) between the two systems. The c 1 3c of methane produced by carbon dioxide reduction depends upon the c 1 3c of the substrate carbon dioxide and the degree of

PAGE 76

66 fractionation induced by the bacteria mediating the reaction. Laboratory studies (Games et al. 1978; Belyaev et al. 1983) indicate that the isotopic fractionation associated with carbon dioxide reduction decreases with increasing as would be expected for temperature-dependent isotopic equilibration of methane and co2 The results of laboratory studies of bacterial sulfate reduction (Jones and Starkey 1957; Kaplan and Rittenberg 1964) imply that the sulfur isotope effect associated with this process is dependent upon the rate of reduction, and that temperature influences fractionation only by affecting the rate of reduction. The rate of methanogenesis is known to vary greatly in response to temperature fluctuations (Zeikus and Winfrey 1976), so the isotope effect associated with co2 reduction (Games et al. 1978; Belyaev et al. 1983) may also be rate-dependent. Extrapolation of the results of these laboratory studies (Games et al. 1978; Belyaev et al. 1983), which were conducted at T 37C and may not be representative of the naturally occurring to Cape Lookout Bight winter temperatures (7 to 23C) implies that temperature (or rate) induced 613c-cH4 variations of about 10 to 30/oo are possible. Blair et al. (1984) reported that 6 13c-Eco2 in Cape Lookout Bight sediments during February of 1984 ranged from -9.5 0.1/oo in the sulfate reduction zone to +6.1 0.1/oo in the methane production zone indicating that variation in substrate isotopic composition could also be responsible for significant 613c-cH4 variations.

PAGE 77

67 METHANE ISOTOPIC COMPOSITIONS IN DEEP-SEA SEDIMENTS Methane associated with present day deep-sea sediments is formed by both microbial and thermal decomposition of buried organic matter. Microbial processes account for most of the methane production in rapidly accumulating (sedimentation rate > 10 m/10 6 years), near-surface (< 1 km depth) deep-sea sediments subjected to normal geothermal gradients ( < 50C/ km depth; Claypool and Kvenvolden 1983). Only in areas of abnormally high heat flow, such as subduction zones, active spreading centers, or near hot igneous intrusions (i. e., Schoell 1982), are near-surface deep-sea sediments likely to attain sufficiently high temperatures (80 to 1 50C) to allow significant thermal decomposition of complex organic matter to methane and other light hydrocarbons (Claypool and Kvenvolden 1983). Biogenic methane production in deep-sea sediments is thought to occur almost exclusively via co 2 reduction (Claypool and Kaplan 1974; Whiticar et al. in press). The principal evidence for this i s that deep-sea profiles of c1 3 c -cH 4 and c13c-Eco2 are generally parallel, presumably as a result of bacterial reduction of C02 with relatively constant kinetic carbon isotopic fractionation (Claypool and Kaplan 1974) Further, available deep -sea 6D-H2 0/6D-CH 4 pairs (Whiticar et al. in press) fit rather well the equation (5) mentioned earlier (Nakai et al. 197l.J; Schoell 1 980) that describes methane production from co2 As discussed previously, the slope of unity in equation (5)

PAGE 78

68 can be interpreted to mean that all four methane hydrogens are supplied by the environmental water, a condition satisfied only by co 2 reduction (Schoell 1980; Daniels et al. 1980) When methane is produced in deep:-sea sediments in sufficient quantities to s upersaturate t h e pore waters unde r app ropriate pressure-temperature conditions, it may combine with water and other gases to form gas hydrates (Claypool and Kaplan 1974) Gas hydrates are a type of clathrate compound in which gas molecules are trapped within and stabilize a framework of hydrogen bonded water m o lecules. The ice-like gas hydrates form at temperatures 10 to 15C above the freezing point of water when certain gases and water are combined at elevated pressures. With respect to pressure-temperature conditions, about 93% of the world s ocean sediments are sui table f o r methane hydrate formation (Trofimuk et al. 1973). Most currently known or suspected hydrate zones occur in continental slope regions and are correlated with bottom-simulating reflectors that result from changes in reflection properties at the gas hydrate-free gas boundary at the base of the gas hydrate zone (Shipley et al. 1979). The presence of gas hydrates in continental slope sediments along the east coast of the U S the western Gul f of Mexico and along the Pacific side of Central and South America was inferred from the presence of bottom -simulating reflectors in seismic profiles by Shipley et al. (1979). Water depths overlying the suspected hydrate zones range from 1a0 to 4,400 m with pressure-temperature conditions for hydrate stability satisfied between 100 and 1 100 m subbottom (Shipley et al. 1979). In this section, measurements of oD-CH4 o13c-cH 4 and oD-H2 0 from several Deep Sea Drilling Project (DSDP) Sites, including some from

PAGE 79

which gas hydrates were recovered, are presented. The samples described here were provided by Dr. James M. Brooks of Texas A&M University and Dr. George Claypool of the United States Geological Survey, Denver. Sample Collection Samples of gas from expansion pockets, which developed within the cores upon retrieval, were transferred to vacutainers using a needle and septum device with a valve to minimize atmospheric contamination (Baltuck et al. 1985). Interstitial waters wer e obtained by squeezing (Manheim and Sayles 1974), and collected in glass ampoules (Site 618 only), or in plastic syringes that wer e sealed with silicone caulking to minimize interaction of the sample with the atmosphere. Results and Discussion DSDP Leg 84 Middle America Tre nch Region 69 Leg 84 of the DSDP sampled landward slope sediments of the Middle America Trench off Costa Rica (Site 565) and Guatemala (Sites 566-570) Geophysical studies indicate that during the Miocene and Pliocene undeformed sediments covering the descending Cocos Plate were not scraped off during subduction but continued into the subduction zone making the Middle America Trench a tectonic type end-member (Auboin and von Huene 1985). Gas hydrates had been encountered in the continental slope sediments of this region during drilling on Legs 66 and 67 in 1979. Isotopic analyses of methane and pore water samples collected during Leg 84 are presented in Table 4. Also listed in Table 4 are the locations where gas hydrates were visually observed in Leg 8 4 sediments by Kvenvolden and McDonald (1985) The paucity and uneven geographical

PAGE 80

Table 4. Isotopic analyses of methane and water from DSDP Leg 84 (in 0/oo). Location of hydrates and% org. c data from Kvenvolden and McDonald (1985); Site locations, water depths, and sed. rates from von Huene, Aubouin et al. (1985). Sample Depth o 13c-CH (core-section) (m) 6D-CH4 oD-H2 0 4 gas gas Site 565 (86 4 'W, 0 sed. rate 0 9 4 3. 7 'N; water depth = 3111 m; (100-300 m) 125 m/m.y.; % org. C (100-300 m) .. 0.5-1. 7) 6-6 58 -192 -73. 1 7-1 59 +2 9-1 78 -1 10-5 94 +3 21-3 195 -200 0 -65.0 23-5 217 -198 -64.8 24-3 223 -193 -63.9 25-4 235 -2 26-4 243 -195 ( ) -64.0 hydrate 285 hydrate 319 34-1 320 -196 (.5) -63. 0 Site 567A (90.9'W,12.0'N; water depth= 5529m; sed. rate ---(230-360 m) = 76 m/m.y.; % erg. C (230-360m) = 0.2-1.5) 4-1 7-2 8-4 17-2 232 255 268 346 -186 -72.7 -183 -72.9 -183 ( 5) -72.9 -174 -73.4 Site 568 (90.0'W, 13.3'N; depth= 2031; sed. rate (0-200 m) -------= 117 m/m.y.,(240-400 m) = 33-66 m/m.y.; % org. C (0-180 m) > 2.6, (180-280 m) 0.6-2.6, (280-400 m) 0.6-1. 3) 2-4 9 -1 3-5 19 -193 -70. 9 16-2 140 0 (.2) 16-4 143 -187 -60.9 20-3 180 -2 (.3) 20-6 184 -187 -54.2 22-5 203 -182 -50.9 23-2 208 -191 -50.2 29-5 270 -166 -46. 4 32-6 300 -172 -46.2 34-6 319 -164 -45.3 35-4 327 -173 -44.9 37-4 346 -172 -43.2 40-3 374 -174 -43.0 42-6 398 -171 -42.3 43-4 404 -175 -43.8 gas hydrate 404 44-4 414 1 ( 1 ) 70

PAGE 81

Table (Cont'd). Sample Depth (core-section) (m) o13c-CH 4 0 0 Site 569 (90 50.4'W, 12 56.3'N; w ater depth= 2800 m ; sed. rate (100---m) = 25 m /m.y.; % org. c m) 12-6 105 -186 -80.6 13-3 115 2 15-3 129 -185 (.5) -78.8 18-2 156 -185 2 -77.9 23-3 207 -179 -73.7 219 -18 2 -71 1 25-2 -178 ( 1) -70.5 26 5 238 -18 1 -69.2 7 1 570 (91 6'W, 13,1'N; water depth= 1718 m; sed. rate (0-230 m) = 130 m /m.y., (220-260 m) = 1 3 m /m.y., (260-330 m) = 66 mlm.y.; % org. C (0-275 m) 1.1-3.2, (275 -350 m) = 0.1-3.0) 3 6 22 -19 3 -79.6 6 1 -188 -71.6 11-1 96 -190 -68.8 15-2 136 -197 (.5) -68.6 16-6 1 5 1 196 -68 2 20-1 183 -187 -65. 4 Hydrate 192 21-2 193 -189 -62 6 24-3 224 -187 -58. 7 25-4 234 -181 -55.3 26-4 245 190 53 0 Hydrate 246 Hydrate 249 Hydrate 254-268 Hydrate 273 294 -186 -41 6 30-4 284 -185 (.8) -42.5 3 1-1 289 -185 -42.8 32-4 302 -19 3 -42.4 Hydrate 303 341 318 -187 -41 6 352 329 -188 (.4) -41 9 36-1 336 -187 -40.9 Hydrate 338 37-1 346 -189 -41 6

PAGE 82

72 and depth distribution of samples available for analysis in this study make detailed interpretation difficult; however, it appears that the methane recovered from the Sites where gas hydrates were either not visually observed at all (567A, 569) or were only observed near the bottom (568) is generally slightly less D-depleted (mean CD-CH4 = 0 0 -179.5 /oo, std. dev. = 7.6 /oo, n = 24) than the methane recovered from the Sites ( 565, 570) where hydrates were visually observed (Kvenvolden and McDonald 1985) at relatively shallower subbottom depths 0 0 (mean OD-CH4 = -190.5 /oo, std. dev = 4.8 /oo, n = 24;Table 4). Claypool et al. ( 1 985) found that significantly more deuterium enrichment in methane occurs at depth at Site 568 from which gas hydrates were recovered only in the deepest part of the section penetrated, than in the methane collected at comparable depths at Site 570 where gas hydrates are massively (3 to 4 meters thick) developed at intermediate (250 m) depths. Galimov and Kvenvolden ( 1 983) suggested that, although the o13c of CH4 and co 2 in gas hydrates should not be directly influenced as a result of their hydrated state, methane in zones of gas hydrate might be slightly 1 2 c -enriched compared to methane from non-hydrated regions because escape of isotopically light methane might be inhibited due to encagement by the water-clathrate structure. If this line of reasoning is extended, it might be expected that methane sampled from gas hydrate deposits could also be relatively more D-depleted. This is apparently observed (Table 4, Claypool et al. 1985) Confinement within the clathrate structure might also reduce the magnitude of possible isotopic fractionation induced by rapid depressurization during sampling. The methane sampled from Sites 567A, 568 and 569 shows no 13c-enrichment relative to that found at Sites

PAGE 83

73 565 and 570, however, reducing the strength of any argument for light-isotope enrichment in gas hydrates that might be made from the 60 data. Further, two separate laboratory studies in which naturally formed gas hydrates were sampled i11termi t tently during controlled decomposition revealed no obvious carbon isotopic fractionation during hydrate decomposition (Brooks et al. 1983; Kvenvolden et al. 1984). Large differences in parameters such as sedimentation rate and percent organic carbon (Kvenvolden and McDonald 1985) among Leg 84 Sites are more likely causes of the slight differences in 6D-CH4 noted above. The o13 c of methane (as heavy as -41/oo) sampled in the lower parts of Sites 568 and 570 (Claypool et al. 1985; Jeffrey et al. 1985; Table 4) implies that the methane was formed from thermal decomposition of co mplex organic matter (Bernard et al. 1977). Profiles of o13 c-cH 4 and o13c-ECO are generally parallel and o13c-ECO as heavy as +35.7 2 2 and +37.5/oo were measured in Sites 568 and 570, however, implying that the anomalously 13c-enriched methane was formed from the reduction of very 1 3c-enr iched CO2 (Claypool et al. 1985). Further, the o 1 3 c difference between the anomalously heavy CH4 and co2 sampled at Sites 568 and 570 is essentially the same as f ound at other DSDP Sites where biogenic methane exhibits a more typical o13c (Claypool et al. 1985). DSDP Leg 87 Nankai Trough Site 583 of DSDP Leg 87 sampled mildly deformed sediments on the lowest structural terrace of the landward slope of the Nankai Trough, which is located off Honshu, Japan. In contrast to the Middle America Trench, most or all of the trench fill covering the descending plate is scraped off and accreted in response to the low rate of subduction (-2 em/yr.) in the Nankai Trough (Karig, Kagami et al. 1983).

PAGE 84

Isotopic analyses of methane and pore water from Site 583 are presented in Table 5. Both oD-CH4 and oD-H2 o are constant with depth, within the analytical precision of the measurements. The oD-CH4 is typical of values normally exhibited by biogenic methane f o rmed from co 2 reduction in deep-sea sediments (Schoell 1983 ) and the oD-H2 o is within the range found in other DSDP pore waters ( Lawrence and Gieskes 1981 ) Claypool et al. (in press) presented depth profiles of o13c-cH 4 and at Site 583 that exhibited similar shapes with 74 most of the variation in these parameters occurring in the upper 30 m of the sediments. They interpreted these results to mean that methane was formed via co2 reduction and that methanogeni c activity probably ceased below depths of 30 to 50 m (Claypool et al. in press) The organic carbon content of these sediments is about 0 4 to 0.6% (Claypool et al. in press), approximately the amount thought necessary to support methane production (Rice and Claypool 1981) The apparently limited extent of methane production in these sediments combined with the great water depth (4618 m) make gas hydrate formation at Site 583 unlikely (Claypool et al. in press). DSDP Leg 96 Orca Basin Orca Basin is a 400 km2 depression on the Louisiana continental slope located in a region where salt diapirism and slumping have created a complex seafloor morphology (Lehner 1969; Shokes et al. 1977). Dissolution by seawater of a near-surface salt deposit on the basin flank apparently accounts for the 200 m thick layer of anoxic, hypersaline (250 g/kg) water that occupies the bottom of the basin (Shokes et al. 1977; Wiesenburg 1980) The brine is near saturation respect to NaCl and is sulfate-enriched (47 2 mM) relative to deep

PAGE 85

7 5 Table 5. Isotopic analyses of methane and water (in /oo). from DSDP Site 583 Sample Depth 6D-CH4 6D-H2o o13c-CH (Hole-core-section) (m) 4 D-3-1 72 -189 -71 1 D-6-2 100 -186 -70.6 D-9-2 128 4 D-11-1 150 -187 -70.6 D-12-2 156 3 D-14-2 178 -188 2 -70.2 D-1 6-2 198 -190 -69.7 D-19-1 228 2 D-22-4 256 3 D-22-6 260 -188 -69.1 D-24-6 274 -185 -68.9 D-29-2 325 4 G-3 325-336 -192 -68.1 G-7-2 370 -188 -66.9 G 9-1 388 -186 -66.5 G-12 411-419 -184 -66.3 G-14 431-439 -189 -66.6 --mean -188 3 Std. dev. 2.2 0.9

PAGE 86

76 Gulf of Mexico seawater (Sheu et al. in press). The dense brine is 'highly stable and mixes very slowly with the overlying seawater. As a result, methane produced in the underlying sediments (Sackett et al. 1979) has accumulated in the brine to a concentration nearly six orders of magnitude greater than in the overlying deep Gulf seawater (Wiesenburg 1980) From consideration of seismic reflection profiles, geotechnical and clay mineralogical properties, and high sedimentation rates (-1 m/ 1000 yr.), Tompkins and Shephard (1979) concluded that slumping of basin flank sediments, both from within and from outside the area covered by brine, is the primary m e chanism of sediment accumulation in the central part of the basin, and that the amount and origin of slumped material varies spatially within the basin. Near-surface basin sediments originally deposited under anoxic conditions are enriched in total organic carbon over normal slope sediments by a factor of 2 to 3 and contain large pieces of undecomposed Sargassum (Sackett et al. 1979; Northam et al. 1981). These observations imply that organic matter in Orca Basin sediments is relatively well-preserved, as a result of reduced rates of biodegradation caused by the hypersalinity (Sackett et al. 1979) and/or the anoxic conditions (Northam et al. 1981 ) Isotopic analyses of Site 618 interstitial methane and water are presented in Table 6. Application of the Woltemate et al. (1984) model (Figure 4, equations (4) and (5)) to these data (Table 6) implies that the acetate dissimilation pathway could account for 15% of the total methane production in Orca Basin sediments. Whelan et al. (in press) report a pore water acetate concentration of about 1 mM at 15m depth

PAGE 87

77 Table 6. Isotopic analyses of methane and water from DSDP Site 618 (in /oo). Sample Depth 6D-CH!! 6D-H2 0 o13c-CH 4 (Core-section) (m) 1-2 3 5 1-3 5 -181 -7!!.9 2-4 1 2 8 3-2 19 8 3-5 2!! -186 -73.3 l!2 28 8 ll-6 3!! -185 -73.1 5-5 ll2 8 5-7 ll5 -18ll -72 8 7-1 52 7 7-3 54 -186 -72.6 8-1 58 5 8-5 6!! -1 85 -72.7 9-2 73 -18!! 5 -72.7 11-1 90 5 11-2 92 -181 -75.7 mean -18!! 7

PAGE 88

78 at Site 618. This concentration is about a factor of two higher than the Km (substrate concentration at which the initial reaction velocity is half maximal) for the acetate-decarboxylating methane-producing bacterium Methanobacterium soehngenii that was isolated from digested sewage sludge (Zehnder et al. 1980) Recently, isolation of a new acetotrophic marine methane-producing bacterium, Methanosarcina acetivorans, from methane evolving sediments was reported by Sowers et al. ( 1984). If an acetotrophic methanogen inhabits Orca Basin sediments, and if its Km is similar in magnitude to that of M. soehngenii, then some production of methane from acetate, at least at 15 m depth, may be possible. The preceding argument assumes that a significant fraction of the acetate detected is 11microbiologically available", a condition that may or may not be met. At 15 m subbottom depth at Site 618 the interstitial salinity was about 115/oo (Bouma, Coleman et al. in press). acetivorans exhibits a maximum specific growth rate at about 0.2 M NaCl concentration (about 12/oo salinity) that decreases rapidly with increase in NaCl concentration (Sowers et al. 1984) Thus, any acetotrophic methanogens living at 15 m depth may be inhibited by the high salinity. This would agree with the suggestion of Sackett et al. (1979) that the rate of overall organic matter destruction may be slowed by the hypersalini ty. Although acetate dissimilation may account for a small percentage of the methane production in Orca Basin sediments, most of the methane probably results from C02 reduction. The salinity (Bouma, Coleman et in press) and oD-H2o (Table 6) profiles for Site 618 imply that the Orca Basin brine has a 60 of about

PAGE 89

79 0 4 /oo. This is heavier than present day Gulf Basin Water (oD = 0.7 to 1.1/oo; Frank 1973) and implies that the Orca Basin brine was formed sometime between the last glacial maximum (18,000 yr. B .P.; Mcintyre et al. 1976) and the end of the subsequent deglaciation period (about 14, 000 to 5,000 yr. B.P.; Barry et al. 1979) when significant quantities of isotopically light water (Dansgaard and Tauber 1969) were tied up in continental ice sheets. It has been estimated that ocean deep waters were about 1 2 (Dansgaard and Tauber 1969) to 1 .65/oo (Shackleton 1977) 18o-enriched at the last glacial maximum relative to present day deep water. This would correspond to a deuterium enrichment of about 8 to 11/oo, assuming the deuterium-18o relationship for North Atlantic Surface Water of Craig (1967) applies. If these estimates of the deep ocean isotopic fractionation and the brine oD (4 to 5/oo) are accurate, then the age of the Orca Brine might be further constrained to the deglaciation period (14,000 to 5,000 yr. B.P.). In support of this, the estimate of Addy and Behrens (1980) of when the brine began to accumulate in the basin (about 8,000 yr. B.P.) falls within this range of time. There is, however, some uncertainty about how much sea level was actually lowered during the last glacial maximum. Blackwelder et al. ( 1979 ) estimated that substantially less ice was present from 17,000 to 10,000 yr. B.P. than would be expected from the estimates of minimum sea level employed by Dansgaard and Tauber (1969) and Shackleton (1977) The interstitial methane o13c values reported in Table 6 and by Pflaum et al. (in press) are similar to values reported by Sackett et al. (1979) for the o 1 3c of methane dissolved in the brine (-74.5 to -72/oo), but quite different from interstitial methane o 13 c (-105 to

PAGE 90

0 -85 I oo) reported by Sackett et al. ( 19 79) very light (<5 13 c < -90/oo) methane has not been widely encountered in sediments sampled by DSDP (Claypool and Kvenvolden 1983). When highly 13c-depleted methane is found, it usually occurs at the base of the sulfate-reduction zone, which is generally acknowledged to be the shallowest depth of rapid production and significant accumulation of methane in deep-sea sediments (Claypool and Kvenvolden 1983). In general, very 13c-depleted methane is probably not of great quantitative significance in deep-sea sediments and may result only under certain geochemical and/or sedimentological conditions. Because the amount and origin of slumped sediments varies spatially within the basin (Tompkins and Shephard 1979), the sediments cored by Sackett et al. ( 1 979) could be quite different from those cored at Site 618. Sackett et al. ( 1979) gave a water depth of 2340 m for their core location; this depth is 80 m more shallow than at Site 618 and may actually be on the basin flank. The probable sediment variability and all of the i sotopic data imply that the very light sedimentary methane sampled by Sackett et al. (1979) is probably a localized, near-surface occurrence that is not isotopically representative of the methane component of either the Orca Basin brine or pore waters. Between the brine-sediment interface and about 3 0 m subbottom depth the interstitial salinity decreases from brine levels to values more characteristic of normal marine sediments (Bouma, Coleman et al. in press). Rapid methane production in sediment layers less influenced by high salinity effects and subsequent upward diffusion probably provides most of the methane dissolved in the brine. Pflaum et al. (in press) observed small, spherical gas hydrates 80

PAGE 91

between 20 and 40m depth at Site 618. The methane that was released upon decomposition of the hydrate had the same 6 13c as methane sampled from gas pockets in the same core section, indicating that no isotopic fractionation of meth ane was associated with formation of the hydrates (Pflaum et al. in press). Deep-sea methane isotopic compositions (mean 6D-CH4 = -185/oo, std. 0 13 0 0 dev. = 7 /oo, n = 75; o C -CH4 = -71.3 /oo, std. dev. = 6.3 /oo, n = 44) determined in this study (Tables 4-6) are, for the most part, within the narrow ranges (6D = -190 to -170/oo; 6 13c = -75 to -60/oo) considered by Schoell (1983) to be characteristic of methane formed by co2 reduction in deep-sea sediments. The mean 6 13c was calculated excluding all data from Sites 568 and 570 because the anomalously heavy methane from those Sites constitutes a relatively much larger fraction of this data set than of the data set containing all known DSDP o13C-CH4 values. The pore waters measured in this study are somewhat more D-enriched (6D = +3/oo, std. dev. = 3/oo, n 0 waters from 24 DSDP Sites (6D = -1.8 /oo, std. dev 26) than pore 0 5 /oo, n 165; Lawrence and Gieskes 1981; Friedman and Hardcastle 1973; Jenden and Gieskes 1 983; Lawrence unpublished data) with a wide geographic distribution. If co2 reduction is by far the predominant source of methane in deep-sea sediments, as proposed by C l ayp o o l and Kaplan 0 (1974) then the mean 6D values for methan e ( 185 /oo) and pore water (6D = -1/oo, n = 191) discussed here indicate that equation (5) describing co2 reduction (Woltemate et al. 1984 ) should be modi fied to: 6D-CH4 0 1 8 4 ( 1 5 ) I oo (8) Alternatively, deviation of deep-sea 6D-CH 4 from equation (5) could be explained by minor production (about 10% of the total) of 81

PAGE 92

82 methane from acetate as discussed previously for DSDP Site 618 Few determinations of acetate concentration in deep-sea pore waters have been made to date. Acetate concentrations that are available for typical sulfate-depleted deep-sea sediments (DSDP Site 619 -Mississippi Fan) range from 21 to 131 )..lM (Whelan et al. in press). Significant (i.e., 10% of the total) methane production from acetate concentrations as low as 100 )..lM would require that the rate constant for acetate conversion to methane be much higher than the rate constant for co2 reduction because deep-sea pore water rco2 concentrations are typically two to three orders of magnitude ( 10 to 100 mM) higher (Claypool a nd Kvenvolden 19. 83; Claypool et al. 1985). Baker et al. (1977) determined the oD of a few biogenic methane samples from DSDP Leg 41 (eastern tropical Atlantic) and obtained values that were almost identical to oD-CH4 from DSDP Site 147 (Cariaco Trench) measured by Lyon (1974) Based on this similarity, Baker et al. (1977) suggested that oD-CH4 in deep-sea pore waters is geographically invariant and proposed that this could result if the deuterium concentration of deep-sea pore waters is uniform and either the isotopic fractionation associated with bacterial methane production is fairly constant or if methane and the interstitial water undergo postgenerative isotopic equilibration. Using available temperature-dependent equilibrium isotope fractionation factors and average oD-CH4 (-178/oo) and oD-H2o (+8/oo; Friedman and Hardcastle 1973 ) Lyon (1974) calculated an equilibrium temperature for Cariaco Trench (Site 147) sediments of 23C, which is about 40C lower than the estimated true temperature. 0 Average oD-CH4 (-185 /oo) and oD-H2o ( +3 /oo) obtained in this study also indicate deep-sea sediment

PAGE 93

83 temperatures of < 20C if evaluated with the equilibrium fractionation factors that were compiled by Friedman and O 'Neil (1977). Further, Nakai et al. ( 1974) found that observed differences in cSD values between methane and water from biogenic gas deposits in Japan were greater by about 50 to 80/oo than would be predicted for isotopic equilibrium. These results imply that the large differences in cSD between methane and water observed in these environments is due to a large kinetic isotope effect associated with bacterial methane formation as proposed by Nakai et al. (1974) Reevaluation of the deep-sea cSD data with recently published equilibrium fractionation factors (Lyon and Hulston 1984), however, yields predicted temperatures that are generally within about 5C of estimated environmental temperatures (Table 7). Thus, low-temperature (0 to 20C) isotopic equilibration between methane and interstitial water during the great length of time (10 5 to 106 years) represented by a deep-sea core may play an important role in determining the cSD of methane associated with deep-sea sediments. In contrast, equilibrium temperatures predicted with cSD data from freshwater and nearshore marine-estuarine sediments are substantially lower than measured temperatures (Table 7), implying that low-temperature isotopic equilibration does not occur on the short time scale represented by these young sediments.

PAGE 94

84 Table 7. Comparison of isotopic equilibrium temperatures (T ) as calculated with the data of Lyon and Hulston (1984)eq. with estimated environmental temperatures (T ). env. Location oD-CH4 oD-H2 0 T eq T env. M T eq. -Ten v. ) ( 0/oo) ( 0/oo) (OC) (OC) (oC) Deep-sea (DSDP) -178a +8b 147 (Cariaco Trench) 12 17 -5 565 (below 200 m) -196 0 -1 5 -6 568 (140-180 m) -190 -1 11 6 5 568 (400 m) -174 +1 25 24 1 569 (100-150 m) -185 +2 9 7 2 618 -184 +7 5 6 -1 Cape Lookout Bight 16 June 1983 115 -222 +3 -33 23 -56 19 August 1983 113 -284 +3 -82 27 -109 2 Fevruary 1984 112 -231 +3 -42 7 -49 Crescent Lake, FL 31 August 1984 112 -304 -7 -92 31 -123 23 January 1985 #2 -263 -10 -61 16 -77 a-from Lyon (1974) b from Friedman and Hardcastle ( 1973)

PAGE 95

85 SUMMARY Methanogenesis, the final step in the anaerobic decomposition of complex organic matter in natural systems, generally occurs via one of two pathways, co2 reduction or acetate dissimilation. Most 14c radiotracer studies (i. e., Lovley and Klug 1982) performed to date have yielded results indicating acetate dissimilation is the dominant methanogenic pathway in freshwater sediments, accounting for 60 to 70% of the total methane production. In contrast, methane production in deep-sea sediments is generally thought to occur almost exclusively via co2 reduction (i.e., Claypool and Kaplan 1974): The data generated in this and other studies (Woltemate et al. 1984; Whiticar et al. in press) indicate that biogenic methane produced in young, freshwater sediments can generally be distinguished from the biogenic methane produced in deep-sea sediments by measurement of oD-CH4 implying that oD-CH4 may be of us e in the genetic characterization of biogenic methanes (Schoell 1 983) and in estimating the relative importance of the two meth anogenic pathways in nature (Woltemate et al. 1984) The isotopic compositions of methane samples collected at eight different DSDP Sites (mea n oD = -185/oo, std. d e v = 7/oo, n = 75; mean o 1 3c = -71.3/oo, std. dev. = 6 .3/oo, n = 44) are generally typical of methane formed via co2 reduction in deep-sea sediments. The absence of wide variation in deep-sea oD-CH4 could be the result of either a relatively constant kinetic hydrogen isotope effect associated

PAGE 96

with co2 reduction or of postgenerati ve isotopic equilibration of methane and pore water during the l ong subsurface burial of these 86 sediments. Resolution of this question through laboratory studies m ay be quite difficult, as this approach would require extrapolation of results obtained under conditions that are vastly different from in situ deep-sea conditions because of the great length of time and, presumably, the slow rate of methane production under natural deep-sea conditions. The isotopic compositions of methane samples collected from gas hydrate zones at several DSDP Sites indicated no obvious isotope effects associated with gas hydrate formation. Production of minor amounts (about 10% of the total) of methane via acetate dissimilation in deep-sea sediments may explain the deviation of the 6D-H2 0/6D-CH 4 pairs from the previousl y published relationship for co2 reduction. The methane isotopic compositions of gas bubbles collected from the sediments of several freshwater environments (mean 6D-CH4 -300/oo, std. dev. = 26/oo, n = 20; 13 0 mean 6 C-CH4 = -60.1 /oo, std. dev. = 6.1 /oo, n = 20) imply that the methane produced in these freshwater sediments is generally more D -depleted and less 13C-depleted than methane produced in deep-sea sediments. Evaluation of the 6DH 2 0/0D -CH4 with the Woltemate et al. ( 1984) model implies that acetate dissimilation is responsible for about 50 to 80% of the methane produced in these freshwater sediments. Methane o 13c and oD are generally inversely correlated, indicating that the large isotopic variability observed in these freshwater sediments is not solely attributable to bacterial methane oxidation, but also to some other parameter such as variability in methanogenic pathway selection.

