Whelan&Fearrington - 1992 - Materia Organica

Organic Matter: Productivity, Accumulation, and Preservation in Recent and Ancient Sediments
EARTHSCAPE DATE: 9/99
Organic Matter: Productivity, Accumulation, and Preservation

in Recent and Ancient Sediments
Jean K. Whelan
and
John W. Farrington
Columbia University Press: New York
1992
Bibliographical data
Acknowledgements
1. Introduction
2. Reflections on the Career and Times of John M. Hunt
3. Modeling Petroleum Generation in Sedimentary Basins
4. Sources, Cycling, and Distribution of Water Column Particulate and

Sedimentary Organic Matter in Northern Newfoundland Fjords and Bays: A
Stable Isotope Study
5. Organic Matter Accumulation, Remineralization, and Burialin an Anoxic Coastal

Sediment
6. Organic Carbon Remineralization and Preservation in Sediments of Skan Bay,

Alaska
7. Preservation of Sargassum Under Anoxic Conditions: Molecular and Isotopic

Evidence
file:///F|/Usuarios/Juan%20carlos/articulos/articulos...anic%20matter/Organic%20Matter%20Productivity,%20.htm (1 de 3)17/01/2006 07:01:03 p.m.

8. Geochemical Features of Organic Matter in Sediment Cores from Lützow-Holm

Bay, Antarctica
9. Sedimentation and Preservation of Amino Compounds and Carbohydrates in

Marine Sediments
10. Hydrodynamic Controls of Anoxia in Shallow Lakes
11. Organic Carbon Accumulation and Preservation in Marine Sediments: How

Important Is Anoxia?
12. Early-Stage Incorporation of Sulfur Into Protokerogens and Possible Kerogen

Precursors
13. Bitumen Classification and Biomarker Correlation Studies Based on Organic

Extracts from Neogene Gulf of California Sediments
14. Resolution of Sediment Hydrocarbon Sources:Multiparameter Approaches
15. Biomarkers in Recent and Ancient Sediments:he Importance of the Diagenetic

Continuum
16. Natural Hydrous Pyrolysis: Petroleum Generation in Submarine Hydrothermal

Systems
17. Stable Carbon Isotope Changes During the Maturation of Organic Matter
18. Source and Biomarker Composition Characteristics of Chinese Nonmarine

Crude Oils
19. Volatile Organic Compounds Associated with Oil Seepage Along the Northern
Continental Slope of the Gulf of Mexico
20. Maturity and Facies-Controlled Composition of the Organic Matter of Selected

Oil Shales
21. Organic Matter Response to Change of Depositional Environment in

Kimmeridgian Shales, Dorset, U.K.
22. The Distribution and Generation of Hydrocarbons in Carbonate Source Rocks.


Organic Matter: Introduction
Organic Matter: Productivity, Accumulation, and Preservation in Recent and Ancient

Sediments, by Jean K. Whelan and John W. Farrington
1. Introduction
This volume includes papers presented at a symposium entitled Organic Matter: Productivity,
Accumulation, and Preservation in Recent and Ancient Sediments, which was given in honor of the
seventieth birthday of our friend and colleague John Meacham Hunt by the Organic Geochemical
Division of the Geochemical Society in November 1988. The introduction of John at the meeting by
his old friend Wally Dow is included here and expresses admirably our motivation for honoring John
in this way. Even the reader well versed in John's extensive career will probably be surprised (as we
were) at the number of organic geochemical "firsts" that have come from his work over the years.
The papers included in the symposium and this volume come from all over the world and reflect the
wide range of John's interests in both ancient and recent sediments. It is characteristic that John
himself is coauthor of a paper in the volume aimed at helping the beginner on estimating realistic
time-temperature characteristics of oil and gas generation in the absence of a computer.
We did not intend that the symposium or the volume that resulted from it to be comprehensive in
covering all subjects of organic geochemistry of highly productive modern ecosystems, sediments
of those ecosystems, organic rich ancient sediments, and petroleum geochemistry. Rather, our
hope was to have a selection of papers in those topical areas that exemplified John's range of
interests during his career to date.
The science of organic geochemistry and petroleum geochemistry has matured with amazing
rapidity during John Hunt's life and exemplifies the explosive and exciting growth of science and
technology in general during those years. One example is the growth and refinement in analytical
methodology to understand the composition of petroleum and the composition of organic matter in
transit in sediments from early deposition to ancient sediment oil and gas. The capability to analyze
relatively rapidly and quantitatively hundreds of compounds in oil, rock, or sediment samples is
taken for granted by students today but is a far cry from the infrared analyses of ground rocks and
bitumens or micro distillation and column chromatography to separate groups of compounds-
elegant techniques in John Hunt's early research.
Perhaps most important, the science of petroleum geochemistry and organic geochemistry has
progressed from a mainly descriptive and qualitative phase to a quantitative phase where theories
are proposed and tested with quantitative data, where numerical models are checked with extensive
field data.
The first section of the book, Recent Sediments and Depositional Environments, provides a series
of papers that combine compound specific and isotopic analyses coupled with an assessment of
physical, chemical, and biological factors in specific ecosystems. They examine controls on
deposition, remineralization, transformation, and eventual deeper burial for organic matter in the
water column and upper one meter or so of sediments, mostly at the sediment-water interface. The
file:///F|/Usuarios/Juan%20carlos/articulos/articulos.../organic%20matter/Organic%20Matter%20Introduction.htm (1 de 5)17/01/2006 07:00:52 p.m.

need to understand modern-day analogs of depositional environments in order to interpret quantity

and composition of organic matter in specific ancient sediment facies is a tenet of modern organic
geochemistry and is amply illustrated by the papers in this section.
Ostrum and Macko provide an example of the use of stable isotopes of carbon and nitrogen to
identify sources of organic matter deposited in sediments of coastal embayments. Martens,
Haddad, and Chanton report on the continuing studies of Professor Martens and collaborators and
students on the carbon cycle and organic geochemistry of Cape Lookout Bight, North Carolina. This
paper is an example of the move toward quantitative studies in organic geochemistry that has
occurred, primarily in the last fifteen to twenty years. Alperin, Reeburgh, and Devol report on similar
studies in Skan Bay, Alaska. Assessment of input and output budgets for surface sediments are
remarkably well constrained in these studies. The relative rates of remineralization of easily
remineralized organic matter contrasted to less easily remineralized organic matter control to a
large measure the type or quality and quantity of organic matter incorporated into ancient
sediments. This, in turn, exerts significant influence on which depositional facies in ancient
sediments will yield oil and gas.
Kennicutt, Macko, Harvey, and Bidigare report on a specific type of algae, Sargassum, and its
organic matter molecular and isotopic composition changes, or lack there of, when deposited in the
highly unique Recent sediment environment of the Orca Basin of the Gulf of Mexico Continental
Margin. In an interesting comparison, the paper of Matsumoto, Matusmoto, Sasaki, and Watanuki
add to our understanding of lipid class compounds in carefully collected cores of surface sediments
from two depositional areas of the coast of Antarctica. These data add to the rapidly growing
knowledge of the biogeochemistry of hydrocarbons, fatty acids, and other lipid compounds (e.g.,
sterols and hopanols in other studies) present in surface sediments of the world's oceans and lakes.
Whelan and Emeis focus on two other classes of compounds, amino compounds and
carbohydrates, and model the rates of disappearance of these compounds in sediments underlying
highly productive waters of the Peru Upwelling area. They revisit another oft-invoked theory in
organic geochemistry that non-biological factors such as kerogen formation by chemical reactions
and adsorption onto mineral surfaces have significant influence on the quality and quantity of
organic matter found in ancient sediments.
The final paper in this section, by Jewell, provides an instructive reminder of the importance of the
physical dynamics of water columns in controlling biogeochemical factors such as nutrient recycling
and oxygen concentration, which are central to the issue of productivity and remineralization.
This section can correctly be interpreted as providing examples of recent results of research on the
initial incorporation of organic matter in surface sediments on its way to becoming ancient sediment
organic matter with a potential to yield oil and gas. However, the research described has
importance in other arenas. The global role of highly productive ecosystems and their sediments in
remineralizing or sequestering and burying significant amountsof carbon in the contemporary and
paleo times has important implications for carbon cycle models related to climate fluctuations.
Furthermore, recycling of carbon is the engine which drives key aspects of nutrient recycling which,
in turn, has implications for sustaining valuable living resources in coastal waters and lakes.
The next section, Transition, begins where the first section ended in addressing the question of the
role in anoxia in the accumulation and preservation of organic carbon in marine sediments in a
provocative paper by Calvert and Pedersen. They conclude that there is very little evidence for the
preferential preservation of organic matter under anoxic conditions as compared to oxic depositional

conditions provided that the supply of fresh organic matter is similar. This conclusion is counter to
the dogma we first encountered when we entered the ranks of organic geochemistry researchers
too many years ago, but is supported by some compelling evidence and persuasive reasoning. We
expect that the debate will continue on this issue for several years.
Philp, Suzuki, and Galvez-Sinibaldi report on the incorporation of sulfur into protokerogens and
possible kerogen precursors. The papers cited in their paper, e.g., the work of Sinninghe Damste
and coworkers, among others as well as the current work reported by the present authors,
investigate the exciting new aspects of the organic geochemistry of sulfur compounds in petroleum
and sediments. This phenomena of renewed interest in sulfur compounds is a classic case of how
major aspects of organic geochemistry have advanced over the years. Without exhaustively
recounting the details of who did what, it suffices for our purposes here to summarize the events by
stating that the proof of the existence of heretofore not reported organic sulfur compounds in the
thousands in petroleum was made possible by a combination of elegant natural product synthesis
and elegant analytical chemistry approaches, primarily by two groups of researchers under the
leadership of Dr. Jan De Leeuw and Professor Peter Schenck at Delft University in the Netherlands-
primarily via the thesis of J. Sinninghe Damste; and Professor Pierre Albrecht at the University
Louis Pasteur in Strassbourg, France, and numerous collaborators with these groups. The key
factors: careful and elegant natural products chemistry, powerful high resolution analytical
chemistry, and a framework of reference of many years of study of sulfur in oils and ancient
sediments by Dr. Wilson Orr among others. The studies to date are but the "tip of the iceberg" in
terms of opening up studies of specific organic sulfur compounds in ancient sediments and
petroleum as exemplified by the paper of Philp, Suzuki, and Galvez-Sinibaldi.
A provocative paper by Comet, McEvoy, and Kennicutt begins with the premise that "attempts to
use biological markers as indicators of ancient environments as well as maturity indices appear to
have reached a crossroads at which several different philosophical paths may follow." The authors
provide a marvelous discussion of the various major approaches to using biomarkers in organic
geochemical studies and use data and interpretations from several ancient sediment samples to
support their use of a "facies" approach to interpreting biomarker records in the correlation of
organic rich strata or oil to source correlations. Kennicutt and Comet follow with a paper that is less
controversial but that provides interesting data and interpretations that exemplify the approaches of
several organic geochemical research groups to using biomarkers and stable isotopes to unravel
sources of organic matter in the environment-in their particular case to understand contributions of
seep oil and gases to hydrocarbons in contemporary and Recent sediments in the Gulf of Mexico.
Brassell takes up one "philosophical path" of Comet, McEvoy, and Kennicutt by illustrating and
discussing the need for careful studies of a diagenetic continuum to unravel and prove putative
precursor-product relationships. The two papers together-that of Brassell and that of Comet,
McEvoy, and Kennicutt-illustrate the intellectual excitement of organic geochemistry in which
hypothesis testing sets forth work yet to be done.
The discovery process of science in all its excitement is exemplified by the subject of B.R.T.
Simoneit's paper on Natural Hydrous Pyrolysis-Petroleum Generation in Submarine Hydrothermal
Systems. The presence of hydrothermal vent areas associated with the ridges at spreading axis/
centers of ocean basins has been one of the more exciting geosciences discoveries of the past two
decades. Given its importance to global geochemical cycles over geologic time scales these
discoveries would have been exciting enough without the associated fascinating occurrence of new,
here-to-fore unknown biologic species in abundances not found elsewhere in the deep ocean

benthos and epibenthos (see references in Simoneit's chapter). Simoneit reports on a third aspect
of the hydrothermal vents that has excited scientists-the occurrence of petroleum formed by the
natural hydrous pyrolysis as heat from the vents acts on immature organic matter in associated
sediments or organic matter in the water and particles drawn into the vent system.
The use of DSDP (Deep Sea Drilling Project) and now IPOD (International Program of Ocean
Drilling) or ODP (Ocean Drilling Program) samples in organic geochemistry has been central to
several organic geochemistry projects over the years. John Hunt was one of the main champions of
organic geochemistry involvement with these programs and thus it is appropriate that Conkright and
Sackett address stable carbon isotope changes during maturation of organic matter in DSDP
samples from the Cape Verde Rise as the last paper in the Transition section of this book.
The final section deals with Ancient Sediments-oil and gas. The thousands of references in the
literature attest to the growth of such studies. Only five papers are presented in this section, but
they illustrate several types of studies. Jiamo Fu and Guoying Sheng report on the use of
biomarkers in studying the source of Chinese nonmarine oils within the context of an interesting
overview about these oils. Macdonald, Kennicutt, Brooks, and Fay report on volatile organic
compounds originating in ancient sediments in the Gulf of Mexico. These compounds are close
enough to the sediment surface of the contemporary environment to be detected in the sediments
at these seep sites and provide clues to the nature of the oil and gas in underlying sediments.
Wehner, Hufnagel, Teschner, and Koester discuss the interrelationships between maturity of
organic matter and facies controlled composition of organic matter of selected oil shales. They use
select maturity parameters derived from biomarker measurements and microscopic palynology type
visual characterizations to compare and contrast oils and various depositional environment-source
rocks.
A detailed study of a 30-m thick section of the Kimmeridge clay outcropping on the south coast of
Dorset, U.K., by geochemical measurements and optical examination is the subject of a paper by
Huc, Lallier-Verges, Bertrand, Carpentier, and Hollander. The results of their study of the organic
matter in this ancient sediment section reinforces a point exemplified by the papers in the second
section of this book: the importance of studies of modern analogs of ancient sediments to unravel
ancient sediment records. Furthermore, the authors raise once again the issue of the role of oxygen
content of bottom waters and surface sediments in controlling organic matter content in ancient
sediments.
The book ends with a paper by Taguchi and Mori on petroleum in carbonate source rocks, a subject
which evokes much debate in the organic geochemistry and petroleum geochemistry community-
and one of several subjects actively researched by John Hunt in the past decade. Various
hypotheses are presented briefly and discussed, and contrasted with hypotheses on hydrocarbon
generation in clay rich sediments.
We hope that the reader will find material in this volume to be thought-provoking. John Hunt loves a
good discussion-he continually disturbs our complacency by poking at the edges of our science. It is
practically mandatory that his birthday volume contain some controversial material! We think it
does.
John has helped and encouraged colleagues all over the world through his courses and personal
correspondence. He has a facility for picking out the important issues in complex problems that
leaves even those of us who work closely with him in awe. He is currently working on a new edition

of his classic text Petroleum Geochemistry and Geology. John tells us that the reason the
enterprise is taking longer than he would like is that organic geochemistry has grown so far and so
fast since the first edition that a new book is required, not simply a revision.

Organic Matter: Chapter 2

2. Reflections on the Career and Times of John M. Hunt
We reflect upon the career and times of John Meacham Hunt, a career that has spanned four
decades and contributed many "firsts" to the emerging science of petroleum geochemistry.
John was born and raised in Cleveland, Ohio. He studied organic chemistry and petroleum refining
at Pennsylvania State University, and in 1947 Standard Oil of New Jersey (SONJ, now Exxon) gave
him a postdoctoral appointment to study motor fuel synthesis. His industry contact, Dick Meinert,
was transferred to Tulsa, Oklahoma, as assistant director of research for Carter Oil Company, a
subsidiary of SONJ, and asked John if he would like to go along. They wanted to start a chemical
group in the geology section.
John had a big decision to make: to become a bench chemist along with hundreds of other Ph.D.'s
at a big eastern laboratory (he had offers from Dupont, Monsanto, and SONJ) or to head west with
a chance to start at ground zero. Adventure won out and he arrived in Tulsa in January 1948.
His new boss, Parke Dickey, took him to his laboratory, a ramshackle building on the back of the
Carter property, and gave him one assistant, a free hand, and a challenge: develop new ways to
find oil. John took a night course in geology at Tulsa University and learned all he could from
geologists in the group: the "geo" was beginning to attach itself to the "chemist."
John's first research project was in the Uinta Basin of Utah. Management thought he might be able
to identify the hydrocarbons (mostly solid bitumens) that were so common there. Oil had not yet
been discovered in that region. In the summer of 1948, John went on a three-month expedition in
Utah to collect samples with Francis Stewart, a recent geology graduate from Oklahoma. It was the
first time John had been west of Oklahoma.
Francis had a big Buick and liked to drive fast. On the way to Vernal, Utah, he was low on fuel, so
he coasted down the mountain roads, recklessly taking the curves. When they got to Vernal it was
rodeo week, and everyone in town was drunk. As they passed a saloon, shots rang out and a
woman ran out screaming. Such sights were never seen by bench chemists from Dupont, and John
knew he had made the right choice. A geochemist was born. During that summer John and Francis
collected several hundred samples of waxes and both liquid and solid bitumens from abandoned
mines and outcrops.
Back in the laboratory, the bitumens and waxes collected were classified by physical properties,
such as refractive index and infrared absorption spectrometry. Infrared was a new technique, and
Dick Meinert bought John the second machine off the Baird assembly line. The differences among
the various bitumens were obvious. John also extracted shales close to the bitumens with carbon
disulfide and compared their infrared spectra with those of the bitumens; they matched perfectly!
This was the first proof that a particular source rock yielded a hydrocarbon, in this case mainly solid
file:///F|/Usuarios/Juan%20carlos/articulos/articulo.../organic%20matter/Organic%20Matter%20Chapter%202.htm (1 de 11)17/01/2006 06:49:48 p.m.

bitumen, found in the reservoir rock. This pioneering work was published in the AAPG Bulletin,
(Hunt, Stewart, and Dickey 1954).
During the study, John and his group found that the infrared spectrophotometer could be used to
analyze a wide range of solid compounds. When ground to a particle size smaller than the infrared
wave length of 2 to 16 microns, characteristic spectra of minerals and chemicals were obtained. His
paper in Analytical Chemistry (Hunt, Wisherd, and Bonham 1950) contained spectra of 64 minerals,
rocks, and inorganic chemicals and proved to be a classic. More than two thousand reprints were
requested because it had a wide range of applications. Incidentally, a coauthor, Larry Bonham, was
a summer student at the time and later became a manager at Chevron Research.
John had pretty well proven that Uinta Basin solid bitumens came from adjacent shales, but most
geologists did not believe similar arguments applied to crude oil. So John and George Jamieson
developed extraction methods to remove oil-like material from source rocks. John got a big fume
hood on the second floor of the new laboratory building that sucked the cool air out of the director's
office down the hall. Despite such setbacks, they developed and patented the extraction technique
using methanol, acetone, and benzene (MAB). It worked!
In the days before gas chromatography, biomarkers, and vitrinite reflectance, John's group used
refining techniques such as distillation and column chromatography to characterize their oils and
rock extracts. They were able to show that shales do contain crude oil and that the amount of
extractable oil in shales far exceeds the amount known in reservoirs. This indicated that the
migration process is very inefficient. The classic Hunt and Jamieson paper showing the first oil-
source correlation was given at the 1955 AAPG meeting and published the following year in the
AAPG bulletin (Hunt and Jamieson 1956). It caused quite a stir and prompted G. T. Phillipi from
Shell to present some of his work at the next International Geological Congress in Mexico City in
1956. His paper had only one reference-the Hunt and Jamieson paper.
In the early 1950s the Carter Oil Company was producing low gravity oil from shallow Tensleep
reservoirs in Wyoming. New seismic developments provided deep resolution, and the company's
geophysicists wanted to drill structures as deep as fifteen thousand feet. Management, however,
said heavy oil would not be economical to produce at that depth. John reasoned that heavy oil
would be made lighter by natural thermal cracking with depth and undertook a study of Wyoming
crude oils to prove his point.
Infrared spectra defined two oil types, one in Mesozoic reservoirs and another in Paleozoic
reservoirs. More important, API (American Petroleum Institute) gravities did seem to increase with
depth as John had predicted, making the oil more valuable so that deeper drilling was justified (Hunt
1953). By the time Hunt's paper was published, Carter had made several Tensleep light oil
discoveries below nine thousand feet. For the first time, management began to think that
geochemists could really help them find oil.
Those were the halcyon days of petroleum geochemistry-a time when insights and discoveries were
made without the techniques we take for granted today. The first commercial gas chromatograph
(GC) did not come along until the early 1960s and mass spectrometry (GC/MS) not until a decade
later. In the early 1960s, D. H. Desty of British Petroleum made a GC column a mile long and used
it to separate 200 compounds from a crude oil in two weeks. We have come a long way in twenty-
five years.

During the same period the question arose about how and when oil migrated out of a shale source
and into a sand reservoir. It was known that there were hydrocarbons in recent sediments, most of
which had been formed directly from organisms. Parke Dickey sent Al Kidwell to the Orinoco delta
in Venezuela to collect cores of sands sealed in shales. John was to do the organic geochemistry
on the cores. The shallow sands had excess hydrostatic pressures from which the direction of fluid
flow could be determined. An increase in hydrocarbons was found in closed sands but not in sands
open to the surface. This was the first demonstration of hydrocarbons beginning to accumulate in a
reservoir and was published in Habitat of Oil (Kidwell and Hunt 1958). The then new carbon-14
analysis was used to show that this early accumulation of oil was less than ten thousand years old.
The Orinoco data included only total C15 + hydrocarbons, classified with column chromatography.
Gas chromatography was not yet in use, so analysis was not possible for individual hydrocarbons. It
now seems likely that the hydrocarbons found by Kidwell and Hunt were biological in origin-not
thermally generated-and therefore did not indicate early generation of crude oil. Many geologists
and managers claimed, however, that the Orinoco data proved that oil was generated "soon after
burial" and that geochemical studies of ancient rocks were a waste of time. This theory, which was
published as AAPG Memoir 5, caused many geologists to conclude that oil was generated in
Recent sediments.
John recognized that there were few gasoline-range hydrocarbons in living organisms but a lot in
crude oil and that the lighter molecular weight range compounds had not been studied in source
rocks. John's group was able to isolate high yields of a whole range of C4 - C8 hydrocarbons from
ancient rocks by using tetralin and ethanol as a solvent. By contrast, recent sediments they
extracted contained only traces of such material. This established that gasoline, and hence crude
oil, was thermally generated only in ancient sediments (Hunt and Dunton 1962).
After all his work on solvent extractions, John wondered from what kind of organic matter crude oil
came. Along with Jim Forsman, he isolated kerogen from rocks with acids and identified two types:
one oily and one coaly. The two types follow different pathways on a carbon, hydrogen, and oxygen
triangular diagram. These are the Type II and Type III kerogens we know today. Forsman and
Hunt's AAPG paper in Habitat of Oil was the first to show there were two basic kerogen types, one
of which generated crude oil and the other only gas (Forsman and Hunt 1958).
In 1962 John visited the Soviet Union on a technical exchange visit with five other Americans,
including Nelson Stevens. John was particularly interested in what the Russians had to say about
surface geochemistry and the inorganic origin of petroleum, topics that had been popular in Russia
since the time of Mendeleeff. They visited many laboratories and oil fields, but the inorganic
chemists were never there. They were obviously being kept under wraps until the visiting organic
chemists left. The Russians had just resumed work on surface geochemistry, and Nelson Stevens'
paper was on the failure of surface prospecting as a direct hydrocarbon indicator.
The first International Meeting on Organic Geochemistry was held in Milan in 1962, and John's
group stopped there on the way to Russia to give their papers. The first Gordon Research
Conference on Organic Geochemistry was also held in 1962, and John and Hollis Hedberg were
featured speakers. John described a pyrolysis technique for distinguishing between oil and gas
source rocks, which he also presented at a geochemical conference in Budapest that same year. It
was published in the conference proceedings (Hunt 1962). John was planning to leave Exxon and
was not able to follow up on the development of his procedure. The French did, and the Rock-Eval
was introduced several years later.

By 1964 John had assembled the first substantive group in petroleum geochemistry, which had
grown in sixteen years from one assistant in a shack to thirty-six people in a modern research
laboratory. But adventure was calling again; John had always wanted to spend half his career in
industry and half in academia. When Ken Emery called him about the job of chairman of a new
chemistry and geology department at Woods Hole Oceanographic Institution (WHOI), he could not
refuse. The chemists did not want a geologist as chairman, and the geologists did not want a
chemist. John was one of the few "geo" chemists around at the time and was perfect for the job.
John enthusiastically embraced oceanography. He went on an Indian Ocean expedition in 1965 and
was chief scientist on a Red Sea cruise in 1966 and also on a major Black Sea expedition
supported by the National Science Foundation (NSF) in 1969. Important contributions on the origin
of hot brines and anoxic depositional environments resulted from these projects. The Red Sea
cruise was the first to collect the brightly colored cores of metallic sulfides from the hot holes.
During this period John also researched carbonate source rocks, the origin of light hydrocarbons,
and oil migration. He even wrote a paper for National Fisherman promoting the nonpolluting virtues
of offshore drilling for gas (Hunt 1971).
John retired as chairman in 1974 after building one of the best chemical oceanography/marine
geochemistry departments in the world at WHOI. In the decade that followed he wrote more than
fifty papers, including his classic textbook Petroleum Geochemistry and Geology (Hunt 1979). He
lost the race to publish the first petroleum geochemistry textbook to Tissot and Welte, but both
books are standard texts on petroleum geochemistry around the world.
I met John for the first time at the sixth International Meeting on Organic Geochemistry in Paris in
1973. John was fifty-four and on his second honeymoon with his new wife, Phyllis. A story went
around about John being accosted by a lady of the evening at the Place Pigalle while Phyllis looked
on in horror. The next evening the Place Pigalle was full of geochemists. But as usual, John had
been there first.
In 1982 John was honored with the Geochemical Society's TREIBS Medal for outstanding career
achievements in organic geochemistry. His career has spanned four decades, probably longer than
any other organic geochemist active today. He has shared his knowledge with us in one hundred
two publications and countless presentations and currently teaches petroleum geochemistry
courses all over the world.
And what about the future? John's career is far from over. As he begins his fifth decade we can
expect more training courses, a new book on petroleum geochemistry, and continued research on
his current interests: pressure seals, light hydrocarbons, and kinetics of petroleum generation.
There are still many important questions remaining to be answered, and we have no doubt that
John will contribute his share of the solutions- for example, his paper in this volume.
John has an emeritus position at W.H.O.I. and works part time on projects along with his colleague,
Jean K. Whelan, who is continuing the program of the organic geochemistry of ancient sediments.
Publications of Dr. John M. Hunt

1949 Hunt, J. M. and T. W. Dennis. Application of certain Instrumental Methods in

Production Research. World Oil, 129 (3):152-158
1950 Hunt, J. M. Infrared spectra of clay minerals. In Infrared Spectra of Reference Clay
Minerals. API Project 49, Clay Mineral Standards Prelim. Report no. 8, sect. 3, pp. 105-122
Hunt, J. M., M. P. Wisherd, and L. G. Bonham. Infrared absorption spectra of minerals and
other inorganic compounds. Anal. Chem. 22(12):1478-1497.
1953 Hunt, J. M. Composition of crude oil and its relation to stratigraphy in Wyoming. AAPG
Bull. 37:1837-1872.
Hunt, J. M. and D. S. Turner. Determination of mineral constituents of rocks by infrared

spectroscopy. Anal. Chem. 25(8):1169-1174.
1954 Hunt, J. M. Chemistry applied to exploration. Preprints 10th Southwest Regional ACS
Mtg., Fort Worth, Texas.
Hunt, J. M., F. Stewart, and P. A. Dickey. Origin of hydrocarbons of Uinta Basin, Utah. AAPG
Bull. 38:1671-1698.
1956 Hunt, J. M. and G. W. Jamieson. Oil and organic matter in source rocks of petroleum.
AAPG Bull. (March) 40:477-488: and Habitat of Oil (AAPG), pp. 735-746, 1958.
1957 Hunt, J. M. and J. P. Forsman. Analytical Techniques in the separation of organic

matter from rocks. Preprints ACS 131st National Mtg., Miami.
Hunt, J. M. and J. P. Forsman. Relation of Crude Oil Composition to Stratigraphy in the Wind
River Basin. Wyoming Geol. Assoc. Field Trip Guidebook, pp. 105-112.
1958 Hunt, J. M. and A. L. Kidwell, Oil migration in recent sediments. World Oil. 147:79-83.
Forsman, J. P. and J. M. Hunt. Insoluble organic matter (kerogen) in sedimentary rocks.

Geochim. Cosmochim. Acta 15:170-182.
Forsman, J. P. and J. M. Hunt. Insoluble organic matter in sedimentary rocks of marine

origin. Habitat of Oil, AAPG, pp. 747-778.
Kidwell, A. L. and J. M. Hunt. Migration of oil in Recent sedmients of Pedernales, Venezuela.

Habitat of Oil, AAPG, pp. 790-817.
1960 Hunt, J. M. Origin of petroleum. Encyclopedia of Science and Technology, pp. 56-57.
New York: McGraw-Hill.
1961 Hunt, J. M. Distribution of hydrocarbons in sedimentary rocks. Geochim. Cosmochim.

Acta 22:37-49.

1962 Hunt, J. M. Geochemical data on organic matter in sediments. Proc. Second Oil Sci.
Int. Oil Producing Conference, Budapest, Hungary.
Hunt, J. M. and M. L. Dunton. Distribution of low molecular weight hydrocarbons in Recent

and ancient sediments. AAPG Bull. 46:2246-2248.
1963 Hunt, J. M. Composition and origin of the Uinta Basin bitumens. Bulletin 54-Oil and
Gas Possibilities of Utah, Utah Geological and Mineralogical Survey.
Bell, K. G. and J. M. Hunt. Native bitumens associated with oil shales. In I. A. Breger, ed.
Organic Geochemistry. New York: Pergamon.
Hurley, P. M., J. M. Hunt, W. H. Pinson, and H. W. Fairbairn. K-Ar age values on the clay
fractions in dated shales. Geochim. Cosmochim. Acta 27:279-284.
1964 Degens, E. T., J. M. Hunt, J. H. Reuter, and W. E. Reed. Data on the distribution of
amino acids and oxygen isotopes in petroleum brine waters of various geologic ages.
Sedimentol. 3:199-225.
1966 Degens, E. T. and J. M. Hunt. Evolutionary trends in calcified tissues of Marine

Organisms. Proc. Second Int. Oceanographic Congr., Moscow, p. 94.
1967 Hunt, J. M. The origin of petroleum in carbonate rocks. In G. V. Chilingar et al., eds.,
Carbonate Rocks, pp. 225-251. New York: Elsevier.
Hunt, J. M. and E. T. Degens. Carbon isotope fractionations in living and fossil organic
matter, Studia Biophysical, 4:179-190.
Ross, D. A. and J. M. Hunt. Third brine pool in the Red Sea. Nature 213(5077):687-688.
Hunt, J. M., E. E. Hays, E. T. Degens, and D. A. Ross. Red Sea: detailed survey of hot brine
areas, Science 156:514-516.
1968 Hunt, J. M. How oil and gas form and migrate. World Oil (October), pp. 140-150.
Degens, E. T. and J. M. Hunt. Data on the distribution of stable isotopes and amino acids in
Indian Ocean sediments. Technical Report, Woods Hole Oceanographic Institution, Ref. no.
68-4.
1969 Deuser, W. G. and J. M. Hunt. Stable isotope ratios of dissolved inorganic carbon in
the Atlantic. Deep Sea Res. 16:221-225.
1970 Hunt, J. M. On Atlantis II cruise #49 to the Black Sea. Oceanus 15(4):5-10.
Hunt, J. M. The significance of carbon isotope variations in marine sediments. In G. D.

Hobson and G. C. Spears, eds., Advances in Organic Geochemistry, pp. 27-35. New York:

Pergamon Press.
Hunt, J. M. and M. Blumer. Oil pollution in marine environment. Proc. Int. Symp.
Hydrogeochem. Biogeochem., Tokyo, pp. 96-97.
Hunt, J. M. and E. T. Degens. Composition and structure of organic matter in Black Sea
sediments. Proc. Int. Symp. Hydrogeochem. Biogeochem., Tokyo, pp. 264-279.
Emery, K. O., J. M. Hunt, and E. E. Hays. Summary of hot brines and heavy metal deposits
in the Red Sea. In E. T. Degens and D. A. Ross, eds., Hot Brines and Recent Heavy Metal
Deposits in the Red Sea, pp. 557-571. New York: Springer Verlag.
1971 Hunt, J. M. Drill gas offshore to meet energy needs safely. National Fisherman
(December), p. 9-C.
Hunt, J. M. Mass attack on your mind. PHP (June), no. 9, pp. 21-23.
1972 Hunt, J. M. Distribution of carbon in crust of Earth. AAPG Bull. 56:2273-2277.
Dickey, P. A. and J. M. Hunt. Geochemical and hydrogeologic methods of prospecting for

stratigraphic traps. AAPG Memoir 16, Stratigraphic Oil and Gas Fields, pp. 136-167.
1973 Hunt, J. M. An examination of petroleum migration processes. Proc. 7th Int. Sci.
Geochem. Conf. on Hydrocarbon Production, Budapest, Hungary, pp. 219-223.
Hunt, J. M. Organic geochemical studies: introduction and summary. In B. C. Heezen et al.,

eds. Initial Reports. DSDP 20:905.
1974 Hunt, J. M. How deep can we find economic oil and gas accumulations? Proc. 1974
Deep Drilling and Production Symp. in Amarillo, Texas, pp. 103-110. Dallas: Society of
Petroleum Engineers of AIME.
Hunt, J. M. Hydrocarbon and kerogen studies, Initial Reports: DSDP 22:673-675.

Washington, D.C.: U.S. Govt. Printing Office.
Hunt, J. M. Hydrocarbon and kerogen studies on Red Sea and Gulf of Aden cores, Initial
Reports: DSDP 24:1165-1167. Washington, D.C.: U.S. Govt. Printing Office.
Hunt, J. M. Hydrocarbon geochemistry of the Black Sea. In: E. T. Degens and D. A. Ross,
eds., The Black Sea-Geology, Chemistry and Biology. AAPG Memoir 20. Tulsa: AAPG.
Hunt, J. M. Organic Geochemistry of the Marine Environment, Proc. Sixth Int. Mtg. Org.
Geochem., pp. 593-604. Paris: Technip.
Hunt, J. M. The petroleum problem. Oceanus 18:4-5.

Emery, K. O. and J. M. Hunt. Summary of Black Sea investigations. In E. T. Degens and D.

A. Ross, eds. The Black Sea-Geology, Chemistry and Biology. AAPG Memoir 20, Tulsa:
AAPG.
1975 Hunt, J. M. Hydrocarbon Studies. In D. E. Karig and J. C. Ingle, eds. Initial Reports.
DSDP. 31:901-903. Washington, D.C.: U.S. Govt. Printing Office.
Hunt, J. M. Is there a geochemical depth limit for hydrocarbons? Petrol. Eng. (March), pp.
112-124.
Hunt, J. M. Origin of gasoline range alkanes in the deep sea. Nature 254(5499):411-413.
Hunt, J. M. Statement on petroleum in the marine environment. Minutes of the Hearing of

1/30/75 of the National Ocean Policy Study Committee of the Senate Commerce Committee,
U.S. Senate. Washington, D.C.: U.S. Govt. Printing Office.
Wilson, E. B., J. M. Hunt, et al. Petroleum in the marine environment. Washington, D.C.:
Report of the Natl. Acad. Sci.
1976 Hunt, J. M. C4-C7 alkane yields. Initial Reports. DSDP. 38:807-808. Washington, D.C.:
U.S. Govt. Printing Office.
Hunt, J. M. Origin of athabasca oil. AAPG Bull. 60:1112.
1977 Hunt, J. M. Distribution of carbon as hydrocarbons and asphaltic compounds in

sedimentary rocks. AAPG Bull. 61:100-104.
Hunt, J. M. Ratio of petroleum to water during primary migration in Western Canada Basin.
AAPG Bull. 61:434-435.
Hunt, J. M. and J. K. Whelan. Light hydrocarbons at Sites 367-370, Leg 41. Initial Reports.
DSDP. 41:859.
1978 Hunt, J. M. Characterization of bitumens and coals. AAPG Bull. 62:301-303.
Hunt, J. M. Light hydrocarbons in Holes 361 and 364, Leg 40, Initial Reports. DSDP, 40
(suppl.):649-650.
Hunt, J. M. Organic sediments. In R. Fairbridge, ed., Encyclopedia of Sedimentology, pp.

515-520.
Hunt, J. M. and J. K. Whelan. Dissolved gases in Black Sea sediments. Initial Reports.
DSDP. 42B:661-665.
Whelan, J. K. and J. M. Hunt. C1-C7 hydrocarbons in Holes 379A, 380/380A and 381. Initial
Reports. DSDP. 42B:673-677.

Hunt, J. M. and J. K. Whelan. Light hydrocarbons in sediments of Leg 44 holes. Initial

Reports. DSDP. 44:651-652.
1979 Hunt, J. M. Petroleum Geochemistry and Geology, San Francisco: W. H. Freeman.
Hunt, J. M. and J. K. Whelan. Volatile compounds in quaternary sediments. Org. Chem.

1:219-224.
Whelan, J. K. and J. M. Hunt. C2-C7 hydrocarbons from IPOD Hole 398D. Initial Reports.
DSDP. 47:561-563.
Whelan, J. K. and J. M. Hunt. Sediment C1-C7 hydrocarbons from IPOD Leg 48-Bay of
Biscay. Initial Reports. DSDP. 48:943-945.
Whelan, J. K. and J. M. Hunt. Sediment C1-C7 hydrocarbons from Deep Sea Drilling Project
Sites 415-416 (Moroccan Basin). Initial Reports. DSDP. 50:623-624.
1980 Hunt, J. M., A. Y. Huc, and J. K. Whelan. Generation of light hydrocarbons in

sedimentary rocks. Nature 288(5792):688-690.
Hunt, J. M., A. Y. Huc, and J. K. Whelan. Genesis of petroleum hydrocarbons in marine

sediments. Science 209:403-404.
Hunt, J. M., R. J. Miller, and J. K. Whelan. Formation of C4-C7 hydrocarbons from bacterial
degradation of naturally occurring terpenoids. Nature 288(5791):577-578.
Huc, A. Y. and J. M. Hunt. Generation and migration of hydrocarbons in offshore South

Texas Gulf Coast sediments. Geochim. Cosmochim. Acta 44:1081-1089.
Whelan, J. K. and J. M. Hunt. C1-C7 volatile organic compounds in sediments from Deep
Sea Drilling Project Legs 56 and 57. Japan Trench. Initial Reports. DSDP. 51, (part 2):1349-
1355.
Whelan, J. K., J. M. Hunt, and J. Berman. Volatile C1-C7 organic compounds in surface
sediments from Walvis Bay. Geochim. Cosmochim. Acta 44:1767-1785.
Whelan, J. K., J. M. Hunt, and A. Y. Huc. Applications of thermal distillation-pyrolysis to

petroleum source rock studies and marine pollution. J. Anal. Appl. Pyrolysis 2:79-96.
1981 Hunt, J. M. The future of deep conventional gas. In R. F. Meyer, ed., Proc. UNITAR
Conference on Long-Term Resources, pp. 343-355. London: Pitman.
Hunt, J. M. The origin of petroleum. Oceanus 24(2):53-57.
Hunt, J. M. 1981. Petroleum geochemistry of the Atlantic margin. AAPG. Proc. Atlantic
Margin Energy Conference, pp. 7-1-7-32.

Hunt, J. M. Source rock characterization by thermal distillation and pyrolysis. In G. Atkinson

and J. Zuckerman, eds., Origin and Chemistry of Petroleum, pp. 57-65. New York:
Pergamon Press.
Huc, A. Y., J. M. Hunt, and J. K. Whelan. The organic matter of a Gulf Coast well studied by
a thermal analysis-gas chromatography technique. J. Geochem. Explor. 15:671-681.
Whelan, J. K. and J. M. Hunt. C1-C8 hydrocarbons in IPOD Leg 63 sediments outer

California and Baja California borderlands. Initial Reports. IPOD. pp. 755-783.
1982 Whelan, J. K., and J. M. Hunt. C1-C8 hydrocarbons in IPOD Leg 64 sediments, Gulf of
California, Initial Reports. IPOD. 64:763-779.
Whelan, J. K., M. E. Tarafa, and J. M. Hunt. Volatile C1-C8 organic compounds in

macroalgae. Nature 299:50-52.
1983 Hunt, J. M. Offshore oil and gas-past, present and future. Oceanus 26(3):2-6.
Peters, K. E., J. K. Whelan, J. M. Hunt, and M. E. Tarafa. Programmed pyrolysis of organic

matter from thermally altered Cretaceous black shales, AAPG Bull. 67:2137-2146.
Tarafa, M. E., J. M. Hunt, and I. Ericsson. Effect of hydrocarbon volatility and adsorption on
source-rock pyrolysis. J. Geochem. Explor. 18:75-85.
Whelan, J. K., M. A. Blanchette, and J. M. Hunt. Volatile C1-C7 organic compounds in an

anoxic sediment core from the Pettaquamscutt River (Rhode Island, USA). Org. Geochem. 5
(1):29-34.
Whelan, J. K. and J. M. Hunt. The organic matter in DSDP Site 504 and 505 sediments
studied by a thermal analysis-gas chromatography technique, Initial Reports. IPOD. Leg 69,
pp. 443-450.
Whelan, J. K. and J. M. Hunt. Volatile C1-C8 organic compounds in sediments from the Peru
Upwelling Region. Org. Geochem. 5(1):13-28.
1984 Hunt, J. M. Generation and migration of light hydrocarbons. Science 226:1265-1270.
Hunt, J. M. and A. P. McNichol. The Cretaceous Austin Chalk of South Texas-a petroleum
source rock. In J. G. Palacas, ed. AAPG Studies in Geology. No. 18. Petroleum
Geochemistry and Source Rock Potential of Carbonate Rocks, pp. 117-125.
Whelan, J. K., J. M. Hunt, J. Jasper, and A. Huc. Migration of C1-C8 hydrocarbons in marine
sediments. Org. Geochem. 6:683-694.
1986 Hunt, J. M. Petroleum, pp. 362-365, New York: McGraw-Hill Yearbook of Science and
Technology.
file:///F|/Usuarios/Juan%20carlos/articulos/articul...organic%20matter/Organic%20Matter%20Chapter%202.htm (10 de 11)17/01/2006 06:49:48 p.m.

1987 Hunt, J. M. Primary and secondary migration of oil. In R. F. Meyer, ed. AAPG Studies
in Geology No. 25. Exploration for Heavy Crude Oil and Natural Bitumen, pp. 345-349.
Hunt, J. M. Geochemistry of petroleum. Short course handout WHOI, Woods Hole, Mass.
1988 Farrington, J. W., A. C. Davis, M. E. Tarafa, M. A. McCaffrey, J. K. Whelan and J. M.

Hunt. Bitumen molecular maturity parameters in the Ikpikpuk Well, Alaskan North Slope.
Org. Geochem. 13 (1-3):303-310.
1989 Jasper, J. P., J. K. Whelan, and J. M. Hunt. Migration of C1 to C8 volatile organic

compounds in sediments from the Deep Sea Drilling Project, Leg 75, Hole 530A, Walvis
Ridge. Initial Reports. DSDP. 75:1001-1008.
1990 Hunt, J. M. The generation and migration of petroleum from abnormally pressured fluid
compartments. AAPG Bull. 74:1-12.
1991 The generation and migration of petroleum from abnormally pressured fluid
compartments: reply. AAPG Bull. 75:328-330, 336-338.
Hunt, J. M., M. K. Lewan, and R. Hennet. Modeling oil generation with time-temperature
index graphs based on the Arrhenius equation. AAPG Bull. 75(4):795-807.
Hunt, J. M. Generation of gas and oil from coal and other terrestrial organic matter. In W. G.
Dow and P. K. Mukhopadhyay, eds., Coal and Terrestrial Organic Matter as a Source Rock
for Petroleum. Org. Geochem. 17:673-680.


3. Modeling Petroleum Generation in Sedimentary Basins
This paper presents a graphic method for defining the oil and gas windows in sedimentary basins
based on the Arrhenius equation. Time-temperature index (TTI) graphs for kerogen Types I, II, and
III plus a graph for the conversion of oil to gas were constructed and tested in various basin
configurations. Application of these graphs showed them to match the oil window indicated by other
maturation parameters more closely than the Lopatin method with fast-reacting kerogens. However,
both methods defined approximately the same oil window in the examples tested for medium- and
slow-reacting Type II kerogens and for a medium-reacting Type I and Type III kerogen at moderate
heating rates.
A large spread was found between the kinetic parameters of different Type I and Type II kerogens,
so it is not valid to use a fixed set of kinetic parameters within each of these groups. Type II
kerogens ranged in activation energies (E) from 144 to 218 kJ/mol. The kerogens with the highest
sulfur contents had the lowest E values and were the fastest in generating oil, while the kerogens
with the lowest sulfur contents had the highest E values and were the slowest in generating oil.
Type I kerogens ranged from 194 to 269 kJ/mol owing to other compositional differences.
A comparison of oil windows in a cratonic basin showed that the slowest reacting Type II kerogen
required 2,800 m more of burial to generate oil than the fastest reacting Type II kerogen did. The
slowest Type I required 2,000 m more of burial than the fastest Type I did to generate oil.
These graphs are simple and accurate enough to have wide application in making preliminary
evaluations of the depth of the oil and gas windows in exploration areas.
This work was supported by the Department of Energy Grant DE-FGO2-86ER13466. This is Woods
Hole Oceanographic Institution Contribution No. 7346
Time-Temperature History of Oil Formation
The carbonization of coal with increasing time and temperature is a well-recognized phenomenon
that was first described by Hilt in 1873. Hilt's law stated that the fixed carbon of coal increased with
increasing depth and temperature. Later, White (1915) showed a relationship between the
occurrence of oil and gas and the rank of coals in the eastern United States. Oil fields were found
where the fixed carbon content of coals was less than 60%, gas fields, in the 60 to 65% range;
above 70% fixed carbon, there were no oil or gas accumulations. These early studies correlated the
origin of oil with the time-temperature formation of coals of different rank. The first direct attempt at
modeling petroleum generation was by Habicht (1964), who constructed a source rock burial history
curve and used Arrhenius kinetics to determine the time and depth of oil generation in the Gifhorn
Trough of northwest Germany. Habicht used an activation energy of 58 kcal/mole and a frequency
factor (A) of 5 x 1013 sec-1 for the kinetics of the Jurassic source rock. His burial profile indicated

that generation started about 90 Ma (million years ago) at 2,610 m, and after further burial and
uplift, the generation of oil was completed about 10 Ma.
Later Philippi (1965) documented the increase in the yield of hydrocarbons from source rocks in the
Los Angeles and Ventura Basins of California with increased time and temperature. Still later a
series of source rock burial history curves was used to describe the origin of hydrocarbons in the
North Sahara Hassi Messaoud area (Poulet and Roucache 1969). These early studies indicated
that temperature alone cannot explain the different examples of oil generation. This gave kinetic
theory involving both temperature and time an important role in evaluating the oil potential of
sedimentary basins. Exposure time that a source rock experiences at varying temperatures with
burial must be taken into account in a subsiding basin. Rapid burial with high geothermal gradients
will not result in the same level of maturation as slow burial with low geothermal gradients.
Coalification and the Formation of Petroleum
Teichmuller's classic study of the Wealden Basin (1958) established the first relationship between
the reflectance of the vitrinite maceral of coal and the occurrence of oil. Since then vitrinite
reflectance has become one of the most widely used maturation parameters for empirically defining
the oil and gas windows. In the thermal alteration of vitrinite, the reflectance increases exponentially
with a linear increase in temperature (Ting 1975). This is why vitrinite reflectance generally plots as
a straight line on a semilog plot (Dow 1977). The increase in reflectance during coalification
involves aromatization and condensation reactions resulting in a loss of methane and a
corresponding decrease in the hydrogen-to-carbon ratio.
This empirical relationship between vitrinite reflectance and petroleum formation was utilized by
Lopatin (1971) to develop a simple method of using both time and temperature to calculate the
thermal maturity of the organic matter in sediments. Lopatin took a burial history diagram and
superimposed on it a grid of isotherms at 10°C intervals as shown in figure 3.1. He then determined
the thermal exposure of the source rock in each time-temperature interval and summed these up to
give the total exposure since the time the source rock was originally deposited. This was called the
time-temperature index (TTI) of maturity (for more details, see Waples 1985). The thermal exposure
was calculated by multiplying each time interval by a temperature factor based on the old chemical
rule that reaction rates double for each 10°C rise in temperature (Bergius 1913). This is why he
used isotherms at 10°C intervals.
Lopatin's technique worked surprisingly well despite the facts that reaction rates in oil generation
increase by more than a factor of two with rising temperatures and they vary considerably between
different kerogens exposed to different thermal conditions. Although Lopatin's initial calibration for
the oil window was based on coals and Type III kerogens, Waples' (1985:135-36) revision enabled
the method to be extended in a very general way to other kerogen types. It became a widely used
technique for defining the oil window in sedimentary basins.
Most techniques begin, however, to show weaknesses as their applications are increased. The
Lopatin method failed to define the oil window accurately in basins with heating rates much higher
or lower than 1°C/million years (Wood 1988). Also, it underestimated the thermal maturity of the oil
window for fast-reacting kerogens such as those in the Monterey Shale of offshore California. In
general, it did not account for the reactivity of widely different kerogens, which is the real strength of
using Arrhenius equation kinetics (Wood 1988). In 1976 Lopatin introduced a nomograph for his

method to minimize the mathematical calculations (see Hunt 1979:289 for details). It was a useful
modification but it still had the problems discussed above.
This lack of correlation between TTI and oil generation in many cases is caused by the differences
in time-temperature relationships between vitrinite reflectance and oil generation. Lewan (1985)
found that while a fixed time-temperature relationship exists for vitrinite reflectance, a significantly
large variation in the time-temperature relationship exists for oil generation from Type II kerogens.
Thus, while the Lopatin/Waples approach estimates the reflectance of minor amounts of vitrinite in a
source rock, it does not estimate oil generation from the predominating oil-prone kerogen
accompanying the vitrinite.
Determining Kinetic Parameters for Oil Generation
Engler (1913) first described the generation of oil from kerogen as a two-step process involving
bitumen as an intermediate. Subsequent research has shown this to be correct for many, but not all,
kerogens (Miknis et al. 1987).
kerogen bitumen oil + gas + residue
In this equation bitumen is defined as the solvent extractable fraction of the organic matter that
remains in the source rock after heating whereas kerogen is the nonextractable fraction. The
residue is the fraction of nonextractable kerogen remaining after maximum heating in the laboratory
or under natural conditions. Several chemical studies cited by Miknis et al. (1987) have shown that
for typical lacustrine hydrogen-rich Type I oil shales, the bitumen decomposition is the rate-
controlling step for oil production. The formation of a bitumen intermediate appears to be typical of
many Type I and Type II kerogens (Shen, Fan, and Castleton 1984).
The exponential temperature dependance of the rate of kerogen decomposition can be expressed
in theoretical terms by the Arrhenius equation which can be written as,
[k] = exp (-[E/RT]) (3.1)

where k is the reaction rate constant (1/my); A is the preexponential, or frequency, factor (1/my); E
is the activation energy (kJ/mol); R is the ideal gas constant; and T is temperature in degrees Kelvin
(°C + 273). The kinetic parameters E and A can be obtained experimentally by heating the source
rock at various temperatures and measuring the yield of hydrocarbons. The two types of
experimental techniques most widely used are an open, nonisothermal programmed temperature
dry pyrolysis such as the Rock-Eval system and a closed, isothermal wet pyrolysis system such as
hydrous pyrolysis. There are advantages and disadvantages to each technique.
The open system is rapid, simple, and uses relatively small sample sizes (<250 mg). Owing to the
ease of operation it is possible to perform several programmed temperature runs at different heating
rates in a relatively short time. The activation energy is determined by a computerized curve-fitting
process that uses a series of E and A values. Its supporters argue that a distribution of activation
energies is obtained that is more realistic than a single value because kerogen breakdown
represents a whole series of reactions (Burnham et al. 1987). Its detractors argue that it does not
monitor exclusively the rate-controlling step of bitumen to oil, but rather, it monitors all the reactions
involving the formation of bitumen, oil, and gases (Lewan, in press). Some investigators have
collected the oil to avoid this problem (Yang and Sohn 1984). However, a dry open system does not

yield a normal crude oil in that it has a disproportionately high percentage of polar compounds and
unsaturates (Lewan, Winters, and McDonald 1979). This appears to be due to the absence of
water. There is growing evidence that the rate controlling step, bitumen to oil, involves water in the
reaction (Lewan 1992). Chemically, water dissolved in bitumen terminates free radical sites by
donating its hydrogen atoms during bitumen decomposition to oil. This enhances generation of
saturates and reduces formation of pyrobitumens (Lewan 1992).
Hydrous pyrolysis is carried out isothermally in a closed system in the presence of excess water. It
yields generated hydrocarbon mixtures very similar to crude oil. The experimental conditions during
hydrous pyrolysis simulate rather closely the physical and chemical conditions encountered in
buried sediments. The presence of mineral, aqueous, gaseous, and organic (kerogen) phases
together in a closed system influences reaction kinetics and reaction pathways in a manner rather
similar to natural burial. Compared with open system pyrolysis, hydrous pyrolysis is believed to
provide more realistic A and E parameters for determining the position of hydrocarbon generating
windows in the history of sedimentary formations.
A disadvantage of hydrous pyrolysis is that it is time-consuming and requires specialized pressure

equipment. However, cross-correlations between open and closed pyrolysis analyses can be made
on key standard samples. Then open pyrolyses may be run to show relative differences within
sample groups. Open pyrolysis also could be used for some reactions that do not appear to involve
a bitumen intermediate such as gas and condensate formation.
Variations in Kinetic Parameters with Kerogen Type
Burnham, Braun, and Samoun (1987) have suggested that activation energies must be known to
within about 13 kJ/mol to predict oil generation to within 10°C in a basin, a constant heating rate of
3°C/my being assumed. This accuracy is required owing to the large extrapolation from the
laboratory to the geological environment. With this limitation it is obviously not possible to assume
that a single set of kinetic parameters could describe all processes of petroleum formation, as has
been attempted by some authors, for example, Quigley, Mackenzie, and Gray (1987). Natural
kerogens are very different in composition, and the range in their kinetic parameters is considerable
(see table 3.1).
In 1974, Tissot et al. defined three major types of kerogen in sedimentary rocks based on their
maturation track in a van Krevelen diagram. Such a diagram is a plot of atomic H/C (hydrogen/
carbon) versus O/C (oxygen/carbon) used originally for coals and coal macerals by the coal
scientist van Krevelen (1961). The Tissot group adopted the van Krevelen diagram for the
dispersed kerogen of sedimentary rocks. The Type I kerogen followed the maturation track of
lacustrine oil shales and boghead coals. The track of Type II was characteristic of most marine oil
source rocks while Type III followed the original coalification track of van Krevelen for the gas-
generating humic coals. These classifications resulted in a separate set of Arrhenius kinetics for
each of these three major types (Tissot and Espitalie 1975).
More recently Lewan (1985, 1989) has shown that time-temperature relationships of oil generation
also vary considerably for kerogens within the Type II class. Therefore, it is not possible to use the
same Arrhenius kinetic parameters for all Type II kerogens. This variability is attributed in part to the
amount of organic sulfur incorporated into the kerogen during its early diagenetic development. In
1985, Orr reported that high sulfur kerogens of the Monterey Formation in the Santa Maria Basin of

California generated low maturity oils at faster reaction rates (lower thermal exposures) than typical
Type II kerogens. He called these kerogens Type IIS and characterized them as having atomic S/C
ratios >0.04. Consequently, the Type II's as defined by Orr are equivalent to the Types IIA and IIB in
table 3.1. Early oil generation by these kerogen types is due to the fact that carbon-sulfur bonds
have lower bond energies than carbon-carbon bonds (Lovering and Laidler 1960). This means that
kerogens with more sulfur in their molecular structures will have lower activation energies. They will
decompose to oil at lower time-temperature increments.
This inverse relationship between organic sulfur content and activation energy (E) for Type II
kerogens is apparent in table 3.1. In this table, kerogens are designated as Type II A, B, C, and D,
depending on their sulfur content and reaction rates. In a previous publication (Hunt, Lewan, and
Hennet 1991) the reaction rates of these four kerogen types were defined as IIA fast, IIB medium
fast, IIC medium, and IID slow. E and A values for the Arrhenius equation plus organic sulfur
contents are shown for kerogens in four source rocks representative of these four types. These are
the Miocene Monterey Formation of California, the Permian Retort Shale Member of the Phosphoria
Formation of Montana, the Cambrian Alum Shale of Sweden, and the Devonian-Woodford Shale of
Oklahoma. Type IIA has the highest organic sulfur content (11%) and the lowest E value. It
decomposes to oil starting at temperatures as low as 50°C, whereas Type IID has the lowest
organic sulfur content (5%) and the highest E value. It starts to decompose at temperatures around
110°C.
There are some Type II kerogens (marine and lacustrine) with organic sulfur contents as low as 2%
(Lewan 1986), but no published kinetic data are yet available. However, at low sulfur contents, other
variations in composition may be as important as sulfur. For example, the Type IB kerogen from the
Green River Formation in the Uinta Basin of Utah in table 3.1 has an E of 219 kJ/mole, an A of
8.87x 1026/my, and an organic sulfur content of 1.3 wt. % (Sweeney, Burnham, and Braun 1987).
The E and A values are almost identical to the Type IID Woodford Shale in table 3.1, even though
the sulfur content is much lower. This might suggest that sulfur contents below approximately 5
percent are no longer the predominant factor affecting the kinetics of oil generation.
Table 3.1 also contains Arrhenius kinetic parameters for three Type I kerogens and one Type III
kerogen, along with their organic sulfur contents if available. Type I kerogens usually have a higher
ratio of aliphatic to aromatic carbons, which results in a higher level of activation energy (E) values
than that of Type II kerogens. The Type I kerogens, which are all of the Green River Formation in
Colorado, Wyoming, and Utah, range in E from 194 to 269 kJ/mole. This range probably reflects
sharp differences in kerogen composition resulting from variations in depositional environments
characteristic of lacustrine and boghead deposits. The Green River lakes fluctuated extensively in
size and in salinity from fresh to hypersaline (Smith 1983). Nonmarine sections compared with
marine, in general, tend to show greater variations in kinetics within a single formation. Marine
sections show more variation between formations.
The source rock of the Type III kerogen in table 3.1 is the Paleocene Tent Island Formation of the
Beaufort-Mackenzie Basin in northern Canada. It is typical of the Type III source rocks throughout
the Beaufort-Mackenzie Basin of Canada (Issler and Snowdon 1990). Its kinetic parameters are
also similar to those of the Miocene coals and coaly shales of the Mahakam Delta, Indonesia
(Tissot, Pelet, and Ungerer 1987).
Kinetic parameters for the conversion of oil to gas (Quigley, Mackenzie, and Gray 1987) are listed in

the caption for figure 3.11. These can be used to define a gas window following the oil window for a
typical crude oil.
The importance of the range in kinetic parameters in table 3.1 is better illustrated by comparing two
examples of subsiding basins with very different geothermal histories. The first example in figure
3.2a represents source rocks subsiding at a rate of 100 m/m.y. in a pull-apart basin with a
geothermal gradient of 45°C/km. Cumulative integration of the Arrhenius equation with the integral
of the first-order rate expression over the subsidence history of this model basin results in the oil
generation curves shown in figure 3.2a. These are for kerogen Types IIA, B, C, and D as
represented by the Monterey, Phosphoria, Alum, and Woodford Shales of table 3.1. Note that oil
generation is complete from kerogen Type IIA in about 1,500 m of burial at temperatures below 100°
C whereas Type IID requires 3,300 m of burial and temperatures above 150°C. The spread
between the oil windows of Types IIA and IID is about 1,800 m of burial equivalent to 81°C.
Figure 3.2b shows oil generation curves for Type I and Type III kerogens in the same basin. Note
that the entire group of curves has shifted to greater depths corresponding to higher temperatures
for oil generation than with the Type II kerogens. The Type III has a slower reactivity, requiring
higher temperatures than all of the Type IIs in figure 3.2a, and only the Type IA and IB samples
overlap the Type II group. Also, there is a significant difference in reactivity within the Type I group
corresponding to as much as 1,100 m (50°C) between the oil windows of the fastest and slowest
reacting kerogens.
Figures 3.2c and d represent source rocks subsiding at a rate of 100 m/my in a cratonic basin with a
geothermal gradient of 25°C/km. Owing to this lower gradient the spread in depth between oil
windows is greater than in the pull-apart basin. Kerogen Type IID in figure 3.2c requires 2,800 m
more of burial than Type IIA for oil generation.
In figure 3.2d there is a 2,000 m spread within the kerogen Type I group. The Type I group
worldwide contains many lacustrine oil shales, and figure 3.2d shows why many of them have never
gone beyond the bitumen stage. The reason is that the time-temperature requirements have been
too great for major oil generation and expulsion. For example, the Tipton Member of the Green
River Formation (Type IC) would require more than 6,000 m of burial at temperatures above 150°C
to form oil in the cratonic basin of figure 3.2d. In summary, figure 3.2 clearly emphasizes the
importance of using different kinetic parameters for different kerogen types in evaluating the
petroleum potential of sedimentary basins.
A Graphic Method of Defining the Oil and Gas Windows
The integration of time and temperature with burial by use of the Arrhenius equation is a task that is
most suited for a computer. However, it is sometimes desirable to make a hand-calculated estimate
of oil generation when a computer and the supporting software are not readily available. The
following discussion presents a graphic approach for determining the progress of oil generation by
use of time-temperature index graphs based on the Arrhenius equation. In addition to providing a
simple calculation, this approach also provides a working understanding of oil generation kinetics,
which minimizes the "black box" mentality fostered by some computer programs. Graphic estimates
of the time-temperature maturation of organic matter have been used since Karweil (1955)
constructed the first time-temperature graph for coalification. Lopatin (1976) used a modification of

his TTI method to construct a time-temperature graph for oil generation. Such graphs can also be
constructed with Arrhenius kinetics.
In 1988 Wood derived a time-temperature index based on the Arrhenius equation (3.1), which he
called the TTIARR to differentiate it from Lopatin's TTI, which was designated TTILOP. He then
compared the two methods for modeling temperature as a function of time in a large-number of
hypothetical burial histories involving different basin configurations and heating rates. Comparisons
were made with fast-, medium-, and slow-reacting kerogens. His conclusion was that the Lopatin
method tends to underestimate thermal maturity relative to the Arrhenius equation for heating rates
significantly greater than 1°C/my but overestimates thermal maturity for heating rates significantly
less than 1°C/my. He also recognized that TTIARR modeled the different kerogen types more
accurately than TTILOP.
The derivation of TTIARR from the Arrhenius equation is explained in Wood (1988) and Hunt,
Lewan, and Hennet (1991). Wood states that his derivation is an approximate analytical solution of
the Arrhenius equation integral and has a solution error less than 1% at temperatures less than 300°
C and E values greater than 50 kJ/mol. His expression assumes a linear heating rate within each
10°C interval.
As soon as the kinetic parameters E and A are known for a source rock, graphs may be constructed
for determining TTIARR along the burial history curve of the rock in 10°C increments (Hunt, Lewan,
and Hennet 1991). Graphs in figure 3.3 through 3.11 have been constructed for this purpose, with
each graph based on the kinetic parameters listed in table 3.1. TTIARR values for each 10°C interval
a source rock experiences are determined graphically on the basis of the amount of time a source
rock resides in each 10°C interval. As an example, with the solid diagonal lines in figure 3.3, a
source rock subjected to a temperature range of 70° to 80°C for 1 my would have a TTIARR of 20 for
that 10°C temperature interval.
Burial history curves sometimes include long time periods at a constant temperature with negligible
subsidence or uplift. In order to account for this more accurately, the TTIARR for a time interval (tn+1
- tn) at constant temperature may be calculated from the following expression
ARR = [|(n+1 - n) exp (-)|] x 100 (3.2)
Alternatively, the TTIARR at constant temperature may be read directly from the graphs in this paper
by using the dashed diagonal lines between the solid lines.
Addition of the TTIARR values for each 10°C interval, or period of constant temperature, in the burial
history curve of a source rock gives a summation index ( TTIARR). This index gives the cumulative
effect of time and temperature on the oil generation of a source rock. Despite different values for the
kinetic parameters, E and A for different source rock types (table 3.1), the TTIARR values are
always related to the same percentage of oil or gas (X%) which can be generated. The expression
is:
% = [1 - exp (- TTIARR/100)|] x 100 (3.3) This expression is shown graphically in figure 3.12.

On this graph, the beginning of oil generation may be equated to a TTIARR of 1 and the end to a
TTIARR of 460. This same range may also be used for determining the beginning and end of oil
cracking to gas when TTIARR is being determined from the oil-to-gas graph (figure 3.11).
The differences in the kinetics of different kerogens such as Types II A, B, C, and D in table 3.1 are
significant and must be taken into account in evaluating the timing of oil generation in a source rock
as discussed for figure 3.2. This point may be further demonstrated from the graphs by considering
the 10°C temperature range required for a source rock to obtain a TTIARR of 1 in one million years
for each set of kinetic parameters given in table 3.1. The Type IIA kinetics require a temperature
range of 50° to 60°C (fig. 3.3), while the Type IID kinetics require a temperature range near 120° to
130°C (fig. 3.6). Accordingly, the Type IIB (fig. 3.4) and Type IIC (fig. 3.5) kinetics require
intermediate temperature ranges of 90° to 100°C and 110° to 120°C, respectively. Therefore, the
use of a fixed set of kinetic parameters to describe oil generation from all source rocks bearing Type
II kerogens is not valid.
Case Examples
The following examples illustrate how the oil windows of source rocks are determined by use of the
TTIARR graphs in figure 3.3 through 3.10. These examples are based on limited data from the
literature and are assumed to be reasonably accurate. The intent of these examples is to illustrate
the utility of graphic estimates rather than to provide rigorous exploration models for the basins or
regions mentioned in the examples.
Kerogen Type IIA Kinetics
Figure 3.1 shows the burial history plot for the Monterey Shale in the Point Conception COST well
from Petersen and Hickey (1987). table 3.2 contains time-temperature data, plus both the Lopatin
and Arrhenius TTI and TTI values for this well. The TTIARR values for the base of the Monterey
were obtained from the graph for the Type IIA kinetics in figure 3.3. Geochemical data from this well
have puzzled geochemists because several chemical maturity indicators reported by Petersen and
Hickey (1987) showed the Monterey to be in the oil window at a depth of 2,134 m (7001 ft)
equivalent to a subsurface temperature of 83°C. But the calculated Lopatin TTI value of 0.3
indicates the Monterey Shale is not in the oil window. Walker et al. (1983) have also reported similar
discrepancies in the Monterey Shale with oil generation occurring at vitrinite reflectance values less
than 0.5% Ro.
In figure 3.3, the intercepts of each solid diagonal temperature-range line with the exposure time
line from the ordinate give the TTIARR values along the abscissa. As shown in this example, the
Monterey Shale was subjected to the 70° to 80°C temperature range for one million years (figure
3.1), which results in a TTIARR value of 20 (figure 3.3, table 3.2). Summation of this value with the
preceding values calculated at the lower 10°C temperature ranges results in a TTIARR of 27 at this
point in its burial history. Based on figure 3.12, this summation value indicates that 26% of the oil
has been generated. By the time the Monterey has been buried to a temperature range of 90°-100°
C, about 98% of the oil has been generated.

Another example of a high-sulfur Type IIA kerogen is the Senonian bituminous rock (SBR) of Israel
described by Tannenbaum and Aizenshtat (1984). Organic sulfur contents of the kerogens in this
rock unit range from 10 to 12 weight percent. These authors used molecular indicators of
maturation such as the isomerization of hopanes and steranes plus CPI, and the C15+ yield to
estimate the level of maturation. Their data showed that the SBR was well into the oil window in the
Amiaz well. As shown in table 3.2, the TTILOP values do not indicate oil generation in this well, but
the TTIARR value of 134 indicates that about 73% of the oil has been generated.
Kerogen Type IIB Kinetics
The Santonian Brown Limestone (SBL) is considered to be one of the main source rocks in the Gulf
of Suez by Chowdhary and Taha (1987). Its classification as Type IIB is based on the organic sulfur
content of its kerogen (Eglinton et al. 1990). Also, crude oils that are known to have been generated
from this rock have moderately high 2.3% sulfur contents. Chowdhary and Taha used a TTILOP of
15 to estimate that the beginning of oil generation occurred about 4.5 Ma when the SBL reached
temperatures in the range of 120° to 130°C. Only about 30% of the oil has been generated from the
SBL, on the basis of the TTILOP calculation from the burial history data for the DD83-1 well (table 3..
Other maturity indicators by these authors indicate, however, that the SBL is well into or past peak
oil generation in the DD83-1 well. Using the burial history data for this well and the graph (figure
3.4) for kerogen Type IIB, table 3.3 shows that oil generation may have started as early as 9 Ma at
a depth of 2,290 m (7500 ft) within the 90°-100°C temperature range and ended 2.5 Ma in the 130°-
140°C range.
Also, the heating rate for the last 25 my in the DD83-1 well has been 4.2°C/my. According to Wood
(1988), the Lopatin method, relative to the Arrhenius equation, underestimates thermal maturity for
heating rates significantly greater than 1°C/my. Our results in the DD83-1 well agree with this.
Kerogen Type IIC Kinetics
The giant Salym oilfield in the West Siberian Basin produces oil from fractures in the Bazhenov
Shale source rock. According to Lopatin (personal communication) the sulfur content of the kerogen
is higher than in most clastics due to some gypsum occurring in the Bazhenov suite. A burial history
curve for Hole 184 in the Salym Field is shown in figure 3.13 (Lopatin, personal communication).
The time-temperature data in table 3.4 were calculated from the curve using kerogen Type IIC
kinetics (figure 3.5). The heating rate for the last 100 my has been close to 1°C/my. Both the
Arrhenius and Lopatin TTI data indicate that generation started when the Bazhenov reached 100°C
around 85 to 90 Ma and ended after maximum burial about 5 to 8 Ma. Wood (1988) concluded that
these two methods tend to agree when the heating rates are not significantly different than 1°C/my.
Kerogen Type IID Kinetics
Table 3.5 contains time and temperature data for the Kimmeridgian source rock in the Central
Graben of the Norwegian North Sea near the Ekofisk Field (Leonard 1989). The burial history curve
is shown in figure 3.14. According to Leonard (1989), oil generation started about 40 Ma, peaked
around 15 Ma and ended about 5 Ma. TTIARR values based on Type IID kinetics generally agree
with this while the TTILOP values show the beginning and ending of generation somewhat earlier.

The Type IID kinetics was chosen based on analyses of the organic sulfur content of Kimmeridgian
Shale samples (Eglinton et al. 1990).
Table 3.5 also contains TTIARR data for the conversion of oil to gas. These numbers are obtained
by use of the first two columns in table 3.5 with the graph in figure 3.11. For example, the 150-160°
C line in figure 3.11 crosses 3.5 my at a TTIARR of 10. The TTIARR of 410 when used on figure
3.12 indicates that most of the oil remaining in this deep part of the source rock has been converted
to gas. Any oil that migrated upward, of course, would not be affected.
The Mowry Shale is a well-known petroleum source rock in the Rocky Mountain region of the
western United States. Its low sulfur kerogen classifies it as Type IID. The time and temperature
data in table 3.6 were derived from the burial history curve in figure 3.15 (Hagen and Surdam,
1984). Both TTILOP and TTIARR values from using figure 3.6 indicate that oil generation from the
Mowry Shale in the Bighorn Basin started around 52 Ma at a temperature of about 125°C. TTILOP
indicates that oil generation ceased about 22 Ma at 140°C while TTIARR suggests that it ended
around 20 Ma at about 145°C. These differences are not significant and they further support our
observation that TTILOP and TTIARR determinations of the oil window tend to agree when kerogen
Type IIC and IID kinetics are used in basins with moderate heating rates. The heating rate of the
Mowry Shale in this basin from surface to maximum depth was about 1°C/my.
Kerogen Type I Kinetics
TTIARR graphs based on kerogen Type I kinetics from table 3.1 are shown in figures 3.7, 3.8, and
3.9. All three of these are from different regions of the Green River Formation (GRF) in the western
United States. The oil generation interval for the GRF in the Uinta Basin of Utah has been studied in
some detail by Sweeney, Burnham, and Braun (1987). Their burial history curve for the base of the
Eocene in the Shell Brotherson 1-11B4 well is shown in figure 3.16. Time-temperature data from
this curve are in table 3.7. The TTIARR graph (figure 3.8) is used to obtain the TTIARR values in
table 3.7. Both the TTILOP and TTIARR show oil generation starting about 35 Ma at a temperature
of 120°C and ending about 24 Ma within the 140°-150°C temperature range.
This agrees with Sweeney, Burnham, and Braun (1987), who concluded from their model study that
peak oil generation was about 30 Ma. It is rather interesting that in this example of a Type I kerogen
the simple Lopatin model, the Arrhenius TTI graphs, and the sophisticated computer model of
Sweeney, Burnham, and Braun (1987) all agree.
Kerogen Type III Kinetics
Figure 3.10 contains a TTIARR graph for the Type III kerogen kinetics of the Tent Island Formation
of the Beaufort-Mackenzie Basin in table 3.1 (Issler and Snowdon 1990). The kinetic parameters
were obtained by open pyrolysis which includes bitumen, oil and gas from kerogen. However, over
60% of the hydrocarbon generation was within E = 230 10 kJ/mol so this E value is assumed to be
for the oil window. Gas generation from kerogen would continue at higher activation energies. The
above assumption is reasonable if the hydrogen index (HI) from open pyrolysis of an immature
source rock is above 200 since this indicates some oil generation and expulsion. At HI values below

200, gas would be the main product.
Figure 3.17 shows a postulated burial history curve for the deepest major coal bed (Layer 4 of
Welte et al. 1984) in the giant Elmworth gas field in the Deep Western Canada Basin of Alberta and
British Columbia. Precise data were not available for the thickness of eroded rock so the maximum
burial depth was estimated from the recent mean vitrinite reflectance by Welte et al. 1984. The
sources of the gas in the Deep Basin are Lower Cretaceous and Jurassic coals and coaly shales.
The Arrhenius TTI data in table 3.8 were calculated from figure 3.10 for the coal bed of Layer 4
shown in figure 3.17. table 3.8 shows oil and gas generation starting at around 120°C about 56 Ma.
When uplifting began about 32 Ma all of the oil and much of the gas had been generated from the
coaly kerogen. Gas generation from kerogen may have continued until the temperature dropped
below 120°C about 9 Ma. Meanwhile, any oil trapped in the coals, coaly shales, or sands would be
converted to gas. When the oil to gas kinetics curve (figure 3.11) is used with the time-temperature
data in table 3.8, it shows additional gas formation starting around 140°C about 45 Ma and ending
due to uplift and cooling about 15 Ma. It is reasonably certain that there has been no gas generation
from Layer 4 since it cooled below 120°C around 10 Ma based on Arrhenius kinetics. The TTILOP
interpretation is similar with gas generation ending at the present depth of Layer 4.
Discussion
The TTIARR graphs in this paper provide a simple and rapid method for using Arrhenius kinetics to
determine the position of the oil- and gas-generating windows in a basin. Their application also
helps a geologist develop a working understanding of oil generation kinetics. Anyone using these
graphs can quickly see how changes in the time-temperature input data for a basin can change the
conclusions regarding the position of the petroleum generation window.
An obvious question that arises from this presentation is, which graph does one use when studying
a specific source rock? The best approach is to experimentally determine the kinetic parameters of
an immature sample of the source rock being considered in a basin evaluation by hydrous pyrolysis
and construct a TTIARR graph by the method described in Hunt, Lewan, and Hennet 1991. Or use
any other laboratory experiment to determine the Arrhenius kinetic parameters for constructing the
TTIARR graph with the understanding that the kinetics apply only to the specific first-order reaction in
the experiment, such as bitumen to oil or oil to gas.
If this is not feasible, a second approach is to have a geochemical service laboratory analyze the
source rock to determine the kerogen maturity and type and follow up with analyzing the kerogen
for organic sulfur (Eglinton et al. 1990). If the kerogen is Type II, then the sulfur analysis will classify
it as IIA, B, C, or D, and the corresponding graph (figures 3.3, 3.4, 3.5 or 3.6) can be used. If it is
Type III, then use figure 3.10. If Type I, use figure 3.8, the graph for the Type IB kerogen with a
medium reaction rate.
Unfortunately, not enough data are available on the variations of Type I kerogen kinetics with
composition to determine which graph to use. It would be expected, however, that Type I kerogens
with the highest hydrogen content would have the most paraffin chains and therefore be the slowest
reacting. By this interpretation, the Type IC graph (figure 3.9) would be used for kerogen

compositions plotting much above the Type I maturation line on a van Krevelen or a HI/OI
(Hydrogen Index/Oxygen Index) diagram. The Type IA graph (figure 3.7) would be used for those
plotting much below the Type I maturation line, and Type IB Uinta (figure 3.8) would be used for
those plotting on the maturation line.
An important consideration in using this general approach is that the composition of a kerogen may
change laterally or vertically within a source rock as its depositional facies changes, so several
samples may need to be analyzed. For example, Lewan (1986) noted that the organic sulfur content
in Type II kerogens of the Monterey Shale vary from 11.2 wt % in the distal pelagic chert-rich facies
to 4.6 wt % in the more proximal clastic facies. This would classify different parts of the Monterey as
Types IIA, B, C, and D. Therefore, one should use caution in overgeneralizing the composition of a
regional source rock.
Another approach applicable to Type II kerogens is based on our observation that source rocks
composed of claystones, mudstones, and siltstones contain Type II kerogens with lower organic
sulfur contents than source rocks composed of biogenic pelagics such as carbonates, diatomites,
and cherts. The reason for this may be that evaporites such as gypsum and anhydrite are more
frequently associated with the latter than the former. With this generalization, biogenic-pelagic
source rocks would have oil generation kinetics similar to those of kerogen Types IIA or IIB, and
clastic source rocks would have oil generation kinetics similar to those of Types IIC or IID. A
concern here is that in a few euxinic clastic environments the buildup of H2S can sometimes
exceed the kdetrital iron content of the sediment, which may result in kerogens with high organic
sulfur contents.
For Type II source rocks in areas where only information on crude oil composition is available, a
third approach based on the sulfur content of crudeoils may be applied. The premise here is that
high-sulfur crude oils are derived from high-sulfur kerogens, and low-sulfur oils from low-sulfur
kerogens. Therefore, oil generation kinetics for the former would be similar to those for kerogen
Types IIA and B and, for the latter, to those for Types IIC and D. An important prerequisite for this
approach is that the oils are not biodegraded and have no secondary sulfur enrichment. When little
subsurface information is available, the graph for a Type IIC kerogen with a medium reaction rate
(figure 3.5) will give a first approximation of the oil window for Type II source rocks.
The application of time-temperature relationships for oil generation has progressed from using one
parameter for all oil generation (TTILOP) to using three for each of the major kerogen types (I, II, III)
and now to using four within the Type II group (IIA, B, C, and D) and three within the Type I group.
An obvious question is how many sets of kinetic parameters will ultimately be required to cover all
potential source rocks?
Concerning the graphs, it is probable that six Type II graphs will suffice for practically all kerogens in
this class. TTIARR values are determined for 10°C intervals on a burial history curve. Note that the
first diagonal line in the upper left hand corner of figures 3.5 and 3.6 for kerogen Types IIC and IID
differ by only 10°C. Type IIC is 90°-100°C and IID is 100°-110°C. It is questionable whether or not a
graph between these two would be a useful refinement in view of all the other estimates in
paleotemperatures and burial history curves. Figures 3.3 and 3.4 (Types IIA and IIB) have a bigger
spread of 20° to 30°C. But the very high sulfur kerogens are not that common. One more graph may
be needed between Types IIB and IIC plus a graph for a slower reacting kerogen than Type IID.
This would provide 6 graphs to cover all Type II kerogens, three for Type I, and one for Type III. Not

enough information is available to describe reaction rate differences within the Type III kerogen
group. If their differences are like Type I, then a total of 12 graphs would probably cover all
kerogens.
In model studies involving a Paleozoic cratonic basin and a Neogene pull-apart basin, a computer
used 2,000 increments along the burial history curves to determine TTIARR with different Type II
kerogens. The same burial curve and kerogen types were used to determine TTIARR graphically
as described in this paper. The TTIARR values determined graphically came within 3% of those
determined by the computer.
The TTIARR graphs in this paper provide a simple and rapid method for using Arrhenius kinetics to
determine the position of the hydrocarbon-generating window in basins containing kerogen Types I,
II, and III as oil and gas sources. Application of these graphs to a variety of known source rocks has
shown that they match the oil window indicated by other maturation parameters more closely than
the Lopatin method for the Type II faster reacting kerogens and for heating rates significantly higher
or lower than 1°C/my. However, both methods gave similar results for medium- and slow-reacting
Type II kerogens and medium-reacting Type I and III kerogens for the source rocks tested. In the
Type I analysis, both methods gave the same result as a sophisticated computer model study.
Type II oil-generating kerogens were designated Types IIA, B, C, and D, depending on their
reaction rates and organic sulfur contents. The fast- and medium-fast-reacting kerogen Types IIA
and B, respectively, had the higher sulfur contents while the medium- and slow-reacting kerogen
Types IIC and D, respectively, had the lower sulfur contents. Activation energies of the kerogens
were inversely proportional to their sulfur contents. The graphs for kerogen Types IIA, B, C, and D
are believed to represent a continuum that will cover most studies concerning the depth of an oil
window formed by source rocks with Type II kerogens.
TTIARR graphs similar to those in this paper may be constructed from Arrhenius kinetic parameters
derived by any method that describes a first-order reaction. figure 3.12 can be used for bitumen to
oil, oil to gas, kerogen to gas, etc. However, the kinetic parameters for making the graphs (figures
3.3 through 3.11) must be derived by methods describing the specific reaction being considered.
References
Bergius, F. 1913. Production of hydrogen from water and coal from cellulose at high
temperatures and pressures. J. Soc. Chem. Ind. 32:462-467.
Burnham, A. K., R. L. Braun, and A. M. Samoun. 1987. Further comparison methods for
measuring kerogen pyrolysis rates and fitting kinetic parameters. Lawrence Livermore Natl.
Lab. UCRL-97352.
Burnham, A. K., R. L. Braun, H. R. Gregg, and A. M. Samoun. 1987. Comparison of methods

for measuring kerogen pyrolysis rates and fitting kinetic parameters. Energy and Fuels
1:452.
Chowdhary, L. R. and S. Taha. 1987. Geology and habitat of oil in Ras Budran field, Gulf of
Suez, Egypt. AAPG Bull. 71:1274-1293.

Dow, W. 1977. Kerogen studies and geological interpretations. J. Geochem. Explor. 7(2):77-
79.
Eglinton, T. I., J. S. Sinninghe Dansté, M. E. L. Kohenen, and J. W. de Leeuw. 1990. Rapid

estimation of organic sulfur content of kerogens, coals and asphaltenes by pyrolysis-gas
chromatography. Fuel 69:1394-1404.
Engler, K. O. V. 1913. Die Chemie und Physik das Erdols, 1:1-37.
Habicht, J. K. A. 1964. Comment on the history of migration in the Gifhorn Trough. Proc.
Sixth World Petrol. Congr. section 1, paper 19-PD2, p. 480.
Hagen, E. S. and R. C. Surdam. 1984. Maturation history and thermal evolution of

Cretaceous source rocks of the Bighorn Basin, Wyoming and Montana: In J. Woodward, F.
F. Meissner, and J. L. Clayton, eds., Hydrocarbon Source Rocks of the Greater Rocky
Mountain Region, pp. 321-338. Denver: Rocky Mtn. Assoc. of Geolgists.
Hilt, C. 1873. Die Beziehungen zwischen der Zusammensetzung und den technischen
Eigenschaften der Steinkohl. Sitzber. Aachener Bezirksvereinigung VDI, 4.
Hunt, J. M. 1979. Petroleum Geochemistry and Geology, pp. 458-462. San Francisco: W. H.
Freeman.
Hunt, J. M., M. D. Lewan, and R. J-C Hennet. 1991. Modeling oil generation with time-
temperature index graphs based on the Arrhenius Equation. AAPG Bull. 75(4):797-807.
Issler, R. D. and L. R. Snowdon. 1990. Hydrocarbon generation kinetics and thermal

modelling, Beaufort-Mackenzie Basin. Bull. Can. Petrol. Geol. 38:1-16.
Karweil, J. 1955. Die Metamorphose der Kohlen vom Standpunkt der physikalischen
Chemie. Z. deutschen geologischen Gesellschaft 107:132-139.
Leonard, R. C. 1989. Generation, migration and entrapment of hydrocarbons on Southern

Norwegian Shelf. AAPG distinguished lecture.
Lewan, M. D. 1985. Evaluation of petroleum generation by hydrous pyrolysis

experimentation. Phil. Trans. Roy. Soc. London 315:123-134.
Lewan, M. D. 1986. Organic sulfur in kerogens from different lithofacies of the Monterey
Formation. Division of Geochemistry, abstract no. 94, 192d ACS Natl. Mtg. September 7-12,
Anaheim, California.
Lewan, M. D. 1989. Hydrous pyrolysis study of oil and tar generation from Monterey shale
containing high sulfur kerogen. Symposium on Geochemistry, abstract, ACS Natl. Mtg. April
9-14, Dallas, Texas.