PAGE 97

87 Nearshore marine-estuarine methanes are isotopically intermediate ( cSD CH 2580/ 0 46 1 3 mean 4 = -oo, std. dev. = 23 /oo, n = ; mean cS c-cH 4 -61 .8/oo, std. dev. = 3 .1/oo, n = 46) between deep -sea and freshwater methanes, implying that acetate dissimilation plays an intermediate role in nearshore mar ine-es tuar i ne methane product ion. Isotopi c analyses of met hane samples collected over the course of a year from the rapidly accumulating, organic-rich marine sediments of Cape Lookout Bight, North Carolina indicate that methane isotopic composition and probably, methanogenic pathway selection are seasonally, and in all likelihood spatially variable. Methane is generall y D-depleted and 13c-enriched during warm summer months (mean cSD = 269 /oo, std. dev. 9/oo, n = 21; mean o13c = -59.5/oo, std. dev. = 4.4/oo, n = 21) as compared to winter months (mean cSD = -236/oo, std. dev. = 4/oo, n = 6; mean o1 3 c-cH 4 0 0 -63.9 loo, std. dev. = 0 5 /oo, n = 6). Evaluation of these data with the Woltem ate et al. (1984) model implies that acetate dissimilation accounts for a relatively greater proportion of total methane production during the summer (about 45 to 60%) than during the winter (about 25 to 40%). The reasonably good agreement of the summer estimates obtained in this manner with those obtained from 14c radiotracer studies (Sansone and Martens 1981; Crill and Martens in press) indicates that application of the Woltemate et al. (1984) model to cSD-CH 4 data can yiel d useful estimates of methanogenic pathway selection in aquatic sediments. As for the freshwater sediments described above, the o13c and cSD of methane produced in Cape Lookout Bight sediments are inversely correlated, implying that both cSD-CH4 and cS 1 3c-CH are influenced by methanogenic pathway and that bacterial 4 methane oxidation plays a relatively minor role in producing the

PAGE 98

88 methane isotopic variability observed in these sediments. The nature and extent of post-depositional bacteria-organic matter interactions occurring prior to rapid methane production probably play a role in determining methanogenic pathway selection and methane isotopic composition in aquatic sediments. Factors t hat likel y influence these interactions are amount and/or type of organic matter deposited, sedimentation rate, concentrations of external electron acceptors and temperature. Land-derived higher plant material with its relatively lower H/C and N/C ratios (Stuermer et al. 1978) and relatively higher concentrations of biologically resistant lignins, resins and waxes (Aizenshtat et al. 1973) is probably not as readily degraded as algae-derived organic matter. Several investigators (Jorgensen 1978; Berner 1980; Westrich and Berner 198 4 ) suggested that decomposable organic matter should be divided into several fractions (two to eight) based on relative degradability. Model profiles of the decomposable organic matter indicate that the various fractions successively disappear based on their relative degradabili ties with depth in the sediment (time), with the result that the remaining fraction becomes progressively more refractory (Berner 1980) A high rate of sediment deposition would favor preservation of the more easily degraded organic matter fractions because both aerobic remineralization occurring at or near the sediment-water interface and the diffusion of other external electron acceptors into the sediments would be minimized by rapid isolation of organic matter from the overlying water (Berner 1980) In marine systems, where dissolved sulfate is by far the most abundant (28 mM) external electron acceptor, sulfate reduction may

PAGE 99

89 account for a significant part of the total organic matter degradation that occurs prior to rapid methane production. Jorgensen (1982) reported that bacterial sulfate reduction accounted for as much organic matter .degradation as did all aerobic processes in Danish coastal sediments. In general, the ratio of [dissolved sulfate]/[organic matter] should decrease in the order deep-sea > nearshore marine-estuarine > freshwater sediments. Because the consumption of the most labile organic matter fractions by aerobic processes and sulfate reduction should be maximized in sediments with low sedimentation rates and high [dissolved sulfate]/[organic matter] ratios, the methane producing zone of deep-sea sediments would be expected to receive a relatively smaller proportion of the more easily degradable organic matter fractions than the methane zone of freshwater sediments. Rapidly accumulating, nearshore marine-estuarine sedim ents which are deposited in waters containing appreciable co ncentrations of dissolved sulfate and receive organic matter of mixed origin, might be expected to fall between deep-sea and freshwater sediments with respect to the relative lability of organic matter reaching the methane zone. In general, nearshore marine-estuarine oD-CH4 measured in this study (mean oD = -258/oo) are intermediate between the freshwater (mean oD -300/oo) a nd deep-sea (mean oD = -185/oo) methanes implying that the lability of the organic matter reaching the methane zone may affect methanogenic pathway selecti on and There is, however, significant overlap between freshwater and nearshore marine-estuarine OD-CH 4 ; this probably reflects the great variability in important environmental parameters (i.e., sedimentation rate, organic matter input) that is inherent in these types of aquatic systems.

PAGE 100

90 The results discussed above imply that the 60 of methane produced in deep-sea sediments is sufficiently different from the 60 of methane produced in shallow aquatic sediments to allow the use of 6D-CH4 in the genetic characterization of commercially significant biogenic methane deposits. Rice and Claypool ( 1981) argue that shallow aquatic sediments are not normally suitable for the accumulation of biogenic gas because rapid production of methane usually begins near the sediment-water interface and hydrostatic pressures are low. As a result, most of the methane is lost from shallow sediments, either by bacterial consumption or by ebullition and escape to the atmosphere (Rice and Claypool 1981). Deep-sea sediments should be more appropriate for biogenic gas accumulation because rapid methane production usually begins deeper in the sediment and higher hydrostatic pressures allow retention of larger quantities of gas in solution in pore waters until the sediments are compacted and traps and seals are formed (Rice and Claypool 1981). Only a few 60-H2 0/6D-CH 4 pairs from a small proportion of the known commercial accumulations of biogenic gas are available in the literature. Surprisingly, most of the waters measured are substantially more 0-depleted than ocean pore waters (Nakai et al. 1974; Schoell 1980); this indicates that the methane may have originated in deep freshwater sediments. The differences in 60 values between methane and water from these commercial biogenic gas deposits is about the same as is found in deep-sea sediments; this was interpreted by Schoell (1980) to mean that the methane in these gas deposits was formed by co2 reduction. If, on the other hand, the observed differences in 60 values between methane and water represent significant post-burial isotopic equilibration in these ancient

PAGE 101

deposits, then measurement of oD-H2 0/oD-CH 4 pairs would probably be of little value in the genetic characterization of commercial biogenic methane deposits. 91 The microbial decomposition of organic matter in water-covered soils is one of the major sources of methane to the atmosphere (Khalil and Rasmussen 1983; Senum and Gaffney 1985), and bubble ebullition is the dominant mode of methane input to the atmosphere from some shallow freshwater (Robertson 1979; Cicerone and Shetter 1981) and marine (Martens and Klump 1980) environments. The results of this and other studies, indicate that methane bubbles formed in shallow aquatic sediments are generally significantly more 13c-depleted than the calculated average (Senum and Gaffney 1985) of all methane sources to the atmosphere. Some of the freshwater sediments studied here and by Woltemate et al. (1984) produce methane bubbles that have a oD near the estimated source average (Senum and Gaffney 1985); however there are presently too few measurements of the oD of atmospheric methane and its sources available to allow evaluation of the role of shallow aquatic sediments in determining the oD of atmospheric methane. Future measurements of: (I) the fluxes (bubble and diffusive) and isotopic compositions (oD and o13c) of methane emitted from various aquatic sediments and other major sources (i.e., ruminants), (II) the isotopic composition of atmospheric methane and, (III) various ancillary parameters (i.e., temperature, salinity, organic matter content) that may affect methane production rate and/or isotopic composition are necessary to allow accurate assessment of the importance of shallow aquatic environments to the global atmospheric methane budget.

PAGE 102

LITERATURE CITED Addy, S. R., and E. W. Behrens 1 980 Time of accumulation of hypersaline anoxic brine in Orca Basin (Gulf of Mexico). Mar. Geol. 37: 241252 Aizenshtat, Z., M J Baedecker, and I. R. Kaplan 1973. Distribution and diagenesis of organic compounds in JOIDES sediment from the Gulf of Mexico and Western Atlantic. Geochim. Cosmochim Acta 37 : 1881-1898. Alperin, M. J., and W. S. Reeburgh. 1984. Geochemical observations supporting anaerobic methane oxidation, p. 282-289. In R. L Crawford and R. s Hanson (Eds ) Microbial Growthon c1 compounds. A merican Society for Microbiology. 92 -------, and ------. 1985. Inhibition experiments on anaerobic methane oxidation. Appl. Environ. Microbial 50: 940-945 Anonymous. 1983. Energy demand, supply, and needed development resources are assessed. Oil and Gas Jour. 81: 71-73. Aubouin, J., and R von Huene. 1985 Summary: Leg 84 Middle America Trench transect off Guatemala and Costa Rica, p 939957. In R. von Huene, J. Aubouin et al., Init. Repts. DSDP 84. U.S. Govt. Printing Office. Bainbridge, A. E., H E Suess, and I Friedman. 1961. Isotope composition of atmospheric hydrogen and methane Nature 192: 6l.J8-649 Baker, E. W., W. Y Huang, J. G. Rankin, J. R. Castano, J. R. Guinn, and A. N Fuex. 1977. E lectron paramagnetic resonance study of thermal alteration of kerogen in deep -sea sediments by basaltic sill intrusion, p. 839-847. In Y Lancelot, E. Seibold, et al., Init. Repts. DSDP 41. U.S. Govt. Printing Office. Baltuck, M., R von Huene, and J Aubouin 1985 Introduction and Explanatory notes, p. 5-19. In R von Huene, J Auboin, et al., Init. Repts. DSDP 84. U.S. Govt. Printing Office. Barnes, R. 0., and E D Goldberg. 1976. Methane production and consumption in anoxic marine sediments. Geology l.J: 297-300. Baross, J. A., M. D. Lilley, and L. I. Gordon. 1982 Is the CH4 H 2 and CO venting from submarine hydrothermal systems prod uced by

PAGE 103

93 thermophilic bacteria? Nature 298: 366-368. -------, and J. W. Deming. 1983. Growth of "black smoker" bacteria at temperatures or at least 250C Nature 303: 423-426. Barry, R., H. Fritts, J. Imbrie, J. Kutzbach, J. M. Mitchell, and S. M. Savin. 1979. Paleoclimatic research: Status and opportunities. Quat. Res. 12: 6-17. Bartlett, K. A. 1981. Macrofauna distribution and seasonal influences on interstitial water chemistry of Cape Lookout Bight, N.C. M.S. thesis, Univ. North Carolina, Chapel Hill. 98p. Begemann, F., and I. Friedman. 1968. Tritium and deuterium in atmospheric methane. Jour. Geophys. Res. 73: 1149-1153. Belyaev, S. S., Z. I. Finkel'shtein, and M. v. Ivanov. 1975. Intensity of bacterial methane formation in ooze deposits of certain lakes. Mikrobiologiya (English translation) 44: 272-275. -------, R. Wolin, W. R. Kenealy, M. J. DeNiro, s. Epstein, and J. G. Zeikus. 1983. Methanogenic bacteria from the Bondyuzhskae oil field: general characterization and analysis of stable carbon isotopic fractionation. Appl. Environ. Microbial. 45: 691-697. Bernard, B. B. 1978. Light hydrocarbons in marine sediments. Ph.D. thesis, Texas A&M Universty, College Station. 144p. ------J. M. Brooks, and W. M. Sackett. 1976. Natural gas seepage in the Gulf of Mexico. Earth Planet. Sci. Lett. 31: 48-54. ------J. Brooks, and W. M Sackett. 1 977. A geochemical model for characterization of hydrocarbon gas sources in marine sediments. Offshore Tech. Conf. OTC 2934: 435-438 Berner, R A. 1980. Early diagenesis a theoretical approach. Princeton Uni v. Press. Blackwelder, B. w., 0 H. Pilkey, and J. D. Howard. 1979. Late Wisconsin sea levels on the southeast U. S. Atlantic shelf based on in-place shoreline indicators. Science 204: 618-620. Blair, N., c s. Martens, and D. Des Marais. 1984. Seasonal o1 3c variations from an anoxic marine sediment. EOS 65: 960 Blunden, G., S.M. Gordon, W F. H. McLean, and M.D. Guiry. 1982. The distribution and possible taxonomic significance of quaternary ammonium and other Dragendorff-positive compounds in some genera of marine algae. Bot. Mar. 25: 563-567 Bouma, A. H., J. M. Coleman, et al. 1985. Site 618 summary, in press. H. Bouma, J. M. Coleman, et al., Init. Repts. DSDP 96. U.S. Govt. Printing Office.

PAGE 104

94 Brock, T. D. 1974. Biology of microorganisms. Prentice-Hall. Brooks, J. M., L.A. Barnard, D. A. Wiesenburg, M. c Kennicutt, and K. A. Kvenvolden. 1983. Molecular and isotopic compositions of hydrocarbons at Site 533, DSDP Leg 76, p. 377-390. In R. E. Sheridan, F. M. Gradstein, et al., Init. Repts. U.S. Govt. Printing Office. Burdige, D. J., and K. H. Nealson. 1985. Microbial manganese reduction by enrichment cultures from coastal marine sediments. Appl. Environ. Microbiol. 50: 491-497. Burke, R. A., J. M. Brooks, and W. M. Sackett. 1981. Light hydrocarbons in Red Sea brines and sediments. Geochim. Cosmochim. Acta 45: 627-634. Cappenberg, Th. E., and R. A. Prins. 1974. Interrelations between sulfate-reducing and methane-producing in bottom deposits of a fresh-water lake. 3. Experiments with C-labeled substrates. Ant. van Leeuwenhoek 40: 457-469. Cavalieri, A. J., and H. C. Huang. 1981. Accumulation of proline and glycinebetaine in Spartina alterniflora Loisel in response to NaCl and nitrogen in the marsh. Oecologia 49: 224-228. Chanton, J. P., C. S. Martens, and G. W. Kipphut. 1983. Lead-210 sediment geochronology in a chang i ng coastal environment. Geochim. Cosmochim. Acta 47: 1791-1804. Chen M., and M. J. Wolin 1977. Influence of CH4 production by Methanobacterium ruminatum on the fermentation of glucose and lactate by Selenomonas ruminatum. Appl. Environ. Microbiol. 34: 756-759. Chung H. M 1976. Isotope fractionation during the maturation of organic matter. Ph.D. thesis, Texas A&M Univ., College Station. 162 p. Claypool, G. E., and I. R. Kaplan. 1974. The or1g1n and distribution of met hane in marine sediments, p. 99-139. In I. R. Kaplan (Ed.), Natural gases in marine sediments. Plenum. -------, and K. A. Kvenvolden. 1983. Methane and other hydrocarbon gases in marine sediments. Ann. Rev. Earth Planet. Sci. 11: 299-327. -------, C. E. Threlkeld, P. M. Mankiewicz, M. A. Arthur, and T. P. Anderson. 1985. Isotopic composition of interstitial fluids and origin of methane in slope sediment of the Middle America Trench, p. 683-691. In R. von Huene, J. Aubouin, et al., Init. Rept. DSDP 84. U.S. Printing Office. -------, A. K. Vuletich, and K. A. Kvenvolden. 1985. Isotopic composition of interstitial fluids in sediment of the Nankai

PAGE 105

95 Trough DSDP Leg 87. In D E. Karig, H Kagami, et al., Init. Repts. DSDP 87. in press. U.S Govt. Printing Office. Cloud P. E., I. Friedman, F D. Sisler, and V H. Dibeler. Microbi.al fractionation of hydrogen isotopes. Science 1394 -1 395. 1958. 127: Coplen, T. B., and R N. Clayton. 1973. Hydrogen isotopic composition of NBS and IAEA stabl e isotope water reference samp les. Geochim. Cos m ochim Acta 37: 23472349 Coleman D. D., J. B. Risatti, and M. Schoell. 1 981. Fractionation of carbon and hydrogen isotopes by methane-oxidizing bacteria. Geochim. Cosmochim Acta 45: 1033-1037 Coleman, M. L., T. J. Shepherd, J. J. Durham, J. E Rouse, and G. R. 1982 Reduction of water with zinc for hydrogen isotope analysis. Anal. Chern. 54: 993 995 Craig, H. 1953. The geochemistry of the stable carbon isotopes. Geochim Cosmochim. Acta 3: 53 -92. -------. 1957. Isotopic standards for carbon and oxygen and correction factors for mass-spectrometric analysis of carbon dioxide. Geochim. Cosmochim. Acta 1 2: 133-149. -------. 1961. Standard for reporting concentrations of deuterium and oxygen-18 in natural wat ers. Science 133: 1 8331 834 -------. 1967. Ocean water, isotopic composition of, p 268 -271. In McGraw-Hill Yearbook Science and Technology McGraw-Hill. -------, and C. C. Chou. 198 2 Methane : the record in polar ice cores. Geophys. Res; Lett. 9: 477-481 Crill, P. M., and C S Martens. 1983 Spatial and temporal fluctuations if methane production in anoxic coastal marine sediments. Limnol. Oceanogr 28: 1117-1130. -------, and -------. 1 985. Methane production from bicarbonate and acetate in the sulfate reduction zone o f an anoxic marine sediment. Geochim Cosmochim Acta in press. Daniels, L., G. Fulton, R. W. Spencer, and W. H Orme-Johnson. 1980. Origin of hydrogen in methane produced by Methanobacterium thermoautotrophicum. J Bacterial. 141: 694698 Dansgaard W., and H. Tauber. 1969. Glacier oxygen-18 content and Pleistocene ocean temperatures. Science 166: 499-502. Davis J B., and H F. Yarbrough. 1966. Anaerobi c oxidation of hydrocarbons by Desulfovibrio Desulfuricans. Chern. Geol. 1: 137-144.

PAGE 106

Deines, P 1980. The isotopic composition of reduced organic carbon, p 329-406 In P Fritz, and J. Ch. Fontes (Eds.), Handbook of environmental isotope geochemistry, vol. 1 The terrestrial environment, A. Elsevier. Devol, A. H. 1983 Methane oxidation rates in the anaerobic sediments of Saanich Inlet. Limnol Oceanogr. 28: 738-742 Ehhalt, D H. 1 973. Methane in the atmosphere, p. 144-158 In G M Woodwell, and E V Pecan (Eds.), Carbo n and t he biosphere. Office of Information Service. -------, and A. Volz. 1976. Coupling of the CH4 with the H and CO cycle: Isotopic evidence, p. 23-33 In H. G Schlegef, G. Gottschalk, and N Pfennig (Eds.), Microbial production and utilization of gases. E. Goltze KG. Epstein, S., C. J. Yapp, and J. H. Hall. 1976. The determination of the D/H ratio of non-exchangeable hydrogen in cellulose extracted from aquatic and land plants. Earth Planet. Sci. Lett. 30: 241-251. Estep, M F., and T. C. Hoering. 1980 Biogeochemistry of the stable hydrogen isotopes. Geochim Cosmochim Acta 44: 1197-1206. Fenchel, T., and Blackburn, T. H 1979. Bacteria and Mineral Cycling. Academic. Frank, D J 1973. Deuterium distribution in the Gulf of Mexico J Phys. Oceanogr 3: 230 236. Friedman, I. 1953 Deuterium content of natural waters. Geochim. Cosmochi m Acta 4: 89-103. -------, A C. Redfield, B. Schoen and J. Harris. 1964. The variation in the deuterium content of natural waters in the hydrologic cycle. Rev. Geophys. 2: 177-224. -------, I and K. Hardcastle. 1973. Interstitial water studies, Leg 15-Isotopic composition of water, p. 901904 In B C Heezen, I. D MacGregor, et al., Init. Repts. DSDP 20. U.S. Govt Printing Office. -------, and J R. O'Neil. 1977. Compilation of stable isotope fractionation factors of geochemica l interest. In Data Geochem., 6th ed. Geol Surv. Prof. Paper 440K. U.S. Govt .-printing Office. Fuex, A. N. 1977. The use of stable isotopes in hydrocarbon exploration. J Geochem. Explor. 7: 155 188. Galimov, E. M. 1968. Isotopic composition of carbon in gases of the crust. Internat. Geol. Rev. 11: 1092-1104. -------, and K A. Kvenvolden. 1983 Concentrations and carbon isotopic

PAGE 107

compositions of CH4 and co2 in gas from sediments of the Blake Outer Ridge, DSDP Leg 76, p. 403-407. In R. E. Sheridan, F. Gradstein, et al., Init. Repts. DSDP-76. U.S. Govt. Printing Office. 97 Games, L. M., J. M. Hayes, and R. P. Gunsalus. 1978. Methane-producing bacteria: natural fractionations of stable carbon isotopes. Geochim. Cosmochim. Acta 42: 1295-1297. Giggenbach, W. F. 1980. Geothermal gas equilibria. Geochim. Cosmochim. Acta 44: 2021-2032. Gold, T., and S. Soter. 1980. The deep earth-gas hypothesis. Scientific American 242: 154-161. Gonfiantini, R. 1978. Standards for stable isotope measurements in natural compounds. Nature 271: 534-536. -------. 1981. The a-notation and the mass-spectrometric measurement techniques, p. 35-84. In J. R. Gat, and R. Gonfiantini (Sci. Eds.), Stableisotope hydrology: Deuterium and oxygen-18 in the water cycle, Tech. Repts. Series #210. IAEA. Gorden, S., and W. A. Mulac. 1975. Reactions of the OH cx2 n) radical produced by the pulse radiolysis of water vapor. Int. J. Chern. Kinet. 7 (Symp. no. 1): 289-299. Gunter, B. D., and B. C. Musgrave. 1971. New evidence on the or1g1n of methane in hydrothermal gases. Geochim. Cosmochim. Acta 35: 113-118. Hagemann, R., G. Nief, and E. Roth. Absolute isotopic scale for deuterium analysis of natural waters. Tellus 22: 712-715. Hameed, S., and R. D. Cess. 1983. Impact of a global warming on biospheric sources of methane and its climatic consequences. Tellus 35B: 1-7. Harriss, R. C., D. I. Sebacher, and F. P. Day. 1982. Methane flux in the Great Dismal Swamp. Nature 297: 673-674. Harrits, S.M., and R. S. Hanson. 1980. Stratification of aerobic methane-oxidizing organisms in Lake Mendota, Madison, Wisconsin. Limnol. Oceanogr. 25: 412-421. Hoefs, J. 1980. Stable isotope geochemistry. Springer-Verlag. Haering, T. c. 1968. Reactions of the organic matter in a Recent marine sediment. Carnegie Inst. Wash. Yearbook 67: 199-201. -------. 1984. Thermal reactions of kerogen with added water, heavy water and pure organic substances. Org. Geochem. 5: 267-278. von Huene, R., J. Aubouin, et al. 1985. Init. Repts. DSDP 84. U.S.

PAGE 108

Govt. Printing Office. Hunt, J. M 1979. Petroleum geochemistry and geology. Freeman. Ianotti, E. L., P. Kafkewitz, M. J. Wolin, and M. P. Bryant. 1973. Glucose fermentation products of Ruminococcus albus grown in continuous culture with Vibrio succinogenes: changes caused by interspecies transfer of H 2 J. Bacterial. 114: 1231-1240. Iversen, N., and B. B Jorgensen. 1985. Anaerobic methane oxidation rates at the sulfate-methane transition in marine sediments from Kattegat and Skagerrak (Denmark). Limnol. Oceanogr. 30: 944-955. Jeffrey, A. W. A., R. C. Pflaum, T. J. McDonald, J. M. Brooks, and K. A. Kvenvolden. 1985. Isotopic analysis of core gases at Sites 565-570 Deep Sea Drilling Project Leg 84, p. 719-726. In R. von Huene, J. Aubouin, et al., Init. Repts. DSDP 84. u.s.-covt. Printing Office. Jenden, P. D., and J. M. Gieskes. 1983. Chemical and isotopic compositions of interstitial water from Deep Sea Drilling Project Sites 533 and 534, p. 453-462. In R. E. Sheridan, F. M. Gradstein, et al., Init. Repts. DSDP 76. u:S. Govt. Printing Office. Jeris, J. S., and P. L. 1965. The biochemistry of methane fermentation using C tracers. J. Water Pollut. Cont. Fed. 37 : 178-192. Jones, G. E., and R. L. Starkey. 1957. Fractionation of stable isotopes of sulfur by microorganisms and their role in deposition of native sulfur. Appl. Microbial. 5: 111-118. Jones, J. G., B. M. Simon, and S Gardner. 1982. Factors affecting methanogenesis and associated anaerobic processes in the sediments of a stratified eutrophic lake. J. Gen. Microbial. 128: 1-11. Jorgensen, B B. 1978. A comparison of methods for the quantification of bacterial sulfate reduction in coastal marine sediments. Geomicrobiol. J. 1: 29-47. -------. 1982. Mineralization of organic matter in the sea bed -the role of sulfate reduction. Nature 246: 643-645. Kaplan, I. R., and S C. Rittenberg. 1964. Microbiological fractionation of sulfur isotopes. J. Gen. Microbiol. 34: 195-212. Karig, D. E., H. Kagami, and DSDP Leg 87 Scientific Party. 1983. Varied responses to subduction in Nankai Trough and Japan Trench Forearcs. Nature 304: 148-151. Khalil, M A. K., and R. A. Rasmussen. 1982. Secular trends of atmospheric methane (CH4). Chemosphere 11: 877-883. -------, and, -------. 1983. Sources, sinks, and cycles of atmospheric

PAGE 109

methane J. Geophys. Res. 88: 513151ijij. King G M. 198ija Utilization of hydrogen acetate, and "noncompetitive" substrates by methanogenic bacteria in marine sediments. Geomicrobiol. J 3: 275-306. ------. 1 98ijb. Metabolis m of trimethylamine choline, a n d glycin e betai ne by sulfate-reducing and methanoge nic bacteria in marine sediments. Appl. E nviron Microb ial. ij8: 7 1 9 725 99 -----, M J. Klug, and D R Lovley. 1983. Metabolism of acetate, methanol, and methylated amines in intertidal sediments of Lowes Cove Maine. Appl Environ. Microbial. ij5: 18ij8-1853 Kipphut, G W., and c. S Martens. organic-rich marine basin-3. methane-saturated sediments. 20ij9-2060. 1982 Biogeochemical cycling in an Dissolved gas transport in Geochim Cosmochim Acta ij6: Koepp M. 1978. D/H isotope exchange reaction between petroleum and water: a contributory determinant for D/H-isotope ratios in crude oils?, p 221222 In R. E. Zartmann (Ed.), Short papers of the Fourth Internationar-conference, Geochronology, Cosmochronology Isotope Geol ogy 1 978 U S Geol. Survey Open File Report 78-7 01. U S. Govt Printing O ffice. Koyama, T. 1963. Gaseous metabolism in lake sediments and paddy soils and the production of atmospheric methane and hydrogen J Geophys. Res 6ij: 3971-3973 Kristjansson, J K., P Schonheit, and R. K Thauer. 1982 Different Ks values for hydrogen of methanogenic bacteria and sulfate-reducing bacteria: An explanation for the apparent inhibition of methanogenesis by sulfate. Arch. Microbial. 1 31: 278 282 Kvenvolden K. A., G E Claypool, C. N. Threlkeld, and E. D. Sloan. 1984. Geochemistry o f a naturally occurring mass i ve marine gas hydrate. Org. Geochem. 6 : 703 7 1 3 ----, a n d T J McDon a ld. 1 985 Gas h y dr a t e s of the M i dd l e A m e rica Trench -Deep Sea Drilling Project Leg 8ij, p. 667-682 In R von Huene J Aubouin et al., Inti. Repts. DSDP 8ij. U.S-.-Govt Printing Office. Lacis, A., J Hansen P Lee T Mitchell, and S. Lebedeff. 1981. Greenhouse effect of trace gases, 1970 1980 Geophys Res Lett. 8: 1035-1038 Lawrence J R. and J M. Gieskes. 1981. Constraints on water transport and alteration in the oceanic crust from the isotopic composition of pore water. J. Geophys Res. 86 : 792ij 793ij. Lovley, D. R., and M J. Klug. 1 982 Intermediary metabolism of organic matter i n the sedi ments of a eutrophi c lake. Appl. Environ.

PAGE 110

100 Microbial. 43: 552-560. -------,and-------. 1983. Methanogenesis from methanol and from hydrogen and carbon dioxide in the sediments of a eutrophic lake. Appl. Environ. Microbial. 45: 1310-1315. Lyon, G. L. 1974. Isotopic analysis of gas from the Cariaco Trench sediments, p. 91-98. In I. R. Kaplan (Ed.), Natural gases in marine sediments. Plenum. -------, and J. R. Hulston. 1984. Carbon and hydrogen isotopic compositions of New Zealand geothermal gases. Geochim. Cosmochim. Acta 48: 11 61-1171. McCarty, P. 1972. Energetics of organic matter degradation, p. 91-118. In R. Mitchell (Ed.), Water pollution microbiology. WileyInterscience. Mcintyre, A., N. G. Kipp A. W. H Be, T. Crowley, T. Kellogg, J. V. Gardner, W. Prell, and W. F. Ruddiman. 1976. Glacial north Atlantic 18,000 ye ars ago: A CLIMAP reconstuction. Geol. Soc. Amer. M e moir 145: 43-76. McNair, H. M., and E. J. Bonelli. 1969. Basic gas chromatography. Varian. M ah, R. A. 1982. Methanogenesis and methanogenic partnerships. Phil. Trans. R. Soc. Lend. B 297: 599-616. Manheim, F. T., and F. L. Sayles. 1974. Composition and or1gw of interstitial waters of marine sediments, based on deep sea drill cores, p. 527-568. In E. D. Goldberg (Ed.), The sea, volume 5: Marine chemistry. Wiley-Interscience. Martens, C. S. 1976. Control of methane sediment-water bubble transport by macro-infaunal irrigation in Cape Lookout Bight, North Carolina. Science 192: 998-1000. -------, and R. A. Berner. 1974. Methane production in the interstitial waters of sulfate-depleted marine sediments. Science 185: 1 067-1069. -------, and-------. 1977. Interstitial water chemistry of anoxic Long Island Sound sediments. 1. Dissolved gases. Limnol. Oceanogr. 22: 10-25. -------, and J. v. Klump. 1980. Biogeochemical cycling in a organic-rich coastal marine basin. 1. Methane sediment-water ex c hange processes. Geochim. Cosmochim. Acta 44: 471-490. -------, G. w. Kipphut, and J. V. Klump. 1980. Sediment-water chemical ex c hange in the c oastal z one traced by in situ Rado n-222 flux measurements. Science 2 08: 285288.