Lewan, M.D. 1992. Primary oil migration and expulsion as determined by hydrous pyrolysis.
Proceedings of the 13th World Petroleum Congress, Buenos Aires, Argentina.
Lewan, M. D. in press. Laboratory simulation of petroleum formation: hydrous pyrolysis. In

M. H. Engel and S. A. Macko, eds., Organic Geochemistry. New York: Plenum.
Lewan, M. D. and B. Buchardt. 1989. Irradiation of organic matter by uranium decay in the
Alum shale, Sweden: Geochim. Cosmochim. Acta 53:1307-1322.
Lewan, M. D., J. C. Winters, and J. H. McDonald. 1979. Generation of oil-like pyrolyzates

from organic rich shale. Science 203:897-899.
Lopatin, N. V. 1971. Temperature and geological time as factors of carbonification. Izvestiya

Akademii Nauk SSSR, Seriya Geologicheskaya, no. 3, pp. 95-106.
Lopatin, N. V. 1976. The determination of the influence of temperature and geologic time on
the catagenic processes of coalification and oil-gas formation. In Issledovaniya
organicheskogo veshchestva sovremennykh i iskopaemykh osadkov (Research on Organic
Matter of Modern and Fossil Deposits), pp. 361-366. Moscow: Akademii Nauk SSSR,
Izdatel'stvo, "Nauka."
Lovering, E. G and K. J. Laidler, 1960, A system of molecular thermochemistry for organic

gases and liquids. II. Extension to compounds containing sulfur and oxygen: Canadian
Journal of Chemistry 38:2367.
Miknis, F. P. and T. F. Turner. 1988. Thermal decomposition of Tipton Member, Green River
Formation of oil shale from Wyoming: Report by Western Research Institute, Laramie, to
Dept. of Energy, September 1988.
Miknis, F. P., T. F. Turner, G. L. Berdan, and P. J. Conn. 1987. Formation of soluble

products from thermal decomposition of Colorado and Kentucky oil shales. Energy and Fuels
1:477-483.
Orr, W. L. 1985. Kerogen/asphaltene/sulfur relationships in sulfur- rich Monterey oils. Org.

Geochem. 10:499- 516.
Petersen, N. F. and P. J. Hickey. 1987. California Plio-Miocene oils: evidence of early

generation. In R. F. Meyer, ed., Exploration for Heavy Crude Oil and Natural Bitumen. AAPG
Studies in Geology, no. 25, pp. 351-359.
Philippi, G. T. 1965. On the depth, time and mechanism of petroleum generation: Geochim.
Cosmochim. Acta 29:1021-1049.
Poulet, M. and J. Roucache. 1969. Etude Geochemique des gisements du Nord-Sahara

(Algerie). Rev. Inst. Francais du Petrole 24:615-644.
Quigley, T. M., A. S. Mackenzie, and J. R. Gray. 1987. Kinetic theory of petroleum

generation. In B. Doligez, ed., Migration of Hydrocarbons in Sedimentary Basins, pp. 649-

665. Paris: Technip.
Shen, M. S., L. S. Fan, and K. H. Castleton. 1984. American Chemical Society, Division of
Petroleum. Chemistry 29(1):127-134.
Smith, J. W. 1983. The chemistry that formed Green River Formation oil shale. In F. P.
Miknis and J. F. McKay, eds., Geochemistry and Chemistry of Oil Shales, pp. 235-248.
Washington, D.C.: ACS.
Sweeney, J. J., A. K. Burnham, and R. L Braun. 1987. A model of hydrocarbon generation

from Type I kerogen: Application to Uinta Basin, Utah: AAPG Bull. 71:967-985.
Tannenbaum, E. and Z. Aizenshtat. 1984. Formation of immature asphalt from organic-rich

carbonate rocks. II. Correlation of maturation indicators: Org. Geochem. 6:503-511.
Teichmuller, M. 1958. Metamorphisme du charbon et prospection du petrole. Rev. Ind.

Minerale, numero special, pp. 1-15.
Ting, T. C. 1975. Reflectivity of disseminated vitrinites in the Gulf Coast region. Petrographie
de la matiere organiques des sediments relations avec la paleo-temperature et le potentiel
petrolier. Paper presented to the Centre National de la Recherche Scientifique, September
15-17, 1973, Paris.
Tissot, B., B. Durand, J. Espitalie, and A. Combaz. 1974. Influence of nature and diagenesis
of organic matter in formation of petroleum. AAPG Bull. 58:499-506.
Tissot, B. and J. Espitalie. 1975. L'evolution thermique de la matiere organiques des

sediments: Application d'une simulation mathematique. Rev. Inst. Francais du Petrole
30:743-777.
Tissot, B., R. Pelet, and P. H. Ungerer. 1987. Thermal history of sedimentary basins,
maturation indices, and kinetics of oil and gas generation. AAPG Bull. 71:1445-1466.
Van Krevelen, D. W. 1961. Coal: Typology, Chemistry, Physics Constitution, p. 295.

Amsterdam: Elsevier.
Walker, A. L., T. H. McCulloh, N. F. Petersen, and R. J. Stewart. 1983. Anomalously low

reflectance of vitrinite, in comparison with other petroleum source-rock maturation indices,
from the Miocene Modele Formation in the Los Angeles Basin, CA. In C. M. Isaacs and R. E.
Garrison, eds., Petroleum Generation and Occurrences in the Miocene Monterey Formation,
California, pp. 185-190. Los Angeles: SEPM, Pac. Sect.
Waples, D. W. 1980. Time and temperature in petroleum formation: Application of Lopatin's

method to petroleum exploration: AAPG Bull. 64:916-926.
Waples, D. W. 1985. Geochemistry in Petroleum Exploration, pp. 121-154. Boston:

International Human Resources Development Corporation.
Welte, D. H., W. Stoessinger, R. G. Schaefer, and M. Radke. 1984. Gas generation and
migration in the deep basin of Western Canada. In J. A. Masters, ed, Elmworth: Case Study
of a Deep Basin Gas Field, pp. 35-47. AAPG Memoir 38.
White, D. 1915. Geology: Some relations in origin between coal and petroleum. J. Wash.
Acad. Sci. 5(6):189-212.
Wood, D. A. 1988. Relationships between thermal maturity indices calculated using

Arrhenius equation and Lopatin method: implications for petroleum exploration: AAPG Bull.
72:115-134.
Yang, H. S. and H. Y. Sohn. 1984. Kinetics of oil generation from oil shale from Liaoning
Province of China. Fuel 63:1511-1514.


4. Sources, Cycling, and Distribution of Water Column

Particulate and Sedimentary Organic Matter in Northern
Newfoundland Fjords and Bays: A Stable Isotope Study
A combined stable isotope and elemental study was conducted to investigate the sources and fate
of organic matter in northern Newfoundland bays. The sediments within one bay were characterized
by a terrigenous signature, which was related to the 13C, 15N and C/N of seston collected
downriver from a paper mill (-25.5 , 2.2 , and 12.3). Freshwater seston collected at other locations
was influenced by phytoplankton as revealed by 13C (-24.9 ), 15N (5.1 ) and C/N (8.1) values.
Three Laminaria solidungula plants were found to have average 13C, 15N and C/N values of -20.3
, 4.6 , and 9.2, respectively, which were used to model the contribution of macroalgae to these
waters. Marine phytoplankton dominated offshore seston samples in August and November, and in
May terrestrial and macroalgae sources were also influential. The average 13C, 15N, and C/N (-
24.6 , 8.2 , and 6.5) of samples not influenced by other end members were used to define the
phytoplankton end member. The geochemistry of marine seston collected from the lower water
column was affected by degradation, resuspension, and changes in productivity. Sedimentary
organic carbon and nitrogen isotope values and ratios were intermediate between phytoplankton
and macroalgae, indicating a mixture of these two sources. These results indicate that organic
matter in continental shelf environments may consist of a complex mixture of several sources.
Furthermore, in these areas the relative contribution of sources and processes affecting the
geochemical composition of seston vary considerably spatially and seasonally.
The natural abundances of the stable isotopes of carbon (C) and nitrogen (N) and C/N have been
well utilized for estimations of the origins of organic matter in marine systems. The successful
separation of sources or end members is dependent on two basic premises. First, sources must
have distinct isotopic and/or elemental compositions. Second, geochemical characteristics should
not be altered during the transport, deposition, and preservation of organic matter.
The basis for the geochemical uniqueness of marine and terrestrial end members is a consequence
of differences in both metabolism and inorganic nutrient sources in primary productivity.
Fractionation during CO2 assimilation in marine and terrestrial C3 plants occurs primarily during
carboxylation by ribulose 1,5-bisphosphate carboxylase (RuBP) and to a lesser extent during CO2
diffusion into the cell. Aquatic plants are more commonly diffusion limited than terrestrial plants, and
this has the effect of reducing the enzymatic fractionation (O'Leary 1981).
Further increases in 13C have been observed in marine phytoplankton during high productivity as
the availability of dissolved CO2 gas is reduced (Deuser 1970). This enrichment may be a
consequence of an increase in the 13C of CO2 as the isotopic equilibrium between dissolved CO2
gas and DIC is unable to keep pace with the chemical equilibrium (Deuser and Degens 1967;
Degens 1969). As the processes of diffusion and CO2-HCO3- isotopic disequilibrium become

dominant, organic matter production favors a greater degree of 13C assimilation than occurs when
fractionation during carboxylation is more thoroughly expressed.
Differences between marine and terrestrial end members in 15N are primarily related to their
inorganic nutrient sources. The ultimate nutrient source for terrestrial productivity is atmospheric
nitrogen, which is made available to plants by nitrogen-fixing soil bacteria. Fractionations during
nitrogen fixation have been shown to be small, and therefore a 15N near that of atmospheric N2 of
0 is expected for the terrestrial biosphere (Hoering and Ford 1960; Delwiche and Steyn 1970;
Macko et al. 1987). Although hydrolyzable soil compounds have been shown to be considerably
more enriched in 15N than atmospheric N2, the refractory fraction is similar and is expected to be
preserved during transport to the sea (Cheng, Bremner, and Edwards 1964; Sweeney, Liu, and
Kaplan 1976). The primary source of nitrogen for marine algae is dissolved inorganic nitrogen in the
form of ammonium or nitrate with 15N values commonly in the range of 6 to 10 (Miyake and
Wada 1967; Cline and Kaplan 1975; Liu 1979). Upon assimilation, phytoplankton frequently have
values similar to their inorganic nitrogen source (Miyake and Wada 1967; Wada 1980).
When assimilation of nutrients by phytoplankton is not limited by availability, a large fractionation

effect is invoked (Macko et al. 1987; Saino and Hattori 1987). This results in depleted 15N values
that may overlap with, or be more depleted in 15N, than the terrestrial end member (Mariotti,
Lancelot, and Billen 1984; Saino and Hattori 1987). Carbon isotope values can assist in
distinguishing these sources. At high latitudes the 13C of phytoplankton and terrigenous matter
can be similar (Sackett et al. 1965; McConnaughey and McRoy 1979; Rau, Sweeney, and Kaplan
1982). An additional indicator of organic matter origin is the carbon-to-nitrogen ratio.
Phytoplankton are abundant in nitrogen-rich proteins and therefore have a low C/N near six (Muller
1977). Terrestrial plants are rich in nitrogen-deficient carbohydrates and lignins and therefore have
a significantly greater C/N than phytoplankton have (Goodell 1972). A combined carbon and
nitrogen isotope and C/N study may more clearly define the origins of organic matter when the
isotopic composition of the end members is similar.
This study was undertaken to investigate the relative inputs of three sources of organic matter to the
nearshore water column and sediments of the continental shelf environment off northern
Newfoundland through isotopic and elemental analyses. In addition to phytoplankton and terrestrial
sources the influence of macroalgae was also evaluated. The relative distribution of organic matter
sources to the marine water column and sediments was based on an isotopic and C/N comparison
with these end members. Changes in geochemistry that potentially occur during transport and
deposition were assessed by comparisons of surface seston with lower water column suspended
and sedimentary organic matter. Further diagenesis within the upper sedimentary layer was
evaluated by comparing surficial sediments with those 3 cm deeper.
Materials and Methods
Waters from riverine drainages were obtained at roadside stations at various locations across the
Island in May 1987 (figure 4.1). Marine water column and sediment samples were taken along the
northern coast of Newfoundland in the summers of 1986 and 1987, during three research cruises
aboard the FRV Wilfred Templeman, CSS Dawson, and CSS Baffin (figure 4.2). All samples were
analyzed for 13C, 15N, and carbon and nitrogen abundances to characterize organic matter

sources and to assess the geochemical effects of diagenesis.
Surficial sediments were collected with a 0.25 m2 Van Veen grab sampler. Once the sediment was
recovered, two samples were taken, one from the upper oxygenated layer and another from a depth
of 3 cm. Surface, middle, and bottom waters were obtained by lowering 5 L Niskin bottles to 5 m,
100 m, and 5 m above bottom, respectively. Riverine waters were collected in PVC (polyvinyl
chloride) buckets and stored for approximately twelve hours until they could be filtered. Seston was
isolated from 5 to 10 L of water by filtration through preashed 1.2 Whatman GF/C glass fiber
filters. All filters were first rinsed with at least 300 mL of distilled water and then acidified (30% HCl)
to remove carbonates.
In preparation for isotopic analysis, sediments were freeze dried, acidified (30% HCl), dried, and
ground into a fine powder. Combustion of organic matter was performed by a modified Dumas
method (Macko 1981). The surficial layer of the POM filter or 200 to 400 mg of sediment was placed
in an ashed quartz tube to which excess precombusted copper oxide and pure copper were added.
Evacuated samples were heated to 850°C and allowed to cool gradually overnight to prevent
formation of carbon monoxide and nitrogen oxides.
Carbon dioxide and nitrogen gas were separated cryogenically from the combustion products.
Purified carbon dioxide and nitrogen isolates from the samples were analyzed for their isotopic
compositions on either a VG 903E or Prism triple collector stable isotope ratio mass spectrometer.
Values for 13C and 15N are reported in per mil ( ) notation relative to PDB (Peedee Belemnite)
and atmospheric nitrogen, respectively.
( ) = (Rsample/Rstandard - 1) x 1000 (4.1)

where R is equal to 15N/14N for nitrogen and 13C/12C for carbon.
Abundance measurements for gas samples of nitrogen were determined a calibrated volume within
the mass spectrometer and for carbon on a calibrated manometer during cryogenic gas separation.
Results and Discussion
Seston from freshwaters were collected from many locations across Newfoundland and
characterized to assist in defining a terrestrial end member for this system (fig. 4.1). All 13C values
for these samples were within or near the range of -26.5 to -24.8 , reported by previous studies for
this end member (Macko 1983; Wada et al. 1987; table 4.1). However the majority of seston
samples were more enriched in 15N than values of 0 to 4 , which have previously been used to
model terrigenous debris (Macko 1983; Macko, Pulchan, and Ivany 1987; table 4.1). In addition,
ratios of carbon to nitrogen were low in comparison with typical terrestrial environments (Goodell
1972; table 4.1). The enriched 15N and low C/N values within lacustrine and riverine systems of
Newfoundland are indicative of productivity originating in freshwater phytoplankton (Muller 1977;
Minigawa and Wada 1984).
The Exploits River seston had a geochemically unique composition relative to other freshwaters of
Newfoundland (table 4.1). The average 13C, -25.5 , and 13N, 2.2 , were more depleted than
nearly all other stations. The average C/N, 12.3, was greater than any value obtained for the other
freshwaters. The cause of the depleted stable isotope and high C/N values may be associated with

effluent from a paper mill located on the Exploits River. The freshwater seston data suggest that
within the northern Newfoundland system the terrestrial end member is influenced by both
freshwater phytoplankton and paper mill effluent. Consequently, the signature of this end member
falls within a range defined by the average 13C and 15N for freshwater seston influenced by
phytoplankton, -24.9 and 5.1 , respectively, and Exploits River seston, -25.5 and 2.2 , respectively
(table 4.2).
The importance of kelp or macroalgae as suppliers of productivity to high-latitude food chains and
sediments has been recognized (Dunton and Schell 1987; Macko, Pulchan, and Ivany 1987; Mayer,
Macko, and Cammen 1988). Previous studies off the north and south coasts of Newfoundland have
indicated that macroalgae are dominant contributors of organic matter to the sediments of these
areas (Pulchan 1987; Troke 1987). Isotope values of macroalgae have been observed to fall within
wide ranges of -20 to -12 and 5 to 10 for carbon and nitrogen, respectively (Miyake and Wada
1967; Stephenson, Tan, and Mann 1984). The 13C and 15N of a dominant species, Laminaria
solidungula, was taken to represent the macroalgae end member within this system (table 4.2).
The isotopic abundances of carbon and nitrogen in marine surface water seston off northern
Newfoundland are similar to those reported by other studies (figures 4.3, 4.4, and 4.5). Seston had
an average 13C of -24.3 , which is similar to values reported for phytoplankton from high-latitude
areas (Sackett et al. 1965; McConnaughey and McRoy 1979; Tan and Strain 1983; Dunton and
Schell 1987). The majority of surface seston 15N values were within the range of 5 to 9 , which is
common for phytoplankton from coastal areas (Wada 1980; Minigawa and Wada 1984; figures 4.3,
4.4, and 4.5). The C/N of surface water seston is similar to the range reported for marine
phytoplankton (Muller 1977; figures 4.3, 4.4, and 4.5).
Filamentous organic matter collected in two phytoplankton tows was separated from copepods and
other zooplankton. The average 15N and C/N of the filamentous material, 9.6 and 4.5
respectively, were within the range of the surface seston (table 4.3). The 13C of this organic
matter, -21.5 , was greater than any of the surface seston values. Zooplankton were more
enriched in 15N and 12C than the filamentous material, which implies distinct origins for these
materials (table 4.3). The increasing enrichments in 15N from filaments to copepods to copepods
eggs is suggestive of a trophic level fractionation effect (DeNiro and Epstein 1981; Minagawa and
Wada 1984). The variation in 13C values among copepods and eggs may be related to differing
relative compositions of 13C-depleted lipids in these samples (Deines 1980).
The distinction in 13C between filamentous material and seston may be a consequence of
differences in the type or source of material collected by the two sampling devices. Niskin bottles
primarily collect suspended particulate organic matter (POM) (McCave 1975; Altabet 1988), which
has been characterized by depleted 13C values relative to the sinking fraction obtained in plankton
tows or sediment traps (Entzeroth 1982). Alternatively, a larger contribution of macroalgae to the
plankton tow material could account for the observed enrichment in 13C relative to seston.
Enrichments in the 13C of phytoplankton have also been attributed to isotopic discrimination
between HCO3- and CO2 in seawater (Deuser 1970). During periods of high productivity there is
increased utilization of CO2, which shifts the equilibrium between HCO3- and dissolved CO2 in
favor of the production of CO2. Under these conditions the normally observed 7 equilibrium

fractionation effect in 13C, between HCO3- and dissolved CO2, is reduced (Deuser and Degens
1967; Degens et al. 1968). This leads to enrichments in 13C of dissolved CO2 and of the
phytoplankton that utilize this gas. The filamentous tow material may not have originated
concurrently with the seston in these areas, although both of these sample types were collected
during the May cruise. Temporal differences in phytoplankton productivity could result in the isotopic
differences between tow material and seston.
The origin of the plankton tow material is uncertain. Isotopic and elemental data suggest that it is
composed of phytoplankton and macroalgae, although there is a possibility that it consists entirely
of phytoplankton that was produced under CO2 limitation. It is difficult to discern the relationship of
this material to the rest of the system because only two samples were obtained. Consequently, the
seston samples are considered to be more representative of marine organic matter.
Variations observed in the isotopic and elemental abundances of surface water seston can be
related to changes in productivity or relative inputs of organic matter from terrestrial, phytoplankton,
and macroalgae detritus. In May, seston from Exploits, Halls, Bonavista, and Trinity Bays was
characterized by lower 15N (2.7 to 6.8 ) and higher C/N (8.7 to 17.0) and organic concentrations
than at other times of the year (figures 4.3, 4.4, and 4.5). These changes indicate an increased
terrestrial influence in May as a consequence of spring runoff.
Depletions in 13C, which normally accompany terrestrial input, were not always observed. At EX4,
EX2, and TB5 the carbon values in May, -23.7 to -23.5 , were greater than at other times of the
year (figure 4.4). This seems to contradict a terrestrial influence since the average value for
freshwater seston from the Exploits River was -25.5 . The most enriched 13C within Exploits Bay, -
22.5 , was obtained at EX1, the station nearest to the mouth of the Exploits River. The carbon and
nitrogen isotope values at EX1 were similar to those of L. solidungula, which suggests that in
addition to the terrestrial-marine mixing evident throughout Exploits Bay, there is a strong
macroalgae contribution as well (table 4.2). A macroalgae influence at HB3 in May and ND2 in
November could also result in the observed enrichments in 13C observed at these stations (figures
4.3 and 4.4).
The low 15N values observed at stations EX6 and ND2 in November not accompanied by high C/N
may be the result of fractionation during low productivity when nutrient concentrations were not rate
limiting (figure 4.3; Mariotti, Lancelot, and Bellen 1984; Macko et al. 1987; Saino and Hattori 1987).
The anomalously high 15N, 11.7 , and low 13C, -29.7 , at WB1 in May appear to be the result of
fractionation during degradation (Gordon 1971; Saino and Hattori 1987; Altabet 1988). However,
the large total suspended matter value of greater than 2 mg/L and low organic concentrations
suggest an allocthonous input whose nature is uncertain (table 4.4).
Except for stations within Exploits, Trinity, and Halls Bays in May and ND2 in November, where a
terrestrial or macroalgae contribution was observed, low 13C values suggested a predominance of
phytoplankton detritus. If these stations are not included, the average for 15N and 13C is 8.2
1.5 and -24.6 1.6, respectively. These values define the phytoplankton end member for the bays
of northern Newfoundland (table 4.2).
The isotopic and elemental composition of seston from lower waters is a function of sources,
degradation, and resuspension. Nitrogen isotope values from all stations and depths were

distributed over a very wide range as a consequence of intense cycling in these waters (figure 4.6).
Midwater column 15N values exhibited the widest range, indicating that seston at this depth is
affected by cycling to a greater extent than in surface or bottom waters. The variations in 13C were
considerably less than those in 15N, the majority of samples from all depths being within a range of
3 to 24 (figure 4.7). Apparently, processes altering the 15N of seston did not have as great an
effect on 13C. Ratios of carbon to nitrogen were widely distributed, though the majority of samples
had values less than 13, which confirms the predominantly phytoplankton origin of seston (fig. 4.8;
Muller 1977).
Degradation of organic matter typically results in increases in 15N and C/N and decreases in 13C
and organic concentrations (Gordon 1971; Muller 1977; Saino and Hattori 1987). Although these
changes are not clearly evident when the seston data are considered as a whole, such trends were
observed at specific stations (figures 4.6, 4.7, and 4.8; table 4.4). Increases in 15N in middle and/
or bottom water seston relative to surface waters were observed at many stations and were
indicative of fractionation during degradation (figures 4.9 and 4.10).
The changes in 13C and C/N of seston with depth were not as consistent as the trends observed in
15N. However, the lack of changes does not exclude the possibility of diagenesis. The smaller
variation in 13C relative to 15N may be explained by a greater utilization and mobilization of
nitrogen than carbon during decay. In addition, microbial activity may increase the nitrogen content
of detritus over time (Tenore 1983). The variable C/N values and organic concentrations of lower
water seston in this study may, therefore, be related to the size of the bacterial community present.
At some stations the 15N of middle-water seston is lower than that of seston in surface and bottom
waters (figure 4.11). Similar depletions in 15N have been observed at the base of the euphotic zone
and were attributed to isotopic discrimination during assimilation of inorganic nutrients under high
concentrations where conditions were nonrate limiting (Saino and Hattori 1980, 1985, 1987).
However, decreases in seston nitrogen isotope values have also been attributed to growth of
bacteria utilizing isotopically depleted ammonium (Wada 1980; Libes and Deuser 1988). Owing to
the lack of data on nutrient concentration and isotopic abundances and information on the bacterial
population present, it was not possible in this study to distinguish between these two processes.
Seston from bottom waters frequently had greater concentrations of total suspended matter and
organic carbon and nitrogen than surface and/or middle waters had (table 4.4). A contribution of
sedimentary material to these samples was probable since the majority of bottom samples were
collected 5 m above the seabed. Near-bottom seston 13C values have been observed to reflect a
mixture of sediments and surface water seston as a consequence of resuspension (Tan and Strain
1979). Similar shifts in bottom-water seston 15N and 13C values were observed at several
stations in this study and indicated resuspension of sediments in these areas (figures 4.10 and
4.12).
These results indicate that the isotopic and elemental composition of seston in lower waters was
affected by productivity, degradation, and resuspension. These processes were observed, in some
cases, to affect one parameter to a greater extent than it affected others. For example, degradation
in middle waters at station HB1 in November resulted in a 5.6 increase in 15N with respect to the
surface but had negligible effects on 13C or C/N (table 4.4). Resuspension is clearly indicated at

GB3 in August by an increase of 2.09 mg/L in total suspended matter of bottom-water seston with
respect to the middle-water sample. Both carbon and nitrogen isotope values at GB3 were shifted
by greater than 1 , but the change in C/N was minor. These data indicate that degradation and
resuspension can affect seston carbon and nitrogen to varying degrees or independently. A clearer
interpretation of the processes affecting the isotopic and elemental composition of seston could be
obtained by more extensive sampling with depth at each station.
The isotopic and elemental composition of surficial sediments within northern Newfoundland bays
should reflect a mixture of the three end members that contribute to the waters of this system. A
mixing equation was applied to the sediment data to estimate the relative contributions of each of
the end members (Harrigan, Zieman, and Macko 1989). However, quantitative determinations of
this nature are hindered by the variations of the end members in this study. Seasonal variation in
the 13C of macroalgae has been shown and might be expected for 15N as well (Dunton and
Schell 1987; Fry, Macko, and Zieman 1987). Nonetheless, the clear distinction among end
members in 15N and 13C permitted a qualitative interpretation of the origins of sedimentary
organic matter.
The average 15N and C/N, 7.5 and 6.8, respectively, for surficial sediments was very similar to
those of phytoplankton. However, the average 13C of these sediments, -21.7 , was closer to that
of the macroalgae end member (table 4.2). The enrichment in 13C of surficial sediments relative to
seston is contradictory to the trend associated with diagenesis and is more accurately interpreted as
a consequence of mixing of phytoplankton and macroalgae (Eadie and Jeffrey 1973; Jeffrey et al.
1983). Phytoplankton, with a higher C/N than that of macroalgae, potentially contribute more
nitrogen to sediments and therefore have a greater effect on sedimentary 15N. Preservation of
macroalgae carbon and 13C values in sediments may be enhanced by the more refractory nature
of this material relative to phytoplankton.
Macroalgae detritus could be preferentially deposited as larger, faster sinking particles that are not
well sampled by water bottle collections (McCave 1975). This hypothesis is supported by the
similarity in 13C between sediments and the phytoplankton tow material (tables 4.3 and 4.5). The
importance of macroalgae as dominant primary producers in high-latitude systems is clearly
illustrated by these results. A preferential contribution of macroalgae detritus to sediments may
explain the observed enrichment of 1 to 2 in 13C of high-latitude sediments with respect to surface
seston that previously has been attributed to a trophic-level fractionation effect (McConnaughey and
McRoy 1979).
The 13C and C/N of surficial sediments at EX1 were very similar to those of the terrestrial end
member (tables 4.2 and 4.6). A fraction of the terrigenous material in Exploits Bay is derived from
paper mill effluent entering into the Exploits River. The grab transect in Exploits Bay clearly
demonstrates mixing of terrestrial and marine sources by increases in 13C and 15N with distance
from the head of the estuary (figures 4.13 and 4.14). A similar transect in Conception Bay does not
show this mixing trend. Isotopic values throughout the Conception Bay are relatively constant and
indicative of a mixture of phytoplankton and macroalgae material.
Diagenesis in sediments must be assessed to determine whether this process obscures the
geochemical signature of the source material. As stated earlier, the differences in 15N and 13C
between seston and surficial sediments were not believed to be a consequence of degradation. The

average isotopic and elemental composition of samples taken at 0 and 3 cm are nearly identical
(table 4.5). This indicates that either diagenesis is not a significant process in the upper 3 cm or that
bioturbation has resulted in a homogenization of this upper sedimentary layer. The presence of
numerous worm tubes and other macrofauna in many of the grab samples suggests that
bioturbation results in mixing of surface sediments to this depth. In either case the geochemical
signature was preserved. Consequently, interpretations based on isotopic and elemental data in this
study provide a clear indication of the origins of organic matter to the fjords and bays of northern
Newfoundland.
The marine continental shelf environment off northern Newfoundland is a complex system in which
organic matter is derived from predominantly three sources and is subject to intense water column
cycling. A proper delineation of the relative contributions of each source to a sample in these
environments can be best accomplished by consideration of isotopic variations of end members
regionally and seasonally. Processes altering the isotopic and elemental composition of marine
seston can differ widely between sampling stations and times. In light of this, it is important to
consider the data from each collection in continental shelf areas separately, for biogeochemical
processes are subject to wide spatial and seasonal variation.
Acknowledgments
We thank the officers and crews of the FRV Wilfred Templeman, CSS Dawson, and CSS Baffin.
This manuscript benefited from the comments of two anonymous reviewers.
References
Altabet, M. A. 1988. Variations in nitrogen isotopic composition between sinking and

suspended particles: implications for nitrogen cycling and particle transformation in the open
ocean. Deep Sea Res. 35:535-554.
Cheng, H. H., J. M. Bremner, and A. P. Edwards. 1964. Variations of nitrogen-15 abundance

in soils. Science 146:1574-1575.
Cline, J. D. and I. R. Kaplan. 1975. Isotopic fractionation of dissolved nitrate during

denitrification in the eastern tropical North Pacific Ocean. Mar. Chem. 3:271-299.
Degens, E. T. 1969. Biogeochemistry of stable carbon isotopes. In G. Eglinton and M. T. J.

Murphy, eds., Organic Geochemistry, pp. 304-329. New York: Springer Verlag.
Degens, E. T., R. R. L. Guillard, W. M. Sackett, and J. A. Hellebust. 1968. Metabolic

fractionation of carbon isotopes in marine phytoplankton I. Temperature and respiration
experiments. Deep Sea Res. 15:1-9.
Deines, P. 1980. The isotopic composition of reduced organic carbon. In P. Fritz and J. Ch.
Fontes, eds., Handbook of Environmental Isotope Geochemistry, 1:329-406. New York:
Elsevier.
Delwiche, C. C. and P. L. Steyn. 1970. Nitrogen isotope fractionation in soils and microbial

reactions. Environ. Sci. Technol. 4:929-935.
DeNiro, M. J. and S. Epstein. 1981. Influence of diet on the distribution of nitrogen isotopes
in animals. Geochim. Cosmochim. Acta 45:341-351.
Deuser, W. G. 1970. Isotopic evidence for diminishing supply of available carbon during
diatom bloom in the Black Sea. Nature 225:1069-1071.
Deuser, W. G. and E. T. Degens. 1967. Carbon isotope fractionation in the system CO2
(gas)-CO2 (aqueous)-HCO3-(aqueous). Nature 215:1033-1035.
Dunton, K. H. and D. M. Schell. 1987. Dependence of consumers on macroalgal (Laminaria

solidungula) carbon in an arctic kelp community: 13C evidence. Mar. Biol. 93:615-625.
Eadie, B. J. and L. M. Jeffrey. 1973. 13C analyses of oceanic particulate organic matter. Mar.
Chem. 1:199-209.
Entzeroth, L. C. 1982. Particulate matter and organic sedimentation on the continental shelf
and slope of the northwest Gulf of Mexico. Ph.D. thesis, Univ. of Texas, Austin, Texas.
Fry, B., S. A. Macko, and J. C. Zieman. 1987. Review of stable isotope investigations of food
webs in seagrass meadows. Florida Marine Res. Publ. no. 42, pp. 190-209.
Goodell, H. G. 1972. Carbon/nitrogen ratio. In R. W. Fairbridge, ed., Encyclopedia of

Geochemistry and Environmental Science, pp. 136-142. New York: Van Nostrand Reinhold.
Gordon, D. C. Jr. 1971. Distribution of particulate organic carbon and nitrogen at an oceanic
station in the Central Pacific. Deep Sea Res. 18:1127-1134.
Harrigan, P., J. C. Zieman, and S. A. Macko. 1989. The base of nutritional support for the
Gray Snapper (Lutjanus griseus): an evaluation based on a combined stomach content and
stable isotope analysis. Bull. Mar. Sci. 44:65-77.
Hoering, T. and H. T. Ford. 1960. The isotope effect in the fixation of nitrogen by
Azotobacter. J. Am. Chem. Soc. 82:376-378.
Jeffrey, A. W. A., R. C. Pflaum, J. M. Brooks, and W. M. Sackett. 1983. Vertical trends in

particulate organic carbon 13C:12C ratios in the upper water column. Deep Sea Res. 9A:971-
983.
Libes, S. M. and W. G. Deuser. 1988. The isotope geochemistry of particulate nitrogen in the
Peru Upwelling Area and the Gulf of Maine. Deep Sea Res. 35:517-533.
Liu, K. -K. 1979. Geochemistry of inorganic nitrogen compounds in two marine

environments: the Santa Barbara Basin and the ocean off Peru. Ph.D. dissertation, Univ. of
California, Los Angeles.

Macko, S. A. 1981. Stable nitrogen isotope ratios as tracers of organic geochemical

processes. Ph.D. dissertation, Univ. of Texas, Austin.
Macko, S. A. 1983. Source of organic nitrogen in Mid-Atlantic coastal bays and continental
shelf sediments of the United States: Isotopic evidence. Carnegie Inst. Wash. Year Book
82:390-394.
Macko, S. A., M. L. Fogel (Estep), P. E. Hare, and T. C. Hoering. 1987. Isotopic fractionation
of nitrogen and carbon in the synthesis of amino acids by microorganisms. Chem. Geol.
65:79-92.
Macko, S. A., K. Pulchan, and D. E. Ivany. 1987. Organic geochemistry of Baffin Island
fjords. Sedimentology of Arctic fjords experiment Geol. Surv. Can., file 1589, pp. 13-1-13-34.
Mariotti, A., C. Lancelot, and G. Billen. 1984. Natural isotopic composition of nitrogen as a
tracer of origin for suspended matter in the Scheldt estuary. Geochim. Cosmochim. Acta
48:549-555.
Mayer, L. M., S. A. Macko, and L. Cammen. 1988. Provenance, concentration and nature of
sedimentary organic nitrogen in the Gulf of Maine. Mar. Chem. 25:291-304.
McCave, I. N. 1975. Vertical flux of particulates in the ocean. Deep Sea Res. 22:491-502.
McConnaughey, T. and C. P. McRoy. 1979. Food web structure and the fractionation of
carbon isotopes in the Bering Sea. Mar. Biol. 53:257-262.
Minigawa, M. and E. Wada. 1984. Stepwise enrichments of 15N along food chains: further
evidence and the relation between 15N and animal age. Geochim. Cosmochim. Acta
48:1135-1140.
Miyake, Y. and E. Wada. 1967. The abundance ratio of 15N/14N in marine environments.
Rec. Oceanogr. Works Japan 9:37-53.
Muller, P. J. 1977. C/N ratios in Pacific deep-sea sediments: effect of inorganic ammonium
and organic nitrogen compounds sorbed by clays. Geochim. Cosmochim. Acta 41:765-776.
O'Leary, M. H. 1981. Carbon isotope fractionation in plants. Phytochem. 20:553-567.
Pulchan, K. 1987. Organic geochemical comparisons of Fortune Bay and Bay d'Espoir. M.
Sc. thesis, Memorial Univ. of Newfoundland, St. Johns, Newfoundland.
Rau, G. H., R. E. Sweeney, and I. R. Kaplan. 1982. Plankton 13C:12C ratio changes with
latitude: differences between northern and southern oceans. Deep Sea Res. 29:1035-1039.
Sackett, W. M., W. R. Eckelmann, M. L. Bender, and A. W. Be. 1965. Temperature

dependence of carbon isotope composition in marine plankton and sediment samples.

Science 148:235-237.
Saino, T. and A. Hattori. 1980. 15N natural abundance in oceanic suspended particulate
matter. Nature 283:752-754.
Saino, T. and A. Hattori. 1985. Variation of 15N natural abundance of suspended organic
matter in shallow oceanic waters. In A. C. Sigleo and A. Hattori, eds., Marine and Estuarine
Geochemistry, pp. 1-13. Chelsea, Mich.: Lewis.
Saino, T. and A. Hattori. 1987. Geographical variation of the water column distribution of
suspended particulate organic nitrogen and its 15N natural abundance in the Pacific Ocean
and its marginal seas. Deep Sea Res. 34:807-827.
Stephenson, R. L., F. C. Tan, and K. H. Mann. 1984. Isotopic variability in marine

macrophytes and its implications to food web studies. Mar. Biol. 81:223-230.
Sweeney, R. E., K. K. Liu, and I. R. Kaplan. 1976. Oceanic nitrogen isotopes and their uses
in determining the source of sedimentary nitrogen. In Stable Isotope Geochemistry. Proc. Int.
Symp., New Zealand Dept, Sci. Ind. Res.
Tan, F. C. and P. M. Strain. 1979. Carbon isotope ratios of particulate organic matter in the
Gulf of St. Lawrence. J. Fish. Res. Board Can. 36:678-682.
Tan, F. C. and P. M. Strain. 1983. Sources, sinks and distribution of organic carbon in the St.
Lawrence Estuary, Canada. Geochim. Cosmochim. Acta 47:125-132.
Tenore, K. R. 1983. What controls the availability to animals of detritus derived from vascular
plants: organic nitrogen enrichment or caloric availability? Mar. Ecol. Prog, series 10, pp.
307-309.
Troke, C. G. 1987. Organic geochemistry of three Newfoundland Bays: Notre Dame Bay-
Trinity Bay-White Bay. B.Sc. thesis, Memorial Univ. of Newfoundland, St. John's,
Newfoundland.
Wada, E. 1980. Nitrogen isotope fractionation and its significance in biogeochemical

processes occurring in marine environments. In E. D. Goldberg, Y. Horibe, and J. K.
Saruhashi, eds., Isotope marine chemistry, pp. 375-398. Tokyo: Uchida Rokakuho.
Wada, E., M. Minigawa, H. Mizutani, T. Tsuji, R. Imaizumi, and K. Karasawa. 1987.

Biogeochemical studies on the transport of organic matter along the Otsuchi River
Watershed, Japan. Est. Coast. Shelf Sci. 25:321-336.



5. Organic Matter Accumulation, Remineralization, and

Burialin an Anoxic Coastal Sediment
Measurements of the fluxes of organic matter and its microbial decomposition end products CO2,
CH4, and dissolved organic carbon (DOC) have been used to establish a quantitative carbon
budget for the anoxic sediments of Cape Lookout Bight, N.C. Of the total incoming particulate
organic carbon (POC) flux of 165 20 mole C/m2yr, 29 5% is decomposed to CO2, CH4, and
DOC largely in the upper 50 cm, or within five years at an accumulation rate of slightly over 10 cm/
yr. Changes in POC and total N concentrations during degradation reveal that the decomposed
organic matter has an atomic C/N ratio of 6.6, resembling that of fresh diatoms, dinoflagellates, and
bacteria. We can account for 64 10% of the carbon lost during decomposition as amino acids (31
14%), lipids (17 7%), and sugars (16 7%); these proportions of biochemicals also resemble
those found in fresh plankton.
The 71% of total incoming POC, which survives degradation during early diagenesis and is
subsequently buried, has an average age of more than 500 years old and appears to be composed
primarily of heavily degraded materials of algal origin with a lesser component of vascular plant
tissue associated with fine-sized sediment particles. This older refractory organic matter has
presumably been through numerous cycles of accumulation and resuspension in surrounding
coastal environments.
The degradation of organic materials in nearshore sediments controls or influences numerous

important biogeochemical processes, including oxidation-reduction reactions and sediment-water
chemical exchange. Materials deposited in fine-grained, organic-rich environments sequentially
pass through a series of redox zones whose chemical compositions are controlled largely by
microbial respiration reactions. The reactions in these zones are fueled by the reactive or
"metabolizable" fraction of the total flux of organic molecules. Our understanding of the coupling of
degradation of organic matter with specific oxidants in the sediment column as well as the role of
specific microbial communities involved has been advanced considerably through the development
of biogeochemical zonation models that predict electron acceptor utilization in the order O2, NO-3,
MnO2, FeOOH, SO42- and CO2 (e.g., Claypool and Kaplan 1974; Martens 1978; Froelich et al.
1979). These models follow earlier thermodynamic predictions (e.g., Stumm and Morgan 1970;
Thorstensen 1970; McCarty 1972; Thauer, Jungermann, and Decker 1977).
Significant progress in measuring rates of degradation processes and resulting chemical fluxes has
resulted from kinetic modeling of specific early diagenetic reactions (e.g., Berner 1974, 1980) and
from in situ rate and flux measurements (see review by Henrichs and Reeburgh 1987). The power
of combining direct flux and rate measurements with such modeling is illustrated in figure 5.1, taken
from a study of anoxic sulfate-reducing and methane-producing sediments by Martens and Klump
(1984). This latter study sought to demonstrate that quantitative knowledge of the fluxes of organic
matter and the primary end products of decomposition (e.g., CO2 and CH4) could be combined
file:///F|/Usuarios/Juan%20carlos/articulos/articulos...a/organic%20matter/Organic%20Matter%20Chapter%205.htm (1 de 9)17/01/2006 06:59:27 p.m.

with kinetic modeling and rate measurements to better describe major degradation pathways as
well as the metabolizable fraction of total organic matter driving the processes involved.
In the present paper we seek to answer a series of questions concerning the composition and fate
of organic matter deposited in the well-characterized coastal sediments of Cape Lookout Bight
located on the Outer Banks of North Carolina (figure 5.2). The approaches illustrated in figure 5.1
are utilized in addressing these questions. The answers are of interest for understanding early
diagenetic processes occurring in any organic-rich sediment.
Study Site and Experimental Methods
The Cape Lookout Bight field site (figure 5.2) has been described in detail elsewhere (e.g., Martens
and Klump 1980, 1984; Haddad and Martens 1987). Muddy sediment found in the interior of the
bight has accumulated at a steady-state rate of 10.3 1.7 cm · yr-1 (4.13 0.67 g dry sed · cm-2 ·
yr-1) for more than two decades (Chanton, Martens, and Kipphut 1983; Canuel, Martens, and
Benninger 1989). Most of the sediment and its organic matter are derived from the backbarrier
island lagoons and marshes landward of the bight and arrive through Barden Inlet. Inorganic and
organic materials delivered to Cape Lookout have thus been exposed to extensive physical and
biological recycling processes within the lagoons. The steady-state sediment supply results in part
from regular delivery via tidal mixing through Barden Inlet.
The anoxic sediments at station A-1 (figure 5.2) are dominated by sulfate reduction and methane
production on a year-round basis (Crill and Martens 1983, 1987; Martens and Klump 1984).
Macroinfauna in the bight interior are limited to small spionid and a few capitellid polychaetes, which
disappear during summer months (Bartlett 1982).
The approaches to quantifying carbon cycling illustrated in figure 5.1 include (1) direct
measurements of fluxes, (2) kinetic modeling of sedimentary organic carbon concentration
distributions, and (3) direct measurements of the rates of in situ degradation processes in the
sediment column. The reader is referred to detailed descriptions of the laboratory and field methods
used to study carbon, sulfur, and nitrogen and phosphorus cycling in Cape Lookout sediments
appearing in Martens and Klump (1984), Chanton, Martens, and Goldhaber (1987) and Klump and
Martens (1987) respectively. On the bases of the radiochronometric sediment accumulation rate
studies mentioned above, on annually reproducible pore water chemical distributions observed
since 1975, and on downcore lignin oxidation product analyses, Haddad and Martens (1987) have
concluded that organic matter accumulation in Cape Lookout Bight has been steady state since at
least 1971. Thus systematic downcore decreases in labile organic matter must result from early
diagenesis rather than from input variations.
In this paper we focus on using the carbon budget for Cape Lookout sediments at station A-1 to
quantify the metabolizable organic matter fraction undergoing remineralization. New data obtained
to achieve these objectives include additional measurements of seasonal variation in the ebullitive
methane gas flux, estimated dissolved organic carbon (DOC) fluxes based on pore water
concentration gradients (M. Alperin, unpublished data), and a detailed seasonal series of
sedimentary organic carbon and nitrogen concentration profiles. In addition, changes in the
concentration of specific organic compound classes (Burdige and Martens 1988; Haddad 1989) and
the radiocarbon activity of organic carbon in the upper 2 meters of the sediment column as well as
methane escaping the sediment have been determined.

Samples for organic carbon and nitrogen as well as radiocarbon analyses were pretreated to
remove carbonates by means of a cold HCl filtration method described by Martens and Klump
(1984). Elemental C and N analyses were made using a Carlo Erba model NA 1500 analyzer.
Determinations of 13C-corrected 14C values were made by Beta Analytic Corporation on our
pretreated samples by means of conventional liquid scintillation benzene counting. Radiocarbon
measurements were also made on methane in two gas bubble samples collected during summer
months. Total hydrolyzable amino acid concentrations were taken from Burdige and Martens
(1988). Concentrations of total hydrolyzable sugars and total extractable lipids were taken from
Haddad (1989).
Results from our investigations of biogeochemical cycling processes in Cape Lookout Bight can be
utilized to address four questions discussed in detail below.
Fraction of Accumulated Organic Matter Remineralized
Under the steady-state sediment accumulation conditions observed in Cape Lookout Bight what
fraction of the incoming POC (particulate organic carbon) is remineralized during early diagenesis?
In order to answer this question we have applied the following mass balance equation, which is
based on steady-state organic matter input and remineralization rates observed since the mid-
1970s (see Martens and Klump 1984; Haddad and Martens 1987):
POC input = POC remineralized + POC buried (5.1)
In practice it has proven easiest to measure fluxes resulting from POC remineralization and burial
and then to calculate POC input by adding these fluxes together. The fluxes are listed in table 5.1
and illustrated in figure 5.3. In this model the incoming POC is either remineralized to dissolved
CO2, CH4, and DOC or buried. The CO2, CH4, and DOC produced during remineralization are
either lost to the water column via sediment-water chemical exchange or buried as a dissolved
component of sediment pore waters. The POC burial rate was found to be 117 19 mole C · m-2yr-
1 by Martens and Klump (1984) using the 210Pb-determined 4.13 0.67 g dry sed · cm-2 · yr-1
sediment accumulation rate of Chanton, Martens, and Kipphut (1983). Sediment-water chemical
exchange accounts for losses of 40.6 5.6 mole C · m-2 · yr-1 as CO2, CH4, and DOC, whereas
7.0 1.1 mole C · m-2 · yr-1 of these dissolved species is buried (figure 5.3). These dissolved
sediment-water exchange and buried fluxes sum to a total POC remineralization rate of 47.6 5.7
mole C · m-2 · yr-1. When the latter value is added to the POC burial rate, a total POC input of 165
20 mole C · m-2 · yr-1 can be calculated via equation 5.1 above. The fraction of this POC input
remineralized is simply calculated as:
[POC remineralized - POC input] 102 = fraction remineralized (5.2)
For station A-1 the fraction of incoming POC remineralized is 29 5% with the flux values given
above.

Composition of POC Remineralized
Which biochemical components of particulate organic carbon are being remineralized and thus are
driving the observed dissolved carbon fluxes?
Important information about the nature of organic matter fueling microbial remineralization
processes in Cape Lookout sediments can be gained from a quantitative knowledge of its atomic C/
N ratio (C/N)a. Profiles of POC and TN (total particulate nitrogen) concentration versus mass depth
(g · cm-2) from a series of sediment cores collected over a one-year period are plotted in figures 5.4
and 5.5. The concentration data are calculated on a carbonate-free basis. Each plot is fitted with an
exponential equation that allows the attenuation coefficient, interface concentration, and asymptotic
concentration at depth (C ) to float. Reaching a constant C value implies that no further
degradation is occurring. The mean depth at which C occurs for the data illustrated in figures 5.4
and 5.5 is 4.4 g · cm-2 or 15 cm at an average porosity of 0.85. Direct measurements of in situ
sulfate reduction (Crill and Martens 1987) and methane production rates (Crill and Martens 1983)
confirm these model results. Degradation reactions are essentially complete by 40 to 50 cm depth
(13.2 to 16.8 g · cm-2 mass depth).
The weight ratio of the remineralized fraction of the total deposited POC can be deduced from the
POC and TN profiles by simply plotting C versus N values for each core at all depths above C .
The slope of the plot for each core yields this C/N weight ratio (see linear regression fits in figure
5.6). Multiplication of these weight ratios by 14/12 yields atomic C/N ratios (C/N)a for each core,
which are summarized in table 5.2.
The significance of these plots may initially be unclear because of their potential sensitivity to any
factors that may cause changes in POC and TN data calculated on a weight percent basis. For
example, dilution of organic matter by nonorganic material would result in POC/ TN relationships
similar to those seen in figure 5.6. One would expect dilution-controlled slopes to extrapolate
through the origin of the plot. Likewise, any input pulses of organic-rich material must be considered
in a manner analogous, albeit opposite, to the dilution effects. In Cape Lookout sediments, dilution
can be directly evaluated by use of sediment grain size distribution data reported by Chanton,
Martens, and Kipphut (1983). These authors demonstrated that sediments above the 1971
Hurricane Ginger level (approximately 120 cm depth in these cores) were relatively constant in
terms of their weight percent sand: silt: clay ratio. On the bases of depth of cores used in this study
(<30 cm) and on visual examination of the cores themselves, we conclude that the slopes derived
from linear regression analyses of the weight percent POC versus weight percent TN plots shown in
figure 5.6 accurately reflect the POC/ TN of the particulate organic matter being utilized to derive
microbial reactions. When calculated on an atomic basis, the average (C/N)a ratio of remineralized
organic matter determined for the eight cores is 6.6 (table 5.2). This value is equal to the Redfield
(C/N)a ratio of 6.6 (106/16) and is also in good agreement with a (C/N)a ratio of 7.2 for
remineralized organic matter determined over the same depth range at station A-1 from porewater
CO2, NH4, and reactive phosphate stoichiometry by Klump and Martens (1987). We conclude that
the organic matter undergoing decomposition and fueling the microbial remineralization process
resembles fresh plankton in its (C/N)a ratio.
Detailed studies by Burdige and Martens (1988) and Haddad (1989) provide key information

concerning downcore degradative losses of specific biochemical components. Changes in the

carbon concentrations of total hydrolyzable amino acids and sugars plus total extractable lipids
between 0 and 5 cm and 95 and 100 cm depths in Cape Lookout sediments (figure 5.7) illustrate
this degradation. By adding the downcore losses (table 5.3) of these three compound classes
calculated by Haddad (1989), we can account for 64 10% of the 47 5.7 mole C · m-2 · yr-1
(figure 5.3) remineralized. Losses of total hydrolyzable amino acids account for 31 14%, while
total extractable lipids and total hydrolyzable sugars account for 17 7 and 16 7% respectively.
We cannot presently account for approximately 36% of the observed total organic carbon loss in
terms of specific organic compounds.
The losses of each biochemical fraction (on a mole:mole basis) in Cape Lookout sediments can be
compared with the biochemical distribution measured in specific organisms. table 5.4 compares the
amino acid:lipid:carbohydrate ratios calculated from the downcore sedimentary losses of these
biochemicals with those determined for bacteria, diatoms, and dinoflagellates. We conclude that the
proportions of biochemicals remineralized in Cape Lookout sediments are similar to those found in
these organisms. Also included in table 5.4 are the ratios of these biochemicals in average woody
and nonwoody vascular plant tissue. These biochemical comparisons also support our earlier
conclusion based on bulk (C/N)a ratios that the organic matter driving sedimentary microbial
reactions in Cape Lookout Bight appears to be of algal or bacterial origin.
Composition of Organic Matter Surviving Early Diagenesis
What is the composition of organic matter surviving the initial ten years of early diagenesis in this
sediment?
The POC surviving early diagenesis appears to be unreactive at least under in situ conditions. The
addition of oxidants such as dissolved sulfate to deeper sediment below 100 cm depth does not
stimulate any measurable further degradation reactions (Haddad 1989). Addition of sulfate plus
reactive organic acids results in degradation of the latter with 2 moles of CO2 produced per mole
of sulfate reduced. This result indicates that the buried organic matter is refractory under in situ
conditions.
Chemical characteristics of organic matter buried in Cape Lookout sediments below the upper 100
cm where most degradation occurs can be utilized to answer this question. These characteristics
can be summarized as follows:
(1) atomic C/N ratio higher than 10, (2) relatively heavy 13C ratio, (3) low aromaticity and
carbohydrate content in dominant humic substances fraction, and (4) low vascular plant tissue
content.
The atomic C/N ratio (C/N)a of buried organic matter (figure 5.8) was obtained from the POC and
TN data for each of the cores in figures 5.4 and 5.5. This ratio consistently approaches a value of
9.5-12 at depth whereas that of degrading material averages 6.6 as discussed above. The 13C
value of this buried organic matter is between -19 and -20 (Blair, Martens, and Des Marais 1987;
figure 5.9), indicating an algal origin or a consistent input of the same mix of C-3- and C-4- (O'Leary
1981) derived materials over the past decade. Unpublished CPMAS 13C NMR Spectra of Haddad
(1989) on the predominant (>80%) humic substances sediment fraction reveal low aromaticity in

buried organic materials consistent with plankton-derived organic matter. The NMR data also show
that little carbohydrate survives early diagenesis to be buried at depth. Lignin oxidation product
(LOP) analysis of the sediment (Haddad and Martens 1987) indicates that only 23 17% of the
buried organic matter is of nonwoody vascular plant origin.
Taken as independent pieces of information, none of the results presented above can be used to
uniquely characterize the origin of the buried organic matter. For example, as suggested above, the
stable carbon isotopic composition of the buried organic matter can be derived from a mixture of
marine C-3 plants and terrestrial C-3 and C-4 plants. Likewise, Haddad and Martens (1987) have
discussed the drawbacks to determining vascular plant tissue sources based solely on sedimentary
LOP results. However, when the various data are considered as a whole, they provide an internally
consistant pattern indicating that organic matter surviving early diagenesis is composed primarily of
heavily degraded algal debris mixed with a smaller component of primarily nonwoody vascular plant
tissue.
Age of Buried Organic Matter Surviving Early Diagenesis
How old is the mixture of organic matter that survives early diagenesis to be buried?
Radiocarbon 14C values and age of POC carbon versus depth are plotted in figure 5.10. Organic
carbon in the upper meter of sediment ranges from 500 to 1000 years in age. Organic carbon below
150 cm depth is more than 1500 years old. The deeper sediment has a high sand content indicating
its inner continental shelf origins prior to enclosure of Cape Lookout Bight (see Chanton, Martens,
and Kipphut 1983). We conclude that the 71% of the initially deposited POC surviving early
diagenesis to be buried (figure 5.3) is old organic matter which has probably been recycled in
surrounding coastal environments for hundreds of years. However, we hypothesize that the 29
5% of remineralized organic matter, which resembles freshly produced plankton in its composition
(table 5.2) and the end products produced during its decomposition, CO2, CH4, and DOC, should
be enriched in bomb radiocarbon.
Although recently produced POC deposited at our site may be enriched in radiocarbon, it is
indistinguishable from the total pool of organic carbon in surficial sediments because of the
presence of a much larger component of older, recycled materials as discussed above. The
dominance of this latter, more recalcitrant fraction is suggested by the high proportion of incoming
organic matter surviving early diagenesis, as indicated by the POC concentration profiles illustrated
in figure 5.4.
However, we have been able to obtain radiocarbon values for methane, one of the principal
decomposition end products (figure 5.3). Methane carbon 14C values obtained from two gas
samples by conventional benzene counting are + 91.5 7 and + 105 7 . These values translate
to a 109 to 110% enrichment of radiocarbon over prebomb levels in methane produced during early
diagenesis. This must result from bomb radiocarbon inputs to recently produced (e.g., since 1955)
plant tissue, which has subsequently been deposited in Cape Lookout sediments and then
decomposed, in part, to methane.
The total input rate of POC to the sediment at our Cape Lookout site is 165 20 mole · m-2 · yr-1.
Of this POC input, 29 5% is remineralized within the upper 50 cm or within 4-5 years at a

sediment accumulation rate of more than cm · yr-1.
The incoming POC appears to consist of two primary components: (1) a freshly produced and easily
degraded component resembling diatoms dinoflagellates, and bacteria in composition; (2) a larger
component of old and heavily degraded organic matter of "algal" origin. This latter component
arrives in a mixture with the freshly produced materials but survives early diagenesis and is buried.
The major implication of the work discussed above is that only a modest fraction of the total organic
matter deposited in rapidly accumulating coastal sediments can be expected to be fresh and readily
metabolizable. This metabolizable fraction will be decomposed rapidly following the now well-
defined sequence of respiration reactions. Moreover, we can expect most of this freshly produced
metabolizable fraction to be consumed regardless of which oxidants and associated microbial
communities are present.
Acknowledgments
We thank Susan Boehme and Neal Blair of North Carolina State University for bringing to our
attention a systematic error in our dissolved inorganic carbon measurements at high concentrations
and the latter for use of unpublished organic carbon 13C data. We thank Dana Reinhold and Steve
Dougherty for meticulous care in elemental C and N sample preparation and analysis. We thank
Jerry Stipp, Murray Tamers, Darden Hood, and their colleagues at the University of Miami and Beta
Analytic, Inc. for cooperative methodological investigations needed to obtain reproducible
radiocarbon results on low-carbon sediment samples and gaseous methane.
Laboratory and boating facilities of the University of North Carolina's Institute of Marine Sciences,
Morehead City, were utilized extensively for this study. Research was supported by NSF Grants
OCE 84-16963, OCE 87-16528, and OCE 90-17979, Marine Chemistry Program and by NASA
Grants NAGW-1455 and NAGW-834, Biospherics Research and Interdisciplinary Research
Programs.
References
Bartlett, K. B. 1982. Macrofauna distribution and seasonal influences on interstitial water

chemistry of Cape Lookout Bight, N.C. M.S. thesis, Univ. of North Carolina, Chapel Hill.
Berner, R. A. 1974. Kinetic models for the early diagenesis of nitrogen, sulfur, phosphorus
and silicon in anoxic marine sediments. In E. D. Goldberg, ed., The Sea, 5:427-450. New
York: Wiley.
Berner, R. A. 1980. Early Diagenesis. A Theoretical Approach. Princeton, N.J.: Princeton

Univ. Press.
Blair, N. E., C. S. Martens, and D. J. Des Marais. 1987. Natural abundances of carbon
isotopes in acetate from a coastal marine sediment. Science 236:66-68.
Burdige, D. J. and C. S. Martens. 1988. Biogeochemical cycling in an organic-rich coastal

marine basin-10. The role of amino acids in sedimentary carbon and nitrogen cycling.

Canuel, E. A., C. S. Martens and L. K. Benninger. 1990. Seasonal variations in beryllium-7

activity in the sediments of Cape Lookout Bight, N.C. Geochim. Cosmochim. Acta 54:237-
245.
Chanton, J. P., C. S. Martens, and G. W. Kipphut. 1983. Lead-210 sediment geochronology

in a changing coastal environment. Geochim. Cosmochim. Acta 47:1791-1804.
Chanton, J. P., C. S. Martens and M. B. Goldhaber. 1987. Biogeochemical cycling in an

organic-rich coastal marine basin-7. Sulfur mass balance, oxygen uptake and sulfide
retention. Geochim. Cosmochim. Acta 51:1187-1199.
Claypool, G. and I. R. Kaplan. 1974. The origin and distribution of methane in marine
sediments. In I. R. Kaplan, ed., Natural Gases in Marine Sediments, pp. 99-139. New York:
Plenum.
Crill, P. M. and C. S. Martens. 1983. Spatial and temporal fluctuations of methane production
in anoxic coastal marine sediments. Limnol. Oceanogr. 28:1117-1130.
Crill, P. M. and C. S. Martens. 1987. Biogeochemical cycling in an organic-rich coastal

marine basin-6. Temporal and spatial variations in sulfate reduction rates. Geochim.
Froelich, P. N., G. P. Klinkhammer, M. L. Bender, N. A. Luedtke, G. R. Heath, D. Cullen, P.

Dauphin, D. Hammond, B. Hartmann, and V. Maynard. 1979. Early oxidation of organic
matter in pelagic sediments of eastern equatorial Atlantic: suboxic diagenesis. Geochim.
Galimov, E. M. 1985. The Biological Fractionation of Isotopes. New York: Academic.
Haddad, R. I. 1989. Sources and reactivity of organic matter accumulating in a rapidly

depositing, coastal marine sediment. Ph.D. dissertation, Univ. of North Carolina, Chapel Hill.
Haddad, R. I. and C. S. Martens. 1987. Biogeochemical cycling in an organic-rich coastal

marine basin-9. Sources and fluxes of vascular plant derived organic material. Geochim.
Henrichs, S. M. and W. S. Reeburgh. 1987. Anaerobic mineralization of marine sediment

organic matter: Rates and the role of anaerobic processes in the oceanic carbon economy.
Geomicrobiol. J. 5:191-237.
Klump, J. V. and C. S. Martens. 1987. Biogeochemical cycling in an organic-rich coastal

marine basin-5. Sedimentary nitrogen and phosphorus budgets based upon kinetic models,
mass balances and the stoichiometry of nutrient regeneration. Geochim. Cosmochim. Acta
51:1161-1173.

Martens, C. S. 1978. Some of the chemical consequences of microbially mediated

degradation of organic materials in estuarine sediments. In E. D. Goldberg, ed.,
Biogeochemistry of Estuarine Sediments, pp. 266-278. Paris: UNESCO.
Martens, C. S. and J. V. Klump. 1980. Biogeochemical cycling in an organic-rich coastal

marine basin-1. Methane sediment-water exchange processes. Geochim. Cosmochim. Acta
44:471-490.
Martens, C. S. and J. V. Klump, 1984. Biogeochemical cycling in an organic-rich coastal

marine basin-4. An organic carbon budget for sediments dominated by sulfate reduction and
methanogenesis. Geochim. Cosmochim. Acta 48:1987-2004.
McCarty, P. L. 1972. Energetics of organic matter degradation. In R. Mitchell, ed., Water

Pollution Microbiology, pp. 91-108. New York: Wiley.
O'Leary, M. H. 1981. Carbon isotope fractionation in plants. Phytochem. 20:553-567.
Parsons, T. R., M. Takahashi, and B. Hargrave. 1977. Biological Oceanographic Processes.

2d ed. New York: Pergamon.
Stumm, W. and J. J. Morgan. 1970. Aquatic Chemistry. New York: Wiley.
Thauer, R. K., K. Jungermann, and K. Decker. 1977. Energy conservation in chemotrophic

anaerobic bacteria. Bacteriol. Rev. 41:100-180.
Thorstensen, D. C., 1970. Equilibrium distribution of small organic molecules in natural

waters. Geochim. Cosmochim. Acta 34:745-770.


6. Organic Carbon Remineralization and Preservation in

Sediments of Skan Bay, Alaska
A carbon budget for the sediments of Skan Bay, Alaska, was determined for better understanding of
the factors that control organic matter remineralization and preservation. The carbon budget was
constructed from (1) concentration-depth distributions for five major carbon reservoirs (particulate
organic carbon [POC], dissolved organic carbon [DOC], dissolved inorganic carbon [DIC], methane
[CH4], and particulate inorganic carbon [PIC]); (2) sulfate reduction and methane oxidation rate
profiles; (3) alkalinity, nitrate, and CH4 benthic fluxes; and (4) 210Pb and 137Cs geochronologies.
The total sediment remineralization rate was calculated by three independent approaches: (1) the
difference between POC deposition and preservation, (2) the quantity of carbon recycled to the
water column and buried at depth, and (3) depth-integrated rates of bacterial metabolism. The three
estimates agree within 10%, indicating that the carbon budget is tightly constrained. The budget
indicates that 84 3% of the organic carbon deposition is remineralized in the upper meter of the
sediment column. Nearly all of the remineralized carbon is recycled to the water column; less than
5% is buried as DIC, DOC, and CH4. DOC comprises a significant fraction ( S17%) of the benthic
carbon flux in this system. Stable isotope and atomic C:N ratios in the sediment POC suggest that
phytoplankton and kelp are the principal sources of organic matter to Skan Bay sediments.
Systematic down core variations in stable isotope ratios of the particulate organic carbon are
consistent with preferential remineralization of organic matter derived from kelp.
Continental margin sediments represent a mere 10% of the sea floor but play a significant role in
the global carbon budget. Because of high water column productivity and shallow depth, coastal
sediments receive 85% of the ocean's organic matter deposition (Romankevich 1984). Sediments
from nearshore environments are biologically active, with organic carbon remineralization rates one
to two orders of magnitude higher than deep-sea sediments (Emerson 1985; Henrichs and
Reeburgh 1987). The continental margins experience high sedimentation rates, leading to burial
and preservation of large quantities of organic carbon (Berner 1982). Coastal marine sediments
clearly represent important sites of organic matter deposition, remineralization, and preservation.
The carbon cycle in nonbioturbated coastal sediments is presented schematically in figure 6.1. The
cycle is fueled by deposition of particulate organic carbon (POC) of aquatic or terrestrial origin.
Following deposition, POC faces one of two fates: remineralization or preservation. Organic matter
remineralization is a complex, multistep process mediated primarily by sediment bacteria. POC is
initially converted to dissolved organic carbon (DOC), a portion of which is oxidized to dissolved
inorganic carbon (DIC). When sediments become depleted of electron acceptors (e.g., O2, NO3-,
and SO42-), DOC is fermented to DIC and methane (CH4). The CH4 can diffuse upward to
sediments containing sulfate where it is oxidized to DIC. The products of organic matter
remineralization (DOC, DIC, and CH4) accumulate in the pore water and are recycled to the water
column via diffusion or buried as sediment accumulates.

Organic matter remineralization often leads to decreasing organic carbon concentrations with
sediment depth. POC concentrations generally do not reach zero but approach an asymptote of
approximately 50 to 70% of the concentration at the sediment surface (Emerson and Hedges 1988).
The remineralized organic matter is referred to as "biologically available," while the remainder is
considered to be "refractory."
The factors that determine whether sediment organic matter is remineralized or preserved are
poorly understood. In this paper, we present concentration, reaction rate, benthic flux, and stable
carbon isotope data for sediments of Skan Bay, Alaska. These data provide a tightly constrained
carbon budget that allows us to quantify what fraction of deposited organic matter is remineralized.
In addition, changes in the isotopic composition of sediment POC provide a means of tracking the
fate of organic matter derived from different sources. This allows us to assess whether source plays
a role controlling the extent of organic matter remineralization.
Experimental Study Site
Skan Bay (57°N, 167°W) is a pristine embayment on the northwest side of Unalaska Island in the
Aleutian Islands (Hattori, Goering, and Boisseau 1978). The bay has a shallow sill (10 m) across the
inlet vegetated with kelp of the genus Alaria pistulosa (Shaw et al. 1984). The shallow sill limits
horizontal advection of oxygen-rich Bering Sea water into the basin. At the time of sampling, bottom
water was O2 depleted (<0.4 mL L-1) and sediments were sulfidic at the interface. Sediment X-
radiography reveals distinct layers, indicating that particles are not mixed by bioturbation. Diatom
tests and kelp fragments are abundant throughout the sediment column.
Skan Bay sediments represent a nearly isothermal environment; temperatures range annually from
1° to 4°C (Alperin 1988). The sediment surface is below the euphotic zone (65 m) and not directly
affected by sunlight.
Sediment Sampling
Sediment was obtained by use of three coring devices: a 30 x 30 cm box corer (Ocean Instruments,
Inc.) capable of sampling to a depth of 40 cm, an in situ benthic chamber (Devol 1987) containing
two 20 x 20 cm box corers capable of sampling to 20 cm, and a gravity corer (Benthos, Inc.)
capable of sampling to 1 meter. After box core retrieval, subcores were taken by gently inserting
clear plastic core liner into the sediment. Subcores showing any sign of disturbance (i.e., turbidity at
the sediment-water interface, CH4 ebullition, etc.) were discarded.
Concentrations
Sediment and bottom water CH4 concentrations were measured at sea by means of a headspace
technique (Alperin and Reeburgh 1985). Sediment cores were obtained by use of core liner with
tape-covered perforations at 3 cm intervals. The sediment was subsampled with tipless 3 mL
syringes by insertion of the syringe barrel into the sediment while the piston was held stationary.
The sample was immediately transferred to a 37.5 mL serum vial containing 5 mL distilled water,
sealed with a butyl rubber stopper, and thoroughly homogenized with a Vortex mixer. Methane was
quantified by gas-chromatography with flame ionization detection. Whole sediment CH4
concentrations were converted to pore water concentration units (mmol L-1 pore water). Bottom

water was collected from a depth of 63 m and analyzed for CH4 as described above.
Sediment DIC, DOC, POC, and PIC (particulate inorganic carbon) concentrations were measured in
the laboratory on samples stored frozen in 600 mL lacquered steel cans. Cores were sampled by
extruding the sediment upward into a 3 cm x 6.6 cm i.d. segment of core liner. A metal shim was
inserted beneath the segment and the sediment was quickly transferred to a N2-flushed can. The
can was immediately sealed and stored frozen until the time of analysis. A brief description of the
analytical methods is given below; complete details are available in Alperin (1988).
Pore water and solid matter were separated by pressure filtration (Reeburgh 1967). The pore water
was filtered through two precombusted 0.6 m glass fiber filters (Whatman EPM 1000) directly into
an N2-flushed reaction vessel. The Pyrex reaction vessel had vacuum o-ring stopcocks on the inlet
and outlet and a sintered-glass gas dispersion tube to accelerate the gas-stripping process.
Precautions were taken to prevent CO2 exchange between pore water and atmosphere.
The pore water was acidified (10 mL carbon-free 2N H3PO4) and DIC was stripped from solution
with ultrapure N2 carrier gas. The H2O vapor was retained in a trap cooled to -89°C and H2S was
trapped in a Pyrex U-tube packed with CuSO4-coated Chromosorb. The CO2 was collected in two
traps in series cooled to -196°C. When pore water stripping was complete, the noncondensible
gases were pumped away and the CO2 was quantified manometrically.
DOC was oxidized by a procedure similar to that of Wilson (1961). Potassium persulfate (Baker
"Instra-Analyzed") was ground to a fine powder and 5.0 g was transferred to the reaction vessel
containing the acidified, stripped pore water. The reaction vessel was sealed and heated to 100°C
for 30 min. A Teflon-coated magnetic stir bar mixed the solid oxidant into solution. Following the
oxidation, CO2 was stripped from solution and quantified manometrically as described above.
Chlorine, produced by Cl- oxidation, was trapped on a column packed with Cuprin (Coleman)
heated to 200°C. Bottom water collected from 10 cm above the sediment surface was analyzed for
DOC as described above.
There is considerable uncertainty regarding the efficacy of persulfate as an oxidant of DOC.

Sugimura and Suzuki (1988) have shown that 30% to 60% of the DOC in North Pacific seawater is
not detected by the persulfate method. They found persulfate to be an effective oxidant of low
molecular weight(<4000) organic compounds but incapable of oxidizing the high molecular weight
fraction. Unfortunately, the accuracy of the persulfate method for sediment DOC has not been
evaluated. Since water column and pore water organics differ both quantitatively and qualitatively,
the results of Sugimura and Suzuki (1988) cannot be directly applied to pore water DOC. Given this
uncertainty in the persulfate oxidation efficiency, DOC concentrations reported in this study should
be regarded as lower limits.
The sediment fraction containing solid material was dried and ground to a fine powder, and an
aliquot was analyzed for PIC by acidification and quantification of the resulting CO2. POC was
determined by combusting acidified sediment in a sealed quartz tube with CuO as oxidant
(Buchanan and Corcoran 1959). An aliquot of the CO2 generated from combustion of the organic
matter was analyzed for carbon stable isotope ratio by mass spectrometry. PIC and POC
concentrations were corrected for the loss of sediment resulting from organic matter
remineralization and are reported as percent carbon in dry sediment.

Kelp was sampled by collecting floating fragments that had broken from the holdfast. Phytoplankton
from a depth of 25 m were collected by pumping 15 liters of seawater through a glass fiber filter.
Kelp and phytoplankton samples were combusted and analyzed for carbon isotope ratio as
described above.
Total alkalinity was determined at sea by micro-Gran titration (Edmond 1970) of bottom water
(collected by in situ benthic chamber; Devol 1987) and pore water (obtained by centrifugation of
sediment collected by box corers on the in situ benthic chamber). Carbonate alkalinity was
calculated by correcting total alkalinity for the contribution from HS- and other sources of excess
alkalinity. DIC concentrations were calculated from carbonate alkalinity and bottom water or
sediment pH.
Sediment 210Pb was measured by alpha spectroscopy of its daughter, 210Po. The 137Cs activity
was determined by gamma spectroscopy. Both techniques are described in Sugai (1985).
Metabolic Rates
Sulfate reduction rates and methane oxidation rates were measured by means of radiotracer
techniques described by Jørgensen (1978a) and Reeburgh (1980), respectively. Sediment samples
were injected with microliter quantities of 35SO42- or 14CH4 and allowed to incubate at in situ
temperature for 1 day. Details of the rate measurements are provided in Alperin and Reeburgh
(1985).
Benthic Fluxes
Fluxes at the sediment-water interface were measured with an in situ benthic chamber (Devol
1987). During each 24-hour deployment, bottom water within the chamber was sampled at 2-3 hr
intervals. Samples were analyzed for total alkalinity and CH4 (described above) and NO3-
(Parsons, Maitia, and Lalli 1984). Total alkalinity was converted to DIC as described above. Benthic
fluxes were calculated from the change of concentration with time in the sealed chamber.
Tritiated water (HTO) was injected into overlying water a short time after closure of the deployed
benthic chamber. Quantitative HTO recovery established that chambers were leak free.
Furthermore, the HTO mixed with pore water at a rate in accordance with molecular diffusion (Devol
1987). This latter result indicates that pore waters were not actively bioirrigated.
Results
Concentration depth distributions for POC, DOC, DIC, and CH4 are shown in figure 6.2. POC
concentrations decrease in an exponential fashion (figure 6.2a), with greater than half the change
occurring in the upper 10 cm. POC concentrations remain relatively constant below 70 cm at about
20% of their initial value.
The most prominent feature of the DOC profile is the large concentration gradient at the sediment-
water interface (figure 6.2b). DOC accumulates to >-.115 mM in the upper 10 cm, indicating net
production within the sediment column. DOC concentrations decrease between 10 and 30 cm,

suggesting net DOC consumption in this zone. Below 30 cm, DOC concentrations show no general
trend with depth.
The DIC concentration profile shows a regular increase with depth, approaching 60 mM at 80 cm
(figure 6.2c). At greater depths, DIC concentrations remain relatively constant. DIC calculated from
shipboard alkalinity measurements agree with laboratory DIC measurements, indicating that
freezing and storage of sediment samples did not significantly affect DIC values.
Methane concentrations are very low in the upper 10 cm (figure 6.2d). The profile shows upward
concavity between 10 and 40 cm, indicating net consumption (Reeburgh 1976; Martens and Berner
1977). Below 40 cm, the profile becomes concave down, suggesting net methane production.
Methane concentrations increase to a maximum of 13 mM, approximately 90% saturation at in situ
temperature and pressure. The scatter in the data from depths greater than 70 cm is probably due
to CH4 bubble formation, which occurred when sediment decompressed after core retrieval.
PIC does not appear to be an active participant in the Skan Bay sediment carbon cycle. Average
PIC concentrations in samples from the sediment surface (0.90 0.11%, n=3) and below 70 cm (0.79
0.11%, n=4) are nearly identical, suggesting no carbonate precipitation or dissolution. Pore waters
below 12 cm are supersaturated with respect to calcite, but stable isotope measurements of the PIC
pool show no indication of authigenic carbonate formation (Alperin 1988).
The 13C value of POC shifts systematically from -19 at the sediment surface to <-20 at depth
(figure 6.3). The shift occurs primarily in the upper 10 cm, where the decrease in POC concentration
is most pronounced (figure 6.2a). Below 10 cm, most 13C-POC values fall within a narrow range (-
20.0 to -20.6 ). Carbon isotope ratios for sediment POC, kelp, and phytoplankton are summarized
in table 6.1.
The relationship between POC concentration and 13C-POC in the upper 10 cm of the Skan Bay
sediment column is shown in figure 6.4. POC concentrations and 13C values are correlated,
suggesting that the down core shift in 13C-POC is related to organic matter remineralization. This
point is discussed further in the section titled "Carbon Isotope Mass Balance."
The 210Pb profile (plotted as natural log excess 210Pb vs. cumulative mass) is approximately linear
(r2=0.94, n=46), suggesting a relatively constant sedimentation rate. The dry sediment
accumulation rate was calculated by dividing the 210Pb decay constant by the slope of a line fitted
to the 210Pb profile. Sediment accumulation rates were also calculated from 137Cs profiles by
assuming the 137Cs maximum to correspond to 1963, the peak year of atmospheric nuclear bomb
testing. The average sediment accumulation rate between 1980 and 1984 was calculated from the
difference in depth of the 137Cs maximum for two cores collected four years apart. Sedimentation
rates calculated from two radiotracers with distinct sources and time functions agree reasonably
well (table 6.2). This agreement provides strong evidence for steady-state sediment deposition.
Sulfate reduction rates (figure 6.5a) were low near the sediment-water interface and had two
maxima, a primary maximum near the surface that coincides with the DOC maximum (figure 6.2b),
and a smaller secondary maximum below 30 cm. Methane oxidation rates were low near the
sediment-water interface, reached a maximum in a narrow subsurface zone, and decreased with

depth (figure 6.5b). The maximum in methane oxidation rate is located just above the secondary
maximum in sulfate reduction rate. Depth-integrated rates of sulfate reduction exceed methane
oxidation by an order of magnitude (table 6.3). Methane concentration gradients in the upper 10 cm
and below 70 cm are relatively small (figure 6.2d), suggesting minimal loss of CH4 via diffusion.
Methane is undersaturated at in situ temperature and pressure, indicating that natural ebullition is
not an important transport mechanism. Since the upper meter of the sediment column approximates
a closed system with respect to CH4, integrated rates of methane production and methane
oxidation must be nearly equal.
Results of the benthic flux experiment are given in table 6.4. DIC and NO3- fluxes from separate
deployments of the in situ benthic chamber over a two-week period agree within 15%, indicating
relatively little horizontal heterogeneity around the study site.
Sediment Carbon Budget
The three components of the Skan Bay sediment carbon budget-(1) deposition, (2) preservation,
and (3) remineralization-are evaluated in the following section.
POC Deposition
The POC deposition rate (2.22 0.48 mmol cm-2 yr-1) was calculated as the product of the dry
sediment accumulation rate (table 6.2) and the POC concentration at the sediment-water interface
(9.5%). The surface concentration was estimated by linear extrapolation of POC data from the
upper 5 cm. This extrapolated concentration is about 10% higher than measured values for the 0 to
3 cm depth interval (8.6 0.3%, n=3). On the assumption that deposited organic matter has a
composition similar to the Redfield molecule [(CH2O)106(NH3)16(H3 cf1PO4)], organic matter
deposition represents 27% of the total sediment accumulation.
POC Preservation
The bulk of the organic matter remineralization in Skan Bay occurs in the upper meter of the
sediment column. Although carbon degradation continues at greater depths, rates are very slow.
Therefore, organic carbon below 100 cm is assumed to be preserved. This assumption is justified
by relatively constant POC, DOC, DIC, and CH4 concentrations between 70 and 100 cm (figure
6.2). The POC preservation rate (0.39 0.09 mmol cm-2 yr-1) was calculated as the product of dry
sediment accumulation rate (table 6.2) and average POC concentration between 70 and 100 cm
(1.7 0.2%, n=4).
POC Remineralization
POC remineralization was calculated by use of three approaches: (1) the difference between POC
deposition and preservation, (2) the quantity of recycled and buried carbon, and (3) depth-
integrated rates of bacterial metabolism. Each approach is based on independent measurements
and requires different assumptions to convert measured variables to components of the carbon
budget. The use of multiple techniques to evaluate remineralization rates provides a check on the
accuracy of the carbon budget.

POC Deposition Minus Preservation
Total carbon remineralization can be calculated from the difference between POC deposition and
preservation, provided that Skan Bay sediment has been at steady state over the time scale
representing the upper meter of the sediment column ( Sone hundred years). As discussed above,
geochronological data suggest that the sediment deposition rate in Skan Bay has been relatively
constant over this time period. The difference between POC deposition and preservation yields a
remineralization rate of 1.83 0.39 mmol cm-2 yr-1.
Recycled and Buried Carbon
Conservation of mass requires that remineralized carbon diffuse to the water column or be buried at
depth. The quantity of carbon recycled to the water column can be directly measured with an in situ
benthic chamber. In nonbioirrigated sediments, benthic fluxes can also be calculated from Fick's
Law (Berner 1980a):
F = Ds (dc/dx)o (6.1)
where F is the diffusive flux, is sediment porosity, Ds is the whole sediment molecular diffusion
coefficient, and (dc/dx)o is the concentration gradient at the sediment-water interface.
Water samples from the benthic chamber were not analyzed for DOC; hence the DOC benthic flux
was determined only from Fick's Law. Although DOC is a complex mixture of molecules, it can be
represented by a single compound having a molecular weight equal to the average for the DOC
reservoir. The DOC concentration gradient was calculated by assuming a linear profile between the
overlying water and the uppermost pore water sample (0 to 3 cm). The logarithm of aqueous
diffusion coefficients for organic compounds varies inversely with the logarithm of molecular weight
(Jost 1960). Diffusion coefficients and benthic fluxes for DOC having various average molecular
weights are summarized in table 6.5.
There is little information on the molecular weight distributions of dissolved organic compounds in
marine pore waters. Several studies have shown that the bulk of the pore water DOC has a
molecular weight greater than 500 (Krom and Sholkovitz 1977; Orem et al. 1986) Numerous studies
of aquatic humic substances suggest molecular weights less than 104 (Thurman 1985). On the
assumption that Skan Bay DOC has an average molecular weight between 103 and 104, the benthic
DOC flux ranges from 0.49 to 0.20 mmol cm-2 yr-1. The mean of these two values is taken as the
best estimate, and the range is taken as the uncertainty.
The DIC concentration gradient was estimated by assuming a linear profile between bottom water
and the first pore water sample (0 to 0.25 cm). The DIC diffusion coefficient (assumed to be the
same as HCO3-) was taken from Li and Gregory (1974), corrected to in situ temperature and
salinity according to the Stokes-Einstein equation (Lerman 1979), and converted to whole sediment
diffusion coefficient by following Ullman and Aller (1982). Porosity at the sediment-water interface
was taken to be 0.99 (Alperin 1988). The DIC flux calculated from Fick's Law (1.96 0.34 mmol cm-2
yr-1) agrees reasonably well with that measured by benthic chamber (1.52 0.24 mmol cm-2 yr-1,

table 6.4). The mean of these two values is taken as the best estimate of the DIC benthic flux.
The CH4 flux calculated by Fick's Law (0.002 0.002 mmol cm-2 yr-1) with a diffusion coefficient
from Sahores and Witherspoon (1970) (corrected to in situ temperature, salinity, and porosity as
described above) agrees with benthic chamber measurements (0.002 0.002 mmol cm-2 yr-1, table
6.4). The CH4 flux is <10% of the integrated oxidation rate (table 6.3), confirming that CH4 is nearly
quantitatively consumed within the sediment column.
The quantity of buried remineralized POC (0.006 0.001 mmol cm-2 yr-1 DOC, 0.042 0.009 mmol
-2
cm-2 yr-1 DIC, and 0.008 0.002 mmol cm yr-1 CH4) was calculated as the product of pore water
burial velocity and average pore water concentration between 70 and 100 cm. Pore water burial
velocity ( ) was calculated from the dry sediment accumulation rate (S) according to the following
equation:
= S/ > [(1 - )] (6.2)
where is solid matter density (2.33 g cm-3; Alperin 1988) and is porosity at a depth of 100 cm
(0.83; Alperin 1988).
Total POC remineralization was calculated from the sum of recycled and buried remineralized
carbon (2.15 0.35 mmol cm-2 yr-1). In Skan Bay sediments, more than 97% of the remineralized
POC is recycled to the water column, the remainder being buried. DOC appears to be a significant
fraction of the benthic carbon flux ( S17%), suggesting that remineralized POC is not quantitatively
converted to CO2 and CH4.
Integrated Metabolic Rates
Sediment bacteria convert organic carbon to CO2 and CH4 using inorganic oxidants as electron
acceptors. This process, known as terminal metabolism, proceeds through a series of reactions in
which oxidants are sequentially consumed in order of decreasing oxidation potential: O2, Mn(II),
NO3-, Fe(III), SO4-2, and HCO3- (Froelich et al. 1979; Reeburgh 1983). In Skan Bay, where bottom
water O2 is highly depleted for a portion of the year and sediments are sulfidic at the interface,
aerobic metabolism is assumed to be relatively unimportant in the sediment carbon cycle. Likewise,
metal oxides have been shown to play a minor role in the organic carbon oxidation process
(Henrichs and Reeburgh 1987). Therefore, rates of terminal metabolism can be estimated as the
sum of nitrate reduction, sulfate reduction, and methane production.
After conversion to carbon equivalents (Froelich et al. 1979), nitrate reduction, sulfate reduction,
and methane production metabolize 0.063 0.001, 1.82 0.10, and 0.058 0.010 mmol C cm-2 yr-1,
respectively. Sulfate reduction is clearly the dominant electron acceptor, accounting for 94% of the
terminal metabolism, while nitrate reduction and methane production each contribute 3%.
As discussed above, S17% of the remineralized organic matter does not undergo terminal
metabolism but rather diffuses from the sediment as DOC. In order to calculate the total
remineralization rate, the recycled and buried DOC must be added to integrated metabolic rates.

This yields a total POC remineralization rate of 2.30 0.18 mmol cm-2 yr-1.
Summary
The sediment carbon budget for Skan Bay is summarized in table 6.6. POC remineralization rates
calculated by three approaches agree quite well. The variability between each estimate is
comparable to the uncertainty associated with each.
The balanced carbon budget allows us to calculate the fraction (expressed as percent) of POC
deposition that is remineralized in the upper meter of the sediment column (%RPOC):
where POCD and POCR represent POC deposition and remineralization rates, respectively. The
POC remineralization rate was calculated by three independent approaches (table 6.6) and is
therefore well constrained. The POC deposition rate is, however, based on a surface POC
concentration that was estimated by extrapolation. A more reliable estimate of POCD can be
calculated by mass balance:
POCD = POCR + POCP (6.4)
where POCP represents the POC preservation rate. The POC preservation rate was calculated
from measured concentrations (figure 6.2a) and sediment accumulation rates (table 6.2) and is
therefore not subject to the uncertainty generated by data extrapolation.
Combining equations [6.3] and [6.4], we calculate that 84 3% of the deposited organic carbon is
remineralized. This fraction is relatively high but within the range found for continental shelf and
estuarine sediments (20 to 90%; Henrichs and Reeburgh 1987).
Carbon Isotope Mass Balance
Stable isotope ratios of Skan Bay sediment POC are intermediate between those of phytoplankton
and kelp (table 6.1). This suggests that POC may be composed of compounds derived from both
sources. Changes in POC isotopic composition with depth suggest that organic compounds derived
from the two sources have different remineralization rates. A carbon isotope mass balance provides
a means of estimating the contribution of phytoplankton and kelp to organic matter deposition and
remineralization. Such a calculation requires the following assumptions:
1. Phytoplankton and kelp are the major sources of POC to Skan Bay sediments.
The atomic C:N ratio for Skan Bay sedimentary POC averaged 7.7 0.1 (n=6) with no systematic
depth variation (S.M. Henrichs, unpublished data), indicating that terrestrial material, with C:N ratios

of 20 to 200 (Hedges et al. 1986), is not a major component. Since sea grasses are not present in
the vicinity of the Skan Bay study site, phytoplankton and kelp appear to be the principal POC
sources.
2. The isotopic compositions of phytoplankton and kelp are not altered during decomposition.
Marine organisms are primarily composed of proteins and carbohydrates with minor amounts of
lipids (Romankevich 1984). Within an organism, the protein and carbohydrate pools have similar
isotopic compositions while the lipid fraction is approximately 5 lighter than the whole organism
(Deines 1980; Galimov 1985). The isotopic conformity in the protein and carbohydrate pools
suggests that isotopic alteration of organic matter by selective decomposition of biochemical
components will be minimal. Since the lipid fraction typically accounts for <10% of the organic
carbon in marine organisms (Romankevich 1984), selective decomposition or preservation of lipid
will have little effect on the isotopic composition of the POC reservoir.
3. The Down core shift in 13C-POC (figure 6.3) is related to remineralization rather than temporal
changes in carbon isotope ratios of organic matter sources.
This assumption is supported by several lines of evidence. First, short-term (seasonal) variability in
the isotopic composition of the source organic matter is integrated by the 3 cm sampling interval,
which represents a time interval of approximately 1.2 years in the upper 10 cm (average porosity of
0.95). Second, long-term (decadal) variability in the isotopic composition of source material is
unlikely given the nearly constant 13C-POC values between 10 and 100 cm (-20.3 0.3 , n=32,
figure 6.3). And third, the correlation between POC concentration and 13C-POC (figure 6.4)
suggests that organic matter concentration and isotope ratios are controlled by the same process, i.
e., organic matter remineralization.
4. Phytoplankton and kelp deposition have been relatively constant over the time span represented
by the upper meter of Skan Bay sediment (the steady-state assumption).
Three independent approaches were used to estimate POC remineralization rates. Each approach
requires that the sediment system be at steady state for a particular time period. POC deposition
and preservation provide an accurate estimate of remineralization only if the system has been
stable over a time scale representing the upper meter of sediment ( S100 years). The quantity of
recycled and buried carbon can be accurately measured or calculated only if concentration
gradients have remained stable over the time scale characteristic of diffusion and chemical reaction
(several years). Rates of terminal metabolism measured during short-duration tracer experiments
provide accurate estimates of annual remineralization only if the system is stable over time scales
less than 1 year. Agreement between the three approaches occurs only if the system has been at
steady state through all these time scales.
Given these four assumptions, the fraction (expressed as percent) of POC deposition derived from
kelp (fkelp) and phytoplankton (fphyto) can be estimated by an isotope mass balance calculation
(Shultz and Calder 1976):

fphyto = 100 - fkelp (6.6)
where 13Cphyto, 13Ckelp, and 13CPOC are 13C values for phytoplankton, kelp, and
deposited POC, respectively (table 6.1). The fraction of preserved POC derived from phytoplankton
and kelp is estimated by analogous equations.
The fraction (expressed as percent) of kelp deposition that is remineralized in the upper meter of
the sediment column (%Rkelp) can be calculated as:
where fP and fD are the fractions of preserved and deposited POC that are derived from kelp. The
fraction of phytoplankton deposition that is remineralized is calculated by an analogous equation.
Results of the isotope mass balance are presented in table 6.7. The calculations suggest
preferential remineralization of organic matter derived from kelp. Kelp represents about half the
organic carbon deposited at the sediment surface but only a third of the organic carbon that is
preserved. Approximately 90% of the kelp- and 80% of the phytoplankton-derived material is
remineralized in the upper meter of the Skan Bay sediment column.
Discussion
We now develop a simple theoretical model to elucidate the key variables that control whether
sediment organic matter is remineralized or preserved in nonbioturbated sediments.
It is widely recognized that sediment POC is a complex mixture of many compounds, each having
different decomposition kinetics (Jørgensen 1978b; Berner 1980b; Westrich and Berner 1984). The
steady-state equation describing the concentration of particulate organic compound i (Ci) as a
function of depth in nonbioturbated, noncompacting sediments is given by Berner (1980a):
- dCi/dx - kiCi = 0 (6.8)
where (cm yr-1) is the whole sediment accumulation rate, x (cm) is vertical distance below the
sediment surface, and ki (yr-1) is the first-order rate "constant" for degradation of compound i. Even
in this theoretical treatment, which considers individual organic compounds, ki may be influenced by
environmental factors such as electron acceptor availability, bacterial population densities, and
interspecies competition and symbiosis. As a first approximation, we assume that depth variation of
the rate "constant" for an individual compound is small relative to variability between rate
"constants" for different components of the sediment organic reservoir and thus may be treated as a

constant.
Given the boundary condition Ci=Coi at x=0 (where Coi is the concentration of i at the sediment-
water interface), the solution to equation (6.8) is:
Ci = Coi exp (-kix/ ) (6.9)
The fraction (expressed as percent) of compound i deposition that is remineralized (%Ri) within a
defined sediment interval (0<x<z) can be calculated from equation (6.9):
%Ri = [1 - exp (-kiz/ )]100 (6.10)
The extent of remineralization of compound i depends on both its rate constant (ki) and the time
interval (z/ ) being considered. It is apparent that "remineralization" and "preservation" are relative
terms. An organic compound will appear to be "biologically available" if ki>> /z. Likewise, an
organic compound will appear to be "refractory" if ki<< /z.
The fraction of compound i deposition that is remineralized in a given time period is proportional to
its decomposition rate constant (figure 6.6). Rate constants primarily depend on the chemical
composition (i.e., molecular weight, presence or absence of heteroatoms, extent of cross-linking or
branching, etc.), which is ultimately controlled by the metabolic pathways of the source organism.
Coastal sediments receive organic carbon from a variety of sources, including phytoplankton, sea
grasses, macroalgae, and terrestrial flora. The extent of organic matter remineralization may be at
least partly controlled by the relative contribution from each source.
POC stable isotope and C:N ratios suggest that kelp and phytoplankton are the main sources of
organic carbon to Skan Bay sediments. "Composite" degradation rate constants for kelp (0.028 yr-1)
and phytoplankton (0.019 yr-1) were calculated from equation (6.10) as follows. The extent of
remineralization for kelp and phytoplankton were taken from table 6.7. The whole sediment
accumulation rate ( = 1.2 cm yr-1) was calculated from the dry sediment accumulation rate (S,
table 6.2), solid matter density ( ), and average sediment porosity ( avg [0.90; Alperin 1988]): = S/
[ (1- avg)]. The zone of active organic matter remineralization was taken as the upper 100 cm.
Although "composite" rate constants apply to a complex, heterogeneous mixture of organic

compounds and have no precise chemical meaning, they do provide a relative measure of the
"biological availability" of organic matter derived from two sources. The average residence time
(calculated as the reciprocal of the "composite" rate constant) for kelp (36 years) is 30% lower than
that of phytoplankton (53 years). The generalization that bacteria prefer nitrogen-rich (e.g.,
phytoplankton, C:N 6.6) to nitrogen-poor (e.g., kelp, C:N 16) organic matter does not appear to be
universally valid. Skan Bay pore waters contain high NH4+ concentrations (>1 mM), which may
satisfy bacterial requirements for reduced nitrogen. The greater reactivity of kelp-derived organic
matter may stem from (1) bacterial preference for carbohydrate-rich kelp material, (2) less chemical
alteration of kelp-derived organic matter prior to deposition, and/or (3) kelp's macroscopic surface,
which is suitable for extensive bacterial colonization.
Degradation rate constants for organic matter derived from kelp and phytoplankton (0.028 and
0.019 yr-1, respectively) are considerably less than rate constants reported by Westrich and Berner

(1984) based on laboratory experiments with sediment amended with fresh and aged plankton (8.8
and 0.84 yr-1, respectively). After correction for temperature differences (assuming Q10=2),
degradation rate constants from the laboratory experiments exceed those of Skan Bay sediment by
one or two orders of magnitude. The discrepancy is at least in part due to the different time scales
being considered. Emerson and Hedges (1988) have shown an inverse correlation between
calculated rate constants and experimental time scale. Thus, laboratory experiments provide
information only on compounds that degrade on rapid time scales while field studies detect only
compounds that degrade on time scales comparable to the age of the sediment.
In Skan Bay sediments, the POC concentration does not decrease below 70 cm, although a
significant quantity of organic carbon remains (1.7%, figure 6.2a). Organic compounds that persist
at this depth must have decomposition rate constants less than 10-2 yr-1 (figure 6.6). Several
hypotheses have been offered to explain the existence of refractory organic matter in shallow
sediments: (1) it is formed in situ by condensation or adsorption reactions involving low molecular
weight DOC (Nissenbaum and Kaplan 1972), (2) water column photosynthetic and/or heterotrophic
organisms produce compounds that are resistent to metabolism by sediment bacteria (Hatcher et
al. 1983), and (3) it is composed of fossil organic matter associated with the fine-grained sediment
particles (Emerson and Hedges 1988). Future studies of DOC molecular weight distributions
coupled with carbon dating of dissolved and particulate organic reservoirs should help to elucidate
which mechanism(s) is (are) responsible for preservation of sediment organic matter.
1. The carbon budget for Skan Bay sediments is well constrained. Organic carbon
remineralization rates calculated by three independent approaches (the difference between
POC deposition and preservation, recycled and buried carbon, and integrated metabolic
rates) agree within 10%. The budget indicates that 84 3% of the organic carbon
deposition is remineralized in the upper meter of the sediment column.
2. Nearly all remineralized POC is recycled to the water column; less than 5% is buried as DIC,
DOC, and CH4. DOC comprises a significant fraction ( S17%) of the benthic carbon flux in
this system.
3. Terminal metabolism is dominated by sulfate reduction. Nitrate reduction and methane

production account for only 6% of the total terminal metabolism by sediment bacteria.
4. A downcore shift in stable carbon isotope ratio of sediment POC is consistent with
preferential remineralization of organic matter derived from kelp. A simple first-order kinetic
model suggests that the degradation rate constant for kelp is S30% higher than that of
phytoplankton.
5. Organic compounds with Ki>10-2 will be quantitatively remineralized in the upper meter of
the Skan Bay sediment column. Likewise, compounds with Ki<10-3 will be essentially
preserved.
Acknowledgments
This research was supported by the Marine Chemistry Program of the National Science Foundation
through grant OCE 84-008674. Susan Sugai provided the geochronology data and Susan Henrichs

provided elemental composition data. We thank Liz Canuel, Ann McNichol, and an anonymous
reviewer for their constructive criticisms. Contribution number 770 from Institute of Marine Science,
University of Alaska.
References
Alperin, M. J. 1988. The carbon cycle in an anoxic marine sediment: concentrations, rates,
isotope ratios, and diagenetic models. Ph.D. dissertation, Univ. of Alaska, Fairbanks.
Alperin, M. J. and W. S. Reeburgh. 1985. Inhibition experiments of anaerobic methane

oxidation. Appl. Environ. Microbiol. 50:940-945.
Berner, R. A. 1980a. Early Diagenesis: a Theoretical Approach. Princeton, N.J.: Princeton

Univ. Press.
Berner, R. A. 1980b. A rate model for organic matter decomposition during bacterial sulfate
reduction in marine sediments. In Biogeochemistry of Organic Matter at the Sediment-Water
Interface, pp. 35-44. CNRS Int. Colloq.
Berner, R. A. 1982. Burial of organic carbon and pyrite in the modern ocean: its geochemical
and environmental significance. Am. J. Sci. 282:451-475.
Buchanan, D. L. and B. J. Corcoran. 1959. Sealed tube combustions for the determination of
carbon-14 and total carbon. Anal. Chem. 31:1635-1638.
Craig, H. 1957. Isotopic standards for carbon and oxygen and correction factors for mass-
spectrometric analysis of carbon dioxide. Geochim. Cosmochim. Acta 12:133-149.
Deines, P. 1980. The isotopic composition of reduced organic carbon. In P. Fritz and J. Ch.
Fontes, eds., Handbook of Environmental Isotope Geochemistry, 1:329-406. New York:
Elsevier.
Devol, A. H. 1987. Verification of flux measurements made with in situ benthic chambers.
Deep-Sea Res. 34:1007-1026.
Edmond, J. M. 1970. High precision determination of titration alkalinity and total carbon
dioxide content of seawater by potentiometric titration. Deep-Sea Res. 17:737-750.
Emerson, S. 1985. Organic carbon preservation in marine sediments. In E. Sundquist and

W. S. Broecker, eds., The Carbon Cycle and Atmospheric CO2: Natural Variations Archean
to Present. Geophysical Monograph 32, American Geophysical Union, Washington, D.C.
Emerson, S. and J. I. Hedges. 1988. Processes controlling the organic carbon content of
open ocean sediments. Paleooceanogr. 3:621-634.
Froelich, P. N., G. P. Klinkhammer, M. L. Bender, N. A. Luedtke, G. R. Heath, D. Cullen, P.

Dauphin, D. Hammond, B. Hartman, and V. Maynard. 1979. Early oxidation of organic matter

in pelagic sediments of the eastern Equatorial Atlantic: sub-oxic diagenesis. Geochim.

Galimov, E. M. 1985. The Biological Fractionation of Isotopes. New York: Academic.
Hatcher, P. G., E. C. Spiker, N. M. Szeverenyi, and G. E. Maciel. 1983. Selective

preservation and origin of petroleum-forming kerogen. Nature 305:498-501.
Hattori, A., J. J. Goering, D. B. Boisseau. 1978. Ammonium oxidation and its significance in
the summer cycling of nitrogen in oxygen depleted Skan Bay, Unalaska Island, Alaska. Mar.
Sci. Comm. 4:139-151.
Hedges, J. I., W. A. Clark, P. D. Quay, P. M. Grootes, J. E. Richey, A. H. Devol, and U. de

M. Santos. 1986. Compositions and fluxes of particulate organic material in the Amazon
River. Limnol. Oceanogr. 31:717-738.
Henrichs, S. M. and W. S. Reeburgh. 1987. Anaerobic mineralization of marine sediment

organic matter: rates and the role of anaerobic processes in the oceanic carbon economy.
Jørgensen, B. B. 1978a. A comparison of methods for the quantification of bacterial sulfate

reduction in coastal marine sediments. I. Measurement with radiotracer techniques.
Jørgensen, B. B. 1978b. A comparison of methods for the quantification of bacterial sulfate

reduction in coastal marine sediments. II. Calculation from mathematical models.
Jost, W. 1960. Diffusion in Solids, Liquids, and Gases. New York: Academic.
Krom, M. D. and E. R. Sholkovitz. 1977. Nature and reactions of dissolved organic matter in
the interstitial waters of marine sediments. Geochim. Cosmochim. Acta 41:1565-1573.
Lerman, A. 1979. Geochemical Processes: Water and Sediment Environments. New York:
Wiley.
Li, Y.-H. and S. Gregory. 1974. Diffusion of ions in seawater and in deep-sea sediments.
Martens, C. S. and R. A. Berner. 1977. Interstitial water chemistry of Long Island Sound
sediments. I. Dissolved gases. Limnol. Oceanogr. 22:10-25.
Nissenbaum, A. and I. R. Kaplan. 1972. Chemical and isotopic evidence for the in situ origin
of marine humic substances. Limnol. Oceanogr. 17:570-582.
Orem, W. H., P. G. Hatcher, E. C. Spiker, N. M. Szeverenyi, and G. E. Macial. 1986.

Dissolved organic matter in anoxic pore waters from Mangrove Lake, Bermuda. Geochim.

Parsons, T. R., Y. Maitia, and C. M. Lalli. 1984. A Manual of Chemical and Biological
Methods for Seawater Analysis. New York: Pergamon Press.
Reeburgh, W. S. 1967. An improved interstitial water sampler. Limnol. Oceanogr. 12:163-

165.
Reeburgh, W. S. 1976. Methane consumption in Cariaco Trench waters and sediments.

Earth Planet. Sci. Lett. 28:337-344.
Reeburgh, W. S. 1980. Anaerobic methane oxidation: rate depth distributions in Skan Bay
sediments. Earth Planet. Sci. Lett. 47:345-352.
Reeburgh, W. S. 1983. Rates of biogeochemical processes in anoxic sediments. Ann. Rev.

Earth Planet. Sci. 11:269-297.
Romankevich, E. A. 1984. Geochemistry of Organic Matter in the Ocean. New York:

Springer Verlag.
Sahores, J. J. and P. A. Witherspoon. 1970. Diffusion of light parrafin hydrocarbons in water

from 2°C to 80°C. In G. D. Hobog and G. C. Spears, eds., Advances in Organic
Geochemistry, 1966, pp. 219-230. New York: Pergamon Press.
Shaw, D. G., M. J. Alperin, W. S. Reeburgh, and D. J. McIntosh. 1984. Biogeochemistry of

acetate in anoxic sediments of Skan Bay, Alaska. Geochim. Cosmochim. Acta 48:1819-
1825.
Shultz, D. J. and J. A. Calder. 1976. Organic carbon 13C/12C variations in estuarine

sediments. Geochim. Cosmochim. Acta 40:381-385.
Sugai, S. F. 1985. Processes controlling trace metal and nutrient geochemistry in two
southeastern Alaskan fjords. Ph.D. dissertation, Univ. of Alaska, Fairbanks.
Sugimura, Y. and Y. Suzuki. 1988. A high-temperature catalytic oxidation method for the
determination of non-volatile dissolved organic carbon in seawater by direct injection of a
liquid sample. Mar. Chem. 24:105-131.
Thurman, E. M. 1985. Organic Geochemistry of Natural Waters. Dordrecht: Martinus Nijhoff/

Dr. W. Junk.
Ullman, W. J. and R. C. Aller. 1982. Diffusion coefficients in nearshore marine sediments.

Limnol. Oceanogr. 27:552-556.
Westrich, J. T. and R. A. Berner. 1984. The role of sedimentary organic matter in bacterial
sulfate reduction: the G model tested. Limnol. Oceanogr. 29:236-249.

Wilson, R. F. 1961. Measurement of organic carbon in sea water. Limnol. Oceanogr. 6:251-
261. <!-------comment------!> </ul> <hr width=25%> <center><a href=index.
html>Organic Matter: Productivity, Accumulation, and Preservation in Recent and Ancient
Sediments</a> </body> </html>


7. Preservation of Sargassum Under Anoxic Conditions:

Molecular and Isotopic Evidence
The Orca Basin represents a unique environment where the molecular and isotopic integrity of
organic matter has been preserved relatively unaltered for hundreds of years. The environmental
conditions that led to organic matter preservation at this site lend clues to understanding the
preservation of organic matter over geological time. Variations in total amounts of biochemicals in
preserved Sargassum such as photosynthetic pigments, amino acids, fatty acids, sterols, and fatty
alcohols can in most part be explained by variability within Sargassum tissues, seasonal variations,
and/or a change in the rate of anabolism to catabolism during settling and deposition. Few
diagenetic alteration products were present. Only minor amounts of chlorophyll a breakdown
products and several minor xanthophyll alteration products were detected. A comparison of fatty
acids, total sterols, and fatty alcohols demonstrates a remarkable similarity between Sargassum
tissues buried in the anoxic, hypersaline sediments and present-day living algae. Whereas some
changes in the lipids were observed, both in concentration and distribution, the environmental
conditions have resulted in the Sargassum lipids remaining largely unchanged. Even the most labile
amino acids are also well preserved. This unusual preservation is most likely due to the anoxic
conditions, the hypersalinity, the low bacterial activity, the low temperature, and the presence of
antibacterial exudates produced by Sargassum.
In the mid-1970s the Orca Basin was recognized as a unique depositional environment located on
the Gulf of Mexico continental slope (Shokes et al. 1977). The Orca Basin was formed when a
group of diapirs coalesced, surrounding a depression. It contains S200 m of hypersaline anoxic
brine that originated from dissolution of a salt diapir. Coring of the anoxic muds recovered layers of
preserved seaweed, identified as Sargassum sp. (Addy and Behrens 1980; Northam et al. 1981;
Wiesenburg, Brooks, and Bernard 1985). It was believed to be the first documented case of
preservation of seaweed in deep sea sediments (Kennett and Penrose 1978). Subsequent coring in
the Orca Basin has revealed that these Sargassum layers are widespread. The seaweed was intact
in the sediment; not only stipes and fronds but also air bladders were observed when the sediment
was visually examined. Carbon-14 dates for the Sargassum range from 380 to 1610 years old,
suggesting that Sargassum deposition is episodic. Initial stable carbon and nitrogen isotopic
analyses demonstrated that the deposited Sargassum was very similar to Sargassum collected
fresh at the ocean surface (14C date 40 ybp), suggesting that the layers of Sargassum were well
preserved. To test this hypothesis, extensive molecular and isotopic analyses were undertaken to
determine the extent of diagenesis and the relative stability of various biochemical constituents.
These analyses and their implications are reported here.
Methods
Sample Collection and 14C Dating Sargassum was recovered from the Orca Basin by conventional
box coring and trawling (10 m Otter Trawl). The material was immediately frozen and maintained at -

20°C prior to analysis. Carbon dating was performed by Beta Analytic, Inc. (Miami, FL).
Stable Isotopes
Samples were acidified with 30% HCl to remove carbonate and dried at 40°C. The dried samples
were ground to powder and mixed with purified (850°C, 1 h) coarsely ground cupric oxide and pure
granular copper (Alpha Resources Inc.) for a modified Dumas combustion (Macko 1981). The
mixed sample was placed in a precombusted quartz tube, which was evacuated, sealed, and
combusted for one hour at 850°C and allowed to slowly cool. The N2 and CO2 combustion products
were cryogenically purified and collected. The stable carbon and nitrogen isotope compositions
were determined on a triple collector PRISM stable isotope ratio mass spectrometer (V. G.
Micromass Ltd.). Isotope compositions are reported as:
where R is the abundance ratio of the heavy to the light isotope. The 13C values are reported
relative to PDB (Peedee Belemnite) and the 15N values are relative to atmospheric N2. Average
precision on replicate samples is approximately 0.1 . Organic carbon content was determined as
the quantity of CO2 gas measured on a calibrated manometer. Nitrogen content was estimated
from the ion intensity of N2 observed in a calibrated volume of the mass spectrometer.
Analytical-Scale Pigment Analysis Determinations
Sargassum sp. ( S 0.9 gram wet weight) was extracted in 4 or 5 mL of 90% acetone by mechanical
disruption with a glass/glass tissue homogenizer. Following homogenization, the extracts were
centrifuged for 5 min to remove cellular debris. Chlorophylls and carotenoids were separated with a
Spectra-Physics Model SP8700 liquid chromatograph equipped with a Radial-PAK C18 column (0.8
x 10 cm, 5 m particle size; Waters Chrom. Div.) at a flow rate of 6.0 mL · min-1. Prior to injection, 1
ml aliquots of the standards and algal extracts were mixed separately with 300 l of ion-pairing
solution (15 g of tetrabutylammonium acetate and 77 g ammonium acetate diluted to 1 liter with
distilled water; Mantoura and Llewellyn 1983). A two-step solvent program was used to separate the
various pigments extracted from the Sargassum (Bidigare 1989). After injection (500 l sample),
mobile phase A (80:15:5; methanol:water:ion-pairing solution) was programmed to mobile phase B
(100% methanol) over a 12 min period. Mobile phase B was then pumped for 18 min for a total
analysis time of 30 minutes. Individual peaks were detected and quantified (by area) with a Waters
Model 440 Fixed Wavelength Detector (436 nm) and a Hewlett-Packard Model 3392A integrator,
respectively. The identities of the peaks from the Sargassum extracts were determined by
comparing their retention times with those of pure standards and extracts prepared from "standard"
plant materials of known pigment composition (Dunaliella tertiolecta and Phaeodactylum
tricornutum). "On-line" diode array spectroscopy (HPLC/DAS; 400-700 nm) was performed with a
Hewlett Packard Model HP8451 Diode Array Spectrophotometer to confirm the identities of the
major chlorophylls and carotenoids.
Pigment standards were obtained from Sigma Chemical Co. (chlorophyll a and -carotene) or
purified from axenic algal cultures by thin-layer chromatography (Jeffrey 1981). Concentrations of

the pigment standards were determined spectrophotometrically in 1 cm cuvettes by use of

published extinction coefficients in the appropriate solvent systems (Jeffrey 1972; Jeffrey and
Humphrey 1975; Davies 1976). Known pigment quantities were injected and resultant peak areas
were used to calculate individual standard response factors ( g pigment · area-1). The concentration
of antheraxanthin was estimated with the violaxanthin response factor. Results are expressed as g
pigment · g-1 wet weight of tissue.
Lipid Analysis
Before extraction of the Sargassum tissues, the exteriors were visually examined and care was
taken to remove any attached organisms (i.e., sponges and bryozoans) Samples were rinsed in
distilled H2O to remove particles. For lipid extraction, equal proportions of berries, leaves, and
stems from both old and modern tissues were cut into small pieces and placed in acid-washed,
solvent-rinsed tubes. Total lipids were extracted three times with ultrasonication (probe) in CH2Cl2-
MeOH (1:1) with tissue in the final extraction allowed to remain in solvent for 12 hours under
refrigeration after sonication. All three extracts were combined and the solvent removed by rotary
evaporation. The total lipid extract was then saponified in 0.5 M KOH/methanol (plus 10% distilled
H2O) under reflux. Additional water was then added and the neutral lipids partitioned into hexane/
ethanol (9:1); the polar fraction was subsequently partitioned in a similar manner after acidification
(pH = 2). The neutral lipid fraction (containing total sterols and fatty alcohols) was treated with
BSTFA reagent to form their trimethylsilyl derivatives. Fatty acids present in the polar lipid fraction
were converted to methyl esters by use of BF3 in MeOH.
Both lipid fractions were analyzed without further purification by capillary gas chromatography (GC)
with an HP 5890 instrument with gases, temperature programming, and columns as previously
described (Harvey et al. 1988). Structural analysis of all components was performed on an HP 5985
GC-MS system under similar conditions as for GC. Electron impact mass spectra (70 eV, 0.5 scan/
sec) were acquired and processed by means of an HP dedicated data system. Molecular
identification for all components was made on the basis of coinjection with authentic standards,
comparison with reference and/or literature spectra (e.g., Brooks, Henderson, and Steel 1973) and
mass spectral interpretations.
Amino Acids
Samples were digested under N2 atmosphere in 1 ml of quartz distilled 6N HCl at 100°C for 24
hours in tubes with Teflon-lined caps. Amino acids were separated with a stepwise isocratic elution
on an ion exchange amino acid analyzer (St. John Assoc., Aldephi, MD). The amino acids are
detected as postcolumn OPA (orthophthalaldehyde) derivatives with an HTV (St. John Associates)
fluorescence detector.
Geologic Setting
The Louisiana slope is geographically complex with hummocky topography created by salt diapirs
forming large mounds, basins, and troughs (Bouma, Martin, and Bryant 1980; Martin 1980). Salt
diapirs in the area have caused extensive growth faulting, slumping, and mass sediment movement
(Sidner, Gartner, and Bryant 1977; Bouma 1978). The Orca Basin, on the northwestern Gulf of
Mexico continental slope, contains anoxic, hypersaline brine similar to that of the Red Sea (McKee

and Sidner 1976; Shokes et al. 1977; Sackett et al. 1979; figure 7.1). Slump scars on the basin
slopes, chaotic reflectors on the basin floor, and sedimentation rates indicate that mass movement
is an important supplier of sediment to the basin (Tompkins and Shephard 1979). The bottom 200
m of the basin is filled with a brine that appears to have originated from dissolution of salt exposures
in surrounding areas. Pore water profiles show decreases in salinity with increasing depth, arguing
against a deeper subsurface source for the brine (Addy and Behrens 1980). It has been estimated
that the brine accumulation began S7900 years b.p. The organic carbon content of basin
sediments is consistently two to three times greater than surrounding aerobic slope sediments
(Northam et al. 1981).
Stable Isotopes
Stable carbon and nitrogen isotopic compositions were determined for whole-tissue and lipid
extracts (table 7.1). The whole-tissue isotope values for 1610 ybp Sargassum of -16.5 is similar to
the previous report of Northam et al. (1981) for Sargassum retrieved from the Orca Basin (-16.0 to -
19.7 ). These values are also very similar to fresh Sargassum (<40 ybp) retrieved at the sea
surface (-16.1 1.3 ). The nitrogen isotopic ratios measured for whole tissue and lipids are also
similar for all Sargassum tissues analyzed. No stable isotopic change or inputs of nonindigenous
carbon or nitrogen are apparent from the stable isotope compositions.
Pigments
The Phaeophyta (brown algae) possess chlorophylls a and c, and fucoxanthin as their primary light-
harvesting pigments (Hager and Stransky 1970). The remaining carotenoids ( -carotene,
violaxanthin, antheraxanthin, and zeaxanthin) are inefficient in energy transfer to reaction center
chlorophyll a and are thought to have a photoprotective function (Larkum and Barrett 1983). In the
Chlorophyta and Phaeophyta, these latter three carotenoids undergo light-induced, reversible
deepoxidation-epoxidation reactions, a process known as "rapid xanthophyll cycling" (figure 7.2).
Although the precise function of these reactions is unclear (Jeffrey 1980), they probably function
together as a photoprotective system (Demming et al. 1987).
The quantitatively important pigments identified in the Sargassum tissues were chlorophylls a and c;
fucoxanthin; violaxanthin; antheraxanthin; zeaxanthin; and -carotene (figure 7.3). The chlorophyll a
content of the <40 ybp and 1610 ybp Sargassum sp. tissues measured in this study ranged from
108-544 g pigment · g-1 and 28-65 g pigment · g-1, respectively (table 7.2). These values are in
good agreement with those published for Sargassum muticum (Lewey and Gorham 1984), which
range from S115 to 375 g pigment · g-1 wet weight (pigment concentrations were converted to
"wet weight" units with a conversion factor of 0.1 g dry weight/1 g wet weight; cf. Lewey and
Gorham 1984). The concentration of the accessory pigments are also similar to those reported in
the literature.
The pigment content of Sargassum tissue indicated only minor diagenetic alterations after 1610
years of preservation under anoxic conditions. While the total amounts of the major pigments (on a
wet weight basis) were S2-10 fold lower than those measured in the <40 ybp Sargassum, the
relative compositions were markedly similar (table 7.2). Such variations can in part be explained by

differences in photoadaptive state, which has a strong seasonal dependence. Lewey and Gorham
(1984), for example, documented S4-fold variations in the pigment content of Sargassum muticum
during spring and early summer. Only small amounts of the chlorophyll a breakdown products,
chlorophyllide a and phaeophytin a, were detected in the 1610 ybp Sargassum. Several minor
unidentified carotenoids were also present in the 1610 ybp Sargassum and probably result from
xanthophyll alteration. Of all the pigments present in Sargassum, violaxanthin appears to be the
most susceptible to diagenetic alteration. On average, violaxanthin accounted for 5% and 1% of the
total pigments extracted from the <40 ypb and 1610 ybp Sargassum, respectively (table 7.2). The
fact that epoxide-containing carotenoids are sensitive to chemical transformation (Repeta 1982,
1989) may explain the reduced levels of violaxanthin in the 1610 ybp Sargassum. One would
expect a priori that violaxanthin would be present at fairly high levels in the 1610 ybp Sargassum
owing to the low light conditions during sinking (see discussion above regarding rapid xanthophyll
cycling). In contrast, -carotene (which lacks epoxide and hydroxyl functional groups) appears to be
the most stable pigment since it comprised 4% and 9% of the total pigments extracted from "new"
and "old" Sargassum, respectively.
Lipid Distributions
The abundance of lipids present in Sargassum was similar for all tissues analyzed. While changes
in concentration among all three lipid classes (fatty acids, sterols, and fatty alcohols) were apparent,
the similarities in lipid distributions were striking.
Fatty acids are generally considered labile components of the total lipid pool, particularly the
polyunsaturated acids (PUFAs), which display rapid decreases with depth in both sedimentary
particulate matter (De Baar, Farrington, and Wakeham 1983; Wakeham et al. 1984) and in marine
surficial sediments (Farrington, Henrichs, and Anderson 1977; Van Vleet and Quinn 1979). Those
rare instances in which significant amounts of PUFAs have been observed in surface sediments (i.
e., Peru) have been attributed to the rapid sedimentation of large amounts of phytoplankton (Smith,
Eglinton, and Morris 1983). For Sargassum, the fatty acid distribution in the tissues analyzed is
similar (table 7.3 and figure 7.4), including the presence of significant amounts of the
polyunsaturated 18:2 (peak 6) and 20:5 (peak 10), even though the total concentration in 380 ybp
material is about half that of modern material. The similarity in fatty acid distribution between
preserved and living plants suggests that little alteration has occurred and that the variability
observed is naturally inherent in the plants. However, a gradual reduction in mono and
polyunsaturated fatty acids, a loss of C18 fatty acids, and a selective preservation of C16 fatty acids
are suggested.
A comparison of total free and esterified sterols between the tissues analyzed indicates a number of
changes in both total concentration and in distribution. Total sterols showed the greatest differences
over time, <40 ybp tissue containing more than three times the total sterol concentration (on a dry
weight basis) of that in 380 ybp tissue (table 7.4 and figure 7.5). Although the reduction in
concentration of a number of sterols contributed to this difference, two important sterols cholest-5-
en-3 -ol and cholest5,22-dien-3 -ol (reduced by 94.5% and 92.6%, respectively) were in large part
responsible. Not all sterols, however, were reduced in concentration, and one sterol, 24-
ethylcholesta-5,24(28)E-dien-3 -ol, showed a high degree of resistance with only a small decrease
in concentration as compared with living tissue. Less substantial changes were observed among
the remaining 11 sterols.

While some differences in Sargassum sterol content might be expected given inherent biological
variability and possible shifts in sterol abundance over time, the large differences among the three
major sterols suggest a differential rate of preservation of 24-ethylcholesta-5,24(28)E-dien-3 -ol as
compared with either cholest-5-en-3 -ol or cholest-5,22-dien-3 -ol. A number of possibilities exist
for the resistance of the 24-ethyl derivative as compared with its nonsubstituted counterparts,
including: (1) a nonuniform distribution of 24-ethylcholesta-5,24(28)Z-dien-3 -ol in structural
portions of Sargassum tissue (e.g., stems), which could provide a mechanical barrier to
degradation, or (2) a preferential removal and/or degradation of cholest-5-en-3 -ol and cholest-5,22-
dien-3 -ol.
Whereas little direct information is available on sterols containing ethyl side chains, previous
investigations of sterol degradation in anaerobic sedimentary environments have demonstrated the
slow transformation of cholest-5-en-3 -ol to more reduced products, including cholestan-3 -ol,
(Gaskell and Eglinton 1975) cholest-4-en-3-one, and even mineralization to CO2 (Taylor, Smith,
and Gagosian 1981). Although one must keep in mind that these studies relied on the introduction
of purified cholest-5-en-3 -ol in solution rather than in plant tissue, it seems reasonable to conclude
that cholesterol decomposition (and by analogy the diunsaturated cholest-5,22-dien-3 -ol) could be
responsible for its reduced concentration in 380 ybp tissue. Surprisingly, however, we did not
observe evidence for an increased abundance in more reduced sterols (table 7.4) or the
appearance of possible degradation products, such as the 3-keto steroids, in Sargassum. In fact,
the concentration of 5 -cholestan-3 -ol, one product of cholest-5-en-3 -ol reduction, was present in
substantially lower amounts in the 380 ybp tissue than in the < 40 ybp tissue. This would suggest
that if sterol decomposition was responsible for the losses in Sargassum, it was dominated by
mineralization to CO2 rather than the formation of more reduced products. This could result from
seasonal variations as well as increased catabolism once the Sargassum settled out of the euphotic
zone.
Fatty alcohols in Sargassum show the greatest reductions among the three lipid classes, 380 ybp
material having only 25.4% of the total alcohol content of <40 ybp tissue (table 7.5 and figure 7.5).
As in many algal tissues, fatty alcohols in Sargassum are dominated by phytol derived from the
chlorophylls and account for 40.7% of total alcohols in <40 ybp tissue. Whereas its absolute
concentration was reduced by 49.5% in 30 ybp material, its relative abundance in older material
increased to 81.2% owing to the complete removal of other important alcohols such as 16:0 and
18:1. These differences in composition may reflect differences within plant tissues similar to that
hypothesized for the sterols or alternatively could be the result of differential degradation. Whereas
phytol occurs as the side chain of chlorophyll a, and is thus protected by structural material, fatty
alcohols are often localized on external surfaces of higher plants, and the distribution of alcohols in
older material could reflect their susceptibility to degradation as the first indication of early
diagenesis. An exception to this scenario is the monounsaturated 20:1, which appears resistant to
degradation and occurs in almost identical concentrations in old and modern tissues. Although its
resistance to degradation appears unusual as compared with other alcohols, which were absent in
older material, it is consistent with the trend we observed for the sterols. On a mass balance
consideration, excess (or preserved) phytol could not be accounted for by an equivalent loss in
chlorophyll a. The resulting breakdown product of chlorophyll a, chlorophyllide a, was not detected
at significant concentrations.
Amino Acids

Amino acid compositions remained essentially unaltered after 1,610 years of preservation, even the
more labile amino acids such as serine, aspartic acid, threonine, and glutamatic acid remaining
unchanged (table 7.6 and figure 7.6). Both bacterial and physical/chemical diagenesis have been
substantially retarded, as evidenced by the nearly complete preservation of the nitrogen-rich amino
acids such as arginine, lysine, and histidine and the large portion of the nitrogen of the plant still
identified as amino-itrogen.
Biochemical reactions occur as the result of chemical diagenesis. Decarboxylation of acidic amino
acids, glutamic and aspartic, yield -alanine and -amino butyric acid. Further, dehydration of serine
and threonine can occur and would yield alanine and -amino-butyric acid. None of these
diagenetic products were noted in any significant quantity, nor did the ratio of the precursor amino
acid to product amino acid change significantly. Further, the amount of nitrogen in all Sargassum
tissues could be essentially accounted for in the nitrogen contribution derived from amino acids,
another indication of the good preservation state of the algae.
The Orca Basin represents a combination of unique environmental conditions that have preserved
Sargassum for as long as 1,610 years. Variations in biochemicals in preserved Sargassum such as
pigments, amino acids, fatty acids, sterols, and fatty alcohols can in large part be explained by
variability within Sargassum tissues, seasonal variations, and/or a shift in anabolism/catabolism as
the algae settled through the water column. Few diagenetic alteration products were present. Only
minor amounts of chlorophyll a breakdown products and several minor xanthophyll alteration
products were detected. A comparison of fatty acids, total sterols, and fatty alcohols demonstrates a
remarkable similarity between living and preserved tissue. Although some changes in all three lipid
classes were observed, both in concentration and distribution, the unique environmental conditions
have preserved Sargassum lipids largely unchanged. Early diagenetic changes do suggest partial
hydrogenation of unsaturated bonds; selective loss of C18 fatty acids and C27 sterols; and selective
preservation of phytol, C16 fatty acids, and C29 sterols. Even the most labile amino acids were well
preserved. This unusual preservation is most likely due to the anoxic conditions, the hypersalinity,
the low bacterial activity, the low temperatures, and the antibacterial exudates produced by the
Sargassum (Conover and Sieburth 1964, 1965; Sieburth and Conover 1965).
The conditions that lead to this preservational event in the Orca Basin suggest that a multiplicity of
factors must have been coincidentally operative. The Sargassum survived a long settling period
through more than 2000 meters of water-owing possibly to protection from microbes by antibacterial
exudates and/or protection provided by the structural components of the Sargassum tissue. A
cataclysmic storm event might have also been involved (i.e., air bladder rupture), resulting in a rapid
transport through the water column. Significant reduction in overall biochemical content, on a dry
weight basis, may have resulted from the ability of the plant to alter its rate of pigment biosynthesis
as it receded from the euphotic zone (photoadaptation) or from the increasing predominance of
catabolism over anabolism. These factors would preserve the Sargassum tissues more effectively
than the cooccurring planktonic pelagic productivity, providing a skewed representation of the actual
water column productivity. The selective preservation of biochemicals (i.e., C29 sterols) also
contributes to nonrepresentative preservation, especially once structurally recognizable organic
matter is no longer present in the sediment.
Orca Basin Sargassum represents a unique example of preservation of organic matter for hundreds
of years. The anoxic, hypersaline conditions have corollaries throughout geologic time, but these
events must also coincide with significant enhancements in water column productivity. A
mechanism for transporting organic matter, relatively intact, to the underlying sediments must also

be present.
References
Addy, S. K. and E. W. Behrens. 1980. Time of accumulation of hypersaline anoxic brine in

Orca Basin (Gulf of Mexico). Mar. Geol. 37:241-252.
Bidigare, R. R. 1989. Photosynthetic pigment composition of the Brown Tide Alga: Unique
chlorophyll anc carotenoid derivatives. In E. J. Cosper, E. Carpenter, and M. Bricelj, M. eds.
Novel Phytoplankton Blooms, Coastal and Estuarine Studies 35:57-75. Berlin: Springer
Verlag.
Bouma, A. H. 1978. Intraslope basins in the northwest Gulf of Mexico. In Framework, Facies,
and Oil Trapping Characteristics of the Upper Continental Margin. AAPG Stud. Geol. 7:289-
302.
Bouma, A. H., R. G. Martin, and W. R. Bryant. 1980. Shallow structure of the upper
continental slope, central Gulf of Mexico. Proc. Offshore Tech. Conf. 12:583-587.
Brooks, C. J. W., W. Henderson, and G. Steel. 1973. The use of thrimethylsilyl ethers in
characterization of neutral sterols and steroid diols by gas chromatography-mass
spectrometry. Biochim. Biophys. Acta 296:431-445.
Conover, J. T. and J. Sieburth. 1964. Effect of Sargassum distribution on its epibiota and
antibacterial activity. Botanica Marina 6:147-157.
Conover, J. T. and J. Sieburth. 1965. Effect of tannins excreted from Phaeophyta on

planktonic animal survival in tide pools. In E. G. Young, and J. L. McLachlan, eds., Proc 5th
Int. Seaweed Symp., pp. 99-100. Oxford: Pergamon Press.
Davies, B. H. 1976. Carotenoids. In T. W. Goodwi, ed., Chemistry and Biochemistry of Plant

Pigments, pp. 38-165. London: Academic.
De Baar, H. J. W., J. W. Farrington, and S. G. Wakeham. 1983. Vertical flux of fatty acids in
the North Atlantic Ocean. J. Mar. Res. 41:19-41.
Demming, B., K. Winter, A. Kruger, and F. Czygan. 1987. Photoinhibition and zeaxanthin
formation in intact leaves. Plant Physiol. 84:218-224.
Farrington, J. W., S. M. Henrichs, and R. Anderson. 1977. Fatty acids and Pb-210
geochronology of a sediment core from Buzzards Bay Massachusetts. Geochim.
Gaskell, S. J. and G. Eglinton. 1975. Rapid hydrogenation of sterols in a contemporary

lacustrine sediment. Nature 254:209-211.
Hager, A. and H. Stransky. 1970. Das carotiniodmuster und die verbreitung des

lichtinduzierten xanthophyllcyclus in verschiedenen algenklassen. V. Einzelne vertreter der

Cryptophyceae, Euglenophyceae, Bacillariophyceae, Chrysophyceae und
Phaeophyphyceae. Arch. Mikrobiol. 73:77-89.
Harvey, H. R., S. A. Bradshaw, S. C. M. O'Hara, G. Eglinton, and E. D. S. Corner. 1988.

Lipid composition of the marine dinoflagellate Scrippsiella trochoidea. Phytochem. 27:1723-
1729.
Jeffrey, S. W. 1972. Preparation and some properties of crystalline chlorophyll C1 and C2

from marine algae. Biochem. Biophys. Acta 279:12-33.
Jeffrey, S. W. 1980. Algal pigment systems. In P. Falkowski, ed., Primary Productivity in the
Sea, pp. 33-58. New York: Plenum Press.
Jeffrey, S. W. 1981. An improved thin-layer chromatographic technique for marine

phytoplankton pigments. Limnol. Oceanogr. 16:191-197.
Jeffrey, S. W. and G. F. Humphrey. 1975. New spectrophotometric equations for determining

chlorophylls a, b, c in higher plants, algae, and natural phytoplankton. Biochem. Physiol.
Acta 167:191-194.
Kennett, J. P. and N. L. Penrose. 1978. The occurrence of fossil holocene seaweed and
attached calcareous polychaetes in an anoxic basin, Gulf of Mexico. Nature 276:172-173.
Larkum, A. W. D. and J. Barrett. 1983. Light-harvesting processes in algae. Adv. Bot. Res.
10:1-219.
Lewey, S. A. and J. Gorham. 1984. Pigment composition and photosynthesis in Sargassum

muticum. Mar. Biol. 80:109-115.
Macko, S. A. 1981. Stable nitrogen isotope ratios as tracers of organic geochemical

processes. Ph.D. dissertation, Univ. of Texas, Austin, Texas.
Mantoura, R. F. C. and C. A. Llewellyn. 1983. The rapid determination of algal chlorophyll

and carotenoid pigments and their breakdown products in natural waters by reverse-phase
high-performance liquid chromatography. Anal. Chim. Acta 151:297-314.
Martin, R. G. 1980. Distribution of salt structures in the Gulf of Mexico: map and description
text. USGS Open File Report 80-1213.
McKee, T. R. and B. R. Sidner. 1976. An anoxic high salinity intraslope basin in the
Northwest Gulf of Mexico. In A. H. Bouma, ed., Beyond the Shelf Break. AAPG Mar. Geol.
Comm. Short Course 2:125.
Northam, M. A., D. J. Curry, R. S. Scalan, and P. L. Parker. 1981. Stable carbon isotope
ratio variations of organic matter in Orca Basin sediments. Geochim. Cosmochim. Acta
45:257-260.

Repeta, D. J. 1982. Transformations of carotenoids in the oceanic water column. Ph.D.

dissertation. Woods Hole, Woods Hole Oceanographic Institution.
Repeta, D. J. 1989. Carotenoid diagenesis in recent marine sediments: II. Degradation of

fucoxanthin to loliolide. Geochim. Acta 53:699-707.
Sackett, W. M., J. M. Brooks, B. B. Bernard, C. R. Schwab, H. Chung, and R. A. Parker.

1979. A carbon inventory for Orca Basin brines and sediments.
Shokes, 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.
Sidner, B. R., S. Gartner, and W. R. Bryant. 1977. Late Pleistocene geologic history of the
outer continental shelf and upper continental slope, northwestern Gulf of Mexico. Texas A &
M University, Dept. of Oceanography, Technical Report 77-5-T.
Sieburth, J. and J. T. Conover. 1965. Sargassum tannin, an antibiotic which retards fouling.
Nature 208:52-53.
Smith, D. J., G. Eglinton, and R. J. Morris. 1983. The lipid chemistry of an interfacial
sediment from the Peru Continental Shelf: fatty acids, alcohols, aliphatic ketones and
hydrocarbons. Geochim. Cosmochim. Acta 47:2225-2232.
Taylor, C. D., S. O. Smith, and R. B. Gagosian. 1981. Use of microbial enrichments for the
study of the anaerobic degradation of cholesterol. Geochim. Cosmochim. Acta 45:2161-
2168.
Tompkins, R. E. and L. E. Shephard. 1979. Orca Basin: depositional processes,

geotechnical properties and clay mineralogy of holocene sediments within an anoxic
hypersaline basin, northwestern Gulf of Mexico. Mar. Geol. 33:221-238.
Van Vleet, E. S. and J. G. Quinn. 1979. Diagenesis of marine lipids in ocean sediments.
Deep Sea Res. 26:1225-1236.
Wakeham, S. G., C. Lee, J. W. Farrington, and R. B. Gagosian. 1984. Biogeochemistry of

particulate organic matter in the oceans: results from sediment trap experiments. Deep Sea
Res. 31:509-528.
Wiesenburg, D. A., J. M. Brooks, and B. B. Bernard. 1985. Biogenic hydrocarbon gases and
sulfate reduction in the Orca Basin brine. Geochim. Cosmochim. Acta 49:2069-2080.



8. Geochemical Features of Organic Matter in Sediment

Cores from Lätzow-Holm Bay, Antarctica
Organic geochemical and geochronological studies of two marine sediment core samples near
Langhovde Glacier (sample No. 112802, Core B, 117cm long, water depth 648 m) and about 40 km
to the east of the Core B sample (sample No. 110306, Core A, 73 cm long, water depth 778 m) in
Lützow-Holm Bay, Antarctica, have been conducted to understand sedimentary sequences and the
distribution, sources, preservation, and destruction of organic matter in an Antarctic setting. The
sediments are composed mainly of poorly sorted glacial clay. Sedimentation rates in the Core A
sample varied from less than 1.7 mm/y for 0-5 cm, greater than 4.4 mm/y for 5-40 cm, and less than
0.01 mm/y for 40-50 cm depths. The vertical distributions of radioisotopic data and organic
compounds suggest considerable reworking by iceberg movements.
The total organic carbon (TOC) contents of the sediments were less than 0.21% of dry sample,
reflecting the low primary productivity in the region. Normal-alkanes with carbon chain length from n-
C13 to n-C37 and isoprenoids with i-C16, i-C18, i-C19 (pristane), i-C20 (phytane), i-C30 (squalane),
and i-C25 diene [after hydrogenation 2, 6, 10, 14-tetramethyl-7-(3-methylpentyl)-pentadecane] were
detected. The i-C25 diene was the dominant hydrocarbon, accounting for up to 80% of the total
hydrocarbons, and was present at extremely high concentrations (0.91 to 610 ppm as carbon/TOC).
It is hypothesized that the i-C25 diene is derived from ice algae (diatoms). Normal-alkanoic acids
ranging in carbon chain length from n-C10 to n-C32 were detected in the samples, in addition to n-
alkenoic and iso- and anteiso-alkanoic acids. The abundances of iso- and anteiso-alkanoic acids in
the Core A sample suggest that bacterial contributions to this sample are considerably greater than
those to the Core B sample. The sources of organic matter are mainly in situ phytoplankton and
zooplankton, including ice algae, and microorganisms, with small amounts of eroded recycled
sediments and the waxes of vascular plants. Although there is a decrease of the i-C25 diene with
depth in the sediment cores, indicative of the microbial degradation, these subzero low-temperature
environments are suitable for the preservation of organic matter.
Antarctic marine environments are characterized by the near absence of terrestrial organic matter,
since terrestrial vegetation is virtually absent in Antarctica. Only two species of vascular plants are
known to occur in the northern part of the Antarctic Peninsula (Greene et al. 1967). Antarctica,
being far removed from industrialized areas, is believed to be one of the least polluted regions
remaining on earth. Thus the composition of organic materials in Antarctic marine and lacustrine
environments can be expected to be very different from that encountered in the mid to lower
latitudes regions. Matsumoto (1989) summarized the features or organic components in Antarctic
lake waters and sediments. Various organic geochemical studies have also been carried out in the
Antarctic and sub-Antarctic marine environments, including the Ross Sea and adjacent regions
(Sackett et al. 1965; Sackett, Eadie, and Exner 1974; Sackett, Poag, and Eadie 1974; McIver 1975;
Wrenn and Beckman 1982; Sackett 1986a,b; Rapp, Kvenvolden and Golan-Bac 1987; Venkatesan
1988a,b), the Bransfield Strait (Whiticar, Suess and Wehner 1985; Venkatesan and Kaplan 1987;

Brault and Simoneit 1988; Venkatesan 1988a,b), and other localities (Mackie, Platt, and Hardy
1978; Platt 1979; Platt and Mackie 1979; Handa and Tanoue 1983; Matsueda and Handa 1986).
Hydrocarbons, such as n-alkanes, pristane, phytane, steranes, and triterpanes; and fatty acids,
including n-alkanoic, n-alkenoic, and iso- and anteiso-alkanoic acids, are ubiquitous compounds in
living organisms. These compounds are commonly distributed throughout geological environments
and have been used to estimate the sources and maturation of organic matter, as well as to
characterize sedimentary depositional environments. Also, branched-C25 and branched-C30
alkenes, with or without rings, and in particular an isoprenoid-C25 diene (i-C25 diene) have been
reported for the water column, fecal pellets, and marine and lacustrine sediments, although their
sources are still unclear (Farrington et al. 1977; Barrick, Hedges and Peterson, 1980; Requejo and
Quinn 1983; Requejo et al. 1984; Matsueda and Handa 1986; Robson and Rowland 1986; Rowland
et al. 1985, 1990).
For the Antarctic and sub-Antarctic marine sediments described above, nonvolatile hydrocarbons: n-
alkanes, pristane, phytane, triterpanes, and/or steranes have been studied by various researchers
(Mackie, Platt, and Hardy 1978; Platt and Mackie 1979; Matsueda and Handa 1986; Kvenvolden,
Golan-Bac, and Rapp 1987; Kvenvolden et al. 1987; Rapp, Kvenvolden, and Golan-Bac 1987;
Venkatesan and Kaplan 1987; Brault and Simoneit 1988; Venkatesan 1988a,b). Although the
distribution of fatty acids in Antarctic marine sediments is scarcely known, Venkatesan and Kaplan
(1987) and Venkatesan (1988b) have reported their distributions in sediments from the Bransfield
Strait and McMurdo Sound. Kvenvolden et al. (1987) reported that hydrocarbons in the Wilkes Land
continental margin and the western Ross Sea sediments are derived from multiple sources, i.e.,
recycled ancient organic matter, atmospheric transport of the waxes of vascular plants, and in situ
biological activity, whereas hydrocarbons and fatty acids in the Ross Sea and Bransfield Strait
sediments are derived nearby exclusively from in situ biological activity (Venkatesan and Kaplan
1987; Venkatesan 1988b).
No one, however, has reported organic components in Lützow-Holm Bay, Antarctica. Here we
report features of total organic carbon (TOC); total nitrogen (TN); hydrocarbons; n-alkanes;
isoprenoids, including a novel i-C25 diene; pentacyclic triterpanes and steranes; and fatty acids in
two sediment core samples in Lützow-Holm Bay, Antarctica. These distributions are discussed in
relation to sedimentary sequences, primary productivity, source organisms, and the preservation
and destruction of organic matter. Also, 234Th, 214Pb, 210Pb, and 137Cs activities in the Core A
sample were analyzed by a gamma spectrometric method to estimate sedimentation rates.
Sampling Sites and Samples
The geographical features of the sea floor in Lützow-Holm Bay have been studied by the Japanese
Antarctic Research Expedition (JARE) members (Yoshida, Murauchi, and Fujiwara 1964; Fujiwara
1971; Moriwaki 1975, 1979; Moriwaki and Yoshida 1983). According to Moriwaki and Yoshida
(1983), a large depression in the central part of the bay is divided along an east/west transect by a
central trough offshore of the Shirase Glacier (figure 8.1). Many drowned glacial troughs exist in the
eastern part of the bay. Large troughs offshore of the Shirase and Telen Glaciers were probably
formed by glaciation along faults, whereas other conspicuous troughs were eroded selectively by
ice streams along the strike of gneissic foliation of basement rocks. The sedimentological studies

reveal that the sediments are mainly found in the central depression, and their thickness is usually
less than 1 m (Fujiwara 1971; Moriwaki 1975, 1979).
More than 30 sediment samples including cores were taken from the depression of the floor of the
troughs in Lützow-Holm Bay by means of a gravity corer (60 mm i.d.), piston corer (60 mm i.d.), and/
or pilot corer (36 mm i.d.) in November 1981 by the 22th JARE wintering members. Two cores were
studied, one off the Langhovde Glacier (sample No. 112802, Core B, 117 cm long, water depth 648
m) and approximately 40 km east of the Core B sampling site (110306, Core A, 73 cm long, water
depth 778 m) with a gravity corer (figure 8.1). These glacial marine sediments are mainly composed
of olive gray, poorly sorted clay with sponge spicules; small amounts of brachiopoda (Laqueus sp.
and Terebulatulina sp.) and balanus (Megabalances sp.); and subangular gravel-size gneiss. The
temperature of the sediment cores was near 0°C. Samples were stored frozen at -20°C until
analysis in July 1985 and 1988. Sediment cores were sectioned into 5 cm lengths.
Analyses
The analyses of 234Th, 214Pb, 210Pb, and 137Cs activities were carried out by the method of E.
Matsumoto (1987). The sediment samples were dried at 110°C and pulverized to fine powder. The
powder samples (each ca. 10 g) were sealed in plastic containers for dating. The gamma rays from
234Th (63 and 93 keV), 210Pb (47 keV), 214Pb (242, 295, and 352 keV), and 137Cs (662 keV) were
measured by a coaxial gamma-X-type pure Ge detector connected to a spectroanalyzer. The
calibration was done with an NBL uranium ore reference material.
TOC and TN were determined by a high-temperature combustion method using a CHN analyzer
(Yanako MT2 CHN Corder) after treatment with 6N hydrochloric acid to remove carbonate carbon.
Analytical methods for hydrocarbons and fatty acids are detailed elsewhere (Matsumoto, Torii, and
Hanya 1979; Matsumoto et al. 1987; Matsumoto, Watanuki, and Torii 1989). Briefly, wet sediments
(20-40 g) were refluxed with 40 mL 0.5M potassium hydroxide in methanol (80°C, two hours) and
centrifuged. The supernatant liquids and residues were acidified and extracted with ethyl acetate.
These extracts were chromatographed through a silica gel column (160 x 5 mm i.d., 100 mesh, 5%
water). Hydrocarbons and fatty acids were eluted with hexane and benzene: ethyl acetate (95:5),
respectively. Fatty acids were methylated with 14% boron trifluoride/methanol (80°C, two hours). To
elucidate the structure of i-C25 diene, the hydrocarbon fraction was hydrogenated with hydrogen
gas in contact with a platinum dioxide catalyst in hexane (Matsumoto, Watanuki, and Torii 1989).
Fatty acid methyl esters and major hydrocarbons were determined with a Shimadzu GC-8A gas
chromatograph equipped with an FID detector connected to a CR-3A Chromatopac data processing
system. Gas chromatographic conditions were as follows: split ratio of 1/30, Hewlett Packard fused
silica capillary column (Ultra performance No. 2: 5% phenyl methyl silicone, 20 m x 0.20 mm i.d.,
film thickness 0.11 m) was programmed from 100° to 300° at 10°C/min; injector and detector
temperatures, 330°C; helium carrier gas, 1.5 mL/min. Normal-alkanes, triterpanes, and steranes
were determined by mass fragmentograms of m/z 71, 191, and 217, respectively, with a Shimadzu
GCMS-QP1000 gas chromatograph-mass spectrometer (GC-MS). Also, mass fragmentography of
molecular ions of triterpanes and triterpenes for C27 to C35 and steranes for C27 to C29 carbon
numbers were run for their identification. The GC-MS conditions are as follows: cooled on-column
injection mode; a fused silica capillary column (J & W Scientific Co. DB-5, 30 m x 0.32 mm i.d., film
thickness 0.25 m) temperature programmed from 70° to 120°C at 25°C/min and then 120° to 310°
C at 6°C/min for n-alkanes, and from 70° to 220°C at 15°C/min and then from 220° to 310°C at 2°C/

min for triterpanes and steranes. The temperatures of the molecular separator and the ion source
were maintained at 320° and 250°C, respectively. The flow rate of helium carrier gas was 4.3 mL/
min. Mass spectra (m/z 50-600) were taken at 70 eV, continuously at 1.3 sec intervals. The
identification of each organic compound was made by comparison of retention sequences and
mass spectra with those of authentic compounds and published data (e.g., Wardroper et al. 1977;
Matsumoto, Torii, and Hanya 1979; Matsumoto et al. 1987; Seifert and Moldowan 1979; Philp 1986;
Venkatesan 1988b; Rowland et al. 1985, 1990). Normal-alkanes, triterpanes-triterpenes, and
steranes-diasteranes were quantified by the measurement of peak heights on the m/z 71, 191, and
217 mass fragmentograms, respectively.
Geochronology and Sedimentary Sequences
The naturally occurring 238U decay series and fallout 137Cs in the sediment column have been
widely used in determining sedimentation rates. The analytical results of 234Th, 214Pb, 210Pb, and
137Cs for the Core A sample are listed in table 8.1. 234Th and 214Pb correspond to 238U and 226Ra,
respectively. The data are insufficient to discuss in detail the sedimentation rates. Since no 137Cs
fallout due to the nuclear weapon testing was observed below 5 cm, sediments below 5 cm should
be older than 1950, giving a sedimentation rate of less than 1.7 mm/y for 0-5 cm section. In the
uppermost 40 cm, 238U decay series are not in equilibrium with 238U. Excess 210Pb
(210Pb/214Pb>1) are observed above 40 cm. This result shows that sediments above 40 cm are
mainly composed of sedimentary matter younger than 1870, and the sedimentation rate for 5-40 cm
is greater than 4.4 mm/y. Below 50 cm, 238U decay series are in equilibrium with 238U. This means
that the sediments below 50 cm are older than 10,000 y b.p. and thus the sedimentation rate for 40-
50 cm is calculated to be less than 0.01 mm/y. A depositional hiatus may have occurred at 40-50
cm.
During September 23, 1981-October 26, 1981, a sediment trap experiment in Lützow-Holm Bay
near Syowa Station revealed a very low sedimentation rate of 0.0087 mm/y (K. Sasaki, unpublished
results). This result suggests that sedimentation rates in the bay show considerable seasonal
variation. That is, biological activity of these months is very small, as evidenced by chlorophyll a
concentration (Hoshiai 1969). Also, iceberg movements mainly occur in summer, because the
surface of the bay is mostly frozen in winter. It is most likely that sedimentation in the bay occurs
mainly in the austral late spring-early autumn periods.
The geochronological data for the Core A sample reveal that the sedimentary sequences are not
smooth, and it is characterized by large changes in sedimentation rates. This may be considered in
the context of the sedimentation processes in the trough from which the core was collected. At least
two possible sedimentation processes can occur at these locations: (1) normal low sedimentation,
(2) slumping from the valley walls due to iceberg strike, etc.
The Core A sampling site is located just in front of the Langhovde Glacier, and thus icebergs often
strike the valley wall and provide sediments intermittently to the valley floor. The presence of small,
poorly sorted gravels in the sediment samples indicates that these processes have occurred in the
trough. The TOC and TN values in the Core A sample generally decreased with depth, especially
dramatically below 50 cm, probably reflecting a depositional hiatus, although the C/N atomic ratios

were near constant (5.9-7.0, figure 8.2). Also, hydrocarbon and fatty acid concentrations generally
decreased with increasing depth, but their peaks were observed at about 30 cm depth, as
discussed below (figures figure 8.2, 8.3, 8.4, tables 8.2 and 8.3).
Consequently, sedimentary sequences in the Core A sample could be summarized as follows: (1)
low sedimentation prevailed at depth below 40 cm, (2) rapid sedimentation due to slumping may
have occurred frequently at about 30 cm depth, (3) this was followed by slow sedimentation plus
some contribution from slumping. The fatty acid maxima (not pronounced) and hydrocarbon
maxima at 30 cm depth may be due to slumped organic matter. All the sediment sections of the
Core A sample contain thermally epimerized steranes, indicating that erosion and/or reworking have
taken place throughout the sedimentary sequence, as discussed below (see table 8.9).
In contrast, the TOC and TN concentrations in the Core B sample decrease to a depth of 50 cm but
increase deeper in the section (figure 8.5). Dramatic decreases in the C/N ratios are observed
below 60 cm depth, suggesting a significant change in the sedimentary inputs. No hydrocarbon
concentration maximum is observed in this section, (figure 8.6). The high fatty acid concentration
below 100 cm (figure 8.7) may be due to changes in organic matter sources. The reasons for this
are unclear. Also, hydrocarbon and fatty acid concentrations generally decrease with depth with
maxima at 30 cm depth, as in the case of the Core A sample (figures 8.6 and 8.7; tables 8.4 and
8.5).
These results reveal that similar processes have occurred in both Core A and Core B sampling sites
and suggest that significant changes in climatic conditions took place at about 1890 in the Lützow-
Holm Bay region.
The sedimentation rate (>3.6 mm/y) obtained for 0-40 cm of the Core A sample is generally similar
to those for the Bransfield Strait (2.7 mm/y, Ventkatesan and Kaplan 1987) and for sub-Antarctic
marine sediments from King Edward Cove (54° 17'S, 36° 30'W; 4.2 mm/y; Platt 1979) and South
Georgia (54°S, 37°W; 4.0 mm/y; Platt and Mackie 1979) but much greater than that in the open
ocean sediments from the Bering Sea (57° 02.9'N, 176° 57.4'W; 0.37-0.90 mm/y; Tanoue and
Handa 1980).
Organic Carbon in Relation to Primary Productivity
Photosynthetic rates and chlorophyll a concentrations (a marker of phytoplankton biomass) in the

Southern Ocean have been studied by many researchers. Those in the coastal and inshore regions
of Antarctica are very high, while those in the Pacific, Atlantic, and Indian oceanic regions are
equivalent to those of other areas in the mid and lower latitudes of the world ocean (e.g., El-Sayed
1968, 1970; El-Sayed and Turner 1977; Fukuchi 1980). The chlorophyll a concentrations of the
Southern Ocean are lower than those in the Bering Sea, which is one of the more productive seas
in the world (Fukuchi 1980). Also, photosynthetic activity in the high-latitude areas is limited to the
late spring-early autumn periods, and the available measurements have been mainly obtained in
the productive season.
Hoshiai (1969) studied seasonal variations of chlorophyll a concentrations, chlorinity, dissolved

oxygen, and pH at several depths at the Kitano-seto Strait near Syowa Station as a function of solar
radiation. The chlorophyll a concentration increased with increasing solar radiation and reached a

maximum value of 60 mg/m2 at the end of January under 24-hour exposure to solar radiation.
Chlorophyll a decreased with a decrease in solar radiation and reached the lowest value of 0.1 mg/
m2 in the middle of August under total darkness. Thus, organic production in the high-latitude areas
averaged over the year is considerably lower than suggested by data from the productive season.
It is known that most organic carbon produced by photosynthetic activity in the water column is
destroyed before sedimentation. Only 2% of the annual influx of organic carbon to the water column
is calculated to be preserved in the sediments of the Aleutian Basin (Tanoue and Handa 1980).
However, sediment TOC concentrations tend to reflect the overlying water's primary productivity.
TOC concentrations in surface marine sediments from various areas in the Northern Hemisphere
are discussed by Tanoue and Handa (1980). Data for several marine sediments from the Northern
and Southern Hemispheres are summarized in table 8.6, together with those in the Lützow-Holm
Bay sediments. The TOC concentrations in the surface sediments of the Core A (0.21%) and the
Core B (0.12%) samples, as well as in the underlying sediments (0.069-0.20%) in Lützow-Holm
Bay, are generally much lower than those in other marine sediments, including the Ross Sea.
However, our TOC concentrations are similar to those in the Wilkes Land continental margin
(Kvenvolden et al. 1987).
Although the Bering Sea is a productive ocean (e.g., Hodd and Kelley 1974), sediment TOC
concentrations are low compared with other marine sediments in the Northern Hemisphere (table
8.6). However, these values are generally similar to those in the Ross Sea sediments. El-Sayed,
Biggs, and Holm-Hansen (1983) reported that primary productivity in McMurdo Sound of the Ross
Sea during the austral summer approaches 1 gC/m2/d. The relatively high TOC concentrations in
the Ross Sea sediments may be a reflection of the high productivity of this region. In contrast, the
low TOC concentrations in the Lützow-Holm Bay sediments suggest that the primary productivity of
the bay is small. It is likely that yearly primary productivity in oceanic environments of the high-
latitude regions is not as high as previously reported.
Features of Hydrocarbons and Fatty Acids
Normal-alkanes with carbon chain length from n-C13 to n-C37 are detected in the gas
chromatogram of the hydrocarbon fraction from the extract of the Core B sediment (25-30 cm
section), together with isoprenoid alkanes i-C16, i-C18, i-C19 (pristane); i-C20 (phytane) and i-C30
(squalane); and branched-C25 alkene, which was identified as an i-C25 diene as discussed below
(figure 8.8; tables 8.2 and 8.4). Also, an unresolved complex mixture (UCM) of hydrocarbons was
present in some sediment sections. The branched-C25 alkene has characteristic peaks at m/z 55,
69 (base peak), 207, 235, 263, 266, 320, and 348 (M+). This mass spectrum is very similar to that of
an i-C25 diene (Requejo and Quinn 1983; Requejo et al. 1984; Rowland et al. 1985) or a branched-
C25 monocyclic monoene from the McMurdo Sound sediments (Venkatesan 1988b). After
hydrogenation intense peaks changed into m/z 57 (base peak), 71, 85, 238, and 239, but no
molecular ion was observed. It is virtually identical to that of i-C25 diene (Requejo and Quinn 1983)
but very different from the branched-C25 monocyclic monoene observed in the McMurdo Sound
sediments. The retention indices (RI) of i-C25 diene before and after hydrogenation were 2082 and
2100, respectively, which are also identical with that of Rowland et al. (1985). Thus we identified
this as the carbon skeleton for the i-C25 diene. The structure of this hydrogenated compound has
been confirmed to be 2, 6, 10, 14-tetramethyl-7-(3-methylpentyl)-pentadecane (Rowland et al.
1990). Rowland et al. (1990) revealed, however, that i-C25 monocyclic monoene from the McMurdo

Sound and Bransfield Strait sediments (Venkatesan and Kaplan 1987; Venkatesan, 1988b) is not
monocyclic and is the result of partial hydrogenation of our i-C25 diene. As suspected, branched
hydrocarbon (figure 8.8) has characteristic mass fragments at m/z 57, 71, 85, 99, etc. but could not
be identified owing to the small quantity. No branched-C30 cyclic alkenes were detected in any of
the sediments.
The concentrations of total hydrocarbons (1056 ng/g of dry sediment) and i-C25 diene (826 ng/g) in
the top (0-10 cm) of the Core B sample were an order of magnitude larger than those in the Core A
sample, in spite of the low TOC and total fatty acid concentrations (tables 8.2, 8.3, 8.4, 8.5).
Surprisingly, the i-C25 alkene concentration in the surface of the Core B sample is comparable with
the n-C16 alkanoic acid concentration, which is the most abundant fatty acid in this sample (table
8.5). Also, the i-C25 diene-C/TOC value (610 ppm) for the surface of the Core B sample is
extremely high, more than an order of magnitude greater than those previously reported (see table
8.9; Barrick, Hedges, and Peterson 1980; Requejo and Quinn 1983; Venkatesan and Kaplan 1987).
The concentrations of n-alkanes in the Core A and Core B samples ranged from 8.9 to 165 ng/g
(tables 8.2 and 8.4), which are similar values to those in the McMurdo Sound sediments
(Venkatesan 1988b) but considerably lower than those in other Antarctic and sub-Antarctic marine
sediments (Mackie, Platt, and Hardy 1978; Platt 1979; Platt and Mackie 1979; Venkatesan and
Kaplan 1987).
Generally n-alkanes in both cores showed biomodal distributions maximizing at n-C16-n-C19 and n-
C27 or n-C29 (tables 8.2 and 8.4). The most dominant n-alkanes in the Core A sample were n-C17,
n-C27, or n-C29, whereas those in the Core B sample were n-C16, n-C18, n-C19, n-C21, or n-C29.
These n-alkanes are considerably different from those in the sediment samples from the Wilkes
Land continental margin and the western Ross Sea; i.e., n-C19 is generally the most abundant
short-chain n-alkane, whereas n-C27 is usually the most abundant long-chain n-alkane (Kvenvolden
et al. 1987). Also, the n-alkane distributions reported here are significantly different from those in
the McMurdo Sound sediments (Venkatesan 1988b) but similar to those in the Bransfield Strait
sediments (Venkatesan and Kaplan 1987).
Isoprenoid-alkanes i-C16, i-C18, pristane, phytane, and squalane were found in all sections of both
cores (tables 8.2 and 8.4). Pristane and phytane are commonly present in various sedimentary
environments, but i-C16 and i-C18 are less common (Matsumoto 1989). These isoprenoid-alkanes,
except for squalane, are found in the sediment samples from the Wilkes Land continental margin
and the western Ross Sea (Kvenvolden et al. 1987), and Bransfield Strait sediment (Venkatesan
and Kaplan 1987; Brault and Simoneit 1988). Squalane is found in several immature marine
sediments (Brassell et al. 1981, Volkman and Maxwell 1986) and in inland hydrothermal sediments
in Japan (Matsumoto and Watanuki 1990). It also occurs in the polluted river waters and sediments
in Tokyo as a major hydrocarbon, derived probably from materials through human activities other
than petroleum-derived materials (Matsumoto and Hanya 1980; Matsumoto 1982).
The i-C25 diene was detected in all the sediments and represented a major component of the
hydrocarbons (tables 8.2 and 8.4). This is the first report of such a high abundance of i-C25 diene
(up to 80%) in natural environments. The i-C25 diene usually co-occurs with other branched-C25
cyclic alkenes, branched-C30 cyclic alkenes and/or other hydrocarbons (e.g., Requejo and Quinn
1983; Requejo et al. 1984; Barrick, Hedges, and Peterson 1986; Venkatesan and Kaplan 1987;
Rowland et al. 1990), although it is possible that these alkanes are not cyclic (Rowland et al. 1990).

Pentacyclic triterpanes, triterpenes, and moretanes with C27 to C35 carbon atoms, including 17
(H), 21 (H)- and 17 (H), 21 (H)-hopanes as well as (22R)-and (22S)-C31-C35 extended hopanes,
were detected in both cores (e.g., figure 8.9, table 8.7). The major compounds were 22, 29, 30-
trisnorhop-17(21)-ene; 17 (H), 21 (H)-30-norhopane; 17 (H), 21 (H)-hopane; neohop-13(18)-ene;
and/or hop-22(29)-ene (diploptene). These distributions are characterized by the coexistence of
triterpenes and mature triterpanes. This is very different from most mature petroleums and
sedimentary rocks of the world (e.g., Philp 1985, 1986) as well as sediments in the Wilkes Land
continental margin and the western Ross Sea (Kvenvolden et al. 1987) but is similar to the
Bransfield Strait sediments (Venkatesan and Kaplan 1987; Brault and Simoneit 1988). Hop-22(29)-
ene, which is often the major pentacyclic triterpane-type compound in recent sediments, is detected
in the McMurdo Sound and Bransfield Strait sediments (Venkatesan and Kaplan 1987; Venkatesan
1988a,b) as well as in Antarctic lacustrine sediments from Ace Lake of the Vestfold Hills (Volkman
et al. 1986), Lake Vanda (Matsumoto, Watanuki, and Torii 1987) and Lake Fryxell (Matsumoto,
Watanuki, and Torii 1989).
Steranes and diasteranes with C27 to C29 carbon atoms, involving (20R)- and (20S)-5 (H), 14 (H),
17 (H) as well as 5 (H), 14 (H), 17 (H) configurations, were detected in both cores (e.g., figure
8.10, table 8.8). These steranes and diasteranes were present in all the sediment sections and are
ubiquitous in petroleums and sedimentary rocks (e.g., Mackenzie et al. 1982; Philp 1985) but are
uncommon in recent sediments.
Normal-alkanoic acids in carbon chain length from n-C12 to n-C28 showed bimodal distributions
maximizing at n-C16 and n-C24, with a predominance of even-carbon numbers, together with n-
alkenoic and iso- and anteiso-branched acids (figure 8.11; tables 8.3 and 8.5). The dominant fatty
acid was n-C16 in all the sediment samples. Interestingly, long-chain n-alkanoic acids were found in
all the samples. The fatty acid distributions are generally similar to those in most contemporary
lacustrine and marine sediments.
The total concentration of fatty acids in the top of the Core A sample (0-5 cm) was 4530 ng/g and is
much higher than that in the Core B sample (2660 ng/g; tables 8.3 and 8.5)). These concentrations
are much lower than those in many marine sediments, including the Bransfield Strait (Venkatesan
and Kaplan 1987) but similar to those in the McMurdo Sound sediments (Venkatesan 1988b),
although they analyzed only solvent extractable free fatty acids.
Organic Sources
Normal-alkane distributions showed generally two maxima at n-C16-n-C19 and n-C25-n-C29,

suggesting that these n-alkanes originate from two different sources (tables 8.2 and 8.4). The
carbon preference indices (CPIHS) for short-chain n-alkanes (n-C15-n-C19) in the Core A sample
(1.0-1.5) are somewhat greater than unity, whereas those in the Core B sample (0.72-1.0) are near
or less than unity, suggesting that source materials in the two sediment cores are different (table
8.9). In contrast, the CPIHL values for long-chain n-alkanes (n-C21-n-C33) in the Core A sample
(1.2-1.9) are generally similar to those in the Core B sample (1.2-1.8). These results suggest that
major n-alkanes come from microorganisms, such as bacteria and some microalgae, and/or erosion
of sedimentary organic matter. The abundance of iso and anteiso-branched acids (discussed
below) is consistent with the proposition that these n-alkanes are derived from bacteria. However,
the occurrence of thermally epimerized triterpanes and steranes supports the presence of reworked

sedimentary organic matter, as discussed below (table 8.9).
A variety of microorganisms, including bacteria, fungi, and yeast species, is suggested as the
source for the predominance of even-carbon-numbered n-alkanes found in various recent and
Miocene sediments, encompassing marine and lacustrine environments, deposited under oxic and
anoxic conditions (Nishimura and Baker 1986; Grimalt and Albaigés 1987). Also, Nachman (1985)
reported n-alkanes with dominant even-carbon numbers in Antarctic organisms Euphausia
crystallorophias, Pleuragramma antarcticum, and Dissostichus mawsoni. The combinations of these
organisms may be responsible for the predominance of even-carbon n-alkanes in the Core B
sample.
Long-chain n-alkanes (n-C20-n-C35) are abundant in all the sediment sections (table 8.9). The
abundances of long-chain n-alkanes with odd-carbon numbers n-C25, n-C27, and n-C29 in both
cores (tables 8.2 and 8.4) may be due to the contribution of the waxes of vascular plants from the
mid and lower latitudes through atmospheric transport and/or oceanic transport, as is the case in
the western Ross Sea and the Wilkes Land continental margin (Kvenvolden et al. 1987), although
their contribution is very small.
Isoprenoid-alkanes i-C16, i-C18, pristane, and phytane can be derived from the phytyl side chain of
chlorophyll a (Didyk et al. 1978; Volkman and Maxwell 1986). In the sediments of Lützow-Holm Bay,
squalane is most likely derived from archaebacteria, such as methanogens and sulfur-oxidizing
bacteria, such as Sulfolobus sp. (Brassell et al. 1981; Volkman and Maxwell 1986), rather than
human products, because there is no marked pollution in this region. The pristane/phytane (Pr/Ph)
ratios for the Core A and Core B samples, ranging from 1.2 to 2.3, are considerably greater than
unity and suggest that these isoprenoids are formed in the oxic sedimentary environments of
Lützow-Holm Bay (Didyk et al. 1978). However, the cooccurrence of i-C25 diene and triterpenes as
well as of mature triterpanes and steranes indicates that pristane and phytane may have originated
from both in situ biological activity and eroded, reworked materials.
Rowland et al. (1985) found i-C25 diene in a green alga Enteromorpha prolifera and proposed that
this isoprenoid alkene in the natural environment is from algae. Interestingly, Matsueda and Handa
(1986) reported that a branched-C25 triene and tetraene, which gave, after hydrogenation, a mass
spectrum very similar to that of 2, 6, 10, 14-tetramethyl-7-(3-methylpentyl)-pentadecane, are
biological markers of the particulate matter excreted by zooplankton. The organic matter of these
fecal pellets was derived from diatoms growing in the surface water layers through zooplankton
grazing. An i-C25 diene and n-C21:6 alkene were found in ice algae (diatoms), such as Amphiprora
sp. and Nitzschia spp. and as the major hydrocarbons from McMurdo Sound (Nichols et al. 1988).
Ice algae are abundant in Lützow-Holm Bay (K. Watanabe, personal communication). Microfossil
analysis of the marine sediment core from the offshore of the Enderby Land close to Lützow-Holm
Bay shows the abundance of diatoms Fragilariopsis spp. (Fukushima and Suzuki 1966). Hence, in
our case ice algae (diatoms) are most likely sources of i-C25 diene. The absence of n-C21:6 alkene
in the sediments could be explained by destruction before sedimentation. This compound has been
detected in sinking particulate matter in the Southern Ocean, but concentrations decreased with
water depth, and it was not detected in the bottom sediments (Matsueda and Handa 1986). The
high concentrations of i-C25 diene and the high i-C25 diene-C/TOC values in the Core B sample
suggest that ice algae-like source organisms are more abundant at the Core B sampling site off the
Langhovde Glacier than at the Core A sampling site (tables 8.2, 8.4, and 8.9).

Hopane-type triterpanes and their precursors, including polyhydroxybacteriohopanes, are widely

distributed among bacteria and/or cyanobacteria but not in higher organisms (Ourisson, Albrecht,
and Rohmer 1979; Rohmer, Boivier-Nave, and Ourisson 1984; Philp 1985). Hop-22(29)-ene is
distributed in prokaryote cyanobacteria (Gelpi et al. 1970; Matsumoto et al., unpublished results)
and microorganisms (Bird et al. 1971; Rohmer, Boivier-Nave, and Ourisson, 1984). Thus, hop-22
(29)-ene and other triterpenes found in our samples most likely originated from microorganisms,
although the exact species are not yet clear.
The ratios of hop-17(21)-ene/hop-22(29)-ene and neohop-13(18)-ene/hop-22(29)-ene tend to

increase with depth in the Core A sample (table 8.9), suggesting diagenetic conversion from hop-22
(29)-ene (Brassell et al. 1980; Brault and Simoneit 1988). However, these tendencies in the Core B
sample are unclear (table 8.9) and may reflect more complex sedimentary sequence in this core.
Steranes are believed to come from sterols and reflect the sterol composition of the source
organisms (e.g., Seifert and Moldowan 1981; Mackenzie et al. 1982). In aquatic environments, C27
sterols, e.g., cholest-5-en-3 -ol, are generally derived from phytoplankton and zooplankton,
although C28 sterols, such as 24-methylcholest-5-en-3 -ol are abundant in certain diatoms,
dinoflagellates, and Prymnesiophycean algae (Volkman 1986; Nichols et al. 1989). Although
vascular plants are generally believed to be major sources of C29 sterols (e.g., 24-ethylcholest-5-en-
3- -ol) in lacustrine and coastal marine environments of the mid and lower latitudes, 24-ethylcholest-
5-en-3 -ol is frequently a major sterol in Antarctic lake waters and sediments from the McMurdo Dry
Valleys of Southern Victoria Land, around Lützow-Holm Bay, and Vestfold Hills of Princess
Elizabeth Land. This compound originated from some cyanobacteria and microalgae in the lakes
(Matsumoto, Torii, and Hanya 1982, 1983; Volkman 1986; Matsumoto 1989). Also, 24-ethylcholest-
5-en-3 -ol is a major sterol in a marine diatom Asterionella glacialis (Jones et al. 1987) and a
diatom community from McMurdo Sound (Nichols et al. 1989). Further, Nichols et al. (1987)
reported that raphidophyte flagellates Heterosigma akashiwo and Chattonella antiqua contain 24-
ethylcholest-5-en-3 -ol as the most abundant sterol. These findings suggest that many kinds of
microalgae and cyanobacteria are important sources of C29 sterols in the world.
(20R)-24-Ethyl-5 (H),14 (H),17 (H)-cholestane (C29 sterane) is the most dominant sterane in the
Core A sample except for 30-35 cm section, whereas (20R)-5 (H),14 (H),17 (H)-cholestane (C27
sterane) is the most abundant sterane in the Core B sample except for the 20-25 cm section (table
8.9). These results indicate again that source materials in two sediment cores differ considerably,
as evidenced by the features of i-C25 diene and short-chain n-alkanes. Of special interest is the
abundance of C29 sterane in the Core A sample, because vascular plants are absent in the
surroundings of the studied sites. These steranes can be attributed to some microalgae and/or
cyanobacteria in the sedimentary environments. However, since the occurrence of steranes in living
organisms is uncommon, these are probably derived from eroded, reworked materials.
Biologically synthesized hopane-type triterpanes generally have the (22R)-17 (H),21 (H)
configurations, but with increasing thermal maturation, (22S)-17 (H),21 (H) configurations appear
until an equilibrium ratio (22S/22R) of about 1.5 is reached (e.g., Mackenzie et al. 1982; Suzuki
1984). Also formed are the moretanes with a 17 (H),21 (H) configuration (e.g., Philp 1985, 1986).
On the other hand, naturally occurring precursor sterols generally possess the (20R)-5 (H),14
(H),17 (H) configuration, which converts into 20S until an equilibrium ratio (20S/20R) of
approximately 1.2 is reached with thermal maturation (e.g., Seifert and Moldowan 1981; Mackenzie
et al. 1982).