PAGE 111

i01 -------and J. V. Klump. 1984. Biogeochemical cycling in a organic-rich coastal marine basin. 4. An organic carbon budget for sediments dominated by sulfate reduction and methanogenesis. Geochim. Cosmochim. Acta 48: 1987-2004 Menzies, R. J., G. T. Rowe, and L. P. Atkinson. 1968. A small basin containing anoxic sediments Cape Lookout Bight, North Carolina. Int. Rev. Hydrobiol. 53: 77-81. Mountfort, D. 0., R. A. Asher, E. L. Mays, and J. M Tiedje. 1980. Carbon and electron flow in mud and sandflat intertidal sediments at Delaware Inlet, Nelson, New Zealand. Appl. Environ. Microbiol. 39: 686-694. Nakai, N., Y. Yoshida, and N. Ando 1974 Isotopic studies on oil and natural gas fields in Japan. Chikyu Kagaku 7/8: 87-98. National Academy of Sciences. 1980. Energy in transition 1985-2010. Freeman. Nier, A. 0. 1947. A mass spectrometer for isotope and gas analysis. Rev. Sci. Instr. 18: 398-411 Nikaido, M. 1 977. On the relation between methane production and sulfate reduction in bottom muds containing sea water sulfate. Geochem. J. 11: 199-206. Northam M A., D. A. Curry, R S Scanlon, and P. L. Parker. 1981. Stable carbon isotope ratio variations of organic matter in Orca Basin sediments. Geochim Cosmochim Acta 45: 257-260. Oana, S., and E. S. Deevey. 1960. Carbon-1 3 in lake waters and its possible bearing on paleolimnology. Am. Jour. Sci. 258A: 253-272. Oremland, R. S., L. M. Marsh, and S Polcin. 1982. Methane production and simultaneous sulfate reduction in anoxic saltmarsh sediments. Nature 296: 143-145. Osmond, C. B., N. Valaane, fJ M Haslam, P. Uotila, and Z. Roksandic. 1981. Comparisons of o C values in leaves of aquatic macrophytes from different habitats in Britain and Finland; some implications for photosynthetic processes in aquatic plants. Oecologia 50: 117 124. Pflaum, R. c., J. M. Brooks, H. B. Cox, M. C. Kennicutt, and D. D. Sheu. 1985. Molecular and isotopic analysis of core gases and gas hydrates from Sites 618 and 619, DSDP Leg 96, in press. In A. H Bouma, J. M. Coleman, et al., Init. Repts. DSDP 96. U.S. Govt. Printing Office. Pine, M. J., and H. A. Barker. 1956 Studies on the methane fermentation 12. The pathway of hydrogen in the acetate fermentation. J. Bacteriol. 71: 644-648

PAGE 112

102 Redding, C. E., M. Schoell, J. C. Monin, and B. Durand. 1980. Hydrogen and carbon isotopic composition of coals and kerogens, p. 711-723. In A. G. Douglas and J R. Maxwell (Eds.), Advances in organic geochemistry 1979. Pergammon. Reeburgh, W. S. 1976. Methane consumption in Cariaco Trench waters and sediments. Earth Planet. Sci. Lett. 28: 337-344. -------. 1980 Anaerobic methane oxidation: rate depth distributions in Ska n Bay sediments. Earth Planet. Sci. Lett. 47: 345-352. -------. 1983. Rates of biogeochemical processes in anoxic sediments. Ann. Rev. Earth Planet. Sci. 11: 269-298 -------, and D. T. Heggie. 1977 Microbial methane consumption reactions and their effect on methane distributions in freshwater and marine environments. Limnol. Oceanogr. 22: 1-9. Rice, D. D., and G E Claypool. 1981. Generation, accumulation, and resource potential of biogenic gas. Am. Assoc. Pet. Geol Bull. 65: 5-25. Rigby, D., B. D. Batts, and J. W. Smith. 1981. The effect of maturation on the isotopic composition of fossil fuels. Org Geochem. 3: 29-36 Robertson, C K. 1979. Quantitative comparison of the significance of methane in the carbon cycles of two small lakes. Arch. Hydrobiol. Beib. Ergebn. Limnol. 12: 123-135. Rudd, J. W. M., and R D. Hamilton. 1978. Methane cycling in a eutrophic shield lake and its effects on whole lake metabolism Limnol. Oceanogr 23: 337-348 -------and C. D Taylor. 1980 Methane cycling in aquatic environments. Adv. Aquat. Microbial. 2: 77-150. -------, A. F. Furrtani, R J. Flett, and R D. Hamilton. 1976. Factors controlling rates of methane oxidation in shield lakes: the role of nitrogen fixation and oxyge n concentration. Limnol. Ocea nogr. 21: 357-364. Rust, F 1981. Ruminant methane <13c;12c) values: relation to atmospheric methane Science 211: 1044-1046 Sackett, W. M. 1978. Carbon and hydrogen isotope effects during the thermocatalytic production of hydrocarbons in laboratory simulation experiments. Geochim. Cosmochim. Acta 42: 571-580. -------. 1983 Stable carbon isotope studies on organic matter in the marine environment. Handbook of environ. isotope geochem in press. -------, G Brooks, M Conkright, L. Doyle and L. Yarbro 1985. Stable

PAGE 113

isotope compositions of organic carbon in Tampa Bay, Florida, U S.A ; implications for evaluating oil contaminations. Appl. Geochem. in press. 103 Sansone, F. J., and C S Martens. 1978. Methane oxidation in Cape Lookout Bight, North Carolina. Limnol. Oceanogr 23: 349-355 -------, and -------. 1981. Methane production from acetate and associated methane fluxes from anoxic coastal sediments. Science 211 : 707-709. -------and -------. 1982. Volatilie fatty acid cycling in organic-rich marine sediments. Geochim Cosmochim. Acta 46: 1575-1589. Schiegl, W. E., and J. C. Vogel. 1970. Deurterium content of organic matter. Earth Planet. Sci. Lett. 7: 307-313. Shoell, M. 1980. The hydrogen and carbon isotopic composition of methane from natural gases of various origins. Geochim. Cosmochim. Acta 44: 649-661. -------. 1982 Stable isotopic analyses of interstitial gases. in Quaternary sediments from the Gul f of California, p 815817. I n J. R Curray, D G Moore, et al., Init. Repts. DSDP 64. u.s Govt Printing Office. -------. 1983. Genetic characterization of natural gases. Am. Assoc. Pet. Geol. Bull. 67: 2225-2238 -------, and C E Redding. 1978. Hydrogen isotopic compositions of selected crude oils and their fractions, p. 384-385. In R. E Zartmann (Ed.), Short papers of the Fourth International Conference, Geochronology, Cosmochronology, Isotope Geology, U. S. Geol. Survey Open File Report 78-701. U .S. Govt. Printing Office. -------, E Faber, and M. L Coleman 1983. Carbon and hydrogen isotopic compositions of the NBS 22 and NBS 21 stable isotope reference materials: an inter-laboratory comparison. Org. Geochem. 5 : 3-6 Schonheit, P., J. A. Kristjansson, and R. K. Thauer. 1982. Kinetic mechanism for the ability of sulfate reducers to outcompete methanogens for acetate. Arch. Microbiol. 132: 285-288. Senum, G I., and J. S. Gaffney 1985. A reexamination of the tropospheric methane cycle: geophysical implications, p. 61-69. In E. T. Sundquist and W. S Broecker (Eds.), The carbon cycle and atmospheric co2 : natural var_iations present, Geophysical Monograph 32. Amer1can Geophys1cal Un1on. Shackleton, N.J. 1977. The oxygen isotope stratigraphic record of the late Pleistocene. Phil. Trans. R Soc. Lond. B 280: 169-182.

PAGE 114

Sheppard, J. C., H. Westberg, J F. Hopper, K. Ganeson, and P. Zimmerman. 1982. Inventory of global methane sources and their production rates. J Geophys. Res 87: 1305 1312. Sheu, D. D., A. Shakur, D. A. Wiesenburg, J. M. Brooks, and H. R 104 Krouse. 1984. Isotopic compositions of sulfur and oxygen in dissolved sulfate from the Orca Basin. Geochim Cosmochim. Acta in press. Shipley, T. H., M H. Houston, R. T. Buffler, J. Shaub K. J. McMillen J W. Ladd, and J L. Worzel 1979 Seismic evidence for widespread possible gas hydrate horizons on continental slopes and rises. Am. Assoc Pet. Geol Bull. 63: 2204-2213. Sha kes, R. F., P. K Trabant, B. J. Presley, and D F. Reid 1977 Anoxic, hypersaline basin in the northern Gulf of Mexico Science 196: 1443-1446. Smith, B. N and S. Epstein. 1970 Biogeochemistry of the stable isotopes of hydrogen and carbon in salt marsh biota. Plant Physiol. 46: 738-742 -------, and-------. 1971. Two categories of 13c;12c ratios for higher plants. Plant Physiol. 47: 380-384. Smith, J. W., D. Rigby, P. W Schmidt, and D A Clark. 1983. D/H ratios of coals and the paleolatitude of their deposition. Nature 302 : 322-323 Smith, P. H., and R A Mah. 1966 Kinetics of acetate metabolism during sludge digestion. Appl Microbial. 14: 368-371. Sowers, K. R., S. F. B aron, and J. G Ferry. 1984. Methanosarcina acetivorans sp. nov., an acetotrophic methane-produ cing bacterium isolated from marine sediments. Appl Environ. Microbial. 47: 971-978 Stahl, w. 1974 Carbon isotope fractionations in natural gases. Nature 251: 134 135. Stevens, C. M., and F. E Rust. 1982. The carbon isotopic composition of atmospheric methane. J. Geophys Res 87: 4879-4882. Stuermer, D. H., K. E. Peters, and I. R. Kaplan 1978. Source indicators of humic substances and proto-kerogen. Stable isotope ratios, elemental compositions and electron resonance spectra. Geochim. Cosmochim. Acta 42: 989 -997. Stumm, w., and J. J. Morgan. 1981. Aquatic Chemistry, 2nd Ed. Wiley. Takai, Y. 1970. The mechanism of methane fermentation in flooded paddy s oil. Soil Sci. Plant Nutr. 16: 238-244. Tompkins, R. E and L E. Shephard. 1979 Orca Basin: depositional

PAGE 115

105 processes, geotechnical properties and clay mineral ogy o f Holocene sediments within an anoxic hypersaline basin, northwest Gulf of Mexico Mar. Geol. 33: 221-238. Trofimuk, A. A., N v. Cherskiy, and V P. Tsarev. 1973 Accumulation of natural gases.in zones of hydrate-formation in the hydrosphere. Doklady Akad. Nauk. SSSR (English translation) 212 : 931-934. Wang, W. C., Y L. Yung, A. A Lacis, T. Mo, and J E. Hansen 1976. Greenhouse effects due to man-made perturbations of trace gases. Science 194: 685 690 . Welhan, J. A 1981. Carbon and hydrogen gases in hydrothermal systems: the search for a mantle source. Ph.D. thesis, Univ Calif., San Diego. 194p -------, and H. Craig. 1979. Methane and hydrogen in East Pacific Rise hydrothermal fluids. Geophys. Res Lett. 6: 829-831. Westrich, J T., and R. A Berner. 198 4 The role of sedimentary organic matter in bacterial sulfate reduction: the G model tested. Limnol. Oceanogr 29: 236 -249. Whelan J. K., R. Oremland, M Tarafa, R. Smith, R Howarth and C. Lee. 1 985 Evi dence for sulfate reducing and meth a n e producing microorganisms in sediments from Sites 618, 619 and 622, in press. In A. H. Bouma, J M. Coleman, et al. !nit. Repts. DSDP 96. U S -.-Govt. Printing Office. Whiticar, M. J., E Faber, and M. Schoen. 1 985 Biogenic methane formation in marine and freshwater environments: C02 reduction v s acetate fermentation -isotopic evidence. Geochim Cosmochim Acta in press. Wiesenburg D A. 1 980. Geoch emistry of d issolved gases in the hypersaline Orca Basin. Ph. D thesi s Texas A&M Univ., College Station. 265p. Winfrey, M R., and J G Zeikus. 1977 Effect of sulfate on carbon and electron flow during microbial methanog enesis in freshwater sediments. Appl. Environ. Microbial. 33: 275-281. -------, and -------. 1 979 Anaerobic metabolism of immediate methane precursors in Lake Mendota Appl Environ Microbial. 37: 244-253 Wolfe, R S 1979. Microbial biochemistry of methane a study in contrasts. Part 1: methanogenesis, p. 267-300. In J. R. Quayle (Ed.) International review of biochemistry, vol. 21, microbial biochemistry. Univ. Park Press. Wolin, M. J 1979. Hydrogen transfer in microbial communities p. 323-356 I n A T Bull and J. H Slater (Eds .), Microbial interactions and communities Academic

PAGE 116

-------1982. The rumen fermentation: a model for microbial interactions in anaerobic ecosystems, p. In M. Alexander (Ed .), Advances in microbial ecology, vol. 3. Plenum. -------, and T. L. Miller. 1982. Interspec ies hydrogen transfer: 1 5 years later. Am. Soc Microbial. News 561565 Woltemate, I., M. J. Whiticar, a n d M. Schoell. Carbon and hydrogen isotopi c composition of bacterial methane in a shallow freshwater lake Limnol. Oceanogr 29: 985-992 106 Yancey, P H M. E. C lark, S C Hand, R D Bowlus, and G M Somero 1982 Living with water stress: evolu tion of osmolyte systems. Science 217: Yapp, C. J., and S. Epstein. 1982. Climatic significance of the hydrogen isotope ratios in tree cellulose. Nature 297: 636-639 Yeh, H W., and S. Epstein. 1981. Hydrogen and carbon isotopes of petroleum and related organic matter. Geochim. Cosmochim Acta 753-762 Zehnder A. J. B and T. D. Brock 1 979 Methane formation and methane oxidation b y methanogenic bacteria. J. Bacterial. 1 3 7: -------, and -------. 1 980 Anaerobic methane oxidation: occurrence and ecology. Appl. Environ Microbial. 39 : -------B. A Huser, T Characterization of non-hydrogen-oxydizing 1 1 1 D. Brock and K Wuhrmann. 1 980 an acetate-decarboxylating, methane bacterium. Arch Microbial. -------, K. Ingvorsen, and T. Marti. 1982. Microbiology of methane bacteria, p. In D. E Hughes D. A. Stafford, 8. I. Wheatley W Baader G Lettinga, E J Nyns, W Verstraete, and R L. Wentworth (Eds.), Anaerobic digestion 1981. Elsevier. Zeikus, J G., and M. R Winfrey. 1 976 T emperature limitati on of methanogenesis in aquatic sediments. Appl. Environ. Microbial. 3 1 : 99-107. Ziegler, H., C. B. Osmond, W Stichler, and P. Trimborn. 1 976 Hydrogen isotope discrimination in higher plants: correlations with photosynthetic pathway and environment. Planta 1 28: 85-92.



PAGE 1

STABLE HYDROGEN AND CARBON ISOTOPIC COMPOSITIONS OF BIOGENIC METHANE$ by Roger Allen Burke, Jr. A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Mari n e Science in the University of South Florida December, 1 985 Major P rofessor : Dr. William M Sackett

PAGE 2

Graduate Council University of South Florida Tampa, Florida CERTIFICATE O F APPROVAL Ph. D Dissertation This is to certify that the Ph.D. Dissertation of Roger Allen Burke Jr. with a major in Marine Sc i e nc e has been app r oved by the Examin i ng Committee o n August 1985 as satisfactory for t h e dissertation r eq u i r ement for t h e Ph.D. degree. Examining Comm i ttee: Member: Dr. Robert M Garrels Member: Dr. Edward S Van Vleet

PAGE 3

ACKNOWLEDGEMENTS I sincerely thank Dr. William M. Sackett for his advice, encouragement, and friendship during my graduate career. I also thank the other members of my committee: Dr. Robert Byrne, Dr. Robert Garrels, Dr. Harold Humm, and Dr. Edward Van Vleet for their time and input into this project. I thank Dr. James M. Brooks of Texas A&M University for providing most of the DSDP samples analyzed in this study. I thank Dr. George Claypool of the USGS, Denver for providing the DSDP Leg 84 samples. I thank Dr. Chris Martens of the University of North Carolina, Chapel Hill and the members of his research group "CH4Ao2sn for collecting and providing the Cape Lookout Bight samples analyzed in this study. I thank Suzy Wicks for the discussions we had about microbiology, for her interest and help with this project, and for her friendship. I especially thank my wife Jenny for her love and support throughout my stay at USF, and my little daughter Lisa, who for the past two and a half years has offered a different perspective on things. This work was financially supported by National Science Foundation grants OCE-8002455, OCE-83089115, and OCE-8417321 (Marine Chemistry Program). I also thank the Poynter Fund (Lake Fellowship) and the Graduate Council of USF for their generous support. ii

PAGE 4

TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ABSTRACT INTRODUCTION Natural Hydrogen Isotopic Compositions Natural Carbon Isotopic Compositions Origin of Natural Methanes Anaerobic Decomposition of Complex Organic Matter Methanogenic Pathways Methane Oxidation Natural Gas Classification Objectives iv v vii 1 3 7 8 9 1 4 16 19 22 EXPERIMENTAL METHODS 24 Preparation of Methane for Isotopic Analysis 24 Reduction of Water and oD Determination 26 + H Correction 27 sJmple Measurement Reproducibility 30 Determination of Ancillary Parameters 32 ISOTOPIC COMPOSITIONS OF METHANES FROM FRESHWATER SEDIMENTS 33 Sample Collection 34 Model of Biogenic Methane Formation 34 Results and Discussion 38 METHANE ISOTOPIC COMPOSITIONS IN CAPE LOOKOUT BIGHT SEDIMENTS 51 Sample Collection 54 Results and Discussion 54 METHANE ISOTOPIC COMPOSITIONS IN DEEP-SEA SEDIMENTS 67 Sample Collection 69 Results and Discussion 69 DSDP Leg 84 Middle America Trench Region 69 DSDP Leg 87 -Nankai Trough 73 DSDP Leg 96 Orca Basin 74 SUMMARY 85 LITERATURE CITED 92 iii

PAGE 5

Table 1. Table 2 Table 3. Table 4 Table 5. Table 6. Table 7. LIST OF TABLES Intercomparison of working standards and comparison of working standards with international reference samples (in /oo). oD-CH4 o13c-cH 4 and ancillary data for several freshwater environments and the Tampa Bay estuary. oD-CH4 o13c-cH 4 and oD-H2 0 from Cape Lookout Bight. Isotopic analyses of methane and water from DSDP Leg 84 (in /oo). Location of hydrates and% org. C data from Kvenvolden and McDonald (1985); Site locations, water depths, and sed. rates from von Huene, Aubouin et al. (1985). Isotopic anal6ses of methane and water from DSDP Site 583 (in /oo). Isotopic anal6ses of methane and water from DSDP Site 618 (in /oo). Comparison of isotopic equilibrium temperatures (T ) as calculated with the data of Lyon and (1984) with estimated environmental temperatures (T ). env iv 28 41 56 70 75 77 84

PAGE 6

Figure 1. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 1. LIST OF FIGURES Hydrogen isotope ranges of hydrogenaceous materials (data from: Friedman et al. 1964; Estep and Hoering 1980; Schiegl and Vogel 1970; Yapp and Epstein 1 982; Smith et al. 1983; Redding et al. 1980; Ziegler et al. 1976; Rigby et al. 1 981; Epstein et al. 1976; Smith and Epstein 1970; Yeh and Epstein 1981; Schoell 1980, 1983; Nakai et al. 1974; Woltemate et al. 1984; Ehhalt 1973; Ehhalt and Volz 1976; Begemann and Friedman 1968; L yon and Hulston 1984; Gunter and Musgrave 1971; Welhan 1981; and Cloud et al. 1958). The anaerobic deco mposition of complex organic matter to methane in sulfate-depleted environments. Percentages represent the flow of energy from the organic and inorganic compounds to methane (after Zehnder et al. 1982) Schematic of the system used for the preparation of methane for isotopic analyses. Model of methane formation. Lines represent the percentage mix of acetate dissimilation (AD) and carbon dioxide reduction (R) (after Woltemate et al. 1984). Plot of methane hydrogen isotopic composition versus environmental water hydrogen isotopic composition for several freshwater environments and the Tampa Bay estuary. The lines are the same as in Figure 4 Symbols: Crescent Lake, FL;_ Mirror Lake FL; A Lake Dias, FL; Kilmer Pond SC; Mississippi River Delta, LA; and $Tampa Bay estuary, FL. Plot of methane hydrogen isotopic composition versus methane carbon isotopic composition for several freshwater environments and the Tampa Bay estuary. Symbols the same as in Figure 5. Cape Lookout Bight, North Carolina, USA. The sampling site was Station A-1 (after Martens and Klump 1980). v 4 1 1 25 35 39 44 52

PAGE 7

Figure 8 Figure 9. Figure 10. variations of temperature, oD-CH4 and 6 C-CH4 in Cape Lookout Bight sediments. The temperature data are from Martens and Klump ( 1980 ) Martens et al. (1980), Crill and Martens (1983), and Martens (unpublished). Plot of methane hydrogen isotopic composition versus pore water hydrogen isotopic composition for samples from Cape Lookout Bight and the samples plotted in Figure 5. Symbols : 41 C a pe Lookout Bight; all others the same as in Figure 5 Plot of methane hydrogen isotopic composition versus methane carbon isotopic composition for the samples from Capl 3 Lookout Bight. The equation: 6D = -6.66 C 626 is a linear regression of all the data in this study. The equation: oD = -4.46 C 528 is a linear regression of data obtained for T 24C vi 55 59 64

PAGE 8

STABLE HYDROGEN AND CARBON ISOTOPIC COMPOSITIONS OF BIOGENIC METHANES by Roger Allen Burke, Jr. An Abstract A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Marine Science in the University of South Florida December 1985 Major Professor: Dr. William M. Sackett vii

PAGE 9

Stable hydrogen and carbon isotopic compositions of biogenic methanes collected from the sediments of several deep-sea, nearshore marine-estuarine, and freshwater environments were determined. The isotopic compositions of methane samples from eight different DSDP Sites (mean oD-CH4 = -185 /oo, std. dev. = 7/oo, n = 75; mean o13c-cH 4 0 0 = -71.3 /oo, std. dev. = 6.3 /oo, n = 44) are generally typical of m ethane formed via co 2 reduction in deep-sea sediments. Methane collected from several freshwater environments was 0 0 D-d epleted (mean oD-CH4 = -300 /oo, std. dev. = 26 /oo, n = 20) and 13c-enriched (mean o13c-cH 4 = -60.1/oo, std. dev. = 6.1/oo, n = 20) compared to the deep-sea methane. Normally, acetate dissimilation is thought to account for about 60 to 70% of the total methane production in freshwater sediments. Nearshore marine-estuarine methanes appear to be isotopically intermediate (mean oD-CH4 = -258/oo, std. dev. = 23/oo, n = 46; mean 0 0 -61.8 /oo, std. dev. = 3.1 /oo, n = 46) between deep-sea and freshwate r methanes. Variation in the relative importance of the two main methanogenic pathways, acetate dissimilation and co2 reduction, is prob ably the single most important factor responsible for the differences in methane isotopic compositions among these three different types of environm ents. Other factors that probably contribute to the methane isotopic differences are temperature, sedimentation rate, organic matter type and amount, concentration of alternate electron acceptors, rate of methane formation and possibly viii

PAGE 10

postgenerative isotopic equilibration. Shallow aquatic sediments are thought to be an important source of methan e to the atmosphere; the methane produced in these systems, including the ones sampled in this study, is generally substantially more 1 3 c-depleted than expected based on the o 13 c of atmospheric methane and the isotopic fractionation associated with the atmospheric sink process. T oo few oD data are available to allow evaluation of the role of shallow aquatic sediments in determining atmospheric oD-CH4 Abstract approved: Professor, Marine Science Date of\ approval ix

PAGE 11

INTRODUCTION The commerc ial production of natural gas, o f which m etha ne is usually the predominant co nstituent, accounted for about 27% of the total domestic energy production of the United States in 1975 (Natl. Acad Sci. 1 980) Natural gas is projected to account for about 20% of global e n e r gy consumption from 1990 to 2000 (Anonymous 198 3) The origi n of virtually all commercially important accumulations o f natura l gas is thought to be t h e degradation o f biologically derive d organic matter in aquatic sedim ents, m ediated eith e r by bacteria or b y the combined effects of time, temperature, an d catalysis after sediment burial. In this study, gas produced directly by bacteria is referred to as biogenic gas, and gas resulting from the effects of time, temperature and catalysis on organic matter is referred to as thermogenic gas. Rice and Claypool ( 1981) recently estima t e d that roughly 20% of the world's discovered gas reserves w ere formed by biogenic processes. In v i e w of the fact that biogenic methane occurs in geologically und erstoo d circu mstances, Rice and Claypool (1981) predict that even more than 20% of future gas discoveries will be of biogenic o rigin. Recent studies (Craig and Chou 1982; Khalil and Rasmussen 1982) of polar ice cores indicate an apparent doubling o f atmosp heric methane co ncen t ration s o v e r the past few hundred years. Recent high precision measurements (Khalil and Rasmussen 1983) indicate that the atmosp heric

PAGE 12

2 methane concentration has been increasing by 1-2% per year during the last several years. Because methane strongly absorbs infrared radiation within the atmospheric window (700 to 1400 em 1 ) that transmits most of the thermal radiation from the earth's surface to outer space (Wang et al. 1976), the atmospheric methane increase is a potential contributor to global warming. investigators (Wang et al 1976; Lacis et al. 1981; Craig and Chou 1982) estimate the contribution of methane to increasing global temperatures to be about 20-40% of that attributed to the carbon dioxide increase. Microbial decomposition of complex organic matter in environments such as aquatic sediments and the intestines of ruminants is thought to be the most important source of methane to the atmosphere, accounting for an estimated 75 to 87% of the total input (Sheppard et al. 1 982; Hameed and Cess 1983). Improved understanding of the geochemistry of biogenic methane may be of practical significance to future energy exploration and climatic predictions. Two useful parameters characterizing natural gas are the stable hydrogen and carbon isotopic compositions of methane. The natural abundances of the stable hydrogen isotopes H and D are 99.9844% and 0.0156%, respectively (Hoefs 1980) The stable isotopes of carbon 12c and 13c have natural abundances of 98.89% and 1 .11%, respectively (Hoefs 1980). Stable isotopic compositions are usually reported as a parts-per-mil (0/oo) deviation (6) from a standard with a known isotopic ratio. The d efinition of the 6 value is: 6 = (R(sample)/R(standard) 1) X 1000 (1) where R is D/H and 13c ; 12c for hydrogen and carbon isotopic compositions, respectively. Values reported in this study are referred

PAGE 13

to the Vienna Standard Mean Ocean Water (V-SMOW; absolute 0/H = 155.76 -6 0.05 X 1 0 ; Hagemann et al. 1970; Gonfiantni 1978) for hydrogen and to the belemnite standard f rom the Cretaceous Peede e Formation of South Carolina (POB; absolute 13c;12c = 0 .01 1 2372; Cra i g 1953, 1957) for carbon. Natural Hydr oge n Isot opic Compositions Isotope effects are fundamentally the result of mass differences. Because hydrogen isotopes have such l a rge relative mass differences, large variations of 60 in nature are expected. Figure 1 shows approximate isotopic ranges for v arious types of hydrogen containing substances found in nature. The deuterium concentration of continental surface waters is determined primarily by the temperature at which local precipitation takes place; temperature i s in turn, influenced by latitude and altit u de (Friedman et al. 1 964) In arid regions the process of evaporati o n can lead to significant deuterium enrichment in standing water (Friedman e t al. 1 964) The 60 of the organically bound hydrogen of plants is largely determined by the 60 of moisture absorbed during g r o wth ( Schieg l and V oge l 1970) an d by the pathway o f photosynthetic carbon metabolism (c3 (Calvin-Benson), c4 (Hatch-Slack) or CAM (Crassulacean Acid Metabolism)) operative in t h e plan t (Ziegler et al. 197 6) The total o rganically b ound hydrogen is usually O-depleted relative to the water that the plants use for growth (Schiegl and Vogel 1970 ) with additional deuterium depletion d uring the formation of lipids (Smith and Epstein 1 970; Estep and Hoering 1980) When CAM plants are deprived of water their organically bound hydrogen can become O-enriched, apparently due to e nhanced evaporation of their tissue water (Ziegler et al. 1976) 3

PAGE 14

-1000 -800 -600 -400 -200 0 SURFACE WATERS HIGH LAT. MID. LAT. (20-50) LOW LAT. (0-20) LAND PLANTS -TOTAL CELLULOSE MARINE ORGANISMS TOTAL LIPIDS CELLULOSE FOSSIL FUELS COAL -CRUDE OIL -METHANE HYDROTHERMAL BIOGENIC ATMOSPHERIC THERMOGENIC HYDROGEN GAS HYDROTHERMAL BIOGENIC -1000 -800 -600 -400 -200 0 bD (%o) Figure 1 Hydrogen isotope ranges o f hydrogenaceous materials (data from : Friedman et al. 1964 ; Estep and Haering 1980; Schiegl and Vogel 1 970 ; Yapp_ and Epstein 1982; Smith et al. 1 983; Redding et al. 1980; Ziegler et al. 1976; Rigby et al. 1981; Epstein et al. 1976; Smith a n d Epstein 1970; Y e h and Epstein 1981; Schoell 1980 1 983; Nakai et al. 1974; Woltemate et al. 1984; Ehhalt 1973 ; Ehhal t and Volz 1976; Begemann and Friedman 1 968 ; Lyon and Hulston 1984; G unter and Musgrave 1971 ; Welhan 1981; and Cloud et al. 1958) 4

PAGE 15

5 Based on their early measurements of the oD of coal, Schiegl and Vogel (1970) suggested that there is some deuterium depletion in the conversion of plant material to coal and that some of the variability noted in coal oD might be attributable to climatic conditions at the time of plant growth Based on the results of a previous study by Redding et al. (1980) which indicated that coal oD is not greatly changed by the one-step formation of methane, and their own data, which indicated an inverse correlation between coal oD and maturity, Smith et al. (1983) concluded that coal oD is primarily source-controlled and that maturational effects are unimportant. Smith et al. (1983) assumed that there is essentially no isotope effect in the conversion of plant material to coal and that, following application of an average water-plant fractionation factor, their coal oD data could be used to estimate the paleolatitude of coal deposition. Based on analyses of over 100 crude oil samples and preliminary results from laboratory experiments, Yeh and Epstein (1981) concluded that the oD of oil is primarily source controlled and that maturity and chemical composition have little effect on the oD of normal crude oils. Yeh and Epstein (1981) based their conclusions on ( 1 ) the observation that there was no correlation between oD-oil and API gravity in their 0 sample set, (2) the range of oD-oil of their samples was only 10 0 loo compared to a oD range of more 250/oo for oil reservoir formation waters and, (3) the lack of detectable change in the oD of an oil sample after it had been in contact with water with a oD of about 0 0 2000 /oo at 1 80 C for two months. Laboratory and field studies by other authors, however, indicate that there is likely to be significant isotopic exchange between water and at least some of the fractions of