Epimerized 17 (H),21 (H)-hopanes were found in all the sediment samples, and thus the 17 (H),21
(H)/17 (H),21 (H)-hopane ratios in the Core A and Core B samples vary from 1.6 to 3.3 and from
0.79 to 10, respectively, suggesting that source materials change largely with depth (table 8.9).
Also, (22S)-C31-C35 hopanes were only found in the Core B sample, whereas (20S)-C27-C29
steranes were detected in all the samples. The (22S/22R)-C32 hopane ratios ranged from 0.75 to
1.5, while the (20S/20R)-C29 sterane ratios for the Core A and Core B samples changed from 0.18
to 0.40 and from 0.20 to 0.97, respectively. The specific heat sources in Lützow-Holm Bay are not
known. Thus, these results indicate that more mature organic matter due to reworking is
contributing significantly to the sedimentary environment. Especially, the near thermal equilibrium
values (1.5) of the (22S/22R)-C32 hopane ratios for the depths between 0 and 20 cm in the Core B
sample together with the occurrence of the UCM hydrocarbons suggest that these surface sediment
sections are influenced by local sources of reworked mature organic matter, rather than widely
occurring petroleum seeps or pollution.
Short-chain n-alkanoic acids (n-C10-n-C19) occur in most living organisms, while long-chain n-
alkanoic acids (>n-C20) are mainly present in the waxes of vascular plants. Iso- and anteiso-methyl
branched alkanoic acids are found in bacterial lipids (Kaneda 1967; Matsuda and Koyama 1977;
Gillan and Johns 1986). Vascular plants are absent in the areas studied, and thus fatty acids
probably originate from phytoplankton and zooplankton and from microorganisms, such as bacteria
and fungi, although it is possible that long-chain n-alkanoic acids are derived from long-range
transport of the waxes of vascular plants from the mid and lower latitudes.
Long-chain n-alkanoic acids are found in Antarctic lake sediments from the McMurdo Dry Valleys,
Lützow-Holm Bay region, and Vestfold Hills (Matsumoto, Torii, and Hanya 1979; Matsumoto et al.
1981; Matsumoto, Torii, and Hanya 1984; Matsumoto, Watanuki, and Torii 1987; Volkman et al.
1988; Matsumoto 1989). These long-chain n-alkanoic acids are believed to be derived-1in situ from
microorganisms, such as bacteria and fungi. The contribution ofatmospheric transport of the waxes
of vascular plants is thought to be negligible. Normal-alkanes n-C27, n-C29, and n-C31 with
vascular plant origin were present at very low concentrations in all sediment sections (tables 8.2,
8.4). Therefore, long-chain n-alkanoic acids found in these sediments are mainly attributed to
microorganisms, e.g., bacteria and fungi. Also, diatoms are suggested to be a source of long-chain
n-alkanoic acids (Venkatesan and Kaplan 1987). The high CPIFS values (7.9-19) for short-chain n-
alkanoic acids (n-C14-n-C18) and CPIFL values (3.3-6.3) for long-chain n-alkanoic acids (n-C20-n-
C28) in both core samples support the hypothesis that these fatty acids mainly originate from recent
organic matter (table 8.9).
Fatty acid composition showed that branched alkanoic acids in the Core A sample were much more
abundant that those in the Core B sample throughout the length of the core. This suggests that
bacterial contributions to the Core A sample are much higher than those to the Core B sample
(figures 8.12 and 8.13).
Preservation and Destruction of Organic Matter
Sediments in the Southern Ocean are expected to provide appropriate environments for the
preservation of organic matter, because of their subzero temperatures. Sediment trap experiments
in the Pacific sector of the Southern Ocean (61° 34.1'S, 150° 23.3'E) show that some degradation
and transformation of hydrocarbons occur during the sinking of particulate matter in the water

column. Pristane and n-C20:6 alkene tend to decrease with depth, while phytane and n-alkanes
tend to increase with depth (Matsueda and Handa 1986). Thus organic components produced in the
overlying water column of Lützow-Holm Bay may be considerably altered before sedimentation.
The concentrations of TOC, TN, hydrocarbons, and fatty acids in both cores generally decrease
with increasing depth (figures 8.2, 8.3, 8.4, 8.5, 8.6, 8.7). However, changes in fatty acid-C/TOC
ratios and n-alkenoic acid percentages with increasing depth are generally small (table 8.9, figures
8.12 and 8.13). Although it is difficult to estimate vertical changes in primary productivity based on
present data, the decrease of i-C25 diene-C/TOC values in both cores as well as the increase of
neohop-13(18)-ene/hop-22(29)-ene and hop-17(21)-ene/hop-22(29)-ene ratios in the Core A
sample suggest that microbial alteration/transformation of labile organic matter has taken place in
subzero temperatures (table 8.9). High abundances of branched fatty acids suggest bacterial
activity is an important process in both cores. Changes in the C/N atomic ratios (6-8) at depths
above 60 cm in both cores are small, suggesting that either remineralization of organic matter is
small or remineralized organic-nitrogen may be preserved as ammonium salts in the sediments
(Tanoue and Handa 1980). Also, Venkatesan and Kaplan (1987) reported that the presence of
labile alkenes, n-alkenoic acids, and hop-22(29)-ene to a depth of 10 m (3600 b.p.) in the Bransfield
Strait sediment core suggested that the subzero water temperatures and the anoxic environmental
conditions of this basin preserve organic matter. Furthermore, Kvenvolden et al. (1987) suggest that
neither the open ocean of the Wilkes Land continental margin nor the more restricted Ross Sea
shelf provides particularly favorable environments for the long-term preservation of organic matter.
It is more likely, therefore, that the diagenesis or remineralization rates in the Antarctic marine
sediments are generally lower than those in the mid and lower latitudes, and thus organic matter
remains relatively well preserved in the Antarctic marine sediments.
TOC, TN, hydrocarbons, and fatty acids in two sediment cores (Core A and Core B) from Lützow-
Holm Bay, Antarctica, were studied and discussed in relation to sedimentary sequences and
production, features, sources, preservation, and destruction of organic matter:
1. Sedimentation rates in the Core A sampling site varied greatly from less than 1.7 mm/y for 0-
5 cm, greater than 4.4 mm/y for 5-40 cm, to less than 0.01 mm/y for 40-50 cm depths.
2. Sedimentary sequences in the troughs of the bay may be characterized by erosion or

reworking due to slumping by iceberg strikes in addition to normal slow sedimentation.
3. The concentrations of TOC and TN in this bay are considerably lower than those other
marine sediments, possibly reflecting low organic matter production.
4. Normal-alkanes in carbon chain length from n-C13 to n-C37 were found with bimodal
distributions at very low concentrations. These n-alkanes are mainly derived from
microorganisms and recycled organic matter.
5. A novel i-C25 diene was the most prominent hydrocarbon (up to 80%) of the hydrocarbons in
all sediment samples, and it is probably derived from ice algae (diatom).
6. Normal-alkanoic acids are found in carbon chain length from n-C10 to n-C28 with the

bimodal distributions maximizing at n-C16 and n-C24 and with the dominant component
being n-C16. Plankton, including ice algae, is a suspected source.
7. The abundance of branched (iso and anteiso) alkanoic acids at the Core A sample implies
that bacterial contributions in the offshore of the Langhovde Glacier are much higher than at
the Core B sampling site.
8. Organic matter in the Core A and Core B sampling sites are served from multiple sources,
including phytoplankton and zooplankton; microorganisms, such as bacteria and fungi; and
small amounts of reworked sediments and the waxes of vascular plants. The relative
importance of these source materials varies with time and location.
9. The remineralization and transformation rates of organic matter at subzero temperatures in

the Antarctic marine environments are generally small, and thus the environments are
relatively favorable for preservation of organic matter. Acknowledgments The authors are
greatly indebted to the 22d JARE wintering members led by Professor Yoshio Yoshida,
National Institute of Polar Research, Japan, for their kind cooperation and support in
collecting sediment samples. Also we thank Dr. J. K. Volkman, CSIRO Division of
Oceanography, Australia, and anonymous reviewers for their useful comments on the
manuscript.
References
Barrick, R. C., J. I. Hedges, M. L. Peterson. 1980. Hydrocarbon geochemistry of the Puget

Sound region-I. Sedimentary acyclic hydrocarbons. Geochim. Cosmochim. Acta 44:1349-
1362.
Bezrukov, P. L. 1960. The bottom sediments of the Okhotsk Sea. Trudy Inst. Okeanol. 32:15-
95 (in Russian, cited in Tanoue and Handa 1980).
Bird, C. W., J. M. Lynch, S. J. Pirt, and W. W. Reid. 1971. The identification of hop-22(29)-
ene in prokaryotic organisms. Tetrahedron Lett. 34:3189-3190.
Bordovskiy, O. K. 1965. Accumulation and transformation of organic substances in marine

sediments. 2. Sources of organic matter in marine basins. Mar. Geol. 3:5-31.
Brassell, S. C., P. A. Comet, G. Eglinton, P. J. Isaacson, J. McEvoy, J. R. Maxwell, I. D.

Thompson, P. J. C. Tibbetts, and J. K. Volkman. 1980. The origin and fate of lipids in the
Japan Trench. In A. G. Douglas and J. R. Maxwell, eds., Advances in Organic Geochemistry
1979, pp. 375-392. Oxford: Pergamon.
Brassell, S. C., A. M. K. Wardroper, I. D. Thomson, J. R. Maxwell, and G. Eglinton, 1981.

Specific acyclic isoprenoids as biological markers of methanogenic bacteria in marine
sediments. Nature 290:693-696.
Brault, M. and B. R. T. Simoneit. 1988. Effects of hydrothermal alteration on organic matter

in sediments from the Bransfield Strait, Antarctica. Org. Geochem. 13:697-705 (Advances in

Organic Geochemistry 1987).
Didyk, B. M., B. R. T. Simoneit, S. C. Brassell, and G. Eglinton, 1978. Organic geochemical

indicators of paleoenvironmental conditions of sedimentation. Nature 272:216-222.
El-Sayed, S. Z. 1968. Primary productivity. In V. C. Bushnell and S. Z. El-Sayed, eds.,

Primary Productivity and Benthic Marine Algae of the Antarctic and Subantarctic, pp. 1-6.
New York: Am. Geogr. Soc. (Antarctic Map Folio Series, folio 10).
El-Sayed, S. Z. 1970. On the productivity of the Southern Ocean (Atlantic and Pacific
sectors). In M. W. Holdgate, ed., Antarctic Ecology 1:119-135. London: Academic.
El-Sayed, S. Z. and J. T. Turner. 1977. Productivity of the Antarctic and tropical/subtropical

regions: a comparative study. In M. D. Dumbar, ed., Polar Oceans, pp. 463-503. Calgary:
Antarctic Inst. North Am.
El-Sayed, S. Z., D. C. Biggs, and O. Holm-Hansen. 1983. Phytoplankton standing crop,

primary productivity, and near-surface nitrogenous nutrient fields in the Ross Sea,
Antarctica. Deep Sea Res. 30:871-886.
Erdman, J. G. and K. S. Schorno. 1978. Geochemistry of carbon: Deep Sea Drilling Project
Legs 42A and 42B. In D. A. Ross, et al. ed., Initial Reports. DSDP(XLII).2, pp. 717-721.
Farrington, J. W., N. M. Frew, P. M. Gschwend, and B. W. Tripp. 1977. Hydrocarbons in

cores of northwestern Atlantic coastal and continental margin sediments. Est. Coastal Mar.
Sci. 5:793-808.
Fujiwara, K. 1971. Soundings and submarine topography of the glaciated continental shelf in
Lützow-Holm Bay, East Antarctica. Antarct. Rec. 41:81-103 (in Japanese with English
abstract).
Fukuchi, M. 1980. Phytoplankton chlorophyll stocks in the Antarctic Ocean. J. Oceanogr.

Soc. Jpn. 36:73-84.
Fukushima, H. and K. Suzuki. 1966. Microfossil analysis of the core collected in the offing of
Enderby Land, Antarctica. Antarct. Rec. 26:10-17 (in Japanese with English abstract).
Gelpi, E., H. Schneider, J. Mann, and J. Oró. 1970. Hydrocarbons of geochemical

significance in microscopic algae. Phytochem. 9:603-612.
Gillan, F. T. and R. B. Johns. 1986. Chemical markers for marine bacteria: fatty acids and
pigments. In R. B. Johns, ed., Biological Markers in the Sedimentary Record. Methods in
Geochemistry and Geophysics, pp. 291-309, Amsterdam: Elsevier.
Gorshkova, T. I. 1960. Sediments of the Norwegian Sea. In Sbornik Dokladov Sovetskikh

Geologov "Morskaya Geologya". Moscow: Izd Akad. Nauk SSSR (in Russian, cited in

Tanoue and Handa 1980).
Greene, S. W., J. L. Gressitt, D. Koob, G. A. Llano, E. D. Rudolph, R. Singer, W. C. Steere,

and F. C. Ugolini. 1967. In Terrestrial Life of Antarctica, V. C. Bushnell, ed., pp. 1-24. New
York: Am. Geogr. Soc. (Antarctic Map Folio Series 5).
Grimalt, J. and J. Albaigés, 1987. Sources and occurrence of C12-C22 n-alkane distributions
with even carbon-number preference in sedimentary environments. Geochim. Cosmochim.
Acta 51:1379-1384.
Handa, N. and E. Tanoue. 1983. Organic compounds of the suspended particles in the
Pacific sector of the Southern Ocean. Mem. Natl Inst. Polar Res., special issue 27:50-63.
Hodd, D. W. and E. J. Kelley, eds. 1974. Oceanography of the Bering Sea. Fairbanks: Univ.
Alaska.
Hoshiai, T. 1969. Seasonal variation of chlorophyll-a and hydrological conditions under sea
ice at Syowa Station, Antarctica. Antarct. Rec. 35:52-67 (in Japanese with English abstract).
Jones, G. J., P. D. Nichols, R. B. Johns, and J. D. Smith. 1987. The effect of mercury and
cadmium on the fatty acid and sterol composition of the marine diatom Asterionella glacialis.
Phytochem. 26:1343-1348.
Kaneda, T. 1967. Fatty acids in the Genus Bacillus. I. Iso- and anteiso-fatty acids as
characteristic constituents of lipids in 10 species. J. Bacteriol. 93:894-903.
Kvenvolden, K. A., M. Golan-Bac, and J. B. Rapp. 1987. Hydrocarbon geochemistry of

sediments offshore from Antarctica: I. Wilkes Land Continental Margin. In S. L. Eittreim and
M. A. Hampton, eds., The Antarctic Continental Margin: Geology and Geophysics of
Offshore Wilkes Land. Circum-Pacific Council for Energy and Natural Resources, Earth
Science Series, 5A:205-213 (cited in Kvenvolden et al. 1987).
Kvenvolden, K. A., J. B. Rapp, M. Golan-Bac, and F. D. Hostettler. 1987. Multiple sources of

alkanes in Quaternary oceanic sediment of Antarctica. Org. Geochem. 11:291-302.
Lisitsyn, A. P., ed. 1966. In Recent Sedimentation in the Bering Sea. Moscow: Izd. Nauka (in
Russian, cited in Tanoue and Handa 1980).
Lisitsyn, A. P. 1972. Sedimentation in the world ocean. Soc. Econ. Palaeontol. Mineral.,
spec. publ. (17), (cited in Tanoue and Handa 1980).
Mackenzie, A. S., S. C. Brassell, G. Eglinton, and J. R. Maxwell. 1982. Chemical fossils: the
geological fate of steroids. Science 217:491-504.
Mackie, P. R., H. M. Platt, and R. Hardy. 1978. Hydrocarbons in the marine environment. II.
Distribution of n-alkanes in the fauna and environment of the sub-Antarctic island of South
Georgia. Est. Coastal Mar. Sci. 6:301-313.

Masuzawa, T. and Y. Kitano. 1977. Cyclic appearance of reduced and oxidized sediments in
a 10-m core from the Japan Sea. J. Earth Sci. Nagoya Univ. 25:1-10.
Matsueda, H. and T. Koyama. 1977. Early diagenesis of fatty acids in lacustrine sediments-I.
Identification and distribution of fatty acids in recent sediment from a freshwater lake.
Matsueda, H. and N. Handa. 1986. Source of organic matter in the sinking particles collected
from the Pacific Sector of the Antarctic Ocean by sediment trap experiment. Mem. Natl. Inst.
Polar Res., spec. issue 40:364-379.
Matsumoto, E. 1987. Pb-210 geochronology of sediments. Studies of the San'in Region.

Natural Environment, no. 3., pp. 187-194.
Matsumoto, G. 1982. Comparative study on organic constituents in polluted and unpolluted

inland aquatic environments-IV. Indicators of hydrocarbon pollution for waters. Water Res.
16:1521-1527.
Matsumoto, G. I. 1989. Biogeochemical study of organic substances in Antarctic lakes.

Hydrobiologia 172:265-289.
Matsumoto, G. and T. Hanya. 1980. Presence of squalane in urban aquatic environments. J.

Chromatogr. 194:199-204.
Matsumoto, G. I. and K. Watanuki. 1990. Geochemical features of hydrocarbons and fatty

acids in sediments of the inland hydrothermal environments, Japan. Org. Geochem. 15:199-
208.
Matsumoto, G., T. Torii, and T. Hanya. 1979. Distribution of organic constituents in lake
waters and sediments of the McMurdo Sound region in the Antarctic. Mem. Natl. Inst. Polar
Res., spec. issue 13:103-120.
Matsumoto, G., T. Torii, and T. Hanya. 1982. High abundance of algal 24-ethylcholesterol in
Antarctic lake sediment. Nature 299:52-54.
Matsumoto, G., T. Torii, and T. Hanya. 1983. Stenols and phytol in lake sediments from
Syowa and Vestfold oases in the Antarctic. Geochem. J. 17:1-8.
Matsumoto, G., T. Torii, and T. Hanya. 1984. Vertical distribution of organic constituents in
an Antarctic lake: Lake Vanda. Hydrobiologia 111:119-126.
Matsumoto, G., F. Fukui, T. Torii, and T. Hanya. 1981. The features of organic constituents
in lake sediments from Syowa Oasis in East Antarctica. Verh. Internat. Verein. Limnol.
21:703-707.
Matsumoto, G. I., K. Watanuki, and T. Torii. 1987. Further study on the vertical distribution of

organic constituents in an Antarctic lake: Lake Vanda. Proc. NIPR Symp. Polar Biol. 1:219-
232.
Matsumoto, G. I., K. Watanuki, and T. Torii. 1989. Vertical distribution of organic constituents
in an Antarctic lake: Lake Fryxell. Hydrobiologia 172:291-303.
Matsumoto, G. I., T. Machihara, N. Suzuki, M. Funaki, and K. Watanuki. 1987. Steranes and
triterpanes in the Beacon Supergroup samples from southern Victoria Land in Antarctica.
McIver, R. D. 1975. Hydrocarbon gases in canned core samples from Leg 28 Sites 271, 272,
and 273, Ross Sea. In D. E. Hayes et al., eds., Initial Reports. DSDP 28:815-817.
Moriwaki, K. 1975. Submarine topography near Syowa Station. Antarctica. Antarct. Rec.
54:101-115 (in Japanese with English abstract).
Moriwaki, K. 1979. Submarine topography of the central part of Lützow-Holm Bay and
around Ongul Islands, Antarctica. Mem. Natl Inst. Polar Res., spec. issue 14:194-209.
Moriwaki, K. and Y. Yoshida. 1983. Submarine topography of Lützow-Holm Bay, Antarctica.

Mem. Natl. Inst. Polar Res., spec. issue 28:247-258.
Nachman, R. J. 1985. Unusual predominance of even-carbon hydrocarbons in an Antarctic

food chain. Lipids 20:629-633.
Nichols, P. D., J. K. Volkman, G. M. Hallegraeff, and S. I. Blackburn. 1987. Sterols and fatty
acids of the red tide flagellates Heterosigma akashiwo and Chattonella antiqua
(Raphidophyceae). Phytochemistry 26:2537-2541.
Nichols, P. D., J. K. Volkman, A. C. Palmisano, G. A. Smith, and D. C. White. 1988.

Occurrence of an isoprenoid C25 diunsaturated alkene and high neutral lipid content in
Antarctic sea-ice diatom communities. J. Phycol. 24:90-96.
Nichols, P. D., A. C. Palmisano, M. S. Rayner, G. A. Smith, and D. C. White. 1989. Changes

in the lipid composition of Antarctic sea-ice diatom communities during a spring bloom: an
indication of community physiological status. Antarct. Sci. 1:133-140.
Nishimura, M. and E. W. Baker. 1986. Possible origin of n-alkanes with a remarkable even-
to-odd predominance in recent marine sediments. Geochim. Cosmochim. Acta 50:299-305.
Ourisson, G., P. Albrecht, and M. Rohmer. 1979. The hopanoids. Palaeochemistry and
biochemistry of a group of natural products. Pure Appl. Chem. 51:709-729.
Philp, R. P. 1985. Biological markers in fossil fuel production. Mass Spectr. Rev. 4:1-54.
Philp, R. P., ed. 1986. Fossil Fuel Biomarkers. Applications and Spectra. Methods in

Geochemistry and Geophysics no. 23. Amsterdam: Elsevier.
Platt, H. M. 1979. Sedimentation and the distribution of organic matter in a sub-Antarctic

marine bay. Est. Coastal Mar. Sci. 9:51-63.
Platt, H. M. and P. R. Mackie. 1979. Analysis of aliphatic and aromatic hydrocarbons in

Antarctic marine sediment layers. Nature 280:576-578.
Prahl, F. G., J. T. Bennett, and R. Carpenter. 1976. The early diagenesis of aliphatic
hydrocarbons and organic matter in sedimentary particulates from Dabob Bay, Washington.
Rapp, J. B., K. A. Kvenvolden, and M. Golan-Bac. 1987. Hydrocarbon geochemistry of

sediments offshore from Antarctica: II. Western Ross Sea. In A. K. Cooper and F. J. Davey,
eds., The Antarctic Continental Margin: Geology and Geophysics of the Western Ross Sea.
Circum-Pacific Council for Energy and Natural Resources. Earth Science Series, 5B:217-224
(cited in Kvenvolden et al. 1987).
Requejo, A. G. and J. G. Quinn. 1983. Geochemistry of C25 and C30 biogenic alkenes in
sediments of the Narragansett Bay estuary. Geochim. Cosmochim. Acta 47:1075-1090.
Requejo, A. G., J. G. Quinn, J. N. Gearing, and P. J. Gearing. 1984. C25 and C30 biogenic
alkenes in a sediment core from the upper anoxic basin of the Pettaquamscutt River (Rhode
Island, U.S.A.). Org. Geochem. 7:1-10.
Robson, J. N. and S. J. Rowland. 1986. Identification of novel widely distributed sedimentary

acyclic sesterterpenoids. Nature 324:561-563.
Rohmer, M., P. Boivier-Nave, and G. Ourisson. 1984. Distribution of hopanoid triterpenes in

prokaryotes. J. Gen. Microbiol. 130:1137-1150.
Rowland, S. J., D. A. Yon, C. A. Lewis, and J. R. Maxwell. 1985. Occurence of 2, 6, 10-

trimethyl-7-(3-methyl-butyl)-dodecane and related hydrocarbons in the green alga
Enteromorpha prolifera and sediments. Org. Geochem. 8:207-213.
Rowland, S. J., S. J. Hird, J. N. Robson, and M. I. Venkatesan. 1990. Hydrogenation

behaviour of two highly branched C25 dienes from Antarctic marine sediments. Org.
Geochem. 15:215-218.
Sackett, W. M. 1986a. 13C signatures of organic carbon in southern high latitude deep sea
sediments; paleotemperature implications. Org. Geochem. 9:63-68.
Sackett, W. M. 1986b. Organic carbon in sediments underlying the Ross Ice Shelf. Org.
Geochem. 9:135-137.
Sackett, W. M., B. J. Eadie, and M. E. Exner. 1974. Stable isotope composition of organic
carbon in Recent Antarctic sediments. In B. Tissot and F. Bienner, eds., Advances in

Organic Geochemistry 1973, pp. 661-671. Paris: Technip.
Sackett, W. M., C. W. Poag, and B. J. Eadie. 1974. Kerogen recycling in the Ross Sea,
Antarctica. Science 185:1045-1047.
Sackett, W. M., W. R. Eckelmann, M. L. Bender, and A. W. H. Be. 1965. Temperature

dependence of carbon isotope composition in marine plankton and sediments. Science
148:235-237.
Seifert, W. K. and J. M. Moldowan. 1979. The effect of biodegradation on steranes and

terpanes in crude oils. Geochim. Cosmochim. Acta 43:111-126.
Seifert, W. K. and J. M. Moldowan. 1981. Paleoreconstruction by biological markers.

Suzuki, N. 1984. Estimation of maximum temperature of mudstone by two kinetic

parameters; epimerization of sterane and hopane. Geochim. Cosmochim. Acta 48:2273-
2282.
Tanoue, E. and N. Handa. 1980. Some characteristic features of the vertical profile of
organic matter in recent sediment from the Bering Sea. J. Oceanogr. Soc. Jpn. 36:1-14.
Venkatesan, M. I. 1988a. Diploptene in the Antarctic sediments. Geochim. Cosmochim. Acta

52:217-222.
Venkatesan, M. I. 1988b. Organic geochemistry of marine sediments in Antarctic region:

marine lipids in McMurdo Sound. Org. Geochem. 12:13-27.
Venkatesan, M. I. and I. R. Kaplan. 1987. The lipid geochemistry of Antarctic marine

sediments: Bransfield Strait. Mar. Chem. 21:347-375.
Volkman, J. K. 1986. A review of sterol markers for marine and terrigenous organic matter.
Org. Geochem. 9:83-99.
Volkman, J. K. and J. R. Maxwell. 1986. Acyclic isoprenoids as biological markers. In R. B.

Jones, ed., Biological Markers in the Sedimentary Record. Methods in Geochemistry and
Geophysics, no. 24, pp. 1-42. Amsterdam: Elsevier.
Volkman, J. K., D. I. Allen, P. L. Stevenson, and H. R. Burton. 1986. Bacterial and algal
hydrocarbons in sediments from a saline Antarctic lake, Ace Lake. Org. Geochem. 10:671-
681 ().
Volkman, J. K., H. R. Burton, D. A. Everitt, and D. I. Allen. 1988. Pigment and lipid
compositions of algal and bacterial communities in Ace Lake, Vestfold Hills, Antarctica.
Hydrobiologia 165:41-57.
Wardroper, A. M. K., P. W. Brooks, M. J. Humberston, and J. R. Maxwell. 1977. Analysis of

steranes and triterpanes in geolipid extracts by automatic classification of mass spectra.

Whiticar, M. J., E. Suess, and H. Wehner. 1985. Thermogenic hydrocarbons in surface

sediments of the Bransfield Strait, Antarctic Peninsula. Nature 314:87-90.
Wrenn, J. H. and S. W. Beckman. 1982. Maceral, total organic carbon, and palynological
analyses of Ross Ice Shelf Project Site J9 cores. Science 216:187-189.
Yoshida, Y., S. Murauchi, and K. Fujiwara. 1964. Submarine topography off the Kronprins
Olav Kyst. In R. D. Adie, ed., Antarctic Geology, pp. 710-714. Amsterdam: North Holland
Publishers.


9. Sedimentation and Preservation of Amino Compounds and

Carbohydrates in Marine Sediments
Rate constants of amino acid and carbohydrate decay are calculated for depth profiles from deep
drill holes in hemipelagic carbonate and opaline sediments. It appears that each of the individual
compound classes (amino acids, carbohydrates) is composed of different fractions (at least 3
fractions in the case of amino acids in the organic-rich upwelling sediments of the Peru area) with
different remineralization rates (with half lives ranging from 50 years to 11 million years). Such
different rates have been predicted by the G-model of organic carbon remineralization, but the rate
constants derived here vary considerably depending on lithology, organic-carbon content, and
depositional environment. Distinct rate constants of amino acid and carbohydrate decay cannot be
derived, but the observed decreases in reactivity of amino acids appear to follow pseudo first-order
kinetics predicted by the "power" model of Middelburg (1989) for bulk organic carbon. The decay of
amino acids and carbohydrates in sediments may be retarded by nonbiological factors such as
kerogen (or protokerogen) formation and adsorption onto mineral surfaces, which each results in
qualitative differences of the residual amino acids and sugars.
In an early publication on amino acid distribution and abundance in geological materials, Degens
and Bajor (1960) showed that labile constituents of organic matter (we refer to proteins and
carbohydrates hydrolyzable in HCl at 60°-110°C to amino acids and monomeric sugars,
respectively, as "labile," as opposed to hydrocarbons, which are geologically more stable) not only
were preserved in organic matrices but also constituted a significant portion of disseminated
sedimentary organic matter in sediments as old as Precambrian in age. In their conclusions, these
authors proposed that a variety of physicochemical conditions (i.e., sedimentary facies and
depositional environment, degree of thermal maturation, permeability, pressure, temperature, and
degree of weathering) affected the quantitative and qualitative preservation of eleven amino acids
measured in their samples. The most important conclusion drawn from their work was that
depositional conditions and diagenesis rather than age controlled preservation of amino acids in the
geologic record.
Realizing that studies of organic matter diagenesis in geologically young sediments hold a key to
better understanding the geological record, Rittenberg et al. (1963) investigated the "perishable"
qualities of sediments of middle Miocene to Holocene age recovered from the experimental Mohole
in the course of the first scientific ocean drilling. They found that the organic matter content of these
pelagic sediments is much lower than in sediments deposited under anaerobic conditions in basins
off southern California and that
Constituents of the organic matter such as hydrocarbons, porphyrin pigments, sugars and amino

acids have been selectively eliminated with respect to total organic matter.|.|.|. The lower ratios [of
these components relative to TOC in pelagic sediments] are characteristic of the surface layer as
well as deeper layers indicating that the selective removal of these components occurs in the water
or soon after deposition (Rittenberg et al. 1963:169).
A distinctive result of this first attempt to elucidate diagenesis of amino acids in deep marine
sediments is that hydrolyzable amino acid content decreases rapidly with depth, whereas
carbohydrates did not show any decrease. Having not detected any microbiological activity in the
section below 1.5 meter depth with methods available at that time, the authors propose that
It is likely that once buried, the amino acids undergo changes to more stable organic nitrogen
compounds with time, for instance by a tight fixation to clay minerals or to high molecular weight
organic condensation products such as kerogen.|.|.|. A steady state may be approached after a
certain period of time, as is suggested by the finding of amino acids in consolidated sediments of
equal or greater age (Rittenberg et al. 1963:170).
Although more than twenty-five years old, this statement includes all of today's explanations for the
preservation of supposedly labile organic matter in marine sediments of all ages. Although many
detailed aspects of organic matter diagenesis have since been clarified, the roles of amino acids
and sugars in the early formation of insoluble sedimentary organic matter and kerogen are still only
sketchily known.
An important point in the early study of amino acids in sediments was the realization that amino
compounds may contribute significantly to kerogen, which at that time was just recognized as the
precursor substance of petroleum hydrocarbons (Forsman and Hunt 1958; Degens et al. 1964).
Interest has continued in thermal decomposition of amino acids (e.g., Abelson and Hare 1971) and
in their contribution to marine humic and fulvic substances (Carter and Mitterer 1978) and to
material adsorbed to clastic and carbonate sedimentary particles (Suess 1970; Carter 1978). Since
the early studies of degradation and preservation of organic matter, much experimental and
geochemical work has been carried out to define pathways and rate constants of amino acid
degradation and preservation in seston intercepted by sediment traps (e.g., Lee and Cronin 1982,
1984; Wefer et al. 1982; Ittekkot, Degens, and Honjo 1984), in marine sediments (e.g., Degens and
Mopper 1976; Whelan 1977; Carter and Mitterer 1978; Henrichs and Farrington 1984), and in
riverine suspended matter (Ittekkot and Arain 1986; Ittekkot 1988). Little research has, however,
been conducted in continuous, thermally immature, well-dated marine sediments that would extend
the pioneering work of Rittenberg et al. (1963) into deeper sediments. In this contribution, we focus
on some recent work on sedimentary organic matter, specifically, the quantities and qualities of
bound carbohydrates and amino acids and their preservation in the geological record. We compare
published data from two marine sedimentary sections recovered by the Ocean Drilling Program
(Tyrrhenian Sea and the Peru margin; Seifert et al. 1990a, 1990b) with time scales of amino acid
and sugar degradation in much younger sediments from coastal environments (Henrichs,
Farrington, and Lee 1984; Liebezeit 1987; Burdige and Martens 1988; Hamilton and Hedges 1988).
We discuss our results in the context of the kinetic models of organic matter degradation
established by Berner (1981) and Westrich and Berner (1984) for geologically young (102 years)
sediments. We focus on three aspects: (1) do these kinetic models adequately describe the
diagenetic behavior of amino acids and carbohydrates in organic-carbon rich and lean sediments?
(2) Do we recognize discrete degradation rates for sugars and amino acids? (3) do organic-
inorganic interaction processes determine the preservation of labile organic matter in sediments as
old as early Pliocene (6 my)?

Input Considerations
Riverine Sources
Ittekkot and Arain (1986) used the concentrations of amino acids, amino sugars, and carbohydrates
(labile particulate organic carbon, LPOC) in suspended matter from the river Indus as criteria for
differentiating between a geochemically labile and a refractory fraction in riverine particulate organic
carbon (POC). In other studies, Ittekkot (1988) used these compounds to differentiate between
annual discharge of refractory organic matter from land (which he estimates to be 150 x 106 tons C
yr-1) and a labile fraction potentially to be degraded in estuaries (81 x 106 tons C yr-1). According to
Ittekkot (1988), the output of refractory organic matter is of a magnitude similar to estimates for the
total organic matter buried annually in the world ocean (Berner 1982). Ittekkot (1988) argues that
most of the labile organic matter from riverine discharge (i.e., amino acids, amino sugars, and
carbohydrates) is either trapped or remineralized in estuaries and riverine cones. From this
argument follows that the majority of carbohydrates, amino acids, and amino sugars deposited in
marine sediments must come from marine primary production.
Marine Sources
Studies of the average composition of plankton show that amino acids contribute between 14 and
35% of total POC (Degens and Mopper 1976). Cowie and Hedges (1984) give a value of 30% for
the contribution of carbohydrates to POC in plankton cultures, and labile POC in plankton samples
from surface waters ranges from 3% to 12% (Wefer et al. 1982; Ittekkot et al. 1984). For this study,
we assume that planktonic POC is composed of 25% carbohydrates and 62% proteins (average of
phytoplankton given in Romankevich 1990). At current estimates of oceanic primary production (60
x 109 tons C yr-1; Romankevich 1990), on the order of 15 x 109t carbohydrates and 37 x 109t
proteins are produced annually by marine plankton.
The bulk of this labile material is remineralized in the euphotic zone (Suess 1980; Lee and Cronin
1982, 1984; Ittekkot et al. 1984). This is illustrated by the results of Wefer et al. (1982), who found
that carbohydrates contribute 3% and amino acids 48% to POC in surface waters of the Drake
Passage, while these compound classes contribute 11% and 23%, respectively, to POC at 2540 m
water depth. This observation suggests that amino acid component of (rapidly decreasing) POC is
remineralized faster than either bulk POC or carbohydrates. The latter appear to be more resistant
than are either bulk POC and amino acids and are thus relatively enriched in seston in this open
marine setting.
Surface sediments show a wide range of total organic carbon (TOC) concentrations, which is an
effect of variables such as productivity, water depth, sedimentation rates, and bottom-water
oxygenation (Müller and Suess 1979; Demaison and Moore 1980). Combined amino acid- and
carbohydrate-carbon range from 3 to 10% of TOC in marine surface sediments such as those from
the Argentine Basin and the Arabian Sea (Degens and Mopper 1976). However, anaerobic surface
sediments from the Black Sea (Mopper 1977) and the Peru Shelf (Henrichs et al. 1984) have higher
concentrations, as do sediments underlying the productive waters of the Panama Basin (11% and
8.5% of TOC for carbohydrates and amino acids, respectively; Ittekkot et al. 1984). Pelagic
sediments extremely low in organic carbon are significantly lower in amino acid contribution (e.g.,
Whelan 1977).

To arrive at a crude mass balance, we estimate the amino acid and carbohydrate component of
organic carbon in the sediment layer at the sediment/water interface to be 15% (3% and 12% of
TOC from amino acids and carbohydrates, respectively; Romankevich 1984). Further, we assume
that sedimentary TOC constitutes 0.5% to 1% of primary production (e.g., Suess 1980). We thus
arrive at a figure of 45 to 90 x 106 tons yr-1 of carbon in amino compounds and sugars from marine
planktonic sources being deposited at the sediment/water interface. According to the models of
Jørgenson (1978) and Berner (1981), this labile component of marine primary production and the
81 x 106t of labile TOC from continental runoff (Ittekkot 1988) may be the prime substrate for
bacterial oxidation of TOC at the expense of interstitial oxygen and sulfate.
Case Studies of Preservation and Rates of Degradation
The data sets used here to derive rate constants of remineralization of amino acids and
carbohydrates in deep marine sediments are from Seifert et al. (1990a, 1990b) and those for
shallow marine shelf sediments are from Henrichs (1980). Details on methodology and further
interpretation are given by the authors. Because we were interested in bulk labile TOC, we do not
discuss any details of compositional differences in the sediments or any downhole changes. The
sedimentary systems may have (and very likely have) varied with regard to productivity and
sedimentation rates over the intervals studied, but the observed long-term consistent trends (see
below) are assumed to be results of diagenesis rather than results of of input variations. Support for
the validity of this assumption is provided by the observation that for sediments from both the Peru
and Oman upwelling areas, pyrolysis-gas chromatography/mass spectroscopy patterns for the high-
molecular-weight organic matter remain very constant in a particular location; variation occurs
predominantly in the amount of pyrolyzable material rather than in the concentrations of individual
components relative to TOC (Emeis and Whelan, in preparation). A change in composition of
pyrolysates indicating a change in the type of source of the sedimentary organic matter was not
observed.
Tyrrhenian Sea
ODP Site 653 (40°15.86'N/ 11°26.99'E) in the Tyrrhenian Sea/Western Mediterranean is located in
2817 m water depth. Sediments from this site are composed of gray and brown nannofossil oozes,
foraminiforal nannofossil oozes, and mud with an average carbonate concentration of 60% and
TOC < 0.3%. The analytical error of TOC measurements exceeded 100% and we thus did not
normalize total hydrolyzable amino acids (THAA) and total hydrolyzable carbohydrates (THCHO) to
TOC. The age model for Site 653 is based on paleontological and magnetostratigraphic work
(Kastens et al. 1987). These sediments may serve as an example of hemipelagic oozes deposited
under well-oxygenated bottom-water conditions, which become anoxic a few centimeters below the
sediment-water interface.
The vertical distributions of THAA (with a range of 0.02 to 1.1 moles/g sediment; up to 2.5% of
organic carbon) and THCHO (range from 0.21 to 1.82 moles/g sediment and about 8% of
sedimentary organic carbon) for Hole 653A show a rapid decrease with depth with only minor
variability in concentrations in the oldest (Pliocene) samples (figure 9.1). The relative downhole
distributions of neutral, acidic, basic, and nonprotein amino acids (figure 9.2) with increasing depth
are increasingly dominated by neutral and acidic amino acids, while nonprotein moieties composed

of ornithine, -alanine, and -aminobutyric acid decrease, and basic amino acids remain at constant
proportions. In the calcareous sediments at Site 653 the variability in amino acid fractions is likely to
be governed by preservation of acidic amino acids through adsorption onto carbonate surfaces
(Suess 1970; Carter and Mitterer 1978; see below). Continued deamination and decarboxylation of
nonprotein amino acids ( -alanine, -amino butyric acid, ornithine), which are produced by microbial
degradation of protein precursors (e.g., aspartic acid, glutamic acid, and argi-nine, respectively),
may result in a relative increase of neutral and acidic amino acids bound to and protected by the
lithogenic matrix. The persistence of a low background value of 0.1-0.2 mole THAA/g sediment
indicates that a residual, nonhydrolyzable fraction accounts for approximately 10% of surface
sediment THAA in these hemipelagic sediments.
In contrast to amino acids, the amount of total hydrolyzable amino sugars (THAS) remains relatively
constant throughout the interval studied but exhibits a wide range of concentrations (figure 9.1).
Hexoseamines (glucoseamine and galactoseamine), which are present in concentrations of tens of
nanomoles per gram sediment, do not exhibit any consistent trends with depth.
Peru Margin, ODP Site 681
Site 681 (10°58.7'S and 77°56.9'W) is located in the southernmost sector of the Salaverry Basin off
Peru, close to the pronounced coastal upwelling around 11°S. Water depth here is only 150 m, and
the site coincides with the upper boundary of the oxygen-minimum zone that impinges on the
Peruvian shelf and upper slope. Paleomagnetic and biostratigraphic correlation show that the 187 m
of sediment recovered are Pleistocene in age, and the sedimentation rate is close to 80 m/my
(Suess et al. 1988).
Total hydrolyzable amino acids were measured in concentrations between 0.37 and 111 mole/g
dry weight, with a mean value of 14.3 mole/g. Values decrease generally with depth but show also
distinct variability that correlates with the downhole distribution of total organic carbon. The
concentrations of THAA and THCHO normalized to TOC decrease rapidly in the upper 30 mbsf
(meters below surface) (figure 9.3). The average composition of the THAA spectra, as shown in
figure 9.4, is dominated by neutral monomers glycine, alanine, and leucine (each with > 10 mole%).
A distinct difference between the THAA composition of the upper and the deeper part is obvious
(figure 9.4). The acidic amino acids aspartic acid and glutamic acid are more abundant in the
shallow samples and uniformly low in concentration below 10 mbsf. Neutral valine, iso-leucine,
leucine, -alanine, and -amino butyric acid show the opposite trend. Analyses of total hydrolyzable
sugars in sediment from Site 681 found a mean concentration of 6.8 mole/g dry weight and range
from 0.8 to 35 mole/g. The average THCHO composition is dominated by glucose, mannose, and
galactose, each of which account for approximately 20 mole%. Xylose, rhamnose, arabinose,
fucose, and fructose concentrations are between 11 and 5.8 mole%. The downhole distribution
pattern of the sugar monomers shows an increase of glucose with depth offset by a decrease of
fucose. The average concentration of total hydrolyzable amino sugars is 4.1 mole/g, ranging
between 0.3 and 17 mole/g with no indication of a downhole decay comparable to that of either
THAA or THCHO. THAS comprise 0.15% to 1.7% (average 0.6%) and 0.7% to 6.1% (average
2.1%) of TOC and TN, respectively (Seifert et al.1990 b).
Rate Constants of Amino Acid and Carbohydrate Degradation

The decrease of THAA and THCHO with increasing depth in the sedimentary column of Site 681
appear to proceed at specific rates that differ from that of bulk TOC, because we normalized the
results to sedimentary organic carbon. Similarly, the total amount of amino compounds and
carbohydrates in sediments from Site 653 appear to decrease fairly regularly with increasing
sediment depth. In the following paragraph, we attempt to model the decrease in "labile"
sedimentary organic matter with age in the sedimentary column in order to investigate the rates of
decay and the relative stability of amino acids and carbohydrates.
According to the multiple-G models of organic matter degradation in marine sediments (e.g.,
Westrich and Berner 1984), the reactivity of organic matter exerts a strong influence on the rates of
organic matter degradation under conditions of sulfate reduction. It is generally believed that fresh
organic matter of marine origin supports bacterial metabolism to a greater extent than organic
matter of terrestrial origin does and that natural microbial communities selectively use the available
substrates according to their energy yield (Alexander 1983; Emerson and Hedges 1988). During
microbially mediated reactions the amount of available substrate decreases with age and limits the
rate of sulfate reduction (Berner 1981). In their attempts to model the distribution of sulfate in
marine sediments and to simulate decomposition processes in vitro, Westrich and Berner (1984)
came to the conclusion that the decrease of organic carbon in coastal marine sediments is not
simply a function of the exponential decay of an initial amount of organic matter through time at one
fixed rate (simple-G model) but rather that the observations are best described by assuming several
pools of organic matter of different lability and with different decay constants (multi-G model). An
important implication of all G-type models is that the reactivity of organic matter decreases with age.
For amino acids, which here represent a relatively labile organic fraction, the multi-G model predicts
that the concentration with time may be represented by an equation (9.1) in the form
where AA1 and AA2 are initial concentrations of amino acid fractions with different decay constants
k1 and k2, AA is a refractory fraction that is not remineralized, and AA(t) is the sum of all these
amino acid fractions at time = t.
Burdige and Martens (1988) employed this multi-G model to explain the observed distribution of
amino acids in anoxic sediments from Cape Lookout Bight. They assumed that two fractions of
hydrolyzable amino acids, a labile fraction dominated by acidic amino acids and a refractory fraction
dominated by neutral amino acids, were present in their sediments. Decay constants for more labile
amino acid moieties derived from their work ranged from 1.1 to 2 yr-1, while a more refractory
component of total amino acids appeared to be constant with depth. The mean residence time of
amino acids in sediments from Cape Lookout Bight was approximately nine months (Burdige and
Martens 1988).
Case Studies
Figure 9.5 shows plots of THAA and THCHO concentrations versus age for those sites and cases
presented in our study. We reexamined Henrichs' (1980) results of total hydrolyzable amino acids
(normalized to TOC) in sediments from Station 5A and plotted the concentration of THAA/g TOC
versus age (figure 9.5a; derived from a mean sedimentation rate of S0.9 cm yr-1; Henrichs and

Farrington 1984). It appears that a simple exponential decay function (eq 9.2) implying two amino
acid fractions in the form
can adequately describe the observed decay of total hydrolyzable amino acids with age in surface
sediments from Station 5A of Henrichs (1980). By assuming a concentration of 0.05 mole/g TOC
for AA derived from results of Site 681, a rate constant k is calculated as 0.02 yr-1. The residence
time t = 1/k of amino acids in surface sediments calculated from the data of Henrichs (1980) is 50
years.
ODP Site 681 is situated in the same general geographic area as Henrichs' (1980) Station 5A on
the Peru margin, but the sections recovered here range back to the early Pleistocene in age (dated
by stable-isotope stratigraphy by Wefer, Heinze, and Suess 1990). The sediments from Site 681
were recovered from a somewhat different depositional environment than those of Henrichs (i.e.,
higher carbonate at Site 681 as compared with almost all siliceous input at 15°S in Henrichs' work).
However, the predominantly marine nature of the organic matter at both locations encouraged us to
calculate the rates of decay of the "labile" organic matter over a longer geological time span at Site
681 for comparison with the Henrichs' surface data.
The data on amino acids and carbohydrates were normalized to TOC, and their concentrations are
plotted versus age in figure 9.5b. In examining the data on amino acids, the THAA can be
represented by a residual fraction of 0.05 mole/g TOC and an initial labile fraction of 1.4 mole/g
TOC in surface sediments that decays at a rate of and 1.6 x 10-5 yr-1. Similarly, carbohydrates can
be represented by two fractions, the more labile of which decays at a rate of 4.4 x 10-6 yr-1. Note
that amino sugars show no consistent decay with depth and that both sugars and amino acids have
not been exhausted after 2 million years.
More than 90% of THAA in the sediment of the Peruvian shelf belong to the rapidly decaying
fraction. Qualitatively, the more labile fraction contains significantly more acidic amino acids when
compared with the recalcitrant fraction in deeper sediments, where acidic AA are relatively depleted
(figure 9.4). Upon closer inspection, the fraction encountered in deep sediments at Site 681, which
is dominated by neutral amino acids, still continues to decrease steadily with age (see figure 9.5b);
it may thus not be equivalent to the stable pool of refractory amino acids seen in other sediments.
Comparison of the relative abundances of different amino acid fractions in deep fractions of both
Site 653 and 681 suggests that neutral amino acids are preferentially enriched over other fractions
(e.g., acidic, basic, and nonprotein). However, the distribution at Site 681 (opaline sediments, figure
9.4) is not the same as the refractory pool of amino acids encountered at depth at Site 653
(calcareous sediments, figure 9.2).
The decay patterns with two distinctly different rates observed for total hydrolyzable amino acids
and sugars at Site 681 (figure 9.5b) indicate that there are at least two forms of these compounds
that are decaying at significantly faster rates (up to 30 times faster) in the surface (to an age of
about 0.3 Ma) in comparison with older sediments. The differences may be caused by any of the
following: (1) differences in primary peptide structure; (2) differences in matrix binding, as discussed
above; (3) microbiological degradation in the surface sediments versus diagenetic or chemical
degradation in deeper sediments; (4) change in macrofauna, microbiological activity, or the

community responsible for degradation in deeper and older sediments; (5) changes in the
depositional environment and thus in primary composition. None of these mechanisms can be ruled
out with the data currently in hand. However, the apparent first-order decay for both parts of the
THAA and THCHO decay curves and the abrupt change in rate of decomposition in sediments older
than 0.3 Ma may argue in favor of successive microbiological degradation. However, even though
no lithologic or faunal and floral break is apparent, we cannot rule out that sedimentation of organic
matter occurred at different rates or in different states of preservation. If, on the other hand, all of
the degradation were considered to be microbiologically mediated, the "deeper" and slower decay
rate might be caused either by a diminishing activity of the microorganisms at depth or by different
metabolic rates of a new type of organism at depths corresponding to sediments older than 0.3 to
0.6 my.
Cragg et al (1990) have shown that anaerobic microbiological activity can be induced in cores
recovered from the Peru shelf to at least 150 mbsf, corresponding to an age of at least 1 my. It has
also been shown that anaerobic (sulfate reduction and methane production) microbiological activity
can be induced in other older sediments (160,000 yrs) from other deep sea locations but at rates
that are several orders of magnitude lower than those observed in surface sediments (Whelan et al.
1986; Tarafa et al. 1987). If this activity also occurs at a slow rate under in situ conditions, then its
effect on biodegradation of sediment organic matter may correspond to the low rates we are
observing here. In this case, the organic matter fraction in the shallow interval, which contains the
larger proportion of acidic amino acids, is degraded faster. If these compounds are a critical factor
in the degradation process (whether biological or chemical), then their depletion below some
threshold value in deeper sediments would cause an overall deceleration of their remineralization.
The observation that the THAS fraction does not undergo a decay pattern similar to that of THAA
and THCHO shows that there is some selectivity in the organic fraction undergoing decay. We
believe that either a difference in mineral matrix binding or selective remineralization by microbial
activity of the THAA and THCHO versus the THAS fraction could be causing these differences.
The data from ODP Site 653 are best described via a residual THAA fraction with no apparent
decay rate that comprises approximately 0.1 moles/g sediment (assumed to represent fraction AA
) and a fraction that decays at a rate of 9.5 x 10-7 yr-1 (figure 9.5c). This latter fraction initially
comprises approximately 0.9 moles/g, so that residual THAA account for approximately 10% of
THAA in surface sediments. THCHO data vary considerably but seem to indicate that sugar
degradation proceeds even after millions of years of diagenesis (figure 9.5c). Seifert et al. (1990a)
attribute the continuing decay of THCHO concentrations to diagenetic remineralization of biogenic
calcareous tests and the concurrent chemical degradation of carbohydrates.
A possible reason for the observed differences in decay rates between the opal-dominated anoxic
sediments from the Peru margin and the carbonate-dominated sediments from the Tyrrhenian Sea
may be that the organic matter deposited in the carbonate sediments underlying the open ocean is
more recalcitrant to begin with than sediments underlying highly productive waters and shallow
water columns. Therefore, the observed variations may primarily reflect differences in the material
deposited at the surface, rather than any diagenetic alteration after burial. The same observation
may hold true for the THAS fraction at Site 681, which showed considerable variability but lacked a
systematic decay pattern with depth.
A very important development of G-type models of organic matter decay was published by

Middelburg (1989) during revision of this manuscript. In an analysis of numerous experimental and
empirical decay constants, he found that the reactivity of bulk organic matter could be described as
decreasing continuously with age rather than with discrete successive rates. Middelburg (1989)
established that a single straight correlation between the rate constant (k) versus time (t) of the form
describes the decrease in reactivity of bulk organic carbon in sediments over more than eight orders
of magnitude in time and concentration. When plotting the limited number of rate constants derived
in this study according to the procedure of Middelburg (1989), we see that the amino acid rate
constants in sediments from the Tyrrhenian Sea and those from the two sites on the Peru margin
follow Middelburg's correlation for the decay of bulk organic matter quite well (figure 9.6). Evaluation
of future data sets will be needed to demonstrate whether or not this general relationship holds for
other compound classes, as well.
General Discussion
From the examples above and many other literature reports it becomes clear that we are faced with
a wide continuum of organic matter decay constants derived from a variety of depositional settings
and experiments (table 9.1). Such extremely wide ranges are not easily reconciled with well-defined
decay rates, even within a single compound class. Variability is not unexpected, when one
considers the wide variety of decay processes possible and other complicating factors, such as
specific mechanisms of biological decay (e.g., macrofaunal vs. aerobic microbial vs. anaerobic
microbial decay), packaging (large particles vs. small particles), and phase differences (i.e., water
column vs. surface sediment vs. deep sediment decay processes). In fact, when one considers the
large potential for variability, it is surprising that all organic matter can apparently be treated
together mathematically via the recently introduced "power" model for bulk organic matter decay
(Middelburg 1989). To account for the decreasing reactivity of amino acids and carbohydrates with
age, various factors need to be examined, as outlined below.
As Rittenberg et al. outlined in 1963 and Degens and Mopper reiterated in 1976, there are
essentially four potential mechanisms that bear on the preservation and accumulation of sugars and
amino acids in organic matter of marine sediments over geological time spans. These mechanisms
are (1) organic-organic interactions, i.e., polymerization, humification, and melanoidin or kerogen
formation from biological precursor substances such as proteins, lipids, and polysaccharides; (2)
mineral-organic interactions, i.e., association of charged amino acids and sugars with siliciclastic
and carbonate particle surfaces and interlayers; (3) preservation in biological matrices as unaltered
biological markers; and (4) metal-organic interactions (Degens and Mopper 1976). The last factor
has long been a matter of controversy, and no conclusive evidence for the role of metal-organic
complexes in the preservation of amino acids and sugars has been produced in the last decade.
Because the subject of preservation of biological molecules in shell matrices of plants and animals
was the original focus of interest with regard to amino acid preservation in geological samples and
thus has been treated extensively in past publications (e.g., Hare 1969), we will in the following
sections briefly review aspects of organic-organic and organic-mineral interactions.
Polymerization Reactions

Natural high-molecular-weight organic substances of complex structure dissolved in seawater and

interstitial waters and sedimentary organic substances (i.e., humic acids, humin, kerogen) have long
been regarded as products of condensation reactions between biological precursor molecules (see
Durand 1980 and contributions therein). More recently, Derenne et al. (1988) and Tegelaar et al.
(1989) have postulated that these geomacromolecules receive a substantial contribution from
chemical and microbial alteration products of biological precursor substances of high molecular
weight and complex structure. On the assumption that both of these mechanisms make a
contribution, early stages of alteration of the resulting macromolecules are mainly mediated by
microorganisms, which sequentially utilize the most reactive substrates available at any particular
time. As a result, the reactivity of sedimentary organic matter decreases with time (e.g., Berner
1981; Westrich and Berner 1984; Emerson and Hedges 1988; Middelburg 1989). Reactions of
various biogenic molecules, such as carbohydrates, quinones, and polyphenols, with amino acids
were recognized early as possible pathways leading to the formation of high-molecular-weight,
nitrogen-containing condensation products (Maillard 1913) that resemble natural humus in
composition and character. Several authors have performed experiments with artificially produced
or naturally occurring macromolecular condensation products of labile monomeric biomolecules to
determine their abilities to chemically bind or adsorb functionalized lipids (e.g., Larter and Douglas
1980), amino acids (e.g., Abelson and Hare 1971), and lignins and phenols (e.g., Huc 1980).
Melanoidin formation from amino acid and carbohydrate polymerization (Maillard 1913), even
though never directly observed to occur naturally in waters or sediments, may indeed be an
important process of humus and humic acid formation. Hedges (1978) measured the rates of
polymer formation in mixtures of glucose and a basic, neutral, or acidic amino acid at variable pH.
His results suggested that the basic amino acid lysine reacts at a higher rate than either neutral
(valine) or acidic (glutamic acid) amino acids at a pH of 8. Basic amino acids are also irreversibly
removed from solution by association with melanoidins and kerogen, and melanoidin formation
occurs much more rapidly with basic amino acids and glucose than with neutral or acidic amino
acids (Hedges 1978).
The results of experiments carried out by Abelson and Hare (1971) and by Larter and Douglas
(1980) on the reaction of kerogen with functional molecules shed some light on the possible
incorporation of amino acids and carbohydrates into kerogen. Based on the differences in carboxyl
and amino groups in the original amino acid, the artificial condensation product or natural kerogen
retains the basic, neutral, or acidic character imparted by the chemically, presumably ester-bound
amino acid. This in turn controls the adsorptive association to carbonate (positively charged) and
silicate (negatively charged) particle surfaces and the facility to incorporate other monomeric
molecules (Abelson and Hare 1971; Hedges 1978; Larter and Douglas 1980).
However, melanoidin formation requires the presence of monomeric, free carbohydrates and amino
acids in high concentrations. Concentrations in the natural environment outside the cells of
decaying microorganisms and similar microenvironments are probably too low for this reaction to
occur in detectable quantity (Henrichs 1980; Mopper et al. 1980). Thus melanoidins may be
quantitatively minor in the formation of the continuum of particles, amorphous matter, colloids, and
chemically bound and physically adsorbed molecules that comprise kerogen. Melanoidin formation
does, however, provide an important potential mechanism to link two of the most abundant
substance classes in fresh organic matter from marine organisms and to form polymers or
aggregates not only with other organic substances, e.g., lipids and phenols, but also with functional
groups and charged surfaces of organic or inorganic particles.

Association with Clastic and Carbonate Particles
Along with reactions and condensations of organic molecules, the interactions of charged organic
molecules with charged particle surfaces can help preserve organic matter in sediments. Carbonate
sediments in particular contain 15 to 20% amino acids in fulvic acid fractions, and 27 to 36% amino
acids in humic substances (Carter and Mitterer 1978). Organic matter in these biogenic deposits
may be preserved either as an adsorbed coating of amino acids or as internal organic matrix that
serves as an aspartic-acid-rich template of calcium carbonate mineralization (Degens 1979). The
fact that acidic amino acids, and aspartic acid in particular, comprise 30 to 40% of total hydrolyzable
amino acids in carbonate sediments suggests that the (positively charged) surface of carbonate
particles preferentially adsorbs (negatively charged) deprotonated carboxyl groups and may even
bind them in ester bonds (Carter and Mitterer 1978).
The preferential adsorption of acidic amino acids observed in carbonate sediments is mirrored in
the preferential adsorption of positively charged basic amino acids onto the negatively charged
surfaces and interlayers of clay minerals (Hedges and Hare 1987). The same principle applies to
the negatively charged surface of biogenic opal. Hedges and Hare (1987) found in laboratory
experiments that basic amino acids lysine, histidine, and arginine adsorb significantly onto
montmorillonite (80% of dissolved basic amino acids). In contrast, the two-layer clay mineral
kaolinite adsorbed significant amounts of anionic aspartic and glutamic acids in addition to basic
molecules. The difference in adsorption characteristics between the two clay minerals is attributed
to significant positive charge at crystal edges of kaolinite, onto which anionic amino acids adsorb.
Because the monomeric amino acids, carbohydrates, and amino sugars analyzed in this work were
all hydrolyzed from the sediment, their primary source before hydrolysis was probably dominantly
biopolymer, geopolymer, and matrix-bound fractions. A significant monomer source seems unlikely
because these compounds are generally biodegraded much more rapidly than the higher molecular
weight fractions and, therefore, generally represent only a minor fraction of sediment organic
matter. Thermal processes generally do not become important until geothermal temperatures of at
least 50°C are reached (Hunt 1979; this volume)-temperatures considerably above any observed
for sediment sections analyzed in this work.
We investigated aspects of three questions: whether kinetic approaches adequately describe the
decay of biological molecules in sedimentary organic matter, whether any "universal" decay
constants can be derived for a relatively labile and ubiquitous fraction of sedimentary organic
matter, and whether mineral-organic interactions influence the decay rates of amino compounds
and sugars in both carbonate-rich and opal-rich sediments.
The first question is the most easily approached with our limited scope and our limited set of
information. The decay of sedimentary amino acids and carbohydrates in deep marine sediments
appears to follow first-order kinetics in decay processes for the lengths of time studied.
Concentrations approach a limit in time spans that are on the order of millions of years.
Theoretically, the early results of the pioneers in organic geochemistry, who must have been thrilled
at discovering traces of amino acids in Precambrian rocks, are thus feasible, because a residue of
amino acids (and sugars) appears to be stable enough to survive billions of years.
The second question is considerably more complex. The shape of the decay curves in surficial
sediments (Station 5A of Henrichs [1980]) is consistent with a single decaying fraction and,

possibly, a refractory or less labile amino acid fraction in the three deepest samples. All other
profiles investigated here in deeper sediments suggest at least two fractions with considerable
differences in decay constants. To compound the problem, a variety of decay constants emerges
from the literature and from the two depositional environments studied in this work (table 9.1). The
potential complexity of decay processes of various organic fractions provides no empirical basis for
the assumption of "universal" rate decay constants. However, during the revision of this manuscript,
Middelburg (1989) published his "power" model of organic matter decay, which appears to simplify
previous G-type models significantly. Testing our data sets with regard to the applicability of the
"power" model showed that the decay of amino acids from two environments and from sediments
ranging from years to millions of years in age can be described surprisingly well by the continuous
decrease in reactivity of bulk organic carbon predicted by Middelburg (1989).
Third, our comparison of amino acid composition in opaline and carbonate sediments suggests that
inorganic phases influence preferential preservation of amino acid classes in sediments, in spite of
the considerable uniformity of amino acid compositions in biological tissue. The preferential
adsorption (basic amino acids onto negatively charged clay mineral and opal surfaces and acidic
amino acids onto positively charged carbonate particles) may lead to different compositional end
members of amino acid degradation (the stable, refractory traces of amino acids encountered even
in geologically old rocks and sediments). Such potential for preferential preservation complicates
even further the search for universal constants that can be used in describing the decay of organic
matter in geological materials.
Acknowledgments
We wish to dedicate this contribution to the memory of the late Egon T. Degens, colleague and
friend of John Hunt. Many of the data used in this study (and much pioneering work on amino acids
and sugars in geological materials) were produced and published by Degens and his coworkers in
Hamburg, Germany. The authors of this contribution include a former student (K.-C.E.) and a
successor and admirer at the Woods Hole Oceanographic Institution (J. K. W.). Both Egon Degens
and John Hunt have been great scientists, teachers, and mentors to both of us! While taking full
responsibility for any controversial statements or "fuzzy thinking," we also gratefully acknowledge
very helpful reviews and comments by Susan Henrichs, Cindy Lee, Erwin Suess, Ludger Mintrop,
and John Farrington.
References
Abelson, P. H. and P. E. Hare. 1971. Reactions of amino acids with natural and artificial
humus and kerogens. Carnegie Inst. Yearbook 69:327-334.
Alexander, M. 1983. Biotechnology Report. Nonbiodegradable and other recalcitrant

molecules. Biotechnol. and Engineering 15:611-617.
Berner, R. A. 1981. A rate model for organic matter decomposition during bacterial sulfate
reduction in marine sediments. In: Biogéochimie de la Matière Organique a l'Interface Eau
Sediment. Marin. Colloq. Int. CNRS 293:35-44.
Berner, R. A. 1982. Burial of organic carbon and pyrite sulfur in the modern ocean: Its
geochemical and environmental significance. Am. J. Sci. 282:451-473.

Burdige, D. J. and C. S. Martens. 1988. Biogeochemical cycling in an organic-rich coastal

marine basin: 10. The role of amino acids in sedimentary carbon and nitrogen cycling.
Carter, P. W. 1978. Adsorbtion of amino acid-containing organic matter by calcite and

quartz. Geochim. Cosmochim. Acta 42:1239-1242.
Carter, P. W. and R. M. Mitterer. 1978. Amino acid composition of organic matter associated
with carbonate and non-carbonate sediments. Geochim. Cosmochim. Acta 42:1231-1238.
Cowie, G. L. and J. I. Hedges. 1984. Carbohydrate sources in a coastal marine environment.

Geochim. Cosmmochim. Acta 48:2075-2087.
Cragg, R., J. Parkes, R. Fry, J. Herbert, J. Wimpenny, and J. Getliff. 1990. Bacterial biomass
and activity profiles within deep sediment layers. In E. Suess, R. von Huene, et al., eds.,
Proc. ODP, Scientific Results, 112:607-619. College Station, Texas (Ocean Drilling
Program).
Degens, E. T. 1979. Why do organisms calcify? Chem. Geol. 25:257-269.
Degens, E. T. and M. Bajor. 1960. Die Verteilung von Aminosäuren in bituminösen

Sedimenten und ihre Bedeutung für die Kohlen- und Erdölgeologie. Glü kauf 24:1525-1534.
Degens, E. T. and K. Mopper. 1976. Factors controlling the distribution and early diagenesis
of organic material in marine sediments. In J. P. Riley and R. Chester, eds., Chem.
Oceanogr. 6:59-113. London: Academic Press.
Degens, E. T., J. M. Hunt, H. Reuter, and W. E. Reed. 1964. Data on the distribution of
amino acids and oxygen isotopes in petroleum brine waters of various geological ages.
Sedimentol. 3:199-225.
Demaison, G. J. and G. T. Moore. 1980. Anoxic environments and oil source bed genesis.
Derenne, S., C. Largeau, E. Casadevall, E. Tegelaar, and J. W. de Leeuw. 1988.

Relationship between algal coals and resistant cell wall biopolymers of extant algae as
revealed by Py-GC-MS. Fuel Processing Technol. 20:93-101.
Durand, B., ed. 1980. Kerogen. Paris: Editions Technip.
Emerson, S. and J. I. Hedges. 1988. Processes controlling the organic carbon content of
open ocean sediments. Paleoceanogr. 3/5:621-634.
Forsman, J. B. and J. M. Hunt. 1958. Insoluble organic matter (kerogen) in sedimentary

rocks of marine origin. In L. G. Weeks, ed., Habitat of Oil, pp. 747-778. Tulsa: AAPG.

Hamilton, S. E., and J. I. Hedges. 1988. The comparative geochemistries of lignins and
carbohydrates in an anoxic fjord. Geochim. Cosmochim. Acta 52:129-142.
Hare, P. E. 1969. Geochemistry of proteins, peptides, and amino acids. In G. Eglinton and
M. T. J. Murphy, eds., Organic Geochemistry: Methods and Results, pp. 438-463. New York:
Springer Verlag.
Hedges, J. I. 1978. The formation and clay mineral reactions of melanoidins. Geochim.
Hedges, J. I. and P. E. Hare. 1987. Amino acid adsorption by clay minerals in distilled water.
Henrichs, S. M. 1980. Biochemistry of dissolved free amino acids in marine sediments. Ph.
D. thesis, Woods Hole Oceanographic Institution.
Henrichs, S. M. and J. W. Farrington. 1984. Peru upwelling region sediments near 15oS. 1.
Remineralization and accumulation of organic matter. Limnol. Oceanogr. 29(1):1-19.
Henrichs, S. M., J. W. Farrington, and C. Lee. 1984. Peru upwelling region sediments near
15oS. 2. Dissolved free and total hydrolyzable amino acids. Limnol. Oceanogr. 29(1):20-34.
Huc, A. Y. 1980. Origin and formation of organic matter in recent sediments and its relation
to kerogen. In B. Durand, ed., Kerogen, pp. 445-474. Paris: Editions Technip.
Hunt, J. M. 1978. Petroleum Geochemistry and Geology. San Francisco: Freeman.
Ittekkot, V. 1988. Global trends in the nature of organic matter in river suspensions. Nature
332:436-438.
Ittekkot, V. and R. Arain. 1986. Nature of particulate organic matter in the river Indus,
Pakistan. Geochim. Cosmochim. Acta 50:1643-1653.
Ittekkot, V., E. T. Degens, and S. Honjo. 1984. Seasonality in the fluxes of sugars, amino
acids, and amino sugars to the deep ocean: Panama Basin. Deep Sea Res. 31(9):1071-
1083.
Jørgenson, B. B. 1978. A comparison of methods for the quantification of bacterial sulfate

reduction in coastal marine sediments-2. Calculations from mathematical models. J.
Microbiol. 1:29-51.
Kastens, K., J. Mascle, C. Auroux, et al. 1987. Proc. ODP, Initial Reports, 107. College
Station, Texas (Ocean Drilling Program).
Knauer, G. A., J. H. Martin, and K. U. Bruland. 1979. Fluxers of particulate carbon, nitrogen
and phosphorus in the upper water column of the northeast Pacific. Deep Sea Res. 26:92-

108.
Larter, S. R. and A. G. Douglas. 1980. Melanoidins-kerogen precursors and geochemical

lipid sinks: a study using pyrolysis gas chromatography (PGC). Geochim. Cosmochim. Acta
44:2087-2095.
Lee, C. and C. Cronin. 1982. The vertical flux of particulate organic nitrogen in the sea:
decomposition of amino acids in the Peru upwelling area and the equatorial Atlantic. J. Mar.
Res. 40:227-251.
Lee C. and C. Cronin. 1984. Particulate amino acids in the sea: Effects of primary
productivity and biological decomposition. J. Mar. Res. 42:1075-1097.
Liebezeit, G. 1987. Early diagenesis of carbohydrates in the marine environment. I.

Sediment trap experiments. In E. T. Degens, E. Izdar, and S. Honjo, eds., Particle Flux in the
Ocean, pp. 279-299. Mitt. Geol.-Paläont. Inst. Univ. Hamburg, 62.
Maillard, L. C. 1913. Formation des matières humiques par action de polypeptides sur
sucres. C. R. Acad. Sci. 156:148-149.
Middelburg, J. J. 1989. A simple rate model for organic matter decomposition in marine
Mopper, K. 1977. Sugars and uronic acids in sediments and water from the Black Sea and
North Sea with emphasis on analytical techniques. Mar. Chem. 5:585-603.
Mopper, K., R. Dawson, G. Liebezeit, and V. Ittekkot. 1980. The monosaccharide spectra of
natural waters. Mar. Chem. 10:55-66.
Müller, P. J. and E. Suess. 1979. Productivity, sedimentation rate, and sedimentary organic
matter in the oceans-I. Organic carbon preservation. Deep Sea Res. 26:1347-1362.
Müller, P. J., E. Suess, and C. A. Ungerer. 1986. Amino acids and amino sugars of surface
particulate and sediment trap material from waters of the Scotia Sea. Deep Sea Res. 33:819-
838.
Rittenberg, S. C., K. O. Emery, J. Hülsemann, E. T. Degens, R. C. Ray, J. H. Reuter, J. R.

Grady, S. H. Richardson, and E. E. Bray. 1963. Biogeochemistry of sediments in
experimental Mohole. J. Sedimentary Petrol. 33:140-172.
Romankevich, E. A. 1984. Geochemistry of Organic Matter in the Ocean. Berlin: Springer

Verlag.
Romankevich, E. A. 1990. Biogeochemical problems of living matter of the present-day

biosphere. In V. Ittekkot, S. Kempe, W. Michaelis, A Spitzy, eds. Facets of Modern
Biogeochemistry, pp. 39-51. Berlin: Springer Verlag.

Seifert, R., K.-C. Emeis, W. Michaelis, and E. T. Degens. 1990. Amino acids and
carbohydrates in sediments and interstitial waters from Site 681, ODP Leg 112, Peru
Continental margin. In E. Suess, R. von Huene, et al. Proc. ODP, Scientific Results, 112:556-
566. College Station, Texas (Ocean Drilling Program).
Seifert, R., K.-C. Emeis, A. Spitzy, K. Strahlendorf, W. Michaelis, and E. T. Degens. 1990.
Geochemistry of labile organic matter in sediments and interstitial water recovered from
Sites 651 and 653, Leg 107 ODP in the Tyrrhenian Sea. In K. A. Kastens, J. Mascle, et al.
Proc. ODP, Scientific Results, 107:591-602. College Station, Texas (Ocean Drilling
Program).
Suess, E. 1970. Interaction of organic compounds with calcium carbonate-I. Association

phenomena and geochemical implications. Geochim. Cosmochim. Acta 34:157-168.
Suess, E. 1980. Particulate organic carbon flux in the oceans-surface productivity and
oxygen utilization. Nature 288:1-3.
Suess, E., R. von Huene et al. 1988. Proc. ODP, Initial Reports, 112: College Station, Texas
(Ocean Drilling Program).
Tarafa, M. E., J. K. Whelan, R. S. Oremland, and R. L. Smith. 1987. Evidence of

microbiological activity in Leg 95 (New Jersey Transect) sediments. In C. W. Poag, A. B.
Watts, et al. eds., Initial Reports. DSDP 95:635-640. Washington, D.C.: U.S. Govt. Printing
Office.
Tegelaar, E., J. W. de Leeuw, S. Derenne, and C. Largeau, 1989. Kerogen revisited.

Wefer, G., P. Heinze, and E. Suess. 1990. Stratigraphy and sedimentation rates from
oxygen isotope composition, organic carbon content, and grain-size distribution at the Peru
upwelling region: Holes 680B and 686B. In E. Suess, R. von Huene, et al., eds., Proc. ODP,
Initial Reports, 112:355-367. College Station Texas (Ocean Drilling Program).
Wefer, G., E. Suess, W. Balzer, G. Liebezeit, P. J. Müller, C. A. Ungerer, and W. Zenk.

1982. Fluxes of biogenic components from sediment trap deployments in circumpolar waters
of the Drake Passage. Nature, London 299:145-147.
sulfate reduction: The G model tested. Limnol. Oceanogr. 29(2):236-249.
Whelan, J. K. 1977. Amino acids in a surface sediment core of the Atlantic abyssal plain.
Geochim. Cosmochim. Acta. 41:803-810.
Whelan, J. K., R. Oremland, M. Tarafa, R. Smith, R. Howarth, and C. Lee. 1986. Evidence
for sulfate-reducing and methane-producing microorganisms in sediments from Sites 618,
619, and 622. In A. H. Bouma, J. M. Coleman, A. W. Meyer, et al., Initial Reports. DSDP
96:767-775. Washington, D.C.: U.S. Govt. Printing Office.



10. Hydrodynamic Controls of Anoxia in Shallow Lakes
The accumulation and preservation of organic carbon in sediments is enhanced by anoxia in the
overlying water column. The causes of anoxia in shallow (< 100 m deep) lakes have been analyzed
with a turbulence closure fluid dynamic model coupled to a simple, three-component geochemical
model which utilizes Michaelis-Menten growth kinetics. The model demonstrates that anoxia is
enhanced by high nutrient concentrations and that lakes of intermediate depth are more likely to
have anoxic water. Turbulent shear energy production (resulting from surface wind stress) and
buoyancy energy production (resulting from surface heat flux) can cause subtle changes in surface
nutrient concentration and therefore photosynthesis. The model successfully reproduces seasonal
time-depth data sets of temperature and oxygen in Esthwaite Water, a eutrophic lake in the English
Lakes district, and Gem Lake, an oligotrophic lake in the Sierra Nevada. The model is less
successful in predicting particulate organic carbon concentrations, in large part due to
simplifications that are inherent in the model development. Preliminary comparison of generalized
model output with paleolimnological reconstruction of open basin lacustrine sediments is
encouraging.
The production, accumulation, and preservation of organic carbon in the sedimentary record is
strongly dependent on the geochemistry and circulation of the overlying water column. The vertical
movement of water transfers nutrients to surface water where they enhance photosynthesis and
hence organic carbon production. Vigorous vertical water exchange also keeps deep water
ventilated, thereby retarding the preservation and accumulation of organic carbon.
The oxygen content of water has long been recognized as a key variable in the carbon cycling of
the earth's surface water. Oxygen, nitrate, sulfate, and other oxidants convert organic carbon into
more soluble inorganic carbon. Depletion of oxygen significantly enhances the amount of organic
carbon preserved in sediments (Blatt, Middleton, and Murray 1980; Emerson 1985). The precise
chemical and physical conditions that cause anoxia have not, however, been clearly established.
Previous studies have focused on two primary factors: (1) surface productivity and the associated
flux of organic carbon to the sediment-water interface and (2) the residence time of deep water.
Note that in this discussion, high organic carbon flux does not imply high organic carbon
accumulation or vice versa. High organic carbon accumulation can occur in oxygenated water and
is a function of organic carbon flux and sedimentation rate as well as oxygen content of the water
(Emerson 1985).
Anoxic settings have been broadly categorized into four types: (1) lakes whose bottom waters are
anoxic as the result of water column stratification; (2) silled marine basins in which freshwater influx
(either rainfall or river inflow) exceeds evaporation; (3) oceanic upwelling zones where deep,
nutrient-rich water flows to the surface adjacent to a continental land mass; and (4) open ocean
anoxia, which typically develops at middepths (200-1000 m) at the edges of some ocean basins
(Demaison and Moore 1980). Various models of oxygen cycling and anoxia have been made for
each of these four settings. A brief description of some of these models within the Demaison and
file:///F|/Usuarios/Juan%20carlos/articulos/articulo...organic%20matter/Organic%20Matter%20Chapter%2010.htm (1 de 14)17/01/2006 06:46:48 p.m.

Moore classification is given below. I then give a detailed discussion of a new model that has been
successfully applied to one particular anoxic setting-shallow lakes.
Previous Numerical Models of Anoxia
Numerical models of dissolved oxygen concentrations are desirable for several reasons.
Continuous spatial and temporal analytical measurements of oxygen are often impossible. On the
other hand, numerical models that have been carefully verified by observational data provide a
continuous representation of the modeling domain. Prognostic dissolved oxygen calculations are
also important to engineers and public officials. For instance, dissolved hydrogen sulfide (a common
component of anoxic water) is very corrosive to hydroelectric turbines in dams. Anoxic bottom water
in reservoirs can therefore have a direct bearing on the economics of hydroelectric projects.
Quantitative models of anoxia can be broadly classified into two endmembers: box models and
continuum models (figure 10.1). Box models assume uniform properties for a given domain (e.g.,
the deep water or surface water of ocean basins). Mass transfer between boxes is expressed as a
series of mass balances and exchange rates. Changing a given property in one box or the
exchange rate between boxes causes a response in the entire model. Box models are attractive
because the computational effort is not burdensome as long as the numbers of boxes and
parameters stay relatively small. Furthermore, box models give a clear picture of the cause and
effect of changing variables. Box models have the disadvantage of being spatially discontinuous,
with the result that many portions of the modeling domain cannot be studied in detail.
Continuum models, on the other hand, solve a differential equation for each quantity that influences
the modeling domain. Each differential equation consists of accumulation, advection, diffusion,
source, and/or sink terms for the variable being considered (figure 10.1). The quantity expressed by
the differential equation varies continuously in space and time. In theory, a continuum model is
simply a box model with an infinite number of "boxes." Continuum models have the advantage of
showing the subtle and/or continuous changes of a property over the entire modeling domain. The
disadvantage of continuum models is the complex solution of realistic problems (generally by some
form of numerical method). This disadvantage becomes particularly acute for the 2- and 3-
dimensional models that are often necessary to realistically represent a domain.
Both box and continuum models have been effectively applied to marine and nonmarine anoxic
regimes. Many early models were prompted by the desire of environmental engineers to predict the
rate of eutrophication in lakes and reservoirs. Typically, these models treat different portions of the
water column (e.g., the hypolimnion and epilimnion) as "continuous stirred tank reactors" (CSTR)
(Snodgrass and O'Melia 1975; Higgins and Kim 1982; Gachter and Imboden 1985). CSTR models
are essentially box models with a time derivative for each variable. For reservoirs and lakes in
which the CSTR representation is appropriate, these models predict dissolved oxygen quite well. A
frequent disadvantage of CSTR models is the large number of empirical constants that must be
determined for each modeling problem (e.g., Snodgrass and O'Melia 1975:943-944).
Open ocean anoxia has been modeled from both box and continuum model perspectives. Southam
and Peterson (1985) describe a COP (carbon-oxygen-phosphate) continuum model that represents
the vertical distribution of chemical components over 0-4 km depths of the open ocean. The model
is used to show that deep sea oxygen profiles can change dramatically over several hundred years
in response to changes in mixed layer nutrient concentrations. These time scales are consistent

with the time rate of change in atmospheric pressure of carbon dioxide (pCO2) recorded in ice
cores.
Shaffer (1989) uses a continuum model to study the relationships among oxygen, nitrogen,
phosphorous, and sulfur in the ocean. The model uses simple physics to predict complex
biogeochemical interactions. Model results show that anoxia in ocean basins is most likely when
biological productivity is phosphorous limited and there is no phosphorite formation on the sea floor.
Under other conditions (i.e., nitrogen limitation of productivity), oceanwide anoxia is difficult to
maintain over geological timescales.
Multiple box models have been used to examine the causes of anoxia over whole ocean basins
(Sarmiento, Herbert, and Toggweiler 1988a). These studies demonstrate that anoxia is not simply a
function of long deep water residence time (which implies slow ocean circulation) or high surface
productivity. Rather, oceanwide anoxia is critically dependent on the ratio of oxygen to reduced
nutrients in areas of deep water formation. This ocean basin box model has been extended to the
prediction of anoxia in the Mediterranean Sea (Sarmiento, Herbert, and Toggweiler 1988b). In this
study, the Eastern and Western Mediterranean are treated as separate silled basins. The present
exchange of water between the Eastern and Western Mediterranean prevents anoxia from
occurring in either basin. In the geologic past, freshening of water in the Eastern Mediterranean
reversed the exchange between these two basins. Under these conditions, anoxia developed in the
Eastern Mediterranean but not in the Western Mediterranean.
Present Study
This paper presents a new type of continuum anoxia model. The model is well suited for studying
anoxia that develops over relatively short time scales (the order of one year or less) in shallow
water (less than 100 m depth). The present study is applied only to 1-dimensional (vertical)
domains. Although the model requires moderate computing resources, it is capable of revealing
many processes in the water column that are difficult or impossible to measure analytically.
One drawback of previous continuum anoxia models has been their inability to predict, a priori, the
vertical mixing coefficient, eddy diffusivity. Eddy diffusivities are usually established on the basis of
chemical tracer studies or a knowledge of the general physical setting of the modeling domain. This
approach has proven satisfactory for studies in a relatively static water column where temporal
variations of eddy diffusivity are small. In settings characterized by rapidly changing physical
forcings, such as estuaries or the seasonal stratification and overturning in lakes, eddy diffusivities
change many orders of magnitude in a short time. Even relatively static water columns have highly
variable eddy diffusivities. For instance, empirical measurements of lake eddy diffusivities range
from 10-5 to 10-9 m2/s (Quay et al. 1980).
An alternative to using empirically derived eddy diffusivities involves computation of these terms
from characteristics of the flow field itself. This approach parameterizes eddy diffusivity and eddy
viscosity in terms of turbulent kinetic energy per unit mass. Solving for turbulent kinetic energy
necessitates some sort of closure scheme, wherein both mean and turbulent variables are solved.
The second-order turbulence closure scheme outlined by Mellor and Yamada (1974, 1982) has
been successfully applied to a variety of settings in the ocean, including studies of the marine
thermocline (Mellor and Durbin 1975), the marine bottom boundary layer (Weatherly and Martin
1978), and coastal estuaries (Oey, Mellor, and Hires 1985).

In the Mellor-Yamada model, the time rate of change of turbulent kinetic energy consists of shear
energy production, buoyancy energy production, and a dissipation term. Shear and buoyancy
energy production of turbulence are critical to understanding oxygen transport in natural waters
(figure 10.2). High winds enhance shear production, which in turn causes oxygen produced in the
photic zone to diffuse to deeper water levels. Surface heating, on the other hand, causes negative
buoyancy production, thereby retarding the mixing of oxygenated surface water with deeper water.
A viable anoxia model must therefore consider the competing effects of shear and buoyancy
production (i.e., wind stress and surface heat flux).
The present study represents a first attempt to couple the turbulence closure model of Mellor and
Yamada to a simple ecosystem model. The goals of the study are (1) to determine whether the
model can accurately reproduce detailed geochemical data sets (specifically, dissolved oxygen and
nutrients) in lakes and (2) to develop the theoretical framework whereby anoxia can be predicted
from physical forcings such as wind and heat flux. Surface temperature and synoptic wind speed
data sets are often readily available near lakes. On the other hand, measuring detailed physical and
chemical properties over an entire water column on a continuing basis is a formidable analytical
task. It is hoped that models such as the one described here can eventually alleviate some of the
need for continuous field measurements in surface waters.
A variety of ecological models has been proposed to describe the distribution of nutrients and living
organisms in natural waters (e.g., Walsh 1975; Jamart et al. 1977; Wroblewski 1977; Kremer and
Nixon 1978; Hoffman and Ambler 1988; Hoffman 1988). These models all use varients of Michaelis-
Menten kinetics to relate nutrient uptake to phytoplankton growth and have been reasonably
successful in their reproduction of field data.
For this study, a simple three-component chemical model consisting of dissolved oxygen (O2),
dissolved phosphate (PO4), and particulate organic carbon (POC) was used. The model is similar
to other simple oxygen models (e.g., the COP model of Southam and Peterson 1985). The simple
geochemical model allows examination of the critical physical and chemical variables that control
anoxia with reliance on a minimum of empirical constants. Details of the physical and chemical
model are given in appendix 10.1.
Lakes were chosen for this initial study because much of the physical, chemical, and biological
dynamics of lakes can be represented by a 1-dimensional model. The 1-dimensional model
described here is now being extended to 2- and 3-dimensional models. It is hoped that these more
sophisticated models will allow examination of the details of anoxic processes in silled basins,
coastal upwelling zones, and other modeling domains of interest.
Idealized Lake Models
The model described in appendix 10.1 has been used to examine the physical and biological
parameters that lead to lake anoxia. In the following discussion, an idealized model run over a
single season is described. The idealized model permits detailed examination of energy balances,
mixing, and surface productivity in a typical lake. Sensitivity analyses are used to examine the effect
of lake depth and nutrient concentration on the development of anoxia. The sensitivity analyses are
simply meant to highlight variables that are most critical to the development of lake anoxia. Absolute
values of these variables should not be extrapolated from this example owing to the idealized

nature of the model.
For the idealized model, an isothermal water column of 6°C was the initial temperature condition.
Dissolved PO4 was uniformly distributed throughout a 15-m-deep water column. Initial PO4 was 2
mole/L. A small amount of POC (0.2 mole/L) throughout the water column was assumed to
represent the seed phytoplankton. Oxygen was set equal to 100% of saturation throughout the
water column. The water column was gradually stratified by increasing surface temperature to 19°C
as a sine function over 125 days. The surface temperature was then decreased to 6°C over an
additional 125 days. Overturn usually homogenized the water column at a temperature of about 8°C
(figure 10.3a). The model was not run for an entire year, owing to complications of modeling an ice-
covered lake. Wind stress was modeled as a daily sine wave with maximum value of 0.2 dynes/
cm2. Additional lake parameters are given in table 10.1. The idealized lake run parameters closely
correspond to those of Esthwaite Water, the eutrophic lake described in more detail below.
Representative Model Run
The influence of physical parameters on POC and dissolved O2 can be seen in figure 10.3. Eddy
diffusivity, Kh, is relatively high in the epilimnion (10-3m2/s) but approaches molecular values (10-6 -
10-7 m2/s) in the hypolimnion (figure 10.3b). In late August and September, the deepening of the
thermocline results in enhanced diffusivity (Kh = 3 x 10-3 m2/s) through much of the water column.
An increase in surface POC occurs in late April (figure 10.3c). By June and July, surface POC
declines significantly. After the surface temperature reaches a maximum in July, surface
productivity and POC steadily increase. This increase is the result of deep water nutrients being
mixed back into the photic zone by increasing eddy diffusivity (figure 10.3b). Maximum POC is
reached when water column overturn mixes deep water back to the surface. Oxygen is gradually
depleted from the hypolimnion after the onset of thermal stratification (figure 10.3d). Bottom water
anoxia (< 5 mole/L) appears in late July and persists until the lake overturn in November.
The idealized model run shows the influence that shear and buoyancy energy production have on
surface nutrients and hence POC production (figure 10.4). The shear energy production rate is at a
maximum during the spring and then steadily decreases. By late August shear production is
approximately 50% of its springtime value. Following this seasonal minimum, shear production
steadily but irregularly increases until lake overturn. The decrease in shear production during
stratification is caused by the inability of energy produced at the surface to be transmitted through
the thermocline. Buoyancy energy production is antithetic to shear production (figure 10.4).
Buoyancy production is negative throughout the period of thermal stratification but becomes positive
once thermal overturn has passed. Note that the absolute amount of shear production (resulting
from wind stress input) is more than an order of magnitude greater than buoyancy production
(which results from surface heat flux). The irregular pattern of shear and buoyancy production
reflects the fact that the model never achieves steady state. This is probably due to the continuously
varying surface wind stress and heat flux boundary conditions (appendix 10.1, equation 15 and 17).
Energy production irregularity is particularly pronounced following the thermal maximum (figure
10.4).
The influence of the combined energy production on surface nutrients and therefore photosynthesis

is reflected in a plot of vertically integrated photic zone PO4 versus time (figure 10.4). The photic
zone is here defined as that portion of the water column in which photosynthesis exceeds
respiration. The dropoff in photic zone PO4 during the first month of the model run is caused by the
spring uptake of POC (figure 10.3c). By late spring, photosynthesis has effectively exhausted photic
zone nutrients. By August, however, photic zone PO4 begins to rise as a result of increased water
column energy (i.e., the combined effect of buoyancy and shear production). In other words, when
surface water passes the thermal maximum, deep water nutrients begin to be mixed back into the
photic zone as the result of enhanced buoyancy and shear production and, therefore, eddy
diffusivity. This effect is extremely subtle. Model phosphorous concentrations increase only slightly
(less than 5 x 10-3 mole/L) as a result of the change. Such small concentrations are well below the
detection limit of routine water analyses, and therefore this effect has probably gone unnoticed in
previous field studies of lakes.
Influence of Nutrient Concentration and Lake Depth
The total nutrient concentration of a lake has long been recognized as a key variable in determining
the oxygen concentration of the bottom water (e.g., Gachter and Imboden 1985). The model
presented here reconfirms this. A series of idealized model runs was conducted in which only initial
PO4 was changed. Low initial PO4 (the order of 0.1 mole/L) causes little change in bottom water
anoxia (figure 10.5a). Concentrations an order of magnitude larger are necessary to bring about
complete bottom water anoxia. Once again, caution should be used in extrapolating these results
beyond the characteristics of this idealized lake (table 10.1).
Since model PO4 (particulate plus dissolved) remains constant (appendix 10.1), the model cannot
evaluate the effect of "PO4 loading," i.e., the yearly influx of nutrients into the lake. PO4 loading is
considered a key variable in determining whether a lake will become eutrophic as the result of
anthropogenic inputs (Vollenweider 1975). This study simply addresses the seasonal changes in
PO4 and O2 without considering how longer term nutrient changes may cause anoxia.
Lake depth plays a critical role in determining the oxygen level of lake bottom water (figure 10.5b).
Lake depth reflects total water volume and, hence, is a measure of the lake's chemical buffering
capacity. In the specific case of seasonal anoxia, large, deep lakes that become oxygenated during
spring overturn have the capacity to oxidize large amounts of organic carbon that sink out of the
photic zone.
The ability of deep lakes to oxidize large amounts of organic matter has been recognized for many
years. What is less obvious from previous studies is the ability of shallow lakes to remain
oxygenated under the same conditions that cause somewhat deeper lakes to go anoxic (figure
10.5b). This is the result of wind-induced turbulence, which can mix oxygen to the bottom of shallow
lakes. In deep lakes, large volumes of deep water oxygen remineralize organic carbon produced in
the photic zone. Lakes are most likely to go anoxic at intermediate depth ranges (15-25 m for the
conditions considered here). This depth range of anoxic bottom water would be considerably wider
for higher average dissolved nutrient concentrations.
Sensitivity analyses of variables other than lake depth and initial PO4 content were conducted. The
results are summarized in table 10.2.

Application to Specific Lakes
The mathematical model described in appendix 10.1 has been applied to two lakes with widely
varying chemical and biological properties. The areal size of the two lakes is small, however. This
helps maintain the simplifications inherent in a 1-dimensional model. A summary of lake
characteristics and modeling parameters is given in table 10.3.
Esthwaite Water
Esthwaite Water is a small biologically productive lake in the English Lakes district of northern
England. The lake displays seasonal anoxia below a well-developed thermocline (Mortimer 1941,
1942; Sholkovitz and Copeland 1982). The lake also has a well-documented seasonal
phytoplankton succession (George and Heany 1978; Heany and Talling 1980). Diatoms are the
dominant phytoplankton in the early spring. In the late spring and summer, the dinoflagellate
Ceratum hiumdimnella and blue-green algae predominate. The dinoflagellates form cysts and
disappear from the water column in the fall.
Wind stress to the Esthwaite Water simulations was set at a maximum of 0.35 dynes/cm2. This
roughly corresponds to the maximum daily wind speeds of 5-8 m/s reported in previous studies
(Heany and Talling 1980). Maximum horizontal velocity at 0.5 m depth predicted by the model (14
cm/s) compares reasonably well with reported values of 9-11 cm/s (George and Heany 1978).
The model simulations for Esthwaite Water were run for a 250-day season (mid-March to early
November). Surface temperature values were taken from Sholkovitz and Copeland (1982). Initial
PO4 was set at 2 mole/L (with the assumption of a Redfield ratio for the NO3 data of George and
Heany [1978]). Two sets of biological parameters were used in the Esthwaite Water simulations.
From mid-March until the end of June, settling velocity, nutrient half saturation constant, and light
half saturation constant typical of diatoms were used (table 10.3). Diatoms are relatively large and
have dense, siliceous tests. Diatom settling rates are therefore higher than those of most other
phytoplankton types. Nutrient half saturation constants are higher for larger phytoplankton (Epply,
Rogers, and McCarthy 1969) and light half saturation constants are generally lower (Parsons,
Takahashi, and Hargrave 1984). After June 30, biological parameters typical of dinoflagellates were
used. Dinoflagellates are smaller than diatoms and therefore have smaller settling velocities and
nutrient half saturation constants (table 10.3).
Model runs show excellent reproduction of the Esthwaite temperature profile (figure 10.6). Rapid
heating occurs in May and early June. The thermocline is well developed at 6-8 m depth during
most of the summer. Overturning occurs in early November. The model deviates from the data by
showing penetration of the 18°C isotherm somewhat deeper than the data indicate.
Oxygen profiles are also reasonably well predicted by the model (figure 10.7). Oxygen depletion
begins to occur in the lower portion of the water column following the initiation of thermal
stratification. This initial anoxia is believed to be the result of oxygen diffusion into the sediments
(Sholkovitz and Copeland 1982). The model runs confirm this observation. Rapidly sinking diatoms
(the dominant phytoplankton type) cause POC to build up in the sediments (equation 10A.19),
which in turn leads to anoxia. The timescale of POC sinking from the photic zone (slightly less than
15 days) is somewhat less than the half-life of model POC below the photic zone (about 20 days;

table 10.1). POC does not, therefore, remineralize prior to reaching the sediments.
Although most oxygen concentrations are well predicted by the model, the upward migration of
completely anoxic water (defined by Sholkovitz and Copeland (1982) as O2 < 5 mole/L) is
somewhat underpredicted. In the model, complete anoxia extends to 11 m depth whereas the data
show anoxia at 8 m depth (figure 10.7). Other studies have indicated this depth varies between 8
and 10 m from year to year (Mortimer 1941, 1942). This discrepency can be explained by
considering the diffusion length scale, ld, in the hypolimnion:
ld (Kht)1/2 (10.1)
In this equation Kh is eddy diffusivity and t is time. The model predicts that total diffusion (molecular
plus eddy) is close to molecular diffusion (i.e., the order of 10-9 m2/s) during thermal stratification.
From the onset of bottom anoxia (mid-June) until seasonal overturn, ld is approximately 3 m. This
matches the limit of upward migration of anoxia predicted by the model (figure 10.7b). It also
suggests that hypolimnion eddy diffusivities are being underpredicted by the model and that they
may be enhanced by a phenomenon such as internal waves.
Sholkovitz and Copeland (1982) show a period of surface oxygen supersaturation (up to 375 mole/
L) in late August that is not reproduced by the model. This may be the result of phytoplankton
patchiness noted by previous researchers of Esthwaite Water (Heany and Talling 1980). This
patchiness might cause short-term oxygen supersaturation.
POC is rather poorly predicted by the model (figure 10.8). Interpretation of POC data has shown the
carbon cycle to be quite complex in Esthwaite Water (Sholkovitz and Copeland 1982). A spring
diatom bloom causes uniform POC of 40-70 mole/L (figure 10.8a). The model shows the bloom but
gives POC values as much as three times higher. A deep water POC increase in midsummer is
attributed to adsorption of dissolved organic carbon onto precipitating Fe-oxide particles (Sholkovitz
and Copeland 1982). The model obviously does not account for this process; in fact the model
shows a midsummer deep water POC minimum (figure 10.8b).
The model reproduces the midsummer dinoflagellate bloom, which causes surface POC to increase
to 150-180 mole/L (figure 10.8a, b). This bloom can logically be attributed to the dinoflagellate's
ability to grow at lower nutrient levels (as shown by its smaller value of Ks (table 10.1)). This bloom
can also be attributed to the increase in flux of nutrients to the surface following the seasonal
temperature maximum that was observed in the generic model runs (figure 10.4).
Comparison of Esthwaite Water model results with the generic model run is instructive (figure 10.3).
The onset of elevated POC concentrations and bottom water anoxia is much more rapid in
Esthwaite Water, despite the fact that initial PO4 and phytoplankton settling velocity are the same in
both model runs. This phenomenon appears to be the result of different temperature inputs for the
two situations. Thermal stratification exists from the outset of the generic model run, effectively
cutting off diffusion of bottom water nutrients to the photic zone (figure 10.3b). The Esthwaite Water
simulation is controlled by real surface temperature data; stratification does not become well
established until early June (figure 10.6). Vertical mixing of nutrients prior to this time is very
efficient and the spring bloom is more vigorous. Preliminary sensitivity analyses indicate that

variable half saturation constants Ks and Ik (table 10.2) are not the cause of the observed
differences in O2 and POC.
Gem Lake
Gem Lake is a small, oligotrophic lake in the high Sierra Nevada of California. Lake elevation is
3300 m and the lake remains ice covered until early summer. The lake has relatively low dissolved
nutrient concentrations. Oxygen depletion, while irregular on a seasonal basis, is usually minor.
Trace metal nutrients may limit phytoplankton growth more severely than nitrogen or phosphorous
(Stoddard 1986).
Gem Lake simulations were conducted for 160 days using the surface temperature and initial
conditions of Stoddard (1986, 1987). The model used data of 1984, a year in which a complete
spring-fall data set was available. No information is available on the phytoplankton type in Gem
Lake. Biological parameters typical of diatoms were used (tables 10.1, 10.3), although sensitivity
analyses have demonstrated that these parameters do not influence the simulations dramatically
(table 10.2).
Reproduction of temperature profiles is good (figure 10.9). Thermal stratification of Gem Lake is
very slight (top to bottom temperature gradient of 1.5°C in 1984). Model simulations indicate that
thermal stratification is very sensitive to the amount of wind stress applied to such shallow lakes.
The 1984 simulation was conducted at very low daily wind stress values (0.1 dynes/cm2). Higher
wind stress values (0.25 dynes/cm2) caused complete thermal homogenization of the water column.
Oxygen and POC profiles were reasonably well predicted by the model (figures 10.10, 10.11). The
model shows higher bottom water oxygen depletion than the data do (figure 10.10a, b). This could
be due to overestimation of POC flux to the bottom (i.e., phytoplankton settling rate) by the model.
Bottom POC is also overpredicted by the model (figure 10.11a, b). With the exception of bottom
concentrations, predicted oxygen values are within 10% of the data, while POC values are within
30%.
Although bottom water oxygen depletion was minor in Gem Lake in 1984, it was virtually absent in
1982 and 1983 (Stoddard 1987, figure 10.5). This seasonal variability in oxygen concentrations
might logically be explained by variable wind shear stress. Although weather patterns over the lake
are probably the same on a seasonal basis, summer storms may have thoroughly mixed the water
column in some years but not others. Such variable mixing would be most likely to occur in shallow
water bodies such as Gem Lake.
Discussion
A simple three-component geochemical model coupled to a sophisticated turbulence closure fluid

dynamic model has a demonstrated ability to reproduce observed synoptic dissolved oxygen
contents of water in shallow lakes. The model should be an effective tool for the study of oxygen
fluxes in other modeling domains.
The model appears to overpredict particulate organic carbon, POC. Accurate prediction of POC

probably must be done with more sophisticated ecological models than the one presented here.
The overprediction in biologically active lakes is probably due to the fact that zooplankton grazing is
not specifically modeled. Zooplankton fecal material tends to be rapidly transported out of the water
column and into the sediment. Incorporation of a zooplankton component into the model will be a
future improvement.
The model used in this study seems to be an effective tool for predicting bottom water anoxia.
Oxygen is transported through the water column primarily by diffusion, whereas POC is transported
by both diffusion and advection (i.e., particle settling). Eddy diffusivity is the quantity that the Mellor-
Yamada turbulence closure scheme predicts very accurately; therefore, oxygen can be more
accurately modeled. This appears to be particularly true for bottom water anoxia, which is a critical
factor in organic carbon preservation.
The model demonstrates the importance that water depth has on the development of anoxia.
Thermally stratified deep lakes have the ability to remain oxygenated by virtue of their deep water
oxygen content. Very shallow lakes tend to stay oxygenated because of wind-induced shear stress.
If all other conditions are constant, lakes of intermediate depth are the most likely to go anoxic
during the summer. Elevated nutrient concentrations expand the depth range over which bottom
water anoxia can occur. However, this generalization applies only to seasonally anoxic lakes.
Permanently stratified lakes such as Lake Tanganyika or saline lakes would clearly not follow this
pattern.
Examination of selected ancient lacustrine sediments in light of the results and qualifications given
above is, however, interesting. Lacustrine depositional environments can be broadly classified into
open basin and closed basin (Allen and Collinson 1986). The former includes freshwater lakes while
the latter usually implies saline lakes, which are often permanently stratified.
In certain open basin lacustrine sequences, paleodepths have been estimated by a variety of
techniques. Lakes with paleodepth estimates less than 100 m generally have some sort of organic-
rich facies or characteristics of anoxic bottom water (table 10.4) associated with them. On the other
hand, Permo-Triassic sediments of the Karoo basin are hypothesized to have been deposited in a
lake at least 150 m deep (table 10.4). Most of the Karoo sediments show evidence of benthic
organisms, indicating that the bottom water was oxygenated.
These examples appear to fit the generalized model trend of lake bottom anoxia and lake depth
(figure 10.5b), although it is certainly imprudent to extrapolate the results of this study to every open
basin lacustrine setting. Future research will concentrate on combining more advanced models with
detailed field studies. Plans are currently under way to expand the model to 2 or 3 dimensions and
to include more sophisticated treatment of bottom sediment and organic carbon accumulation
boundary conditions. These improvements should permit generalized patterns of organic facies to
be constructed as a function of lake depth and shape. Future field studies of lacustrine sequences
might easily be integrated into the modeling effort. For instance, if the formation of wave-induced
sedimentary structures implies a given wind shear stress and/or lake depth, the model might predict
whether anoxia may have been present above time-equivalent sediments elsewhere in the basin.
Appendix 10.1
Model Description
file:///F|/Usuarios/Juan%20carlos/articulos/articul...rganic%20matter/Organic%20Matter%20Chapter%2010.htm (10 de 14)17/01/2006 06:46:48 p.m.

Acknowledgments
Computing time for this research was provided by the Program in Atmospheric and Oceanic
Sciences, Princeton University, and the Geophysical Fluid Dynamics Laboratory, National Oceanic
and Atmospheric Administration. Discussions with George Mellor and Bob Stallard guided the
modeling effort. The manuscript benefited from reviews by Jorge Sarmiento, Ray Najjar, George
Mellor, Jean Whelan, Eileen Hofmann, and David Glover.
References
Allen, P. A. 1981a. Devonian lake margin and processes, SE Scotland: J. Geol. Soc. Lond.
138:1-14.
Allen, P. A. 1981b. Wave generated structures in the Devonian lacustrine sediments of

southeast Shetland and ancient wave conditions. Sedimentol. 28:369-379.
Allen, P. A. and J. D. Collinson. 1986. Lakes. In H. G. Reading ed., Sedimentary

Environments and Facies, pp. 63-94. Oxford: Blackwell.
Batchelor, G. K. 1967. An Introduction to Fluid Dynamics. Cambridge: Cambridge University

Press.
Berner, R. A. 1980. Early diagenesis. Princeton, N.J.: Princeton Univ. Press.
Blatt, H., G. Middleton, and R. Murray. 1980. Origin of sedimentary rocks. Englewood Cliffs,
N.J.: Prentice Hall.
Broecker, W. S. and T. H. Peng. 1982. Tracers in the sea. Palisades, N.Y.: Eldigo.
Carpenter, J. H. 1966. New measurements of oxygen solubility in pure and natural water.
Limnol. Oceangr. 11:264-277.
Chemical Rubber Company (CRC). 1977. Handbook of Chemistry and Physics. Cleveland,
Ohio: Chemical Rubber Company Press.
Am. Assoc. Petrol. Geol. Bull. 64:1179-1209.
Emerson, S. 1985. Organic carbon preservation in marine sediments. In E. T. Sundquist and

W. S. Broecker, eds., The Carbon Cycle and Atmospheric CO2: Natural Variations Archean
to Present, pp. 78-88. Washington, D.C.: Am. Geophys. Union Geophys. Monograph series,
vol. 32.
Epply, R. W. 1972. Temperature and phytoplankton growth in the sea. Fish Bull. 70:1063-
1085.

Epply, R. W., J. N. Rogers, and J. J. McCarthy. 1969. Half saturation constants for uptake of
nitrate and ammonium by marine phytoplankton. Limnol. Oceanogr. 14:912-920.
Gachter, R. and D. M. Imboden. 1985. Lake restoration. In W. Stumm, ed., Chemical

Process in Lakes, pp. 363-388. New York: Wiley.
George, D. G. and S. I. Heany. 1978. Factors influencing spatial distribution of phytoplankton

in a small productive lake. J. Ecol. 66:133-155.
Goldman, J. C. and E. J. Carpenter. 1974. A kinetic approach to the effect of temperature on

algal growth. Limnol. Oceangr. 19:756-766.
Hanton, J. T. 1969. Algal phosphate uptake, kinetic growth rates and limiting phosphate
concentrations. M. S. thesis, Univ. of North Carolina, Chapel Hill, N.C.
Heany, S. I. and J. F. Talling. 1980. Dynamic aspects of dinoflaggelate distribution patterns

in a small productive lake. J. Ecol. 68:75-94.
Hentz, T. F. 1985. Early Jurassic sedimentation of a rift-valley lake: Culpepper basin,

northern Virginia. Geol. Soc. Am. Bull. 96:92-107.
Higgins, J. M. and B. R. Kim. 1982. DO model for discharge from deep impoundments. Proc.
ASCE, J. Env. Eng. Div. 108 (EE1):107-122.
Hofmann, E. E. 1988. Plankton dynamics on the outer southeastern U. S. continental shelf.

III. A coupled physical-biological model. J. Mar. Res. 46:919-946.
Hofmann, E. E. and J. W. Ambler. 1988. Plankton dynamics on the outer southeastern U. S.

continental shelf. II. A time dependent biological model. J. Mar. Res. 46:883-917.
Hubert, J. F., A. A. Reed, P. J. Carey. 1976. Paleogeography of the East Berlin Formation,
Newark Group, Connecticut Valley. Am. J. Sci. 276:1183-1207.
Jamart, B. M., D. F. Winter, K. Bause, G. C. Anderson, and R. K. Lam. 1977. A theoretical

study of phytoplankton growth and nutrient distribution in the Pacific Ocean off the
northwestern U. S. coast. Deep Sea Res. 24:753-773.
Kremer, J. N. and S. W. Nixon. 1978. A coastal marine ecosystem. Berlin: Springer Verlag.
Li, Y.-H. and S. Gregory. 1974. Diffusion of ions in seawater and in deep sea sediments.
Geoch. Cosmochim. Acta 38:703-714.
Mellor, G. L. and P. A. Durbin. 1975. The structure and dynamics of the ocean surface mixed
layer. J. Phys. Oceanogr. 5:718-728.
Mellor, G. L. and T. Yamada. 1974. A hierarchy of turbulence closure models for planetary

boundary layers. J. Atmos. Sci. 31:1791-1806.
Mellor, G. L. and T. Yamada. 1982. Development of turbulence closure models for

geophysical fluid dynamics problems. Rev. Geophys. Space Phys. 20:851-875.
Mortimer, C. H. 1941. The exchange of dissolved substances between mud and water in
lakes, parts I, II. J. Ecol. 29:280-329.
Mortimer, C. H. 1942. The exchange of dissolved substances between mud and water in
lakes, parts III, IV. J. Ecol. 30:147-201.
Oey, L.-Y., G. L. Mellor, and R. I. Hires. 1985. A three-dimensional simulation of the Hudson-
Raritan estuary. Part I. Description of the model and model simulations. J. Phys. Oceangr.
15:1676-1692.

New York: Pergamon Press.
Quay, P. D., W. S. Broecker, R. H. Hesslein, and D. W. Schindler. 1980. Vertical diffusion

rates determined by tritium tracer experiments in the thermocline and hypolimnion of two
lakes. Limnol. Oceangr. 25:201-218.
Riley, G. A. 1956. Oceangraphy of Long Island Sound, 1952-1954. IX. Production and
utilization of organic matter. Bull. Bingham Oceangr. Coll. 15:324-343.
Sarmiento, J. L., T. D. Herbert, and J. R. Toggweiler. 1988a. Causes of anoxia in the world
ocean. Global Biogeochem. Cycles 2:115-128.
Sarmiento, J. L., T. D. Herbert, and J. R. Toggweiler. 1988b. Mediterranean nutrient balance

and episodes of anoxia. Global Biogeochem. Cycles 2:427-444.
Schindler, D. W. 1974. Eutrophication and recovery in experimental lakes: implications for

lake recovery. Science 184:897-899.
Shaffer, G. 1989. A model of biogeochemical cycling of phosphorous, nitrogen, oxygen, and

sulfur in the ocean: one step toward a global climate model. J. Geophys. Res. 94:1979-2004.
Sholkovitz, E. R. and D. Copeland. 1982. The chemistry of suspended matter in Esthwaite

Water, a biologically productive lake with seasonally anoxic hypolimnion. Geochim.
Cosmochim. Acta 46:46,393-410.
Smayda, T. J. 1970. The suspension and sinking of phytoplankton in the sea. Oceangr. Mar.
Biol. Ann. Rev. 8:353-414.
Smithsonian Meterological Tables. 1951. Publication 4014. Washington, D.C.: The

Smithsonian Institution.

Snodgrass, W. J. and C. R. O'Melia. 1975. Predictive model for phosphorous in lakes.

Environ. Sci. Technol. 9:937-944.
Southam, J. R. and W. H. Peterson. 1985. Transient response of the marine carbon cycle. In
E. T. Sundquist and W. S. Broeker, eds., The Carbon Cycle and Atmospheric CO2: Natural
Variations Archean to Present, pp. 89-98. Am. Geophys. Union Geophys. Monograph series,
v. 32, Washington, D.C.: Am. Geophys. Union.
Stoddard, J. L. 1986. Nutritional status, microcrustacean communities, and susceptibility to

acid precipitation of high elevation lakes in the Sierra Nevada, California. Ph.D. thesis, Univ.
Calif., Santa Barbara.
Stoddard, J. L. 1987. Alkalinity dynamics in an unacidified alpine lake, Sierra Nevada,

California. Limnol. Oceangr. 32:825-839.
VanDijk, D. E., D. K. Hobday, and A. J. Tankard. 1978. Permo-Triassic lacustrine deposits in

the eastern Karoo Basin, Natal, South Africa. In A. Matter and M. E. Tucker, eds., Modern
and Ancient Lake Sediments, pp. 223-238. Int. Assoc. Sedimentologists, spec. publ. 2.
Vollenweider, R. A. 1975. Input-output models. Schwiez. Z. Hydrol. 37:53-84.
Walsh, J. J. 1975. A spatial simulation of the Peruvian upwelling ecosystem. Deep Sea Res.
22:201-236.
Weatherly, G. and P. J. Martin. 1978. On the nature and dynamics of the ocean bottom
boundary layer. J. Phys. Oceangr. 8:557-570.
sulfate reduction: the G-model tested. Limnol. Oceangraph 29(2):236-249.
Wroblewski, J. S. 1977. A model of phytoplankton plume formation during variable Oregon

upwelling. J. Mar. Res. 35:357-394.

Organic Matter: Productivity, Accumulation, and Preservation in Recent and Ancient Sediments,
by Jean K. Whelan and John W. Farrington
11. Organic Carbon Accumulation and Preservation

in Marine Sediments: How Important Is Anoxia?
This paper reviews the available information on the carbon contents of anoxic and oxic sediments,
factors controlling the preservation of organic carbon in marine sediments and the origin of glacial
carbon maxima in the deep sea. Sediments accumulating in oxic and anoxic basins in similar settings
have very similar carbon contents. On continental slopes, carbon maxima appear to be due to the
complex interplay among the supply of carbon to the sea floor, the texture of the sediment, the dilution
of the carbon by other sediment components, and the decreasing settling flux of carbon in the deeper
waters of the open ocean. There does not appear to be a consistent relationship between the zone of
high organic carbon on such slopes and the position of the oxygen minimum.
Available information suggests that the degradation of sedimentary organic matter by aerobes and
sulfate reducers is very similar where the supply of fresh organic matter is similar. Hence, there is little
evidence for the preferential preservation of organic matter under anoxic conditions. Terrestrial organic
matter appears, however, to be degraded to a lesser extent by sulfate reducers. The burial of carbon
below the surficial, oxygenated horizons of a sediment removes the easily oxidized fractions, leaving
material that may be less susceptible to attack by sulfate reducers.
The sedimentary carbon maxima in Pleistocene glacial horizons are probably due to the increased
settling flux of organic matter brought about by climatically induced increases in upwelling in the
equatorial and marginal areas of the ocean. Changes in bottom water oxygen levels during these
periods played a minor role in producing these signals.
Previous work, which used information from the modern ocean to claim that anoxic bottom waters were
prevalent during the accumulation of organic-rich black shales in the geological record, requires
reevaluation. Such facies possibly reflect marked climatically induced changes in ocean circulation and
fertility and can provide important information on the feedbacks involved in the evolution of the Earth's
climate system.
Studies of the concentration and distribution of organic matter in Recent marine sediments have been
carried out for a number of different purposes. Early workers were interested in identifying modern
analogues for the environment of formation of petroleum source rocks, and this led to the compilation
of information on the areal distribution of organic carbon in surficial sediments in discrete areas or
defined basins of sedimentation. Other workers have investigated the central role of organic matter
sources and degradation in controlling the wide range of diagenetic chemical reactions in modern
sediments. More recently, there has been considerable interest in the carbon cycle in the ocean, which
involves studies of the settling and burial fluxes of carbon and associated biogenic elements at discrete
oceanic sites. The latter has important ramifications for the problem of the long-term consequences of
the increase of anthropogenic CO2 in the atmosphere.

Much of the recent resurgence of interest in the factors that lead to the formation of organic-rich facies
in the ocean and have led to the preservation of carbonaceous rocks in the geological record is
probably due to the discovery of especially carbon-rich black shales in the lower and middle
Cretaceous sections of the deep Atlantic and Pacific Oceans. The explanation for the formation of such
rocks has most frequently been by means of the preferential preservation of deposited organic matter
under anoxic conditions. Depending on the depth and areal distributions of the carbon-rich facies, this
has required the wholesale stagnation of the Atlantic Ocean (de Graciansky et al. 1984), alternating
periods of oxygenation and stagnation over an extended time period (Stow and Dean 1984), or the
accumulation of sediments in oxygen minima (Schlanger and Jenkyns 1976; Thiede and van Andel
1977).
In contrast to the emphasis in a great deal of the geological literature that anoxia promotes the
accumulation of organic matter in marine sediments (see Calvert 1987 for a review), recent
oceanographic research has tended to focus on the settling flux of organic matter to the sea floor as
the principal control of the formation of organic-rich deposits (Müller and Suess 1979; Müller,
Erlenkeuser and Grafenstein 1983; Pedersen 1983). Hence, attention has been turned to the variability
of primary production in the near-surface layers of the sea and the factors that ultimately cause
variations in the fertility of the ocean. This information has been used to explain the occurrence of
carbon maxima in late Pleistocene sections in the deep sea by variations in production. These are
thought to have been brought about, in turn, by climatically controlled variations in the intensity of
upwelling in the open ocean. While this mechanism appears to be generally accepted, there is
discussion about the possibility that lower oxygen contents in deep waters could also have caused a
higher degree of preservation of the deposited carbon during glacial maxima (Boyle and Keigwin 1982;
Curry and Lohmann 1983; Emerson 1985; Boyle 1988).
In this chapter we review the factors that control the accumulation of organic matter in marine
sediments, describe the distribution of organic carbon in modern marine oxic and anoxic sediments,
examine the evidence for the preferential preservation under anoxic conditions, and review the
evidence for and against the existence of oxygen-poor conditions in deep water during glacial maxima.
Concentration of Organic Matter in Oxic and AnoxicMarine Sediments
Measurements of the organic carbon or nitrogen content of Recent marine sediments have been
carried out as part of the characterization of marine deposits for a long time period. The most extensive
set of data is provided by Soviet workers, and this has been summarized by Bordovskiy (1965) and
Romankevich (1977). Together with other, more limited data sets, this information has been compiled
into a series of maps by Premuzic et al. (1982). This shows (figure 11.1) that sediments containing
more than 0.5% by weight organic carbon are confined to areas bordering the deep ocean, that is,
areas of the continental shelf, marginal basins, and the continental slope. Over most of the abyssal sea
floor, carbon contents are generally less than 0.25% and only slightly higher in the equatorial, sub-
Arctic and sub-Antarctic regions.
The concentrations of organic matter observed in modern marine sediments are controlled by the rate
of supply of organic matter, the degree of preservation of the deposited material, and the degree to
which this fraction is diluted by other sediment components. The latter factor indicates that valid
comparisons between sediments from different depositional settings can be made only if the
accumulation rates of all sediment components are available. This approach was emphasized by
Koczy (1950) but has been largely ignored. Hence, the data required to assess the relative
accumulation rates of organic carbon and other sediment components in different depositional settings
are very scanty. It is for this reason, and because the discussion in the literature is concerned with the

factors controlling the concentration of organic carbon in sediments, that we also discuss in this paper
sedimentary carbon contents rather than accumulation rates.
Supply of Organic Matter to Marine Sediments
Organic matter is supplied to marine sediments from both marine and terrestrial sources. Studies of the
carbon isotopic composition of sedimentary organic matter (Sackett and Thompson 1963; Sackett
1964) and the distribution of lignin in marine sediments (Hedges and Parker 1976; Hedges, 1981)
suggest that the influence of terrestrial sources is limited to areas within a few tens of km of the
coastline so that marine organic matter is thought to be the dominant component in most marine
deposits. Terrestrial organic matter may constitute a significant proportion of the total in some shelf and
slope areas of active modern sedimentation. In the Gulf of Mexico, for example, Hedges and Parker
(1976) showed that the proportion of terrestrial organic matter in the shelf sediments can reach 50% of
the total in areas close to the Mississippi River. This proportion decreases steadily offshore, and
terrestrial material is virtually absent in the midshelf regions. In the midshelf depositional center off the
Columbia River on the Washington shelf, the proportion of terrestrial carbon ranges between 46 and
61%, as shown by the distribution of lignins (Hedges and Mann 1979), or between 47 and 87% as
shown by the abundance of n-alkanes (Prahl and Muehlhausen 1989). On the continental slope off
Washington, Hedges and Mann (1979) concluded that <10% of the organic fraction on the same slope
region was terrestrial, whereas the distribution of n-alkanes indicates that >20% of the total organic
carbon (TOC) is of terrestrial origin (Prahl and Muehlhausen 1989), suggesting that the influence of the
Columbia River extends across this shelf region.

Berner (1982), Meybeck (1982), and Ittekkot (1988) have estimated the mass of particulate organic
carbon supplied to the ocean by rivers; the results range from 135 to 231 x 1012 g/yr. This compares
with the estimate of the total primary production in the ocean of 27 x 1015 g/yr (Berger et al. 1988). On
the assumption that 1% of the primary production reaches the sea floor at water depths of 5000 m
(Suess 1980), it can be seen that marine and terrestrial sources of organic carbon are equally
important on a global scale. However, Berner (1982) argued that almost all of the river-borne
particulate carbon accumulates in deltaic and shelf environments. Hence, planktonic carbon is probably
more important in deep-sea sediments, and terrestrial carbon, depending on the location of river
sources and the intensity of shelf production, would be more important in marginal environments.
In the deep ocean, the general consensus has indeed been that the influence of terrestrial organic
matter is minor (Degens 1969; Hedges and Parker 1976). However, the presence of vascular plant
lipids in many deep-sea sediments (Lee and Wakeham, 1989) suggests that terrestrial organic matter

does occur in this environment. This has recently been emphasized by Prahl and Muehlhausen (1989)
and Prahl, Muehlhausen, and Lyle (1989), who have shown that the abundance of n-alkanes derived
from terrestrial plant waxes strongly indicates a significant input of terrigenous organic matter to the
open ocean. At Manop Site C, in the central equatorial Pacific, for example, Prahl, Muehlhausen, and
Lyle (1989) conclude that ca. 20% of the total organic carbon is introduced by long-range aeolian
transport from the continents, the remainder being derived from marine productivity. Hence, terrigenous
organic matter may be more important in some deep-sea sediments than previously thought. These
intriguing results are not consistent with the earlier conclusions on the sources of organic matter in
deep-sea sediments. For example, Emerson et al. (1987) have shown that the 14C activity of the
organic matter in some deep-sea sediment is not consistent with the presence of a large fraction of
"old," presumably terrestrially derived, organic carbon and that, consequently, most of the organic
matter in these deposits has been recently derived from the plankton. Further work is clearly required
to resolve these conflicting results.
Marine organic matter is derived overwhelmingly from the plankton within the upper 100 m of the sea
surface. Productivity (autotrophic carbon fixation) is controlled by light and nutrient supply (Ryther
1963). Sustained production occurs only if nutrients are continuously supplied to the near-surface
layers of the ocean where solar radiation is adequate for photosynthesis. In the centers of the
midlatitude oceanic gyres, where solar radiation is sufficient at all times of the year, production is
nutrient limited because the stable stratification of the water column limits the transport of deeper,
nutrient-rich water to the surface. In some coastal regions, upwelling refertilizes the surface layers
either seasonally or more or less continuously, so that production is only light limited. Finally,
production is strongly seasonal in high latitudes because of insufficient light and intense vertical mixing
in the winter months, whereas the fertilized surface layers experience increased light and stability
during the summer months.
A substantial fraction of the organic matter produced in the photic zone is recycled by grazing and
excretion. Only a small fraction settles out of this layer as biogenic particles. This so-called export flux,
as a proportion of the total production, is evidently greater in the productive nearshore and upwelling
zones than it is in the open ocean (Parsons, Takahashi, and Hargrave 1977; Eppley and Peterson
1979), probably because of the inefficient consumption of phytoplanktonic organic matter under
conditions of high production. Hence, it is estimated that less than 10% and around 30% of the total
production is exported in oligotrophic and eutrophic conditions, respectively (Eppley and Peterson
1979). The settling flux, measured by moored particle interceptor traps (Soutar et al. 1977), decreases
with water depth (Suess 1980) so that the flux of carbon at 200, 1000, and 5000 m depth is around
10%, 3%, and 1% of the production, respectively (Berger et al. 1988). The relationship between the
areal variations in production and bathymetry means that, in spite of the much smaller area of the
marginal regions (shelf and slope and sub-Arctic areas) relative to the open ocean, at least half of the
global settling flux of organic matter from the surface layers is deposited in these areas (Deuser 1979;
Berger et al. 1988).
The distribution of organic carbon in surficial sediments of the Pacific Ocean (figure 11.1) is readily
explained by the combination of the marked regional variation in primary production and the depth-
controlled settling flux of organic matter. The marginal areas are generally much more productive so
that a high absolute flux and a higher flux relative to the production rate will occur here. In contrast,
abyssal sediments are carbon poor because the centers of the oceanic gyres are poorly productive and
the absolute and relative settling fluxes are the lowest at great water depths. Walsh et al. (1981, 1985)
have suggested that a substantial fraction of the carbon produced over the continental shelves is
exported to deeper water beyond the shelf edge, thereby augmenting the settling flux of organic carbon
to these areas. The upper parts of the continental slope could, therefore, constitute important sinks for
organic carbon. Although Rowe et al. (1986), using more recent data from a study of shelf-edge

exchange processes, showed that the amount of carbon exported from the shelves is probably much
less than that estimated by Walsh et al. (1985), Hargrave (1986) has pointed out that the proportion of
deposited carbon actually buried in the sediments reaches a maximum in water depths between 200 to
500 m. Hence, the upper continental slope may well be the site of accumulation of a large fraction of
the carbon produced photosynthetically in the marginal areas of the ocean (Walsh 1989). This is
consistent with estimates that show that the inventory of reduced carbon stored in marginal
environments is higher by a factor of six than that of the deep ocean and the difference in the
preanthropogenic settling fluxes to the sea floor are higher by a factor of 20 in the marginal areas
compared with the deep ocean (Holser et al. 1988). The fact that organic carbon probably accumulates
at a maximal rate in upper slope sediments is relevant to the problem of the apparent relationship
between the carbon content of continental slope sediments and the position of the oxygen minimum,
which we examine later.
Anoxic Basins
For the purposes of this review, anoxic basins are defined as those environments where the bottom
water contains free hydrogen sulfide. Such conditions are found in areas where the replenishment of
deep water is restricted and the oxygen content of the resident waters is completely used by the
oxidation of particulate organic matter settling from shallower waters. Examples include some of the
fjords of Norway (Strom 1936) and British Columbia (Richards 1965), the Cariaco Trench (Richards
and Vaccaro 1956), and the Black Sea (Andrusov 1890; Caspers 1957). These are all marginal basins
and should be compared with other marginal areas where oxic conditions prevail.
The concentration and distribution of organic carbon in the sediments of some of the fjords of Norway
have been reported by Doff (1969), Hamilton-Taylor and Price (1983), Skei (1988), and Skei, Loring,
and Rantala (1988). The carbon contents of an oxic and an anoxic basin in the inner part of Oslo Fjord,
where the texture, the bulk composition, and the accumulation rates of the sediments are similar, are
not significantly different (3.15 0.8% and 3.08 0.8%, respectively (Doff 1969)). The bulk of this
organic material is probably planktonic (C/N = 8-9), reflecting the high production in the inner parts of
this fjord system. In Drammensfjord, a highly restricted, anoxic fjord contiguous with Oslo Fjord, the
sedimentary organic carbon concentration is 2.47 0.7%, and this is overwhelmingly terrestrial (C/N =
12-17). In Bolstadfjord, north of Bergen, Norway, the organic carbon content of the sediments of the
inner, anoxic basin lies between 6 and 7%, generally similar to the sediments of the outer, oxic basin
with 5 to 7% carbon (Hamilton-Taylor and Price 1983).
Sediments deposited during the last century in Framvaren, a permanently anoxic basin in south
Norway with a very low input of terrigenous detritus, contain 9-18% organic carbon, almost entirely of
marine origin (Skei, Loring, and Rantala 1988). Interpretation of these results is hampered by the lack
of comparative sediments accumulating in a sediment-starved setting under oxic conditions.
Gucluer and Gross (1964) and Francois (1988) have determined the distribution of organic carbon in
the sediments of Saanich Inlet, British Columbia, a 200-m-deep, intermittently anoxic basin on
Vancouver Island. Maximum concentrations of 4.5% to 5% are found here, the highest values occurring
toward the head of the inlet. Francois (1988) explained this distribution by the progressive dilution of
the deposited carbon by aluminosilicate detritus toward the mouth of the inlet, the major source of such
material, even though primary production is much higher at the mouth. These concentrations may be
compared with those found in neighboring Jervis Inlet, a 600-m-deep, fully oxygenated fjord on the
British Columbia mainland 100 km distant from Saanich Inlet (Calvert 1987), where organic carbon
contents reach 6% by weight. The measured settling fluxes of organic carbon and the sediment
accumulation rates in the two inlets are very similar (Calvert, unpublished); hence, the very different

bottom oxygen concentrations in the two inlets do not appear to cause any significant difference in the
organic carbon content of the sediments.
In the case of the Black Sea, which is considered the "type" anoxic basin (Glenn and Arthur 1984) and
which is often used as a modern analogue for the environment of formation of black shales
(Woolnough 1937), the modern sediment facies has 2 to 4% organic carbon, with isolated maximum
concentrations of 5% (Glagoleva 1961; Strakhov 1962; Ross and Degens 1974; Shimkus and Trimonis
1974; Calvert, Vogel, and Southon 1987; Hay 1988). As shown above, such values do not appear to be
exceptional and are often found in many oxic muds in marginal environments. Moreover, the data of
Calvert, Vogel, and Southon (1987) show that the relationship between the accumulation rate of
organic carbon and the sedimentation rate in the modern facies conforms closely with that shown by
oxic sediments discussed by Müller and Suess (1979, fig. 4); their Black Sea data were taken from an
earlier study wherein the accumulation rates were poorly constrained. In addition, the ratio of the
carbon accumulation rate in the modern facies to the primary production rate (taken from Deuser 1979)
and the bulk sedimentation rate does not deviate from the relationship for oxic sediments deduced by
Müller and Suess (1979). In contrast, if the higher average primary production of Sorokin (1983) is
taken, the Black Sea sediments actually have an anomalously low burial of organic carbon relative to
the production rate. Hence, the burial rate of organic carbon in this anoxic basin does not appear to be
higher than in oxic basins, given similar water depths, sedimentation rates, and primary productivity.
A pre-Recent sapropel, where organic carbon contents can reach 20% by weight, occurs below the
modern facies (Degens and Ross 1972). However, it is important not to confuse this facies with the
modern coccolith marls, which have lower organic carbon contents. The sapropel apparently
accumulated under quite different conditions when the Black Sea was evolving from the Pleistocene
oxic lake phase to the modern anoxic marine phase. In fact, Calvert (1990) has shown that the
distributions of Mn, I, and Br in the sapropel can be explained only by the presence of dissolved oxygen
in the bottom waters during the accumulation of the sapropel. Consequently, the geochemical evidence
for oxic conditions and the new information on the burial rate of organic carbon both suggest that the
Black Sea is not a good modern analogue for the formation of carbonaceous shales in anoxic marine
basins.
The Oxygen Minimum
The oxygen minimum impinges on the continental slope between depths of around 100 to 1500 m. In
some areas, the concentration of oxygen at these depths approaches zero, and free hydrogen sulfide
has occasionally been detected in oxygen minima, for example, in the Arabian Sea (Ivankenkov and
Rozanov 1961). Such environments have been invoked to explain the occurrence of carbonaceous
shales at intermediate palaeodepths in the Atlantic and Pacific (Schlanger and Jenkyns 1976; Thiede
and van Andel 1977; Dow 1978; Demaison and Moore 1980). It is instructive, therefore, to examine the
distribution of organic carbon on continental slopes to test this hypothesis. It must be borne in mind
during this discussion that, as noted earlier, the upper parts of the continental slope are often the sites
of high carbon deposition because of the intrinsically high rates of production in the marginal ocean and
the loss of shelf carbon to the slope regions by sediment bypassing.
The first reported correlation between the concentration of bottom water oxygen and the carbon
content of the sediments was that of Richards and Redfield (1954). In the northern Gulf of Mexico, a
zone of high carbon contents, reaching 1% by weight, coincides with the depth of the oxygen minimum
where the lowest values reach 130 mole/L. Calvert (1987) observed that the oxygen levels at the
depth of the carbon maximum are as high as those in large areas of the deep Atlantic and Pacific
Oceans, where fully oxic conditions are recognized to prevail. Moreover, the apparent correlation

between bottom water oxygen and sedimentary carbon is probably induced by the interplay between
the texture of the sediments, the supply (settling flux) of carbon, and the rapidly changing water depth
on this slope. It is well known that coarse-grained sediments invariably contain less carbon than their
fine-grained counterparts (Trask 1953) because of the hydraulic equivalence of organic debris and fine
clay particles. In the Gulf of Mexico, the sediments above the carbon maximum are coarser grained
than those lower on the slope (figure .11.2), so that there would inevitably be an increase in the carbon
content of the sediments progressing from the shelf edge to deeper water. The decrease in the carbon
contents below the carbon maximum is probably produced by the offshore decrease in primary
production coupled with the decreased settling flux of carbon in deeper water, as discussed previously.
The evidence from this case for the preferential preservation of organic matter in oxygen minima is
therefore equivocal.
Gross (1967) reported that the sediments of the upper part of the continental slope off Oregon and
Washington contained between 2 and 3% by weight organic carbon, the values decreasing in water
depths shallower than 500 m and deeper than 3000 m. Gross suggested that this maximum is
generally inversely correlated with the oxygen content of the bottom water. However, the organic
carbon maximum is much broader than the oxygen minimum (Gross 1967, fig. 4); although an intense
minimum is centred at 1000 m depth, oxygen values at 3000 m, where the sedimentary carbon content
is more than 2%, are at least 90 mole/L. In addition, the organic carbon contents are clearly inversely
related to grain-size (Gross 1967, fig. 3), and the relict sands on the outer shelf and upper slope
(McManus 1972; Smith and Hopkins 1972) are likely to be deficient in organic carbon. We suggest that
the distribution of organic carbon on this margin is also controlled by sediment texture, the supply of
particulate organic matter from the photic zone, and inputs from the Columbia River and that the
carbon maximum on the slope is not controlled by the oxygen content of the bottom waters.
Farther south, on the continental slope off central California, the distribution of sedimentary organic
carbon passes through a minimum within the oxygen minimum zone, which lies between 500 and 1000
m depth and has oxygen levels less than 20 M (Vercoutere et al. 1987). Values increase in the
surface sediments deeper than the oxygen minimum zone, probably because of the increasing clay
contents of the deeper sediments. Hence, a textural control is also dominant on this continental margin.
Premuzic et al. (1982) provided a summary of the extensive information on the distribution of organic
carbon, nitrogen, and CaCO3 on the U.S. Atlantic continental margin. figure 11.3 shows that, on
average, there are two organic carbon and nitrogen maxima on the slope, one centred at around 1600
m, confirming reports by Emery and Uchupi (1972) and Miller and Lohmann (1982), and a smaller one
at around 300 m. The carbon maximum between 1500 and 2000 m depth on the slope is divorced from
the oxygen minimum, which lies between 150 and 600 m. The carbon content of the sediments is,
however, strongly and positively correlated with the silt and clay content. The shallow maxima also
occur where the CaCO3 content of the sediment is low (around 10%); carbonate values increase
abruptly moving into deeper water, where concentrations reach around 90%. This maximum coincides
with the organic carbon and nitrogen minima. In still deeper water, the carbonate content decreases
abruptly to values around 10 to 20%, while the carbon and nitrogen contents increase in proportion.
This rather complex pattern confirms not only the close interrelationship between the concentrations of
all major components of marine sediments but also the primary relationship between texture and the
organic matter concentration.


The intense oxygen minimum in the Arabian Sea where it intersects the western Indian continental
slope is also used as an example of a site where the accumulation of carbon in bottom sediments is
controlled by the oxygen content of the overlying water (Thiede and van Andel 1977; Slater and
Kroopnick 1984). The data of Marchig (1972) and von Stackelberg (1972) show that a zone of high
organic carbon values, with concentrations reaching 6.9 wt. %, lies on the upper and middle slope
between water depths of 200 and 1500 m, roughly coincident with the oxygen minimum. A close
examination of the data from this area suggests, however, that the sedimentological and

oceanographic controls identified in the Gulf of Mexico, the northeast Pacific, and the northwest Atlantic
are also probably involved here. figure 11.4 shows that the sediments occurring shallower than the
carbon maximum are generally coarse grained and have very high CaCO3 contents. The amount of
sand decreases seaward, so that the upper slope sediments are significantly finer grained. Hence, the
increase in the sedimentary carbon content from the shelf to the upper slope is probably caused by the
depletion of organic carbon in the calcareous sands of the shelf, the higher carbon content of fine-
grained sediments, and the bypassing of the fine-grained material from the shelf to the slope. The
settling flux of carbon decreases seaward because of the increase in water depth and lower primary
productivity as distance from the margin increases, as described by Suess (1980), so that the carbon
content of the sediments decreases in deeper water.
A somewhat similar situation prevails on the western margin of the Arabian Sea, where a carbon
maximum occurs between approximately 500 and 1500 m depth (figure 11.5). Once again, the
distribution of CaCO3 over the region as a whole shows that the sediments shallower than the carbon
maximum have high carbonate contents. In addition, a carbonate minimum coincides with the organic
carbon maximum on the upper slope. These observations are corroborated by more recent research.
Shimmield, Price, and Pedersen (1990), who focused on a geographical area smaller than that studied
earlier by Wiseman and Bennett (1940), have shown that the shelf off southern Oman is covered by
well-sorted quartzose sands. Off central Oman, a tectonically complicated area south of Masirah Island
is characterized by a series of NE-SW trending valleys that separate fault blocks. On the outer shelf in
this region, occasional canyons provide conduits for the transport of terrigenous sands into deeper
water, thereby allowing fine-grained, organic-rich diatomaceous ooze with lower CaCO3 contents to
accumulate in the valleys. Some of these basins are silled, lie within the oxygen minimum, contain
nearly anoxic bottom water, and are floored by faintly to moderately laminated deposits. Off north-
central Oman, north of Masirah Island, a series of ENE-WNW valleys also contains diatomaceous silts,
although these are unlaminated and host a moderately active benthos.
The distribution of organic carbon on the Oman Margin broadly reflects the location of major upwelling
centers (Shimmield, Price, and Pedersen 1990), although local physical reworking, as on bank tops, or
deposition in the quiescent intrashelf basins has an important influence on the organic matter content.
The highest organic carbon concentrations occur in the anoxic sediments of the intrashelf basins and
the valleys north of Masirah Island, where the sediments are very fine grained. As shown in figure
11.6a, however, there is little correlation between the concentration of O2 in bottom water and the
organic carbon distribution, but, as we demonstrate below, there is a reasonably well-defined
relationship between sediment texture and the nature and quantity of the associated organic material.
Because nitrogen is lost preferentially relative to carbon during the early diagenesis of marine organic
matter (Rosenfeld 1981), the C/N ratio potentially provides an index of the extent of degradation of
sedimentary organic material (see Blackburn 1987; Kristensen and Blackburn 1987). We discount the
influence of adsorbed ammonium-N in these organically rich sediments (see Müller 1977). It follows
that if a low bottom water oxygen concentration fosters enhanced preservation of organic matter on the
Oman Margin, a positive correlation should be seen between the C/N ratio and dissolved O2. figure
11.6b, incorporating data from Shimmield, Price, and Pedersen (1990), shows that no correlation exists
between the C/N ratio of the sediments and the bottom water O2 concentration and that there is no
evidence that the organic matter in the laminated silts deposited in the intrashelf basins is less
degraded (lower C/N ratio) than in sediments from other locations. Hence, based on figures 11.6a and
11.6b, the conclusion can be drawn that the impingement of low-oxygen waters on Oman Margin
sediments promotes neither enhanced preservation of organic matter nor a higher organic carbon
content in the bottom sediments. The C/N ratio does, however, correlate with the Cr concentration
(normalized to 0% CaCO3) in the sediments (figure 11.6c). Chromium occurs in the sediments largely

as the heavy resistate mineral chromite, which is derived from weathered ophiolite terranes in eastern
Oman and on Masirah. The element is therefore concentrated in deposits where winnowing has
removed the Cr-poor, fine-grained-size fraction. Thus, its concentration should show a positive
correlation with the C/N ratio because in these deposits it is a proxy measure of sediment texture. Low
C/N ratios therefore characterize the finer size fractions deposited under hydrodynamically quiet
conditions, where degradation of organic material is subdued relative to more physically dynamic
settings. The general negative correlation evident in a plot of organic carbon against Cr (figure 11.6d) is
consistent with this textural interpretation: coarse-grained, Cr-enriched deposits contain less (but more
degraded, as shown by higher C/N ratios) organic matter. Taken together, these results suggest that,
on this margin, the concentration of O2 in bottom water does not appear to influence either the degree
of degradation of the organic matter or the carbon content of the sediments.
A maximum in the organic carbon content of the upper slope sediments also occurs in the Sulu Sea, a
deep, silled basin in the South China Sea. The partial isolation of this 5000-m-deep basin is
responsible for the very uniform hydrographic properties of the waters below 500 m depth, and the low
oxygen content (40 to 50 mol/L) of these waters (van Riel 1943). Exon et al. (1981) showed that the
sedimentary organic carbon content reaches a maximum between depths of 500 and 760 m. The

oxygen levels at the carbon maximum and in deeper water are identical, even though the carbon
contents decrease progressively with depth, presumably because of the decrease in the supply of
organic material with increasing water depth and distance from the surrounding shelves. Kuenen
(1942) earlier found, with one exception, that there was no relationship between the sedimentary
organic carbon content and the bottom oxygen levels in the basins of the Indonesian archipelago
sampled during the Snellius expedition. The exception was the relatively high organic matter content of
the sediments of anoxic Kau Bay, a 500-m-deep, semienclosed nearshore basin on Halmahera Island,
but in this case the relative carbon and nitrogen contents show that terrestrial organic matter is the
predominant constituent of the organic fraction. As we discuss in the next section, although the data
are equivocal, terrestrial organic matter may be less susceptible to microbial degradation under anoxic
conditions.
The organic carbon maximum on the continental slope off northwest Africa, which occurs between
1000 and 2000 m depth, occurs well below the center of the oxygen minimum at around 400 m
(Sarnthein et al. 1982). This is analagous to the situation on the northwest Atlantic slope described
earlier (see Miller and Lohmann 1982). Within the carbon maximum off northwest Africa, moreover,
separate maxima are found below the upwelling centers off Cap Blanc and off the mouth of the
Senegal River (Sarnthein et al. 1982, their fig. 10). These relationships imply that the accumulation of
organic matter on this slope is linked with the settling flux of organic matter and not with the oxygen
content of the bottom waters.
Finally, Calvert (1987) has compared the organic carbon contents of the varved and homogeneous
sediments of the Gulf of California. The varved sediments accumulate in a virtually anoxic oxygen
minimum where a burrowing benthos is absent, whereas the homogeneous sediments, having the
same bulk composition as the varved sediments, accumulate under higher oxygen conditions where an
active burrowing fauna is present (Calvert 1964). Although the varved sediments are intensely anoxic
almost to the surface layer (Berner 1964), their carbon contents are indistinguishable from the
homogenized equivalents in shallower and deeper water. In this case, the carbon input to the two
sediment types and the resulting carbon contents are similar, even though the environments of
deposition are quite different.
Organic Matter Preservation in Marine Sediments
The reactions involved in the decomposition of organic matter in marine sediments have been
intensively studied over the past decade, and a good understanding of the principal controls on this
process is now available. The oxidation of organic matter proceeds via a well-defined progression of
reactions where different electron acceptors are utilized (Froelich et al. 1979). Thus, oxygen is the
preferred electron acceptor during organic matter oxidation in all oxic environments. Oxygen is not
completely consumed in abyssal sediments because of the very small oxidant demand of the small
residue of recalcitrant organic matter that survives settling through the deep water column, which
means that oxygen can diffuse to great depths below the sea floor. In hemipelagic and almost all
nearshore environments, however, oxygen does become exhausted below a variable depth in the
sediment, so that deposited organic matter eventually becomes buried in an anoxic environment.
Oxidation of the residual buried organic debris continues, of course, using alternative electron
acceptors. In anoxic environments, nitrate is quickly consumed and sulfate becomes the dominant
oxidant.
The view that organic matter oxidation does not occur, or is greatly retarded, in the absence of oxygen
does not appear to be supported by recent work. Experimental evidence has shown that the oxidation
of organic matter by sulfate reduction can be as rapid and as efficient as that under oxygenated

conditions (Foree and McCarty 1970; Jewell and McCarty 1971; Otsuki and Hanya 1972a, b; Westrich
and Berner 1984; Kristensen and Blackburn 1987). Moreover, the observations of Murray, Grundmanis,
and Smethie (1978) and Martens and Klump (1984) of the exponential decrease in the concentration of
organic carbon with depth in anoxic sediments necessarily imply that organic matter degradation must
occur in such environments. This is supported by the observations of Aller (1980) and Aller and Yingst
(1980) that the amount of easily utilizable organic matter decreases rapidly below the sediment surface
in anoxic sediments. Jorgensen (1982) has demonstrated that the mass of organic matter mineralized
by sulfate reducers is roughly equal to that degraded by aerobes in coastal sediments. More recently,
Mackin and Swider (1989) showed that the maximum contribution of O2 to total organic matter
decomposition in a coastal mud flat on Long Island Sound was 14% of the degradation, compared with
a minimum contribution of 65-85% from SO42- reduction. Henrichs and Farrington (1987) demonstrated
that the rate of carbon oxidation in the sediments of an anoxic and an oxic basin in the Pettaquamscutt
River estuary were very similar, a result consistent with the earlier observations of Orr and Gaines
(1974) of a remarkably high rate of sulfide production in the same anoxic basin. Hence, both the
anaerobic and aerobic rates of oxidation of organic carbon as well as the relative proportions degraded
by aerobes and anaerobes are probably not significantly different if the supply of readily metabolizable
organic matter and the availability of carbon oxidants are the same.
Although Müller and Mangini (1980) determined that the decay constant for organic matter
decomposition was much smaller at a given sedimentation rate in subsurface anoxic sediments and
interpreted this as a direct reflection of the greater energy efficiency of aerobic bacteria, Reimers and
Suess (1983) have reinterpreted this difference by arguing that such rate constants are overestimated
in pelagic sediments, where sedimentation rates are very low, because of the effects of bioturbation,
which cause a steepening of the metabolizable organic carbon profiles (the carbon content at the
surface is increased by bioturbation). In subsurface anoxic sediments, on the other hand, in cases
where the sedimentation rate is much higher, the model rate constants can easily be underestimated
because of the flattening of the gradient of carbon (carbon is spread over a larger depth interval) by the
transport of metabolizable organic matter below the surface by burrowing fauna. Hence, it is apparently
not possible to compare directly the model rate constants for organic matter degradation derived from
the profiles of buried carbon or porewater constituents.
The effect of macrofaunal bioturbation on the preservation or degradation of organic matter in bottom
sediments is evidently highly variable and the general conclusions are equivocal (Blackburn 1987;
Tenore 1987). Burrowing fauna have several roles: they can irrigate the sediment and thereby transport
oxidants to depths below the sediment-water interface (Aller 1982); they can directly consume some of
the sedimentary detritus on and in the sediment surface; and they are thought to promote increased
microbial activity in the sediments by providing fresh surfaces for microbial colonization (Aller and
Yingst 1985). In the experiments of Kristensen and Blackburn (1987), macrofaunal feeding was
probably responsible for roughly 20% of the particulate organic carbon loss in an oxic microcosm. In
the natural sediments of the Bering and Chukchi Seas, Blackburn (1987) reported that 5 to 70% of the
organic carbon mineralization was due to macrofauna, depending on their population density.
Martin and Sayles (in press) have carried out a modeling study of the influence of the degradation rate,
the mixing (bioturbation) rate, and the bulk sedimentation rate on the burial efficiency (proportion of
settled organic carbon that is actually buried in the sediment) and have concluded that increasing the
biological mixing rate has no effect on the burial efficiency in continental margin sediments and actually
increases the burial efficiency in pelagic sediments. In the latter case, mixing has the effect of
transporting organic matter below the depth over which degradation reactions are important; in the
shallower water sites, the deposited organic matter is much more reactive, degradation occurs over a
larger depth interval, and secondary oxidants are more important. Hence, the degradation rate
constants are greater than those at abyssal depths so that physical mixing becomes less important in

sequestering organic carbon. This study shows that although the effects of macrofaunal feeding on the
degradation of organic matter can be identified in experimental (Kristensen and Blackburn 1987) and
field studies (Blackburn 1987), the mechanism is only one of a number of controls on the degree of
preservation of organic matter in marine sediments, and because of the interrelatedness of other
factors, its effect is highly variable.
Henrichs and Reeburgh (1987) have reviewed the role of anoxic diagenesis in the degradation of
organic matter in marine sediments. Their principal conclusion is that there is no consistent difference
between the rates of carbon oxidation under oxic and anoxic conditions. Canfield (1989) has extended
this analysis by comparing rates of carbon oxidation in oxygenated and anoxic sediments and has
reached broadly similar conclusions. However, in order to explain an apparently enhanced burial
efficiency of organic carbon (fraction of organic carbon delivered to the sediment surface that escapes
degradation) in some slowly accumulating Black Sea sediments, Canfield (1989) requires the depth-
integrated rates of sulfate reduction to be half those for oxic respiration. Although somewhat lower
rates can be included within the scatter in the data available to Canfield, there is, as argued earlier, no
direct evidence that carbon oxidation rates by oxygen and sulfate reduction are significantly different.
Moreover, these results are at variance with the conclusions of Calvert, Vogel, and Southon (1987)
discussed earlier.
A possibly important factor controlling the burial of organic matter in anoxic sediments may be the
limited ability of anaerobic bacteria to metabolize the full range of organic compounds present in
marine sediments. Several workers have specifically indicated that aromatic compounds, in particular
those derived from lignin, are broken down very slowly, if at all, by sulfate reducers (Zeikus 1980;
Crawford 1981; Benner, Maccubbin, and Hodson 1984; Henrichs and Reeburgh 1987). Hence,
terrestrially derived organic matter would have a greater chance of survival in marine sediments, and
anoxic sediments in particular. However, Hamilton and Hedges (1988) found that the degradation rate
constant of lignin phenols was very similar to that of total organic carbon, nitrogen, and neutral sugars
in the anoxic sediments of Saanich Inlet, a result that they recognized was not consistent with other
information. At distances of more than a few tens of km from the coastline, of course, the dominant
constituent of sedimentary organic material is planktonic, so that we need information on the
preservation potential of this material. Emerson and Hedges (1988) have pointed out that there is some
evidence for the selective preservation of bacterial lipids (Harvey, Fallon, and Patton 1986), aliphatic
polymers (Hatcher et al. 1983), and other hydrophobic compounds (Atlas, Boehm, and Calder 1981) in
modern anoxic sediments. In addition, Harvey, Fallon, and Patton (1986) determined experimentally
that increasing concentrations of total organic matter had the effect of inhibiting organic matter
degradation, owing possibly to the protective effect of the adsorption of lipids by the bulk organic
matrix. Emerson and Hedges (1988) suggest that the organic matter in abyssal sediments has a
molecular composition that is much more easily attacked by aerobes than it is by nitrifying or sulfate-
reducing bacteria.
On the other hand, Jorgensen (1982) has pointed out that some strains of sulfate-reducing bacteria
recently isolated from coastal sediments are capable of oxidizing all the major fermentation products
produced in the sediments completely to CO2 and H2O. Further information on this problem is clearly
required. However, the fact that lignin and similar structural polymers do not constitute a substantial
fraction of sediments accumulating more than a few tens of km from the coastline, the possibility that
high organic matter burial fluxes will have the effect of protecting more labile organic materials from
microbial attack, and the experimental evidence that freshly deposited organic matter suffers oxidation
at roughly the same rate under oxic and anoxic conditions lead us to the conclusion that there is no
significant difference in the overall rate of organic matter oxidation in oxygenated and anoxic
environments.

The seminal paper of Emerson (1985) has frequently been used to contend that it is the oxygen
content of the bottom water that is the dominant control on the accumulation and preservation of
organic matter in marine sediments (see, for example, Howell et al. 1988; Kump 1988; Sarmiento,
Herbert, and Toggweiler 1988). Note, however, that Emerson's model was based on the simplifying
assumption that oxidants other than oxygen can be neglected in estimating the fraction of organic
matter likely to be preserved under a range of oxygen concentrations. Hence, this model applies only to
the situation in the deep ocean, where oxygen can diffuse to great depth in the sediments. In this
situation, the rate of diffusion of oxygen into, and therefore its availability in, the sediment will, of
course, depend on the bottom water O2 concentration. Emerson (1985) claimed that there was a
convincing correlation between carbon preservation and the bottom water oxygen content in the
California Borderland basins. In making this analysis, he was forced to assume that the flux of carbon
to the various basins was very similar and that terrestrial carbon constituted a minor fraction of the total
organic matter in the sediments. However, the data of Eppley et al. (1977) show that the standing
stocks of chlorophyll and particulate organic carbon are highly variable in the region considered by
Emerson (1985) and that there are very strong offshore gradients in these parameters. This implies
that there is indeed a variable settling flux of organic matter to the various offshore basins. Moreover,
Jahnke (1990) has reviewed more recent data on the burial fluxes of carbon in the Borderland basins
and has shown that there is, in fact, no consistent relationship between bottom water oxygen and the
burial efficiency of carbon. There is also no correlation between the burial flux and the distance from
the coastline, suggesting that terrestrial carbon, if present in significant amounts, is not an important
factor.
In hemipelagic and nearshore sediments and in all anoxic basins, oxidants other than oxygen are
consumed during the degradation of organic matter. The importance of the secondary oxidants in the
overall carbon degradation has been pointed out by Jorgensen (1982), who compared the measured
rates of oxygen uptake and sulfate reduction in a range of coastal and shelf sediments. More recently,
Sayles and Curry (1988) have shown, using measurements of the carbon isotopic composition of
porewater CO2 in sediments from the slope, a productive marginal basin, and the deep ocean, that
suboxic metabolism (nitrate and manganese oxide reduction) can be an important component of the
overall rate of carbon oxidation in some offshore sediments. The range estimated is from less than 1%
in the western Guatemala Basin, where no nitrate consumption could be detected, to 10% on the Nova
Scotia slope at around 3500 m depth, and around 40% of the total oxidation rate in the Panama Basin
at 3300 m depth. These results are entirely consistent with the order of variation in the settling flux of
carbon at these sites, which is highest in the Panama Basin (9.5 mgC/m2/d: Honjo, Manganini, and
Cole 1982) and lowest in the western part of the Guatemala Basin (2.4 mgC/m2/d: Dymond and Lyle
1985). The small amount of oxygen available is rapidly consumed in the Panama Basin, and instead of
the oxidation of carbon ceasing or slowing down appreciably, the other oxidants are being used to
degrade the carbon that has survived burial below the surficial few cm of the sediment.
In strictly anoxic basins, sulfate is the major electron acceptor for the oxidation of the organic matter
settling to the surface sediments. In this situation, the organic debris has been exposed to oxygen for a
short time period only in the waters overlying the anoxic water mass; hence, relatively fresh material is
exposed to a rather large sulfate reservoir. Based on a comparison of the concentration of oxygen and
sulfate in normal bottom waters and the electron activity of the respective oxidation reactions involving
these two species, the oxidizing capacity of sulfate can be shown to be up to 200 times greater than
that of oxygen. That active carbon oxidation is taking place in the sediments of anoxic basins is shown
by the exponential decrease in the concentration of organic carbon with depth (Murray, Grundmanis,
and Smethie 1978) and the complete consumption of porewater sulfate within the top 40 cm of the
sediments in anoxic Saanich Inlet (Devol and Ahmed 1981). Likewise, sulfate is almost completely
depleted in the upper 1 to 2 m of the sediments of the Black Sea (Manheim and Chan 1974).

A particularly interesting example of the oxidant demand engendered by the deposition of relatively
organic-rich sediments in pelagic settings is furnished by the sporadic and essentially instant input of
organic-rich turbidites from the NW African and European continental margins to the abyssal plains in
the NE Atlantic. The compositional, diagenetic, and sedimentological character of such deposits has
been described in a number of papers (Colley, Wilson, and Higgs 1984; Thomson et al. 1984, 1987;
Colley and Thomson 1985; Wilson et al. 1985, 1986; Jarvis and Higgs 1987). The turbidites are
emplaced onto oxidizing pelagic clays or marls, which establishes a nonsteady-state diagenetic regime
as the contained organic matter is progressively degraded. Wilson et al. (1985) observed that oxidation
of the organic material is essentially confined to a discrete horizon representing the depth of
penetration of a downward-propagating "oxidation front." This locus is marked by quantitative depletion
in interstitial water of downward-diffusing oxygen and nitrate and the reduction of solid-phase
manganese and iron oxides. As noted above, the latter act as preferred electron acceptors after O2
and NO3- have been depleted. Thus, organic matter in the turbidite horizons is oxidized by a suite of
oxidants, not just oxygen. Colley, Wilson, and Higgs (1984) estimated that dissolved nitrate supported
up to about 30% of the total oxidation observed. Sulfate reduction appears to be absent in these
particular deposits. Emerson and Hedges (1988) have perceptively noted that the organic carbon in
these turbidites appears to have been preserved in the presence of sulfate for thousands of years. Why
this should be the case is not clear. It may be that organic matter that has previously been degraded in
a different environment is less prone to attack by sulfate-reducing bacteria but still labile with respect to
aerobes and denitrifiers. Indeed, Emerson and Hedges (1988) observe that structural alteration of
biochemicals will inhibit subsequent enzymatically catalyzed degradation. Partial oxidation of the
organic matter at its original site of deposition prior to slumping, transport, and turbidite emplacement
may have left behind a refractory organic matter fraction that cannot be metabolized by sulfate
reducers.
In contrast to the situation in the turbidites off NW Africa, sulfate reduction, albeit minor (<2mM), does
occur in the pore water in a buried organic carbon maximum (0.5% carbon) on the Porcupine Abyssal
Plain (Wallace et al. 1988) in the North Atlantic. This carbon-rich zone is similar in character to the
turbidites on plains to the south except that its occurrence is attributed to an increased settling flux of
carbon during the last glacial maximum. The presence of some sulfate reduction at the Porcupine site
may reflect greater postburial lability associated with essentially direct settling of marine organic matter
in contrast to the initial deposition, partial degradation, slumping, and redeposition, which preceded the
onset of diagenesis in the turbidites. The Porcupine Abyssal Plain example is particularly noteworthy in
that some sulfate reduction, in addition to quantitative depletion of O2, NO3-, and manganese and iron
oxides, is fostered by a buried carbon maximum, which occurs at a water depth of 4.5 km and contains
a maximum of only S0.5% organic carbon. Clearly, oxidants other than O2 can be important in
organic matter degradation even in deep ocean sediments.
An Anoxic Ocean During Glacial Maxima?
It is well known that an increased rate of accumulation of organic matter, represented by maxima in
profiles of the sedimentary organic carbon concentration, occurred in sediments deposited during the
last glacial maximum under upwelling zones along the equatorial divergence in the Atlantic and Pacific
(Arrhenius 1952; Pedersen 1983; Morris, McCartney and Weaver 1984; Lyle et al. 1988) and
associated with eastern boundary currents (Müller, Erlenkeuser, and von Grafenstein 1983; Zahn,
Winn, and Sarnthein 1986). In all of these cases, the carbon maxima have been interpreted as
reflecting enhanced production induced by increased mean wind speeds and concomitant upwelling
during the last glacial. Finney, Lyle, and Heath (1988), Lyle et al. (1988), and Pedersen (1988) have
shown that in the eastern equatorial Pacific, the association between glacial stages and increased
organic carbon accumulation rates is consistent for at least the last 500,000 years.

It has been suggested that the carbon maxima could represent enhanced preservation resulting from
lower concentrations of oxygen in deep water during glacials (e.g., Emerson, 1985; Boyle, 1988). Boyle
and Keigwin (1982) and Curry and Lohmann (1983) had earlier found evidence for reduced advection
of deep waters into the Atlantic during the last glacial maximum (LGM), which would have had the
effect of reducing the dissolved oxygen content of the deep waters. However, Pedersen et al. (1988)
used four lines of micropalaeontological and geochemical data to discount the possibility that glacial
oxygen levels could have been substantially depleted in the Panama Basin, where the bottom water
oxygen concentration is currently about 120 mol/L. First, both the abundance and size of benthic
foraminifera of the genus Uvigerina were greater in LGM sediments in the Panama Basin, most
probably because of the higher organic matter content; the increased population and larger mean test
size imply that the upper centimeter of the sediments remained oxygen replete, at least where the
infaunal population flourished. Second, the number of meiofaunal fecal pellets increased markedly
during the LGM, which also argues against oxygen depletion and reinforces the idea that the infauna
were more active. Third, there is no significant decrease in the iodine: Corg ratio in the LGM sediments.
Because iodine is depleted relative to carbon in anoxic basins but is enriched in association with
organic matter in oxic environments, the absence of a depletion in the I:Corg ratio during the LGM is
evidence that Panama Basin bottom water remained oxic during that time. Fourth, molybdenum is not
enriched in Panama Basin sediments deposited during the LGM.
This contrasts with the high Mo contents seen in sediments accumulating at present in anoxic basins,
such as Saanich Inlet (Francois 1988) and the Black Sea (Calvert 1990), and deposits in coastal areas
where anoxia prevails within millimetres of the sediment surface (Pedersen, Waters, and Macdonald
1989). In these cases, Mo is extracted from seawater by sulfide phases, which precipitate in the
surface sediments where the water column is anoxic or at very shallow depths in anoxic sediments,
which underlie an oxic water column. Thus, the absence of Mo enrichments in the glacial Panama
Basin sediments, coupled with the other data presented by Pedersen et al. (1988), argues against
significant oxygen depletion in either bottom water or the uppermost portion of the sediment column
during the last glacial period when the accumulation rate of organic carbon was considerably higher
than it is today.
In another approach, Lyle et al. (1988) used carbon isotope data obtained from benthic foraminifera in
the same general area to construct a 300,000-year model record of the bottom water oxygen content in
the eastern equatorial Pacific and showed that there was no correlation between O2 and the organic
carbon accumulation rate during the Late Quaternary. The evidence strongly supports the conclusion
that the increased carbon accumulation is not the result of a low concentration of bottom water oxygen
during glacials. Productivity increases, driven by enhanced atmospheric circulation and upwelling
during glacials, remain the most plausible explanation for the observed carbon maxima. The profiles of
the Cd/Ca ratio in Galapagos Islands corals reported by Shen et al. (1988) show that the mixed layer
was significantly more nutrient rich during the Little Ice Age in the equatorial Pacific. This is interpreted
as a reflection of a more intense upwelling regime in response to a strengthening of the Trade Winds
caused, in turn, by a steeper equator-to-pole temperature gradient during this period. This marked
change in the oceanography of the equatorial Pacific can be used by analogy to support the
interpretation of the origin of the glacial sedimentary carbon maxima in the same region by increased
primary production brought about by a change in ocean climate.
The abundance and distribution of organic carbon in marine sediments are generally consistent with
the dominant controls by the primary production in the overlying water column, the decreasing settling
flux of particulate organic matter through the water column with increasing water depth, the texture of
the sediments (finer grained sediments containing more carbon than their coarser grained equivalents),
and the dilution of the deposited carbon by other sediment components. An examination of the

available information reveals that there is no consistent difference in the concentration of organic
carbon between sediments accumulating under fully anoxic and oxic conditions in basins of similar
settings. Moreover, the frequent occurrence of an organic carbon maximum on many continental
slopes may be due to the complex interplay of the factors that govern the carbon content of sediments
and not the oxygen content of the water in contact with the sea floor.
The oxidation of organic matter in marine sediments probably depends only to a minor extent on the
oxygen content of the bottom water. In other words, the secondary oxidants, nitrate, Mn and Fe
oxyhydroxides, and sulfate, could be equally effective in degrading freshly deposited carbon, so that
the degree of oxygenation of a given basin of deposition is not necessarily the dominant control of the
organic carbon content of sediments and rocks. Further work is required on the nature of the organic
matter that survives burial below the uppermost oxygenated horizons of sediments and the extent to
which this organic fraction can be attacked by sulfate reducers. This point should not, however,
obscure the main conclusion of this analysis, which is that, given similar inputs of organic and other
sedimentary components, the carbon contents of the sediments of anoxic basins (i.e., those with free
H2S in the bottom waters) and oxygenated areas of the sea floor will be very similar.
This conclusion may have important implications for the interpretation of the sedimentary record of
carbon burial in Pleistocene and older sediments and sedimentary rocks. Hence, the enhanced
accumulation of carbon during glacial maxima is consistent with increases in the production of
planktonic organic matter in the ocean, and this is climatically controlled via increases in equatorial and
marginal upwelling because of higher wind speeds during these periods. Such events may therefore
provide valuable palaeoceanographic information. In addition, it appears that the occurrence of organic-
rich black shales in the geological record may not be due to the stagnation of the ocean or the
expansion or intensification of the oxygen minimum. Other explanations for these occurrences are
equally plausible, for example, climatically induced increases in oceanic fertility, as argued for the
Cretaceous by Pedersen and Calvert (1990).
Acknowledgments
We thank N. B. Price and G. Shimmield for permission to use their unpublished data on the
geochemistry of the sediments of the Oman margin, and R. Jahnke for an early copy of his paper on
carbon burial in the California Borderland Basins. K-C. Emeis and an anonymous reviewer provided
very helpful comments and criticisms of the manuscript. This paper was prepared with the support of
the Natural Sciences and Engineering Research Council of Canada.
References
Aller, R. C. 1980. Diagenetic processes near the sediment-water interface of Long Island
Sound. I. Decomposition and nutrient element geochemistry (S,N,P). Adv. in Geophysics,
22:238-350.
Aller, R. C. 1982. The effects of macrobenthos on chemical properties of marine sediment and
overlying water. In P. L. McCall and M. J. S. Trevesz, eds., Animal-Sediment Relations, pp. 53-
102. New York: Plenum.
Aller, R. C. and J. Y. Yingst. 1980. Relationships between microbial distributions and the
anaerobic decomposition of organic matter in surface sediments of Long Island Sound, U.S.A.
Mar. Biol. 56:29-42.

Aller, R. C. and J. Y. Yingst. 1985. Effects of the marine deposit-feeders Heteromastus

filiformis (Polychaeta), Macoma balthica (Bivalvia), and Tellina texana (Bivalvia) on averaged
sedimentary solute transport, reaction rates, and microbial distributions. J. Mar. Res. 43:615-
645.
Andrusov, N. I. 1890. Preliminary account of participation in the Black Sea deep-water

expedition of 1890. Izvestiya Vsesoyuznogo Geograficheskogo Obshchesteva 26:380-409 (in
Russian).
Arrhenius, G. O. 1952. Sediment cores from the East Pacific. In H. Petterson, ed., Reports of
the Swedish Deep-Sea Expedition, vol. 5.
Atlas, R. M., P. D. Boehm, and J. A. Calder. 1981. Chemical and biological weathering of oil,
from the Amoco Cadiz spillage, within the littoral zone. Est. Coastal and Shelf Sci. 12:589-608.
Benner, R., A. D. Maccubbin, and R. E. Hodson. 1984. Anaerobic biodegradation of the lignin
and polysaccharide components of lignocellulose and synthetic lignin by sediment microflora.
Appl. Environ. Microbiol. 47:998-1004.
Berger, W. H., K. Fischer, C. Lai, and G. Wu. 1988. Ocean carbon flux: global maps of primary
production and export production. In C. R. Agegian, ed., Biogeochemical Cycling and Fluxes
Between the Deep Euphotic Zone and Other Oceanic Realms, pp. 131-176. NOAA National
Undersea Research Program, Res. Rept. 88-1.
Berner, R. A. 1964. Distribution and diagenesis of sulfur in some sediments from the Gulf of
California. Mar. Geol. 1:117-140.
Berner, R. A. 1980. Early Diagenesis: A Theoretical Approach. Princeton, N.J.: Princeton Univ.
Press.
Berner, R. A. 1982. Burial of organic carbon and pyrite sulfur in the modern ocean: its
geochemical and environmental significance. Am. J. Sci. 282:451-473.
Blackburn, T. H. 1987. Microbial food webs in sediments. In: M. A. Sleigh, ed., Microbes in the
Sea, pp. 39-58. New York: Wiley.
Bordovskiy, O. K. 1965. Accumulation and transformation of organic substances in marine

sediments. Mar. Geol. 3:1-114.
Boyle, E. A. 1988. The role of vertical chemical fractionation in controlling Late Quaternary
atmospheric carbon dioxide. J. Geophys. Res. 93:15701-15714.
Boyle, E. A. and L. Keigwin. 1982. Deep circulation of the North Atlantic over the last 200,000
years: geochemical evidence. Science 218:784-787
Calvert, S. E. 1964. Factors affecting distribution of laminated diatomaceous sediments in the

Gulf of California. In T. H. van Andel, and G. G. Shor, eds., Marine Geology of the Gulf of

California, pp. 311-330. AAPG Memoir 3.
Calvert, S. E. 1987. Oceanographic controls on the accumulation of organic matter in marine

sediments. In: J. Brooks and A. J. Fleet, eds., Marine Petroleum Source Rocks, pp. 137-151.
London: Blackwell.
Calvert, S. E. 1990. Geochemistry and origin of the Holocene sapropel in the Black Sea. In V.
Ittekkot, S. Kempe, W. Michaelis, and A. Spitzy, eds., Facets of Modern Biochemistry, pp. 328-
353. Berlin: Springer Verlag.
Calvert, S. E., J. C. Vogel, and J. R. Southon. 1987. Carbon accumulation rates and the origin
of the Holocene sapropel in the Black Sea. Geology 15:918-922.
Canfield, D. E. 1989. Sulfate reduction and oxic respiration in marine sediments: implications
for organic carbon preservation in euxinic environments. Deep Sea Res. 36:121-138.
Caspers, H. 1957. Black Sea and Sea of Azov. In J. W. Hedgpeth, ed., Treatise on Marine
Ecology, part 1, pp. 803-890. Geol. Soc. Am. Memoir 67.
Colley, S. and J. Thomson. 1985. Recurrent uranium relocations in distal turbidites emplaced
in pelagic conditions. Geochim. Cosmochim. Acta 49:2339-2348.
Colley, S., J. Thomson, T. R. S. Wilson, and N. C. Higgs. 1984. Post-depositional migration of

elements during diagenesis in brown clay and turbidite sequences in the northeast Atlantic.
Crawford, R. L. 1981. Lignin Biodegradation and Transformation, New York: Wiley.
Curry, W. B. and G. P. Lohmann. 1983. Reduced advection into Atlantic Ocean deep eastern
basins during last glacial maximum. Nature 306:577-580.
Degens, E. T. 1969. Biogeochemistry of the stable carbon isotopes. In G. Eglinton and M. T.

Murphy, eds., Organic Geochemistry: Methods and Results, pp. 304-329. Berlin: Springer
Verlag.
Degens, E. T. and D. A. Ross. 1972. Chronology of the Black Sea over the last 25,000 years.
Chem. Geol. 10:1-16.
de Graciansky, P. C., G. Deroo, J. P. Herbin, L. Montadert, C. Müller, A. Schaaf, and J. Sigal.