PAGE 16

oil (Koepp 1978; Schoell and Redding 1978), kerogen (Haering 1968; 1984), and coal (Rigby et al. 1981) during subsurface burial. 0 Field (Schiegl and Vogel 1970; Schoell 1980; 1 983) and laboratory pyrolysis (Chung 1976; Sackett 1978) studies indicate that methane formed from the thermal degradation of precursors (i. e., kerogen, coal, oil) is D-depleted relative to the precursor and that the magnitude of the depletion decreases with increasing precursor maturity. Field and laboratory studies (Nakai et al. 1 974; Schoell 1980; Woltemate et al. 1984) indicate that biogenic methane is D-depleted relative to the water in which the methanogenic bacteria grew and that oD-CH4 is affected by oD-H2o, oD of the precursor organic matter, and the pathway used in methane generation. Available measurements (Bainbridge et al. 1961; Begemann and Friedmann 1968; Ehhalt 1973) indicate that atmospheric methane is greatly D-enriched compared to its known sources. There appears, however, to be a large kinetic hydrogen isotope effect associated with the destruction of methane in the atmosphere by reaction with hydroxyl radicals (Senum and Gaffney 1985). Lyon and Hulston ( 1984) listed four possible origins for the methane in hydrothermal systems: (1) primordial or abiogenic methane from the mantle (i. e., Welhan and Craig 1 979; Gold and Soter 1980); (2) the high temperature decompositi on of organic matter (i.e., Welhan 1981; Burke et al. 1981); (3) chemical reaction between carbon dioxide and hydrogen (Fisher-Tropsch reaction, i.e., Hunt 1979; Giggenbach 1980) and (4 ) biological production (i. e., Burke et al. 1981; Baross et al. 1982; Baross and Deming 1983). Due to rapid isotopic equilibration between hydrogen-water and hydrogen-methane, any oD differences in

PAGE 17

methane and hydrogen formed by these processes are obscured, and oD-H 2 0 and oD-CH4 reflect the temperature (100-1200 C) of the water at the time of sampling and Musgrave 1971; Welhan 1981) Natural Carbon Isotopic Compositions The distribution of the stable carbon isotopes in natural 7 substances has been the subject of many reviews (i.e., Craig 1953; Fuex 1977; Deines 1980; and others) to which the interested reader is referred for more detail than will be presented here. Marine limestones typically have the greatest relative concentration s of 13c, with o 13 c values generally between 5 and +5 loo (Fuex 1977) Terrestrial plants that utilize the c 3 pathway are generally 13C-depleted ( -32 to -22/oo) relative to terrestrial plants fixing carbon dioxide via the c 4 pathway ( -1 8 to -9/oo; Deines 1980). Terrestrial plants that are known to use the CAM pathway exhibit o 13c that span the range of c 3 and c 4 plants (-32 to -11/oo) as a result of their unique ability to accomplish net carbon dioxide fixation via either the c 3 or c 4 pathway (Deines 1980). Most marine phytoplankton from tropical and temperate waters have a o 13 c of about 20 loo (Sackett in press). The o 1 3c of coals, crude oils, and kerogens are largely inherited from their precursors and are less apt to change as a result of maturational processes than are oD (Rigby et al. 1 981; Yeh and Epstein 1981). Crude oils (mode -21 to -30/oo) are generally 1 3c -depleted relative to coal (mode -23 to -26/oo) and kerogen, probably as a result of preferential retention o f isotopically light lipid components during oil formation (Deines 1980). Methane produced from the thermal decomposition of organic matter generally has a o 1 3c ranging from 60

PAGE 18

8 to -25/oo, whereas biogenic methane is generally more 13 c-depleted ( 90 to -50/oo; Fuex 1977) The o13c of the abiogenic methane emitted from the East Pacific Rise ranges from -17.6 to -15/oo (Welhan 1981) Origin of Natural Methanes Interest in primordial or abiogenic methane has increased recently as a result of observations that methane along with other mantle-derived gases ( 3He, co2 H 2), is injected into the deep sea from active spreading centers (Welhan and Craig 1979; Welhan 1981). These authors stated that it is uncertain whether the methane is an intrinsic mantle component that is extracted from the basalts by the circulating seawater or whether it represents a chemical equilibrium with co 2 and H 2 (Fisher-Tropsch reaction) under high pressure and decreasing temperatures (Welhan and Craig 1 979; Welhan 1981) Gold and Soter (1980) suggested that degradation of complex hydrocarbons, formed at the time of earth' s accretion under conditions of high temperature and pressure, is responsible for this deep-earth gas. Gold and Soter (1980) further suggested that the quantity of methane produced in thiS manner is enormous and that methane from this source is responsible for most of the gas and oil (through polymerization) in commercial accumulations. Most geochemists, on the other hand, think that thermal and biological decomposition of photosynthetically derived organic matter in marine or lacustrine sediments accounts for virtually all significant accumulations of natural gas. Early diagenetic (sed. T < 50C) reactions, primarily mediated by microorganisms, convert labile chemica l groups to gases (i.e., C02 CH4 ) and cause the remaining organic matter to be converted into a complex polymeric form termed

PAGE 19

kerogen. At temperatures between 50 and 200C (catagenesis) the combined effects of time, temperature, and catalysis lead to the formation of gas and oil or coal from the kerogen (Hunt 1979) Between temperatures of about 60 and 150C oil and some gas is formed by catalytic and thermal cracking of kerogen (Hunt 1979). Peak methane generation from kerogen, oil, and coal occurs between about 150 and 200C, with only a small amount of methane generation at higher (metamorphic) temperatures (Hunt 1979). In catalytic cracking, which is believed to be the dominant process in petroleum generation at 0 temperatures of up to about 125 C (Hunt 1979), positively charged, highly reactive, carbonium ions are generated by hydride removal from paraffin molecules a t Lewis acid centers of clay minerals (Sackett 1978). Carbonium ion reactions can then give rise to a large variety of products. In thermal cracking, which becomes increasingly important at temperatures above 125C (Hunt 1979), the reactive intermediate is a free radical that arises from the homolytic cleavage of a carbon to carbon bond in which each carbon retains one of the two shared electrons. As with carbonium ion reactions, a wide spectrum of products is possible when free radicals are created. Biogenic methane is produced by a specialized group of obligately anaerobic bacteria that are limited to a narrow range of low molecular weight substrates for growth and e nergy production. Known methanogenic substrates include carbon dioxide a nd hydrogen, acetate, methanol, and methylamines (Zehnder e t al. 1982). Methanogenic bacteria thus rely on other bacterial species to decompose complex organic compou nds to substrates suitable for their growth. Anaerobic Decomposition of Complex Organic Matter 9

PAGE 20

10 The microbial degradation of complex organic matter to carbon dioxide and methane in anaerobic environments such as organic-rich sediments and sewage sludge digesters is a multi-step process requiring the participation of at least three different trophic groups of bacteria (Zehnder et al. 1982). In environments in which carbon dioxide and protons are the only available inorganic electron acceptors, this process is believed to proceed as illustrated in Figure 2. In the first step, biological polymers such as polysaccharides, proteins, and lipids are digested by extracellular enzymes to soluble materials that are then fermented to organic acids and alcohols by the acid-forming or chemoheterotrophic fermentative hydrogen-producing bacteria (Brock 1974; Mah 1982). In the next step organic acids larger than acetate and alcohols larger than methanol are converted to acetate, carbon dioxide, and hydrogen by the chemoheterotrophic non-fermentative obligate proton reducers (Mah 1982). The reducing equivalents (hydrogen) and acetic acid generated during the first two steps are disposed of in the final step by the methane-producing bacteria. The pathways for oxidation of organic compounds and ATP generation under anaerobi. c conditions can be divided into two major groups, fermentation and anaerobic respiration (Brock 1 974) In fermentation a compound at an intermediate oxidation level is disproportioned both to compounds that are more oxidized and compounds that are more reduced than the starting substrate in the abscence of an external electron acceptor. For example, the fermentation of glucose by Ruminococcus albus produces ethanol, acetate, hydrogen, and carbon dioxide (Wolin 1 979) Fermentation releases only a small part of the available

PAGE 21

BIOLOGICAL POLYMERS HYDROLYTIC AND FERMENTATIVE MICROORGANISMS ALCOHOLS, CARBOXYLIC ACIDS (except acetate) OBLIGATE PROTON REDUCERS Figure 2. The anaerobic decomposition of complex organic matter to methane in environments. Percentages represent the flow of energy from the organic and inorganic compounds to methane (after Zehnder et al. 1982). 11

PAGE 22

12 substrate energy; e.g., in the fermentation of glucose albus about 90% of the glucose energy remains in the end -products (Mah 1982). In anaerobic respiration, an external electron acceptor other than oxygen (i.e., nitrate, sulfate, protons, or carbon dioxide) is involved in the oxidation of organic compounds for energy production (Brock 1974). Examples of anaerobic respirations involved in the degradation of complex organic matter (Figure 2) are the production of acetate, hydrogen, and carbon dioxide by the obligate proton reducers, and methane production from carbon dioxide and hydrogen. Without the anaerobic respiration of the obligate proton reducers a significant portion of the complex organic matter would be only partially degraded to higher fatty acids and alcohols (Figure 2) Because degradatio n of these higher fatty acids and alcohols requires the use of enzymes that are strongly inhibited by hydrogen (Wolin and Miller 1982), the obligate proton reducers can operate only if methanogens, or some other hydrogenotroph, efficiently consume the hydrogen they produce. This process, referred to as interspecies hydrogen transfer (Iannott i et al. 1973; Wolin and Miller 1982), may also ensue when chemoheterotrophic fermentative hydrogen-producing bacteria are cultured with hydrogenotrophs (Wolin 1982) In this case interspecies hydrogen transfer is not obligatory because the fermentative organism has an alternate means of electron disposal. In monocul ture, Selenomonas rumina tum fermented lactate to propionate, acetate, and carbon dioxide (Chen and Wolin 1977). Addition of s. ruminatum to a pregrown culture of the methanogen Methanobrevibacter smithii (formerly Methanobacterium ruminatum) by a sequential transfer procedure caused a shift in lactate fermentation products to more

PAGE 23

1 3 acetate and less propionate with concomitant production of methane from carbon dioxide and hydrogen (Chen and Wolin 1977). Thus, interspecies hydrogen transfer is important to the anaerobic degradation process (Figure 2) not only because it creates an environment that is conducive to the growth of the obligate proton reducers, but also because it can cause a shift in the electron and carbon flow (Mah 1982 ) Electron flow is diverted away from the soluble reduced electron sink products (i.e., propionate) to proton reduction (hydrogen formation) with subsequent methane production, and carbon is channeled into the production of methane and the more oxidized soluble end products (i.e. acetate) at the expense of the more reduced soluble compounds such as propionate (Mah 1982). This is advantageous to the hydrogenotrophs ( esp.ecially methanogens) because they gain energy sources (hydrogen, acetate), a nd often to the hydrogenogens because they may obtain more ATP from their substrates due to the relatively more complete oxidation of the starting substrate (Wolin 1982). Other electron acceptors such as nitrate, iron and manganese oxides, and oxidized sulfur compounds are preferentially used (compared to protons and carbon dioxide) in anaerobic respirations (Zehnder et al. 1982), a nd when they are present in sufficient quantities carbon and electron flow (Figure 2) may be diverted away from methane production. Because of its high concentration in seawater, sulfate is the predominant inorganic electron acceptor in upper, anoxic marine sediments (Fenchel and Blackburn 1979; Reeburgh 1983) Consequently, many studies (e.g., Martens and Berner 1974; Claypool and Kaplan 1974 ; Ni kaido 1977) f o und that optimum rates of methane production and near-saturation methane concentrations are not encountered in marine

PAGE 24

sediments until sulfate is largel y (80-90%) depleted. This inhibition of methane production by high concentrations of dissolved sulfate is apparently due to the ability of sulfate-reducing bacteria to outcompete methanogens for substrates such as hydrogen and acetate (Winfrey and Zeikus 1977). Claypool and Kaplan (1974) proposed that sulfate reducers gain competitive advantage by virtue of the fact that sulfate reduction yields relatively more energy than methane production; however, McCarty (1972) argued that if two reactions are energetically feasible, then from a strictly thermodynamic viewpoint, one reaction can not inhibit the other simply because it is energetically more rewarding. Recent studies (Kristjansson et al. 1982; Schonheit et al. 1982) have shown that sulfate reducing bacteria have higher substrate affinities (lower K ) by factors of 5 and 15 for s hydrogen and acetate, respectively, than do methanogens. This difference in substrate affinities does not imply that methane production and sulfate reduction from these substrates are mutually exclusive (Kristjansson et al. 1982; Schonheit et al. 1982), but it does provide a satisfactory explanation for the vertical separation of the sulfate-reducing zone from the methane-producing zone that is generally observed in marine sediments (Claypool and Kvenvolden 1983) Methanogenic Pathways In sulfate-depleted marine sediments, as well as freshwater sediments, the dominant terminal process of anaerobic organic matter decomposition is methane production. Methanogenesis is generally 14 thought to occur mainly via two pathways, co2 reduction (2) and acetate dissimilation (3): + 8e + + ( 2)

PAGE 25

+ ( 3) The results of 4c labeling studies implied that acetate normally accounts for about 60 to 70% of the total methane production in sewage sludge digesters (Jeris and McCarty 1965; Smith and Mah 1966), paddy soils (Koyama 1963; Takai 1970), and some freshwater lake sediments (Cappenberg and Prins 1974; Winfrey and Zeikus 1979; Lovley and Klug 1982), with co2 reduction responsible for the remainder. In other freshwater lake sediments, however, the co2 reduction pathway is apparently predominant, accounting for about 90% of the total methane production in Russian Lake Kuznechikha (Belyaev et al. 1975) and about 75% of the methane generated in the sediments of Blelham Tarn in the English Lake District (Jones et al. 1982). Incubation of sediment samples from Cape Lookout Bight, North Carolina, a small, nearshore marine basin with rapidly accumulating, organic-rich sediments, with 15 ( 1 2-14c) sodium acetate indicated that more than 50% of the Cape Lookout Bight summertime methane flux could be accounted for by acetate dissimilation (Sansone and Martens 1981). If sediments receive a fairly specific type of organic matter input or if methanogens are competing with sulfate-reducing bacteria for H 2 and acetate, a significant fraction of the total methane production can result from other substrates such as methanol and trimethylamine (Oremland et al. 1982; King et al. 1983). All of the aforementioned studies used 14c labeled substrates as tracers to estimate the relative contributions of the methanogenic pathways to total m ethane production. An alternate method was proposed for estimating relative pathway contribution based on measurements of

PAGE 26

16 3 1 + when dissolved inorganic nitrogen (DIN = NH4 + + N03 -) concentration s are low bu t not when DIN concentrati o ns are above 1 5 to (Rudd et al. 1976; Sansone and Marte ns 1978). The aerobic oxidation process may require relatively high methane concentrati o n s For exa m p l e methane co ncentratio n s l imited the

PAGE 27

17 rate of methane -oxidizer population development in Lake Mendota Wisconsin (Harrits and Hanson 1980) and Sansone and Martens (1978) proposed that aerobic _methane oxidation in Cape Lookout Bight waters should be limited by other factors at methane concentrations Pure cultures of methanogenic bacteria (Zehnder and Brock 1979), freshwater lake sediments, and sewage sludge (Zehnder and Brock 1980) anaerobically converted 1 4cH 4 to 1 4co2 while producing methane; however, net methane production was not observed and the methane oxidation process was inhibited by the same conditions that inhibited methane production (i.e., presence of oxygen) implying that methanogenic bacteria were responsible for the methane conversion. Profiles of methane concentration in anoxic marine sediments frequently show extreme upward concavity in the zone of sulfate reduction (Barnes and Goldberg 1976; Reeburgh 1976; Martens and Berner 1977; Bernard 1978; Alperin and Reeburgh 1984 ) Barnes and Goldberg (1976) and Reeburgh (1976) suggested that net consumption of upward diffusing methane in the sulfate-reducing zone is required to explain the extreme upward concavity. Further, successful fits of diagenetic models to methane profiles from various anoxic marine sediments were < obtained only if the models included a methane consumption term (Martens and Berner 1977; Bernard 1978; Alperin and Reeburgh 1984 ) Reeburgh (1976, 1980) proposed that methane consumption in anoxic marine sediments primarily takes place in a relatively thin layer (estimated to be about 10 em in Cariaco Trench sediments) at the base of the sulfate-reducing zone where the profiles of methane, sulfate, and EC02 all typically exhibit slope changes. Several recent studies (Reeburgh 1980; Devol 1983; Alperin and Reeburgh 1984; Iversen and

PAGE 28

Jorgensen 1985) demonstrated the conversion of 14cH4 to 14co2 in relatively thin layers (about 15 em to 40 em thick) near the base of 18 the sulfate-reducing zone of various marine sediments. The location of the anaerobic methane consumption layer at depths where the concentrations of both sulfate and methane are appreciable, the demonstrated ability of the sulfate reducer Desulfovibrio desulfuricans to co-metabolize under anaerobic conditions (Davis and Yarbrough 1966), and the co-occurrence of peaks in methane oxidation rate and sulfate reduction rate (i.e., Iversen and Blackburn 1985), have been cited as evidence that anaerobic methane consumption is coupled to sulfate reduction, either directly (i.e., Reeburgh 1976) or indirectly via reactions coupled to other substrates (i.e., Devol 1983) The anaerobic conversion of 14cH4 to 14co2 in minimally disturbed marine sediments from the main anaerobic consumption zone of Skan Bay, Alaska was not inhibited by specific inhibitors of either methanogenesis (2-bromoethanesulfonic acid) or sulfate reduction (molybdate), implying that anaerobic methane consumption in marine sediments is not directly mediated by either methanogenic bacteria or sulfate-reducing bacteria (Alperin and Reeburgh 1985). Alperin and Reeburgh (1985) suggested that their results are consistent with two possibilities: anaerobic methane consumption in marine sediments is either mediated by an unknown organism, which possibly uses iron oxides, manganese oxides, or reduced sulfur compounds as an external electron acceptor, or methane consumption is accomplished by a bacterial consortium that includes an unknown organism capable of converting methane to carbon dioxide and hydrogen and sulfate-reducing bacteria that consume the hydrogen. Alperin and Reeburgh ( 1985)

PAGE 29

19 pointed out that anaerobic methane consumption would not be inhibited by molybdate if other hydrogen-consuming organisms replaced sulfate-reducers in the consortium. It is not clear, however, why this modified consortium does not function at greater depths where methane concentrations are e ven great er. The confinement of substa ntial methane consumption t o a thin layer implies that some component of the sulfate-reduction zone perhaps som e trace product of the sulfat e reduction process, is directly involved in anaerobic methane consumption. The impo rtance of iron and manganese oxides in anoxic sedi ments to the microbial decomposition of organic matter in general and methane in particular, is unknown because these metal oxides may be used as an e lectron acceptor directly by orga n isms o r by metabolic prod ucts (i.e. sulfide) of organisms ( Burdige and Nealson 1985), including a variety o f organic substances (Stumm and Morgan 1 981) Natural Gas Classificati on Isotopic and molecular compositional variations have been used frequently to develop models that describe the origin and subsurface history of natura l gas. All of these models assume that the observed isotopic and compositional variations of natural gases predominantly reflect processes (primarily kinetic isotope effects and source maturity) that occur d urin g gas formation (Schoell 1 983) althou gh this is not always explicitly stated. Galimov ( 1 968) presented data tha t indicated methan e becomes less 13c-depleted with increasing depth in the sediment, with most of the c h ange occurring during the initial stages of burial. To explain his data, Galimov (1968) develope d a model in which the sediment column was divided i nto three zones: (I) the biochemical zo n e (T < 25C) in which

PAGE 30

20 bacteria mediate the production of methane and its isotopic equilibration with carbon dioxide, (II) the catalytic zone (25C < T < 200C) in which methane is l3C-depleted relative to its precursor due to temperature-dependent kinetic isotope effects but is not in isotopic equilibrium with carbon dioxide because temperatures are too low for sufficient inorganic co 2 -cH4 isotopic exchange and bacterial activity ( ) ( C < T < 300C) 1 n is practically absent and III the thermal zone 200 which temperatures are sufficient for control of o13 c -cH4 by inorganic C02 -cH 4 isotopic equilibration. Stahl ( 1974) presented a model in which three separate stages of gas formation are apparent from a plot o f o13c-cH 4 versus c 1/[Cn (ratio of methane over all saturated gaseous hydrocarbons). In the initial stage the gas is formed predominantly by bacterial action and is characterized by a o13 c -cH4 ranging between about 80 and -68/oo and a c 1/[Cn near unity. The second stage represents thermocatalytic alterati on of marine source material as it is subjected to higher temperatures and time during deeper burial. Gas formed during the second stage maturational trend is characterized by progressive increases in o13c-CH4 from about -68 to 42/oo and progressive decreases in the c,;[cn ratio from a bout 1.0 to abou t 0.8. Upon further burial the source material and its hydrocarbo n products are subjected to increasing thermal stress; this results in gas that continually becomes more methane-and 13c-enriched. Gas formed in this third stage is typified by o13c-cH 4 between -42 and -35/oo and a C1/[Cn ratio of 0.8 to 1.0 (Stahl 1974). The results of Stahl (197 4 ) also indicate that o 13 c -CH 4 is dependent upon the type of source material from which it originates. Methane produced from predominantlY

PAGE 31

21 terrestrial source material during coal formation displays the same trends as noted above for marine source material, but with less 13c-depletion as high as -20/oo; Stahl 1974). The apparent decrease in c 1 /ECn with increasing thermal maturity during the second stage observed by Stahl (1974) is at odds with the laboratory pyrolysis results of Sackett ( 1978) that indicated the opposite trend, i.e., methane initially accounts for a small (<10%) fraction of volatile pyrolysis products and becomes increasingly more important with longer pyrolysis. Sackett's ( 1978) experiments were carried out at higher temperatures (400-500C) than would be found under natural conditions and, as pointed out by Bernard (1978), may not accurately simulate natural processes. Also, Hoering (1984) argued that because water is ubiquitous in sediments, dry pyrolysis experiments may be unrealistic in that some hydro gen transfer reactions in which water plays a role under natural conditions may be blocked. On the other hand, the carbon and hydrog e n isotopic compositions of several natural methanes associated with crude oils were qualitatively comparable to compositions that would be predicted from the results of Sackett (1978) implying that the dry pyrolysis process bears a close resemblance to gas formation in nature (Schoell 1980). Bernard et al. ( 1977) proposed that natura l gases could be efficiently characterized by a model in which c13 c-cH 4 was plotted against the ratio c 1/(C2 + c 3 ) (methane divided by the sum of ethane and propane) Bernard (1978) pointed out that molecular ratios expressed in this way vary by more than four orders of magnitude for natural samples, so this parameter should yield greater resolution than the C 1 /ECn ratio of Stahl ( 1974). Bernard et al. ( 1977) suggested

PAGE 32

22 that, in general, biogenic gas has a c1;(c2 + c3 ) ratio greater than 1000 and a o13c -cH4 more negative than -60/oo, whereas thermogenic gases generally have G1/(C2 + c3 ) ratios of less than 50 and o13c -cH4 0 greater than 50 /oo. The Bernard et al. (1977) model also takes into account the isotopic and compositional changes, with comcomitant loss of resolution, that may result from substrate depletion during biogenic methane formation, the mixing of biogenic and thermogenic gases, and subsurface migration of thermogenic gas. Schoell (1980, 1983) developed a more extensive gas characterization model in which o13c-cH4 is separately correlated wit h 13 the three parameters: 6D-CH4 6 C-C2 H 6 and c2+ (concentration of saturated gaseous hydrocarbons larger than methane). Use of the two 13 new parameters 6D-CH4 and 6 C C 2 H 6 made possible in part by recently improved analytical capabilities, allows a clearer distinction to be made between different genetic types of gas in some cases. For example Schoell (1980 1983) suggested that measurement o f 6D-CH4 might allow differentiation of biogenic gas produced under marine conditions from biogenic gas produced in a terrestrial system when o13c-CH alone is insufficient to make the distinction. 4 The isotopic and compositional variations that may result from secondary processes such as oxidation, mixing, and migration are also indicated in the Schoell ( 198 3) model. For example, migration may appreciably change the c2+ concentration but would generally not be expected to alter significantly the isotopic composition of methane (Schoell 1 983) Objectives My research had the following two objectives: (1) to add to the meager 6D-CH4 data base by measuring methane samples from a variety of

PAGE 33

23 aquatic environments such as deep-sea, nearshore marine-estuarine, and freshwater lake sediments. The resulting inhanced understanding of natural 6D-CH4 variations should then be of use in the improvement of natural gas characterization models and, in conjunction with concurrent o13c-cH4 measurements, eventually aid in gaining a clearer understanding of the relative contributions of the major methane sources to the atmosphere; (2) to use 6D-CH4 measurements to estimate the relative importance of acetate dissimilati on and co2 reduction to methane production in various aquatic sediment types. In the past, pathway importance was generally estimated from the results of 14c radiotracer studies. T o accomplish these objectives, it was necessary to construct a preparation system for the separation of methane from other gases and its conversion to products suitable for stable isotopic measurement The details of such a system h ave not been previously published and few investigators have made oD-CH4 measurements.

PAGE 34

24 EXPERIMENTAL METHODS Preparation of Methane for Isotopic Analysis Methane gas samples were prepared for isotopic analysis with the system illustrated in Figure 3 Preparation involves gas chromatographic (GC) separation of methane from other gases followed by combustion to co2 and water in a vacuum line. The GC used is a Hewlett-Packard 5710A equipped with a thermal conductivity detector and two 6mm OD stainless steel columns (3m grade 12 silica gel and 2m molecular sieve 5A) connected in series through a Valco ten port switching valve configured for column sequence reversal and backflush of the silica ge l column to the detector. This GC configuration allows baseline separation of H 2 o2 N 2 CH4 and co2 The helium carrier gas flow rate is 50ml/min and the oven temperature is held constant at 90C. Methane is combusted in a cupric oxide furnace (850C) to which 5-1 0 torr of oxygen is added before each run, as it is swept through the furnace by the GC carrier gas. Addition of oxygen to the system before each run is necessary to assure complete combustion of the methane. The isotopic composition of incompletely combusted methane was found to be significantly lighter than the actual composition; 0 3.5 /oo, respectively, were observed. The C02 and water resulting from the combustion of methane are condensed by liquid nitrogen ( 196C) in the trap immediately downstream from the combustion oven the excess oxygen

PAGE 35

S WITCHING VALVE tVACUUM VACUUM VACUUM \ GAUGE --H 2 0 COLLECTION C02 COLLECTION VENT-Figure 3. Schema tic o f the system used for the prep aratio n of methane for isotopfc a n a lyses. 02JN f\) V1

PAGE 36

26 and helium are pumped away, and the co2 and water are cryogenically (-89C) separated. The a13c of the co2 is then measured in a Finnigan nee Varian MAT 250 iso_tope ratio mass spectrometer (IRMS) by comparison With the NBS-21 standard (a13c = -28.16/oo; Schoell et al. 1983) Reduction of Water and aD Determination The water from the combustion of methane is then collected in a sample tube containing approximately 0.25g of treated zinc shot and reduced to H 2 gas by reaction with the zinc at 450C for 30 minutes (Coleman et al. 1982). The sample tube consisted of either a 15cm length of 9mm OD Pyrex tubing that was sealed at one end or a high vacuum teflon plug, 90 pattern stopcock (Model 8195-47 Ace Glass Inc., Vineland, NJ) with one arm sealed. The reducing agent used is AnalaR zinc shot (BDH Chemicals Ltd., Poole, England; presently obtainable from Gallard/Schlesinger Chemical Mfg. Corp., New York) that was sieved so that it passes through size 14 mesh but is retained by size 60 mesh. After sieving, the dull, gray oxide coating is removed from the zinc shot by reaction with dilute nitric acid until the zinc becomes quite shiny. This activated zinc is then washed several times With distilled water, dried and outgassed under vacuum overnight at about 300C. When not in use, the treated zinc is stored under vacuum The aD of H 2 gas is measured in the IRMS by comparison with a H 2 gas of known 6D. Through the assistance of Dr. Thomas Hoering of the Geophysical Laboratory, Carnegie Institution of Washington several high pressure cylinders of H 2 gas of differing 6D were prepared for use as routine working standards. Deuterium-depleted methane samples were always compared to Tank Standard-C (TS -C), which was assigned a oD of 220 .4/oo based upon comparison with the water reference sample

PAGE 37

Greenland Ice Sheet Precipitation (GISP) and Tank Standard-A (TS-A) which was assigned a oD of -31.9/oo from comparison with the NBS-1 water reference sampl e (Table 1) Comparison of these two tank standards with a third (TS-B) indicated that at least the isotopic difference between TS-A and TS-C is correct (Table 1) When isotopically heavier water samples wer e run, TS-A was used as the working standard. Each day that samples were run, the tank standard serving as the working standard was used in determining the oD of at least one of the other tank standards. The isotopic differences (TS-A vs. TS-C (242.4/oo) and TS-C vs. TS-A (-194.1/oo) were not quite internally consistent (Table 1) implying that H 2 gas admitted to the mass spectrometer through the sample-side of the inlet system is fractionated by about 0 .6/oo relative to gas flowing in through the standard-side capillary. Because the magnitude of this apparent fractionation is comparable to the uncertainty involved in determining the oD of H 2 gas with the IRMS and is small compared to the overall uncertainty involved with oD analysis of water (Table 1) and methane (see below) and would have no effect on any interpretations made here, n o correction for this effect has been applied to the data presented here. H + 3 Correction When mixtures of H 2 and HD gas a r e ionized in the mass 27 + + + spectrometer source, H 2 and HD 1ons are formed. Reaction of H 2 with + H 2 also results in the formation of H 3 ions that cannot be resolved from HD + ions by the mass spectrometer used in this study. Because d t + / + + e ermination of the ratio HD H 2 1s desired, the contribution of H 3 + to the mass-3 signal must be removed In this study the H 3 correction

PAGE 38

2 8 Table 1. Intercomparison of working standards and comparison of working standards with international reference samples (in /oo). Sample Tank Standard A Tank Standard c 00TS-A 00SMOW 00TS-C 00SMOW TS-A -----------242 4 3 1.9 (aO o7, n) TS-B -83o 1 -112o4 138o8 -112.2 (a .. o.1, n=7) (aO o6, n) TS-C -194o1 -220oll ----------(a=Oo5, n"'42) V-SMOW 31.3 -1.6 -----------(a=1o 2 n=21) NBS-1 -15.7 -47 01 ------------( a-1 o 1 n=16) GISP ------------39.8 -189o 4 (a = 2 o3, n)