1984. Ocean-wide stagnation episode in the late Cretaceous. Nature 308:346-349.
Demaison, G. J. and G. T. Moore, 1980, Anoxic environments and oil source bed genesis:
AAPG Bull. 64:1179-1180.
Deuser, W. G. 1979. Marine biota, nearshore sediments and the global carbon balance. Org.
Geochem. 1:243-247.
Devol, A. H. and S. I. Ahmed. 1981. Are high rates of sulfate reduction associated with

anaerobic oxidation of methane? Nature 291:407-408.
Doff, D. H. 1969. Geochemistry of the sediments of Oslo Fjord, Norway. Doctoral dissertation,
Univ. of Edinburgh, Scotland.
Dow, W. G. 1978. Petroleum source beds on continental slopes and rises. AAPG Bull. 62:1584-
1606.
Dymond, J. and M. Lyle. 1985. Flux comparisons between sediments and sediment traps in the
eastern tropical Pacific: implications for atmospheric CO2 variations during the Pleistocene.
Limnol. Oceanogr. 30:699-712.
Emerson, S. 1985. Organic carbon preservation in marine sediments. In E. T. Sundquist and

W. S. Broecker, eds. The Carbon Cycle and Atmospheric CO2: Natural Variations Archaean to
Present, pp. 78-87. Geophys. monograph series, vol. 32, Am. Geophys. Union.
Emerson, S. and J. I. Hedges. 1988. Processes controlling the organic carbon content of open
ocean sediments. Paleoceanogr. 3:621-634.
Emerson, S., C. Stump, P. M. Grootes, M. Stuiver, G. W. Farwell, and F. H. Schmidt. 1987.

Organic carbon in surface deep-sea sediments: C-14 concentration. Nature 329:51-54.
Emery, K. O. and E. Uchupi. 1972. Western North Atlantic Ocean; topography, rocks, structure,
water, life and sediments. AAPG Memoir 17:1-532.
Eppley, R. W. and B. J. Peterson. 1979. Particulate organic matter flux and planktonic new
production in the deep ocean. Nature 282:677-680.
Eppley, R. W., W. G. Harrison, S. W. Chisholm, and E. Stewart. 1977. Particulate organic

matter in surface waters off Southern California and its relationship to phytoplankton. J. Mar.
Res. 35:671-696.
Exon, N. F., F. W. Haake, M. Hartmann, F. C. Kögler, P. J. Müller, and M. J. Whiticar. 1981.

Morphology, water characteristics and sedimentation in the silled Sulu Sea, southeast Asia.
Mar. Geol. 39:165-195.
Finney, B. P., M. W. Lyle, and G. R. Heath. 1988. Sedimentation at Manop Site H (Eastern
Equatorial Pacific) over the past 400,000 years: climatically induced redox variations and their
effects on transition metal cycling. Paleoceanogr. 3:169-190.
Foree, E. G. and P. L. McCarty. 1970. Anaerobic decomposition of algae. Environ. Sci.

Technol. 4:842-849.
Francois, R. 1988. A study on the regulation of the concentrations of some trace metals (Rb,
Sr, Zn, Pb, Cu, V, Cr, Ni, Mn and Mo) in Saanich Inlet sediments, British Columbia, Canada.
Mar. Geol. 83:285-308.
Froelich, P. N., G. P. Klinkhammer, M. L. Bender, N. A. Leudtke, G. R. Heath, D. Cullen, P.

Dauphin, D. Hammond, and B. Hartman. 1979. Early oxidation of organic matter in pelagic
sediments of the eastern equatorial Atlantic: suboxic diagenesis. Geochim. Cosmochim. Acta
43:1075-1090.
Glagoleva, M. A. 1961. On the geochemistry of the sediments of the Black Sea. In N. M.

Strakhov, ed. Recent Sediments of Seas and Oceans, pp. 448-476. Moscow: Izd. Akademie
Nauk SSSR.
Glenn, C. R. and M. A. Arthur 1984. Sedimentary and geochemical indicators of productivity

and oxygen contents in modern and ancient basins: The Holocene Black Sea as the "type"
anoxic basin. Chem. Geol. 48:325-354.
Gross, M. G. 1967. Organic carbon in surface sediment from the northeast Pacific Ocean. Int.
J. Oceanol. Limnol. 1:46-54.
Gucluer, S. M. and M. G. Gross. 1964. Recent marine sediments in Saanich Inlet, a stagnant
marine basin. Limnol. Oceanogr. 9:359-376.
Hamilton, S. E. and J. I. Hedges. 1988. The comparative geochemistries of lignins and

carbohydrates in an anoxic fjord. Geochim. Cosmochim. Acta 52:129-142.
Hamilton-Taylor, J. and N. B. Price. 1983. The geochemistry of iron and manganese in the
waters and sediments of Bolstadfjord, S.W. Norway. Est. and Coastal Shelf Sci. 17:1-19.
Hargrave, B. T. 1986. Vertical export of particulate matter from the upper ocean and the
relationship to organic carbon accumulation in sediments at different depths. In R. W. Eppley
and H. Ducklow, eds., U.S. GOFS Report 3, Woods Hole, Mass: U.S. Planning Office, pp. 121-
123.
Harvey, H. R., R. D. Fallon, and J. S. Patton. 1986. The effect of organic matter and oxygen on
the degradation of bacterial membrane lipids in marine sediments. Geochim. Cosmochim. Acta
50:795-804.
Hatcher, P. G., E. G. Spiker, N. M. Szeverenyi, and G. E. Maciel. 1983. Selective preservation

and origin of petroleum-forming aquatic kerogen. Nature 305:498-501.
Hay, B. J. 1988. Sediment accumulation in the central western Black Sea over the past 5100
years. Paleoceanogr. 3:491-508.
Hedges, J. I. 1981. Chemical indicators of organic carbon sources in rivers and estuaries. In G.
E. Likens, et al., eds., Flux of Organic Carbon by Rivers to the Oceans, pp. 109-141.
Washington, D.C.: U.S. Dept. of Energy.
Hedges, J. I. and D. C. Mann, 1979. The characterization of plant tissues by their lignin
oxidation products. Geochim. Cosmochim. Acta 43:1803-1807.
Hedges, J. I. and P. L. Parker. 1976. Land-derived organic matter in surface sediments from
the Gulf of Mexico. Geochim. Cosmochim. Acta 40:1019-1029.

Henrichs, S. M. and J. W. Farrington. 1987. Early diagenesis of amino acids and organic
matter in two coastal marine sediments. Geochim. Cosmochim. Acta 51:1-15.
Henrichs, S. M. and W. S. Reeburgh, 1987. Anaerobic mineralization of marine organic matter:

Rates and the role of anaerobic processes in the oceanic carbon economy. Geomicrobiol. J.
5:191-237.
Holser, W. T., M. Schindlowski, F. T. Mackenzie, and J. B. Maynard. 1988. Biogeochemical

cycles of carbon and sulfur. In C. B. Gregor, R. M. Garrels, F. T. Mackenzie, and J. B.
Maynard, eds., Chemical Cycles in the Evolution of the Earth, pp. 105-173 New York: Wiley.
Honjo, S., S. J. Manganini, and J. J. Cole. 1982. Sedimentation of biogenic matter in the deep
ocean. Deep Sea Res. 29:609-625.
Howell, M. W., R. Thunell, E. Tappa, D. Rio, and R. Sprovieri. 1988. Late Neogene laminated
and opal-rich facies from the Mediterranean region: geochemical evidence for mechanisms of
formation. Palaeogeogr. Palaeoclimatol. Palaeoecol. 64:265-286.
Ittekkot, V. 1988. Global trends in the nature of organic matter in river suspensions. Nature
332:436-438.
Ivankenkov, V. N. and A. G. Rozanov. 1961. Hydrogen sulfide contamination in the

intermediate layers of the Arabian Sea and the Bay of Bengal. Okeanologiya, no. 1, pp. 443-
449.
Jahnke, R. A. 1990. Early sediment diagenesis and recycling of biogenic debris in a deep
continental margin basin. J. Mar. Res. 48:413-436.
Jarvis, I. and N. C. Higgs. 1987. Trace element mobility during early diagenesis in distal
turbidites: Late Quaternary of the madeira Abyssal Plain, North Atlantic. In P. P. E. Weaver and
J. Thomson, eds., Geology and Geochemistry of Abyssal Plains, pp. 179-213. Geol. Soc.
special publ. 31.
Jewell, W. J. and P. L. McCarty, 1971. Aerobic decomposition of algae. Environ. Sci. Technol.
5:1023-1031.
Jorgensen, B. B. 1982. Mineralization of organic matter in the sea bed-the role of sulfate
reduction. Nature 296:643-645.
Koczy, F. 1950. Zur Sedimentation und Geochemie im Aequatorialen Atlantischen Ozean.

Medd. Oceanog. Inst. Goteborg, no. 17.
Kristensen, E. and T. H. Blackburn. 1987. The fate of organic carbon and nitrogen in
experimental marine sediment systems: influence of bioturbation and anoxia. J. Mar. Res.
45:231-257.
Kuenen, P. H. 1942. Geological results, Repts Snellius Expedition, vol. 5, part 3, section 1.

Leiden: E. J. Brill.
Kump, L. R. 1988. Terrestrial feedback in atmospheric oxygen regulation by fire and

phosphorus. Nature 335:152-154.
Lee, C. and S. G. Wakeham. 1989. Organic matter in seawater: biogeochemical processes. In

J. P. Riley and R. Chester, eds., Chemical Oceanography, 9:1-51. New York: Academic.
Lyle, M., D. M. Murray, B. P. Finney, J. Dymond, J. M. Robbins, and K. Brooksforce. 1988. The
record of Late Pleistocene biogenic sedimentation in the Eastern Tropical Pacific Ocean.
Paleoceanogr. 3:39-59.
Mackin, J. E. and K. T. Swider. 1989. Organic matter decomposition pathways and oxygen
consumption in coastal marine sediments. J. Mar. Res. 47:681-716.
Manheim, F. T. and K. M. Chan. 1974. Interstitial waters of Black Sea sediments: new data and
review. In E. T. Degens and D. A. Ross, eds., The Black Sea-Geology, Chemistry and Biology,
pp. 155-180. AAPG Memoir 20.
Marchig, V. 1972. Zur Geochimie rezenter Sedimente des Indisches Ozeans. Meteor
Forschungs Ergebnisse C, 11:1-104.
Martens, C. S. and J. V. Klump. 1984. Biogeochemical cycling in an organic-rich coastal

marine basin. 4. An organic carbon budget for sediments dominated by sulfate reduction and
methanogenesis. Geochim. Cosmochim. Acta 48:1987-2004.
Martin, W. R. and F. L. Sayles (in press). Seafloor diagenetic fluxes. In Global Surficial
Geofluxes. Washington, D. C.: Geophysics Study Committee, Natl. Acad. Sci.
McManus, D. A. 1972. Bottom topography and sediment texture near the Columbia River
mouth. In A. T. Pruter, and D. L. Alverson, eds., The Columbia River Estuary and Adjacent
Ocean waters, pp. 241-253. Seattle: Univ. Washington Press.
Meybeck, M. 1982. Carbon, nitrogen and phosphorus transport by world rivers. Am J. Sci.
282:401-450.
Miller, K. G. and G. P. Lohmann. 1982. Environmental distribution of Recent benthonic

foraminifera on the northeast United States continental slope. Geol. Soc. Am. Bull. 93:200-206.
Morris, R. J., M. J. McCartney, and P. P. E. Weaver. 1984. Sapropelic deposits in a sediment

from the Guinea Basin, South Atlantic. Nature 309:611-614.
Müller, P. J. 1977. C/N ratios in Pacific deep-sea sediments: Effect of inorganic ammonium and
organic nitrogen compounds sorbed by clays. Geochim. Cosmochim. Acta 41:765-776.
Müller, P. J. and A. Mangini. 1980. Organic carbon decomposition rates in sediments of the
Pacific manganese nodule belt dated by 230Th and 231Pa. Earth Planetary Sci. Lett. 51:94-114.

Müller, P. J. and E. Suess. 1979. Productivity, sedimentation rate and sedimentary organic
matter in the oceans. I. Organic carbon preservation. Deep Sea Res. 26:1347-1362.
Müller, P. J., H. Erlenkeuser, and R. von Grafenstein. 1983. Glacial-interglacial cycles in ocean
productivity inferred from organic carbon contents in eastern North Atlantic sediment cores. In
J. Thiede and E. Suess, eds., Coastal Upwelling, part B, pp. 365-398. New York: Plenum.
Murray, J. W., V. Grundmanis, and W. M. Smethie. 1978. Interstitial water chemistry in the
sediments of Saanich Inlet. Geochim. Cosmochim. Acta 42:1011-1026.
Orr, W. L. and A. G. Gaines. 1974. Observations on the rate of sulfate reduction and organic
matter oxidation in the bottom waters of an estuarine basin: the upper basin of the
Pettaquamscutt River (Rhode Island). In B. Tissot and F. Bienner, eds., Advances in Organic
Geochemistry, pp. 791-812. Proc. 6th Int. Org. Geochem. Congress. Paris: Technip.
Otsuki, A. and T. Hanya. 1972a. Production of dissolved organic matter from dead algal cells. I.
Aerobic microbial decomposition. Limnol. Oceanogr. 7:248-257.
Otsuki, A. and T. Hanya. 1972b. Production of dissolved organic matter from dead algal cells.
II. Anaerobic microbial decomposition. Limnol. Oceanogr. 7:258-264.

Oxford: Pergamon Press.
Pedersen, T. F. 1983. Increased productivity in the eastern equatorial Pacific during the last
glacial maximum (19,000 to 14,000 yr b.p.). Geology 11:16-19.
Pedersen, T. F. 1988. Late Pleistocene carbon enrichments in the Panama Basin: frequency
and causes. Chem. Geol. 70:111 (abstract).
Pedersen, T. F. and S. E. Calvert. 1990. Anoxia versus productivity: What controls the
formation of organic carbon-rich sediments and sedimentary rocks? AAPG Bull. 74:456-466.
Pedersen, T. F., M. Pickering, J. S. Vogel, J. N. Southon, and D. E. Nelson. 1988. The

response of benthic foraminifera to productivity cycles in the eastern equatorial Pacific: faunal
and geochemical constraints on glacial bottom water oxygen levels. Paleoceanogr. 3:157-168.
Pedersen, T. F., R. D. Waters, and R. W. Macdonald. 1989. On the natural enrichment of

cadmium and molybdenum in the sediments of Ucluelet Inlet, British Columbia. Sci. Total
Environ. 79:125-139.
Prahl, F. G. and L. A. Muehlhausen. 1989. Lipid biomarkers as geochemical tools for

paleoceanographic study. In W. H. Berger, V. S. Smetacek, and G. Wefer, eds., Productivity of
the Ocean: Present and Past, pp. 271-289. New York: Wiley.
Prahl, F. G., L. A. Muehlhausen, and M. Lyle. 1989. An organic geochemical assessment of

oceanographic conditions at MANOP Site C over the past 26,000 years. Paleoceanogr. 4:495-
510.

Premuzic, E. T., C. M. Benkovitz, J. S. Gaffrey, and J. J. Walsh. 1982. The nature and
distribution of organic matter in the surface sediments of the world oceans and seas. Organic
Geochem. 4:63-77.
Reimers, C. E. and E. Suess. 1983. The partitioning of organic carbon fluxes and sedimentary
organic matter decomposition rates in the ocean. Mar. Chem. 13:903-910.
Richards, F. A. 1965. Anoxic basins and fjords. In J. P. Riley and G. Skirrow, eds., Chemical
Oceanography 1:611-645. New York: Academic Press.
Richards, F. A., and A. C. Redfield. 1954. A correlation between the oxygen content of sea
water and the organic content of marine sediments. Deep Sea Res. 1:279-281.
Richards, F. A., and R. F. Vaccaro. 1956. The Cariaco Trench, an anaerobic basin in the
Caribbean. Deep Sea Res. 3:214-228.
Romankevich, E. A. 1977. Geochemistry of Organic Matter in the Ocean, Moscow: Nauka.
Rosenfeld, J. K. 1981. Nitrogen diagenesis in Long Island Sound sediments. Am. J. Sci.
281:436-462.
Ross, D. A. and E. T. Degens. Recent sediments of the Black Sea. In E. T. Degens, and D. A.
Ross, eds., The Black Sea-Geology, Chemistry and Biology, pp. 183-199. AAPG Memoir 20.
Rowe, G. T., S. Smith, P. Falkowski, T. Whitledge, R. Theroux, W. E. Phoel, and H. Ducklow.

1986. Do continental shelves export organic matter? Nature 324:559-561.
Ryther, J. H. 1963. Geographic variations in productivity. In M. N. Hill, ed., The Sea 2:347-380.
New York: Wiley.
Sackett, W. M. 1964. The depositional history and isotopic organic carbon composition of
marine sediments. Mar. Geol. 2:173-185.
Sackett, W. M. and R. R. Thompson. 1963. Isotopic organic carbon composition of recent

continental derived clastic sediment of eastern Gulf of Mexico. AAPG Bull. 47:525-531.
Sarmiento, J. L., T. Herbert, and J. R. Toggweiler. 1988. Causes of anoxia in the world ocean.
Global Biogeochem. Cycles 2:115-128.
Sarnthein, M., J. Thiede, U. Pflaumann, H. Erlenkeuser, D. Fütterer, B. Koopman, H. Lange,

and E. Seibold. 1982. Atmospheric and oceanic circulation patterns off Northwest Africa during
the past 25 million years. In U. von Rad, K. Hinz, M. Sarnthein, and E. Seibold eds., Geology of
the Northwest African Continental Margin, pp. 545-604. Berlin: Springer Verlag.
13
Sayles, F. L. and W. B. Curry. 1988. C, TCO2 and the metabolism of organic carbon in
deep sea sediments. Geochim. Cosmochim. Acta 52:2963-2978.

Schlanger, S. O., and H. C. Jenkyns. 1976. Cretaceous oceanic anoxic events: causes and
consequences. Geologie en Mijnbouw 55:179-184.
Shen, G. T., R. B. Dunbar, M. W. Colgan, and P. W. Glynn. 1988. El Niño and Little Ice Age
effects on upwelling in the eastern tropical Pacific. Chem. Geol. 70:198 (abstract).
Shimkus, K. M., and E. S. Trimonis. 1974. Modern sedimentation in the Black Sea, In E. T.
Degens and D. A. Ross, eds., The Black Sea-Geology, Chemistry and Biology, pp. 249-278.
AAPG Memoir 20.
Shimmield, G. B., N. B. Price, and T. F. Pedersen. 1990. The influence of hydrography,

bathymetry and productivity on sediment type and composition on the Oman Margin and in the
northwest Arabian Sea. In A. H. F. Robertson, ed., The Geology and Tectonics of the Oman
Region, pp. 759-769. Geol. Soc. spec. publ.
Skei, J. 1988. Framvaren-environmental settling. Mar. Chem. 23:209-218.
Skei, J., D. H. Loring, and R. T. T. Rantala. 1988. Partitioning and enrichment of trace
elements in a sediment core from Framvaren, south Norway. Mar. Chem. 23:269-281.
Slater, R. D., and P. Kroopnick. 1984. Controls on dissolved oxygen distribution and organic
carbon deposition in the Arabian Sea., In B. U. Haq and J. D. Milliman, eds. Marine Geology
and Oceanography of Arabian Sea and Coastal Pakistan, pp. 305-313. New York: Van
Nostrand Reinhold.
Smith, J. D. and T. S. Hopkins. 1972. Sediment transport on the continental shelf off of
Washington and Oregon in light of recent current measurements. In D. J. P. Swift, D. B.
Duane, and O. H. Pilkey, eds., Shelf Sediment Transport: Process and Pattern, pp. 143-180.
Stroudsburg, Pa: Dowdon, Hutchinson and Ross.
Sorokin, Y. I. 1983. The Black Sea. In B. H. Ketchum, ed., Ecosystems of the World, vol. 26.
Esturaries and Enclosed Seas. Amsterdam: Elsevier. pp. 253-292.
Soutar, A., S. A. Kling, P. A. Crill, E. Duffrin, and K. W. Bruland. 1977. Monitoring the marine
environment through sedimentation. Nature 277:136-139.
Stow, D. A. V., and W. E. Dean. 1984. Middle Cretaceous black shales at Site 530 in the
southeastern Angola Basin: Initial Reports. DSDP 75:805-818.
Strakhov, N. M. 1962. Principles of Lithogenesis. Edinburgh: Oliver and Boyd.
Strom, K. M. 1936. Land-locked waters: Skrifter Norsk. Viden-Akademie Oslo 1:1-84.
Suess E. 1980. Particulate organic carbon flux in the oceans-surface productivity and oxygen
utilization. Nature 288:260-263.
Tenore, K. R. 1987. Nitrogen in benthic food chains. In T. H. Blackburn, and J. Sorensen, eds.,

Nitrogen Cycling in Coastal Marine Environments. New York: Wiley.
Thiede, J. and T. J. van Andel, 1977. The paleoenvironment of anaerobic sediments in the late
Mesozoic south Atlantic Ocean. Earth Planetary Sci. Lett. 33:301-309.
Thomson, J., T. R. S. Wilson, F. Culkin, and D. J. Hydes. 1984. Non-steady state diagenetic
record in eastern equatorial Atlantic sediments. Earth Planetary Sci. Lett. 71:23-30.
Thomson, J., S. Colley, N. C. Higgs, D. J. Hydes, T. R. S. Wilson, and J. Sorensen. 1987.

Geochemical oxidation fronts in N.E. Atlantic distal turbidites and their effects in the
sedimentary record. In P. P. E. Weaver, and J. Thomson eds., Geology and Geochemistry of
Abyssal Plains, Geol. Soc. spec. publ. 31:167-177.
Trask, P. D. 1953. Chemical studies of the sediment of the western Gulf of Mexico. Papers in
Physical Oceanography and Meteorology 12:49-120. Boston: M.I.T.
Van Riel, P. M. 1943. Introductory remarks and oxygen content. Reports of the Snellius
Expedition, vol. 2, part 5, pp. 1-77. Leiden. E. J. Brill.
Vercoutere, T. L., H. T. Mullins, K. McDougall, and J. B. Thompson. 1987. Sedimentation

across the central California oxygen minimum zone: An alternative coastal upwelling
sequence. J. Sedim. Petrol. 57:709-722.
Von Stackelberg, U. V. 1972. Faziesverteilung in Sedimenten des indisch-pakistanischen

Kontinentalrandes. Meteor Forschungs Ergebnisse C, 9:1-73.
Wallace, H. E., J. Thomson, T. R. S. Wilson, P. P. E. Weaver, N. C. Higgs, and D. J. Hydes.

1988. Active diagenetic formation of metal-rich layers in N.E. Atlantic sediments. Geochim.
Walsh, J. J. 1989. How much shelf production reaches the deep sea? In W. H. Berger, V. S.
Smetacek, and G. Nefer, eds., Productivity of the Ocean: Present and Past, pp. 175-191. New
York: Wiley.
Walsh, J. J., G. T. Rowe, R. L. Iverson, and C. P. McRoy. 1981. Biological export of shelf
carbon is a sink for the global CO2 cycle. Nature 291:196-201.
Walsh, J. J., E. T. Premuzic, J. S. Gaffney, G. T. Rowe, G. Harbottle, R. W. Stoenner, W. L.

Balsam, P. R. Betzer, and S. A. Macko. 1985. Organic storage of CO2 on the continental slope
of the mid-Atlantic Bight, the southeastern Bering Sea and the Peru coast. Deep Sea Res.
32:853-883.
sulfate reduction: The G model tested. Limnol. Oceanogr. 29:236-249.
Wilson, T. R. S., J. Thomson, S. Colley, D. J. Hydes, N. C. Higgs, and J. Sorensen. 1985.

Early organic diagenesis: the significance of progressive subsurface oxidation fronts in pelagic

Wilson, T. R. S., J. Thomson, D. J. Hydes, S. Colley, F. Culkin, and J. Sorensen. 1986.

Oxidation fronts in pelagic sediments: diagenetic formation of metal-rich layers. Science
232:972-975.
Wiseman, J. D. H. and H. Bennett. 1940. The distribution of organic carbon and nitrogen in the
sediments from the Arabian Sea. Repts. John Murray Expedition 3(4):193-221.
Woolnough, W. G. 1937. Sedimentation in barred basins and source rocks of oil. Bull. AAPG
21:1101-1157.
Zahn, R., K. Winn, and M. Sarnthein. 1986. Benthic foraminiferal 13C and accumulation rates
of organic carbon: Uvigerina peregrina group and Cibicidoides wuellerstorfi. Paleoceanogr.
1:27-42.
Zeikus, J. G. 1980. Fate of lignin and related aromatic substances in anaerobic environments.
In T. K. Kirk, ed., Lignin Biodegradation 1:101-109. Boca Raton, Fla., CRC Press.


12. Early-Stage Incorporation of Sulfur Into Protokerogens

and Possible Kerogen Precursors
The formation of organosulfur compounds in the early stages of diagenesis has been studied by
reaction of H2S with recently deposited cyanobacterial mats from Baja, California, and artificial
melanoidin polymers under mild laboratory reaction conditions (40°C, 2 to 30 days). Reaction
products have been characterized by pyrolysis-GC using both a FPD (flame photometric detector)
and a FID (flame ionization detector) as well as pyrolysis-GC/MS (gas chromatography and mass
spectrometry).
Incorporation of sulfur from H2S into recently deposited cyanobacterial mats occurs relatively
readily. Humic acid fractions were found to incorporate sulfur more rapidly than protokerogens, but
no sulfur was incorporated into the corresponding lipid fraction. Variations in the pyrolysis products
of the melanoidins (glucose + lysine) after H2S treatment were similar to those observed in the algal
mat experiments. Incorporation of H2S sulfur into sedimentary organic matter in the early stages of
diagenesis may be an important controlling factor governing the formation of melanoidin-like
geopolymers in recent sediments. Geochemical characteristics of kerogens in carbonate rocks may
be rationalized by rapid reaction between H2S and reducing sugars in the early stages of
diagenesis.
The study of organosulfur compounds in geological samples, of all ages, has become a significant
and important area of geochemical research. Many papers in the literature cite lists of organosulfur
compounds identified in crude oils (for example, Hughes 1984; Poirier and Smiley 1984; Payzant,
Montgomery, and Strausz 1986; Arpino, Ignatiadis, and Gael De Rycke 1987; Sinninghe Damste et
al. 1987), liquids produced by liquefaction of oil shales (for example, Willey et al. 1981; Wong,
Crawford, and Burnham 1984; Braekman-Danheux 1985), or coals (for example, Kong et al. 1982;
Wasterman et al. 1983; Herod and Smith 1985). Their origins or mechanisms of formation have also
been discussed (for example; Casagrande et al. 1979; Whelan, Hunt, and Berman 1980; Brassell et
al. 1986; Schmid, Connan, and Albrecht 1987; Sinninghe Damste et al. 1988, 1989 a and b).
The abundance of organically bound sulfur in living biomass is very low relative to that present in
oils, source rocks, and sediments. It is unlikely, therefore, that biochemical organic sulfur
compounds can account for all the organically bound sulfur in fossil organic matter. In addition,
structures of biochemical sulfur compounds are quite different from those of geochemical origin.
It is now generally accepted that reduced sulfur species, hydrogen sulfide, elemental sulfur and/or
polysulfides are major donors of sulfur in reactions with organic molecules (Orr 1978). Several
studies have indicated an enrichment of sulfur in macromolecular sedimentary organic matter
during early stages of diagenesis or simulation experiments (Nissenbaum and Kaplan 1972;
Casagrande and Ng 1979; Casagrande et al. 1979; Aizenshtat et al. 1983; Francois 1987). Further
evidence for the early stage incorporation of abiogenic sulfur comes from the identification of

hopanoids and isoprenoids containing thiophene rings in relatively immature sediments (Valisolalao
et al. 1984; Brassell et al. 1986). Various organosulfur compounds with biologically abundant
carbon skeletons such as steroids and isoprenoids have also been identified in sedimentary rocks
(Sinninghe Damste et al. 1988). Possible mechanisms for incorporation of the sulfur include
microbial activity at the time of deposition, chemical reactions with evaporites, or other sulfur-
containing species during diagenesis, or sulfurization of crude oils in the reservoirs.
H2S is generally present in recent marine sediments deposited under anoxic conditions. Reactions
between H2S and sedimentary organic matter, under mild reaction conditions, are considered to be
important pathways for the formation of sedimentary organosulfur compounds such as lower
molecular weight thiols (Vairavamurthy and Mopper 1987) and sulfur-containing structures in humic
acids and kerogens (Sinninghe Damste et al. 1989b; Suzuki and Philp 1989). The reaction between
H2S and reducing sugars, possible precursors of melanoidin-like geopolymers, has been observed
to occur fairly readily under laboratory conditions (Moers et al. 1988; Suzuki and Philp 1989).
The origin of some organosulfur compounds in geological samples is probably closely related to the
environmental conditions present at the time of deposition of the organic matter. For example
enrichment of organosulfur compounds appears to correlate with a marine-carbonate type of
depositional environment. Examples of such depositional environments are those responsible for
the formation of oils from the Middle East and Lake Maracaibo (Gransch and Posthuma 1974).
Other examples include hypersaline depositional environments such as those responsible for the
formation of Rozel Point Oil (Sinninghe Damste et al. 1987) and various oils from China (Philp, Li,
and Lewis 1989). In a closed environment the oxygen supply will be rapidly utilized and sulfate-
reducing bacteria will extract the sulfate from saline water and reduce it to H2S, which in turn acts
as an alternative source of sulfur. In carbonate muds where the availability of iron, for the formation
of iron sulfides, is limited, various forms of sulfur remain free and hence available for incorporation
into residual organic matter during diagenesis.
Our recent experiments have focused on the formation of organosulfur compounds in the early
stages of diagenesis. Laboratory simulation experiments have been performed to investigate
possible incorporation of sulfur from H2S or elemental sulfur into recently deposited cyanobacterial
mats from Baja California under mild reaction conditions (40°C, 2 to 30 days) and artificial
melanoidin polymers (D-glucose + L-lysine). The newly formed sulfur-containing structures, or
reaction products, from the simulation experiments have been characterized by pyrolysis-GC (gas
chromatography) with FPD (flame photometric detector)/FID (flame ionization detector) and
pyrolysis-GC/MS (gas chromatography and mass spectrometry). The results observed in these
studies provide a basis for interpretation of possible mechanisms for sulfur incorporation into
naturally occurring complex organic geopolymers in the early stages of diagenesis.
Experimental Samples
The samples of cyanobacterial mat were obtained from the bottom surface of Exportadora de Sal, S.
A. concentration pond, Baja California, in 1985 with a steel spatula and aluminum foil that had been
precleaned in an annealing oven. This salt pond is situated in a hypersaline coastal pond and the
salinity of sampling location ranges from 75 to 95 per mil. The cyanobacterial mat was frozen at -20°
C prior to analysis.
A melanoidin polymer was prepared from an aqueous solution consisting of 0.30 mole D-glucose

and 0.15 mole L-lysine at pH = 7. Aliquots (8 mL) of the solution were sealed in Pyrex tubes and
heated at 80°C for 20 days under N2 at 80°C for 30 days in the presence of H2S. The brown-
colored suspension that precipitated from the D-glucose and L-lysine mixture under N2 was
recovered by centrifugation and defined as P-N2. The mixture of D-glucose and L-lysine reacted
under H2S conditions (80°C, 30 days) showed no major visual change apart from a slight change in
color to light brown. The mixture of D-glucose and L-lysine reacted under H2S conditions (80°C, 30
days) was dried at 40°C for 2 days under air. This dried residue was recovered and defined as DR-
H2S. A part of the original aqueous solution of D-glucose and L-lysine (2:1) was similarly dried at
40°C for 2 days under air. The dried original glucose and lysine mixture was defined as DR-ORI.
Reaction of Cyanobacterial Mats with H2S and Elemental S
Cyanobacterial mats (50 mg), or melanoidins, were sealed in Pyrex tubes (ca. 50 mL) with Teflon-
lined screw caps in the presence of ca. 45 mL of H2S (99.99% purity, Fisher) both with and without
distilled water (2 mL). The reaction tubes were kept at 40 1°C for 2 to 30 days at a pH of 7. The
samples of cyanobacterial mats were also reacted with water (2 mL), supersaturated with elemental
sulfur at 40°C for 30 days under a N2 atmosphere.
The residues from the experiments between the cyanobacterial mats, melanoidins, and H2S were
dried and extracted with methylene chloride/methanol. Humic acids and fulvic acids were extracted
by 0.1N NaOH. Humic acids were isolated by acidification of the alkaline extracts. The solvent
extracts, humic acids, and extracted residues from the cyanobacterial mats, and the melanoidins,
before and after reaction with H2S, were subsequently characterized by pyrolysis-GC (py-GC) and
pyrolysis-GC/MS (py-GC/MS).
Pyrolysis GC and Pyrolysis GC/MS
The py-GC characterization was performed with a Varian 3300 GC with a Chemical Data System
(CDS) pyroprobe system under conditions similar to those previously described by Philp and Bakel
(1988). The py-GC/MS analyses were performed with the CDS pyrolysis unit and Finnigan Triple
Stage Quadrupole mass spectrometer (TSQ 70) system equipped with a Varian 3400 GC. The
samples were pyrolyzed with the coil probe at a temperature setting of 800°C for 20 seconds and
analyzed on a DB-5 (30 m x 0.25 mm i.d.) fused silica column programmed from -25°C to 300°C at
4°C/min. The TSQ 70 system was operated with a filament current of 200 A, ionization energy of
70 eV, ion source temperature of 200°C, and electron multiplier voltage of 1600 V. GC conditions
were similar to those described above in Philp and Bakel (1988).
Variations in Pyrolysates from Recently Deposited Cyanobacterial Mats After Reaction with H2S
H2S in depositional environments is generally produced by sulfate-reducing bacteria in the upper
part of the anoxic bottom sediments. Early stage sulfur incorporation by H2S into organic debris is
thought to occur under very low reaction temperatures and in the presence of water. Hence,
laboratory simulation experiments were performed under mild temperature conditions to prevent the
occurrence of side reactions. The simulation experiments in this work were performed at 40°C for
varying periods of time (two to thirty days).

Chromatograms illustrating the pyrolysis products (FID traces) produced from the solvent-extracted
(SE) mat residues (unaltered and altered by H2S or elemental S at 40°C for 0 to 30 days) are
shown in figure 12.1 and products are listed in table 12.1. Changes in organosulfur compounds
distributions (FPD traces) for the same residue are shown in figure 12.2.
The FID chromatogram of the original SE mat residue (figure 12.1) shows toluene to be the most
abundant pyrolysis product (other than gaseous products), along with a relatively high abundance of
nitriles and pyridine, which are proposed to be derived from proteins or nucleic acids (Simmonds,
Shulman, and Stembridge 1969; Bayer and Morgan 1984). A possible explanation for the presence
of furaldehydes and methylfuraldehydes is that they are derived from carbohydrates in the original
mat (Simmonds, Shulman, and Stembridge 1969; Bayer and Morgan, 1984). An alternative and
perhaps more probable explanation is that the furan compounds are derived from the uronic acid-
rich sheath materials and slimes from algae and (photosynthetic) bacteria. Furaldehyde formation
by pyrolysis from a uronic acid polymer (pectins) has been reported by Aries, Gutteridge, and Laurie
(1988). The FID and FPD chromatograms show that the original algal mat is poor in organosulfur-
containing structures that can be released by pyrolysis but rich in unstable organic structures
potentially capable of acting as receptors for sulfur. The FID chromatogram of the pyrolysis
products from the H2S-altered SE algal mat (40°C, 30 days; fig. 12.1-vi) was markedly different
from the original unaltered sample (figure .12.1-ii). In particular, the relative abundance of
thiophene, methylthiophenes, 2-methylpyridine, benzofuran, methylbenzofuran, benzene, and
methylbenzenes increased markedly. On the other hand, the relative abundance of the
furaldehydes, methylfuraldehydes, phenol, and methylphenols decreased. The distributions of
alkanes, alkenes, indenes, indoles, nitroles, and pyridine do not appear to change significantly with
increasing time. A comparison of the top two chromatograms (fig. 12.1-i with 12.1-ii and 12.2-i with
12.2-ii) shows that very little thermal decomposition and alteration by H2O or elemental S or organic
matter has occurred under these reaction conditions (40°C, 2 to 30 days, hydrous).
Pyrolysis of the SE original mat sample itself produced few organosulfur compounds (figure 12.2-ii),
the most intense peak being identified as a thiazole. Trace amounts of carbon disulfide, thiophene,
and methylthiophenes were also present in the pyrolysate of the original SE algal mat. Thiophene
and methylthiophenes are known to be characteristic pyrolysis products of taurine and cystine,
respectively (Merritt and Robertson 1967). Thiazole may be derived from sulfur-containing amino
acids and polypeptides.
The FPD traces obtained from pyrolysis of the H2S-altered algal mats also changed markedly with
increasing reaction time (figure .12.2). Several organosulfur compounds appeared to be formed by
reaction with H2S and the cyanobacterial mat under both hydrous and anhydrous conditions.
Carbon disulfide, thiophene, and C1-, C2-, and C3-thiophenes were identified as the organosulfur
pyrolysis products in the FPD chromatograms that were formed under these reaction conditions. In
comparison with the pyrolysates compositions from kerogens in older sediments (Philp and Bakel
1988; Sinninghe Damste et al. 1989a), the pyrolysates from H2S-altered cyanobacterial mats are
characterized by the fact that 3-methylthiophene is more abundant than 2-methylthiophene and that
thiophene is more abundant than methylthiophenes. With longer reaction times there is a marked
increase of relative abundance of 3-methylthiophene (figures 12.1 and 12.2). In the FID
chromatogram (figure 12.1-vi) of H2S-altered extracted mat (40°C, 30 days), 3-methylthiophene
(peak 26) is a major component relative to the hydrocarbons.
Whelan, Hunt, and Barnan (1980) proposed a mechanism for the formation of 3-methylthiophene

from isoprenes that could be derived by breakdown of the higher molecular weight sedimentary
organic matter. It was proposed that the reaction between 2-methyl-5-hydroxylbutanal and H2S
could form 3-methylthiophene in the very early stages of diagenesis. If thiophene rings containing
isoprenoid substitutes at the C-3 position can be formed in humic acids or protokerogens under mild
reaction conditions, as suggested by Sinninghe Damste et al. (1989b), then the direct cleavage of
such units could form 3-methylthiophene. However, the relative abundance of thiophene in the
hydrolysates of the present samples suggests 3-methylthiophene. The relative abundance of
thiophene suggests, however, that the H2S is mainly incorporated into the organic units poor in
alkyl side chains.
Pyrolysates of the humic acid fractions isolated from algal mats are characterized by a high
abundance of 2-methylpyridine (peak 31) relative to toluene (peak 24) (figure 12.3). The 2-
methylpyridine was not detected in the FID chromatogram of the pyrolysates of the organic extract
or the alkaline-insoluble residue (protokerogen). The organosulfur compounds are markedly more
abundant in the pyrolysates from the humic acid fraction than in the pyrolysates from the
protokerogen fraction of the same H2S-altered mat. This observation indicates that sulfur
incorporation into humic acids probably occurs faster than it does into the protokerogen fraction. In
addition, it was observed that 3-methylthiophene (peak 26) is twice as abundant as 2-
methylthiophene (peak 25) in pyrolysates of H2S-altered humic acid (figure 12.3). No sulfur
compounds, except for elemental sulfur, were detected in the FPD chromatogram of pyrolysates
from the solvent extracts of the cyanobacterial mats.
Casagrande et al. (1979) previously reported that H2S was incorporated into wet peat at room
temperature. The humic acid and humin (protokerogen in this paper) fraction from the peat showed
the largest degree of organic sulfur incorporation. Our results are consistent with this observation.
Douglas and Mair (1964) suggested that elemental sulfur would react with steroids and terpenoids
to form organosulfur compounds in crude oil. Casagrande and Ng (1979) demonstrated that
elemental sulfur could be incorporated into the humic acid fraction of coal by refluxing under
chloroform for 6 to 48 hours. Significant incorporation of sulfur in the cyanobacterial mat material
was not observed in our studies, which were performed under mild conditions. The elemental sulfur
incorporation experiments of Douglas and Mair (1964) and Casagrande and Ng (1979) were
performed at relatively high temperatures (135° to 150°C and refluxing chloroform, respectively).
The results from these different experiments may vary owing to experimental conditions. However,
it is evident that sulfur incorporation by H2S into recent cyanobacterial mats, under mild reaction
conditions, occurs much faster than the rate at which elemental sulfur is incorporated in the
cyanobacterial mats.
Sinninghe Damste et al. (1989b) have suggested that sulfur incorporation by H2S into precursors
with double bonds will lead to formation of organic sulfur compounds. The reaction of unsaturated
lipids with H2S is favorable for the geochemical formation of thiophene-ring-containing compounds
with biologically abundant carbon skeletons. However, significant sulfur incorporation into the lipid
fraction was not observed in the present study. Casagrande et al. (1979) have also shown that the
reaction of H2S with peat extracts did not proceed significantly under ambient temperature. Higher
reaction temperatures or longer reaction times may be necessary for the reaction between
unsaturated lipids and H2S in order for the formation of thiophene rings to occur.
Sulfur Incorporation from H2S into the Cyanobacterial Mats

From the experiments described above it has been demonstrated that sulfur derived from H2S
could be incorporated into algal mat material under relatively mild reaction conditions. Sulfur derived
from H2S was incorporated into the humic acid fraction much faster than into the kerogen, or
protokerogen fractions, and no sulfur incorporation at all was observed in the free lipid fractions.
These observations are in good agreement with the fact that D-glucose, a major humic acid
precursor (Nissenbaum and Kaplan 1972; Hoering 1973), is reactive with H2S under mild reaction
conditions (Moers et al. 1988; Suzuki and Philp 1989).
It is proposed that the first stage of sulfur incorporation via the formation of the organosulfur
compounds in these experiments parallels a marked decrease in the relative concentration of
furaldehyde (figure .12.4). Furans do not decrease in relative abundance as much as the
furaldehydes, but both furaldehyde and furan pyrolysis products are probably derived predominantly
from uronic acid-rich sheath materials from the algae and bacteria. The selective decrease of
furaldehydes in pyrolysates from the algal mats is considered to be related to the sulfur
incorporation mechanisms described later.
The second stage of sulfur incorporation (after 8 days' reaction time) is signified by changes in the
composition of the pyrolysate as illustrated by an increase of the thiophene, 3-methylthiophene, and
2-methylpyridine concentrations and a decrease of pyridine, toluene, and phenol concentrations
(figure 12.1). These changes cannot be interpreted only by the reaction between H2S and
carbohydrates, since the relative abundance of furaldehydes and furans in the pyrolysis products
does not change significantly, and moreover, carbohydrates do not contain nitrogen.
Reactions Between H2S and Melanoidins
Melanoidins, a carbohydrate-amino acid complex polymer, have been considered as an important

precursor of various geopolymers such as fulvic acids, humic acids, and protokerogen (Nissenbaum
and Kaplan 1972; Hoering and Hare 1973). This nitrogen-containing artificial polymer probably
contains carbohydrate structures that are also reactive with H2S. Reducing sugars and amino acids
will react readily under mild conditions to form complex polymers called melanoidins, which possess
chemical characteristics similar to natural humic substances in soils, natural waters, and recent
sediments (Hoering 1973).
In the present study significant amounts of dark-brown-colored suspensions appeared from the D-
glucose and L-lysine mixture under N2 after 20 days. However, the mixture of D-glucose and L-
lysine mixture under H2S showed no major visual change after 20 days. This mixture of D-glucose
and L-lysine was kept in a Pyrex tube at 80°C for an additional 10 days but again showed no major
visual change apart from a slight darkening in color to light brown. No changes were observed
under acidic or basic conditions, indicating that the formation of melanoidins from D-glucose and L-
lysine proceeds very slowly, or not at all, in the presence of H2S. The dried residues from the
various melanoidin experiments were characterized by py-GC and selected chromatograms shown
in figure 12.5 (peak identifications listed in table 12.2). All of the pyrolysates obtained in the
presence of N2 (DR-ORI and DRR-N2) apart from the precipitate (P-N2) (i.e., absence of H2S) are
characterized by the high abundance of 2-methylfuran (peak 2), 2,5-dimethylfuran (peak 6), 2-
furaldehyde (peak 13), and 5-methyl-2-furaldehyde (peak 20). It has previously been observed that
furans and furaldehydes can be derived from the D-glucose structure (Simmonds, Shulman, and
Stembridge 1969; Bayer and Morgan 1984), and this is their probable origin in these samples as
well. The precipitate formed under N2 (P-N2 in figure 12.5d) is relatively rich in aromatic

compounds such as toluene (peak 8), xylenes (peaks 16 and 17), and phenol (peak 21), suggesting
that this precipitate has a more highly condensed structure consistent with its insolubility in both
organic solvents and water. The pyrolysates from the dried residue formed under H2S (figure 12.5b)
are characterized by various organosulfur compounds such as thiophene (peak 4 in figure 12.5 and
peak 28 in figure 12.6) and alkylthiophenes (peaks 9 and 15 in figure .12.5 and peaks, 4s, 5s, and
7s in figure 12.6). This indicates that a significant amount of sulfur from H2S has been incorporated
into the organic phase of the mixed glucose and lysine solutions to form organosulfur compounds
during heating (80°C, 30 days). The pyrolysate from the residue formed in the presence of H2S is
very poor in furaldehydes (peaks 13 and 20 in figure 12.5b), compared with the pyrolysates from the
original dried residue (figure 12.5a).
Aldehyde functional groups of furaldehydes in pyrolysates are thought to be derived from carbonyls
at C-1 of glucosic compounds (Byrne, Gardiner, and Holmes 1966; Shafizadeh 1968). This
observation can be interpreted on the basis of incorporation of sulfur from H2S mainly occurring by
the reaction between H2S and reactive carbonyls present in the glucose structures. Lalonde,
Ferrara, and Hayes (1987) have previously shown that H2S is reactive with activated carbonyl
groups.
The formation of melanoidin from reducing sugars and basic amino acids generally occurs readily.
However, in view of the reaction rates of melanoidin formation given by Hedges (1978), the rate of
condensation of glucose and lysine is considerably slower than the rate of the reaction between D-
glucose and H2S. The simulation experiments have shown that the reaction between H2S and
reducing sugars proceeds fairly readily under mild reaction conditions (40°C, 2 to 30 days, hydrous)
(Suzuki and Philp 1989). The solution of D-glucose and L-lysine (2:1 mol ratio) was heated at the
same temperature of 40°C for 30 days under N2, but no visual changes were observed, indicating
that higher reaction temperatures or longer reaction times are necessary for the significant
melanoidin formation.
The carbonyl-amine condensation is an important feature in the formation of melanoidins from

reducing sugars and amino acids. The formation of organosulfur compounds by the rapid reaction
of H2S with reactive carbonyl groups may hinder, or prevent, the progress of the condensation
between amino acids and reducing sugars.
The very slow formation of melanoidins under H2S conditions is of significant importance for various
geochemical processes. H2S in depositional environments is generally produced by sulfate-
reducing bacteria in bottom sediments, which are anoxic. In carbonate muds, where the availability
of iron is reduced, various forms of sulfur remain free and available for incorporation into residual
organic matter, especially reducing sugars. Marine-carbonate-type crude oils such as Middle East
and Lake Maracaibo oils are generally rich in organic sulfur, and similarly, kerogens in carbonate
source rocks are also rich in organic sulfur (Gransch and Posthuma 1974). In addition, carbonate
rocks have anomalously high extractable organic matter-total organic carbon ratios, indicating that
carbonate rocks are generally poorer in the kerogens than clastic rocks such as clayey shales (Hunt
1961; Taguchi 1982; Palacas 1983). High sulfur contents and lower abundance of kerogens in
carbonate rocks seem to parallel the observations of sulfur incorporation into the residues formed in
the presence of H2S and the slow formation of melanoidins under laboratory conditions. The rapid
incorporation of sulfur from H2S into reducing sugars to form organosulfur compounds in the very
early stages of diagenesis can be another important controlling factor governing the formation of
sulfur-containing geopolymers in recent sediments.

Laboratory experiments simulating the incorporation of sulfur from H2S, or elemental sulfur, into
various types of organic matter were performed to reveal the geochemical formation of organosulfur
compounds in the early stages of diagenesis. The following conclusions could be drawn from this
study:
1. Sulfur incorporation, from H2S, into recently deposited algal mats occurs fairly readily under mild
reaction conditions (40°C, two to thirty days, and hydrous). However, elemental sulfur does not
react as readily with the cyanobacterial mats under the same conditions.
2. Sulfur incorporation, from H2S, into humic acids occurs more rapidly than into protokerogens. No
sulfur incorporation into the lipid fractions could be observed.
3. Changes in the distribution of pyrolysis products before and after H2S alteration of melanoidin (D-
glucose + L-lysine) were very similar to those observed in the cyanobacterial mats. The reaction
between H2S and melanoidin polymers is one of the useful model processes of sulfur from H2S
being incorporated into naturally occurring sedimentary organic matter.
Acknowledgment
This work was supported by National Science Foundation Grant No. EAR8608820. We thank Dr.
David Des Marias, NASA Ames, for providing Exportadora de Sal bacterial mats.
References
Aizenshtat, Z., A. Stoler, Y. Cohen, and H. Nielsen, 1983. The geochemical sulfur
enrichment of recent organic matter by polysulfides in the Solar Lake. In M. Bjorøy et al.,
eds., Advances in Organic Geochemistry 1981, pp. 279-288. New York: Wiley.
Aries, R. E., C. S. Gutteridge, and W. A. Laurie. 1988. A pyrolysis-mass spectrometry

investigation of pectin methylation. Anal. Chem. 60:1498-1502.
Arpino, P. J., I. Ignatiadis, and Gael De Rycke. 1987. Sulphur-containing polynuclear

aromatic hydrocarbons from petroleum. J. Chromatogr. 390:329-348.
Bayer, F. L. and S. L. Morgan, 1984. The analysis of biopolymers by analytical pyrolysis gas
chromatography. In S. A. Liebman and E. J. Levy, eds., Pyrolysis and GC in Polymer
Analysis, pp. 277-338, New York: Marcel Dekker.
Braekman-Danheux, C. 1985. Pyrolysis-gas chromatography-mass spectrometry of model

compounds from coal hydrogenates. Part 3. Sulphur compounds. J. Anal. Appl. Pyrol. 7:315-
322.
Brassell, S. C., C. A. Lewis, J. W. De Leeuw, F. De Lange, and J. S. Sinninghe Damsté.

1986. Isoprenoid thiophenes: novel products of sediment diagenesis? Nature 320:160-162.
Byrne, G. A., D. Gardiner, and F. H. Holmes. 1966. The pyrolysis of cellulose and the action

of flame-retardants. II. Further analysis and identification of products. J. Appl. Chem. 16:81-
88.
Casagrande, D. J. and L. Ng. 1979. Incorporation of elemental sulphur in coal as organic

sulphur. Nature 282:598-599.
Casagrande, D. J., G. Idowu, A. Friedman, P. Richert, K. Siefert, and D. Schlenz. 1979. H2S
incorporation in coal precursors: origins of organic sulphur in coal. Nature 282:599-600.
Douglas, A. G. and B. J. Mair. 1964. Sulfur: role in genesis of petroleum. Science 147:499-
500.
Francois, R. 1987. A study of sulphur enrichment in the humic fraction of marine sediments
during early diagenesis. Geochim. Cosmochim. Acta 51:17-27.
Gransch, J. A. and J. Posthuma. 1974. On the origin of sulfur in crudes. In B. Tissot and F.
Bienner, eds., Advances in Organic Geochemistry 1973, pp. 727-739. Paris: Technip.
Hedges, J. I. 1978. The formation and clay mineral reactions of melanoidins. Geochim.
Herod, A. A. and C. A. Smith. 1985. Release of oxygen and sulphur compounds from coal.
Fuel 64:281-283.
Hoering, T. C. 1973. A comparison of melanoidin and humic acid. Carnegie Inst. Wash. Year
Book 72:682-690.
Hoering, T. C. and P. E. Hare. 1973. Comparison of natural humic acids with amino acid-
glucose reaction products. AAPG Bull. 57:784.
Hughes, W. B. 1984. Use of thiophenic organosulphur compounds in characterizing crude

oils derived from carbonate versus siliciclastic sources. In J. G. Palacas, ed., Petroleum
Geochemistry and Source Rock Potential of Carbonate Rocks, pp. 181-196. AAPG Studies
in Geology no. 18.
Hunt, J. M. 1961. Distribution of hydrocarbons in sedimentary rocks. Geochim. Cosmochim.

Acta 22:37-49.
Kong, R. C., M. L. Lee, Y. Tominaga, R. Pratap, M. Iwao, R. N. Castle, and S. A. Wise.

1982. Capillary column gas chromatographic resolution of isomeric polycyclic aromatic sulfur
heterocycles in a coal liquid. J. Chromatogr. Sci. 20:502-510.
Lalonde, R. T., L. M. Ferrara, and M. P. Hayes. 1987. Low temperature, polysulfide reactions
of conjugated ene carbonyls: a reaction model for the geologic origin of S-heterocycles. Org.
Geochem. 11:563-571.
Merritt, C. and D. H. Robertson. 1967. The analysis of proteins, peptides and amino acids by

pyrolysis-gas chromatography and mass spectrometry. J. Gas. Chromatogr. 5:96-102.
Moers, M. E. C., J. W. De Leeuw, H. C. Cox, and P. A. Schenck. 1988. Interaction of glucose

and cellulose with hydrogen sulphide and polysulphides. In L. Mattavelli and L. Novelli, eds.,
Advances in Organic Geochemistry 1987, pp. 1087-1091. Oxford: Pergamon Press.
Nissenbaum, A. and I. R. Kaplan. 1972. Chemical and isotopic evidence for the in situ origin
of marine humic substances. Limnol. Oceanogr. 17:570-582.
Orr, W. L. 1978. Sulphur in heavy oils, oil sands and oil shales. In O. P. Strausz and E. M.
Lown, eds., Oil Sand and Oil Shale Chemistry, pp. 223-243. New York: Verlag Chemie Int.
Palacas, J. G. 1983. Geochemical and geological factors controlling generation of petroleum

in carbonate rocks. Proc. 11th World Petrol. Congr., panel discussion 1, paper 3, p. 3.
Payzant, J., D. S. Montgomery, and O. P. Strausz. 1986. Sulfides in Petroleum. Org.

Geochem. 9:357-369.
Philp, R. P. and A. Bakel. 1988. Heteroatomic compounds produced by pyrolysis of

asphaltenes, coals and source rocks. Energy and Fuel 2:59-64.
Philp, R. P., J. Li, C. A. Lewis. 1989. A geochemical investigation of crude oils from
Shanganning, Jianghan, Chaidamu, and Zhungeer Basins, China. Org. Geochem. 14(4):447-
460.
Poirier, M. A. and G. T. Smiley. 1984. A novel method for separation and identification of
sulphur compounds in naphtha (30-200°C) and middle distillate (200-350°C) fractions of
Lloydminster heavy oil by GC/MS. J. Chromatogr. Sci. 22:304-309.
Schmid, J. C., J. Connan, and P. Albrecht. 1987. Identification of long-chain

dialkylthiacyclopentanes in petroleum. Nature 329:54-56.
Shafizadeh, F. 1968. Pyrolysis and combustion of cellulosic materials. In M. L. Wolfrom and

R. S. Tipson, eds., Advances in Carbohydrate Chemistry, pp. 419-474. New York: Academic
Press.
Simmonds, P. G., G. P. Shulman, and C. H. Stembridge. 1969. Organic analysis by pyrolysis-

gas chromatography mass spectrometry, a candidate experiment for the biological
exploration of Mars. J. Chromatogr. Sci. 22:36-41.
Sinninghe Damsté, J. S., A. C. Kock-Van Dalen, and J. W. De Leeuw. 1988. Identification of

long-chain isoprenoid alkylbenzenes in sediments and crude oils. Geochim. Cosmochim.
Acta 52:2671-2677.
Sinninghe Damsté, J. S., T. I. Eglinton, J. W. De Leeuw, and P. A. Schenck. 1989a6.

Organic sulphur in macromolecular sedimentary organic matter: I. Structure and origin of
sulphur-containing moieties in kerogen, asphaltenes and coals as revealed by flash

pyrolysis. Geochim. Cosmochim. Acta 53:873-889.
Sinninghe Damsté, J. S., W. I. C. Rijpstra, A. C. Kock-Van Dalen, J. W. De Leeuw, and P. A.

Schenck. 1989b. Quenching of labile functionalised lipids by inorganic sulphur species:
Evidence for the formation of sedimentary organic sulphur compounds at the early stages of
diagenesis. Geochim. Cosmochim. Acta 53:1343-1355.
Sinninghe Damsté, J. S., J. W. De Leeuw, A. C. Kock-Van Dalen, M. A. De Leeuw, F. De

Lange, W. I. C. Ripjstra, and P. A. Schenck. 1987. The occurrence and identification of
series of organic sulphur compounds in oils and sediment extracts. I. A study of Rozel Point
Oil (U.S.A.). Geochim. Cosmochim. Acta 51:2369-2391.
Suzuki, N. and R. P. Philp. 1989. Formation of melanoidins in the presence of H2S. Org.
Geochem. (accepted).
Taguchi, K. 1982. On the characteristics of carbonate sediments as petroleum source rocks-

A review. Sci. Rep. Tohoku Univ., series 3, 15:149-161.
Vairavamurthy, A. and K. Mopper. 1987. Geochemical formation of organosulphur

compounds (thiols) by addition of H2S to sedimentary organic matter. Nature 329:623-625.
Valisolalao, J., N. Perakis, B. Chappe, and P. Albrecht. 1984. A novel sulfur containing C35
hopanoid in sediments. Tetrahedron Lett. 25:1183-1186.
Wasterman, D. W. D., S. S. Ketti, M. W. Vogizang, L. I. Chen-Lie, B. C. Gates, and L.

Petrakis. 1983. Capillary column gas chromatography with sulphur- and nitrogen- specific
Hall detectors for determination of kinetics of hydroprocessing reactions of individual
compounds in coal-liquid fractions. Fuel 62:1376-1378.
Whelan, J. K., J. M. Hunt, and J. Berman. 1980. Volatile C1-C7 organic compounds in
surface sediments from Walvis Bay. Geochim. Cosmochim. Acta 44:1767-1785.
Willey, C., M. Iwao, R. M. Castle, and M. L. Lee. 1981. Determination of sulfur heterocycles
in coal liquids and shale oils. Anal. Chem. 53:400-407.
Wong, C. M., R. W. Crawford, and A. K. Burnham. 1984. Determination of sulfur-containing

gases from oil shale pyrolysis by triple quadruple mass spectrometry. Anal. Chem. 56:390-
395.


13. Bitumen Classification and Biomarker Correlation Studies

Based on Organic Extracts from Neogene Gulf of California
Sediments
The present work attempts to extend the facies concept to apply to source rocks, bitumen, and
petroleum. Facies has been defined as "the aspect, appearance and characteristics of a rock unit,
usually reflecting the conditions of its origin; especially as differentiating the unit from adjacent or
associated units .|.|.; Latin `face, form, aspect, condition'|" (Bates and Jackson 1980). Description
according to both original input and subsequent diagenesis is detailed in this paper with the
extensive data base provided by McEvoy (1983).
Attempts to use "biological markers" as indicators of ancient environments as well as maturity

indices appear to have reached a "crossroads" at which several different philosophical paths may
be followed. The present work aims to specify one direction and show where this path could lead.
De Leeuw (1988) notes that "in environmental and diagenetic studies it is not clear anymore to what
kind of information content of the organic molecules is referred." Therefore, it is necessary to
reevaluate the "marker concept in organic geochemistry." Four approaches recommended by de
Leeuw (1988) are to recognize: (1) Biological specificity (uniqueness in organisms) of a given
organic compound(s). He links the structural specificity of sedimentary organic molecules to various
precursor molecules and notes that these transformations can be used to unravel diagenesis. De
Leeuw (1988) further notes that: (2) Distribution patterns such as n-alkane patterns can be used to
recognize inputs (e.g., plant wax derived from higher plants) and degree of diagenesis (n-alkane
odd over even predominance). He also notes that: (3) Mode of occurrence of lipids, i.e., whether
free or bound, may be significant in delineating paleoenvironments and diagenetic pathways.
Finally: (4) Total profiles can be used to give relative concentrations of individual compounds or
classes of compounds directly. These profile data can be used to determine relative lipid
contributions from various organisms and the extent of lipid diagenesis.
Mackenzie et al. (1988) concentrate on the description of maturity alone (as opposed to early
diagenetic and environmental studies). Mackenzie et al. (1988) recognize five maturity boundary
zones based on first appearances (in order of increasing maturity) of (1) 22S 17 (H), 21 (H)
hopanes (> 5%); (2) triaromatic steroids (> 10%); (3) 5 (H), 14 (H), 17 (H) 20S steranes (> 5%),
(4) abundant 5 (H), 14 (H), 17 (H) steranes (> 25%), (5) abundant C20 triaromatic steroids (>
60%). They claim to follow the convention established by metamorphic petrologists in delineating
these five zonal boundaries by the first appearance of a single new biomarker isomer (as opposed
to mineral assemblages). They do not follow Eskola's (1939) method in determining metamorphic
boundaries using the total mineralogy. However, the major thrust of the work of Mackenzie et al.
(1988) is to reconstruct the burial history of a basin by using biomarkers.
Seifert and Moldowan (1986), in their review of petroleum maturity and source correlation

parameters, praise the work of Mackenzie and McKenzie (1983) but note the problems of assigning
product/precursor relationships. For instance, the present authors note that the simple assumption
that triaromatic steroids derive from monoaromatic steroids is open to question. Other factors, such
as the differential catalytic effects of various mineral matrices upon isomerization and aromatization
reactions, also require further discussion in Mackenzie and McKenzie's (1983) work, even though
Mackenzie et al. (1988) note that "a different catalytic activity .|.|. with respect to the reactions could
change the appropriate kinetic constants." However, the studies of Mackenzie and Mckenzie (1983)
and Mackenzie et al. (1988) deal with rocks of far greater maturity than those reported in this paper.
Comet et al. (1981), Comet (1982), and Comet and Eglinton (1987) attempted to provide a
descriptive framework for organic geochemistry by applying the principle of relative proportions of
different carbon skeletons as being more environmentally diagnostic than the differing functionalities
exhibited by those skeletons. The former schemes devised by de Leeuw (1988) and Mackenzie et
al. (1988) address other purposes and are not readily suitable for the comprehensive descriptive
approach attempted here. Comet and Eglinton (1987) employed a simple grid system using
quantitative lipid data. The differing carbon skeletons formed one axis, and the differing common
functionalities (alcohols, ketones, alkenes, etc.) formed the other axis. Because of the complexity of
sedimentary lipid distributions, the many different lipid components were first classified by use of a
hierarchy of refined subsets based on Ruzicka's rules of terpenoid cyclization (Eschenmoser et al.
1955; Ruzicka 1959). These subsets were arranged in such a way as to indicate relationships
among the various structural groups. With this system some sixty sediments were crudely
quantitated. Three major sedimentary facies types could be recognized: (1) marine, (2) terrigenous,
and (3) bacterial. Further progress in environmental description was not possible, partly because of
sampling bias in the data set (mainly deep-sea marine sediments were analyzed). Diagenetic facies
were not described by Comet and Eglinton (1987).
Facies description can be subdivided into the following broad categories: (1) sedimentary facies, (2)
biofacies, (3) igneous facies, and (4) metamorphic facies. All of these categories form useful
philosophical frameworks for comparing organic extracts and oils. No descriptive framework of any
kind exists at present for organic biogeochemistry; it is simply not possible to describe a sediment in
terms of its soluble organic components.
Earlier work (de Leeuw 1988, Mackenzie et al. 1988) attempted to relate biological marker
distributions to specific theories such as product/precursor ratios or biological specificity; i.e., they
used deductive reasoning (using a specific model or "law" to explain relationships in a complex data
base). The present "geological" approach attempts to use inductive reasoning, i.e., to initially
formulate a descriptive methodology and subsequently derive any specific relationships or "laws"
from the classified data. In the early nineteenth century geologists started to "make sense" of
geological strata by classifying and correlating them by their composition and the fossils they
contained. An attempt is made here to apply similar methodology.
Geological Background
McEvoy (1983) and McEvoy and Maxwell (1983) analyzed a series of Miocene-Pleistocene
samples from offshore California (San Miguel Gap) obtained on Leg 63 of the Deep Sea Drilling
Project (DSDP) (Site 467). These nine samples consist mainly of diatomaceous clays and
claystones with one major interbedded tuff. Eight of these samples were chosen for the construction
of a sedimentary and diagenetic facies framework because of the very comprehensive nature of the
data base available. Acids, alcohols, ketones, as well as aromatic and aliphatic hydrocarbons, were

analyzed from every sample.
The quantitated lipid data from McEvoy (1983) are grouped and summed by the method of Comet
et al. (1981) and Comet and Eglinton (1987). Analytical methodology, details of paleoenvironment,
and diagenetic transformations within specific compound classes are discussed by McEvoy and
Maxwell (1983) and McEvoy, Eglinton, and Maxwell (1981).
Sediments (DSDP Site 467) from the San Miguel Gap are characterized by the very high
depositional rates and high total organic carbon values associated with an upwelling area.
Sedimentation rates of 75 m/my for the Quaternary and 150 m/my for the Late Pliocene have been
calculated (Yeats et al. 1981a). At nearby Site 471, recorded geothermal gradients between 70°
and 154°C/km were obtained. A value of 81°C/km was taken as minimum mean (Yeats et al.
1981b). The lithologies sampled consisted mainly of clay and claystone. The major clay constituents
are of the montmorillonite group. Many of the sediments are organic rich. This richness is due not
only to high productivity but also to the presence of an oxygen minimum in the overlying water
column that inhibits microbial and benthic activity within the sediment.
Description of Geochemical Data
Living systems are usually categorized by means of a taxonomy of groups and subgroups using the
Linnaeus binomial nomenclature. The methodology of Comet (1982) and Comet and Eglinton
(1987) outlines a first lipid (biomarker) taxonomy. The present work attempts to develop this
taxonomic scheme into a classification of organic-rich extracts in which original input and thermal
maturity are kept as separate descriptive parameters.
By application of the system of quantitation of structural groups (Comet et al. 1981) and structural
and functional summation (Comet and Eglinton 1987), the following patterns of lipid groupings are
described from the DSDP Site 467 sediments (McEvoy 1983).
The strikingly consistent patterns for steroid side chains downhole was a feature of Site 467.
Quantitation of Carbon Skeleton Groupings
Sample 63-467-3-3. Sample is principally characterized by abundant straight-chain (i.e., n-

paraffinic) lipid material of higher plant as well as algal origin; terpenoids are minor, consisting
essentially of steroids and triterpenoids, the latter mainly of the hopane type (figures 13.1 and 13.2).
Chlorophyll derivatives were minor. The distribution of steroid side chains was complex, comprising
several different side chains. Thus, in the desmethyl series, C26 (one side chain); C27 (two side
chains), C28 (one side chain), C29 (two side chains), and C30 (one side chain) were all present. In
the 4-methyl series both C29 and C30 (one side chain each) were also present. Other steroid
structures were also present.
Sample 63-467-18-5.
Sample contained a much larger proportion of steroidal components than straight-chain

components (13.1 and 13.2). Perylene was also more abundant, whereas chlorophyll derivatives
were present in relative concentrations similar to those of Sample 3-3; hopanoid triterpenoids were

relatively minor. The distribution of steroid side chains was almost identical to 63-467-3-3; however,
all samples investigated except Sample 467-63-2 had the same general pattern of steroid side
chains. In general, Sample 63-467-18-5 was typical of Site 467 in both its distribution of major
carbon skeleton groups and steroid side chains.
Sample 63-467-36-2.
Sample has a structural distribution similar to that of 18-5.
Sample 63-467-54-2.
Sample has a structural distribution similar to that of 18-5.
Sample 64-467-63-2.
Sample was the most unusual, for the single compound 17 (H), 18 (H), 21 (H)-28,30
bisnorhopane, present in minor relative concentration in the majority of other samples, was the
major component in 63-467-63-2. This compound has been described as being present in
abundance in Monterey shale and Monterey shale-sourced petroleum (Seifert et al. 1978). Sample
63-467-63-2 is of Upper Miocene age, whereas the onshore organic-rich shale member of the
Monterey is Middle Miocene (Ingle 1981). It may be that the bisnorhopane-rich shale facies is time-
transgressive (diachronous) and deposition started slightly earlier at the present onshore locations
before it developed in the Site 467 area. Possibly anoxic conditions developed more extensively
farther to the northeast. Alternatively, the `bisnorhopane facies' may have occurred several times
and at a number of locations in the Californian marine Middle to Late Miocene. More sampling
would be required to map the lateral and chronological extent and unity of this remarkable facies.
The distribution of steroid side chains was also apparently different in this sample as steroid
diversity was reduced; the norsteroids were absent, i.e., the 24-norcholestane and 27-nor-24-
methylcholestane derivatives.
Sample 63-467-74-1
Sample 63-467-74-1 is much richer in chlorophyll derivatives than the other samples. This sample
consists of a tuff (i.e., volcanic ash) and shows some evidence of enhanced diagenetic changes as
compared with the other samples investigated. The different distribution of skeletons as compared
with the other samples may reflect the activity of oxidation within this sample. However, the steroid
side-chain pattern was the same as that of the majority of other samples. Similarly, the enhanced
abundance of chlorophyll derivatives links with work undertaken on the same sample suite by
Louda and Baker (1981), who noted an increase in the pigment index at this particular stratigraphic
horizon (Middle-Upper Miocene boundary). The increase occurs in association with a tuff horizon,
and it is possible that the high relative tetrapyrrole content is linked to increased productivity
occurring in response to the nutrients released from volcanic emanations.
Sample 63-467-97-2.
Sample has a structural distribution similar to that of 63-467-18-5.

Sample 63-467-110-3.
Sample has a structural distribution similar to that of 63-467-18-5.
Diagenesis of DSDP Site 467 Sediments
Only steroids and triterpenoids are discussed in detail here. Brassell et al. (1984), and references
therein, reviewed diagenesis of saturated and aromatic steroidal hydrocarbons from the McEvoy
(1983) data base. However, Brassell et al. (1984) did not discuss functionalized steroids or attempt
to correlate the steroidal hydrocarbon data with optically based maturity data. The present work
attempts to build on the work of Brassell et al. (1984) and references therein. Brassell et al. (1984)
noted the following reactions: (1) sterene isomerization, i.e., 2-sterenes converting to 4- and 5-
sterenes with depth; (2) diasterene formation; the diasterenes forming by clay-catalyzed
rearrangement of 4>- and 5-sterenes via a possible 8 intermediate; (3) diasterene isomerization;
the 20S diasterenes steadily increase in relation to the 20R isomer with depth; (4) spirosterene
formation; backbone rearrangement of the 4- and 5-sterenes generates spirosterenes as a
secondary by-product; (5) spirosterene isomerization; 20S spirosterenes become enhanced relative
to the 20R with depth; (6) B-ring monoaromatic anthrasteroid formation; these compounds may
originate from 3,5-steradienes; (7) B-ring monoaromatic anthrasteroid isomerization; the B-ring
monoaromatic anthrasteroid 14 (H) isomer is replaced by the more stable 14 (H) isomer with
depth; (8) A-ring monoaromatic steroid formation; these compounds are also thought to originate
from steradienes; and (9) C-ring monoaromatic formation (from unknown diene precursors).
Brassell et al. (1984) noted that certain of these nine reactions were restricted to specific depth
ranges but did not attempt to correlate the data with vitrinite reflectance measurements. Instead
they used heat from downhole measurements as an indication of the degree of thermal stress to
which the sediments had been subjected. Some of the results were inconsistent. Thus Brassell et
al. (1984) noted that C27 diasterene isomerization (at 43°C) ranges between 10% and 45%,
depending on the basin from which the samples were taken. This variation in isomerization might
reflect the age of the sediments (in the older sediments the reaction having progressed further for a
given temperature than in younger sediments).
Because of the complexity and number of diagenetic reactions undergone by lipids in sediments, it
is proposed here to restrict discussion to just two major structural types that are typical of the many
other compounds present in sediments. These are the (1) ethylcholestane skeleton (derived from
C29 sterols) and (2) hopane skeleton (derived from various hopane triterpenoids). They are used
here to illustrate the sequence of transformations exhibited by the majority of other sedimentary
steroids and triterpenoids.
Gross Transformations of the Ethylcholestane Skeleton
The 24-ethylcholestane skeleton occurs in a variety of guises in Site 467 sediments (figure 13.3).
These include alcohols, ketones, and acids, as well as saturated, unsaturated, and aromatic
hydrocarbons. The sequence of transformations with increasing maturity exhibited by compounds
with the ethylcholestane skeleton, typical of other desmethylsteroids, are summarized below.

1. At low maturities (Ro = 0.27 - 0.33%) sterols and steroidal ketones are relatively abundant
(Samples 63-467-3-3, 63-467-18-5, and 63-467-36-2).
2. At vitrinite reflectance (Ro) of 0.35% the sterols disappear and are replaced by abundant
sterenes. Diasterenes also appear and steroidal ketones persist (Sample 63-467-54-2).
3. At the slightly higher vitrinite reflectance of about 0.37% the steroidal ketones also disappear to
be replaced by sterenes and steranes (63-467-63-2).
4. At vitrinite reflectance of 0.45% sterenes start to be replaced by diasterenes and steranes (63-
467-97-2).
5. At slightly higher Ro (about 0.48%) the diasterenes and steranes start to become minor as the
major steroid fraction begins to be dominated by steranes (63-467-110-3).
Vitrinite reflectance data are taken from Rullkötter, von der Dick, and Welte (1981).
Diagenesis of Ethylcholestenes and Ethylcholestanes
It is possible to recognize a detailed maturity progression by examination of the aliphatic steroid

hydrocarbons alone (figure .13.4). Thus sample 63-467-3-3 is characterized by the presence of
abundant 2-sterenes, 3,5 and 2,24(28)-steradienes (Ro = 0.27%). This "top" sample is polluted
with mature petroleum [i.e., 20S 5 (H), 14 (H), 17 (H) steranes, 5 (H), 14 (H), 17 (H) steranes
and diasteranes] hydrocarbons, though whether these are formed from natural seepage or
anthropogenic pollution is unclear at present.
In Sample 63-467-18-5 (Ro = 0.30%) changes are already apparent. These include the
disappearance of the 2,24(28)-steradienes and the appearance of 4 and 5-sterenes; 2 and 3,5-
sterenes persist. The C29 5 (H), 14 (H), 17 (H) 20R sterane is dominant. In Sample 467-36-2 (Ro
= 0.33%), 4,24(28) and 5,24(28)-steradienes occur; 4 and 5-sterenes become very abundant.
Indigenous 5 (H) steranes appear. 5 (H) 20R ethylcholestane is abundant: 2-sterene and 3,5-
steradienes persist in minor amounts.
At Ro = 0.35% (63-467-54-2) the 2 and 3,5-sterenes and steradienes disappear; diasterenes

make their first appearance and 4,24(28), 5,24(28)-steradienes, and 5B(H) steranes persist.
Sample 63-467-63-2 is very similar to Sample 36-2 (Ro = 0.37%), but all the steradienes have
disappeared.
In Sample 63-467-74-1 (Ro = 0.39%), the ash layer, maturation appears to have been locally
accelerated as 5 (H), 14 (H), 17 (H) steranes (both 20R and 20S isomers), and 5 (H), 14 (H), 17
(H) 20S steranes are present. No sterenes or steradienes are present. The 20R isomers are
dominant.
In Sample 63-467-97-2, diasterenes are the dominant C29 steroid hydrocarbon type, though 4-
and 5-sterenes persist (Ro = 0.45%), and 5B(H) and 5 (H) ethylcholestanes remain abundant

while 20S steranes are absent.
At Ro = 0.48% (Sample 63-467-110-3), three appearances occur simultaneously. These are: (1)
spirosterenes; (2) 20S 5 (H), 14 (H), 17 (H) steranes; and (3) 5 (H), 14 (H), 17 (H) (20R and S)
ethylcholestanes. However, sterenes and diasterenes persist and the 5 (H) 20R ethylcholestane
isomer still remains quantitatively dominant over the 20S 5 (H) ethylcholestane isomer. 5 (H)
ethylcholestane is also present. It is possible that Ro = 0.48% may be the vitrinite reflectance level
marking the first "true" appearance of the 20S isomers in the sterane series.
Using sterenes and steranes, three diagenetic "facies" assemblages can be recognized:
1. Assemblages characterized by the presence of sterenes with unsaturation in the side chain (Ro =
0.27 - 0.35%).
2. Assemblages characterized by the presence of diasterenes and the absence of sterenes with
unsaturation in the side chain (Ro = 0.37 - 0.48%). Both diasterenes and sterenes with unsaturation
in the side chain co-occur at Ro = 0.35%.
3. Assemblages characterized by the presence of 5 (H), 14 (H), 17 (H) steranes and 20S 5 (H),
14 (H), 17 (H) steranes (onset of oil generation).
Diagenesis of Aromatic Steroids
The top sample (63-467-3-3) is probably contaminated with petroleum-derived aromatics. Thus,
C29 C-ring monoaromatics and C28 triaromatic steranes are of equivalent abundance and the
mono/triaromatic sterane ratio closely resembles mature petroleum (figure 13.5). Samples 63-467-
18-5, 63-467-36-2, and 63-467-54-2 (Ro = 0.3 - 0.35%) all show similar (immature) aromatic steroid
profiles. All of these three samples contain 14 (H) and 14 (H) anthrasteroids B-ring anthrasteroids
with unsaturation in the side chain ( 22), and C-ring aromatics also with unsaturation in the side
chain. With maturation the 14 (H) anthrasteroids are gradually replaced by 14 (H) anthrasteroids
until, at Ro = 0.37%, all 14 (H) anthrasteroids have disappeared (Sample 63-467-63-2). Also in
Sample 63-467-63-2 all aromatic steranes with unsaturation in the side chain have been eliminated.
In Sample 63-467-74-1 (Ro = 0.39%) the assemblage of aromatics is substantially different from the
overlying samples. Thus diaromatic AB-ring anthrasteroids make their first appearance and
triaromatic steroids become abundant.
In Samples 63-467-97-2 and 63-467-110-3 (Ro = 0.45 - 0.48%) further (unknown) diaromatic
steranes make their first appearance and triaromatic steranes continue to become more abundant.
Other aromatic steroids are present throughout the sequence (with the exception of Sample 63-467-
3-3). These include A-ring monoaromatics. Thus the whole sequence marks two major maturity
"facies" assemblages:
1. Assemblages characterized by the presence of aromatic steranes with unsaturation in the side
chain and 14 (H) anthrasteroids (Ro = 0.27 - 0.35%).
2. Assemblages marked by the absence of 14 (H) anthrasteroids and aromatic steranes with no

unsaturation in the side chain. Instead, various diaromatics and the 14 (H) anthrasteroids are
present. Triaromatics become abundant (Ro = 0.37 - 0.48%). 5 (20S) and 5 (H), 14 (H), 17 (H)
steranes, 22S 17 (H), 21 (H) hopanes are present in the lower part of this zone.
Diagenesis of Hopanoids
In general, the diagenetic transformations of hopanoids follow the same sequence as the steroids
(figure 13.6). Thus the top sample (63-467-3-3) contains mature (petroleum-derived), hopane
hydrocarbons, such as the extended 17 (H), 21 (H) 22S hopanes. These mature hopanes are
mixed with the various hopanols and hopanoid ketones of highly immature, indigenous origins(Ro =
0.27%). Samples 63-467-18-5 and 63-467-36-2 are similar to 63-467-3-3 (Ro = 0.30 - 0.33%) but
lack the mature hydrocarbon assemblage.
Sample 63-467-54-2 marks the disappearance of hopanols but contains an abundance of hopenes
and hopanes as well as hopanones (Ro = 0.35%). By contrast, Sample 63-467-63-2 contains
predominantly 17 (H), 18 (H), 21 (H)28, 30-bisnorhopane, which probably marks a major
paleoenvironmental event rather than diagenetic modification. Hopanones have disappeared (Ro =
0.37%).
Sample 63-467-74-1, composed of volcanic ash (Ro = 0.39%), contains abundant hopanoid acids
and a selection of hopenes and hopanes similar to that of 63-467-63-2; however, by contrast, the
bisnorhopane is minor. Sample 63-467-74-1 also contains the 17 (H), 21 (H) 22S homohopane.
The anomalous presence of this compound signifies enhanced diagenesis, possibly related to the
volcanic-derived, metastable mineralogy of the deposit.
Samples 63-467-97-2 and 63-467-110-3 are very similar; they contain few functionalized
components. Sample 63-467-110-3 is of particular interest, for it demonstrates the coexistence of 17
(H), 21 (H) hopanes, 17 (H), 21 (H) 22S extended hopanes, and numerous hopenes at Ro =
0.48% (approaching the beginning of oil generation).
In summary, the diagenesis of hopanoids follows a sequence of transformations very similar to

those exhibited by the steroids; two major facies assemblages can be recognized:
1. functionalized assemblages characterized by the presence of hopanols and hopanones and
2. defunctionalized assemblages characterized by the absence of hopanols and hopanones These

two maturity assemblages correspond in depth to a major transformation among the steroid
assemblages (compare figure 13.6 with figures 13.3, 13.4, and 13.5).
Application of the Facies Concept to Leg 63 Sedimentary Bitumens
Sedimentary Facies Description
By means of the concept of carbon skeleton diagenetic stability (Eglinton and Calvin 1967; Comet
et al. 1981; Comet and Eglinton 1987) the differing proportions of carbon skeleta in organic extracts
and oils can be applied to the numerical description of biomarker inputs to sediments and oils.

Sediments, originally rich in sterols, will yield oils or source rocks also rich in steranes. In particular,
the bisnorhopane-rich horizon (Sample 63-467-63-2) would be anticipated to yield bisnorhopane-
rich oils very deficient in normal paraffins. This has been noted by Curiale, Cameron, and Davis
(1985) for certain of the Monterey rock extracts.
The differing proportions of steroids, hopanoids, etc. can be assigned as a "primary" feature,
possibly reflecting the original depositional imprint. A nomenclature for organic-rich rocks and
bitumens can then be constructed based on the relative amounts of various major terpenoid types.
An example would be Sample 63-467-63-2; this could be termed "bisnorhopanate" (to crudely follow
geological terminology). The "bisnorhopanate" can be correlated with a similar "bisnorhopanate"
onshore Monterey (Seifert et al. 1978; Giger and Schaffner 1980).
This first attempt at a system to describe organic inputs to sediments is strongly influenced by both
sedimentary petrology and coal petrography classification. Hence, just as the principal components
of a sedimentary rock allow one to distinguish argillaceous facies (c.f. claystone), or a quartz-rich
facies (sandstones and siltstone), a similar approach can be applied to the bitumen component
fractions of a sediment. Mixtures in sedimentology are referred to by the use of combined names, e.
g., calcareous clays, etc. A similar methodology is applied in coal petrography. Thus, by applying
the existing geological philosophy of sedimentary and metamorphic facies, it becomes possible to
compare, contrast, and describe organic-rich rocks.
Problems occur with the description of sediments containing a diverse assemblage of terpenoids,
for any nomenclature must take into account both the principal (i.e., most abundant) and secondary
(i.e., quantitatively minor) components. With the McEvoy (1983) data, Sample 63-467-110-3 could
be termed a "steranic polyterpenate," whereas Sample 63-467-3-3 could be named as a
"paraffinate." A nomenclature that bears some relationship to terminology already widely used by
geochemists and geologists stands the best chance of becoming widely accepted. The problems of
devising a descriptive geochemical terminology are dealt with in greater detail in a subsequent
section.
Diagenetic Facies Description
In describing diagenetic facies in the bitumen fractions of sediments, preexisting terminology used
for describing metamorphic transformations is adapted for use with the description of biomarker
diagenesis. Miyashiro (1975) has summarized the metamorphic transformations of sediments and
igneous rocks:
It is conventional to define a metamorphic facies using boundaries at which major mineral changes
occur. Each of the zones, based on progressive mineral changes, corresponds to definite ranges of
temperature, rock pressure and chemical potential of H2O.
A metamorphic facies is characterized by the total assemblage of minerals and not necessarily by
any single mineralogical transformation. It is also important to recognize that metamorphic
transformations are mainly governed by the thermodynamic equilibria, whereas diagenesis (and
maturation) is governed by kinetics. It is probable that most of the assemblages described here are
not in thermodynamic equilibrium.
The transition temperature between diagenesis and low-grade metamorphism has been placed

around 150°C, i.e., toward the bottom of the oil window (Miyashiro 1975). By applying the same
descriptive methodology as utilized in metamorphic zonation (Eskola 1939; Miyashiro 1975), it may
be possible to extend the metamorphic zonation into a low-temperature diagenetic zonation based
on the major transformations of total assemblages of biomarkers. As yet it is unknown to what
extent these transformations are accelerated or retarded by pressure; also the mineral matrix plays
a major role in determining the extent and type of transformation. Another "unknown" is the relative
importance of heating rate in determining assemblage type. Thus it is unknown at present whether
or not the proposed facies boundaries will always occur at the same vitrinite reflectance values.
Appreciation of the significance of all these aspects will require further fundamental work.
The present work has allowed the formal definition of two major diagenetic facies zones and a
number of "subfacies" to be constructed. These two facies zones follow the philosophy of Eskola
(1939) in utilizing the total assemblage of components in delineating a facies type. Both molecular
"appearances" and "disappearances" are used in the construction of zonal boundaries (see figures
13.3, 13.4, 13.5, and 13.6).
The two major diagenetic facies characterized are:
Zone A, Steroid and triterpenoid ketones and alcohols are present and diagnostic (they
progressively lose water to generate various alkenes).
Zone B, Defunctionalization of alcohols and ketones is complete (but carboxylic acids remain). The
upper limit of the zone is delimited by the presence of steroid and triterpenoid ketones and alcohols;
its lower limit is defined by the disappearance of 17 (H), 21 (H) hopanes. 5 (H), 14 (H), 17 (H)
20S and 5 (H), 14 (H), and 17 (H) steranes appear in this zone.
Within the aliphatic and aromatic hydrocarbons these two major facies zones can be subdivided by
their characteristic compound assemblages. These are as follows:
Zone A1, Ro = 0.25 - 0.29% (Represented by Sample 63-467-3-3).
Triterpenoid and steroid ketones and alcohols are present. Triterpenes (hop-22(29)-ene) and
sterenes are also present, particularly 2-sterenes and 2,24(28)-steradienes. Aromatic steroids with
unsaturation in the side chain also are important; 14 (H) anthrasteroids are present.
As above, but characterized by the absence of 2,24(28)-sterenes and the persistence of 3,5-
steradienes. The 4 and 5-sterenes appear.
As above, but characterized by the coexistence of 2-sterenes and 3,5, 4,24(28), 5,24(28)-
steradienes.

The "transitional zone," characterized by the presence of steroid and triterpenoid ketones (and 4-
methylsterols) but lacking desmethyl stanols and sterols. 2- and 3,5-sterenes are absent, but 4,24
(28)- and 5,24(28)-steradienes are present. Diasterenes (in clay-rich facies) make their first
appearance, and 4- and 5-sterenes are present in large amounts.
Zone B1, Ro = 0.37 - 0.45% (Represented by Samples 63-467-63-2, 63-467-74-1, and 63-467-97-
2).
This is characterized by the presence of abundant steranes and triterpanes with the original
biosynthetic stereochemistry preserved [e.g., 5 (H)20R steranes and 22R 17 (H), 21 (H) extended
hopanes]. Steroid and triterpenoid alcohols and ketones are no longer present. Steradienes, 2-
sterenes, and 14 (H) anthrasteroids, and 22 aromatic steroids are also absent, but AB-ring
diaromatic anthrasteroids are present and diagnostic of this zone (the absence of AB diaromatic
anthrasteroids in Sample 63-467-63-2 may be due to environmental factors).
Zone B2 (Sample 63-467-110-3).
Similar to Zone B1 but additionally characterized by the appearance of 5 (H), 14 (H), 17 (H) and 5
(H), 14 (H), 17 (H), and 20S steranes as well as spirosterenes and 22S 17 (H), 21 (H) extended
hopanes. Zone B2 probably marks the "onset of oil generation" (Ro = 0.48%).
Development of a First Descriptive Classification of Bitumens and Oils Using Site 467 Samples
Combining the methodologies for sedimentary facies (organic input) description with diagenetic
facies description has been accomplished by using a system of suffixes similar to that used in
sedimentary and metamorphic petrology; e.g., quartz-ite is the name given to a metamorphic rock
composed essentially of quartz. From organic petrology, examples of the use of suffixes are the
compositionally descriptive terms such as cutin-ite, or exin-ite (depending on the origin of the
particular material encountered). Cutinite derives from plant cuticle, and exinite derives from spore
and pollen exines. Utilizing a similar geological nomenclature for extracts, the following rules of
classification are applied.
Description of Organic Input
Description of organic inputs into sediments may be integrated into the classification by using the
various major lipid-based structural components as descriptive prefixes (see table 13.2).
1. The principal structural group is used as the preliminary adjective; this word is terminated the
suffix -ic (see table 13.2).
2. The secondary structural group is terminated by the suffix -oid (see table 13.3).
3. Quantitative data may be supplied, if desired, by a "numerical qualifier," expressed as a

percentage, following each structural group (see table 13.3).
4. Where large (and approximately equal) amounts of different terpenoid groups co-occur in a

bitumen or oil, the descriptive prefix polyterpen may be used. This term may be used as either a
primary or secondary qualifying name. Similarly, where a complex mixture of different (possibly
unknown) triterpane skeletons is encountered, the term triterpan may be used in a similar "general"
fashion. The other major terpenoid components are dealt with in the same way (steroids,
diterpenoids, sesquiterpenoids, etc.).
5. Where one single compound (or closely related group of compounds) is overwhelmingly
predominant, then the name of the compound with the suffix ic can be used instead of the structural
group to which the compound belongs. Similarly if the structural class to which a given very
abundant compound (or compound group) is of secondary quantitative importance, then the name
of this compound, completed by the suffix -oid, should go after the principal structural group; e.g.,
Sample 63-467-63-2 is called a "bisnorhopanic polyterpenoid isomerizate" (I) because
"bisnorhopane" is overwhelmingly predominant in the hopane class of compounds. If the hopane
group (as represented by the bisnorhopane) had been somewhat less abundant than the other
structural groups present, the bisnorhopane would be given "second place" and the name of the
extract would have been a "polyterpenic bisnorhopanoid isomerisate (I)."
Description of Diagenetic Stage
1. According to the maturation stage reached by the oil or bitumen, the generalized suffix -ate may
be modified (see table 13.1).
2. The description of maturation stage may be further qualified by the application of the diagenetic
zonation developed for the Leg 63 sediments. Thus the ketolates can be further subdivided into 4
subzones based on the Site 467 samples described in this paper. "Isomerizate 1" corresponds to
Zone B1 and "Isomerizate 2" corresponds to Zone B2 as applied to the Site 467 samples. The other
"isomerisates" and "secanates" await further subdivision. This descriptive combination of structural
types and their diagenetic derivatives allows the classification of the Leg 63 sediments (see table
13.3).
The idea of using a systematic nomenclature of chemically based descriptive terms in combination
with a simple system of prefixes and suffixes is useful for rapid verbal communication of the
environmental and diagenetic affinities (or facies) of an extract or oil. For written communication, it
might be more meaningful to use a composite system of quantitated abbreviations (see table 13.3).
A system of this kind would allow rapid environmental comparison of sediments of differing
diagenetic history but presupposes minimal migrated bitumen.
It is possible to apply and combine the various theories and concepts of organic geochemistry
developed by Seifert and Moldowan (1986), de Leeuw (1988), Mackenzie et al. (1988), and others
into a more nearly comprehensive synthesis. The combination of the various theories is made
possible by applying the facies concept to organic geochemical data. Application of the facies
concept allows formal objective description of organic extracts in terms of their lipid-based
molecules. De Leeuw's (1988) major objectives were the reconstruction of paleoenvironment using
(1) the biological (taxonomic) specificity of "biological markers," (2) their distribution pattern, (3)
mode of occurrence, and (4) total lipid profiles. This environmental approach can be contrasted with
the present authors' "geological" philosophy for the reconstruction of ancient environments by (1)
classifying and quantitating different assemblages of carbon skeleta in the sediments of various
described environments and (2) attempting to correlate these assemblages of carbon skeletons with

oils or sediments (of unknown environment) in order to establish their depositional

paleoenvironment by comparison. This comparative approach extends the use of chemical fossils in
determining depositional environment by the application of a concept closely related to
uniformitarianism (the present is the key to the past). The data base outlined by Comet and Eglinton
(1987) marks a first attempt to outline the molecular assemblages of carbon skeletons characteristic
of various environments for comparative purposes. The present work shows how the recognition of
a given organic facies can then be successfully used for the correlation of organic-rich strata or oil
to source correlation. In particular the recognition of the deep ocean bisnorhopane horizon in the
Monterey shale offshore (Leg 63) equivalent may be significant. Also, the fact that the
paleoenvironmental signature, as exemplified by the steroids, is stable and independent of
diagenesis is noteworthy.
The work of Mackenzie et al. (1988) summarizes current thinking in the field of maturation studies
and emphasizes the rate of various organic reactions as a basis for assessing the heating rate of
various sediments. It describes five maturity zones on the basis of these reactions but uses a
convention different from the one adopted here by using the first appearance of a single new
diagenetic reaction product in the delineation of diagenetic zones. The approach utilized here
adopts the mineralogical approach of Eskola (1939) and emphasizes transformation of the total
assemblage, rather than individual product/precursor reactions. Here, a useful comparative tool for
measuring the progress of organic maturation has been devised by comparing the 9 "biomarker
assemblage maturity zones" with vitrinite reflectance measurements. Nine biomarker assemblage
maturity zones are proposed for describing the progress of organic diagenesis. By linking this
zonation to a system for the description of sedimentary lipids using biomarker structural ratios, a
comprehensive scheme for the description of organic bitumen facies (both diagenetic and
environmental) is proposed.
The present work has shown that it is possible to compare environments in spite of their very
different degree of diagenesis. Thus, Samples 63-467-18-5 and 63-467-97-2 are of similar
environment of deposition, though Sample 63-467-18-5 is characterized by abundant functionalized
lipids and Sample 63-467-97-2 contains mainly hydrocarbons. Without classifying, grouping, and
quantitating the various molecules, this correlation would not be readily apparent. The parameters
used can routinely classify and correlate petroleums.
In the absence of a system of molecular grouping, the environmentally diagnostic data from
immature sediments cannot be satisfactorily compared with oils and source rocks. In particular,
those sediments that are partially mature (vitrinite reflectance equivalent of 0.3 - 0.45%) contain
mixtures of diagenetically derived alkenes and alkanes, very different from most petroleum or the
original lipid assemblage, and characterized by extreme complexity. This kind of data requires a
format, such as the one proposed, that would allow data comparison between recent sediments and
petroleum.
The present classification scheme is not to be considered the final word on relating lipid biomarkers
to geological processes; rather it is intended as a stimulant for consideration of better classification
schemes in the future. However, the approach used here represents a formal attempt to apply
classical geological philosophy to a very large biomarker database and to interpret the derived data
in terms of the geological objectives of correlation of similar facies.
Appendix 13.1.

Reviewer Comments by Jan de Leeuw
In my view this sort of paper is coming twenty or thirty years too early. It gives the impression that in
organic geochemistry everything is sorted out and that the time has come to refer to sedimentary
organic matter in terms suggested in tables 13.1, 13.2, and 13.3.
I have the feeling that we cannot (yet) do this. Organic geochemistry is still in its early stages, and
just at the moment, the situation is chaotic; old ideas are shown to be wrong, our concept of
kerogen is changing dramatically, the influence of organic sulfur can hardly be overseen presently,
and even with the simple biomarkers like hopanes and especially steranes, the recent reviews of
five years ago are hopelessly out of date.
The major problem was and still is that we are still in the process of finding out how diagenesis/
maturation parameters can be separated from input (depositional environment) parameters. I will try
to demonstrate that by just three examples from Delft work, because I am most familiar with that.
Steroids
Very recently it became clear that the accepted isomerization pathway of sterenes ( 2 / 3 / 4 / 5,
etc.) cannot be true. Hence, we had to formulate an alternative. From that it became clear that, for
example, (14)-, 14-, spirosterenes and 14 , 17 71 steranes, at least in immature sediments,
originate exclusively from 7-sterols. The proposed alternative pathway has been proved several
times now.
Thus, -steranes can reflect 7-sterol contribution (of specific organisms) confusing the whole idea
of isomerization during diagenesis. To some extent this may hold for the hopanes as well if hopenes
are abundantly present originally. In general, more data are becoming available indicating that the
Mackenzie approach/theory is not valid. In conclusion, the recent studies of steroids show that
diagenesis and input have not been separated rightly over the last decades.
Resistant Biopolymers
It is becoming clear that the Tissot-Welte/Hunt/Durand approaches to kerogen formation-via

depolymerization, random polymerization, and condensation-are entirely wrong. New, resistant-
especially aliphatic-biopolymers have been encountered in many algal cell walls, plant cuticles,
barks, dinoflagellate cysts and resins. Simulation experiments with these biopolymers show
surprisingly well where oil and coal substituents come from. These recent findings are going to
change dramatically our ideas of kerogen formation and, hence, relationships between bitumens
and kerogens.
The example of the Messel Shale kerogen is very illuminating; over the years the organic matter of
Messel has been considered of mainly bacterial origin (many hopanes, etc.). It is clear now that
95% of the organic matter (the kerogen) consists solely of a resistant part of an algal (T. minimum)
cell wall. Thus, on the basis of the bitumen, we may consider Messel organic matter as a
"bacteriate," based on the kerogen as an "alginite." It is, of course, clear that the organic matter is of
algal origin, for the far greater part, with long n-alkyl moieties dominant.

Thus, one cannot define the type of organic matter on bitumen characteristics at all. We must do
that on the characteristics of the bulk of the organic matter; otherwise, the results will be frivolous.
Organic Sulfur Compounds
Recent research on organic sulfur compounds in sediments, oils, and coals has made clear that the
lipid distributions in bitumen can be heavily biased by incorporation of inorganic sulfur into specific
lipids during very early diagenesis; e.g., phytane/pristane ratios are heavily influenced, carotenoids
can be totally absent as hydrocarbons though originally present overwhelmingly, etc. As a
consequence we must conclude that the environment of deposition (marine, lacustrine, hypersaline)
heavily influences the lipid distributions we do see in sediments.
Thus, once again, the classification of the organic matter based on the bitumen characteristics can
lead to erroneous results and complete misapprehensions.
By means of these three examples I hope to have made clear that the approach by the authors to
classifying sedimentary organic matter is very dangerous and can be highly misleading.
At the moment the kerogen classifications may, unfortunately, still be most reliable because these
kerogens represent the bulk of the organic matter.
References
Bates, R. L. and J. A. Jackson, eds. 1980. Glossary of Geology, 2d ed. Falls Church, Va:
American Geological Inst.
Brassell, S. C., J. McEvoy, C. F. Hoffmann, N. A. Lamb, T. M. Peakman, and J. R. Maxwell.

1984. Isomerisation, rearrangement and aromatisation of steroids in distinguishing early
stages of diagenesis. In P. A. Schenck, J. W. de Leeuw, and G. W. M. Lijmbach, eds.,
Advances in Organic Geochemistry 1983, pp. 11-23. Oxford: Pergamon.
Comet, P. A. 1982. The use of lipids as facies indicators. Unpublished thesis, Univ. of Bristol
(U.K.).
Comet, P. A. and G. Eglinton. 1987. The use of lipids as facies indicators. In J. Brooks and
A. J. Fleet, eds., Marine Petroleum Source Rocks, pp. 99-117. Geolog. Soc. spec. publ. no.
26.
Comet, P. A., J. McEvoy, S. C. Brassell, G. Eglinton, J. R. Maxwell, and I. D. Thomson.

1981. Lipids of an Upper Albian limestone, section 465A-38-3. In J. Thiede, T. Vallier, et al.,
eds., Initial Reports of the Deep Sea Drilling Project, 62:923-937. Washington, D.C.: U.S.
Govt. Printing Office.
Curiale, J. A., D. Cameron, and D. V. Davis. 1985. Biological marker distribution and
significance in oils and rocks of the Monterey Formation, California. Geochim. Cosmochim.
Acta 49:271-288.

de Leeuw, J. W. 1988. In: M. Sohn, ed., ACS Symposium Series 305: Organic Marine
Geochemistry, pp. 33-61. Washington, D.C.: ACS, 1986.
Eglinton, G. and M. Calvin. 1967. Chemical fossils. Scientific American 216(21):32-43.
Eschenmoser, A. E., L. Ruzicka, O. Jeger, and D. Arigoni. 1955. Zur kenntnis der triterpene.
Helv. Chim. Acta 38:1890.
Eskola, P. 1939. Die enstehung der gesteine, ein lehrbuch des petrogenese, bearkeit. In T.
F. W. Barth, C. W. Covens, et al., Herausgegeben von dr. Carl W. Covens; mit 210
textabbddungen, vol. 8. Berlin: Springer Verlag.
Giger, W. and C. Schaffner. 1980. Molecular fossils in Miocene Monterey shales, California
(abstract). Proc. 26th Geol. Congr. 2:772.
Ingle, J. C., Jr. 1981. Cenozoic depositional history of the northern continental borderland of
southern California and the origin of associated Miocene diatomites. In C. Isaac, ed., Guide
to the Monterey Formation in the California Coastal Area, Ventura to San Luis Obispo:
Pacific Section, 52:1-8. AAPG.
Louda, J. W. and E. W. Baker. 1981. Geochemistry of tetrapyrrole, tetraterpenoid and

perylene pigments in sediments from the San Miguel gas (Site 467) and Baja California
Borderlands (Site 471): DSDP/IPOD Leg 63. In B. Haq, R. S. Yeats, et al., eds., Initial
Reports: DSDP 63:758-818. Washington, D.C.: U.S. Govt. Printing Office.
Mackenzie, A. S. and D. McKenzie. 1983. Isomerisation and aromatisation of hydrocarbons

in sedimentary basins formed by extension. Geology 120:417-470.
Mackenzie, A. S., D. Leythaeuser, F. J. Altebaumer, U. Disko, and J. Rullkötter. 1988.

Molecular measurements of maturity for Lias w shales in N. W. Germany. Geochim.
McEvoy, J. 1983. The origin and diagenesis of organic lipids in sediments from the San
Miguel Gap. Ph. D. thesis, Univ. of Bristol, U. K.
McEvoy, J. and J. R. Maxwell, 1983. Diagenesis of steroidal compounds in sediments from

the Southern California Bight (DSDP Leg 63, Site 467). In M. Bjoroy et al., Advances in
Organic Geochemistry, 1981, pp. 449-464. Chichester: Willey.
McEvoy, J., G. Eglinton, and J. R. Maxwell. 1981. Preliminary lipid analyses of sediments
from Sections 467-3-3 and 467-97-2. In B Haq, R. S. Yeats, et al., eds., Initial Reports'
DSDP 63:763-774. Washington, D. C.: U. S. Govt. Printing Office.
Miyashiro, A. 1975. Metamorphism and Metamorphic Belts. London: George Allen and
Unwin.

Rullkötter, J., H. von der Dick, and D. H. Welte. 1981. Organic petrography and extractable
hydrocarbons of sediments from the eastern North Pacific Ocean, Deep Sea Drilling Project
Leg 63. In B. Haq, R. S. Yeats, et al., ed Initial Reports: DSDP 63:819-836. Washington, D.
C.: U.S. Govt. Printing Office.
Ruzicka, L. 1959. History of the isoprene rule. Proc. Chem. Soc., p. 347.
Seifert, W. K. and J. M. Moldowan, 1986. Use of biological markers in petroleum exploration.

In E. B. Johns, ed., Biological Markers in the Sedimentary Record, pp. 261-290. Amsterdam:
Elsevier.
Seifert, W. K., J. M. Moldowan, G. W. Smith, and L. V. Whitehead. 1978. First proof of

structure of a C28-pentacyclic triterpane in petroleum. Nature (London), 271:436-437.
Yeats, R. S., B. U. Haq, et al. 1981a. Site 467: San Miguel Gap. In B. Haq, R. S. Yeats, et
al., eds., Initial Reports: DSDP 63:23-112. Washington, D.C.: U.S. Govt. Printing Office.
Yeats, R. S., B. U. Haq, et al. 1981b. Site 471: Offshore Magdalena Bay. In B. Haq, R. S.
Yeats, et al., eds. Initial Reports: DSDP 63:269-349. Washington, D.C.: U.S. Govt. Printing
Office.


14. Resolution of Sediment Hydrocarbon Sources:

Multiparameter Approaches
The complexity of hydrocarbon mixtures in nature requires a comprehensive analytic approach to

unravel a sediment's history. Advances in modern analytical techniques suggest that sensitive and
definitive fingerprints are possible. Supporting or ancillary evidence and an understanding of the
geologic setting are essential for the proper evaluation of detailed analyses. The discussion
presented here has been primarily restricted to aliphatic and aromatic marker compounds that are
widely used at the present time. However, a wide array of other classes of compounds are
potentially sensitive indicators including acyclic isoprenoids, diterpanes, 4-methyl steranes, and
heteroatomic compounds (porphyrins, sulfur-containing compounds, etc.). Ratios between groups
of compounds may also be informative when high-quality quantitative analyses are performed. The
advent of molecular stable isotopic measurements promises to offer an even more powerful
approach and a new dimension to deconvoluting the complex inputs represented by sedimentary
organic matter. The complete characterization of a sediment extract from bulk to molecular
composition and ultimately molecular isotopic composition enhances the reliability of the inferred
interpretations.
Hydrocarbons in the environment are composed of a complex mixture of compounds derived from
multiple sources. From an environmental chemistry, as well as an organic geochemical standpoint,
differentiation of these sources has become increasingly important over the years. The simple
determination of the presence or absence of petroleum pollution is often insufficient to assess the
environmental effect and fate of petroleum releases. A more finely resolved distinction of natural
seepage, biological inputs, refined products, crude oil, combustion products, recycled sediments,
and atmospheric fallout is necessary. The ability to resolve human-related, or human-induced,
hydrocarbon contamination from the biological background is an essential requirement of any
analytical approach. The increased use of molecular markers to interpret depositional environments
and paleoclimates as a function of organic matter input has also led to increased interest in
hydrocarbon markers that can be related to distinct, definable sources. But do hydrocarbons have
unique chemistries and distributions that lend themselves to full and complete resolution of multi-
endmember inputs, and what is the effect of diagenesis and catagenesis on original source
signatures? There is a need to understand which sedimentary hydrocarbons are primary in the
sense that they are directly biosynthesized and which are the products of diagenetic transformation.
A second, more difficult problem is delineating the presence of natural seepage from anthropogenic
inputs, especially when natural seepage and oil exploration/production coincide and in effect
produce the same hydrocarbon mixture.
The advent of high-resolution capillary gas chromatography with mass spectrometric detection has
produced sensitive and reliable hydrocarbon source indicators. Multiparameter approaches,
including bulk parameters (such as stable isotopes and extractable organic matter content),
molecular compositions, quantitative determinations, and areal as well as vertical distributions need

to be used to resolve the origins of complex environmental mixtures of hydrocarbons. No single

parameter or set of parameters is definitive for all situations, and each analytical strategy must be
evaluated in light of the goals of each study. The following summarizes several studies that illustrate
the use of multiparameter approaches to distinguish hydrocarbon sources and also describes
progressively more detailed analytical protocols. This report does not try to review the extensive
literature available on hydrocarbons in the environments but is designed to illustrate the application
of the principles derived from these studies (the reader is referred to several excellent review
articles). The studies summarized here emphasize the recognition of natural seepage and recycled
mature continental sediments as distinguished from anthropogenic and recent biological inputs.
Concurrent Processes
Processes associated directly or indirectly with seepage have often been useful in first suggesting
that natural seepage has occurred as contrasted with deposition of oil from the overlying water
column. For example, oil seeps are often associated with brine seepage (figure 14.4.1). Other
circumstantial evidence for seepage includes the presence of large amounts of gaseous
hydrocarbons and/or hydrates. In cases of massive seepage, as on the Gulf of Mexico slope,
sediment and hydrate gases closely match gases reservoired several thousand meters deep in the
subsurface (table 14.1; Kennicutt, Brooks, and Denoux 1988). Authigenic carbonate, identifiable by
its light isotopic composition and increased concentration, is often associated with seepage (figure
14.2; Kennicutt et al. 1989). Another seepage-associated feature is the presence of dense
biological communities that depend on a relatively continuous, long-term source of hydrocarbons
and anoxic conditions (Kennicutt et al. 1988). The association of chemical, biological, and physical
parameters and hydrocarbon seepage has been extensively documented on the northern Gulf of
Mexico continental slope. Often ancillary or circumstantial evidence confirms interpretations based
on more specific and diagnostic procedures, but the former is usually insufficient in itself to define
sources or resolve multiple inputs. For example, seep-associated phenomena are generally present
at massive macroseep sites and may not be present at more subtle microseep locations.
Bulk Parameters
Organic-solvent-extractable organic matter (EOM), also referred to as lipids and/or bitumens, is

increased in sediments contaminated by petroleum (figure 14.3; Kennicutt et al. 1989). EOM
concentration is useful in identifying the more extreme cases of macroseepage and/or massive oil
spillage. For highly contaminated sediments, EOM content can be used to indicate the amount of oil
present. However, EOM content alone cannot distinguish natural seepage from oil pollution, though
vertical distributions may be informative. Increased amounts of resins and asphaltenes are
collaborative evidence of petroleum contamination since organisms produce little or none of these
materials. Most sediments contain a measurable biological lipid component, and normalization to
organic carbon content is often necessary. Higher organic carbon content samples are often
associated with higher concentrations of biological lipids. The origin of anomalously high EOM
concentrations cannot be inferred from gravimetric measurements alone, though elevated levels do
suggest an additional EOM input beyond recent biological debris.
Stable Isotopes
The stable isotope composition of organic matter is potentially useful for recognizing hydrocarbon

sources. Carbon, sulfur, nitrogen, and deuterium stable isotope compositions can, on occasion,
indicate a unique source. The utility of stable isotopic composition is highly dependent on whether
sufficient resolution of end members is possible. In general a multi-isotope approach is more
definitive than any single isotopic composition. In macroseepage areas the carbon isotope of the
extractable organic matter approaches that of the suspected source material (figure 14.4.4;
Kennicutt et al. 1989). Carbon isotope ratios have been shown to be relatively resistant to alteration,
though some fractionation or partitioning was suspected during mousse formation at the IXTOC
blowout (Sweeney, Haddad, and Kaplan 1980; Kennicutt 1988). Carbon isotope ratios of
subfractions of the total extract (i.e., aliphatics, aromatics) can provide additional information (figure
14.5; Kennicutt, Brooks, and Denoux 1988). In general, stable isotope compositions can be used
only to establish that samplesare different, but they are more ambiguous for establishing that
samples are identical. Stable carbon isotope ratios of petroleum (liquids) worldwide only vary over
about a 15 parts per thousand range, and this limits the ability to resolve multiple sources.
Gas Chromatography-Flame Ionization Detection
Aliphatic hydrocarbon compositions have been extensively used by many investigators to estimate
the relative importance of hydrocarbon sources at a given location. The primary mode of analysis is
gas chromatography with flame ionization detection (GC/FID) of solvent-extracted hydrocarbons.
The use of aliphatic hydrocarbons as indicator compounds is based on the premise that
assemblages of normal and branched alkanes are associated with specific sources. In nature,
however, few unique aliphatic end-members occur. Several reviews of alkane hydrocarbon
distributions and their significance are available, and this material will be only briefly reviewed
(Brassell et al. 1978; Philp 1985a; Boehm and Requejos 1986). Basically, petroleum has little or no
carbon preference among alkanes, whereas biological sources preferentially synthesize alkanes
with an odd number of carbons. Plankton generally produce a simple mixture of hydrocarbons
dominated by n-C15,17,19 and pristane, so the presence of these compounds can be useful as
plankton indicators (Clark and Blumer 1967; Blumer, Mullin and Guillard 1970; Goutx and Saliot
1980; table 14.2). Petroleum also contains these compounds but usually contains nearly equal
amounts of n-C16,18,20 and phytane as well (Farrington et al. 1973; Farrington and Tripp 1977).
Straight-chain biowaxes with 25, 27, 29, and 31 carbons have been used extensively as indicators
of terrestrial or land-derived organic matter (Philp 1985a and references therein). Wax-derived
normal alkanes are also found in petroleum but are accompanied by near equal amounts of n-
C24,26,28,30. Immature sediments also contain these hydrocarbons but in general exhibit a
significant odd carbon preference similar to the original input. Weathered petroleum also contains
enhanced naphthenic components (termed the UCM or unresolved complex mixture) owing to loss
of paraffins by microbial degradation.
These relationships were used to determine hydrocarbon sources on the Gulf of Mexico slope as a
function of water depth, location, and time of sampling (Kennicutt et al. 1987a). Aliphatic
hydrocarbon indicator parameters, as measured by GC/FID, were evaluated in relation to the UCM,
an indicator of petroleum input; carbon preference index, an indicator of maturity and terrigenous
inputs; and bulk sediment parameters. Aliphatic hydrocarbon distributions suggest that Gulf of
Mexico slope sediments contain a mixture of terrigenous, petroleum, and plankton hydrocarbons.
The influence of river/land-derived material on the slope is widespread, as evidenced by the
presence and abundance of C23-C31 odd normal alkanes (figure 14.6). The dominance of plant
biowaxes may be due to their resistance to degradation and/or their entrapment in clays and/or
terrestrial humic complexes. Plankton and petroleum hydrocarbons are apparently more readily

available to microorganism attack and dissolution. Preferential preservation of hydrocarbons can

create a skewed picture of original source inputs. Since locations far removed from direct river
deposition often contain significant amounts of biowaxes, the transport of terrestrial material to the
slope is probably by secondary sediment movement and particle transport. Aeolian transport cannot
be ruled out as a source but its input is difficult to assess. Petroleum hydrocarbons, as recognized
by a complete suite of n-alkanes and a UCM, were detected at all locations and have a suggested
dual source in natural seepage and river-associated particulates. Natural seepage is widespread on
the Gulf of Mexico slope and is characterized by the co-occurrence of high concentrations of EOM,
UCM, and alkanes (figure 14.7; Kennicutt et al. 1988). Normal alkane distributions can be
substantially altered by microbial degradation and thus may be ambiguous indicators of petroleum,
though a significant UCM is usually still present (figure 14.8). Vertical hydrocarbon distributions, the
widespread detection of visible oil staining of cores, and subsurface geology suggest that seepage
is a significant input of petroleum hydrocarbons to Gulf of Mexico continental slope sediments
(Brooks et al. 1984, 1987; Lacerda, Kennicutt, and Brooks 1987; Kennicutt, Brooks and Denoux
1988; Kennicutt et al. 1988). Hydrocarbon distributions in general are patchy owing to the
nonuniform distribution of seepage on the slope. Terrigenous hydrocarbons are often preferentially
associated with argillaceous, organic-rich sediments, suggesting an association with river-derived
material.
Gas Chromatography-Flame Photometric Detection
Sulfur-containing compounds, particularly aromatics, are very limited in organisms but are widely
occurring in petroleum. The extractable and volatile sulfur fractions of sediments are conducive to
gas chromatographic analysis. Sediment extracts from 81 piston cores taken in water depths
between 70 and 1200 m on the Gulf of Mexico continental shelf and slope were analyzed for
dibenzothiophenes (DBT) by capillary gas chromatography with flame photometric detection (FPD;
Lacerda, Kennicutt, and Brooks 1987). The major aromatic sulfur compounds detected were
dibenzothiophene; methyl, ethyl, and propyl dibenzothiophenes; two unidentified sulfur-containing
compounds; and a series of benzothiophenes (figure 14.9). Vertical distributions generally showed
significant increases in DBT concentrations with depth. DBT distributions in sediment extracts were
similar to oils produced in the northern Gulf of Mexico (figure 14.10). Variations from this
composition may be due to microbial degradation in near surface sediments, but in general, DBTs
are relatively resistant to microbial degradation. The vertical and molecular distributions of DBT, as
well as the regional geologic setting, suggest that the source of DBT in the Gulf of Mexico
sediments is upward migrating petroleum. This study demonstrates that element-specific detection
(FPD) of aromatic sulfur compounds is a useful technique for recognizing petroleum as well as for
providing additional fingerprinting information for suggesting sources.
Fluorescence
Aromatic hydrocarbons have been analyzed by a number of techniques including ultraviolet (UV)
and fluorescence spectroscopy and gas chromatography. Spectroscopic methods are generally the
most rapid and inexpensive but suffer from varying degrees of nonspecificity and interferences. As
a screening tool, fluorescence analyses have been extensively used to detect microseepage and
macroseepage (Brooks, Kennicutt, and Carey 1986). Total scanning fluorescence provides
semiquantitative estimates of aromatic compound concentrations as well as an indication of ring
number distribution and has been discussed in detail elsewhere (Brooks, Kennicutt and Carey
1986; Kennicutt, Brooks, and Denoux 1986). Fluorescence and gas chromatographic analyses are

complementary analytical techniques that aid in the recognition of hydrocarbon sources. In figure
14.11, GC/FID analyses demonstrate the presence of petroleum-related normal alkanes in
sediment extracts from the Gulf of Mexico continental slope. Fluorescence analysis confirmed the
presence of petroleum-related aromatic hydrocarbons as well as perylene, an indigenous aromatic
hydrocarbon of uncertain origin. In the Barents Sea a combination of fluorescence fingerprinting and
gas chromatography/mass spectrometry confirmed the presence of petroleum-derived aromatic
hydrocarbons and demonstrated the similarity between oils produced in the area and the sediment
extracts (figure 14.12; Brooks, Kennicutt, and Carey 1986).
Gas Chromatography-Mass Spectrometry
Flame ionization detection, as described above, is relatively nonspecific and can suffer from
significant nonhydrocarbon interferences, though sample purification can alleviate some of these
problems. More definitive source information can be provided by gas chromatographic analysis with
detection by mass spectrometry.
Aromatic Hydrocarbons
Aromatic hydrocarbons are widely used as an indicator of petroleum contamination in

environmental samples. In general, environmental studies consider only polyaromatic hydrocarbons
consisting of condensed rings and simple alkylations (i.e., naphthalenes, phenanthrenes,
fluoranthenes, chrysenes, etc.). Other, more complex aromatic hydrocarbons such as aromatized
steroids are considered in the next section. Aromatics are ubiquitous in sedimentary environments
and can have multiple sources in petroleum, biosynthesis, early diagenesis, coal, combustion, and
immature/mature sediments (table 14.3). Molecular compositions based on parent compound, alkyl
homologue, and ring number distributions can be used to recognize the various sources. The use of
aromatic distributions has been extensively reviewed elsewhere and will not be repeated here (La
Flamme and Hites 1978; Hites et al. 1980; Wakeham, Schaffner, and Giger 1980a,b; and
references therein).
In a study of hydrocarbons in Gulf of Mexico Mississippi Fan Pleistocene sediments, two major
sources of hydrocarbons were recognized (1) terrestrial organic matter, transported to the site from
the continent, and (2) thermogenic hydrocarbons, migrating upward from much deeper in the
sediment column (Kennicutt et al. 1987b). The aromatic hydrocarbon, biological marker, and
aliphatic hydrocarbon distributions were interpreted as demonstrating the mature nature of the
hydrocarbons detected in these Fan and basin sediments. Aromatic hydrocarbons, as determined
by gas chromatography/mass spectrometry in a selected ion-monitoring mode, increased in
concentration with increasing depth (figure 14.13). Further supporting evidence by Marzi and
Rullkötter (1986), based on extractable organic matter concentrations, Rock-Eval analysis, n-alkane
patterns, and biological marker distributions for similar samples from Leg 96 DSDP/IPOD,
confirmed that migrated, mature hydrocarbons are present at these locations. The terrestrial
hydrocarbon component is primarily represented by normal alkane biowaxes with 21 to 33 carbons
(figure 14.14). Co-occurring with this transported material is a higher molecular weight aromatic
fraction represented by some portion of the phenanthrene, methyl phenanthrenes, and other less
volatile aromatic hydrocarbon compounds, which had a random distribution with depth. A secondary
recycled vitrinite particle population around Ro = 0.8% was observed by Marzi and Rullkötter
(1986), but the prominent maximum of unaltered primary vitrinite was in the range of 0.3 to 0.4%
Ro. Phenanthrene concentrations did not covary with terrigenous n-alkanes, suggesting a separate

detrital source for these thermogenic hydrocarbons. The major portion of thermogenic hydrocarbons
at these sites is represented by an envelope of normal and isoprenoid alkanes with fewer than 20
carbons, naphthalene, methylnaphthalene, ethylnaphthalene, and other aromatic hydrocarbon
compounds of similar volatility (figure 14.14). These volatile components are most likely not
recycled, since they would be substantially removed during transport to the present site of
deposition by weathering, evaporation, dissolution, and microbial degradation.
The presence of microseepage, or low levels of petroleum hydrocarbons in sediments, is often

difficult to unambiguously ascribe to a given source. Requejo and coworkers have pointed out that,
at the Leg 96 sites, in particular the intraslope basin sites, the EOM values are low and not
significantly above the biological debris background (Requejo, Whelan, and Boehm 1986; Whelan
1986). They interpret the mature hydrocarbon signatures as being reflective of material transported
from the continents by rivers and/or possible redeposition of seepage from other areas adsorbed
onto particulates in the water column. The presence of significant river-transported material is
supported by the discussion above, but the relative importance of the various processes is difficult
to resolve. The presence of highly volatile hydrocarbons can argue for upward migration, but the
degree of alteration of hydrocarbons exposed subareally is highly variable and dependent on
numerous environmental conditions, and no conclusive argument on the degree of alteration
expected is possible.
Saturated Hydrocarbons-Biological Markers
Techniques that directly detect the more complex hydrocarbon components of petroleum fluids
often provide the most direct link to hydrocarbon sources. Any source indicator should uniquely
identify the hydrocarbon fluid and be relatively resistant to alteration by near-surface phenomena
such as microbial degradation and differential solution in water as well as be recognizable from the
biological background. Living organisms consist of a relatively few structurally defined biochemicals
containing primarily carbon, hydrogen, oxygen, nitrogen, and sulfur. It is now widely accepted that
petroleum results from the action of temperature on the remains of organisms that have been
preserved and deposited in sedimentary rocks. One line of evidence for this is the wide array of
compounds present in petroleum that could only have a biological origin, i.e., porphyrins, steranes,
isoprenoids, etc. Any compound in petroleum that can be unambiguously linked to a heteroatomic,
precursor biochemical found in living organisms is considered to be a molecular fossil or "biological
marker." The progressive alterations caused by diagenesis and catagenesis can lead to the loss of
functional groups and unsaturations in the biochemical precursor until a fully saturated hydrocarbon,
consisting of carbon and hydrogen only, results (table 14.4).
Aliphatic hydrocarbon biological marker compounds are widely used in organic geochemistry for oil-
oil and oil-source rock correlations, as depositional environment indicators as well as indicators of
thermal history or maturation. Several extensive reviews of the occurrence, significance, and
methods of analysis of biological markers have appeared in recent years (Philp 1985a,b; and
references therein). For an extensive treatment of biological marker chemistry the reader is referred
to these reviews. The complex mixture of compounds that results from the exposure of organic
matter to elevated temperatures provides unique and sensitive fingerprints that can be used to
distinguish hydrocarbon mixtures. The advent of fused silica capillary gas chromatography and
mass spectrometry allows the analyst to resolve these complex mixtures into detailed fingerprints.
The interconversion of compounds within this fingerprint can be related to thermal history. Thus, the
composition of a hydrocarbon mixture can be used to distinguish mature, petroleum-related

compounds from immature or recent biological lipids (Dastillung and Albrecht 1976; Albaiges 1980;
Albaiges and Albrecht 1980; Jones et al. 1983, 1986; and references therein). Reactions thought to
be directly related to increasing thermal exposure are isomerization, aromatization, and bond
cleavage. Sterane and triterpane isomers are initially in the biological configuration (R) and
isomerize to the S isomer as they are heated. Monoaromatized steranes are aromatized to
triaromatized steranes, and long-chain compounds are thermally cracked to shorter chain
compounds. Though this simplistic approach is increasingly questioned, i.e., whether a product/
reactant relationship truly exists in these reactions, ratios developed from these relationships have
been demonstrably related to thermal alteration (Philp 1985b). Many relatively labile compounds, i.
e., alkenes, are also rapidly lost with burial (table 14.4).
The use of biological marker compounds to detect and fingerprint petroleum sources is based on
the recognition of representative groups of compounds that can be directly linked to a definable
thermal history. The earliest diagenesis (Ro < 0.35%) results in the loss of unsaturated compounds
such as sterenes and hopenes, as well as transformation of the more labile stereoisomers such as
hopanes (table 14.4). These compounds change with time and the and stereoisomers
become dominant early in the burial of a sediment. Steranes, 22S extended hopanes, and
aromatized steranes appear after some period of thermal exposure, and many of these compounds
are generated at thermal maturities that can occur only deep in the subsurface during petroleum
formation (exceptions include hydrothermal systems and magmatic intrusions). The co-occurrence
of mature biomarkers such as hopanes, 22S extended hopanes, 20S steranes, and diasteranes
with immature biomarker assemblages such as sterenes, hopenes, and hopanes in a recent
sediment can be interpreted only as representing a mixture of sources including nonindigenous
hydrocarbons. The most labile compound present in a mixture gives an indication of the highest
temperature to which the sediment could have been exposed.
At the macroseepage areas in the Gulf of Mexico, near-surface bitumen, sea slick, and tar ball
biological marker distributions were shown to be similar to oil reservoired deep in the subsurface
(Kennicutt, Brooks, and Denoux 1988; figure 14.15). Triterpane and sterane compositions were
similar for various fluids analyzed, though differences, probably due to microbial degradation, were
apparent. This represents the first evidence for tarball formation in association with deep-water
natural seepage. Biological marker distributions in Mississippi Fan and intraslope basin sediments
suggested the presence of mature, thermogenic hydrocarbons overprinted with immature biological
markers (Kennicutt et al. 1987b; figure 14.16). The immature overprint was recognized by the
abundance of hopanes, excess 22R extended hopanes, and excess 20R steranes. Mature
hydrocarbons were recognized by the presence of abundant hopanes, isomeric ratios of
hopanes and steranes near thermal equilibrium (except in cases of overprinting), the presence of
steranes and diasteranes (including and C27 to C29 steranes), and the abundance and
distribution of aromatized steranes. Biological marker distributions also suggested that the mature
hydrocarbons in Mississippi Fan sediments were atypical for known northern Gulf of Mexico
production, suggesting a separate source (figures 14.16 and 14.17). In particular, triaromatic
sterane distributions were strikingly different and the relative abundance of diasteranes was greatly
reduced in the sediment extracts relative to Gulf Coast oils.
Near-surface liquid hydrocarbon seeps near offshore southern California production were analyzed
for detailed molecular and isotopic information to determine whether the near-surface seepage was
similar to known production and thus presumably had a common source (Kennicutt and Brooks
1988). In this study the analysis of sediment hydrocarbons was used not only to determine the

presence of petroleum hydrocarbons but also to indicate the type and quality of deeper seated
hydrocarbons. Near-surface sediment bitumens collected over known production in the offshore
Santa Maria basin were shown to be similar to reservoired petroleum in the adjacent onshore
(figure 14.18). The prominent C28 bisnorhopane typical of California oils was ubiquitous. Triterpane,
sterane, and aromatized sterane compositional similarities between oils and near-surface sediment
bitumens suggest a common source in the Monterey Formation. The near-surface seeps were also
shown to be overprinted by in situ immature biological markers (figure 14.19). Increases in
petroleum-related parameters with increasing depth in the sediments suggest that a major source of
these compounds is upward migration. Concurrent geologic data confirm that this seepage is
preferentially associated with deep subsurface faults. Extensive redeposition of Monterey formation
in these areas through erosion and resuspension is also a potential source of material that cannot
be ruled out as a source of sediment hydrocarbons. Eganhouse and Kaplan (1988) also suggest
that the oil seepage into the overlying water might be adsorbed onto particulate matter and
transported to the sediment and thus redistribute seepage in these areas. The magnitude of these
effects is unclear. Clearly similar data can be interpreted as representing different processes,
including natural seepage, redistribution of seepage, and recycling of continental material. Even
with quantitative data the relative importance of these effects may be difficult to resolve unless
unique portions of the fingerprints can be linked to a specific hydrocarbon input.
A study by Kennicutt and Brooks (1990) detected aliphatic and aromatic hydrocarbons in sediments
at all locations sampled in the Bering Sea. The hydrocarbons were interpreted to be a mixture of
marine biological debris (bacteria, algae, zooplankton, phytoplankton), terrestrial plant biowaxes
(normal alkanes), "recycled" or exposed immature sediments, petroleum (natural seepage), and
combustion sources. The presence of a complete suite of normal alkanes and isoprenoids, an
unresolved complex mixture, petroleum-related PAHs (polynuclear aromatic hydrocarbons), mature
biological markers ( 22S hopanes), and vertical distributions of hydrocarbons confirms the
presence of petroleum at stations in the western and southern Bering Sea. Immature, pentacyclic
triterpanes ( ) found in near-surface sediments have been overprinted with mature 22S
hopanes (figures figure 14.20 and figure 14.21). This petroleum is most likely derived from natural
seepage from deep source rocks and/or reservoired fluids, though a recycled sediment or water
column tarball source cannot be ruled out.
References
Albaiges, J. 1980. Fingerprinting petroleum pollutants in the Mediterranean Sea. In Analytical

Techniques in Environmental Chemistry, pp. 69-81, Oxford: Pergamon Press.
Albaiges, J. and P. Albrecht, 1980. Fingerprinting marine pollutant hydrocarbons by

computerized gas chromatography-mass spectrometry. Int. J. Environ. Anal. Chem. 6:13-15.
Anderson, R. K., R. S. Scalan, P. L. Parker, and E. W. Behrens, 1983. Seep oil and gas in
Gulf of Mexico slope sediments. Science 222:619-621.
Blumer, M., M. Mullin, and R. Guillard. 1970. A polyunsaturated hydrocarbon (3, 6, 9, 12, 15,
18-heneicosahexaene) in the marine food web. Mar. Biol. 6:226-235.
Boehm, P. D. and A. G. Requejo, 1986. Overview of the Recent sediment hydrocarbon

geochemistry of Atlantic and Gulf Coast over continental shelf environments. Est. Coast.

Shelf Sci. 23:29-58.
Brassell, S. C., G. Eglinton, J. R. Maxwell, and R. P. Philp, 1978. Natural background of

alkanes in the aquatic environment. In O. Huntzinger, L. H., van Lelyveld, and B. C. J.,
Zoetman, eds., Aquatic Pollutants, Transformations and Biological Effects, pp. 69-86.
Brooks, J. M., M. C. Kennicutt II, and R. C. Carey, 1986. Strategies in offshore surface
geochemical exploration. Oil and Gas Journal 84:66-72.
Brooks, J. M., M. C. Kennicutt II, R. R. Fay, T. J. McDonald, and R. Sassen. 1984.

Thermogenic gas hydrates in the Gulf of Mexico. Science 225:409-411.
Brooks, J. M., H. B. Cox, W. R. Bryant, and M. C. Kennicutt II. R. G. Mann, and T. J.

McDonald. 1986. Association of oil seepage and gas hydrates in the Gulf of Mexico. Org.
Geochem 10:221-234.
Clark, Jr., R. and M. Blumer. 1967. Distribution of n-paraffins in marine organisms and
sediments. Limnol. Oceanogr. 12:79-87.
Dastillung, M. and P. Albrecht. 1976. Molecular test for oil pollution in surface sediment. Mar.
Poll. Bull. 1:13-15.
Eganhouse, R. P. and I. R. Kaplan. 1988. Depositional history of recent sediments from San
Pedro Shelf, California: reconstruction using elemental abundance, isotopic composition,
and molecular markers. Mar. Chem. 24:163-191.
Farrington, J. W. and B. W. Tripp. 1977. Hydrocarbons in western North Atlantic surface

Farrington, J. W., J. M. Teal, J. G. Quinn, T. L. Wade, and K. Burns. 1973. Intercalibration of

analyses of recently biosynthesized hydrocarbons and petroleum hydrocarbons in marine
lipids. Bull. Environ. Contamin. Toxicol. 10:129-146.
Farrington, J. W., S. G. Wakeham, J. B. Sarmiento, B. W. Tripp, and J. M. Teal. 1986.

Aromatic hydrocarbons in New York Bight polychaetes: Ultraviolet fluorescence analysis and
gas chromatography/gas chromatography-mass spectrometry analyses. Environ. Sci.
Technol. 20:69-72.
Goutx, M. and A. Saliot. 1980. Relationship between dissolved and particulate fatty acids
and hydrocarbons, chlorophyll a and zooplankton biomass in Villefranche Bay,
Mediterranean Sea. Mar. Chem. 8:299-318.
Hites, R. A., R. E. Laflamme, J. G. Windsor, Jr., J. W. Farrington, and W. G. Deuser. 1980.

Polycyclic aromatic hydrocarbons in an anoxic sediment core from the Pettaquamscutt River
(Rhode Island, USA). Geochim. Cosmochim. Acta 44:873-878.

Jones, D. M., S. J. Rowland, and A. G. Douglas. 1986. Steranes as indicators of petroleum-

like hydrocarbons in marine surface sediments. Mar. Poll. Bull. 17:24-27.
Jones, D. M., A. G. Douglas, R. J. Parkes, J. Taylor, W. Giger, and C. Schaffner. 1983. The
recognition of biodegraded petroleum-derived aromatic hydrocarbons in recent sediments.
Kennicutt II, M. C. 1988. The effect of biodegradation on crude oil bulk and molecular level
composition. Oil and Chemical Pollution 4:89-112.
Kennicutt II, M. C. and J. M. Brooks. 1988. Relation between shallow sediment bitumen and
deeper reservoired hydrocarbons, offshore Santa Maria basin, California, U.S.A. Appl.
Geochem. 3:573-582.
Kennicutt II, M. C., and J. M. Brooks. 1990. Hydrocarbons in Bering Sea sediments. Org.
Geochem. (in press).
Kennicutt II, M. C., J. M. Brooks, and R. A. Burke, Jr., 1989. Hydrocarbon seepage, gas
hydrates, and authigenic carbonate in the northwestern Gulf of Mexico. OTC 5952, 21st
Annual OTC, pp. 649-654.
Kennicutt, M. C. II, Brooks, J. M. and Denoux, G. J. 1986. Carbon isotope, gas

chromatography, and fluorescence techniques applied to the North Slope of Alaska
correlation study. In: Alaska North Slope Oil-Rock Correlation Study: Analysis of North Slope
Crude, (eds. Magoon, L. B. and Claypool, G. E.), AAPG Studies in Geology Series #20, pp.
639-650.
Kennicutt II, M. C., J. M. Brooks, and G. J. Denoux. 1988. Leakage of deep, reservoired
petroleum to the near surface on the Gulf of Mexico continental slope. Mar. Chem. 24:39-59.
Kennicutt II, M. C., J. M. Brooks, R. R. Bidigare, and G. J. Denoux. 1988b. Gulf of Mexico
hydrocarbon seep communities-I. Regional distribution of hydrocarbon seepage and
associated fauna. Deep-Sea Res., 35:1639-1651.
Kennicutt II, M. C., G. J. Denoux, J. M. Brooks, and W. A. Sandberg. 1987. Hydrocarbons in

Mississippi Fan and intraslope basin sediments. Geochim. Cosmochim. Acta 51:1457-1466.
Kennicutt, M. C., II, Sericano, J., Wade, T. L., Alcazar, F., and Brooks, J. M. 1987. High-
molecular weight hydrocarbons in Gulf of Mexico continental slope sediments. Deep-Sea
Res. 34(3):403-424.
Lacerda, C. P., M. C. Kennicutt II, and J. M. Brooks. 1987. The distribution of

dibenzothiophenes in Gulf of Mexico sediments. Applied Geochem. 2(3):297-304.
La Flamme, R. E. and R. A. Hites. 1978. The global distribution of polycyclic aromatic

hydrocarbons in Recent sediments. Geochim. Cosmochim. Acta 42:289-303.
Marzi, R. and J. Rullkötter. 1986. Organic matter accumulation and migrated hydrocarbons

in deep sediments of the Mississippi fan and adjacent intraslope basins, Northern Gulf of
Mexico. In E. T. Degens, E. T. Myers, and S. C. Brassess, eds., Biogeochemistry of Black
Shales, pp. 359-379. Mitt. Geol.-Paleontol. Inst. Univ. Hamburg 60.
Philp, R. P. 1985a. Biological markers in fossil fuel production. Mass Spectrometry Rev. 4:1-
54.
Philp, R. P. 1985b. Fossil Fuel Biomarkers: Application and Spectra. Methods in

Geochemistry and Geophysics, vol. 23. New York: Elsevier.
Requejo, A. G., J. K. Whelan, and P. D. Boehm. 1986. Hydrocarbon geochemistry and

biological markers in Orca and Pigmy Basin sediments (Sites 618 and 619). In A. H. Bouma,
J. M. Coleman, A. W. Meyer, et al. eds., Initial Reports: DSDP 96:785-793. Washington, D.
Sweeney, R. E., R. I. Haddad, and I. R. Kaplan, 1980. Tracing the dispersal of the IXTOC-I
oil using C, H, S, and N stable isotope ratios. In D. K. Atwood, ed., Proc. Symp. on
Preliminary Results from the September 1979 Researcher/Pieree IXTOC-I Cruise pp. 89-
115. U.S. National Oceanic and Atmosphere Administration, Washington, D.C.
Wakeham, S. G., C. Schaffner, and W. Giger. 1980a. Polycyclic aromatic hydrocarbons in

Recent lake sediments-I. Compounds having anthropogenic origins. Geochem. Cosmochim.
Acta 44:403-413.
Wakeham, S. G., C. Schaffner, and W. Giger. 1980b. Polycyclic aromatic hydrocarbons in

recent lake sediments-II. Compounds derived from biogenic precursors during early
Whelan, J. K. 1986. Geochemistry Summary Leg 96-The Mississippi Fan. In A. H. Bouma, J.

M. Coleman, A. W. Meyer, et al., eds. Initial Reports: DSDP 96:691-695. Washington, D.C.:


15. Biomarkers in Recent and Ancient Sediments:he

Importance of the Diagenetic Continuum
Precursor/product relationships for biomarkers are typically deduced or inferred from a combination
of information, principally the structural comparability of specific compounds, the patterns of their
occurrence in sediments of different maturity, and the viability of their diagenetic conversion as
demonstrated by laboratory simulation experiments. For many proposed relationships the evidence
from structural and stereochemical considerations is persuasive, even where the maturity range of
the sediments containing the compounds is disparate rather than contiguous and also where the
postulated precursor/product transformation is neither firmly established nor adequately validated.
For example, -carotane, botryococcane, and lycopane are generally held to derive from -
carotene, botryococcene, and lycopene, respectively, but a diagenetic pathway that explicitly
explains the process for reduction of all of the double bonds in these molecules has yet to be
confirmed.
Given such concerns the examination of successive changes in biomarker distributions in immature
sediment sequences can play a vital role in the elucidation and verification of such precursor/
product relationships. For example, discrepancies between the observed occurrences of 4-
methylsteranes in ancient marine and freshwater sediments and those of their 4-methylsterol
counterparts in contemporary sediments suggest that the dinoflagellate marker sterol dinosterol is
not a precursor of dinosteranes. This conclusion implies that 22 sterols are not reduced to their
saturated counterparts but experience an alternative fate, such as transformation into aromatic
steroidal hydrocarbons, incorporation into insoluble organic matter, or interaction with sulfur and
subsequent liberation as steroids with short side chains.
Significant factors in the preservation of sedimentary organic matter are the scope, nature, and
selectivity of diagenetic processes. A wide variety of transformation pathways for individual
biomarkers, their homologous series, and structural classes has been proposed. Typically, these
can involve a diverse range of chemical reactions, including loss or alteration of functional groups,
structural modification or rearrangement, stereochemical configurational changes, or aromatization
(e.g., Ensminger et al. 1974; Rubinstein, Sieskind, and Albrecht 1975; Laflamme and Hites 1979;
Seifert and Moldowan 1980; Wakeham, Schaffner, and Giger 1980; Mackenzie et al. 1982; Hussler
and Albrecht 1983; Brassell et al. 1984; Mackenzie 1984; Brassell 1985; de Leeuw and Baas 1986;
Simoneit et al. 1986; Volkman and Maxwell 1986; Wolff, Lamb, and Maxwell 1986; Peakman and
Maxwell 1988; Volkman 1988). During the earliest phases of diagenesis, including effects occurring
in the water column, such processes appear to be mediated by biological activity; subsequently they
are primarily driven by thermodynamic constraints as their burial temperature increases.
In some instances gaps or discrepancies exist between the constituents of organisms and their
attributed products in mature sediments and petroleums. These hiatuses in the geological record of
biological markers may arise from (1) anomalies or inconsistencies in the reported occurrences of

the compounds, or (2) the incorrect assignment of precursors to product relationships, or (3) the
limited information on the relationships between extractable compounds and their analogues bound
in kerogen, or (4) the incompleteness of the proposed diagenetic transformation reaction(s), often
attributable to a lack of knowledge of the detailed pathways and processes involved. All four factors
benefit from improvements in the documentation of the geological occurrences of biological markers
in soluble and insoluble organic matter and from assessment of the continuity of their sedimentary
records. Several of these aspects of the diagenetic alteration of biological markers are examined
below, with specific reference to selected transformations of isoprenoids and steroids.
Precursor-to-Product Relationships: Confirmed and Unconfirmed
Ideally, precursor/product relationships should be based on a combination of: (1) the structural
comparability and compatibility of the components (e.g., Albrecht and Ourisson 1971; Maxwell,
Pillinger, and Eglinton 1971; Ensminger et al. 1974; Mackenzie et al. 1982; Brassell, Eglinton, and
Maxwell 1983; Mackenzie 1984; Volkman 1988); (2) evidence for their interconversion via
microbially mediated transformations (e.g., Gaskell and Eglinton 1975; Taylor, Smith, and Gagosian
1981) or attributable to effects associated with food web processes, for example, selective uptake
and alteration during grazing (e.g., Prahl et al. 1984; Harvey et al. 1987); (3) verification of the
possibility of their diagenetic interconversion by laboratory simulation experiments involving
chemical treatments (e.g., Peakman and Maxwell 1988), or pyrolysis at elevated temperatures (e.g.,
Van Dorsselaer, Albrecht, and Connan 1977), perhaps in the presence of appropriate catalysts
such as clays (e.g., Rubinstein, Sieskind, and Albrecht 1975; Riolo and Albrecht 1985; Larcher,
Alexander and Kagi 1988) or sulfur (e.g., Abbott, Lewis, and Maxwell 1985); (4) results from the
studies of biomarkers released as pyrolysis products from discrete precursors (e.g., Goossens et al.
1984), or from macromolecular biopolymers (e.g., Goth et al. 1988), or by chemical treatment of
insoluble organic matter (e.g., Chappe, Michaelis, and Albrecht 1980; Mycke and Michaelis 1986);
(5) empirical evidence from direct comparison of their occurrences in uniform sequences of
sediments of different maturity (e.g., Mackenzie et al. 1980; Mackenzie, Hoffmann, and Maxwell
1981; Hussler and Albrecht 1983; Rullkötter and Welte 1983; Brassell et al. 1984; Brassell 1985;
Aquino Neto et al. 1986; Noble, Alexander, and Kagi 1987); and (6) consistency with
thermodynamic considerations of their comparative thermal stability (e.g., Seifert and Moldowan
1979), including molecular mechanics calculations (e.g., van Grass et al. 1982; de Leeuw et al.
1989).
Given that such a complete set of information is rarely available, precursor/product relationships
and diagenetic transformations are typically based on structural comparisons and on generally
accepted sedimentary reaction processes, such as defunctionalization and reduction. Often,
subsequent evidence confirms and substantiates such proposals, as with the formation of
diasterenes (XII; Rubinstein, Sieskind, and Albrecht 1975) and diasteranes (XVI; Ensminger, Joly,
and Albrecht 1978) from sterols via 4- and 5-sterenes (VIII and IX, respectively; figure 15.1). In
some cases, however, conflicting evidence emerges. For example, initial interpretations of sterane
diagenetic pathways (figure 15.1) postulated that 5 (H), 14 (H), 17 (H)-steranes (XVII) were
products of the isomerization of 5 (H),14 (H),17 (H)-steranes (XIII; e.g., Seifert and Moldowan
1979; Mackenzie et al. 1980, 1982), an interpretation supported by the generation of 5 (H),14
(H),17 (H)-cholestanes among the isomerization products of 5 (H),14 (H),17 (H)-cholestane
(Seifert and Moldowan 1979). However, sedimentary sterane distributions include many examples
in which 5 (H),14 (H),17 (H)-steranes occur as 20S and 20R isomers in similar abundance
whereas the accompanying 20R 5 (H),14 (H),17 (H)-steranes greatly dominate their 20S

counterparts (ten Haven et al. 1986). Such differences in the extent of isomerization at C-20
between the two structural isomers at the same level of maturity seem to refute the possibility of
their direct diagenetic interconversion and hence their relationship as precursors and products,
unless the process involved also results in isomerization and equilibration at C-20. It is now evident
that both 20S and 20R 5 (H),14 (H),17 (H)-steranes and 20S 5 (H),14 (H),17 (H)-steranes can
be formed from 7-sterol (III) precursors via 2,8(14)- and 2,14-steradiene, 8(14)- and 14-sterene,
and spirosterene intermediates (VI, VII, X, XI, and XIV, respectively; figure 15.1; ten Haven et al.
1986; Peakman and Maxwell 1988; Peakman et al. 1989) at a much earlier stage of diagenesis
than that required for the isomerization of 5 (H),14 (H),17 (H)-steranes at C-20. This interpretation
provides an explanation for the origin of 5 (H),14 (H),17 (H)-steranes that is more consistent with
the geological occurrences and distributions of steranes, including differences between the relative
abundance of the carbon number homologues of 5 (H),14 (H),17 (H)-steranes and 5 (H),14
(H),17 (H)-steranes in sediments and petroleums where they cannot be attributed to mixed sources
(e.g., Mackenzie et al. 1982; Shi Ji-Yang et al. 1982).
Molecular mechanics calculations (de Leeuw et al. 1989) suggest that the transformation of 2-
sterenes (V) into 4- and 5-sterenes (VIII,IX, respectively; figure 15.1) proposed on the basis of
their sedimentary occurrences and laboratory simulations (e.g., Rubinstein, Sieskind, and Albrecht
1975; Mackenzie et al. 1982; Brassell et al. 1984; Brassell 1985) is unlikely; rather, 2-sterenes
should be reduced directly to 5 (H),14 (H),17 (H)-steranes (XIII; figure .15.1). This modification of
generally accepted diagenetic pathways suggests, in turn, that 5 (H)-stanols (II), which are the
recognized precursors of 2-sterenes (e.g., Dastillung and Albrecht 1977; Gagosian and Farrington
1978; Mackenzie et al. 1982; Brassell et al. 1984), cannot undergo transformation to diasterenes
(XII), unlike 5-sterols (I; figure .15.1), which are the diagenetic precursors of 3,5-steradienes (IV; e.
g., Gagosian and Farrington 1978; Mackenzie et al. 1982) and thence, apparently, of 4- and 5-
sterenes (VIII, IX, respectively; de Leeuw et al. 1989). In general, such improvements in the
understanding of steroid diagenesis can help explain the variations in their distributions among
different compound classes (cf. Brassell and Farrimond 1986; Curiale 1987). In addition, molecular
mechanics calculations for sterenes (de Leeuw et al. 1989) suggest that 2-sterenes should co-
occur with equilibrium mixtures of 4- and 5-sterenes, especially in immature sediments, which
contain no diasterenes. Despite several previous reports to the contrary, reexamination of Neogene
marine sediments, including samples from the Cariaco Trench, Japan Trench, and Walvis Ridge,
reveals that mixtures of 2-, 4- and 5-sterenes are indeed present in these comparatively
immature sediments in addition to their documented co-occurrence in Cretaceous black shales
recovered by the Deep Sea Drilling Project (Brassell and Farrimond 1986; Rullkötter,
Mukhopadhyay, and Welte 1987).
These cases confirm the need for caution in proposing precursor-to-product relationships and also
demonstrate that seemingly established diagenetic pathways may require revision as new
information emerges. Also, direct structural comparability between precursors and products cannot
be assumed, as illustrated by the recent recognition that taraxer-14-ene (XVIII; figure 15.2) can
undergo diagenetic transformation and rearrangement to olean-12-ene (XIX) and, hence, to olean-
13(18)-ene (XX) and olean-18-ene (XXI; ten Haven and Rullkötter 1988). This structural change
involving methyl migration demonstrates the potential for interconversion of specific structurally
related, yet different, biological marker skeletons.
Proposed diagenetic pathways include examples where both precursors and products are well
documented but where the alteration process remains unclear-a consideration that is addressed

below in discussion of selected isoprenoid alkanes. Other constituents of sediments appear to be

diagenetic products derived from as yet unknown precursors. An example of a compound in the
latter category is 17 (H),21 (H)-pentakishomohopanoic acid (XXII; figure 15.3) which has been
recognized as a constituent of immature sediments (Brassell, Eglinton, and Maxwell 1983). The
formation of this component from the best described C35 hopanoid precursor, bacteriohopanetetrol
(XXIII), seems chemically improbable, since it requires dehydration of three hydroxyl groups (at C-
32, C-33, and C-34) and oxidation of the fourth (at C-35). Perhaps its sedimentary occurrence can
be taken to suggest that there are other, less functionalized, biological precursors for sedimentary
C35 hopanoids. Other sedimentary components, such as sterol ethers (e.g., 3-nonyloxycholest-5-
ene, XXIV; figure 15.3; Boon and de Leeuw 1979; Brassell et al. 1980a, b; Brassell, Eglinton, and
Maxwell 1983) and A-nor sterones (XXV; figure 15.3; McEvoy and Maxwell 1986), cannot be readily
explained as the diagenetic products of recognized precursors (Brassell and Eglinton 1986).
Perhaps they are unidentified biosynthetic constituents and the inability to relate them to their
biological source reflects the paucity of data on the lipid compositions of environmentally significant
species. Certainly, the number and range of similar "biomarker orphans" in the geochemical
literature illustrate that there are undoubtedly numerous precursors of sedimentary compounds that
have yet to be identified, or recognized, among the lipid constituents of living organisms. Progress
in confirming the biological source(s) of biomarkers and their reaction intermediates continues. For
example, tetrahymenol (XXVI; figure 15.4) has long been proposed as the biological precursor of
gammacerane (XXVIII), first identified in a lacustrine oil shale (Hills et al. 1966) and subsequently
found to occur in marine sediments and petroleums (Moldowan, Seifert, and Gallegos 1985; Mello
et al. 1988a, b) and to be especially abundant in sediments laid down in saline depositional
environments and in their derived petroleums (Moldowan et al. 1985; ten Haven, de Leeuw, and
Schenck 1985; Fu et al. 1986; Brassell et al. 1988; ten Haven et al. 1988; Mello et al. 1988a, b).
Only recently, however, has tetrahymenol been recognized as a widespread component of
immature marine sediments (ten Haven et al. 1989; Venkatesan 1989), together with its dehydration
product gammacer-2-ene (XXVII; ten Haven et al. 1989), which provides substantive evidence of
the precursor/product link between tetrahymenol and gammacerane.
Sedimentary Isoprenoid Alkanes: Origins and Occurrences
There are two principal ways to account for the presence of sedimentary alkanes: either as direct
contributions of compounds biosynthesized by organisms or as diagenetic products formed by
microbial or sedimentary transformation reactions, including compounds generated from biological
debris or biopolymers bound into insoluble matter (e.g., Goth et al. 1988). There are instances,
however, when neither of these possibilities appears, at present, to satisfactorily explain the
occurrence of particular compounds. For example, no direct confirmation of the precursor-to-
product transformation pathway has yet been made for several sedimentary isoprenoid alkanes,
notably botryococcane, -carotane, and lycopane (XXX, XXXIV, and XXXVI, respectively;
figure .15.5), despite their evident structural relationship to biologically occurring terpenoid
polyenes.
The high degree of structural specificity of botryococcane (XXX) and its homologues (e.g., XXXII)
provides compelling evidence that their presence in petroleums (Moldowan and Seifert 1980; Seifert
and Moldowan 1981; Moldowan, Seifert, and Gallegos 1985; McKirdy et al. 1986) and sediments
(Brassell, Eglinton, and Fu 1986; Curiale 1987) can be attributed to contributions from the green
alga Botryococcus braunii, which is the only organism known to biosynthesize the structurally
analogous alkene botryococcene (XXIX; Maxwell et al. 1968; Cox et al. 1973; Metzger et al. 1985a,

b; figure 15.5) and its homologues (e.g., XXXI; Metzger and Casadevall 1983; Metzger et al.
1985a). However, the reduction of botryococcene to botryococcane requires a series of specific
hydrogenations at six positions of unsaturation. To date, no partially reduced intermediates with
botryococcane-type structures have been observed among the extractable hydrocarbons of
immature sediments, suggesting that any such hydrogenation process must be concerted, or
proceed via functionalized components that are not hydrocarbons, or involve insoluble organic
matter. The highly aliphatic biopolymers of freshwater algae can act as biological sources for n-
alkanes with varying carbon number distributions (Largeau et al. 1984, 1986; Goth et al. 1988),
illustrating the potential for alkane generation from such sources. However, the B race of B. braunii,
which contains botryococcene and its homologues among its extractable hydrocarbons, does not
appear to incorporate these acyclic isoprenoid alkenes within its biopolymer (Laureillard, Largeau,
and Casadevall 1988). Furthermore, such resistant biopolymeric material is unlikely to break down
during early diagenesis liberating, and thereby accounting for, homologues of botryococcane in
immature sediments (Brassell, Eglinton, and Fu 1986; Curiale 1987).
-carotane (XXXIV) occurs as a prominent component of many lacustrine sediments and

petroleums of different ages (e.g., Murphy, McCormick, and Eglinton 1967; Shi Ji-Yang et al. 1982;
Hall and Douglas 1983; Jiang and Fowler 1986; Brassell et al. 1988; Duncan and Hamilton 1988;
Fu, Sheng, and Liu 1988; Jiang, Philp, and Lewis 1988; Peters et al. 1989; Yang Bin and Yang
Jianqiang 1989) but has also been observed in some marine environments (Mello et al. 1988a, b).
Its most logical and likely precursor is -carotene (XXXIII, figure 15.5; Murphy, McCormick, and
Eglinton 1967; Chou and Wood 1986). However, this reaction requires selective hydrogenation of
an even greater number of double bonds than reduction of botryococcene to botryococcane. In
addition, as for botryococcane, no partially reduced carotenes have been reported in immature
sediments.
Various explanations can be offered to account for the occurrences of alkanes in sediments that
cannot be readily explained by known diagenetic reactions. First, alkanes that occur as minor
constituents in living organisms may become concentrated by the selective bacterial degradation of
more abundant alkene analogues (cf. Robson and Rowland 1988). Such a process could occur
during passage of the compounds through the food web or during transportation through the water
column. Alternatively, the selective reduction of alkenes could be effected by microbes during very
early stage diagenesis, either prior to sediment deposition or in the upper few centimeters of wet
unconsolidated sediment.
Second, the alkanes may be biosynthesized by their source organisms during growth stages that
are rarely evaluated, for example, during algal senescence. Indeed, the lipid biochemistry of
organisms in culture may differ significantly from that of their natural populations, thereby limiting
the value of culturing experiments as definitive indicators of lipid compositions of particular biota.
Also, many organisms exhibit changes, sometimes systematic, in their lipid composition during their
growth cycle, or in response to differences in environmental factors, such as variations in salinity,
temperature, or nutrient supply (e.g., Harwood and Russell 1984). Hence, the hydrocarbon
compositions of organisms, especially algae, can be expected to depend on environmental or
experimental conditions, and no single analysis or series of analyses can provide a wholly
representative indication of the range of lipids that they produce.
Third, the alkanes may be directly synthesized by living organisms but have yet to be detected
among their constituents. For example, the occurrence of squalane (XXXIX; figure 15.5) in Nigerian
petroleums (e.g., Gardner and Whitehead 1972) was attributed to geochemical reduction of

squalene (XXXVIII) prior to its recognition as a constituent of archaebacteria (e.g., Holzer, Oro, and
Tornabene 1979) and as a component of recent and immature sediments (e.g., Brassell et al.
1981). Direct biosynthesis of lycopane (XXXVI; figure 15.5) in archaebacteria, specifically
methanogenic bacteria, has been suggested (Brassell et al. 1981) to explain its occurrence in a
wide variety of recent and ancient marine (e.g., Dastillung and Corbet 1975; Brassell et al. 1981;
Brassell, Maxwell, and Eglinton 1981; Comet et al. 1981; McEvoy, Eglinton, and Maxwell 1981;
Thomson et al. 1982; von der Dick, Rullkötter, and Welte 1983; Brassell 1984; Brassell and
Farrimond 1986; Volkman and Maxwell 1986; Farrimond et al. 1988, 1989; Farrington et al. 1988)
and lacustrine (Kimble et al. 1974) sediments. However, the abundance of lycopane in a marine
water column containing substantial concentrations of sulfate is intriguing since methanogens would
not be expected to be present under such environmental conditions (Wakeham and Ertel 1988).
The recognition of lycopa-14E,18E-diene (XXXVII; figure 15.5) as the dominant hydrocarbon of the
L race of Botryococcus braunii (Metzger and Casadevall 1987; Derenne et al. 1989) provides the
possibility for formation of lycopane from a green alga via a markedly less extensive series of
geochemical reductions. It is perhaps significant that unsaturated analogues of lycopane have been
recognized in both lacustrine and marine sediments, although the number and sites of their
unsaturation were not determined (Kimble et al. 1974; Wardroper 1979). The possibility of
determination of carbon isotopic measurements for individual molecules (Hayes et al. 1987) would
permit the distinction of lycopane derived from algal and methanogenic bacterial sources since the
latter would be strongly depleted in 13C. Indeed, the application of such techniques reveals a
carbon isotopic signature for lycopane (-20.8 ) in the Messel shale, which suggests that it is of
algal origin (Freeman et al. 1990). Similar investigations of the carbon isotopic characteristics of
biosynthetic lipids and sedimentary biomarkers should aid the recognition of the biological sources
of the latter and the elucidation of precursor/product relationships linking biologically and
geologically occurring organic compounds.
Finally, a further possibility to explain the occurrence of isoprenoid alkanes in sediments is the
existence of biological or geochemical processes, as yet unrecognized, which effect the efficient
and concerted (and stereospecific?) hydrogenation of polyenes in sediments. It is noteworthy that
both botryococcane and -carotane typically occur in lacustrine sediments and their derived
petroleums (e.g., Murphy, McCormick, and Eglinton 1967; Moldowan and Seifert 1980; Seifert and
Moldowan 1981; McKirdy et al. 1986; Shi Ji-Yang et al. 1982; Jiang and Fowler 1986; Brassell et al.
1988; Duncan and Hamilton 1988; Jiamo Fu et al. 1988; Jiang, Philp, and Lewis 1988; Mello et al.
1988a,b; Peters et al. 1989; Yang Bin and Yang Jianqiang 1989), perhaps suggesting a link
between the hydrogenation process and depositional environment. Recent results suggest that
interaction with sulfur, perhaps cross linking at sites of unsaturation, may play a significant role in
the formation of -carotane from -carotene and in the generation of alkanes. This interpretation is
supported by the liberation of -carotane as a major alkane among the products of desulfurization
(Raney Ni) of the polar fraction of bitumen from a Miocene marl (Sinninghe Damsté et al. 1990) and
of an insoluble, highly aliphatic sulfur cross-linked polymer isolated from sulfur-rich petroleums
(Mycke, Schmid, and Albrecht, unpublished data). The other products of the desulfurization
treatment of the petro-leum include lycopane and numerous other acyclic isoprenoid alkanes
(Mycke, Schmid, and Albrecht, unpublished data). The comparative weakness of carbon-sulfur
bonds suggests that such a polymer might readily break down with increasing maturity, liberating -
carotane and a suite of acyclic isoprenoids. This process can help explain the occurrence of such
isoprenoid alkanes in petroleums but is less satisfactory as an explanation for their presence in
comparatively immature sediments.
The interaction of sulfur with alkenes in polymeric material may account for the occurrences of

alkanes, including n-alkanes, derived from functionalized precursors, notably alkenes, in sulfur-rich
systems, especially iron-limited hypersaline environments (e.g., Sinninghe Damsté et al. 1986,
1987; Schmid, Connan, and Albrecht 1987). This process of sulfur incorporation and subsequent
liberation of alkanes can also occur in environments of normal marine salinity. Specifically, the
enhanced abundances of C37 and C38 n-alkanes in ancient sediments and petroleums, including
Monterey shales (McEvoy et al. 1981), their hydrous pyrolysates (Sofer 1988), and their derived
petroleums (Zumberge 1987), may derive from reduction of thiolanes and thianes (e.g., XL, XLI,
respectively; figure 15.6) formed by sulfur incorporation into their precursor alkenes (Sinninghe
Damsté et al. 1989c). In general, the wide variability in the sulfur contents of sediments and
petroleums suggests that the process for alkane formation via organosulfur intermediates may not
be widely applicable but occurs only for specific compounds with appropriate structural
characteristics (i.e., unsaturation) in a limited range of depositional environments. Evidence for the
formation of organic sulfur compounds in freshwater lacustrine environments is poorly documented,
although acyclic isoprenoid thiophenes (e.g., XLII; figure 15.6; Brassell et al. 1986b) do occur
(Cranwell, personal communication). It remains uncertain, therefore, whether such a process can
provide an explanation for the formation of botryococcane and its homologues, given that they are
present in freshwater to brackish lacustrine sediments and in their derived petroleums with low
sulfur contents (e.g., Moldowan and Seifert 1980; Seifert and Moldowan 1981; Brassell, Eglinton,
and Fu 1986a; McKirdy et al. 1986; Curiale 1987).
Overall, the differences in the geochemical occurrences and characteristics of botryococcane, -

carotane and lycopane suggest that there is no general solution that can explain all the processes
of alkene-to-alkane transformations in sediments. The diverse possibilities for the formation of
isoprenoid alkanes, and other significant classes of biomarkers, also demonstrate the fundamental
and unresolved problems that complicate efforts to understand their geochemical origin. However,
investigation of the carbon isotopic signatures of such biomarkers offers a novel and useful tool in
the clarification of the biological origins and precursor-to-product transformations of such
compounds.
4-Methylsteroids and 22-Steroids
The 4,23,24-trimethylcholestanes (XLIIIa; figure 15.7) appear to occur only in significant amounts in
marine sediments and petroleums (Summons, Volkman, and Boreham 1987; Goodwin, Mann, and
Patience 1988), yet their presumed precursor dinosterol (4,23,24-trimethyl-5 (H)-cholest-22-en-3 -
ol, XLIVb; e.g., Summons, Volkman, and Boreham 1987) is a prominent constituent of both marine
(e.g., Boon et al. 1979; Wakeham et al. 1980; Brassell, Maxwell, and Eglinton 1981; Brassell and
Eglinton 1981, 1983a, b, 1986; McEvoy, Eglinton, and Maxwell 1981; Thomson et al. 1982; Smith et
al. 1982; de Leeuw et al. 1983; McEvoy and Maxwell 1983; Smith, Eglinton, and Morris 1983, 1986;
Shaw and Johns 1985, 1986; ten Haven et al. 1987a-d; Venkatesan and Kaplan 1987; Volkman,
Farrington, and Gagosian 1987; Brault and Simoneit 1988) and freshwater (e.g., Mermoud et al.
1982; Cranwell 1984; Robinson et al. 1984a, b, 1986; Cranwell, Eglinton, and Robinson 1987;
Robinson, Eglinton, and Cranwell 1987; Goossens et al. 1989) systems. This discrepancy strongly
suggests that dinosterol is not the precursor of sedimentary 4,23,24-trimethylcholestanes. The lack
of evidence for the diagenetic conversion of dinosterol to dinosteranes raises questions about the
fate of 22-4-methylsterols (XLIVb-e) since it suggests that they do not undergo diagenetic
conversion to their saturated counterparts. In addition, it seems likely that 22-sterols (XLVb-g,
XLVIb-g) would not be transformed into their sterane counterparts given the comparability in the
diagenetic pathways of 4-methylsteroids and steroids (e.g., Mackenzie et al. 1982), except where

the methyl substitution at C-4 exerts an influence on the comparative stability of diagenetic
intermediates or products, such as with 4- and 5-4-methylsterenes and 4 - and 4 -
methyldiasterenes (LIII, LIV, LV, and LVI, respectively; figure .15.8; Wolff et al. 1986; Peakman and
Maxwell 1988).
Survival of 4,22- and 5,22- steradienes (XLVIIc-e, XLVIIIc-e; figure 15.7; e.g., Giger and Schaffner
1981; Rullkötter, von der Dick, and Welte 1981, 1982; von der Dick et al. 1983; Rullkötter,
Mukhopadhyay, and Welte 1984, 1987; Rullkötter et al. 1984, 1986; Farrimond, Eglinton, and
Brassell 1986) and of 22 B-ring monoaromatic anthrasteroids (e.g., XLIXd; Brassell, unpublished
data) demonstrates that such components survive early diagenetic processes with their side chain
unsaturation intact. The comparability between steroidal and 4-methylsteroidal diagenetic alteration
pathways is illustrated by the retention of 22 unsaturation in 4-methylsteroids through early
diagenetic processes. For example, dinostera-4,22-diene (4,23,24-trimethylcholesta-4,22-diene; Lb;
figure 15.7) occurs in Cretaceous marine black shales (Comet et al. 1981; Brassell and Farrimond
1986), and dinoster-22-ene (4,23,24-trimethylcholest-22-ene; XLIIIb; figure 15.7) has been reported
in Miocene sediments from the California Borderland (McEvoy, Eglinton, and Maxwell 1981;
McEvoy and Maxwell 1983).
22 unsaturation in sterenes is ultimately lost during diagenesis. The distributions of 4- and 5-

sterenes (VIII, IX; figure 15.1) and 4,22-steradienes (XLVIIc-e; figure 15.6) found in immature
Monterey shales are not apparent in more mature samples containing diasterenes (XII; figure 15.1;
Giger and Schaffner 1981). A similar difference in sterene distributions with increasing maturity is
observed for many marine sediments recovered by the Deep Sea Drilling Project (McEvoy and
Maxwell 1983; Brassell et al. 1984; Rullkötter, Mukhopadhyay, and Welte 1984; Rullkötter, et al.
1984; Brassell 1985; Farrimond, Eglinton, and Brassell 1986). However, these observations do not
provide direct evidence for saturation of 22-sterenes, raising the question of their diagenetic fate.
Possible fates for 22-steroids and 22-4-methylsteroids include: (1) migration into ring C of the
steroid nucleus and formation of ring C monoaromatic steroidal hydrocarbons (LVII; figure 15.8;
Moldowan, Sundararaman, and Schoell 1986; Peakman 1986) (2) interaction with sulfur to form
steroidal thiophenes and thiolanes (e.g., LVIII, LIX; figure .15.8; Sinninghe Damsté et al. 1989a,b),
though this process may be restricted to sulfur-rich systems; (3) incorporation into kerogen via C-22
or C-23; and (4) oxidation, though this seems a more plausible explanation for the degradation of
22-steroids in the water column than for the apparent disappearance of 22-sterenes in immature
sediments. The last three of these potential fates for 22-steroids (i.e., interaction with sulfur,
incorporation into kerogen, and oxidation) all offer the possibility for the formation of short side chain
sterols (e.g., pregnane; LX; figure 15.8) as degradation products of 22-steroids. In several respects
this explanation seems more satisfactory than the suggestion that these steranes are diagenetic
products of short side chain 5 (H)-stanols (e.g., XLVh-k; figure 15.7), given the comparatively few
reports of their occurrence in sediments (e.g., Brassell et al. 1980a,b; Brassell and Eglinton 1981).
Whatever the precise diagenetic fate of 22-sterols (XLVb-g, XLVIb-g; figure 15.7), it clearly
represents a significant pathway for steroidal compounds given their abundance, variety, and
apparently ubiquitous occurrence in marine sediments (e.g., Lee, Gagosian, and Farrington 1977,
1980; Huang and Meinschein 1978; Wardroper, Maxwell, and Morris 1978; Boon et al. 1979;
Gagosian, Lee, and Heiner 1979; Lee, Farrington, and Gagosian 1979; Brassell et al. 1980a,b,
1981; Gagosian et al. 1980; Thomson et al. 1982; Smith et al. 1982; Brassell and Eglinton 1983a,b,
1986; de Leeuw et al. 1983; McEvoy and Maxwell 1983; Smith, Eglinton, and Morris, 1983, 1986;

Shaw and Johns 1985, 1986; ten Haven et al. 1987a-d; Venkatesan and Kaplan 1987; Volkman,
Farrington, and Gagosian 1987; Brault and Simoneit 1988), water column particulates (Wakeham et
al. 1980, 1984; Gagosian, Smith, and Nigrelli 1982; Gagosian, Volkman, and Nigrelli 1983;
Gagosian, Nigrelli, and Volkman 1983; Wakeham 1987; Wakeham and Canuel 1988; Wakeham
and Lee 1989), and lacustrine sediments (Gaskell and Eglinton 1976; Nishimura and Koyama 1976;
Nishimura 1977; Matsumoto, Torii, and Hanya 1982; Mermoud et al. 1982; Cranwell 1984;
Robinson et al. 1984a,b, 1986; Cranwell, Eglinton, and Robinson 1987; Robinson, Eglinton, and
Cranwell 1987; Wü#che, Gülaçar, and Buchs 1987; Wü#che, Mendoza, and Gülaçar 1988;
Goossens et al. 1989). A similar fate may be expected for 22-sterones and 22-steryl esters (e.g.,
LIb-e, LIIb-e, respectively; figure 15.8), which are also significant steroidal constituents in both
marine (e.g., Gagosian and Smith 1979; Wakeham et al. 1980; Comet et al. 1981; Gagosian et al.
1982; Smith et al. 1982; Wakeham 1982, 1987; Wakeham and Frew 1982; McEvoy and Maxwell
1983; Smith, Eglinton, and Morris 1983, 1986; Brassell, Eglinton, and Howell 1987; Brault and
Simoneit 1988; Wakeham and Lee 1989) and lacustrine (Cranwell and Volkman 1981; Robinson et
al. 1984b, 1986; Cranwell, Eglinton, and Robinson 1987; Robinson, Eglinton, and Cranwell 1987)
environments.
The best opportunity to explore, evaluate, and validate the transformations of organic compounds in
sediments and to understand their origin(s) and fate(s) emerges from examination of biological
marker occurrences in samples that encompass the full maturity range from recent sediments to
mature petroleums. Assessment of variations in biomarker distributions through the complete
spectrum of their diagenetic and catagenetic alteration provides the opportunity to relate the
components of ancient sediments and petroleums, not only to the constituents present in organisms
that are known to be biosynthetic products, but also to their geologically occurring counterparts and
to appropriate series of intermediates in immature sediments. In addition, a thorough appraisal of
both microbial effects on biomarker distributions and later thermally induced modifications should
facilitate an appreciation of the nature, variety, and consistency of the reactions that influence their
compositions, which, in turn, can help resolve the many unresolved questions concerning the
precursor(s) and product(s) of specific sedimentary biomarkers. The determination of the carbon
isotopic signatures of individual biomarkers offers an extra dimension to such studies.
Acknowledgments
I thank my former colleagues and students from the Organic Geochemistry Unit at the University of
Bristol for helpful discussions over many years that have contributed to the themes expressed in
this paper. I also acknowledge the support of Chevron and Unocal toward the development of
molecular organic geochemical research at Stanford University, and the David and Lucile Packard
Foundation for a Fellowship for Science and Engineering.
References
Abbott, G. D., C. A. Lewis, and J. R. Maxwell. 1985. The kinetics of specific organic
reactions in the zone of catagenesis. Phil. Trans. Roy. Soc. Lond. A 315:107-122.
Albrecht, P. and G. Ourisson. 1971. Biogenic substances in sediments and fossils.