PAGE 39

29 was determined by the method of Nier (1947) using software supplied with the HP 9815A calculator that controls operation of the IRMS. + Because the H3 rate production is proportional to the square of the + hydrogen partial pressure, whereas the.rates of generation of the H 2 + and HD ions are directly proportional to the hydrogen partial pressure, measurement of the mass-3/mass-2 ratio (i3 +1i 2+) at different + pressures (proportional to i 2 ; Friedman 1953) yields a straight line (Nier 1947). + The product of the slope of this line and i 2 at the pressure at which an isotopic analysis is made is the contribution to + + + the ratio i 3 / i 2 resulting from the formation of H 3 and is simply + + subtracted out to yield the desired i (HD)/i (H2 ) ratio. Determination of this slope or H 3-factor requires a minimal investment of time (five minutes) and operator effort; this makes periodic redetermination practical if it is suspected that IRMS operating conditions have fluctuated. + The H 3 contribution is large (about 30%) at typical operating pressures employed here (mass-2 signal > 5V or sample size > 2cc(STP) H 2 ) but is fairly constant (within about 1 %) It can be calculated that a 1% fluctuation in the H 3-factor should cause an uncertainty of about 0.5/oo when a sample-standard pair differing in 0 cSD by about 200 loo is compared. This is close to the long term uncertainty that was observed for the comparison of TSA to TSC (Table + 1) further implying that the H 3 contribution remains constant to within 1% under typical operating conditions. Because the sample-standard pairs measured in this study were generally much closer in cSD than 200/oo, measurement uncertainty due to H 3-factor fluctuation should generally be less than 0. 5 /oo. At very low operating pressures the Nier ( 1947) relationship may deviate from

PAGE 40

30 linearity (Whelhan 1981) and may not be applicable when samples much smaller than the smallest measured here (about 2cc(STP) H 2 ) are analyzed. Sample Measurement Reproducibility The GC-combustion system (Figure 3) described here is a flow-through system in which methane is combusted as it is swept through the furnace by the GC carrier gas. There is no Toepler pump or bellows to cycle the gas repeatedly through the furnace. As a result, it is reasonable to expect that the time the methane spends in the oven which is determined by the GC carrier gas flow rate, could be an important variable. To test this, a methane working standard was analyzed several times using carrier gas flow rates of 25ml/min and 50ml/min. Isotopic values obtained at 50ml/min (mean OD-CH4 = 0 0 13 0 -164 /oo, std. dev. = 1.0 /oo, n = 12; mean o C-CH4 = -44. 2 /oo, std. 0 dev. = 0 .14 /oo, n = 12) were not significantly different from those obtained at a GC flow of 25ml/min (mean oD-CH4 = -164/oo, std. dev. = 1 .7/oo, n = 6; mean o 13 c-cH 4 = -44.1/oo, std. dev. = 0.17/oo, n = 10). Repeated analyses of a second methane working standard, analyzed daily during the course of this work, yielded standard deviations (mean oD-CH4 = -189.3/oo, std. dev. = 1.7/oo, n = 36; mean o 1 3c-cH 4 0 -39.6 /oo, std. dev. 0 21/oo, n = 32) that are similar to those obtained above for the GC flow rate test and are comparable to those reported by Schoell (1980). The combustion efficiency of the GC-combustion system (Figure 3) was tested by comparing the methane concentration of the gas atmosphere in the vacuum line resulting from the injection into the GC of 2 ml of methane under normal operating conditions (three separate trials) to

PAGE 41

31 the methane concentration of the vacuum line gas atmosphere after injection of 2 ml of methane into the GC with the combustion furnace turned off (three trials). The trap immediately downstream from the combustion oven was immersed in a Dewar flask containing an isopropyl alcohol slush (-89C) The resulting meth a ne concentrations, which were measured with a GC equipped with dual flame ionization detectors, indicated a combustion efficiency >99%. Water samples were prepared for oD determination by the zinc metal method (Coleman et al. 1982). In this case, water samples were simplY injected into the sample tubes with a syringe as described by Coleman et al. ( 1982). The long term accuracy and precision of oD-H2o determination are indicated by the analyses of water reference samples shown in Table 1. The oD-H2o value obtained here for NBS-1 by comparison with TS-A is essentially identical to that listed by Gonfiantini (1978). The GISP oD value of -189.4/oo determined by comparison with TS-C in this study is within analytical uncertainty of the value (-189.8/oo) reported for GISP by Gonfiantini (1981). The water reference sample V-SMOW was prepared from distilled Pacific Ocean water (Gonfiantini 1978) such that its oD would match the SMOW standard proposed by Craig (1961) as closely as possible. According to Hagema nn et al. (1970) V-SMOW is about -0.1/oo (relative to SMOW), whereas Coplen and Clayton (1973) obtained a oD of -1.2 0.4/oo (relative to SMOW) for V-SMOW and stated that the reason for the discrepancy is not known. Due to this apparent uncertainty in the oD of V-SMOW, the good agreement of TS-A and TS-C with NBS-1 and GISP, respectively, and the uncertainty (o = 1-2/oo) involved with determining oD-H2o, I have elected not to redefine the oD of TS-A such that the oD of V-SMOW

PAGE 42

32 0 0 measured here (-1 6 loo; Table 1 ) becomes 0 /oo. Instead, the 60 of water samples analyzed by comparison with TS-A and with a 60 near zero were corrected by adding 1 .6/oo to their measured 60. Possible errors associated with this method of correction are relatively s mall and have no significant effect on any interpretations made in this study. Determination of Ancillary Parameters Preparation of carbon dioxide gas for analysis by IRMS invol ved cryogenic separation of the co2 from the other gases. This was accomplished by injecting an aliquot of the sample gas through a rubber septum into a vacuum line with a trap cooled by liquid nitrogen (LN2 ) and then simply pumping away the LN2 non -condensable gases (02 N 2 CH4 ) Water vapor was then removed cryogenically (-89C) from the co2 Sedimentary organic matter was prepared for 6 13c analysis by IRMS in a Craig type combustion apparatus (Craig 1953). The methane and carbon dioxide concentrations (mol%) of the gas samp les were calculated from the GC-integration units (HP-3390A Integrator) following application of appropriate T. C. Weight Factors as given by McNair and Bonelli (1969).

PAGE 43

31 ISOTOPIC COMPOSITIONS OF METHANES FROM FRESHWATER SEDIMENTS Biogenic methane is a common constituent of anoxic sediments of freshwater environments such as lakes and swamps when large quantities of organic matter are vigorously decomposed (Figure 2) When methane production rates are high enough to supersaturate the pore waters of shallow sediments, bubbles may form and be released to the overlying water and then to the atmosphere. Few measurements of the isotopic compositi on of biogenic methane formed in freshwater sediments are available in the literature. Oana and Deevey (1960) determined the o1 3c of methane and carbon dioxide obtained in the gaseous state following agitation of the sediments of several Connecticut lakes. The methane was 13c-depleted (6 1 3c-cH 4 ranged from -80 to -57/oo) relative to both the gaseous carbon dioxide (o13c-co ranged from -25 to -4/oo) and the sedimentary organic matter 2 (6 1 3c -soM ranged from 33 to -27/oo; Oana and Deevey 1960). Stevens and Rust (1982) reported that methane collected from the bottom of a small c 3-plant dominat ed slough in Northern Illinois had a o1 3c of -57.8 to -54.7/oo, whereas that collected at the surface was somewhat less 13C-depleted (o13c -cH4 ranged from -51.3 to -49.4/oo) implying that there was partial oxidation of the methane during its transfer from the bottom to the surface. Fuex ( 1 977) measured 17 methane samples obtained from marsh sediments in Minnesota and reported o1 3c 0 ranging from -70.2 to 55 6 loo. Meth ane collected by agitating the

PAGE 44

sediments of Wurmsee, a shallow West German lake, was 13c-depleted (o 1 3c-cH 4 ranged from -64 to -52/oo) relative to the coexisting carbon dioxide gas (o 13c-co2 ranged from -21 to -3/oo) and was D-depleted (oD-CH4 ranged from -338 to -294/oo) relative to the pore water 0 (oD-H2 o ranged from -28 to -20 loo; Woltemate et al. 1984). In this section, measurements of oD-CH4 oD-H2o, o 13 c-CH 4 o13c-co2 and ancillary parameters from several freshwater environments and from a few locations within the Tampa Bay estuary are presented. Using these measurements and the model proposed by Woltemate et al. (1984), the relative importance of the two primary methanogenic pathways, acetate dissimilation and co 2 reduction, to methane production in these systems can be estimated. Sample Collection Gas samples were obtained by agitating the sediment with a rod and funneling the released gases into 125ml serum bottles (Wheaton Scientific, Millville, NJ) that were initially filled with lake water. The bottles were filled as completely as possible with gas and stoppered with a black rubber stopper (no. 2048-11800, Bellco Glass, Inc., Vineland, NJ) that was then secured with an aluminum crimp seal (Wheaton). The samples were stored on ice during transit back to the laboratory where they were kept in a freezer until analysis. Water samples for oD determination were collected in screw cap vials, the mouths of which were covered with parafilm to retard evaporation. Sediment samples were obtained by scraping surficial material into Whirlpak plastic bags and were frozen until analysis. Model of Biogenic Methane Formation Figure 4 is a model consisting of lines that describe predicted

PAGE 45

-0 ?fl. -J: () I 0 .0 -200 -100 Figure 4. Model of methane formation. Lines represent the percentage mix of acetate dissimilation (AD) and carbon dioxide reduction (R) (after Wolt e mate et al. 1 984) 35

PAGE 46

36 cSD-H20/ OD-CH4 isotopic pairs resulting from varying the relativ e contributions to methane production of the acetate dissimilation and co2 reduction pathways. This model was proposed by Woltemate et al. (1984) and used in that study to estimate that methyl group transfer (from acetate or other methyl group donors such as methanol) was responsible for about 76% of the total methane production in the sediments of Wurmsee, a shallow lake near Hannover, FRG. Lovley and Klug (1983) reported that methanol and methylamines were the precursors for les s than 5 and 1%, respectively, of total meth ane production in the sediments of eutrophic Lake Wintergreen, Michigan. The likely explanation for this is the low abundance of methanol and methylamine precursors relative to H 2 and acetate precursors in the organic matter input to the sediments (Lovley and Klug 1983) Low rates of methane productio n from methanol and methylamines in near-surface, sulfate-rich marine sediments have been reporte d recently (King et al. 1983; King 1984a) Although methane production from these "noncompeti ti ve11 s ub strates (King 198 4a) is of interest for several reasons, I would argue that it is likely to generate large quantities of methane (e.g., a significant fraction of that r equired to achieve saturation) only under r elatively unusual conditions. Many marine organisms use m ethylamines or methylamine precursors (King 1984b) in osmoregu lation (Yancey et al. 198 2). In response to salt stress, the halophyte Spartina alterniflora concentrated up to 30% of its total leaf nitrogen in glycine betaine ( Cavalieri and Huang 1981), which can b e metabolized t o acetate and trimethylamine in marine sediments (King 1984b). Oremland et al. ( 198 2 ) reported that metabolism of methanol and trimethylamine could account for the bulk of meth ane produce d in salt

PAGE 47

37 marsh sediments containing 8ml of methane per liter of wet sediment. experiments that yielded these results were conducted in flasks containing 60ml of San Francisco Bay water, 40ml of sediment, and 10ml of homogenized Spartina foliosa materials (Oremland et al. 1982). Although known (macroalgae; Blunden et al. ( 1982)) and potential (mangroves, seagrasses) contributors of methylamine precursors (unknown concentrations) account for some organic input, sediments as rich in organic matter as those used above (Oremland et al. 1982) are rarely encountered (Sackett et al. in press) in the Tampa Bay estuary. From all of the preceding, I conclude that methane production from methanol and methylamines is quantitatively unimportant in the samples analyzed during this study; therefore, in applying the model (Woltemate et al. 1984) to the data presented in this section, I consider acetate to be by far the major source of methyl groups to the methanogens active in these sediments. In constructing this model, Woltemate et al. (1984) assumed that for the co2 reduction pathway all four methane hydrogens are supplied by the environmental water (Schoell 1980; Daniels et al. 1980), and that for the acetate dissimilation pathway only one hydrogen comes from water with the remaining three coming from the methyl group of acetate (Pine and Barker 1 956) As a result, the slopes o f the 0:100 (acetate dissimilation: co2 reduction) and the 100:0 lines are 1 and 0 .25, respectively. The slopes of the intermediate lines are obtained by multiplying the relative amount of each pathway by the appropriate slope (1 or 0.25) and summing. For example, the slope of the 80:20 line would be : 0 .80(0.25) + 0.20(1) = 0.4. They-intercept of the 80:20 line was obtained from the sewage sludge incubation experiment of

PAGE 48

38 Schoell (1980). In that experiment, aliquots of sewage sludge were incubated in plastic.bottles spiked with different amounts of 0 20. The results indicated a linear correlation between 6D-H20 and 6D-CH4 that was described (Schoell 1980) by the equation: OD-CH4 0 323 /oo (4). The y -intercept of the 0:100 line resulted from the observation that OD-H2 0/6D-CH 4 pairs obtained from the measurement of natural samples, in which methane was presumably formed via co2 reduction (Schoell 1980) fit the relationship (Schoell 1980; Nakai et al. 1974): 6D-CH4 (5). The equations of the remaining lines (Woltemate et al. 1984) are obtaine d by assuming that all of the lines pass through the point of intersection of equations (4) and (5). Equations (4) and (5) are based o n analyses of meth ane formed under widely different env ironmental conditions. Variations in the magnitude of isotopic fractionation, due to variation in param eters such as rate of reaction and time (degree of isotopic equilibration), may introduce error into predictions of r e lative methanog enic pathway selection obtained with the model It appears, however that the model yields predictions that are generally valid (e.g., Woltemate e t al. 1984), and that application of the model to data presented in this study should yield useful interpretation. Results and Discussion Measu r e ments of oD-H2 0/oD-C H 4 pairs from various freshwater and estuarine sediments presented in Figure 5 imply that about 5 0 to 80% of the total methane production in these sediments can be attributed to acetate dissimilation. This agrees with some earlier studies (Cappenbe r g and Prins 1974; Winfr ey and Zeikus 1979; Lovley and Klug

PAGE 49

-0 ?fl. -.::t I () I 0 .0 -250 .& .& -270 -290 ... .& -* -330 -40 -20 0 20 Figure 5. Plot of methane hydrogen isotopic composition versus environmental water hydrogen isotopic composition for several freshwater environments and the Tampa Bay estuary. The lines are the same as in Figure 4. Symbols: Lake, FL.; Mirror Lake, FL; Dias, FL; Kilmer Pond, SC; Mississippi River Delta, LA: and Tampa Bay estuary, FL. 39

PAGE 50

1982; Woltemate et al. 1984), but is in conflict with others that .indicated co2 reduction dominates methane production in other freshwater lake sediments (Belyaev et al. 1975; Jones et al. 1982) 40 The variation in pathway importance between individual lakes and even between different locations within the same lake (Figure 5) may be attributable to natural qualitative and quantitative variations in the organic matter and bacteria occurring in these sediments (i.e., Belyaev et al. 1975). The predominance of the acetate dissimilation pathway under nearshore marine-estuarine (S = 6-34/oo; Table 2) conditions indicated in Figure 5 was also found in other nearshore-marine circumstances (Sansone and Martens 1981; Mountfort et al. 1980). In contrast, the biogenic methane found in deep-sea sediments is thought to originate almost exclusively from co2 reduction (Schoell 1 980; Claypool and Kaplan 1974) and is isotopically distinguishable (oD = -170 to -190/oo; Schoell 1983) from the marine methane measured in this study (Figure 5). The 6D-CH4 values reported here support the idea that with respect to methane production, highly productive brackish and marine sediments can bear a greater resemblance to freshwater than deep-sea sediments if sulfate depleted near the sediment-water interface due to high rates of respiration (Rudd and Taylor 1980; King 1983). As discussed previously, partial oxidation of a quantity of methane, mediated by aerobic bacteria, can leave the remaining methane markedly less 13c-and D-depleted (Coleman et al. 1981). If the percentage of methane oxidation and the degree of isotopic fractionation varied depending upon location within a given freshwater lake, for example, isotopic variation (Figure 5) could occur. The

PAGE 51

1 3 and ancillary data for several freshwater environments and Table 2. oD-CH4 o C-cH4 the Tampa Bay estuary. LOCATION/ oD (0/oo) o13c (0 /oo) r.c Sed. Cone. (mol % ) DATE u CH4 H 2 0 CH4 C02 SOMa SOM T(C) aCb aDC CH4 C02 A 08/3l/84 1 297:!:1. 0 -7:t0. 9 -61. 3:!:0.1 -7.2 31 1.058 1.413 59.8 6.5 It 3 304.5 -58.8:!:0.1 -6. 9 1.061 1.427 55.4 2 6 10/02/84 1 -316:!:0.6 '-58.8.1 -11.6 26 1.050 45.9 2.2 11/17/84 1 -298 -62.0 -8.1 22 1.057 50.5 2.9 12/21/84 1 -296 -11:!:0. 7 -62.9 -8. 8 23 1.058 1.405 46.2 2.5 ... 2 -267:!:0.4 -75.0:!:0.3 -7.2 1.073 1.349 50.9 7.5 01/23/85 1 -10:!:0.5 -27.7 3.3 16 2 -263 -71. 4 -5.0 -27.9 3.5 1.072 1.343 59.1 4 3 B 12/2l/84 1 -285 3:!:0.4 55 1 -2.2 21 1.056 1.394 85.3 7.2 II 2 -314 -52.2 -6.8 1.048 1.453 85.8 2 9 01/23/85 1 -292 -2:!:0.2 -56.4 -10.8 -22. 5 0.8 14 1.048 1.410 65.0 .1.4 2 -286 -55.8 -9.0 1.050 1.398 77.0 3.5 c 12/25/84 1 -292 +5:!:0.2 -60.3 -3.7 -27.5 11.1 21 1 060 1.419 60.8 8.4 II 2 -256 -66.7 -3.7 -28.0 33.6 1.068 1.351 49.0 13.1 It 3 -265:!:0.5 -64.8 -2. 4 -26.9 0.5 1.067 1.367 74.4 8.0 D 04/23/82 2 -335 -48:!:0.2 -59.9:!:0.1 1.432 63.9 1.2 5 346 -60.0 1.456 61.0 0.6 II 6 -342:!:0 5 -60.4. 1 -12.2 1.051 1.447 69.4 3.6 E 10/14/84 1 -315 -53.6 -10.9 24 1.045 41.0 2.8 II 2 -319 -51.5 -12.0 1.042 34. 0 1.9 12/12/84 1 -311 -16:!:0.6 -55.3 9 1.428 33.6 0.7

PAGE 52

Table 2 (Cont'd). LOCATION/ cSD (0/oo) 613c (0 /oo) DATE u CH4 H 2 0 CH4 C02 sm1 a F 09/16/84 1 -323:!:1. 2 +11 70 .8:!:0.1 10/11/84 2 -296 +11 -63.3 12/03/84 3 288 +7:!:0.4 -64.7 23 2 12/03/84 4 -291 0:!:0.2 -63.8 TABLE LEGEND Lake, 12ha, fresh, St. Petersburg (SP), FL Mirror Lake, 7ha, fresh, SP, FL f -Lake Dias, 300ha, fresh, Volusia County, FL -Mississippi River Delta, fresh, LA Pond, fresh, near Columbia, S.C. -Tampa Bay estuary, FL 0 1-Ft. DeSoto Pk., S = /oo 2 Bishop Harbor, S = 320/oo 3 Lake Seminole, S = 20 /oo 4 -Feather Sound S = 6/oo % C SOM 1.6 Sed. Cone (mol%) T("C) aCb aDC CH4 29 1.493 54.9 28 1.436 79.3 24 1.414 70.7 21 1.4 10 74.1 a SOM = Sedimentary organic matter b aC = c -aD = 13 (C02 /cH 4 ) a 6 C02 + 1000 13 6 C-CH4 +1000 (H20/CH4) = 6 D-H20 + 1000 6 D -CH4 + 1000 co2 2.0 13.5 7.5' 6.9 +:"" [\)

PAGE 53

43 generally negative correlation between o 1 3c-cH 4 and oD-CH4 indicated in Flgure 6 implies that the isotopic variation is not solely due to methane consumption. The negative correlation does not mean that these samples were totally unaffected by in situ methane oxidation, but it does imply that some additional process (i.e., variation in path way) contributed to the isotopic variation. Methane concentration profiles in the upper 1 0 to 1 5 em of anoxic freshwater sedi ments are generally linear, implying that production and transport are of greater importance than anaerobic methane consumption in controlling the methane distributions (Reeburgh and Heggie 1977) Methanogenic bacteria are strict anaerobes and produce methane only under highly reducing conditions (Mah 1982). Methane produced under these conditions could only be oxidized by the aerobic process if it either migrated (as a bubble or in solution) to a less reducing location or if the depth of oxygen penetration increased. A bubble rising from the depth at which it was formed would likely exit the sediments entirely, because the distance between the methane production zone and the sediment-water interface is short (50 em or less), and would not be sampled by the methods employed here. Likewise, methane that dissolved would not be sampled, and because the process of partial dissolution itself would not be expected to induce much isotopic fractionation (Bernard et al. 1976), any methane remaining in bubble form should maintain its isotopic signature. Deeper penetration of oxygen into the sediments because of a temperature-induced decrease in oxygen consumption might allow partial oxidation of trapped bubbles. This could be responsible for the slight increase (Table 2) in the isotopic compositions of methane sampled from Crescent Lake (#2) on 1/23/85

PAGE 55

45 compared to that sampled at the same location on 12/21/84. This slight difference has very little effect on any interpretations, and when considered along with the other discussion advanced above 1 argues against extensive isotopic fractionation of samples reported here (Figures 5 ,6) by methane consumption All of the isotopic and ancillary data considered in this section are listed in Table 2 Concentrations of methane and carbon dioxide ranged from 33 to 86% and from 4 to 13%1 respectively, in the samples that were analyzed. The carbon isotopic composition of sedimentary organic matter (SOM) ranged from -26.9 to -28.0/oo for all but one of the freshwater sediment samples (Table 2); this implies that higher land plants (c3 pathway) and low 13c;12 c aquatic plants (Smith and Epstein 1971) dominate organic matter input to these sediment s For instance, a n unidentified emergent aquatic macrophyte from Lake Dias (#3) yielded a o13c of -28.1/oo. A o13c of -22.5/oo was determined for the SOM of Mirror Lake (Table 2); this is substantially heavier than the other freshwater values. Possible explanations for this include a relatively greater contribution of isotopically heavy c 4 pathway plant material (i.e., domestic Bermuda grass; Smith and Epstein 1971) and utilization of isotopically heavier source carbon by the submerged aquatic plants (Osmond et al. 1981). There is no obvious relationship between either o13c-SOM or %C-SOM and o13 c-cH 4 for the samples reported here (Table 2). For example, although the o13 c-cH 4 of Crescent Lake #2 is 8-12/oo lighter than the o1 3 c-cH 4 of Crescent Lake #1 I there is no substantial difference between the two sites with respect to o1 3c-SOM or %C-SOM. Factors other than amount and o1 3c of the source organic matter, such as pathway selection (Figure 5), must

PAGE 56

46 also be involved in determining o 13c-CH 4 Fractionation factors ac(a(CO/CH4)) and aD(a(H 20/CH4)) of the freshwater samples, ca,lculated from measured isotopic compositions, ranged from 1 042 to 1 073 and from 1 343 to 1 456 respectively (Table 2). These values are similar to those determined for German lake Wurmsee (Woltemate et al. 1 984) although the ranges reported here are greater. The greater ranges (higher aC, lower aD) probably reflect the larger relative contribution of the co2 reduction pathway in some of these samples (Figure 5) compared to the Wurmsee samples (Woltemate et al. 1984) The fractionation factors calculated from typical deep -sea isotopic compositions (Rice and Claypool 1981; Claypool and Kaplan 1974; Schoell 1980) are significantly larger (aC-1 08) and smaller (aD1.25) than the factors listed in Table 2; this implies that both isotopes are affected by methanogenic pathway and indicates the potential applicability of these parameters (particularly together) to the genetic characterization of biogenic methane deposits. As discussed earlier, recent studies (Craig and Chou 1982; Khalil and Rasmussen 1982) indicated a doubling of atmospheric methane concentrations during the past few hundred years. Furthermore, during the last 3 to 4 years, atmospheric methane concentrations have increased at rates of 1 to 1 .9% per year (Khalil and Rasmussen 1983) Decomposition of organic matter in water-covered soils, intestinal fermentation in ruminants, biomass burning, and direct anthropogenic input (i.e. leakage of fossil fuels) are thought to be the major sources of methane to the atmosphere contributing approximately 46, 22, 5, and 7% of the total flux, respectively (Khalil and Rasmussen 1983) The major sink of atmospheric methane is thought to be reaction with OH

PAGE 57

47 radicals in the troposphere (Khalil and Ras mussen 1983). A recent study by Harriss et al. ( 1982) indicated that periodically inundated soils may also be a site of atmospheric methane removal during dry periods. Recent measurements indicate that the o1 3c of atmospheric methane is about -47. 0 0 .3/oo (Stevens and Rust 1982). The average isotopic 0 fractionation associated with the sink process i s -2.5 1 5 /oo, and there is a +0.3/oo isotope effect resulting from the nonsteady state increasing methane concentrations (Stevens and Rust 1982). This implies that the average o13c -CH4 for all sources is -49. 2 1 .5/oo; this is significantly less 13 c -depleted than most known biogenic methane sources (Stevens and Rust 1982), and is about 1 o0 /oo less negative than the average o13c-cH 4 of all sources (-61 0 3 /oo) calculated by Senum and Gaffney ( 1985). Biomass burning and fossil fuels are the on l y known significant methane sources mor e 13c-enriched than -49/oo (Stevens and Rust 1982), but they apparently account for a relatively small fraction of the total input. Bubble ebullition is the dominant mode of methane input to the atmosphere from some shallow freshwater (Robertson 1979; Cicerone and Shetter 1981) and marine (Martens and Klump 1980) environments. All of the gas samp les analyzed for this section were collected as bubbles from sediments covered by very shallow (1m or less) water. Three of the freshwater lakes (Crescent Lake, Mirror Lake and Kilmer Pond) hav e been observed to release gas as bubbles from their sediments and it is very likely that active bubbling also occurs in the other environments. There is a good deal of variation ( -24 /oo) in the o 13 c -cH4 data presented in Table 2; however, most of the samples are substantially

PAGE 58

48 lighter than -49/oo, in agreement with other data from shallow aquatic environments (Stevens and Rust 1982). Though limited in number, these data imply that ebullition from these sediments is unlikely to provide an atmospheric input of methane as 13c-enriched as the source average. Based on the results of a study by Rust (1981) which demonstrated that the o 13 c of methane produced by ruminants depends on the of the plants in their diet, Stevens and Rust ( 1982 ) proposed that the anaerobic decomposition of isotopically heavy c 4 plant debris in large wetlands, such as the Sudd marshes and the Florida Everglades, might be a significant source of isotopically heavy methane to the atmosphere. As mentioned before, no such simple relationship between the o 1 3c-cH 4 and of aquatic sediments is indicated by the data presented in Table 2. According to Wolfe (1979), four well-defined groups of bacteria are involved in the conversion of complex organic matter to methane in sediments, whereas in the rumen only two of the groups are involved and there is no significant production of methane from acetate in the rumen. Also, substrate input, product output, and environmental conditions are more closely controlled in the rumen than in sediments (Wolin 1982). Thus, in addition to other factors such as pathway selection and possibly kinetic effects are probably important in setting the o 13 c-cH 4 of aquatic sediments. Escape of dissolved methane across the air-water interface, rather than ebullition, may be the more likely means of tranferring isotopically heavy methane from shallow aquatic environments to the atmosphere. Dissolved methane should be more readily available for aerobic methane oxidation (Rudd and Taylor 1980) and isotopic fractionation (Coleman et al. 1981); however, assessment of the potential contributions of ebullition and

PAGE 59

dissolved methane to atmospheric 613c-cH4 is hindered by a lack of relevant data. Using data obtained in 1960 and 1970 (Bainbridge et al. 1961; Begemann and Friedman 1968; Ehhalt 1973), Senum and Gaffney (1985) noted an apparent decrease in atmospheric methane 60 with time that they extrapolated to the present to yield an estimate of -104 4/oo for the 60 of present-day atmospheric methane. The extrapolated 60 value was corrected for kinetic isotopic fractionation resulting from the reaction of methane with hydroxyl radicals (the magnitude of the fractionation was attributed to Gorden and Mulac 1975) in the troposphere to yield an estimate of -322/oo as the present-day average 60 of all methane sources to the atmosphere (Senum and Gaffney 1985) Analyses reported in Table 2 and by Woltemate et al. (1984) indicate that methane as O-depleted as -322/ oo is produced in some freshwater sedi ments Organic matter decomposition in sewage sludge may also yield methane as O -depleted as 322 / oo (Schoell 1980 ) According to the genetic characterization model of Schoell (1983), which is based on approximately 500 natural gas analyses, thermogenic methane is generally more 0-enriched than -300/oo. To my knowledge, there are no estimates of the 60 of methane produced from either the intestinal fermentation in ruminants or biomass burning. Although an important source of atmospheric methane with a 60 near the estimated source average (Senum and Gaffney 1985 ) has been identified (freshwater sediments) there are presently too few measurements of the 60 of atmospheric methane and its sources available to determine whether or not this apparent agreement has any real significance in regard to evaluating the role of shallow aquatic sediments in the atmospheric

PAGE 60

methane budget and in regard to determining the 6D of atmospheric methane. 50

PAGE 61

5 1 METHANE ISOTOPIC COMPOSITIONS IN CAPE LOOKOUT BIGHT SEDIMENTS Cape Lookout Bight is a small, partially enclosed marine basin located on the Outer Banks of North Carolina (Figure 7). The b ight interior is somewhat protected from high energy winds a nd waves, and serves as an efficient trap of fine-grained sediments and organic debris that exit the biologically productive barrier island lagoonal system to the north through Barden's Inlet in response to tidal flushing and winter storms (Martens and Klump 1980, 1984) Organic matter in bight sediments comes f rom plankton, salt marsh and seagrass detritus eroding shoreline peat deposits (Martens and Klump 1 984) and terrigenous materials washed into t h e barrier isl and lagoonal system by freshwater runoff. Rapid remineralization of organic carbon in the rapidly accu mulating (8.4 to 11.8 em/yr ; Chanton et al. 1983), organic-rich (3.3 t o 5% organic carbon) sediments of the b ight interior (Station A-1; Figure 7) creates anoxic conditions at or near (0.5 to 1.0 em s ubbottom) the sediment-water interface (Mar tens and Klump 1 984) The a noxic conditions prevent inhabitation of the sediments by macrobenthic infauna (Menzies et al. 1968), which pump sulfate-rich seawater into their burrows (Martens 1976) except during winter months (Bartlett 1981 as cited by Chanton et al. 1983; Martens and Klump 1984). As a result, dissolved sulfate is quickly depleted (about 1 mM) and meth a ne is produced rapidly enough to supersaturate the interstitial waters

PAGE 62

CAPE 38' N \ 1000 M ATLANTIC OCEAN Figure 7 Cape Lookou t Bight, North Carolina USA. The sam p ling s ite was Stati o n A 1 (aft e r Martens an d Klump 1980). 52

PAGE 63

53 near (about 8-15 em subbottom) the sediment-water interface during warm (16-29C) summer months (late May to mid November; Martens and Klump 1980; Sansone and Martens 1982; Kipphut and Martens 1982) Reduction of hydrostatic pressure associated with ebb tide (tidal range about 1 m) triggers expansion and upward migration of trapped gas bubbles (82 to 90% methane, mean 86%) through sedimentary "bubble tubes" (Martens 1976) to the water co lumn and finally, to the troposphere (Martens and Klump 1980). An estimated 70 to 80% of the yearly flux of methane ejected to the atmosphere from the Cape Lookout Bight system comes from direct bubble transport, with the remaining 20 to 30% contributed by escape of dissolved methane from the waters across the air-sea interface (Martens and Klump 1980). Blair et al. (1984) reported that methane bubbles produced in bight sediments display a seasonal o 1 3c variation (o13 c-cH 4 ranging between -64 and -58/oo), with ebullition of the more 13c-enriched methane o ccurring during the summer when bubbling rates are maxima l In this section, measurements of the oD and o13c of methane gas samples from Cape Lookout Bight sediments collected on fifteen different dates from 1 6 June 1983 to 17 November 1984, and oD measurements of three pore water samples from various times and depths are presented. All of the samples described here were collected by the research group "CH4Ao2sn under the direction of Dr. Christopher S. Martens of the University of North Carolina, Chapel Hill. These measurements provide an opportunity to co mpare the estimates of relative methanogenic pathway importance obtained with the Woltemate et al. (1984) model to those derived from 14c labeling studies reported by Sansone and Martens ( 1981) and Crill and Martens (in press).

PAGE 64

54 Further, the large range of temperatures (7 to 29C) at which these samples were collected allows assessment of seasonal influences on methanogenic pathway selection. Sample Collection Trapped gas bubbles were displaced by physical disturbance from different depths within the upper 20 to 40 em of the sediments into a large inverted cone placed about 1 m above the bottom (water depth about 8 m), and transferred at depth from the cone to 125 ml (except for the 19 September 1984 samples (30 ml)) Wheaton serum bottles that were initially filled with water. The serum bottles were filled as completely as possible with gas, stoppered, and c rim ped with an aluminum seal. During sample collection, gas was continuously released from the sediment to the cone as gas was withdrawn from the cone. As a result, replicate samples, particularly during summer whe n up to five 125 ml samples were taken, did not all come from exactly the same aliquot of sedimentary gas. The samples were kept in a freezer until analysis, except during transit between North Carolina and Florida. Results and Discussion Methane isotopic compositions over the course of a year and sediment temperatures from several years (Martens and Klump 1980; Martens et al. 1980; Crill and Martens 1983; Martens unpublished data) are plotted versus sampling date in Figure 8. The data (Figure 8, Table 3) indicate great isotopic variability among replicate sample bottles during the summer, and indicate that oD-CH4 and o13c-cH 4 vary in an opposite manner in response to change in season, with methane more D-depleted and more 13c-enriched (also Blair et al. 1984) during 0 summer months (July to September; mean oD-CH4 = 269 /oo, std. dev. =

PAGE 65

30 ,..... (.) 0 1-w 20 za: w :::> :::E 1-<( 8 ffi10 (/) (L :::E w 1-cfi -...; :-250 (.) I 0 .0 -270 ,..... 0 cfi -...; ' .. l res> (2) J A 0 D F BOTTLE +: 2 3 4 + 5 ... t... .... J. ......... '\ ...... ... ...... '\. I 1 (2) (2) (2) A 1984 J (4)J: & y I A ...... -\4 ... ... 1 (2) 1(3) 0 D Figure 8. Seasonal variations o f temperature, oD-CH4 and o13c-cH 4 in Cape Lookout Bight sediments. The temperature data are from Martens and Klump (1980), Martens et al. (1980), Crill and Martens (1983) and Martens (unpub lished). V1 V1

PAGE 66

56 Table 3 oDC H 4 o 1 3c-cH4 a nd o DH 2 0 from Cape Look o u t Bight. goc H 4 o 1 3 C C H Date Bottle IF ( /oo) (o/oo) 4 T(C) 6/16/83 1 22 7 63.5 23 2 -225 63 7 3 -227 63.8 4 -222 63 9 5 2 2 6 63 8 8/19/83 1 2 5 9 -62.0 27 2 264 -60. 5 3 2 84 -56. 8 4 2 77 -58 9 5 258 62 2 9/14/83 1 26 1 61.4 27.8 2 263 6 1 9 10/16/83 1 265 60 0 22 8 2 2 65 60.3 2/2/84 -238 64 3 7.3 2 23 1 -64.3 4/7/84 1 24 1 63 6 14 2 239 63 0 5/6/84 1 -231 -64. 0 1 8 2 233 63 9 5/31 I 8 4 23 1 -68.4 23 2 228 -68. 3 6/14/84 234 65 1 23 2 233 64. 1 7/2/84 -270 59 2 24 2 -271 59 3 7/18/84 25 1 -62.1 26 2 263 61.3 8/10/84 1 277 57 0 25 2 276 56 0 3 275 57 9 4 275 58 3 8/31/84 -278 -58.3 26 2 258 61.2 3 -281 -57.4 4 260 60 7

PAGE 67

57 Table 3 (Cont 'd). 8D-CH4 o 13c-CH Date Bottle II ( loo) (oloo) 4 T(C) 9119184 276 -58.9 26 2 -273 -59.0 1 1 I 17 I 84 250 -60. 4 1 6 2 257 -60. 5 3 -260 -61.0 4 -262 60.4 Pore wat e r
PAGE 68

o .r 13C-CH o o 9 /oo, n = 21; mean v 4 = -59.5 /oo, std. dev. = 1.9 /oo, n= 21) when bubbling rates are maximal (Martens and Klump 1980), as compared to winter months (December to April; mean oD-CH4 = -236/oo, std. dev. 0 13 0 0 = 4.4 /oo, n = 6; mean o C-CH4 = -63.9 /oo, std. dev. = 0.5 /oo, n 6) when Cape Lookout Bight sediments are not observed to bubble (Martens and Klump 1980, 1984). The indicated seasonal fluctuation of oD-CH4 (Figure 8) probably results from seasonal variation in methanogenic pathway selection, with acetate dissimilation accounting for a greater proportion of the total methane production during summer than during winter. Evaluation of these data with the Woltemate et al. (1984) model (Figure 9) implies that acetate dissimilation accounts for about 45 to 60% of the total methane production during summer months, a n d about 25 to 40% of the total during winter months Sansone and Martens (1981) measured the 1 4 production rates of carbon dioxide and methane from C-labeled acetate in the 5 em depth interval immediately below the 1 mM sulfate isopleth of bight sediments at various times of the year, and estimated that over 50% of the summertim e methane flux could be accounted for by acetate dissimilation. In contrast, measurements made from January to May indicated that the rate of methane production from acetate in this layer was near zero during the winter (Sansone and Martens 1981). Crill and Martens (in press) measured methane production rates from 14 C-labeled bicarbonate a n d acetate at various depths in Cape Lookout Bight sediments during the summer of 1983, and found that the relative importance of the acetate dissimilation pathway to total methane production, integrated over the upper 30 e m of the sediments, increased over the course of the summer from 23% on 21 July to 30% on 30 August.

PAGE 69

0 '#. :r: () I Cl .0 I -230 I -250 I I !.& --270 -290 A -. -330 -20 0 20 bD-H20 ( %o ) Figure 9 Plot of methane hydrogen isotopic composition versus pore water hydrogen isotopic composition for samples from Cape L ookout Bight and the samples plotted in Figure 5. Symbols: ecape Lookout Bight; all other the same as in Figure 5 59

PAGE 70

In the sulfate-reducing zone of bight sediments the acetate methyl is oxidized only to carbon dioxide (Sansone and Martens 1982). Although appreciable rates of methane production from carbon dioxide reduction were measured in Cape Lookout Bight sediments at all depths except the top 2 em (including some in which high rates of sulfate reduction were measured), acetate was a significant precursor of methane only in sulfate-depleted (< 1 mM) layers; further, acetate dissimilation accounted for an increasing percentage of the total methane production with increasing depth to a maximum of 48% in the 20 to 22 em layer on 7 August 1983 (Crill and Martens in press). Net 60 methane production rates as determined by the tube incubation technique of Crill and Martens (1983) were nearly equivalent to the sum of the methane production rates from 14c-labeled bicarbonate and 'acetate during the summer of 1983, implying that nearly all of the methane production in Cape Lookout Bight sediments during the summer results from carbon dioxide reduction and acetate dissimilation (Crill and Martens in press). Measurements of in situ concentrations of alternate methanogenic substrates, such as methanol and methylamines, and 14c radiotracer studies will be required for a more accurate assessment of the quantitative signficance of alternate methanogenic pathways in bight sediments. The interpretation presented here (Figure 9) that acetate dissimilation is relatively less important during the winter in Cape Lookout Bight sediments qualitatively agrees with the aforementioned results of Sansone and Martens (1981). Further, if significant methane production from acetate occurs only when sulfate concentrations are < mM (Crill and Martens in press), then acetate dissimilation would be

PAGE 71

61 expected to be of less importance during the winter because penetration of the mM sulfate isopleth is 20 to 22 em in winter and 7 to 10 em in summer (Sansone and Martens 1982). Because more than 98% of the calculated total remineralization of metabolizable organic matter takes place within three years in Cape Lookout Bight sediments (Martens and Klump 1984 ) which corresponds to about 30 em depth (Chanton et al. 1983), the microbes living at depths where acetate dissimilation to methane should be possible during winter would have to subsist on the more refractory organic matter fractions. Lowered rates of methane production from acetate during winter would be expected due to the combination of lowered temperatures (Zeikus and Winfrey 1976) and poorer nutrition. The logical prediction that acetate dissimilation is relatively less important during the winter and the general agreem ent of the estimates of summer Cape Lookout Bight methanogenic pathway selection presented in Figure 9 with those from 14c radiotracer studies (Sansone and Martens 1981; Crill and Martens in press) imply that interpretations made using the model (Woltemate et al. 1984) employed in this study are reasonable. Variation of the relative importance of acetate dissimilation with depth in'the sediment (Crill and Martens in press) is probably the factor most responsible for the great isotopic variability observed among replicate samples (Figure 8 Table 3) taken during the summer (19 August 198 3 and 31 August 1984) when bubbling rates are maximal (Martens and Klump 1980, 1984) Because the method of sample collection employed in this study did not assure that all "replicates" sampled exactly the same aliquot of sedimentary gas, any isotopic inhomogeneity (i.e., with depth) of gas bubbles trapped within the

PAGE 72

62 sediment could express itself in the replicates analyzed here. Kipphut and Martens (1982) observed 222Rn profiles in Cape Lookout Bight 222 sediments during early and late summer that exhibited a sharp Rn maximum at 6 to 8 em depth and a 222Rn deficit below. The unusual shape of the 222Rn profiles was attributed to stripping of 222Rn by methane bubbles at depth followed by partial dissolution of the bubbles at 6 to 8 em depth, particularly during times of the year when bubbling rates are relatively low and the residence time of the bubbles in the sediment should be longer (Kipphut and Martens 198 2) The results of Bernard et al. (1976) indicate that very little isotopic fractionation occurs when methane gas is partially dissolved in water; therefore, it is unlikely that variation in the extent of bubble dissolution with depth in the sediment or season would induce much isotopic variability. If all of the methane produced in Cape Lookout Bight sediments had the same isotopic composition and was then subjected to varying degrees of bacterial oxidation, depending upon season and/or depth, differential isotopic fractionation (Coleman et al. 1981) and isotopic variability could result. The zone of oxic conditions is very thin (0.5 to 1.0 em; Martens and Klump 1984) in bight sediments during the winter and essentially absent during the summer, greatly decreasing the possibility of extensive oxidation of sedimentary methane bubbles by the assimilatory, microaerophilic process (Martens and Klump 1980) The potential significance of dissimilatory anaerobic methane oxidation in Cape Lookout Bight sediments is difficult to assess as there are no direct radiotracer measurements of the process (i.e. Alperin and Reeburgh 1984) available for this system. Because appreciable anaerobic methane oxidati on in marine sediments is apparently confined

PAGE 73

63 to a thin layer at the base of the sulfate-reducing zone (Reeburgh 1980; Devol 1 983; Alper in and Reeburgh 1984; Iversen and Jorgensen 1985) methane bubbles shou l d not be subject to extensive oxidation during entrapment below the sulfate-reducing zone The linear to convex-up pore water methane concentration profiles that are observed during summer and winter (Martens and Klump 1980; Kipphut and Martens 1982) can be interpreted to mean that the rate of anaerobic methane oxidation is significantly less than the rate of methane transport out of the sediment (Martens and' Klump 1984) The negative linear correlation between Cape Lookout Bight oD-CH4 and o 13 c-cH 4 (all data): oD-CH4 = -6.1 o13C -CH4 626/oo (r = 0 .90, n = 42) (6) indicated in Figure 10 is good evidence that the isotopic variability is not simply the result of differential oxidation of a singl e isotopic end-member, as partial bacterial consumption would likely enrich the residual methane with respect to both isotopes (i. e Coleman et al. 1981). Thus it is unlikely that any secondary process such as partial dissolution or bacterial oxidation could account for all of the seasonal and replicate sample isotopic variability indicated in Figure 8 Linear regression of oD-cH4 and o13c-cH 4 data obtained during the main Cape Lookout Bight bubbling period (1 July to 1 October, sediment T 24C, Martens and Klump 1984) yields the equation: oD-CH4 528/oo (r = 0.92, n = 21) ( 7) With the exception of the 3 1 May 1 984 samples, the methane isotopic compositions of winter samples taken from February to mid -June (Table 3) deviate significantly from the line (7) derived from the summer data. Partial bacterial oxidation of trapped methane bubbles resulting

PAGE 74

0 ?fl. -'It J: 0 I 0 .0 -240 -260 -280 : bD=-4.4 b 13 C-5{8 T>24C, n=21, r=0.92 bD=-6.1 b 13 C -626 n=42, r=0.90 CAPE LOOKOUT BIGHT, N.C -68 -66 -64 -62 -60 -58 -56 Figure 10. Plo t of methane hydrogen isotopic composition versus meth a ne carbon isotopic composition for from Cape Lookout Bight. The equation: oD = -6.16 C626 is a linear regression of all the1gata considered in this study. The equation : oD = -4.4 C 528 is a linear regressi on of the data obtained for T 24C. 64

PAGE 75

65 from deeper penetration of sulfate during the winter (Sansone and Martens 1982) would be expected to shift the methane isotopic compositions in the direction required to explain the deviation. Blair et al. ( 1984) suggested that near-sur.face methane oxidation could account for the variation in pore water methane c 1 3c with depth (-61 .0 0.6 /oo at 2-7 em depth to -65 0.1/oo at 35-47 em depth) that they observed during February of 1984. The c13c-CH of the winter 4 samples discussed here range from -65.1 to -63.0/oo, implying that they originated in deeper layers where methane oxidation is probably of less significance. The disagreement of the winter isotopic compositions with equation ( 7) probably results from the greater_ relative importanc e of the carbon dioxide reduction pathway to methane production during the winter as discussed above. The biogenic methane found in deep-sea sediments, which is thought to originate almost exclusively from carbon dioxide reduction (Claypool and Kaplan 1974; Schoell 1980), is generally significantly less D-depleted (cD = -190 to -170/oo, Schoell 1983) than the methane produced during the summer in 0 Cape Lookout Bight sediments (mean cD = -269 /oo). It is reasonable to expect that methane produced by carbon dioxide reduction in Cape 0 Lookout Bight sediments would be less D-depleted than 269 /oo, although it might be more D-depleted than the methane produced in deep-sea sediments as a result of environmental differences (i.e., rate of organic matter input and remineralization, sedimentation rate, methane production rate, extent of isotopic exchange) between the two systems. The c 1 3c of methane produced by carbon dioxide reduction depends upon the c 1 3c of the substrate carbon dioxide and the degree of

PAGE 76

66 fractionation induced by the bacteria mediating the reaction. Laboratory studies (Games et al. 1978; Belyaev et al. 1983) indicate that the isotopic fractionation associated with carbon dioxide reduction decreases with increasing as would be expected for temperature-dependent isotopic equilibration of methane and co2 The results of laboratory studies of bacterial sulfate reduction (Jones and Starkey 1957; Kaplan and Rittenberg 1964) imply that the sulfur isotope effect associated with this process is dependent upon the rate of reduction, and that temperature influences fractionation only by affecting the rate of reduction. The rate of methanogenesis is known to vary greatly in response to temperature fluctuations (Zeikus and Winfrey 1976), so the isotope effect associated with co2 reduction (Games et al. 1978; Belyaev et al. 1983) may also be rate-dependent. Extrapolation of the results of these laboratory studies (Games et al. 1978; Belyaev et al. 1983), which were conducted at T 37C and may not be representative of the naturally occurring to Cape Lookout Bight winter temperatures (7 to 23C) implies that temperature (or rate) induced 613c-cH4 variations of about 10 to 30/oo are possible. Blair et al. (1984) reported that 6 13c-Eco2 in Cape Lookout Bight sediments during February of 1984 ranged from -9.5 0.1/oo in the sulfate reduction zone to +6.1 0.1/oo in the methane production zone indicating that variation in substrate isotopic composition could also be responsible for significant 613c-cH4 variations.

PAGE 77

67 METHANE ISOTOPIC COMPOSITIONS IN DEEP-SEA SEDIMENTS Methane associated with present day deep-sea sediments is formed by both microbial and thermal decomposition of buried organic matter. Microbial processes account for most of the methane production in rapidly accumulating (sedimentation rate > 10 m/10 6 years), near-surface (< 1 km depth) deep-sea sediments subjected to normal geothermal gradients ( < 50C/ km depth; Claypool and Kvenvolden 1983). Only in areas of abnormally high heat flow, such as subduction zones, active spreading centers, or near hot igneous intrusions (i. e., Schoell 1982), are near-surface deep-sea sediments likely to attain sufficiently high temperatures (80 to 1 50C) to allow significant thermal decomposition of complex organic matter to methane and other light hydrocarbons (Claypool and Kvenvolden 1983). Biogenic methane production in deep-sea sediments is thought to occur almost exclusively via co 2 reduction (Claypool and Kaplan 1974; Whiticar et al. in press). The principal evidence for this i s that deep-sea profiles of c1 3 c -cH 4 and c13c-Eco2 are generally parallel, presumably as a result of bacterial reduction of C02 with relatively constant kinetic carbon isotopic fractionation (Claypool and Kaplan 1974) Further, available deep -sea 6D-H2 0/6D-CH 4 pairs (Whiticar et al. in press) fit rather well the equation (5) mentioned earlier (Nakai et al. 197l.J; Schoell 1 980) that describes methane production from co2 As discussed previously, the slope of unity in equation (5)

PAGE 78

68 can be interpreted to mean that all four methane hydrogens are supplied by the environmental water, a condition satisfied only by co 2 reduction (Schoell 1980; Daniels et al. 1980) When methane is produced in deep:-sea sediments in sufficient quantities to s upersaturate t h e pore waters unde r app ropriate pressure-temperature conditions, it may combine with water and other gases to form gas hydrates (Claypool and Kaplan 1974) Gas hydrates are a type of clathrate compound in which gas molecules are trapped within and stabilize a framework of hydrogen bonded water m o lecules. The ice-like gas hydrates form at temperatures 10 to 15C above the freezing point of water when certain gases and water are combined at elevated pressures. With respect to pressure-temperature conditions, about 93% of the world s ocean sediments are sui table f o r methane hydrate formation (Trofimuk et al. 1973). Most currently known or suspected hydrate zones occur in continental slope regions and are correlated with bottom-simulating reflectors that result from changes in reflection properties at the gas hydrate-free gas boundary at the base of the gas hydrate zone (Shipley et al. 1979). The presence of gas hydrates in continental slope sediments along the east coast of the U S the western Gul f of Mexico and along the Pacific side of Central and South America was inferred from the presence of bottom -simulating reflectors in seismic profiles by Shipley et al. (1979). Water depths overlying the suspected hydrate zones range from 1a0 to 4,400 m with pressure-temperature conditions for hydrate stability satisfied between 100 and 1 100 m subbottom (Shipley et al. 1979). In this section, measurements of oD-CH4 o13c-cH 4 and oD-H2 0 from several Deep Sea Drilling Project (DSDP) Sites, including some from

PAGE 79

which gas hydrates were recovered, are presented. The samples described here were provided by Dr. James M. Brooks of Texas A&M University and Dr. George Claypool of the United States Geological Survey, Denver. Sample Collection Samples of gas from expansion pockets, which developed within the cores upon retrieval, were transferred to vacutainers using a needle and septum device with a valve to minimize atmospheric contamination (Baltuck et al. 1985). Interstitial waters wer e obtained by squeezing (Manheim and Sayles 1974), and collected in glass ampoules (Site 618 only), or in plastic syringes that wer e sealed with silicone caulking to minimize interaction of the sample with the atmosphere. Results and Discussion DSDP Leg 84 Middle America Tre nch Region 69 Leg 84 of the DSDP sampled landward slope sediments of the Middle America Trench off Costa Rica (Site 565) and Guatemala (Sites 566-570) Geophysical studies indicate that during the Miocene and Pliocene undeformed sediments covering the descending Cocos Plate were not scraped off during subduction but continued into the subduction zone making the Middle America Trench a tectonic type end-member (Auboin and von Huene 1985). Gas hydrates had been encountered in the continental slope sediments of this region during drilling on Legs 66 and 67 in 1979. Isotopic analyses of methane and pore water samples collected during Leg 84 are presented in Table 4. Also listed in Table 4 are the locations where gas hydrates were visually observed in Leg 8 4 sediments by Kvenvolden and McDonald (1985) The paucity and uneven geographical

PAGE 80

Table 4. Isotopic analyses of methane and water from DSDP Leg 84 (in 0/oo). Location of hydrates and% org. c data from Kvenvolden and McDonald (1985); Site locations, water depths, and sed. rates from von Huene, Aubouin et al. (1985). Sample Depth o 13c-CH (core-section) (m) 6D-CH4 oD-H2 0 4 gas gas Site 565 (86 4 'W, 0 sed. rate 0 9 4 3. 7 'N; water depth = 3111 m; (100-300 m) 125 m/m.y.; % org. C (100-300 m) .. 0.5-1. 7) 6-6 58 -192 -73. 1 7-1 59 +2 9-1 78 -1 10-5 94 +3 21-3 195 -200 0 -65.0 23-5 217 -198 -64.8 24-3 223 -193 -63.9 25-4 235 -2 26-4 243 -195 ( ) -64.0 hydrate 285 hydrate 319 34-1 320 -196 (.5) -63. 0 Site 567A (90.9'W,12.0'N; water depth= 5529m; sed. rate ---(230-360 m) = 76 m/m.y.; % erg. C (230-360m) = 0.2-1.5) 4-1 7-2 8-4 17-2 232 255 268 346 -186 -72.7 -183 -72.9 -183 ( 5) -72.9 -174 -73.4 Site 568 (90.0'W, 13.3'N; depth= 2031; sed. rate (0-200 m) -------= 117 m/m.y.,(240-400 m) = 33-66 m/m.y.; % org. C (0-180 m) > 2.6, (180-280 m) 0.6-2.6, (280-400 m) 0.6-1. 3) 2-4 9 -1 3-5 19 -193 -70. 9 16-2 140 0 (.2) 16-4 143 -187 -60.9 20-3 180 -2 (.3) 20-6 184 -187 -54.2 22-5 203 -182 -50.9 23-2 208 -191 -50.2 29-5 270 -166 -46. 4 32-6 300 -172 -46.2 34-6 319 -164 -45.3 35-4 327 -173 -44.9 37-4 346 -172 -43.2 40-3 374 -174 -43.0 42-6 398 -171 -42.3 43-4 404 -175 -43.8 gas hydrate 404 44-4 414 1 ( 1 ) 70

PAGE 81

Table (Cont'd). Sample Depth (core-section) (m) o13c-CH 4 0 0 Site 569 (90 50.4'W, 12 56.3'N; w ater depth= 2800 m ; sed. rate (100---m) = 25 m /m.y.; % org. c m) 12-6 105 -186 -80.6 13-3 115 2 15-3 129 -185 (.5) -78.8 18-2 156 -185 2 -77.9 23-3 207 -179 -73.7 219 -18 2 -71 1 25-2 -178 ( 1) -70.5 26 5 238 -18 1 -69.2 7 1 570 (91 6'W, 13,1'N; water depth= 1718 m; sed. rate (0-230 m) = 130 m /m.y., (220-260 m) = 1 3 m /m.y., (260-330 m) = 66 mlm.y.; % org. C (0-275 m) 1.1-3.2, (275 -350 m) = 0.1-3.0) 3 6 22 -19 3 -79.6 6 1 -188 -71.6 11-1 96 -190 -68.8 15-2 136 -197 (.5) -68.6 16-6 1 5 1 196 -68 2 20-1 183 -187 -65. 4 Hydrate 192 21-2 193 -189 -62 6 24-3 224 -187 -58. 7 25-4 234 -181 -55.3 26-4 245 190 53 0 Hydrate 246 Hydrate 249 Hydrate 254-268 Hydrate 273 294 -186 -41 6 30-4 284 -185 (.8) -42.5 3 1-1 289 -185 -42.8 32-4 302 -19 3 -42.4 Hydrate 303 341 318 -187 -41 6 352 329 -188 (.4) -41 9 36-1 336 -187 -40.9 Hydrate 338 37-1 346 -189 -41 6

PAGE 82

72 and depth distribution of samples available for analysis in this study make detailed interpretation difficult; however, it appears that the methane recovered from the Sites where gas hydrates were either not visually observed at all (567A, 569) or were only observed near the bottom (568) is generally slightly less D-depleted (mean CD-CH4 = 0 0 -179.5 /oo, std. dev. = 7.6 /oo, n = 24) than the methane recovered from the Sites ( 565, 570) where hydrates were visually observed (Kvenvolden and McDonald 1985) at relatively shallower subbottom depths 0 0 (mean OD-CH4 = -190.5 /oo, std. dev = 4.8 /oo, n = 24;Table 4). Claypool et al. ( 1 985) found that significantly more deuterium enrichment in methane occurs at depth at Site 568 from which gas hydrates were recovered only in the deepest part of the section penetrated, than in the methane collected at comparable depths at Site 570 where gas hydrates are massively (3 to 4 meters thick) developed at intermediate (250 m) depths. Galimov and Kvenvolden ( 1 983) suggested that, although the o13c of CH4 and co 2 in gas hydrates should not be directly influenced as a result of their hydrated state, methane in zones of gas hydrate might be slightly 1 2 c -enriched compared to methane from non-hydrated regions because escape of isotopically light methane might be inhibited due to encagement by the water-clathrate structure. If this line of reasoning is extended, it might be expected that methane sampled from gas hydrate deposits could also be relatively more D-depleted. This is apparently observed (Table 4, Claypool et al. 1985) Confinement within the clathrate structure might also reduce the magnitude of possible isotopic fractionation induced by rapid depressurization during sampling. The methane sampled from Sites 567A, 568 and 569 shows no 13c-enrichment relative to that found at Sites

PAGE 83

73 565 and 570, however, reducing the strength of any argument for light-isotope enrichment in gas hydrates that might be made from the 60 data. Further, two separate laboratory studies in which naturally formed gas hydrates were sampled i11termi t tently during controlled decomposition revealed no obvious carbon isotopic fractionation during hydrate decomposition (Brooks et al. 1983; Kvenvolden et al. 1984). Large differences in parameters such as sedimentation rate and percent organic carbon (Kvenvolden and McDonald 1985) among Leg 84 Sites are more likely causes of the slight differences in 6D-CH4 noted above. The o13 c of methane (as heavy as -41/oo) sampled in the lower parts of Sites 568 and 570 (Claypool et al. 1985; Jeffrey et al. 1985; Table 4) implies that the methane was formed from thermal decomposition of co mplex organic matter (Bernard et al. 1977). Profiles of o13 c-cH 4 and o13c-ECO are generally parallel and o13c-ECO as heavy as +35.7 2 2 and +37.5/oo were measured in Sites 568 and 570, however, implying that the anomalously 13c-enriched methane was formed from the reduction of very 1 3c-enr iched CO2 (Claypool et al. 1985). Further, the o 1 3 c difference between the anomalously heavy CH4 and co2 sampled at Sites 568 and 570 is essentially the same as f ound at other DSDP Sites where biogenic methane exhibits a more typical o13c (Claypool et al. 1985). DSDP Leg 87 Nankai Trough Site 583 of DSDP Leg 87 sampled mildly deformed sediments on the lowest structural terrace of the landward slope of the Nankai Trough, which is located off Honshu, Japan. In contrast to the Middle America Trench, most or all of the trench fill covering the descending plate is scraped off and accreted in response to the low rate of subduction (-2 em/yr.) in the Nankai Trough (Karig, Kagami et al. 1983).

PAGE 84

Isotopic analyses of methane and pore water from Site 583 are presented in Table 5. Both oD-CH4 and oD-H2 o are constant with depth, within the analytical precision of the measurements. The oD-CH4 is typical of values normally exhibited by biogenic methane f o rmed from co 2 reduction in deep-sea sediments (Schoell 1983 ) and the oD-H2 o is within the range found in other DSDP pore waters ( Lawrence and Gieskes 1981 ) Claypool et al. (in press) presented depth profiles of o13c-cH 4 and at Site 583 that exhibited similar shapes with 74 most of the variation in these parameters occurring in the upper 30 m of the sediments. They interpreted these results to mean that methane was formed via co2 reduction and that methanogeni c activity probably ceased below depths of 30 to 50 m (Claypool et al. in press) The organic carbon content of these sediments is about 0 4 to 0.6% (Claypool et al. in press), approximately the amount thought necessary to support methane production (Rice and Claypool 1981) The apparently limited extent of methane production in these sediments combined with the great water depth (4618 m) make gas hydrate formation at Site 583 unlikely (Claypool et al. in press). DSDP Leg 96 Orca Basin Orca Basin is a 400 km2 depression on the Louisiana continental slope located in a region where salt diapirism and slumping have created a complex seafloor morphology (Lehner 1969; Shokes et al. 1977). Dissolution by seawater of a near-surface salt deposit on the basin flank apparently accounts for the 200 m thick layer of anoxic, hypersaline (250 g/kg) water that occupies the bottom of the basin (Shokes et al. 1977; Wiesenburg 1980) The brine is near saturation respect to NaCl and is sulfate-enriched (47 2 mM) relative to deep

PAGE 85

7 5 Table 5. Isotopic analyses of methane and water (in /oo). from DSDP Site 583 Sample Depth 6D-CH4 6D-H2o o13c-CH (Hole-core-section) (m) 4 D-3-1 72 -189 -71 1 D-6-2 100 -186 -70.6 D-9-2 128 4 D-11-1 150 -187 -70.6 D-12-2 156 3 D-14-2 178 -188 2 -70.2 D-1 6-2 198 -190 -69.7 D-19-1 228 2 D-22-4 256 3 D-22-6 260 -188 -69.1 D-24-6 274 -185 -68.9 D-29-2 325 4 G-3 325-336 -192 -68.1 G-7-2 370 -188 -66.9 G 9-1 388 -186 -66.5 G-12 411-419 -184 -66.3 G-14 431-439 -189 -66.6 --mean -188 3 Std. dev. 2.2 0.9

PAGE 86

76 Gulf of Mexico seawater (Sheu et al. in press). The dense brine is 'highly stable and mixes very slowly with the overlying seawater. As a result, methane produced in the underlying sediments (Sackett et al. 1979) has accumulated in the brine to a concentration nearly six orders of magnitude greater than in the overlying deep Gulf seawater (Wiesenburg 1980) From consideration of seismic reflection profiles, geotechnical and clay mineralogical properties, and high sedimentation rates (-1 m/ 1000 yr.), Tompkins and Shephard (1979) concluded that slumping of basin flank sediments, both from within and from outside the area covered by brine, is the primary m e chanism of sediment accumulation in the central part of the basin, and that the amount and origin of slumped material varies spatially within the basin. Near-surface basin sediments originally deposited under anoxic conditions are enriched in total organic carbon over normal slope sediments by a factor of 2 to 3 and contain large pieces of undecomposed Sargassum (Sackett et al. 1979; Northam et al. 1981). These observations imply that organic matter in Orca Basin sediments is relatively well-preserved, as a result of reduced rates of biodegradation caused by the hypersalinity (Sackett et al. 1979) and/or the anoxic conditions (Northam et al. 1981 ) Isotopic analyses of Site 618 interstitial methane and water are presented in Table 6. Application of the Woltemate et al. (1984) model (Figure 4, equations (4) and (5)) to these data (Table 6) implies that the acetate dissimilation pathway could account for 15% of the total methane production in Orca Basin sediments. Whelan et al. (in press) report a pore water acetate concentration of about 1 mM at 15m depth

PAGE 87

77 Table 6. Isotopic analyses of methane and water from DSDP Site 618 (in /oo). Sample Depth 6D-CH!! 6D-H2 0 o13c-CH 4 (Core-section) (m) 1-2 3 5 1-3 5 -181 -7!!.9 2-4 1 2 8 3-2 19 8 3-5 2!! -186 -73.3 l!2 28 8 ll-6 3!! -185 -73.1 5-5 ll2 8 5-7 ll5 -18ll -72 8 7-1 52 7 7-3 54 -186 -72.6 8-1 58 5 8-5 6!! -1 85 -72.7 9-2 73 -18!! 5 -72.7 11-1 90 5 11-2 92 -181 -75.7 mean -18!! 7

PAGE 88

78 at Site 618. This concentration is about a factor of two higher than the Km (substrate concentration at which the initial reaction velocity is half maximal) for the acetate-decarboxylating methane-producing bacterium Methanobacterium soehngenii that was isolated from digested sewage sludge (Zehnder et al. 1980) Recently, isolation of a new acetotrophic marine methane-producing bacterium, Methanosarcina acetivorans, from methane evolving sediments was reported by Sowers et al. ( 1984). If an acetotrophic methanogen inhabits Orca Basin sediments, and if its Km is similar in magnitude to that of M. soehngenii, then some production of methane from acetate, at least at 15 m depth, may be possible. The preceding argument assumes that a significant fraction of the acetate detected is 11microbiologically available", a condition that may or may not be met. At 15 m subbottom depth at Site 618 the interstitial salinity was about 115/oo (Bouma, Coleman et al. in press). acetivorans exhibits a maximum specific growth rate at about 0.2 M NaCl concentration (about 12/oo salinity) that decreases rapidly with increase in NaCl concentration (Sowers et al. 1984) Thus, any acetotrophic methanogens living at 15 m depth may be inhibited by the high salinity. This would agree with the suggestion of Sackett et al. (1979) that the rate of overall organic matter destruction may be slowed by the hypersalini ty. Although acetate dissimilation may account for a small percentage of the methane production in Orca Basin sediments, most of the methane probably results from C02 reduction. The salinity (Bouma, Coleman et in press) and oD-H2o (Table 6) profiles for Site 618 imply that the Orca Basin brine has a 60 of about

PAGE 89

79 0 4 /oo. This is heavier than present day Gulf Basin Water (oD = 0.7 to 1.1/oo; Frank 1973) and implies that the Orca Basin brine was formed sometime between the last glacial maximum (18,000 yr. B .P.; Mcintyre et al. 1976) and the end of the subsequent deglaciation period (about 14, 000 to 5,000 yr. B.P.; Barry et al. 1979) when significant quantities of isotopically light water (Dansgaard and Tauber 1969) were tied up in continental ice sheets. It has been estimated that ocean deep waters were about 1 2 (Dansgaard and Tauber 1969) to 1 .65/oo (Shackleton 1977) 18o-enriched at the last glacial maximum relative to present day deep water. This would correspond to a deuterium enrichment of about 8 to 11/oo, assuming the deuterium-18o relationship for North Atlantic Surface Water of Craig (1967) applies. If these estimates of the deep ocean isotopic fractionation and the brine oD (4 to 5/oo) are accurate, then the age of the Orca Brine might be further constrained to the deglaciation period (14,000 to 5,000 yr. B.P.). In support of this, the estimate of Addy and Behrens (1980) of when the brine began to accumulate in the basin (about 8,000 yr. B.P.) falls within this range of time. There is, however, some uncertainty about how much sea level was actually lowered during the last glacial maximum. Blackwelder et al. ( 1979 ) estimated that substantially less ice was present from 17,000 to 10,000 yr. B.P. than would be expected from the estimates of minimum sea level employed by Dansgaard and Tauber (1969) and Shackleton (1977) The interstitial methane o13c values reported in Table 6 and by Pflaum et al. (in press) are similar to values reported by Sackett et al. (1979) for the o 1 3c of methane dissolved in the brine (-74.5 to -72/oo), but quite different from interstitial methane o 13 c (-105 to

PAGE 90

0 -85 I oo) reported by Sackett et al. ( 19 79) very light (<5 13 c < -90/oo) methane has not been widely encountered in sediments sampled by DSDP (Claypool and Kvenvolden 1983). When highly 13c-depleted methane is found, it usually occurs at the base of the sulfate-reduction zone, which is generally acknowledged to be the shallowest depth of rapid production and significant accumulation of methane in deep-sea sediments (Claypool and Kvenvolden 1983). In general, very 13c-depleted methane is probably not of great quantitative significance in deep-sea sediments and may result only under certain geochemical and/or sedimentological conditions. Because the amount and origin of slumped sediments varies spatially within the basin (Tompkins and Shephard 1979), the sediments cored by Sackett et al. ( 1 979) could be quite different from those cored at Site 618. Sackett et al. ( 1979) gave a water depth of 2340 m for their core location; this depth is 80 m more shallow than at Site 618 and may actually be on the basin flank. The probable sediment variability and all of the i sotopic data imply that the very light sedimentary methane sampled by Sackett et al. (1979) is probably a localized, near-surface occurrence that is not isotopically representative of the methane component of either the Orca Basin brine or pore waters. Between the brine-sediment interface and about 3 0 m subbottom depth the interstitial salinity decreases from brine levels to values more characteristic of normal marine sediments (Bouma, Coleman et al. in press). Rapid methane production in sediment layers less influenced by high salinity effects and subsequent upward diffusion probably provides most of the methane dissolved in the brine. Pflaum et al. (in press) observed small, spherical gas hydrates 80

PAGE 91

between 20 and 40m depth at Site 618. The methane that was released upon decomposition of the hydrate had the same 6 13c as methane sampled from gas pockets in the same core section, indicating that no isotopic fractionation of meth ane was associated with formation of the hydrates (Pflaum et al. in press). Deep-sea methane isotopic compositions (mean 6D-CH4 = -185/oo, std. 0 13 0 0 dev. = 7 /oo, n = 75; o C -CH4 = -71.3 /oo, std. dev. = 6.3 /oo, n = 44) determined in this study (Tables 4-6) are, for the most part, within the narrow ranges (6D = -190 to -170/oo; 6 13c = -75 to -60/oo) considered by Schoell (1983) to be characteristic of methane formed by co2 reduction in deep-sea sediments. The mean 6 13c was calculated excluding all data from Sites 568 and 570 because the anomalously heavy methane from those Sites constitutes a relatively much larger fraction of this data set than of the data set containing all known DSDP o13C-CH4 values. The pore waters measured in this study are somewhat more D-enriched (6D = +3/oo, std. dev. = 3/oo, n 0 waters from 24 DSDP Sites (6D = -1.8 /oo, std. dev 26) than pore 0 5 /oo, n 165; Lawrence and Gieskes 1981; Friedman and Hardcastle 1973; Jenden and Gieskes 1 983; Lawrence unpublished data) with a wide geographic distribution. If co2 reduction is by far the predominant source of methane in deep-sea sediments, as proposed by C l ayp o o l and Kaplan 0 (1974) then the mean 6D values for methan e ( 185 /oo) and pore water (6D = -1/oo, n = 191) discussed here indicate that equation (5) describing co2 reduction (Woltemate et al. 1984 ) should be modi fied to: 6D-CH4 0 1 8 4 ( 1 5 ) I oo (8) Alternatively, deviation of deep-sea 6D-CH 4 from equation (5) could be explained by minor production (about 10% of the total) of 81

PAGE 92

82 methane from acetate as discussed previously for DSDP Site 618 Few determinations of acetate concentration in deep-sea pore waters have been made to date. Acetate concentrations that are available for typical sulfate-depleted deep-sea sediments (DSDP Site 619 -Mississippi Fan) range from 21 to 131 )..lM (Whelan et al. in press). Significant (i.e., 10% of the total) methane production from acetate concentrations as low as 100 )..lM would require that the rate constant for acetate conversion to methane be much higher than the rate constant for co2 reduction because deep-sea pore water rco2 concentrations are typically two to three orders of magnitude ( 10 to 100 mM) higher (Claypool a nd Kvenvolden 19. 83; Claypool et al. 1985). Baker et al. (1977) determined the oD of a few biogenic methane samples from DSDP Leg 41 (eastern tropical Atlantic) and obtained values that were almost identical to oD-CH4 from DSDP Site 147 (Cariaco Trench) measured by Lyon (1974) Based on this similarity, Baker et al. (1977) suggested that oD-CH4 in deep-sea pore waters is geographically invariant and proposed that this could result if the deuterium concentration of deep-sea pore waters is uniform and either the isotopic fractionation associated with bacterial methane production is fairly constant or if methane and the interstitial water undergo postgenerative isotopic equilibration. Using available temperature-dependent equilibrium isotope fractionation factors and average oD-CH4 (-178/oo) and oD-H2o (+8/oo; Friedman and Hardcastle 1973 ) Lyon (1974) calculated an equilibrium temperature for Cariaco Trench (Site 147) sediments of 23C, which is about 40C lower than the estimated true temperature. 0 Average oD-CH4 (-185 /oo) and oD-H2o ( +3 /oo) obtained in this study also indicate deep-sea sediment

PAGE 93

83 temperatures of < 20C if evaluated with the equilibrium fractionation factors that were compiled by Friedman and O 'Neil (1977). Further, Nakai et al. ( 1974) found that observed differences in cSD values between methane and water from biogenic gas deposits in Japan were greater by about 50 to 80/oo than would be predicted for isotopic equilibrium. These results imply that the large differences in cSD between methane and water observed in these environments is due to a large kinetic isotope effect associated with bacterial methane formation as proposed by Nakai et al. (1974) Reevaluation of the deep-sea cSD data with recently published equilibrium fractionation factors (Lyon and Hulston 1984), however, yields predicted temperatures that are generally within about 5C of estimated environmental temperatures (Table 7). Thus, low-temperature (0 to 20C) isotopic equilibration between methane and interstitial water during the great length of time (10 5 to 106 years) represented by a deep-sea core may play an important role in determining the cSD of methane associated with deep-sea sediments. In contrast, equilibrium temperatures predicted with cSD data from freshwater and nearshore marine-estuarine sediments are substantially lower than measured temperatures (Table 7), implying that low-temperature isotopic equilibration does not occur on the short time scale represented by these young sediments.

PAGE 94

84 Table 7. Comparison of isotopic equilibrium temperatures (T ) as calculated with the data of Lyon and Hulston (1984)eq. with estimated environmental temperatures (T ). env. Location oD-CH4 oD-H2 0 T eq T env. M T eq. -Ten v. ) ( 0/oo) ( 0/oo) (OC) (OC) (oC) Deep-sea (DSDP) -178a +8b 147 (Cariaco Trench) 12 17 -5 565 (below 200 m) -196 0 -1 5 -6 568 (140-180 m) -190 -1 11 6 5 568 (400 m) -174 +1 25 24 1 569 (100-150 m) -185 +2 9 7 2 618 -184 +7 5 6 -1 Cape Lookout Bight 16 June 1983 115 -222 +3 -33 23 -56 19 August 1983 113 -284 +3 -82 27 -109 2 Fevruary 1984 112 -231 +3 -42 7 -49 Crescent Lake, FL 31 August 1984 112 -304 -7 -92 31 -123 23 January 1985 #2 -263 -10 -61 16 -77 a-from Lyon (1974) b from Friedman and Hardcastle ( 1973)

PAGE 95

85 SUMMARY Methanogenesis, the final step in the anaerobic decomposition of complex organic matter in natural systems, generally occurs via one of two pathways, co2 reduction or acetate dissimilation. Most 14c radiotracer studies (i. e., Lovley and Klug 1982) performed to date have yielded results indicating acetate dissimilation is the dominant methanogenic pathway in freshwater sediments, accounting for 60 to 70% of the total methane production. In contrast, methane production in deep-sea sediments is generally thought to occur almost exclusively via co2 reduction (i.e., Claypool and Kaplan 1974): The data generated in this and other studies (Woltemate et al. 1984; Whiticar et al. in press) indicate that biogenic methane produced in young, freshwater sediments can generally be distinguished from the biogenic methane produced in deep-sea sediments by measurement of oD-CH4 implying that oD-CH4 may be of us e in the genetic characterization of biogenic methanes (Schoell 1 983) and in estimating the relative importance of the two meth anogenic pathways in nature (Woltemate et al. 1984) The isotopic compositions of methane samples collected at eight different DSDP Sites (mea n oD = -185/oo, std. d e v = 7/oo, n = 75; mean o 1 3c = -71.3/oo, std. dev. = 6 .3/oo, n = 44) are generally typical of methane formed via co2 reduction in deep-sea sediments. The absence of wide variation in deep-sea oD-CH4 could be the result of either a relatively constant kinetic hydrogen isotope effect associated

PAGE 96

with co2 reduction or of postgenerati ve isotopic equilibration of methane and pore water during the l ong subsurface burial of these 86 sediments. Resolution of this question through laboratory studies m ay be quite difficult, as this approach would require extrapolation of results obtained under conditions that are vastly different from in situ deep-sea conditions because of the great length of time and, presumably, the slow rate of methane production under natural deep-sea conditions. The isotopic compositions of methane samples collected from gas hydrate zones at several DSDP Sites indicated no obvious isotope effects associated with gas hydrate formation. Production of minor amounts (about 10% of the total) of methane via acetate dissimilation in deep-sea sediments may explain the deviation of the 6D-H2 0/6D-CH 4 pairs from the previousl y published relationship for co2 reduction. The methane isotopic compositions of gas bubbles collected from the sediments of several freshwater environments (mean 6D-CH4 -300/oo, std. dev. = 26/oo, n = 20; 13 0 mean 6 C-CH4 = -60.1 /oo, std. dev. = 6.1 /oo, n = 20) imply that the methane produced in these freshwater sediments is generally more D -depleted and less 13C-depleted than methane produced in deep-sea sediments. Evaluation of the 6DH 2 0/0D -CH4 with the Woltemate et al. ( 1984) model implies that acetate dissimilation is responsible for about 50 to 80% of the methane produced in these freshwater sediments. Methane o 13c and oD are generally inversely correlated, indicating that the large isotopic variability observed in these freshwater sediments is not solely attributable to bacterial methane oxidation, but also to some other parameter such as variability in methanogenic pathway selection.

PAGE 97

87 Nearshore marine-estuarine methanes are isotopically intermediate ( cSD CH 2580/ 0 46 1 3 mean 4 = -oo, std. dev. = 23 /oo, n = ; mean cS c-cH 4 -61 .8/oo, std. dev. = 3 .1/oo, n = 46) between deep -sea and freshwater methanes, implying that acetate dissimilation plays an intermediate role in nearshore mar ine-es tuar i ne methane product ion. Isotopi c analyses of met hane samples collected over the course of a year from the rapidly accumulating, organic-rich marine sediments of Cape Lookout Bight, North Carolina indicate that methane isotopic composition and probably, methanogenic pathway selection are seasonally, and in all likelihood spatially variable. Methane is generall y D-depleted and 13c-enriched during warm summer months (mean cSD = 269 /oo, std. dev. 9/oo, n = 21; mean o13c = -59.5/oo, std. dev. = 4.4/oo, n = 21) as compared to winter months (mean cSD = -236/oo, std. dev. = 4/oo, n = 6; mean o1 3 c-cH 4 0 0 -63.9 loo, std. dev. = 0 5 /oo, n = 6). Evaluation of these data with the Woltem ate et al. (1984) model implies that acetate dissimilation accounts for a relatively greater proportion of total methane production during the summer (about 45 to 60%) than during the winter (about 25 to 40%). The reasonably good agreement of the summer estimates obtained in this manner with those obtained from 14c radiotracer studies (Sansone and Martens 1981; Crill and Martens in press) indicates that application of the Woltemate et al. (1984) model to cSD-CH 4 data can yiel d useful estimates of methanogenic pathway selection in aquatic sediments. As for the freshwater sediments described above, the o13c and cSD of methane produced in Cape Lookout Bight sediments are inversely correlated, implying that both cSD-CH4 and cS 1 3c-CH are influenced by methanogenic pathway and that bacterial 4 methane oxidation plays a relatively minor role in producing the

PAGE 98

88 methane isotopic variability observed in these sediments. The nature and extent of post-depositional bacteria-organic matter interactions occurring prior to rapid methane production probably play a role in determining methanogenic pathway selection and methane isotopic composition in aquatic sediments. Factors t hat likel y influence these interactions are amount and/or type of organic matter deposited, sedimentation rate, concentrations of external electron acceptors and temperature. Land-derived higher plant material with its relatively lower H/C and N/C ratios (Stuermer et al. 1978) and relatively higher concentrations of biologically resistant lignins, resins and waxes (Aizenshtat et al. 1973) is probably not as readily degraded as algae-derived organic matter. Several investigators (Jorgensen 1978; Berner 1980; Westrich and Berner 198 4 ) suggested that decomposable organic matter should be divided into several fractions (two to eight) based on relative degradability. Model profiles of the decomposable organic matter indicate that the various fractions successively disappear based on their relative degradabili ties with depth in the sediment (time), with the result that the remaining fraction becomes progressively more refractory (Berner 1980) A high rate of sediment deposition would favor preservation of the more easily degraded organic matter fractions because both aerobic remineralization occurring at or near the sediment-water interface and the diffusion of other external electron acceptors into the sediments would be minimized by rapid isolation of organic matter from the overlying water (Berner 1980) In marine systems, where dissolved sulfate is by far the most abundant (28 mM) external electron acceptor, sulfate reduction may

PAGE 99

89 account for a significant part of the total organic matter degradation that occurs prior to rapid methane production. Jorgensen (1982) reported that bacterial sulfate reduction accounted for as much organic matter .degradation as did all aerobic processes in Danish coastal sediments. In general, the ratio of [dissolved sulfate]/[organic matter] should decrease in the order deep-sea > nearshore marine-estuarine > freshwater sediments. Because the consumption of the most labile organic matter fractions by aerobic processes and sulfate reduction should be maximized in sediments with low sedimentation rates and high [dissolved sulfate]/[organic matter] ratios, the methane producing zone of deep-sea sediments would be expected to receive a relatively smaller proportion of the more easily degradable organic matter fractions than the methane zone of freshwater sediments. Rapidly accumulating, nearshore marine-estuarine sedim ents which are deposited in waters containing appreciable co ncentrations of dissolved sulfate and receive organic matter of mixed origin, might be expected to fall between deep-sea and freshwater sediments with respect to the relative lability of organic matter reaching the methane zone. In general, nearshore marine-estuarine oD-CH4 measured in this study (mean oD = -258/oo) are intermediate between the freshwater (mean oD -300/oo) a nd deep-sea (mean oD = -185/oo) methanes implying that the lability of the organic matter reaching the methane zone may affect methanogenic pathway selecti on and There is, however, significant overlap between freshwater and nearshore marine-estuarine OD-CH 4 ; this probably reflects the great variability in important environmental parameters (i.e., sedimentation rate, organic matter input) that is inherent in these types of aquatic systems.

PAGE 100

90 The results discussed above imply that the 60 of methane produced in deep-sea sediments is sufficiently different from the 60 of methane produced in shallow aquatic sediments to allow the use of 6D-CH4 in the genetic characterization of commercially significant biogenic methane deposits. Rice and Claypool ( 1981) argue that shallow aquatic sediments are not normally suitable for the accumulation of biogenic gas because rapid production of methane usually begins near the sediment-water interface and hydrostatic pressures are low. As a result, most of the methane is lost from shallow sediments, either by bacterial consumption or by ebullition and escape to the atmosphere (Rice and Claypool 1981). Deep-sea sediments should be more appropriate for biogenic gas accumulation because rapid methane production usually begins deeper in the sediment and higher hydrostatic pressures allow retention of larger quantities of gas in solution in pore waters until the sediments are compacted and traps and seals are formed (Rice and Claypool 1981). Only a few 60-H2 0/6D-CH 4 pairs from a small proportion of the known commercial accumulations of biogenic gas are available in the literature. Surprisingly, most of the waters measured are substantially more 0-depleted than ocean pore waters (Nakai et al. 1974; Schoell 1980); this indicates that the methane may have originated in deep freshwater sediments. The differences in 60 values between methane and water from these commercial biogenic gas deposits is about the same as is found in deep-sea sediments; this was interpreted by Schoell (1980) to mean that the methane in these gas deposits was formed by co2 reduction. If, on the other hand, the observed differences in 60 values between methane and water represent significant post-burial isotopic equilibration in these ancient

PAGE 101

deposits, then measurement of oD-H2 0/oD-CH 4 pairs would probably be of little value in the genetic characterization of commercial biogenic methane deposits. 91 The microbial decomposition of organic matter in water-covered soils is one of the major sources of methane to the atmosphere (Khalil and Rasmussen 1983; Senum and Gaffney 1985), and bubble ebullition is the dominant mode of methane input to the atmosphere from some shallow freshwater (Robertson 1979; Cicerone and Shetter 1981) and marine (Martens and Klump 1980) environments. The results of this and other studies, indicate that methane bubbles formed in shallow aquatic sediments are generally significantly more 13c-depleted than the calculated average (Senum and Gaffney 1985) of all methane sources to the atmosphere. Some of the freshwater sediments studied here and by Woltemate et al. (1984) produce methane bubbles that have a oD near the estimated source average (Senum and Gaffney 1985); however there are presently too few measurements of the oD of atmospheric methane and its sources available to allow evaluation of the role of shallow aquatic sediments in determining the oD of atmospheric methane. Future measurements of: (I) the fluxes (bubble and diffusive) and isotopic compositions (oD and o13c) of methane emitted from various aquatic sediments and other major sources (i.e., ruminants), (II) the isotopic composition of atmospheric methane and, (III) various ancillary parameters (i.e., temperature, salinity, organic matter content) that may affect methane production rate and/or isotopic composition are necessary to allow accurate assessment of the importance of shallow aquatic environments to the global atmospheric methane budget.

PAGE 102

LITERATURE CITED Addy, S. R., and E. W. Behrens 1 980 Time of accumulation of hypersaline anoxic brine in Orca Basin (Gulf of Mexico). Mar. Geol. 37: 241252 Aizenshtat, Z., M J Baedecker, and I. R. Kaplan 1973. Distribution and diagenesis of organic compounds in JOIDES sediment from the Gulf of Mexico and Western Atlantic. Geochim. Cosmochim Acta 37 : 1881-1898. Alperin, M. J., and W. S. Reeburgh. 1984. Geochemical observations supporting anaerobic methane oxidation, p. 282-289. In R. L Crawford and R. s Hanson (Eds ) Microbial Growthon c1 compounds. A merican Society for Microbiology. 92 -------, and ------. 1985. Inhibition experiments on anaerobic methane oxidation. Appl. Environ. Microbial 50: 940-945 Anonymous. 1983. Energy demand, supply, and needed development resources are assessed. Oil and Gas Jour. 81: 71-73. Aubouin, J., and R von Huene. 1985 Summary: Leg 84 Middle America Trench transect off Guatemala and Costa Rica, p 939957. In R. von Huene, J. Aubouin et al., Init. Repts. DSDP 84. U.S. Govt. Printing Office. Bainbridge, A. E., H E Suess, and I Friedman. 1961. Isotope composition of atmospheric hydrogen and methane Nature 192: 6l.J8-649 Baker, E. W., W. Y Huang, J. G. Rankin, J. R. Castano, J. R. Guinn, and A. N Fuex. 1977. E lectron paramagnetic resonance study of thermal alteration of kerogen in deep -sea sediments by basaltic sill intrusion, p. 839-847. In Y Lancelot, E. Seibold, et al., Init. Repts. DSDP 41. U.S. Govt. Printing Office. Baltuck, M., R von Huene, and J Aubouin 1985 Introduction and Explanatory notes, p. 5-19. In R von Huene, J Auboin, et al., Init. Repts. DSDP 84. U.S. Govt. Printing Office. Barnes, R. 0., and E D Goldberg. 1976. Methane production and consumption in anoxic marine sediments. Geology l.J: 297-300. Baross, J. A., M. D. Lilley, and L. I. Gordon. 1982 Is the CH4 H 2 and CO venting from submarine hydrothermal systems prod uced by

PAGE 103

93 thermophilic bacteria? Nature 298: 366-368. -------, and J. W. Deming. 1983. Growth of "black smoker" bacteria at temperatures or at least 250C Nature 303: 423-426. Barry, R., H. Fritts, J. Imbrie, J. Kutzbach, J. M. Mitchell, and S. M. Savin. 1979. Paleoclimatic research: Status and opportunities. Quat. Res. 12: 6-17. Bartlett, K. A. 1981. Macrofauna distribution and seasonal influences on interstitial water chemistry of Cape Lookout Bight, N.C. M.S. thesis, Univ. North Carolina, Chapel Hill. 98p. Begemann, F., and I. Friedman. 1968. Tritium and deuterium in atmospheric methane. Jour. Geophys. Res. 73: 1149-1153. Belyaev, S. S., Z. I. Finkel'shtein, and M. v. Ivanov. 1975. Intensity of bacterial methane formation in ooze deposits of certain lakes. Mikrobiologiya (English translation) 44: 272-275. -------, R. Wolin, W. R. Kenealy, M. J. DeNiro, s. Epstein, and J. G. Zeikus. 1983. Methanogenic bacteria from the Bondyuzhskae oil field: general characterization and analysis of stable carbon isotopic fractionation. Appl. Environ. Microbial. 45: 691-697. Bernard, B. B. 1978. Light hydrocarbons in marine sediments. Ph.D. thesis, Texas A&M Universty, College Station. 144p. ------J. M. Brooks, and W. M. Sackett. 1976. Natural gas seepage in the Gulf of Mexico. Earth Planet. Sci. Lett. 31: 48-54. ------J. Brooks, and W. M Sackett. 1 977. A geochemical model for characterization of hydrocarbon gas sources in marine sediments. Offshore Tech. Conf. OTC 2934: 435-438 Berner, R A. 1980. Early diagenesis a theoretical approach. Princeton Uni v. Press. Blackwelder, B. w., 0 H. Pilkey, and J. D. Howard. 1979. Late Wisconsin sea levels on the southeast U. S. Atlantic shelf based on in-place shoreline indicators. Science 204: 618-620. Blair, N., c s. Martens, and D. Des Marais. 1984. Seasonal o1 3c variations from an anoxic marine sediment. EOS 65: 960 Blunden, G., S.M. Gordon, W F. H. McLean, and M.D. Guiry. 1982. The distribution and possible taxonomic significance of quaternary ammonium and other Dragendorff-positive compounds in some genera of marine algae. Bot. Mar. 25: 563-567 Bouma, A. H., J. M. Coleman, et al. 1985. Site 618 summary, in press. H. Bouma, J. M. Coleman, et al., Init. Repts. DSDP 96. U.S. Govt. Printing Office.

PAGE 104

94 Brock, T. D. 1974. Biology of microorganisms. Prentice-Hall. Brooks, J. M., L.A. Barnard, D. A. Wiesenburg, M. c Kennicutt, and K. A. Kvenvolden. 1983. Molecular and isotopic compositions of hydrocarbons at Site 533, DSDP Leg 76, p. 377-390. In R. E. Sheridan, F. M. Gradstein, et al., Init. Repts. U.S. Govt. Printing Office. Burdige, D. J., and K. H. Nealson. 1985. Microbial manganese reduction by enrichment cultures from coastal marine sediments. Appl. Environ. Microbiol. 50: 491-497. Burke, R. A., J. M. Brooks, and W. M. Sackett. 1981. Light hydrocarbons in Red Sea brines and sediments. Geochim. Cosmochim. Acta 45: 627-634. Cappenberg, Th. E., and R. A. Prins. 1974. Interrelations between sulfate-reducing and methane-producing in bottom deposits of a fresh-water lake. 3. Experiments with C-labeled substrates. Ant. van Leeuwenhoek 40: 457-469. Cavalieri, A. J., and H. C. Huang. 1981. Accumulation of proline and glycinebetaine in Spartina alterniflora Loisel in response to NaCl and nitrogen in the marsh. Oecologia 49: 224-228. Chanton, J. P., C. S. Martens, and G. W. Kipphut. 1983. Lead-210 sediment geochronology in a chang i ng coastal environment. Geochim. Cosmochim. Acta 47: 1791-1804. Chen M., and M. J. Wolin 1977. Influence of CH4 production by Methanobacterium ruminatum on the fermentation of glucose and lactate by Selenomonas ruminatum. Appl. Environ. Microbiol. 34: 756-759. Chung H. M 1976. Isotope fractionation during the maturation of organic matter. Ph.D. thesis, Texas A&M Univ., College Station. 162 p. Claypool, G. E., and I. R. Kaplan. 1974. The or1g1n and distribution of met hane in marine sediments, p. 99-139. In I. R. Kaplan (Ed.), Natural gases in marine sediments. Plenum. -------, and K. A. Kvenvolden. 1983. Methane and other hydrocarbon gases in marine sediments. Ann. Rev. Earth Planet. Sci. 11: 299-327. -------, C. E. Threlkeld, P. M. Mankiewicz, M. A. Arthur, and T. P. Anderson. 1985. Isotopic composition of interstitial fluids and origin of methane in slope sediment of the Middle America Trench, p. 683-691. In R. von Huene, J. Aubouin, et al., Init. Rept. DSDP 84. U.S. Printing Office. -------, A. K. Vuletich, and K. A. Kvenvolden. 1985. Isotopic composition of interstitial fluids in sediment of the Nankai

PAGE 105

95 Trough DSDP Leg 87. In D E. Karig, H Kagami, et al., Init. Repts. DSDP 87. in press. U.S Govt. Printing Office. Cloud P. E., I. Friedman, F D. Sisler, and V H. Dibeler. Microbi.al fractionation of hydrogen isotopes. Science 1394 -1 395. 1958. 127: Coplen, T. B., and R N. Clayton. 1973. Hydrogen isotopic composition of NBS and IAEA stabl e isotope water reference samp les. Geochim. Cos m ochim Acta 37: 23472349 Coleman D. D., J. B. Risatti, and M. Schoell. 1 981. Fractionation of carbon and hydrogen isotopes by methane-oxidizing bacteria. Geochim. Cosmochim Acta 45: 1033-1037 Coleman, M. L., T. J. Shepherd, J. J. Durham, J. E Rouse, and G. R. 1982 Reduction of water with zinc for hydrogen isotope analysis. Anal. Chern. 54: 993 995 Craig, H. 1953. The geochemistry of the stable carbon isotopes. Geochim Cosmochim. Acta 3: 53 -92. -------. 1957. Isotopic standards for carbon and oxygen and correction factors for mass-spectrometric analysis of carbon dioxide. Geochim. Cosmochim. Acta 1 2: 133-149. -------. 1961. Standard for reporting concentrations of deuterium and oxygen-18 in natural wat ers. Science 133: 1 8331 834 -------. 1967. Ocean water, isotopic composition of, p 268 -271. In McGraw-Hill Yearbook Science and Technology McGraw-Hill. -------, and C. C. Chou. 198 2 Methane : the record in polar ice cores. Geophys. Res; Lett. 9: 477-481 Crill, P. M., and C S Martens. 1983 Spatial and temporal fluctuations if methane production in anoxic coastal marine sediments. Limnol. Oceanogr 28: 1117-1130. -------, and -------. 1 985. Methane production from bicarbonate and acetate in the sulfate reduction zone o f an anoxic marine sediment. Geochim Cosmochim Acta in press. Daniels, L., G. Fulton, R. W. Spencer, and W. H Orme-Johnson. 1980. Origin of hydrogen in methane produced by Methanobacterium thermoautotrophicum. J Bacterial. 141: 694698 Dansgaard W., and H. Tauber. 1969. Glacier oxygen-18 content and Pleistocene ocean temperatures. Science 166: 499-502. Davis J B., and H F. Yarbrough. 1966. Anaerobi c oxidation of hydrocarbons by Desulfovibrio Desulfuricans. Chern. Geol. 1: 137-144.

PAGE 106

Deines, P 1980. The isotopic composition of reduced organic carbon, p 329-406 In P Fritz, and J. Ch. Fontes (Eds.), Handbook of environmental isotope geochemistry, vol. 1 The terrestrial environment, A. Elsevier. Devol, A. H. 1983 Methane oxidation rates in the anaerobic sediments of Saanich Inlet. Limnol Oceanogr. 28: 738-742 Ehhalt, D H. 1 973. Methane in the atmosphere, p. 144-158 In G M Woodwell, and E V Pecan (Eds.), Carbo n and t he biosphere. Office of Information Service. -------, and A. Volz. 1976. Coupling of the CH4 with the H and CO cycle: Isotopic evidence, p. 23-33 In H. G Schlegef, G. Gottschalk, and N Pfennig (Eds.), Microbial production and utilization of gases. E. Goltze KG. Epstein, S., C. J. Yapp, and J. H. Hall. 1976. The determination of the D/H ratio of non-exchangeable hydrogen in cellulose extracted from aquatic and land plants. Earth Planet. Sci. Lett. 30: 241-251. Estep, M F., and T. C. Hoering. 1980 Biogeochemistry of the stable hydrogen isotopes. Geochim Cosmochim Acta 44: 1197-1206. Fenchel, T., and Blackburn, T. H 1979. Bacteria and Mineral Cycling. Academic. Frank, D J 1973. Deuterium distribution in the Gulf of Mexico J Phys. Oceanogr 3: 230 236. Friedman, I. 1953 Deuterium content of natural waters. Geochim. Cosmochi m Acta 4: 89-103. -------, A C. Redfield, B. Schoen and J. Harris. 1964. The variation in the deuterium content of natural waters in the hydrologic cycle. Rev. Geophys. 2: 177-224. -------, I and K. Hardcastle. 1973. Interstitial water studies, Leg 15-Isotopic composition of water, p. 901904 In B C Heezen, I. D MacGregor, et al., Init. Repts. DSDP 20. U.S. Govt Printing Office. -------, and J R. O'Neil. 1977. Compilation of stable isotope fractionation factors of geochemica l interest. In Data Geochem., 6th ed. Geol Surv. Prof. Paper 440K. U.S. Govt .-printing Office. Fuex, A. N. 1977. The use of stable isotopes in hydrocarbon exploration. J Geochem. Explor. 7: 155 188. Galimov, E. M. 1968. Isotopic composition of carbon in gases of the crust. Internat. Geol. Rev. 11: 1092-1104. -------, and K A. Kvenvolden. 1983 Concentrations and carbon isotopic

PAGE 107

compositions of CH4 and co2 in gas from sediments of the Blake Outer Ridge, DSDP Leg 76, p. 403-407. In R. E. Sheridan, F. Gradstein, et al., Init. Repts. DSDP-76. U.S. Govt. Printing Office. 97 Games, L. M., J. M. Hayes, and R. P. Gunsalus. 1978. Methane-producing bacteria: natural fractionations of stable carbon isotopes. Geochim. Cosmochim. Acta 42: 1295-1297. Giggenbach, W. F. 1980. Geothermal gas equilibria. Geochim. Cosmochim. Acta 44: 2021-2032. Gold, T., and S. Soter. 1980. The deep earth-gas hypothesis. Scientific American 242: 154-161. Gonfiantini, R. 1978. Standards for stable isotope measurements in natural compounds. Nature 271: 534-536. -------. 1981. The a-notation and the mass-spectrometric measurement techniques, p. 35-84. In J. R. Gat, and R. Gonfiantini (Sci. Eds.), Stableisotope hydrology: Deuterium and oxygen-18 in the water cycle, Tech. Repts. Series #210. IAEA. Gorden, S., and W. A. Mulac. 1975. Reactions of the OH cx2 n) radical produced by the pulse radiolysis of water vapor. Int. J. Chern. Kinet. 7 (Symp. no. 1): 289-299. Gunter, B. D., and B. C. Musgrave. 1971. New evidence on the or1g1n of methane in hydrothermal gases. Geochim. Cosmochim. Acta 35: 113-118. Hagemann, R., G. Nief, and E. Roth. Absolute isotopic scale for deuterium analysis of natural waters. Tellus 22: 712-715. Hameed, S., and R. D. Cess. 1983. Impact of a global warming on biospheric sources of methane and its climatic consequences. Tellus 35B: 1-7. Harriss, R. C., D. I. Sebacher, and F. P. Day. 1982. Methane flux in the Great Dismal Swamp. Nature 297: 673-674. Harrits, S.M., and R. S. Hanson. 1980. Stratification of aerobic methane-oxidizing organisms in Lake Mendota, Madison, Wisconsin. Limnol. Oceanogr. 25: 412-421. Hoefs, J. 1980. Stable isotope geochemistry. Springer-Verlag. Haering, T. c. 1968. Reactions of the organic matter in a Recent marine sediment. Carnegie Inst. Wash. Yearbook 67: 199-201. -------. 1984. Thermal reactions of kerogen with added water, heavy water and pure organic substances. Org. Geochem. 5: 267-278. von Huene, R., J. Aubouin, et al. 1985. Init. Repts. DSDP 84. U.S.

PAGE 108

Govt. Printing Office. Hunt, J. M 1979. Petroleum geochemistry and geology. Freeman. Ianotti, E. L., P. Kafkewitz, M. J. Wolin, and M. P. Bryant. 1973. Glucose fermentation products of Ruminococcus albus grown in continuous culture with Vibrio succinogenes: changes caused by interspecies transfer of H 2 J. Bacterial. 114: 1231-1240. Iversen, N., and B. B Jorgensen. 1985. Anaerobic methane oxidation rates at the sulfate-methane transition in marine sediments from Kattegat and Skagerrak (Denmark). Limnol. Oceanogr. 30: 944-955. Jeffrey, A. W. A., R. C. Pflaum, T. J. McDonald, J. M. Brooks, and K. A. Kvenvolden. 1985. Isotopic analysis of core gases at Sites 565-570 Deep Sea Drilling Project Leg 84, p. 719-726. In R. von Huene, J. Aubouin, et al., Init. Repts. DSDP 84. u.s.-covt. Printing Office. Jenden, P. D., and J. M. Gieskes. 1983. Chemical and isotopic compositions of interstitial water from Deep Sea Drilling Project Sites 533 and 534, p. 453-462. In R. E. Sheridan, F. M. Gradstein, et al., Init. Repts. DSDP 76. u:S. Govt. Printing Office. Jeris, J. S., and P. L. 1965. The biochemistry of methane fermentation using C tracers. J. Water Pollut. Cont. Fed. 37 : 178-192. Jones, G. E., and R. L. Starkey. 1957. Fractionation of stable isotopes of sulfur by microorganisms and their role in deposition of native sulfur. Appl. Microbial. 5: 111-118. Jones, J. G., B. M. Simon, and S Gardner. 1982. Factors affecting methanogenesis and associated anaerobic processes in the sediments of a stratified eutrophic lake. J. Gen. Microbial. 128: 1-11. Jorgensen, B B. 1978. A comparison of methods for the quantification of bacterial sulfate reduction in coastal marine sediments. Geomicrobiol. J. 1: 29-47. -------. 1982. Mineralization of organic matter in the sea bed -the role of sulfate reduction. Nature 246: 643-645. Kaplan, I. R., and S C. Rittenberg. 1964. Microbiological fractionation of sulfur isotopes. J. Gen. Microbiol. 34: 195-212. Karig, D. E., H. Kagami, and DSDP Leg 87 Scientific Party. 1983. Varied responses to subduction in Nankai Trough and Japan Trench Forearcs. Nature 304: 148-151. Khalil, M A. K., and R. A. Rasmussen. 1982. Secular trends of atmospheric methane (CH4). Chemosphere 11: 877-883. -------, and, -------. 1983. Sources, sinks, and cycles of atmospheric

PAGE 109

methane J. Geophys. Res. 88: 513151ijij. King G M. 198ija Utilization of hydrogen acetate, and "noncompetitive" substrates by methanogenic bacteria in marine sediments. Geomicrobiol. J 3: 275-306. ------. 1 98ijb. Metabolis m of trimethylamine choline, a n d glycin e betai ne by sulfate-reducing and methanoge nic bacteria in marine sediments. Appl. E nviron Microb ial. ij8: 7 1 9 725 99 -----, M J. Klug, and D R Lovley. 1983. Metabolism of acetate, methanol, and methylated amines in intertidal sediments of Lowes Cove Maine. Appl Environ. Microbial. ij5: 18ij8-1853 Kipphut, G W., and c. S Martens. organic-rich marine basin-3. methane-saturated sediments. 20ij9-2060. 1982 Biogeochemical cycling in an Dissolved gas transport in Geochim Cosmochim Acta ij6: Koepp M. 1978. D/H isotope exchange reaction between petroleum and water: a contributory determinant for D/H-isotope ratios in crude oils?, p 221222 In R. E. Zartmann (Ed.), Short papers of the Fourth Internationar-conference, Geochronology, Cosmochronology Isotope Geol ogy 1 978 U S Geol. Survey Open File Report 78-7 01. U S. Govt Printing O ffice. Koyama, T. 1963. Gaseous metabolism in lake sediments and paddy soils and the production of atmospheric methane and hydrogen J Geophys. Res 6ij: 3971-3973 Kristjansson, J K., P Schonheit, and R. K Thauer. 1982 Different Ks values for hydrogen of methanogenic bacteria and sulfate-reducing bacteria: An explanation for the apparent inhibition of methanogenesis by sulfate. Arch. Microbial. 1 31: 278 282 Kvenvolden K. A., G E Claypool, C. N. Threlkeld, and E. D. Sloan. 1984. Geochemistry o f a naturally occurring mass i ve marine gas hydrate. Org. Geochem. 6 : 703 7 1 3 ----, a n d T J McDon a ld. 1 985 Gas h y dr a t e s of the M i dd l e A m e rica Trench -Deep Sea Drilling Project Leg 8ij, p. 667-682 In R von Huene J Aubouin et al., Inti. Repts. DSDP 8ij. U.S-.-Govt Printing Office. Lacis, A., J Hansen P Lee T Mitchell, and S. Lebedeff. 1981. Greenhouse effect of trace gases, 1970 1980 Geophys Res Lett. 8: 1035-1038 Lawrence J R. and J M. Gieskes. 1981. Constraints on water transport and alteration in the oceanic crust from the isotopic composition of pore water. J. Geophys Res. 86 : 792ij 793ij. Lovley, D. R., and M J. Klug. 1 982 Intermediary metabolism of organic matter i n the sedi ments of a eutrophi c lake. Appl. Environ.

PAGE 110

100 Microbial. 43: 552-560. -------,and-------. 1983. Methanogenesis from methanol and from hydrogen and carbon dioxide in the sediments of a eutrophic lake. Appl. Environ. Microbial. 45: 1310-1315. Lyon, G. L. 1974. Isotopic analysis of gas from the Cariaco Trench sediments, p. 91-98. In I. R. Kaplan (Ed.), Natural gases in marine sediments. Plenum. -------, and J. R. Hulston. 1984. Carbon and hydrogen isotopic compositions of New Zealand geothermal gases. Geochim. Cosmochim. Acta 48: 11 61-1171. McCarty, P. 1972. Energetics of organic matter degradation, p. 91-118. In R. Mitchell (Ed.), Water pollution microbiology. WileyInterscience. Mcintyre, A., N. G. Kipp A. W. H Be, T. Crowley, T. Kellogg, J. V. Gardner, W. Prell, and W. F. Ruddiman. 1976. Glacial north Atlantic 18,000 ye ars ago: A CLIMAP reconstuction. Geol. Soc. Amer. M e moir 145: 43-76. McNair, H. M., and E. J. Bonelli. 1969. Basic gas chromatography. Varian. M ah, R. A. 1982. Methanogenesis and methanogenic partnerships. Phil. Trans. R. Soc. Lend. B 297: 599-616. Manheim, F. T., and F. L. Sayles. 1974. Composition and or1gw of interstitial waters of marine sediments, based on deep sea drill cores, p. 527-568. In E. D. Goldberg (Ed.), The sea, volume 5: Marine chemistry. Wiley-Interscience. Martens, C. S. 1976. Control of methane sediment-water bubble transport by macro-infaunal irrigation in Cape Lookout Bight, North Carolina. Science 192: 998-1000. -------, and R. A. Berner. 1974. Methane production in the interstitial waters of sulfate-depleted marine sediments. Science 185: 1 067-1069. -------, and-------. 1977. Interstitial water chemistry of anoxic Long Island Sound sediments. 1. Dissolved gases. Limnol. Oceanogr. 22: 10-25. -------, and J. v. Klump. 1980. Biogeochemical cycling in a organic-rich coastal marine basin. 1. Methane sediment-water ex c hange processes. Geochim. Cosmochim. Acta 44: 471-490. -------, G. w. Kipphut, and J. V. Klump. 1980. Sediment-water chemical ex c hange in the c oastal z one traced by in situ Rado n-222 flux measurements. Science 2 08: 285288.

PAGE 111

i01 -------and J. V. Klump. 1984. Biogeochemical cycling in a organic-rich coastal marine basin. 4. An organic carbon budget for sediments dominated by sulfate reduction and methanogenesis. Geochim. Cosmochim. Acta 48: 1987-2004 Menzies, R. J., G. T. Rowe, and L. P. Atkinson. 1968. A small basin containing anoxic sediments Cape Lookout Bight, North Carolina. Int. Rev. Hydrobiol. 53: 77-81. Mountfort, D. 0., R. A. Asher, E. L. Mays, and J. M Tiedje. 1980. Carbon and electron flow in mud and sandflat intertidal sediments at Delaware Inlet, Nelson, New Zealand. Appl. Environ. Microbiol. 39: 686-694. Nakai, N., Y. Yoshida, and N. Ando 1974 Isotopic studies on oil and natural gas fields in Japan. Chikyu Kagaku 7/8: 87-98. National Academy of Sciences. 1980. Energy in transition 1985-2010. Freeman. Nier, A. 0. 1947. A mass spectrometer for isotope and gas analysis. Rev. Sci. Instr. 18: 398-411 Nikaido, M. 1 977. On the relation between methane production and sulfate reduction in bottom muds containing sea water sulfate. Geochem. J. 11: 199-206. Northam M A., D. A. Curry, R S Scanlon, and P. L. Parker. 1981. Stable carbon isotope ratio variations of organic matter in Orca Basin sediments. Geochim Cosmochim Acta 45: 257-260. Oana, S., and E. S. Deevey. 1960. Carbon-1 3 in lake waters and its possible bearing on paleolimnology. Am. Jour. Sci. 258A: 253-272. Oremland, R. S., L. M. Marsh, and S Polcin. 1982. Methane production and simultaneous sulfate reduction in anoxic saltmarsh sediments. Nature 296: 143-145. Osmond, C. B., N. Valaane, fJ M Haslam, P. Uotila, and Z. Roksandic. 1981. Comparisons of o C values in leaves of aquatic macrophytes from different habitats in Britain and Finland; some implications for photosynthetic processes in aquatic plants. Oecologia 50: 117 124. Pflaum, R. c., J. M. Brooks, H. B. Cox, M. C. Kennicutt, and D. D. Sheu. 1985. Molecular and isotopic analysis of core gases and gas hydrates from Sites 618 and 619, DSDP Leg 96, in press. In A. H Bouma, J. M. Coleman, et al., Init. Repts. DSDP 96. U.S. Govt. Printing Office. Pine, M. J., and H. A. Barker. 1956 Studies on the methane fermentation 12. The pathway of hydrogen in the acetate fermentation. J. Bacteriol. 71: 644-648

PAGE 112

102 Redding, C. E., M. Schoell, J. C. Monin, and B. Durand. 1980. Hydrogen and carbon isotopic composition of coals and kerogens, p. 711-723. In A. G. Douglas and J R. Maxwell (Eds.), Advances in organic geochemistry 1979. Pergammon. Reeburgh, W. S. 1976. Methane consumption in Cariaco Trench waters and sediments. Earth Planet. Sci. Lett. 28: 337-344. -------. 1980 Anaerobic methane oxidation: rate depth distributions in Ska n Bay sediments. Earth Planet. Sci. Lett. 47: 345-352. -------. 1983. Rates of biogeochemical processes in anoxic sediments. Ann. Rev. Earth Planet. Sci. 11: 269-298 -------, and D. T. Heggie. 1977 Microbial methane consumption reactions and their effect on methane distributions in freshwater and marine environments. Limnol. Oceanogr. 22: 1-9. Rice, D. D., and G E Claypool. 1981. Generation, accumulation, and resource potential of biogenic gas. Am. Assoc. Pet. Geol Bull. 65: 5-25. Rigby, D., B. D. Batts, and J. W. Smith. 1981. The effect of maturation on the isotopic composition of fossil fuels. Org Geochem. 3: 29-36 Robertson, C K. 1979. Quantitative comparison of the significance of methane in the carbon cycles of two small lakes. Arch. Hydrobiol. Beib. Ergebn. Limnol. 12: 123-135. Rudd, J. W. M., and R D. Hamilton. 1978. Methane cycling in a eutrophic shield lake and its effects on whole lake metabolism Limnol. Oceanogr 23: 337-348 -------and C. D Taylor. 1980 Methane cycling in aquatic environments. Adv. Aquat. Microbial. 2: 77-150. -------, A. F. Furrtani, R J. Flett, and R D. Hamilton. 1976. Factors controlling rates of methane oxidation in shield lakes: the role of nitrogen fixation and oxyge n concentration. Limnol. Ocea nogr. 21: 357-364. Rust, F 1981. Ruminant methane <13c;12c) values: relation to atmospheric methane Science 211: 1044-1046 Sackett, W. M. 1978. Carbon and hydrogen isotope effects during the thermocatalytic production of hydrocarbons in laboratory simulation experiments. Geochim. Cosmochim. Acta 42: 571-580. -------. 1983 Stable carbon isotope studies on organic matter in the marine environment. Handbook of environ. isotope geochem in press. -------, G Brooks, M Conkright, L. Doyle and L. Yarbro 1985. Stable

PAGE 113

isotope compositions of organic carbon in Tampa Bay, Florida, U S.A ; implications for evaluating oil contaminations. Appl. Geochem. in press. 103 Sansone, F. J., and C S Martens. 1978. Methane oxidation in Cape Lookout Bight, North Carolina. Limnol. Oceanogr 23: 349-355 -------, and -------. 1981. Methane production from acetate and associated methane fluxes from anoxic coastal sediments. Science 211 : 707-709. -------and -------. 1982. Volatilie fatty acid cycling in organic-rich marine sediments. Geochim Cosmochim. Acta 46: 1575-1589. Schiegl, W. E., and J. C. Vogel. 1970. Deurterium content of organic matter. Earth Planet. Sci. Lett. 7: 307-313. Shoell, M. 1980. The hydrogen and carbon isotopic composition of methane from natural gases of various origins. Geochim. Cosmochim. Acta 44: 649-661. -------. 1982 Stable isotopic analyses of interstitial gases. in Quaternary sediments from the Gul f of California, p 815817. I n J. R Curray, D G Moore, et al., Init. Repts. DSDP 64. u.s Govt Printing Office. -------. 1983. Genetic characterization of natural gases. Am. Assoc. Pet. Geol. Bull. 67: 2225-2238 -------, and C E Redding. 1978. Hydrogen isotopic compositions of selected crude oils and their fractions, p. 384-385. In R. E Zartmann (Ed.), Short papers of the Fourth International Conference, Geochronology, Cosmochronology, Isotope Geology, U. S. Geol. Survey Open File Report 78-701. U .S. Govt. Printing Office. -------, E Faber, and M. L Coleman 1983. Carbon and hydrogen isotopic compositions of the NBS 22 and NBS 21 stable isotope reference materials: an inter-laboratory comparison. Org. Geochem. 5 : 3-6 Schonheit, P., J. A. Kristjansson, and R. K. Thauer. 1982. Kinetic mechanism for the ability of sulfate reducers to outcompete methanogens for acetate. Arch. Microbiol. 132: 285-288. Senum, G I., and J. S. Gaffney 1985. A reexamination of the tropospheric methane cycle: geophysical implications, p. 61-69. In E. T. Sundquist and W. S Broecker (Eds.), The carbon cycle and atmospheric co2 : natural var_iations present, Geophysical Monograph 32. Amer1can Geophys1cal Un1on. Shackleton, N.J. 1977. The oxygen isotope stratigraphic record of the late Pleistocene. Phil. Trans. R Soc. Lond. B 280: 169-182.

PAGE 114

Sheppard, J. C., H. Westberg, J F. Hopper, K. Ganeson, and P. Zimmerman. 1982. Inventory of global methane sources and their production rates. J Geophys. Res 87: 1305 1312. Sheu, D. D., A. Shakur, D. A. Wiesenburg, J. M. Brooks, and H. R 104 Krouse. 1984. Isotopic compositions of sulfur and oxygen in dissolved sulfate from the Orca Basin. Geochim Cosmochim. Acta in press. Shipley, T. H., M H. Houston, R. T. Buffler, J. Shaub K. J. McMillen J W. Ladd, and J L. Worzel 1979 Seismic evidence for widespread possible gas hydrate horizons on continental slopes and rises. Am. Assoc Pet. Geol Bull. 63: 2204-2213. Sha kes, R. F., P. K Trabant, B. J. Presley, and D F. Reid 1977 Anoxic, hypersaline basin in the northern Gulf of Mexico Science 196: 1443-1446. Smith, B. N and S. Epstein. 1970 Biogeochemistry of the stable isotopes of hydrogen and carbon in salt marsh biota. Plant Physiol. 46: 738-742 -------, and-------. 1971. Two categories of 13c;12c ratios for higher plants. Plant Physiol. 47: 380-384. Smith, J. W., D. Rigby, P. W Schmidt, and D A Clark. 1983. D/H ratios of coals and the paleolatitude of their deposition. Nature 302 : 322-323 Smith, P. H., and R A Mah. 1966 Kinetics of acetate metabolism during sludge digestion. Appl Microbial. 14: 368-371. Sowers, K. R., S. F. B aron, and J. G Ferry. 1984. Methanosarcina acetivorans sp. nov., an acetotrophic methane-produ cing bacterium isolated from marine sediments. Appl Environ. Microbial. 47: 971-978 Stahl, w. 1974 Carbon isotope fractionations in natural gases. Nature 251: 134 135. Stevens, C. M., and F. E Rust. 1982. The carbon isotopic composition of atmospheric methane. J. Geophys Res 87: 4879-4882. Stuermer, D. H., K. E. Peters, and I. R. Kaplan 1978. Source indicators of humic substances and proto-kerogen. Stable isotope ratios, elemental compositions and electron resonance spectra. Geochim. Cosmochim. Acta 42: 989 -997. Stumm, w., and J. J. Morgan. 1981. Aquatic Chemistry, 2nd Ed. Wiley. Takai, Y. 1970. The mechanism of methane fermentation in flooded paddy s oil. Soil Sci. Plant Nutr. 16: 238-244. Tompkins, R. E and L E. Shephard. 1979 Orca Basin: depositional

PAGE 115

105 processes, geotechnical properties and clay mineral ogy o f Holocene sediments within an anoxic hypersaline basin, northwest Gulf of Mexico Mar. Geol. 33: 221-238. Trofimuk, A. A., N v. Cherskiy, and V P. Tsarev. 1973 Accumulation of natural gases.in zones of hydrate-formation in the hydrosphere. Doklady Akad. Nauk. SSSR (English translation) 212 : 931-934. Wang, W. C., Y L. Yung, A. A Lacis, T. Mo, and J E. Hansen 1976. Greenhouse effects due to man-made perturbations of trace gases. Science 194: 685 690 Welhan, J. A 1981. Carbon and hydrogen gases in hydrothermal systems: the search for a mantle source. Ph.D. thesis, Univ Calif., San Diego. 194p -------, and H. Craig. 1979. Methane and hydrogen in East Pacific Rise hydrothermal fluids. Geophys. Res Lett. 6: 829-831. Westrich, J T., and R. A Berner. 198 4 The role of sedimentary organic matter in bacterial sulfate reduction: the G model tested. Limnol. Oceanogr 29: 236 -249. Whelan J. K., R. Oremland, M Tarafa, R. Smith, R Howarth and C. Lee. 1 985 Evi dence for sulfate reducing and meth a n e producing microorganisms in sediments from Sites 618, 619 and 622, in press. In A. H. Bouma, J M. Coleman, et al. !nit. Repts. DSDP 96. U S -.-Govt. Printing Office. Whiticar, M. J., E Faber, and M. Schoen. 1 985 Biogenic methane formation in marine and freshwater environments: C02 reduction v s acetate fermentation -isotopic evidence. Geochim Cosmochim Acta in press. Wiesenburg D A. 1 980. Geoch emistry of d issolved gases in the hypersaline Orca Basin. Ph. D thesi s Texas A&M Univ., College Station. 265p. Winfrey, M R., and J G Zeikus. 1977 Effect of sulfate on carbon and electron flow during microbial methanog enesis in freshwater sediments. Appl. Environ. Microbial. 33: 275-281. -------, and -------. 1 979 Anaerobic metabolism of immediate methane precursors in Lake Mendota Appl Environ Microbial. 37: 244-253 Wolfe, R S 1979. Microbial biochemistry of methane a study in contrasts. Part 1: methanogenesis, p. 267-300. In J. R. Quayle (Ed.) International review of biochemistry, vol. 21, microbial biochemistry. Univ. Park Press. Wolin, M. J 1979. Hydrogen transfer in microbial communities p. 323-356 I n A T Bull and J. H Slater (Eds .), Microbial interactions and communities Academic

PAGE 116

-------1982. The rumen fermentation: a model for microbial interactions in anaerobic ecosystems, p. In M. Alexander (Ed .), Advances in microbial ecology, vol. 3. Plenum. -------, and T. L. Miller. 1982. Interspec ies hydrogen transfer: 1 5 years later. Am. Soc Microbial. News 561565 Woltemate, I., M. J. Whiticar, a n d M. Schoell. Carbon and hydrogen isotopi c composition of bacterial methane in a shallow freshwater lake Limnol. Oceanogr 29: 985-992 106 Yancey, P H M. E. C lark, S C Hand, R D Bowlus, and G M Somero 1982 Living with water stress: evolu tion of osmolyte systems. Science 217: Yapp, C. J., and S. Epstein. 1982. Climatic significance of the hydrogen isotope ratios in tree cellulose. Nature 297: 636-639 Yeh, H W., and S. Epstein. 1981. Hydrogen and carbon isotopes of petroleum and related organic matter. Geochim. Cosmochim Acta 753-762 Zehnder A. J. B and T. D. Brock 1 979 Methane formation and methane oxidation b y methanogenic bacteria. J. Bacterial. 1 3 7: -------, and -------. 1 980 Anaerobic methane oxidation: occurrence and ecology. Appl. Environ Microbial. 39 : -------B. A Huser, T Characterization of non-hydrogen-oxydizing 1 1 1 D. Brock and K Wuhrmann. 1 980 an acetate-decarboxylating, methane bacterium. Arch Microbial. -------, K. Ingvorsen, and T. Marti. 1982. Microbiology of methane bacteria, p. In D. E Hughes D. A. Stafford, 8. I. Wheatley W Baader G Lettinga, E J Nyns, W Verstraete, and R L. Wentworth (Eds.), Anaerobic digestion 1981. Elsevier. Zeikus, J G., and M. R Winfrey. 1 976 T emperature limitati on of methanogenesis in aquatic sediments. Appl. Environ. Microbial. 3 1 : 99-107. Ziegler, H., C. B. Osmond, W Stichler, and P. Trimborn. 1 976 Hydrogen isotope discrimination in higher plants: correlations with photosynthetic pathway and environment. Planta 1 28: 85-92.