Angewandte Chem. (Int. Ed.) 10:209-225.
Aquino Neto, F. R. de, J. N. Cardoso, R. Rodrigues, and L. A. F. Trindade. 1986. Evolution

of tricyclic alkanes in the Espirito Santo Basin. Geochim. Cosmochim. Acta 50:2069-2072.
Boon, J. J. and J. W. de Leeuw. 1979. The analysis of wax esters, very long mid-chain
ketones and sterol ethers isolated from Walvis Bay diatomaceous ooze. Mar. Chem. 7:117-
132.
Boon, J. J., W. I. C. Rijpstra, F. de Lange, J. W. de Leeuw, M. Yoshioka, and Y. Shimizu.

1979. The Black Sea sterol-a molecular fossil for dinoflagellate blooms. Nature 277:125-127.
Brassell, S. C. 1984. Aliphatic hydrocarbons of a Cretaceous black shale and its adjacent
green claystone from the southern Angola Basin, Deep Sea Drilling Project Leg 75. In W. W.
Hay, J.-C. Sibuet, et al., eds., Initial Reports DSDP 75:1019-1030. Washington D.C.: U.S.
Brassell, S. C. 1985. Molecular changes in sediment lipids as indicators of systematic early

diagenesis. Phil. Trans. Roy. Soc. Lond. A 315:57-75.
Brassell, S. C. and G. Eglinton. 1981. Biogeochemical significance of a novel C27 stanol.

Nature 290:579-582.
Brassell, S. C. and G. Eglinton. 1983a. The potential of organic geochemical compounds as

sedimentary indicators of upwelling. In E. Suess and J. Thiede, eds., Coastal Upwelling, Its
Sediment Record, Part A, pp. 545-571. New York: Plenum.
Brassell, S. C. and G. Eglinton. 1983b. Steroids and triterpenoids in deep sea sediments as
environmental and diagenetic indicators. In M. Bjorøy et al., eds., Advances in Organic
Geochemistry 1981, pp. 684-697. Chichester: Wiley.
Brassell, S. C. and G. Eglinton. 1986. Molecular geochemical indicators in sediments. In M.

Sohn, ed. Organic Marine Geochemistry, pp. 10-32. Washington D.C.: ACS Sympos. Series
305.
Brassell, S. C. and P. Farrimond. 1986. Fluctuations in biological marker compositions within

a Cenomanian black shale from the Angola basin. In E. T. Degens, P. A. Meyers, and S. C.
Brassell, eds., Biogeochemistry of Black Shales, pp. 311-338. Mitt. Geol.-Palaontol. Inst.
Univ. 60.
Brassell, S. C., G. Eglinton, and Fu Jiamo. 1986. Biological marker compounds as indicators
of the depositional history of the Maoming oil shale. In D. Leythaeuser and J. Rullkötter,
eds., Advances in Organic Geochemistry 1985. Org. Geochem. 10:927-941. Oxford:
Pergamon.
Brassell, S. C., G. Eglinton, and V. J. Howell. 1987. Palaeoenvironmental assessment for

marine organic-rich sediments using molecular organic geochemistry. In J. Brooks and A. J.
Fleet, eds., Marine Petroleum Source Rocks, pp. 79-98. Oxford: Blackwell.
Brassell, S. C., G. Eglinton, and J. R. Maxwell. 1983. The geochemistry of terpenoids and

steroids. Biochem. Soc. Trans. 11:575-586.
Brassell, S. C., J. R. Maxwell and G. Eglinton. 1981. Preliminary lipid analyses of two
Quaternary sediments from the Middle America Trench, southern Mexico transect, Deep Sea
Drilling Project Leg 66. In J. S. Watkins, J. C. Moore, et al., eds., Initial Reports. DSDP
66:557-580. Washington D.C.: U.S. Govt. Printing Office.
Brassell, S. C., Sheng Guoying, Fu Jiamo, and G. Eglinton. 1988. Biological markers in
lacustrine Chinese oil shales. In A. J. Fleet, K. Kelts, and M. R. Talbot, eds., Lacustrine
Petroleum Source Rocks, pp. 299-308. Oxford: Blackwell.
Brassell, S. C., C. A. Lewis, J. W. de Leeuw, F. de Lange, and J. S. Sinninghe Damsté.

1986. Isoprenoid thiophenes: novel products of sediment diagenesis? Nature 320:160-162.
Brassell, S. C., A. M. K. Wardroper, I. D. Thomson, J. R. Maxwell and G. Eglinton. 1981a.

Specific acyclic isoprenoids as biological markers of methanogenic bacteria in marine
Brassell, S. C., J. McEvoy, C. F. Hoffmann, N. A. Lamb, T. M. Peakman, and J. R. Maxwell.

1984. Isomerisation, rearrangement and aromatisation of steroids in distinguishing early
stages of diagenesis. In P. A. Schenck, J. W. de Leeuw, and G. M. W. Lijmbach, eds.,
Advances in Organic Geochemistry 1983. Org. Geochem. 6:11-23. Oxford: Pergamon.

Thomson, P. J. C. Tibbetts, and J. K. Volkman. 1980a. The origin and fate of lipids in the
Japan Trench. In A. G. Douglas and J. R. Maxwell, eds., Advances in Organic Geochemistry

Thomson, P. J. C. Tibbetts, and J. K. Volkman. 1980b. Preliminary lipid analyses of Sections
440A-7-6, 440B-3-5, 440B-8-4, 440B-68-2 and 436-11-4: Legs 56 and 57, Deep Sea Drilling
Project. In Scientific Party, eds. Initial Reports. DSDP, 56, 57(2):1367-1390. Washington D.
Brault, M. and B. R. T. Simoneit. 1988. Steroid and triterpenoid distributions in Bransfield

Strait sediments: Hydrothermally-enhanced diagenetic transformations. In L. Matavelli and L.
Novelli, eds., Advances in Organic Geochemistry 1987. Org. Geochem. 13:583-592. Oxford:
Pergamon.
Chappe, B., W. Michaelis, and P. Albrecht. 1980. Molecular fossils of Archaebacteria as

selective degradation products of kerogen. In A. G. Douglas and J. R. Maxwell, eds.,
Chou, M. M. and K. V. Wood. 1986. New aromatic biomarkers and possible maturity
indicators found in New Albany Shale extracts. Org. Geochem. 9:351-356.
Comet, P. A., J. McEvoy, S. C. Brassell, G. Eglinton, J. R. Maxwell, and I. D. Thomson.

1981. Lipids of an Upper Albian limestone, Deep Sea Drilling Project Site 465, Section 465A-
38-3. In J. Thiede, T. L. Vallier, et al., eds. Initial Reports DSDP 62:723-937 Washington D.
Cox, R. E., A. L. Burlingame, D. M. Wilson, G. Eglinton, and J. R. Maxwell, 1973.

Botryococcene-a tetramethylated acyclic triterpenoid of algal origin. J. Chem. Soc. Chem.
Commun. pp. 284-285.
Cranwell, P. A. 1984. Lipid geochemistry of sediments from Upton Broad, a small productive
lake. Org. Geochem. 7:25-37.
Cranwell, P. A. and J. K. Volkman. 1981. Alkyl and steryl esters in a Recent lacustrine
sediment. Chem. Geol. 32:29-43.
Cranwell, P. A., G. Eglinton, and N. Robinson. 1987. Lipids of aquatic organisms as potential
contributors to lacustrine sediments-II. Org. Geochem. 11:513-527.
Curiale, J. 1987. Steroidal hydrocarbons of the Kishenehn formation, northwest Montana.

Dastillung, M. and P. Albrecht. 1977. 2-sterenes as diagenetic intermediates in recent

Dastillung, M. and B. Corbet. 1975. La géochemie organique des sédiments marins profonds
I. Hydrocarbures saturés et insaturés des sédiments. In A. Combaz and R. Pelet, eds.,
Géochemie Organique des Sédiments Marins Profonds. Orgon II Atlantique-N.E. Brésil, pp.
295-326. Paris: Éditions du CNRS.
Derenne, S., C. Largeau, E. Casadevall, and C. Berkaloff. 1989. Occurrence of a resistant

biopolymer in the L race of Botryococcus braunii. Phytochem. 28:1137-1142.
von der Dick, H., J. Rullkötter, and D. H. Welte. 1983. Content, type, and thermal evolution of
organic matter in sediments from the eastern Falkland Plateau, Deep Sea Drilling Project,
Leg 71. In W. J. Ludwig, V. A. Krasheninikov, et al., eds., Initial Reports DSDP, 71(2):1015-
1032. Washington, D.C.: U.S. Govt. Printing Office.
Duncan, A. D. and R. F. M. Hamilton. 1988. Palaeolimnology and organic geochemistry of

the Middle Devonian in the Orcadian Basin. In A. J. Fleet, K. Kelts, and M. R. Talbot, eds.,
Lacustrine Petroleum Source Rocks, pp. 173-201. Oxford: Blackwell.
Ensminger, A., G. Joly, and P. Albrecht. 1978. Rearranged steranes in sediments and crude
oils. Tetrahedron Lett. 18:1575-1578.
Ensminger, A., A. Van Dorsselaer, C. Spyckerelle, P. Albrecht, and G. Ourisson. 1974.

Pentacyclic triterpenes of the hopane type as ubiquitous geochemical markers: origin and
significance. In B. Tissot and F. Bienner, eds., Advances in Organic Geochemistry 1973, pp.
245-260. Paris: Technip.

Farrimond, P., G. Eglinton, and S. C. Brassell. 1986. Geolipids of black shales and
claystones in Cretaceous and Jurassic sediment sequences from the North American Basin.
In C. P. Summerhayes and N. J. Shackleton, eds., North Atlantic Palaeoceanography, pp.
347-360. Oxford: Blackwell.
Farrimond, P., G. Eglinton, S. C. Brassell, and H. C. Jenkyns. 1988. The Toarcian black
shale event in Northern Italy. In L. Matavelli and L. Novelli, eds., Advances in Organic
Geochemistry 1987. Org. Geochem. 13:823-832. Oxford: Pergamon.
Farrimond, P., G. Eglinton, S. C. Brassell, and H. C. Jenkyns. 1989. Toarcian anoxic event in
Europe: an organic geochemical study. Mar. Petrol. Geol. 6:136-147.
Farrington, J. W., A. C. Davis, J. Sulanowski, M. A. McCaffrey, M. McCarthy, C. H. Clifford,

P. Dickinson, and J. K. Volkman. 1988. Biogeochemistry of lipids in surface sediments of the
Peru Upwelling Area at 15°S. In L. Matavelli and L. Novelli, eds., Advances in Organic
Freeman, K. H., J. M. Hayes, J.-M. Trendel, and P. Albrecht. 1990. Evidence from carbon
isotope measurements for diverse origins of sedimentary hydrocarbons. Nature 343:254-
256.
Fu, Jiamo, Sheng Guoying, and Liu Dehan. 1988. Organic geochemical characteristics of
major types of terrestrial petroleum source rocks in China. In A. J. Fleet, K. Kelts, and M. R.
Talbot, eds., Lacustrine Petroleum Source Rocks, pp. 279-289. Oxford: Blackwell.
Fu, Jiamo, Sheng Guoying, Peng Pingan, S. C. Brassell, G. Eglinton, and Jiang Jigang.
1986. Peculiarities of salt lake sediments as potential source rocks in China. In D.
Leythaeuser and J. Rullkötter, eds., Advances in Organic Geochemistry 1985. Org.
Geochem. 10:119-126. Oxford: Pergamon.
Gagosian, R. B. and J. W. Farrington. 1978. Sterenes in surface sediments from the

southwest African shelf and slope. Geochim. Cosmochim. Acta 42:1091-1101.
Gagosian, R. B. and S. O. Smith. 1979. Steroid ketones in surface sediments from the south-
west African shelf. Nature 277:287-289.
Gagosian, R. B., C. Lee, and F. Heinzer. 1979. Processes controlling the stanol/stenol ratio
in Black Sea seawater and sediments. Nature 280:574-576.
Gagosian, R. B., G. E. Nigrelli and J. K. Volkman. 1983. Vertical transport and

transformation of biogenic organic compounds from a sediment trap experiment off the coast
of Peru. In E. Suess and J. Thiede, eds., Coastal Upwelling, Its Sediment Record, Part A,
pp. 241-272. New York: Plenum.
Gagosian, R. B., S. O. Smith, and G. E. Nigrelli. 1982. Vertical transport of steroid alcohols
and ketones measured in a sediment trap experiment in the equatorial Atlantic Ocean.

Gagosian, R. B., J. K. Volkman, and G. E. Nigrelli. 1983. The use of sediment traps to
determine sterol sources in coastal sediments off Peru. In M. Bjorøy et al., eds., Advances in
Organic Geochemistry 1981, pp. 369-379. Chichester: Wiley.
Gagosian, R. B., S. O. Smith, C. Lee, J. W. Farrington, and N. M. Frew. 1980. Steroid

transformation in recent marine sediments. In A. G. Douglas and J. R. Maxwell, eds.,
Gardner, P. M. and E. V. Whitehead. 1972. The isolation of squalane from a Nigerian

petroleum. Geochim. Cosmochim. Acta 36:259-263.
Gaskell, S. J. and G. Eglinton. 1975. Rapid hydrogenation of sterols in a contemporary

lacustrine sediment. Nature 254:209-211.
Gaskell, S. J. and G. Eglinton. 1976. Sterols of a contemporary lacustrine sediment.

Giger, W. and C. Schaffner. 1981. Unsaturated steroid hydrocarbons as indicators of

maturation in immature Monterey shales. Naturwissenschaften 68:37-39.
Goodwin, N. S., A. L. Mann, and R. L. Patience. 1988. Structure and significance of C30 4-
methylsteranes in lacustrine shales and oils. Org. Geochem. 12:495-506.
Goossens, H., J. W. de Leeuw, P. A. Schenck, and S. C. Brassell. 1984. Tocopherols as

likely precursors of pristane in ancient sediments and crude oils. Nature 312:440-442.
Goossens, H., R. R. Düren, J. W. de Leeuw, and P. A. Schenck. 1989. Lipids and their mode
of occurrence in bacteria and sediments-II. Lipids in the sediment of a stratified, freshwater
lake. Org. Geochem. 14:27-41.
Goth, K., J. W. de Leeuw, W. Püttmann, and E. W. Tegelaar. 1988. Origin of Messel oil
shale kerogen. Nature 336:759-761.
van Graas, G., J. M. A. Baas, B. van der Graaf, and J. W. de Leeuw. 1982. Theoretical
organic geochemistry: I. The thermodynamic stability of several cholestane isomers
calculated by molecular mechanics. Geochim. Cosmochim. Acta 46:2399-2402.
Hall, P. B. and A. G. Douglas. 1983. The distributions of cyclic alkanes in two lacustrine
deposits. In M. Bjorøy et al., eds., Advances in Organic Geochemistry 1981, pp. 576-587.
Chichester, Wiley.
Harvey, H. R., G. Eglinton, S. C. M. O'Hara and E. D. S. Corner. 1987. Biotransformation

and assimilation of dietary lipids by Calanus feeding on a dinoflagellate. Geochim.
Harwood, J. L. and N. J. Russell. 1984. Lipids in Plants and Microbes. London: Allen &

Unwin.
ten Haven, H. L. and J. Rullkötter. 1988. The diagenetic fate of taraxer-14-ene and oleanene
isomers. Geochim. Cosmochim. Acta 52:2543-2548.
ten Haven, H. L., J. W. de Leeuw, and P. A. Schenck. 1985. Organic geochemical studies of
a Messinian evaporitic basin, northern Appenines (Italy) I: Hydrocarbon biological markers
for a hypersaline environment. Geochim. Cosmochim. Acta 49:2181-2191.
ten Haven, H. L., J. W. de Leeuw, T. M. Peakman, and J. R. Maxwell. 1986. Anomalies in

steroid and hopanoid maturity indices. Geochim. Cosmochim. Acta 50:853-855.
ten Haven, H. L., M. Rohmer, J. Rullkötter, and P. Bisseret. 1989. Tetrahymenol, the most
likely precursor of gammacerane, occurs ubiquitously in marine sediments. Geochim.
ten Haven, H. L., M. Baas, J. W. de Leeuw, J. M. Maassen, and P. A. Schenck. 1987a.

Organic geochemical characteristics of sediments from the anoxic brine-filled Tyro basin
(eastern Mediterranean). Org. Geochem. 11:605-611.
ten Haven, H. L., M. Baas, M. Kroot, J. W. de Leeuw, P. A. Schenck, and J. Ebbing. 1987b.
Late Quaternary Mediterranean sapropels I-On the origin of organic matter in sapropel S7. In
J. E. van Hinte, M. B. Cita, C. H. van der Weijden, eds., Extant and Ancient Anoxic Basin
Conditions in the Eastern Mediterranean. Mar. Geol. 75:137-156. Amsterdam: Elsevier.
ten Haven, H. L., M. Baas, M. Kroot, J. W. de Leeuw, P. A. Schenck, and J. Ebbing. 1987c.
Late Quaternary Mediterranean sapropels II-Organic geochemistry of S1 sapropels and
associated sediments. Chem. Geol. 64:149-167.
ten Haven, H. L., M. Baas, M. Kroot, J. W. de Leeuw, P. A. Schenck, and J. Ebbing. 1987d.
Late Quaternary Mediterranean sapropels III-Assessment of source of input and
paleotemperature as derived from biological markers. Geochim. Cosmochim. Acta 51:803-
810.
ten Haven, H. L., J. W. de Leeuw, J. S. Sinninghe Damsté, P. A. Schenck, S. E. Palmer, and

J. E. Zumberge. 1988. Application of biological markers in the recognition of
palaeohypersaline environments. In A. J. Fleet, K. Kelts, and M. R. Talbot, eds., Lacustrine
Petroleum Source Rocks, pp. 123-140. Oxford: Blackwell.
Hayes, J. M., R. Takigiku, R. Ocampo, H. J. Callot, and P. Albrecht. 1987. Isotopic

compositions and probable origins of organic molecules in the Eocene Messel Shale. Nature
329:48-51.
Hills, I. R., E. V. Whitehead, D. E. Anders, J. J. Cummins, and W. E. Robinson. 1966. An

optically active triterpane, gammacerane, in Green River, Colorado, oil shale bitumen. J.
Chem. Soc. Chem. Commun., pp. 752-754.

Holzer, G., J. Oro, and T. G. Tornabene. 1979. Gas chromatographic-mass spectrometric

analysis of neutral lipids from methanogenic and thermoacidophilic bacteria. J. Chromatogr.
186:795-809.
Huang, W.-Y. and W. G. Meinschein. 1978. Sterols in sediments from Baffin Bay, Texas.
Hussler, G. and P. Albrecht. 1983. C27-C29 monoaromatic anthrasteroid hydrocarbons in

Cretaceous black shales. Nature 304:262-263.
Jiang, Z. S. and M. G. Fowler. 1986. Carotenoid-derived alkanes in oils from northwestern

China. In D. Leythaeuser and J. Rullkötter, eds., Advances in Organic Geochemistry 1985.
Org. Geochem. 10:831-839. Oxford: Pergamon.
Jiang, Z. S., R. P. Philp, and C. A. Lewis. 1988. Fractionation of biological markers in crude
oils during migration and the effects on correlation and maturation parameters. In L.
Matavelli and L. Novelli, eds., Advances in Organic Geochemistry 1987. Org. Geochem.
13:561-571. Oxford: Pergamon.
Kimble, B. J., J. R. Maxwell, R. P. Philp, G. Eglinton, P. Albrecht, A. Ensminger, P. Arpino,

and G. Ourisson. 1974. Tri- and tetraterpenoid hydrocarbons in the Messel oil shale.
Laflamme, R. E. and R. A. Hites. 1979. Tetra- and pentacyclic naturally-occurring, aromatic

hydrocarbons in recent sediments. Geochim. Cosmochim. Acta 43:1687-1691.
Larcher, A. V., R. Alexander, and R. I. Kagi. 1988. Differences in reactivities of sedimentary

hopane diastereomers when heated in the presence of clays. In L. Matavelli and L. Novelli,
Pergamon.
Largeau, C., E. Casadevall, A. Kadouri, and P. Metzger. 1984. Comparative study of

immature torbanite and of the extant alga Botryococcus braunii. In P. A. Schenck, J. W. de
Leeuw, and G. M. W. Lijmbach, eds., Advances in Organic Geochemistry 1983. Org.
Geochem. 6:327-332, Oxford: Pergamon.
Largeau, C., S. Derenne, E. Casadevall, A. Kadouri, and N. Sellier. 1986. Pyrolysis of

immature torbanite and of the resistant biopolymer (PRB A) isolated from extant alga
Botryococous braunii. Mechanism of formation and structure of torbanite. In D. Leythaeuser
and J. Rullkötter, eds., Advances in Organic Geochemistry 1985. Org. Geochem. 10:1023-
1032. Oxford: Pergamon.
Laureillard, J., C. Largeau, and E. Casadevall. 1988. Oleic acid in the biosynthesis of the
resistant biopolymers of Botryococcus braunii. Phytochem. 27:2095-2098.
Lee, C., J. W. Farrington, and R. B. Gagosian. 1979. Sterol geochemistry of sediments

fromthe western North Atlantic Ocean and adjacent coastal areas. Geochim. Cosmochim.

Acta 43:35-46.
Lee, C., R. B. Gagosian, and J. W. Farrington. 1977. Sterol diagenesis in Recent sediments
from Buzzards Bay, Massachusetts. Geochim. Cosmochim. Acta 41:985-992.
Lee, C., R. B. Gagosian, and J. W. Farrington. 1980. Geochemistry of sterols in sediments

from the Black Sea and the southwest African shelf and slope. Org. Geochem. 2:103-113.
de Leeuw, J. W. and M. Baas. 1986. Early-stage diagenesis of steroids. In R. B. Johns, ed.,

Biological Markers in the Sedimentary Record, pp. 101-123. Amsterdam: Elsevier.
de Leeuw, J. W., W. I. C. Rijpstra, P. A. Schenck, and J. K. Volkman. 1983. Free, esterified

and residual bound sterols in Black Sea Unit 1 sediments. Geochim. Cosmochim. Acta
47:455-465.
de Leeuw, J. W., H. C. Cox, G. van Graas, F. W. van de Meer, T. M. Peakman, J. M. A.

Baas, and B. van de Graaf. 1989. Limited double bond isomerization and selective
hydrogenation of sterenes during early diagenesis Geochim. Cosmochim. Acta 53:903-909.
Mackenzie, A. S. 1984. Applications of biological markers in petroleum geochemistry. In J.

Brooks and D. Welte, eds., Advances in Petroleum Geochemistry, pp. 115-214. London:
Academic.
Mackenzie, A. S., C. F. Hoffmann, and J. R. Maxwell. 1981. Molecular parameters of

maturation in the Toarcian shales, Paris Basin, France-III. Changes in aromatic steroid
Mackenzie, A. S., S. C. Brassell, G. Eglinton, and J. R. Maxwell. 1982. Chemical fossils-the

geological fate of steroids. Science 217:491-504.
Mackenzie, A. S., R. L. Patience, J. R. Maxwell, M. Vandenbroucke, and B. Durand. 1980.

Molecular parameters of maturation in the Toarcian shales, Paris Basin, France-I. Changes
in the configurations of acyclic isoprenoid alkanes, steranes and triterpanes. Geochim.
Matsumoto, G., T. Torii, and T. Hanya. 1982. High abundance of algal 24-ethylcholesterol in
Antarctic lake sediment. Nature 299:52-54.
Maxwell, J. R., C. T. Pillinger, and G. Eglinton. 1971. Organic Geochemistry. Quart. Rev.
25:571-628.
Maxwell, J. R., A. G. Douglas, G. Eglinton, and A. McCormick. 1968. The botryococcenes-

hydrocarbons of novel structure from the alga Botryococcus braunii, Kutzing. Phytochem.
7:2157-2171.
McEvoy, J. and J. R. Maxwell. 1983. Diagenesis of steroidal compounds in sediments from

the Southern California Bight (DSDP Leg 63, Site 467). In M. Bjorøy et al., eds., Advances in

Organic Geochemistry 1981, pp. 449-464. Chichester: Wiley.
McEvoy, J. and J. R. Maxwell. 1986. A-norsteroidal ketones in deep sea sediments. Org.
Geochem. 9:101-104.
McEvoy, J., G. Eglinton, and J. R. Maxwell. 1981. Preliminary lipid analyses of sediments
from Sections 467-3-3 and 467-97-2. In R. S. Yeats, B. U. Haq, et al., eds., Initial Reports
DSDP. 63:763-774. Washington D.C.: U.S. Govt. Printing Office.
McKirdy, D. M., R. E. Cox, J. K. Volkman, and V. J. Howell. 1986. Botryococcane in a new

class of Australian non-marine crude oils. Nature 320:57-59.
Mello, M. R., P. C. Gaglianone, S. C. Brassell, and J. R. Maxwell. 1988a. Geochemical and

biological marker assessment of depositional environments using Brazilian offshore oils.
Mar. Petrol. Geol. 5:205-219.
Mello, M. R., N. Telnaes, P. C. Gaglianone, M. I. Chicarelli, S. C. Brassell, and J. R.

Maxwell. 1988b. Organic geochemical depositional palaeoenvironments of source rocks and
oils in Brazilian marginal basins. In L. Matavelli and L. Novelli, eds., Advances in Organic
Mermoud, F., F. O. Gülaçar, S. Siles, B. Chassaing, and A. Buchs. 1982. 4-Methylsterols in

recent lacustrine sediments: terrestrial, planktonic or some other origin? Chemosphere
11:557-567.
Metzger, P. and E. Casadevall. 1983. Structure de trois nouveaux "botryococcènes"

synthétisés par une souche de Botryococcus braunii cultivée en laboratoire. Tetrahedron
Lett. 24:4013-4016.
Metzger, P. and E. Casadevall. 1987. Lycopadiene, a tetraterpenoid hydrocarbon from new

strains of the green alga Botryococcus braunii. Tetrahedron Lett. 28:3931-3934.
Metzger, P., C. Berkaloff, E. Casadevall, and A. Coute. 1985a. Alkadiene- and

botryococcene-producing races of wild strains of Botryococcus braunii. Phytochem. 24:2305-
2312.
Metzger, P., E. Casadevall, M. J. Pouet, and Y. Pouet. 1985b. Structures of some

botryococcenes: branched hydrocarbons from the B-race of the green alga Botryococcus
braunii. Phytochem. 24:2995-3002.
Moldowan, J. M. and W. K. Seifert. 1980. First discovery of botryococcane in petroleum. J.

Chem. Soc., Chem. Commun., pp. 912-914.
Moldowan, J. M., W. K. Seifert, and E. J. Gallegos. 1985. The relationship between

petroleum composition and the environment of deposition of petroleum source rocks. AAPG
Bull. 69:1255-1268.

Moldowan, J. M., P. Sundararaman, and M. Schoell. 1986. Sensitivity of biomarker

properties to depositional environment and/or source input in the Lower Toarcian of SW-
Germany. In D. Leythaeuser and J. Rullkötter, eds., Advances in Organic Geochemistry
1985. Org. Geochem. 10:915-926. Oxford: Pergamon.
Murphy, M. T. J., A. McCormick, and G. Eglinton. 1967. Perhydro- -carotene in Green River
Shale. Science 157:1040-1042.
Mycke, B. and W. Michaelis. 1986. Molecular fossils from chemical degradation of

macromolecular organic matter. In D. Leythaeuser and J. Rullkötter, eds., Advances in
Organic Geochemistry 1985. Org. Geochem. 10:847-858. Oxford: Pergamon.
Nishimura, M. 1977. The geochemical significance in early sedimentation of geolipids

obtained by saponification of lacustrine sediments. Geochim. Cosmochim. Acta 41:1817-
1823.
Nishimura, M. and T. Koyama. 1976. Stenols and stanols in lake sediments and diatoms.
Chem. Geol. 17:229-239.
Noble, R. A., R. Alexander, and R. I. Kagi. 1987. Configurational isomerization in

sedimentary bicyclic alkanes. Org. Geochem. 11:151-156.
Peakman, T. M. 1986. Synthesis, occurrence, and low temperature diagenesis of steroid

hydrocarbons. Ph.D. thesis, Univ. of Bristol, U.K.
Peakman, T. M. and J. R. Maxwell. 1988. Early diagenetic pathways of steroidal alkenes. In

L. Matavelli and L. Novelli, eds., Advances in Organic Geochemistry 1987. Org. Geochem.
13:583-592. Oxford: Pergamon.
Peakman, T. M., H. L. ten Haven, J. R. Rechka, J. W. de Leeuw, and J. R. Maxwell. 1989.

Occurrence of (20R)- and (20S)- 8(14) and 14 5 (H)-steranes and the origin of 5 (H),14
(H), 17 (H)-steranes in an immature sediment. Geochim. Cosmochim. Acta. 53:2001-2009.
Peters, K. E., J. M. Moldowan, A. R. Driscole, and G. J. Demaison. 1989. Origin of Beatrice

oil by co-sourcing from Devonian and Middle Jurassic source rocks, Inner Moray Firth,
United Kingdom. AAPG Bull. 73:454-471.
Prahl, F. G., G. Eglinton, E. D. S. Corner, S. C. M. O'Hara and T. E. V. Fosberg. 1984.

Changes in plant lipids during passage through the gut of Calanus. J. Mar. Biol. Assoc. U.K.
64:317-334.
Riolo, J. and P. Albrecht. 1985. Novel rearranged ring C monoaromatic steroid hydrocarbon
in petroleum. Tetrahedron Lett. 26:2701-2704.
Robinson, N., G. Eglinton, and P. A. Cranwell. 1987. Sources of the lipids in the bottom
sediments of an English oligo-mesotrophic lake. Freshwater Biol. 17:15-33.

Robinson, N., G. Eglinton, S. C. Brassell, and P. A. Cranwell. 1984a. Dinoflagellate origin for
sedimentary 4 -methylsteroids and 5 (H)-stanols. Nature 308:439-442.
Robinson, N., P. A. Cranwell, B. J. Finlay, and G. Eglinton. 1984b. Lipids of aquatic

organisms as potential contributors to lacustrine sediments. In P. A. Schenck, J. W. de
Leeuw, and G. M. W. Lijmbach, eds., Advances in Organic Geochemistry 1985. Org.
Geochem. 6:143-152. Oxford: Pergamon.
Robinson, N., P. A. Cranwell, G. Eglinton, S. C. Brassell, C. L. Sharp, M. Gophen, and U.

Pollingher. 1986. Lipid geochemistry of Lake Kinneret. In D. Leythaeuser and J. Rullkötter,
eds., Advances in Organic Geochemistry 1985. Org. Geochem. 10:733-742 Oxford:
Pergamon.
Robson, J. N. and S. J. Rowland. 1988. Biodegradation of highly branched isoprenoid

hydrocarbons: a possible explanation of sedimentary abundance. In L. Matavelli and L.
Novelli, eds., Advances in Organic Geochemistry 1987. Org. Geochem. 13:691-695. Oxford:
Pergamon.
Rubinstein, I., O. Sieskind, and P. Albrecht. 1975. Rearranged sterenes in a shale:

occurrence and simulated formation. J. Chem. Soc. Perkin Trans I, pp.1833-1836.
Rullkötter, J. and D. H. Welte. 1983. Maturation of organic matter in areas of high heat flow:
a study of sediments from DSDP Leg 63, offshore California and Leg 64, Gulf of California.
In M. Bjorøy et al., eds., Advances in Organic Geochemistry 1981. pp. 438-448. Chichester:
Wiley.
hydrocarbons of sediments from the eastern North Pacific Ocean, Deep Sea Drilling Project
Leg 63. In R. S. Yeats, B. U. Haq, et al., eds., Initial Reports DSDP 63:819-836. Washington
D.C.: U.S. Govt. Printing Office.
hydrocarbons of sediments from the Gulf of California, Deep Sea Drilling Project Leg 64. In
J. R. Curray, G. Moore, et al., eds., Initial Reports DSDP 64(2):837-853. Washington D.C.: U.
S. Govt. Printing Office.
Rullkötter, J., P. K. Mukhopadhyay and D. H. Welte. 1984. Geochemistry and petrography of

organic matter in sediments from Hole 530A, Angola Basin, and Hole 532, Walvis Ridge,
Deep Sea Drilling Project. In W. W. Hay, J.-C. Sibuet, et al., Initial Reports DSDP 75:1069-
1087. Washington D.C.: U.S. Govt. Printing Office.
Rullkötter, J., P. K. Mukhopadhyay, and D. H. Welte. 1987. Geochemistry and petrography

of organic matter in sediments from Deep Sea Drilling Project Site 603, lower continental rise
off Cape Hatteras. In J. E. Van Hinte, S. W. Wise, et al., eds., Initial Reports DSDP 93:1163-
1176. Washington D.C.: U.S. Govt. Printing Office.
Rullkötter, J., P. K. Mukhopadhyay, R. G. Schaefer, and D. H. Welte. 1984. Geochemistry

and petrography of organic matter in sediments from Deep Sea Drilling Project Sites 545 and

547, Mazagan Escarpment. In K. Hinz, E. L. Winterer, et al., eds., Initial Reports DSDP
79:775-806. Washington D.C.: U.S. Govt. Printing Office.
Rullkötter, J., P. K. Mukhopadhyay, U. Disko, R. G. Schaefer, and D. H. Welte. 1986. Facies

and diagenesis of organic matter in deep sea sediments from the Blake Outer Ridge and the
Blake Bahama Basin, Western North Atlantic. In E. T. Degens, P. A. Meyers, and S. C.
Brassell. eds., Biogeochemistry of Black Shales. Mitt. Geol.-Palaontol. Inst. Univ. 60, pp. 179-
203.
Schmid, J. C., J. Connan, and P. Albrecht. 1987. Identification of long-chain

dialkylthiacyclopentanes in petroleum. Nature 329:54-56.
Seifert, W. K. and J. M. Moldowan. 1979. The effect of biodegradation on steranes and

terpanes in crude oils. Geochim. Cosmochim. Acta 43:111-126.
Seifert, W. K. and J. M. Moldowan. 1980. The effect of thermal stress on source rock quality
as measured by hopane stereochemistry. In A. G. Douglas and J. R. Maxwell, eds.,
Seifert, W. K. and J. M. Moldowan. 1981. Paleoreconstruction by biological markers.

Shaw, P. M. and R. B. Johns. 1985. Comparison of the lipid composition of sediments from
three sites in the Venezuela Basin. In D. K. Young and M. D. Richardson, eds., Benthic
Ecology and Sedimentary Processes of the Venezuela Basin: Past and Present. Mar. Geol.
68:205-216. Amsterdam: Elsevier.
Shaw, P. M. and R. B. Johns. 1986. The identification of organic input sources of sediments
from the Santa Catalina Basin using factor analysis. In D. Leythaeuser and J. Rullkötter,
Pergamon.
Shi Ji-Yang, A. S. Mackenzie, R. Alexander, G. Eglinton, A. P. Gowar, G. A. Wolff, and J. R.

Maxwell. 1982. A biological marker investigation of petroleums and shales from the Shengli
oilfield, the People's Republic of China. Chem. Geol. 35:1-31.
Simoneit, B. R. T., J. O. Grimalt, T. G. Wang, R. E. Cox, P. G. Hatcher, and A. Nissenbaum.

1986. Cyclic terpenoids of contemporary resinous plant detritus and of fossil woods, ambers
and coals. In D. Leythaeuser and J. Rullkötter, eds., Advances in Organic Geochemistry
Sinninghe Damsté, J. S., H. L. ten Haven, J. W. de Leeuw, and P. A. Schenck. 1986.

Organic geochemical studies of a Messinian evaporitic basin, Northern Apennines (Italy) II:
Isoprenoid and n-alkyl thiophenes and thiolanes. In D. Leythaeuser and J. Rullkötter, eds.,
Advances in Organic Geochemistry 1985. Org. Geochem. 10:791-805. Oxford: Pergamon.
Sinninghe Damsté, J. S., J. W. de Leeuw, A. C. Kock-van Dalen, M. A. de Zeeuw, F. de

Lange, W. I. C. Rijpstra and P. A. Schenck. 1987. The occurrence and identification of series
of organic sulphur compounds in oils and sediment extracts. I. A study of Rozel Point Oil
(USA). Geochim. Cosmochim. Acta 51:2369-2391.
Sinninghe Damsté, J. S., T. I. Eglinton, J. W. de Leeuw, and P. A. Schenck. 1989a. Organic

sulphur in macromolecular sedimentary organic matter: I. Structure and origin of sulphur-
containing moieties in kerogen, asphaltenes and coals as revealed by flash pyrolysis.
Sinninghe Damsté, J. S., W. I. C. Rijpstra, J. W. de Leeuw, and P. A. Schenck. 1989b. The

occurrence and identification of organic sulphur compounds in oils and sediment extracts: II.
Their presence in samples from hypersaline and non-hypersaline palaeoenvironments and
possible application as source, palaeoenvironmental and maturity indicators. Geochim.
Sinninghe Damsté, J. S., W. I. C. Rijpstra, A. C. Kock-van Dalen, J. W. de Leeuw, and P. A.

Schenck. 1989c. Quenching of labile functionalised lipids by inorganic sulphur species:
evidence for the formation of sedimentary organic sulphur compounds at the early stages of
Sinninghe Damsté, J. S., T. I. Eglinton, W. I. C. Rijpstra, and J. W. de Leeuw. 1990.

Characterization of organically bound sulfur in high molecular weight, sedimentary organic
matter using flash pyrolysis and Raney Ni desulfurization. In W. L. Orr and C. M. White, eds.,
Geochemistry of Sulfur in Fossil Fuels, pp. 486-528. Washington D.C.: ACS Symp. Ser. 429.
Smith, D. J., G. Eglinton, R. J. Morris, and E. L. Poutanen. 1982. Aspects of the steroid
geochemistry of a recent diatomaceous sediment from the Namibian Shelf. Oceanol. Acta
5:365-378.
Smith, D. J., G. Eglinton, and R. J. Morris. 1983. Aspects of the steroid geochemistry of an
interfacial sediment from the Peruvian upwelling. Oceanol. Acta 6:211-219.
Smith, D. J., G. Eglinton, and R. J. Morris. 1986. The lipid geochemistry of a recent sapropel
and associated sediments from the Hellenic Outer Ridge, eastern Mediterranean Sea. Phil.
Trans. Roy. Soc. Lond. A 319:375-415.
Sofer, Z. 1988. Hydrous pyrolysis of Monterey asphaltenes. In L. Matavelli and L. Novelli,

Pergamon.
Summons, R. E., J. K. Volkman, and C. J. Boreham. 1987. Dinosterane and other steroidal
hydrocarbons of dinoflagellate origin in sediments and petroleum. Geochim. Cosmochim.
Acta 51:3075-3082.
Taylor, C. D., S. O. Smith, and R. B. Gagosian. 1981. Use of microbial enrichments for the
study of the anaerobic degradation of cholesterol. Geochim. Cosmochim. Acta 45:2161-
2168.

Thomson, I. D., S. C. Brassell, G. Eglinton, and J. R. Maxwell. 1982. Preliminary lipid

analysis of Section 481-2-2. In J. R. Curray, D. G. Moore, et al., eds., Initial Reports DSDP 64
(2):913-919. Washington D.C.: U.S. Govt. Printing Office.
Van Dorsselaer, A., P. Albrecht, and J. Connan. 1977. Changes in composition of polycyclic
alkanes by thermal maturation (Yallourn lignite, Australia). In R. Campos and J. Goni, eds.,
Advances in Organic Geochemistry 1975, pp. 53-60. Madrid: Enadisma.
Venkatesan, M. I. 1989. Tetrahymenol: Its widespread occurrence and geochemical

significance. Geochim. Cosmochim. Acta 53:3095-3101.
Venkatesan, M. I. and I. R. Kaplan. 1987. The lipid geochemistry of Antarctic marine

sediments: Bransfield Strait. Mar. Chem. 21:347-375.
Volkman, J. K. 1988. Biological marker compounds as indicators of the depositional

environments of petroleum source rocks. In A. J. Fleet, K. Kelts, and M. R. Talbot, eds.,
Lacustrine Petroleum Source Rocks, pp. 103-122. Oxford: Blackwell.
Volkman, J. K. and J. R. Maxwell. 1986. Acyclic isoprenoids as biological markers. In R. B.

Johns, ed., Biological Markers in the Sedimentary Record, pp. 1-42. Amsterdam: Elsevier.
Volkman, J. K., J. W. Farrington, and R. B. Gagosian. 1987. Marine and terrigenous lipids in
coastal sediments from the Peru upwelling region at 15°S: sterols and triterpene alcohols.
Wakeham, S. G. 1982. Organic matter from a sediment trap experiment in the equatorial
North Atlantic: wax esters, steryl esters, triacylglycerols and alkyldiacylglycerols. Geochim.
Wakeham, S. G. 1987. Steroid geochemistry in the oxygen minimum zone of the eastern
tropical North Pacific Ocean. Geochim. Cosmochim. Acta 51:3051-3069.
Wakeham, S. G. and E. A. Canuel. 1988. Organic geochemistry of particulate matter in the

eastern tropical North Pacific Ocean. J. Mar. Res. 46:183-213.
Wakeham, S. G. and J. Ertel. 1988. Diagenesis of organic matter in suspended particles and
sediments in the Cariaco Trench. In L. Matavelli and Novelli, eds., In Advances in Organic
Wakeham, S. G. and N. M. Frew. 1982. Glass capillary gas chromatography-mass

spectrometry of wax esters, steryl esters and triacylglycerols. Lipids 17:831-843.
Wakeham, S. G. and C. Lee. 1989. Organic geochemistry of particulate matter in the ocean:
the role of particles in oceanic sedimentary cycles. Org. Geochem. 14:83-96.
Wakeham, S. G., C. Schaffner, and W. Giger. 1980. Polycyclic aromatic hydrocarbons in

Recent lake sediments-II. Compounds derived from biogenic precursors during early

Wakeham, S. G., C. Lee, J. W. Farrington, and R. B. Gagosian. 1984. Biogeochemistry of

particulate organic matter in the oceans: results from sediment trap experiments. Deep Sea
Res. 31:509-528.
Wakeham, S. G., J. W. Farrington, R. B. Gagosian, C. Lee, H. DeBaar, G. E. Nigrelli, B. W.

Tripp, S. O. Smith, and N. M. Frew. 1980. Organic matter fluxes from sediment traps in the
equatorial Atlantic Ocean. Nature 286:798-800.
Wardroper, A. M. K. 1979. Aspects of the geochemistry of polycyclic isoprenoids. Ph.D.

thesis, Univ. of Bristol, U.K.
Wardroper, A. M. K., J. R. Maxwell, and R. J. Morris. 1978. Sterols of a diatomaceous ooze

from Walvis Bay. Steroids 32:203-221.
Wolff, G. A., N. A. Lamb, and J. R. Maxwell. 1986. The origin and fate of 4-methylsteroid
hydrocarbons. I. Diagenesis of 4-methylsterenes. Geochim. Cosmochim. Acta 50:335-342.
Wü#che, L., F. O. Gülaçar, and A. Buchs. 1987. Several unexpected marine sterols in a
freshwater sediment. Org. Geochem. 11:215-219.
Wü#che, L., Y. A. Mendoza, and F. O. Gülaçar. 1988. Lipid geochemistry of a post-glacial

lacustrine sediment. In L. Matavelli and L. Novelli, eds., Advances in Organic Geochemistry
Yang Bin and Yang Jianqiang. 1989. Effects of biodegradation on crude oils from Karamay
oilfield. Chinese J. Geochem. 8:15-24.
Zumberge, J. E. 1987. Terpenoid biomarker distributions in low maturity crude oils. Org.
Geochem. 11:479-496.


16. Natural Hydrous Pyrolysis: Petroleum Generation in

Submarine Hydrothermal Systems
The conversion of organic matter to petroleum products by hydrothermal activity is a geologically

rapid process, occurring in nature in many types of submarine environments. Geologically immature
organic matter of marine sediments is being altered to petroleum by this process, which is
analogous to laboratory hydrous pyrolysis. This activity has been studied in Guaymas Basin (Gulf of
California), Escanaba Trough and Middle Valley (NE Pacific), Bransfield Strait (Antarctica), and
Atlantis II Deep (Red Sea). Petroleum-like products are formed by the same process from
contemporary organic detritus and/or viable microorganisms when they become entrained by
turbulent mixing into the discharging vent waters, resulting in an instantaneous hydrous pyrolysis.
This latter case is ubiquitous in hydrothermal systems and has been studied in sediment-starved
fields of hydrothermal vents emanating directly from oceanic ridges, such as on the East Pacific
Rise at 13°N and 21°N and on the Mid-Atlantic Ridge at 26°N. The hydrocarbon products (methane
to asphalt) generated in all these areas have been compared in terms of composition, organic
matter sources, and analogy to reservoir petroleum. Preliminary data from laboratory hydrous
pyrolysis studies indicate that some of the organic matter interconversions observed in nature can
be duplicated and thereby studied in greater detail. Hydrous pyrolysis is a natural process in
hydrothermal vent systems generating petroleum, and the associated fluids are efficient solvents for
primary and secondary migration of that petroleum. In addition, such petroleum represents a major
input of carbon to the primary chemosynthetic productivity around hydrothermal vent systems.
Organic matter of marine sedimentary basins is derived from the syngenetic residues of biogenic
debris that originates from both autochthonous (marine) and allochthonous (continental) sources (e.
g., Simoneit 1978, 1982a; Hunt 1979). The preservation of organic matter in sediments depends on
the initial diagenetic processes, which involve microbial degradation and chemical conversions,
together with the acidity and redox potential of the environment (e.g., Didyk et al. 1978; Demaison
and Moore 1980). Subsequent sediment maturation and lithification cause metamorphism of the
organic matter, ultimately yielding petroleum products through the effects of temperature, pressure,
and petrology (Hunt 1979; Tissot and Welte 1984). However, the action of hydrothermal processes
on such sedimentary organic matter has been found to generate petroleum products in Guaymas
Basin almost "instantaneously" in terms of geological time (Simoneit and Lonsdale 1982). The
present paper reviews this process using examples from the extensively studied Guaymas Basin
and other areas of hydrothermal activity at both sedimented and bare-rock spreading centers and
compares the products with those obtained from laboratory simulation using hydrous pyrolysis.
The analytical techniques of organic geochemistry have been used extensively to examine the
character of organic matter in the geologic record in terms of its structural and compositional
makeup (e.g., Simoneit 1978; van de Meent et al. 1980; Tissot and Welte 1984; Johns 1986). The
sources, diagenetic and catagenetic histories, and migration mechanisms of this organic matter can
be evaluated from such data. In the following discussion, organic matter is classified as gas, lipids

(bitumen), asphaltenes, and kerogen (with the kerogen precursor, sometimes called
"pseudokerogen" or "protokerogen"), and the characteristics of this organic matter have been
described elsewhere (e.g., Simoneit 1983, 1988).
Nature of Organic Matter in Maturing Basins
Immature organic matter in recent sediments is composed of minor amounts (based on total organic
carbon content) of biogenic gas (CH4 and CO2, sometimes H2S), significant lipid residues of
terrigenous and/or marine origins, and a major macromolecular fraction consisting of fulvic and
humic acids and particulate detritus (e.g., biopolymer fragments, cell membranes, and
miscellaneous carbonaceous matter). The lipids and macromolecular material undergo alteration
and diagenesis (both thermal and microbial) according to the environmental conditions during
transport and deposition, i.e., oxidative degradation in high-energy, oxygenated environments and
reductive changes in low-energy anaerobic environments (Didyk et al. 1978; Demaison and Moore
1980; Simoneit 1983).
After early diagenesis, thermal maturation of organic matter commences at about 50°C (Hunt 1979)
as burial increases with a concomitant rise in temperature due to the geothermal gradient. This
process produces some low-temperature cracking products from the kerogen (nonextractable
macromolecular organic matter), such as natural gas (C1-C8+) and bitumen (C8-C40+, including a
large envelope of an unresolved complex mixture of compounds or UCM), which are added to the
endogenous biogenic gas and lipid residues. Maturation of protokerogen is generally believed to
occur via molecular rearrangements and addition of geomonomers by copolymerization. Further
heating during catagenesis (to about 150°C) generates additional amounts of bitumen and gas,
which far exceed the original concentrations of endogenous lipids and gas, thus erasing the bulk
compositional signatures of the latter and resulting in the characteristic distributions of petroleum
compounds (Hunt 1979; Simoneit 1983; Tissot and Welte 1984). Late stages of catagenesis and
the subsequent high-temperature alteration phase called metagenesis (very deep burial) primarily
generate methane from both bitumen and kerogen.
Nature of Organic Matter in Hydrothermal Systems
Hydrothermal systems can also act on sedimentary organic matter and on entrained ambient
organic detritus in water or in talus; these processes result in "instantaneous" diagenesis and
catagenesis, thus producing petroleum products analogous to those from the slower acting
geothermal changes described above (Simoneit and Lonsdale 1982; Simoneit 1983; 1984a,b; 1985;
1988; 1990). Gas (C1-C8+, CO2 and H2S) and bitumen (C8-C40+, with a large UCM) are cracked
from the pseudokerogen and biopolymers and are added to the endogenous gas and lipids. The
bitumen additionally contains products characteristic of elevated thermal processes (e.g.,
polynuclear aromatic hydrocarbons or PAH, stabilized molecular markers such as 17 (H)-hopanes,
etc.). The spent kerogen remains as disseminated, amorphous "activated" carbon (Curray et al.
1982; Simoneit 1982b; Simoneit et al. 1984).
The effects of pressure, temperature, and time on the chemistry of the organic matter are all
interrelated. Temperature and time, which primarily effect petroleum generation, control
mineralogical interactions and the products derived from the various compositions of kerogen.
Migration processes are understood less thoroughly. In sedimentary basins, migration of petroleum
seems to occur primarily by oil phase flow; migration by molecular diffusion and in water solution is

minor (Hunt 1979; Tissot and Welte 1984; Durand 1988). In hydrothermal areas, migration
proceeds with extreme efficiency by thermally driven diffusion, in solution (H2O/CH4/CO2
cosolvent) and by advection and mass transport as oil and/or emulsions (Kawka and Simoneit
1987; Simoneit, Kawka, and Brault 1988; Didyk and Simoneit 1989; Simoneit et al. 1990). Despite
these differences, the overall result may be the same for both regimes. Pressure in both cases aids
mainly the solubility of the petroleum in the H2O/CH4/CO2 fluids (Price, Clayton, and Rumen 1981;
Price et al. 1983).
Some of the recent results from this laboratory on hydrocarbon compositions from hydrothermal
systems are summarized below. Various basic organic geochemical fractionation procedures used
for sample analysis, described in detail elsewhere (e.g., Kawka and Simoneit 1987; Simoneit et al.
1979, 1981, 1984, 1987; Simoneit, Kawka, and Brault 1988; van de Meent et al. 1980; Simoneit
1981, 1982a,b) have been applied here with only minor modifications prior to instrumental analyses
to aid the intercomparability of the resultant data.
Seafloor Spreading Centers and HydrothermalPetroleum Generation
Although this overview discusses the occurrence of petroleum products in eight geographic
locations of hydrothermal activity (figure 16.1), it is my opinion that petroleum generation and
migration is a ubiquitous process associated with hydrothermalism and metallic mineral formation in
global rift systems. All that is really required are organic source material and temperatures in excess
of 50°C in an aqueous medium under a confining pressure. There are at present about 100 known
hydrothermal mineral occurrences at oceanic ridges and rifts (Rona 1988). The quantities of
petroleum associated directly with submarine hydrothermal systems are generally negligible in
terms of economic interest owing to the low sediment overburden, the lack of structural traps, and
often the poor source-rock potential (i.e., low organic carbon content). These systems do, however,
provide a "natural laboratory" for studying the processes of petroleum generation and the behavior
of petroleum in high-temperature fluids. The following data illustrate the diverse suite of
hydrothermal samples from different environments and the wide ranges of organic carbon source
material.
Guaymas Basin, Gulf of California
Guaymas Basin is an actively spreading oceanic basin, which is part of the system of spreading
axes and transform faults that extend from the East Pacific Rise to the San Andreas fault (Curray et
al. 1982; Lonsdale 1985). Ocean-plate accretion occurs by dyke and sill intrusions into the
unconsolidated sediments, leading to high conductive heat flow (Einsele et al. 1980; Curray et al.
1982; Einsele 1985; Lonsdale and Becker 1985). Sediments accumulate rapidly (>2 m/1000 yr) and
have covered the rift floors to a depth of S400 m (Curray et al. 1982). The organic matter of these
recent sediments is derived primarily from diatomaceous and microbial detritus, and the sediments
contain on the average about 2% total organic carbon. Influx of terrigenous organic matter is low
owing to the low continental runoff from the deserts that border the Gulf. Thermal stress causes
rapid maturation with concomitant petroleum generation and expulsion in sediments exposed to
temperatures in excess of about 50°C; the "oil window" seems to migrate upward as the magmatic
heat front rises in the sedimentary column (Simoneit 1984a; Simoneit et al. 1984). Petroleum
products have been characterized in samples obtained by shallow gravity coring in two troughs
(Simoneit et al. 1979), piston coring in the northern trough (Simoneit 1983) and deep coring by Leg
64 of the Deep Sea Drilling Project (DSDP) in the southern trough (Curray et al. 1982). Petroleum-

bearing samples have also been recovered from the seafloor by dredging operations (Simoneit and
Lonsdale 1982) and submersible sampling with the Alvin in 1982, 1985, and 1988 (figure 16.2;
Bazylinski, Farrington, and Jannasch 1988; Simoneit 1984a, b, 1985; Von Damm et al. 1985a;
Kawka and Simoneit 1987; Simoneit and Kawka 1987; Simoneit, Kawka, and Brault 1988).
Leg 64 of the DSDP encountered sills and hydrothermal alteration at depth in all holes drilled in the
basin (figure 16.2). Thermogenic hydrocarbon gas, H2S, and CO2 were identified at all sites. Lipids
(bitumen) were thermally altered close to and below the sills, especially at Site 477 in the southern
trough (Simoneit and Philp 1982; Simoneit et al. 1984). This alteration is indicated by: (1) loss of the
odd carbon number predominance of the n-alkanes (CPI approaches 1.0*); (2) appearance of a
broad hump of UCM (naphthenes); (3) isomerization of various biomarkers; (4) presence of large
amounts of terminal olefins (alk-1-enes), isoprenoid hydrocarbons and elemental sulfur; and (5)
appearance of PAH.
The petroleum products have migrated away from heat sources (e.g., sills or magma at depth) by
advection, diffusion, distillation, and especially by cosolution in hydrothermal fluids (Simoneit and
Philp 1982; Simoneit 1982b; Simoneit et al. 1984). Cosolution in this context refers to the solution in
any proportion of petroleum (bitumen), itself a solution of a large number of compounds, in
hydrothermal water, i.e., the stage before emulsion. The kerogen (i.e., insoluble higher molecular
weight organic matter in sediments) of the shallow DSDP core samples is typical of unaltered
marine organic matter (Simoneit and Philp 1982; Simoneit et al. 1984). In deeper sections (> S180
m depth at Sites 477 and 478), the kerogen reflects essentially complete expulsion of pyrolysate (i.
e., no additional hydrocarbons can be generated by pyrolysis); in these zones of high thermal stress
(entering greenschist facies, 300 50°C, Kastner 1982) the kerogen residues resemble activated
amorphous carbon (Simoneit 1982b), and on the basis of 13C data, this carbon is residual,
thermally matured marine organic carbon (Jenden, Simoneit, and Philp 1982; Simoneit et al. 1984).
Numerous hydrothermal mounds rise to 20-30 m above the southern trough floor (water depth
about 2000 m), and most are actively discharging vent fluids with water temperatures up to 315°C
at S 200 bars, i.e., 0.2 x 108 Pa (Lonsdale 1985; Lonsdale and Becker 1985; Merewether, Olsson,
and Lonsdale 1985). Typical samples from these mounds are stained and cemented with
petroleum; the samples have a strong odor reminiscent of diesel fuel (Simoneit and Lonsdale 1982).
Samples have very diverse petroleum contents and hydrocarbon distributions (figures 16.3, figure
16.4a; Simoneit 1984a, b, 1985; Kawka and Simoneit 1987; Simoneit and Kawka 1987; Simoneit,
Kawka, and Brault 1988) but are similar to those described for bitumens at depth in the DSDP holes
(Simoneit 1983, 1984b). The n-alkanes range from methane to greater than n-C40, with usual
maxima in the mid-C20 region and no carbon number predominance (CPI = 1.0, table 16.1). These
data and the kinetic parameters of the biomarkers indicate that the petroleums were generated by
rapid and intense heating.
An example of a gas chromatographic (GC) trace of an aromatic/naphthenic fraction (F2) of a

sample is shown in figure 16.4b. The major resolved peaks are PAH, a group of compounds
uncommon in petroleums but ubiquitous in higher temperature pyrolysis residues (Geissman, Sim,
and Murdoch 1967; Blumer 1975). The dominant compounds are the pericondensed aromatic
series, for example, phenanthrene, pyrene, benzopyrenes, perylene, benzoperylene, and coronene
(Kawka and Simoneit 1990). A pyrolytic origin is also supported by the presence of five-membered
alicyclic rings (e.g., acenaphthene, methylenephenanthrene, fluorene, fluoranthene, etc.). These
hydrocarbons are found in all pyrolysates from organic matter; once formed, they do not easily

revert to pericondensed aromatic hydrocarbons (Blumer 1975, 1976; Scott 1982). These fractions
also contain significant amounts of toxic PAH, e.g., the benzopyrenes (Kawka and Simoneit 1990).
In addition, perylene is present; it is the predominant PAH of unaltered lipids in sediments deposited
under oxygen-minimum environments in the Gulf (Simoneit 1982a, b; Simoneit and Philp 1982).
Thus, the chemical composition of the aromatic fractions indicates derivation through high-
temperature pyrolysis and rapid quenching, presumably by hydrothermal fluids.
Hydrothermal petroleum migration in Guaymas Basin seems to occur mainly as bulk-phase (pure
petroleum) and to a lesser extent as cosolute fluid and aqueous solution (high-temperature solution
of predominantly n-alkanes only) upward to the seabed, where it condenses (solidifies) and collects
in the conduits and vugs of hydrothermal mineral mounds in response to ambient temperatures.
PAH and sulfur accumulate in the hot vent regions; waxes crystallize in intermediate temperature
areas (20-80°C), and volatile petroleum partly collects in cold areas (3°C) and emanates into the
sea water as plume discharges (Simoneit 1984a,b, 1985; Merewether, Olsson, and Lonsdale 1985;
Simoneit et al. 1990). The activated amorphous carbon residue from the kerogen does not migrate
to the seafloor, although at greater subbottom depths in the DSDP holes this carbon is not evident,
indicating that it may have reacted to oxidative products (e.g., CS2, COS, etc.) or undergone limited
migration (Simoneit 1982b). Both the extensive maturation of organic matter to bitumen in the
DSDP holes and the significant accumulations of petroleum at the rift floor confirm the importance
of hydrothermalism as a feasible mechanism for the formation of petroleum.
Escanaba Trough, Northeastern Pacific
The Escanaba Trough represents the southern extension of the Gorda Ridge, an active oceanic
spreading center about 300 km long, bounded on the north and south by the Blanco and Mendocino
fracture zones, respectively (figure 16.1; McManus et al. 1970). The Trough is filled with up to 500
m of Quaternary turbidite sediments (McManus et al. 1970).
Petroleum cements the sediments and sulfide deposits that blanket the ridge axis and is also
derived from hydrothermal alteration of sedimentary organic matter (Kvenvolden et al. 1986).
Typical GC traces of the saturated and aromatic hydrocarbon fractions are shown in figure .16.5.
The organic source material for these petroleums is mainly terrigenous, on the basis of CPI, carbon-
number range (table 16.1), biomarker composition, and sedimentological considerations
(Kvenvolden and Simoneit 1990).
In the aliphatic hydrocarbon fraction, the n-alkanes range from C14 to C40, with a carbon-number
maximum at n-C27. A predominance of odd carbon numbers > n-C25 (CPI = 1.25, figure 16.5a) is
typical of a contribution from terrestrial, higher plants. Homologs of a marine origin (<n-C25) are
less concentrated. However, these may also have been preferentially biodegraded or expelled by
migration.
The PAH of these samples are mainly the pericondensed nonalkylated series with phenanthrene,
pyrene, and benzopyrenes as the predominant components (figure 16.5b). The overall PAH
compositions are quite variable for these samples, owing probably to the differential solubilities of
the PAH compounds in hot fluids and the temperature ranges of their formation (Simoneit 1984a).
This petroleum was probably generated by intense heating of short duration, as indicated by kinetic
parameters of the biomarkers and by the high concentrations of unsubstituted PAH (figure 16.5b,
Kvenvolden and Simoneit 1990). For example, the aromatic hydrocarbons were interpreted to have

formed at temperatures >400°C when compared with data from laboratory simulations (Kvenvolden
and Simoneit 1990).
Bransfield Strait, Antarctica
Bransfield Strait is a marginal basin located in a heavily sedimented back-arc ridge system (figure
16.1), which is both tectonically and hydrothermally active (Suess et al. 1990). Gravity cores
recovered on the R/V Polarstern cruise ANT IV/2 (1985) were sampled in order to evaluate the
effects of thermal stress on the sedimentary organic matter and to further substantiate the
preliminary data described from a prior cruise (Whiticar, Suess, and Wehner 1985). The general
core lithologies consist of olive-gray clay, which represents thermally unaltered sediment deposited
under generally oxic conditions and black clay/silt sections that are enriched in sulfide, owing
probably to fluid migration. The gray clays near the core bottoms had a slight petroliferous odor, but
migration of heavier hydrocarbons into the sections was not evident.
The major compositional differences of the lipids between the upper and lower sections (subbottom
depths of 5-7 m) of the Bransfield Strait sediments appear to result from an increased rate of
diagenesis in the lower sections (figure 16.6 a,b; Brault and Simoneit 1988). There the temperature
seems to have been higher owing to mild hydrothermal stress, probably from low-temperature fluid
transgression into the sedimentary section. The n-alkane distributions are similar, but the total
concentrations increase slightly with depth. The generation of phytane and other isoprenoids from
phytol via phytenes is evident. Archaebacterial isoprenoids do not appear to be an important input.
Biomarker signature changes in the lower sections include the disappearance of diploptene and the
appearance of hop-17(21)-ene, suggesting diagenetic migration of the double bond.
High abundances of steroidal components in the lipids reflect high productivity and good
preservation (Brault and Simoneit 1988). The 3,5-steradiene (C28) is the dehydration product of
the 5-sterol counterpart. The 2-sterene series is present and seems to have isomerized in the
lower sections to the 4- and 5-sterenes, which are more stable. Diasterenes with the 20R
configuration formed by rearrangement from the sterenes.
Thus, most of the lipid compounds identified at depth in the cores are derived from accelerated
diagenesis of primary biogenic components. The dehydrations of phytol to phytenes and sterols to
sterenes and steradienes are the major indicators of lipid diagenesis with depth, enhanced by mild
hydrothermal stress (Comet, this volume). The minor complex components (e.g., figure 16.6b)
probably represent water-soluble and/or volatile compounds, including gasoline range and aromatic
hydrocarbons (Whiticar, Suess, and Wehner 1985; Whiticar, private communication), brought into
the deeper sections by warm, but not hot, fluid transgression.
East Pacific Rise, 13°N and 21°N
Hydrothermal activity and associated massive sulfide deposits are found on the unsedimented axis
of the East Pacific Rise (EPR) in the region of 13°N (figure 16.1). Abundant faunal communities are
also associated with this activity (Hékinian et al. 1983). Aliphatic hydrocarbons have been analyzed
in hydrothermal plumes and in metalliferous sediments near the active vents and at the base of an
inactive chimney (Brault et al. 1985, 1988). Hydrocarbons from metalliferous sediments have
characteristics of immature organic matter, which was recently biosynthesized and microbiogically

degraded, as indicated by the abundance of low-molecular-weight (C25) n-alkanes and phytane; a

contribution of continental higher plant material is shown by the presence of high-molecular-weight
n-alkanes with an odd carbon number predominance (figure 16.6c, table 16.1). The immature
character of the organic matter is also indicated by the presence of biomarker hydrocarbons derived
from steroids and triterpenoids, which are the result of low-temperature alteration, as might be
expected in the talus of an extinct vent system. The contents of a sediment trap deployed in the
area within S20 m of the vents are characterized by biologically derived material and also by
biomarker compounds affected by thermal stress (Brault et al. 1985), which indicates that higher
temperature degradation of entrained organic detritus is an important process near hydrothermal
discharge sites. Thermally matured compounds (figure 16.6d) are also present at trace levels in
waters collected within S1 km above the hydrothermal vents. The hydrocarbon pattern of these
waters is indicative in many cases of pyrolysis of bacterial matter present in the ocean water that is
entrained during the turbulent cooling of the discharging vent fluids (Brault et al. 1988).
Extensive hydrothermal activity and associated sulfide deposits have also been described at 21°N
on the EPR (figure 16.1) where the crust is unsedimented (Spiess et al. 1980; Ballard et al. 1981;
Von Damm et al. 1985b). The hydrocarbon contents of massive sulfides from vent chimneys are
extremely low (table 16.1) but definitely thermogenic. A GC trace for total hydrocarbons is shown in
figure 16.6e (Brault, Simoneit, and Saliot 1989). The n-alkanes in massive sulfides range from C14
to greater than C40, with no carbon number predominance. The n-alkanes in sulfide with a pyritized
tube worm from a chimney have a slight odd carbon number predominance. All samples contain
PAH, providing evidence for hydrothermal activity. These data, coupled with the carbon-number
maxima at n-C27 or higher, indicate that the hydrocarbons were entrapped/condensed in a high-
temperature regime such as an active chimney. These hydrocarbons were, however, subjected to
these high temperatures for only a short time. The sample with pyritized tube worm residues (figure
16.6e) also contains hydrothermally altered derivatives of biomarkers (e.g., cholestenes, hopenes)
from the vent biota, i.e., mainly tube worms and bacteria. These biomarkers are not present in the
vent biota (Brault and Simoneit, unpublished data).
Atlantis II Deep, Red Sea
The Atlantis II Deep (figure 16.1) contains stratified brine layers, the deepest of which is at a
temperature of 62°C (Hartmann 1980, 1985). Bulk organic matter and hydrocarbons have been
analyzed in two sediment cores (No. 84 and 126, CHAIN 61 cruise) from the Atlantis II Deep
(Simoneit et al. 1987). Although the brine overlying the coring areas is reported to be sterile,
sedimentary organic material derived from autochthonous marine planktonic and microbial inputs
and minor terrestrial sources is present. The organic input derived from the water column above the
brine is further metabolized by microorganisms, and the reworked compounds with organic detritus
are apparently then incorporated into the sediments under the brine by sinking adsorbed or bound
together with particles of metallic oxide precipitates (Simoneit et al. 1987).
Low-temperature maturation in the sediments results in petroleum generation, even from low
amounts of organic matter (average 0.14 wt. percent total organic carbon [TOC]). Both steroid and
triterpenoid hydrocarbons (biomarkers, especially neohop-13(18)-enes) show that extensive acid-
catalyzed reactions are occurring in the sediments (Simoneit et al. 1987). In comparison with other
hydrothermal systems (e.g., Guaymas Basin; Cape Verde Rise, Simoneit et al. 1981), sediments in
the Atlantis II Deep exhibit a lower degree of thermal maturation, as is clearly shown by the
elemental composition (H/C and N/C) of the kerogens and the absence of pyrolytic PAH in the

bitumen.
The lack of carbon-number preference among the n-alkanes (CPI = 1.0) suggests, especially in the
case of the long-chain homologs (e.g., figure 16.7a, table 16.1), that the organic matter has been
affected by catagenesis or that it never had an odd carbon number predominance. However, the
yields of hydrocarbons with respect to sediment weight are much lower than those observed in
other hydrothermal areas. The low temperature and low organic carbon content of the source
sediments in the Atlantis II Deep seem to be responsible for this difference.
Mid-Atlantic Ridge, TAG Area 26°N
The Trans-Atlantic Geotraverse (TAG) hydrothermal field on the Mid-Atlantic Ridge crest at 26°N is
one of two active vent systems known for slow-spreading oceanic ridges (Rona et al. 1984;
Thompson et al. 1988). Hydrothermal deposits lying directly on oceanic crust have been dredged
from the area (TAG 1985-1). Three sulfide-rich samples, consisting mainly of anhydrite, sphalerite,
and chalcopyrite, respectively contained minor amounts of the more volatile (C10-C22)
hydrothermal petroleums (Brault and Simoneit 1989). The saturated and aromatic hydrocarbon
fractions separated from the extract of the sphalerite are shown in figure 16.7 b,c. The n-alkanes
range from C11 to C22 with a CPI = 1.0; pristane and phytane are present, and the UCM maximizes
at the GC retention time for n-C17. This alkane pattern is analogous to that observed for samples
from the EPR at 13°N and is derived from autochthonous marine organic matter. The aromatic
fraction, which contains naphthalene, phenanthrene, their alkyl homologs, and sulfur aromatic
compounds, supports a hydrothermal origin from marine organic matter.
Hydrous Pyrolysis Simulation
Because hydrothermal petroleums are admixtures of products generated over a wide temperature
window, it is important to try to understand the processes through laboratory simulations. A first
attempt using thermally unaltered mud from the seabed of Guaymas Basin (sample 1176-PC2, 10-
20 cm) is described here. The hydrous pyrolysis was carried out in a chromium-lined, stainless steel
vessel with 100 g mud (with 34% pore water) plus 500 g simulated seawater. The temperature
program was 20-330°C over 5 hours, at 330°C for 1 hour, and cooling to 30°C over 2 hours. The
pyrolysis products were extracted from the aqueous and solid phases, separated by liquid
chromatography and analyzed by GC and GC-MS with the conventional procedures.
Homologous Compounds
The n-alkanes generated by the simulation are shown in figure .16.8c and range from C14-C36 with
a minor odd carbon number predominance (CPI = 1.09) and Cmax at C21. This distribution pattern
is completely unlike that of the unaltered sediment (e.g., sample 30G-I, 102-105 cm, figure 16.8a)
but does compare well with that of a hydrothermal petroleum (sample 1172-4, figure 16.8b).
Pristane and phytane are generated with a Pr/Ph of 0.56, which is lower than that for many of the
hydrothermal petroleums, where this ratio is about 1.0. However, the ratios of Pr/n-C17 and Ph/n-
C18 are 0.59 and 0.93, respectively, for the simulation, which more closely approach the same
ratios for the hydrothermal petroleums (e.g., 1.09 and 0.95, respectively, for sample 1172-4). The
naphthenic hump or UCM of the hydrothermal petroleums is generally broad, extending over the full
GC retention range, whereas the lipids of the unaltered sediments exhibit a narrow UCM centered

at about C19 elution time (figure 16.8a, b). The UCM of the simulation bitumen is similar to that
observed for the hydrothermal petroleums.
Most of the hydrothermal petroleums contain a full suite of alkylmonoaromatic hydrocarbons

composed mainly of n-alkylbenzenes, n-alkyltoluenes, and n-alkylxylenes, ranging from C16 to C31
with a Cmax at C23 and no carbon number predominance (figure 16.9a-c). These compounds are
not detectable in the lipids of the unaltered sediment. The simulation bitumen shows n-
alkylbenzenes and n-alkyltoluenes, ranging from C13 to C30 with a lower Cmax at C16 and a slight
odd carbon number predominance (figure 16.9d, e).
Biomarkers
The triterpenoid and steroid hydrocarbon biomarkers are summarized in figure 16.10. The unaltered
sediment contains primarily triterpenes, the 17 (H), 21 (H)-hopane (figure 16.10a) and sterenes ( 2
and 4). A minor amount of the 17 (H), 21 (H)-hopane ( ) series is also present in this repeat
analysis carried out in 1988. The previous GC-MS analysis of 1978 did not exhibit the series
Simoneit et al. 1979; (Kawka and Simoneit 1987). Over the ten years of storage the extract has
changed, losing a major amount of the triterpenes and generating the -hopanes. However, this
does not invalidate the simulation data, because new sample material was used for the
experiments. The mature hydrothermal petroleums contain primarily the -hopane series, ranging
from C27 to C34 (no C28 and traces of C35), with essentially complete maturity for the extended
homologs (C31 22S/R S 1.3, figure 16.10b) (Kawka and Simoneit 1987).
The triterpanes of the simulation experiment are of an intermediate maturity (figure 16.10c). The -
hopanes are predominant (thermally most stable configuration), but a significant amount of the
series remains and a major amount of the intermediate 17 (H),21 (H)-hopane series ( ,
moretanes) is also present. Note that -norhopane is the dominant triterpane in the simulation
sample (C29/C30 = 1.7) as compared with -hopane in the hydrothermal petroleums (C29/C30 =
0.45), and the C29 and C30 moretanes are also abundant in the simulation sample. This
observation appears to indicate a mixture of triterpane biomarkers partially converted to full
maturity. The extended -hopane isomer ratios for C31 and C32 22S/R are 0.3 and 0.5,
respectively, also indicating the immature nature of the triterpanes from the simulation (Ensminger
et al. 1974).
The steranes of the hydrothermal petroleums consist of the C27-C29 homologs ranging in maturity
from dominantly the 5 (H), 14 (H), 17 (H)-20R and lesser amounts of the 5 (H) configuration to
the more rearranged assemblage (e.g., figure .16.10d; Kawka and Simoneit 1987). Diasteranes are
also prominent in the more mature samples (e.g., figure 18.1d). The steranes of the 330°C hydrous
pyrolysis consist mainly of the relatively immature assemblage of the 5 (H), 14 (H), 17 (H)-20R
and the lesser 5 (H) series (figure .16.10e). Diasteranes are not detectable, which may indicate that
clay catalysis by montmorillonite as simulated for Jouy shale (Rubinstein, Sieskind, and Albrecht
1975) is not active in this sediment sample.
Polynuclear Aromatic Hydrocarbons
The hydrothermal petroleums of Guaymas Basin contain PAH consisting primarily of the

unsubstituted parent compounds (Simoneit 1984a; Kawka and Simoneit 1990), whereas the
unaltered sediments have only perylene from diagenetic sources (Simoneit et al. 1979). The PAH of
the simulation experiment are compared with the hydrothermal oils in figures 16.11 and 16.12.
Phenanthrene and anthracene as well as the alkylphenanthrenes and methylenephenanthrene are
generated in the pyrolysis (figure 16.11d-f) with relative concentrations that are similar but not
identical to typical Guaymas Basin hydrothermal petroleum (figure 16.11a-c). Hydrothermal
petroleums from Escanaba Trough also contain PAH (Kvenvolden and Simoneit 1990), and the
signature of these tricyclic PAH is different (figure 16.11g-i). Anthracene is not detectable and
methylenephenanthrene is more prominent, indicating a higher formation temperature (Kawka and
Simoneit 1990). Diels' hydrocarbon [1,2-(3'-methylcyclopenteno)phenanthrene] was generated in
the simulation in amounts comparable to that found in the Guaymas petroleums. This result
indicates that aromatization of biomarker precursors (e.g., steroids) proceeds rapidly (Simoneit et
al. 1990).
The higher molecular weight PAH are present in the hydrothermal petroleums from Guaymas Basin
and Escanaba Trough (Simoneit 1984a; Kawka and Simoneit 1990; Kvenvolden and Simoneit
1990). For example, the m/z 252 group of benzofluoranthene, benz(e)pyrene, benz(a)pyrene, and
perylene is a significant component in the oils (figure 16.12a-c). Only perylene is a major member of
the m/z 252 series in the oil from the simulation (figure 16.12d), which indicates that higher
temperatures and/or longer heating times are necessary to generate such heavy PAH and that the
generation of perylene from its unknown precursors was enhanced by the conditions of this
simulation.
Discussion
Organic Matter Type
The constitution of the initial organic matter determines the types of petroleum products that form in
basins with a normal geothermal gradient and those with hydrothermal systems. The major source
of petroleum compounds is kerogen, the sedimentary macromolecular organic detritus, which
generally constitutes the bulk of the total organic carbon content (Hunt 1979; Tissot and Welte
1984). In general, terrestrial detritus from mainly vascular plants yields an aromatic kerogen (e.g.,
coal) that has a natural gas potential, whereas marine/lacustrine organic matter from primarily
microbial and planktonic residues yields an aliphatic kerogen (e.g., sapropel) that has a paraffinic
petroleum potential (Hunt 1979; Tissot and Welte 1984). Kerogens in sedimentary basins are
generally mixtures of these inferred endmembers.
Syngenetic sedimentary lipid matter undergoes alteration during diagenetic and early catagenetic
processes that changes the hydrocarbon signature. During oil generation, however, large amounts
of additional hydrocarbons are superimposed on these syngenetic lipids, thus obscuring the earlier
signature (e.g., loss or reduction of the odd carbon number predominance of n-alkanes > C25, cf.
figure 16.8a vs. b). Syngenetic lipids of sediments can usually be utilized to elucidate the various
sources of the total organic matter (e.g., Simoneit 1978, 1981, 1982a). For example, lipid
hydrocarbons of normal sediment in Guaymas Basin (figure 16.8a) contain n-alkanes, mostly <
C21, and a minor series of homologs > C23 with an odd carbon number predominance. The lower
carbon number composition is typical of a predominantly marine planktonic and microbial origin with
a minor influx of terrigenous vascular plant wax (> C25). By contrast, lipid hydrocarbons of sediment

in the Escanaba Trough (Kvenvolden et al. 1986; Simoneit 1988) contain n-alkanes, mostly > C23
with an odd carbon number predominance. This signature is probably derived mainly from
terrigenous vascular plant wax, with a minor microbial component (Kvenvolden et al. 1986). In both
the Guaymas and Escanaba sediment samples, the compositions of the kerogens support these
interpretations based on the nature of the lipid hydrocarbons (Simoneit 1978; Simoneit et al. 1979).
Petroleum Generation
The principal zone of petroleum formation in sedimentary sequences under normal geothermal
gradients extends from about 1 to 3 km depth (e.g., Hunt 1979; Tissot and Welte 1984). This depth
corresponds to a temperature range of 50°-120°C. The effect of pressure on this process has not
been quantified (Tissot and Welte 1984). The further cracking of organic matter to natural gas is
thought to take place at higher temperatures of 150°-250°C, although gas is also generated within
the "oil window" (e.g., Kartsev et al. 1971; Vassoevich, Akramkhodzhaev, and Geodekyan 1974;
Hunt 1979). These proposed temperature regimes for the oil and gas "windows" need some upward
adjustment to accommodate the recent data on hydrocarbon presence at higher temperatures in
hydrothermal systems and ultradeep wells.
The "instantaneous" petroleum generation in hydrothermal systems is a natural and efficient

process, which occurs at temperatures approaching a maximum of 400°C. At such high
temperatures, organic matter is only partly destroyed, probably because the thermogenic products
are rapidly removed from the hot zone. Formation of hydrothermal petroleum seems to commence
in low-temperature areas, generating the more aliphatic products (Simoneit, Kawka, and Brault
1988), and as the temperature regime rises, products are derived from more refractory organic
matter and may even be "resynthesized" (e.g., PAH compounds).
For example, the phenanthrene-to-methylphenanthrenes (P/MP) ratio can be used to estimate

formation temperatures for these compounds by comparison with laboratory data (figure 16.11).
The P/MP for the Guaymas Basin oils range from 0.12 to 0.95, with an average of 0.58 (excluding a
low and high outlier), and for the Escanaba Trough oils, from 2.7 to 3.3 (Kawka and Simoneit 1990;
Kvenvolden and Simoneit 1990). These ratios can be compared with data from laboratory
maturation studies of immature Tanner Basin (California Bight) kerogen (Ishiwatari and Fukushima
1979), a process analogous to hydrothermal organic matter alteration. The phenanthrene-to-
methylphenanthrene ratios ranged from 0.59-0.83 at 310°C for 18 to 100 hours, and 0.9-5.0 at 410°
C for 5 to 32 hours. Based on this ratio, the GB oils fit with a temperature window of around 310°C
and the ET oils with temperatures approaching 410°C.
Organic matter associated with deeper hydrothermal systems (e.g., epithermal ores in volcanic
terranes) is typically more asphaltic and has a high PAH content. Such organic matter is widely
distributed (e.g., California mercury deposits: Geissman, Sim, and Murdoch 1967; Blumer 1975;
other hydrothermal sulfides: Germanov and Bannikova 1972). Deep well drilling (>7000 m) has
intersected Cretaceous shales that were at in situ temperatures of about 260°-300°C (e.g., Price,
Clayton, and Rumen 1981; Price 1982). These samples were rich in bitumen components and their
kerogens still had significant hydrocarbon generation potential. These observations indicate that in
situ petroleum is stable at much higher temperatures and pressures than those discussed above
and probably also over longer geologic time periods.
The major similarities and differences between hydrothermal petroleums and reservoir petroleums

have been summarized (Simoneit 1988) and are briefly recapped here. table 16.2 presents this
summary in terms of compound classes and bulk parameters. Most hydrocarbon products occur in
both types of petroleum; a major difference is the enhanced content of PAH and sulfur-PAH in the
hydrothermal products.
Migration Processes
Petroleum migration in conventional sedimentary basins is proposed to occur mainly by oil phase
flow, and migration in water solution and/or by molecular diffusion is of minor importance (Durand
1988). However, in hydrothermal systems the migration of petroleum occurs by a different
combination of the processes mentioned above, and the following laboratory measurements on
petroleum solubility in water (Price 1976; Price et al. 1983) may be applicable. The aqueous
solubility of petroleums and various hydrocarbon fractions increases exponentially from 100°C to
180°C, and solubilities are high enough to possibly account for the formation of petroleum
reservoirs by migration of molecular or cosolutions (Price 1976). Increased salinities of 150 NaCl
cause dramatic exsolution of the petroleum and at 350 almost total "salt-out" occurs (Price 1976).
This finding is consistent with the requirement for the separation of petroleum from migrating
solutions in the salty waters of reservoir sands. It has been demonstrated that methane in the
presence of water is an even better carrier for petroleum than water or methane alone (Price et al.
1983). Both increases in pressure (to about 1800 bar, i.e., 1.8 x 108 Pa) and in temperature (to 250°
C) raised the solubility of petroleum. Cosolubility was found at rather moderate conditions (e.g., 100°
C at 108 Pa, 200°C at 0.5 x 108 Pa). The addition of other gases such as CO2 and ethane to this
mixture also increases the solubility of petroleum. Under these experimental conditions, which are
similar to those in hydrothermal systems, CH4, C2H6, and CO2 are all supercritical, and H2O
approaches its near-critical state; thus, the gases and H2O are all mutually soluble by the reduced
hydrogen bonding in water and are a good cosolvent for petroleum. Therefore, in hydrothermal
systems petroleum expulsion and primary migration appear to proceed as gas/fluid and aqueous
solution phases and secondary migration as bulk phase and to a lesser extent in cosolution,
emulsion, and by molecular diffusion (Simoneit 1985, 1988; Kawka and Simoneit 1987; Simoneit,
Kawka and Brault 1988; Didyk and Simoneit 1989; Simoneit et al. 1990).
Simulation
The results of the laboratory hydrothermal alteration of Guaymas Basin sediment are summarized
in table 16.3. Generally, most of the compounds found in the hydrothermal petroleums were
generated by the simulation experiment. This has also been observed for other hydrous pyrolysis
experiments where the match of crude oil to pyrolysis products from the source rock is essentially
identical (e.g., Lewan 1985; Lewan, Bjorøy, and Dolcater 1986). Thus, the bulk of the hydrothermal
petroleum compounds (i.e., n-alkanes, naphthenes, isoprenoids) form readily over a range
commencing at low temperatures to the high temperatures of this simulation. The hopane and
sterane biomarkers were not converted to the fully mature configurations, and sterane maturation
was slower than hopane maturation. This is probably due to the brief heating time and is as
expected from other studies.
The alkylarenes appeared at the onset of generation in the simulation, and the lower molecular
weight PAH, such as for example, the phenanthrene series, are also fully represented. Triaromatic
steroid hydrocarbons, including Diels' hydrocarbon, were generated in the hydrous pyrolysis.
However, the higher molecular weight PAH, excluding perylene, were not generated in the

simulation. This indicates that aromatization of steroids and other tricyclic precursors proceeds
readily, but higher temperatures, longer contact times, and/or unknown reactions are required to
form the heavy PAH.
Recent immature sediments receive biogenic detritus, which upon deposition, undergoes diagenetic
and additional microbial alteration. Increasing burial in sedimentary basins results in the onset of
organic matter maturation, which generates some volatile products from the kerogen (easily
cracked moieties) that become added to the endogenous lipid residues. This is the beginning of
petroleum formation. As the depth of burial (i.e., temperature) increases, catagenesis commences
and major petroleum generation takes place. At still greater depths of burial the metagenetic stage
is reached, where extensive cracking, disproportionation, and reforming of the organic matter, both
petroleum and kerogen residues, occur to yield primarily gases and residual carbon.
In the case of hydrothermal systems, these processes are compressed into an "instantaneous"
geological time frame. At seafloor spreading axes, hydrothermal systems operating below a
sediment blanket (e.g., Guaymas Basin and Escanaba Trough) generate petroleum from generally
immature organic matter in the sediments. This petroleum then migrates upward, leaving behind a
spent carbonaceous residue (amorphous "activated" carbon). The Guaymas petroleums are
composed of: (1) gasoline-range hydrocarbons (C1-C12); (2) a broad distribution of n-alkanes (C12-
C40+) with no carbon number predominance; (3) a naphthenic hump, UCM; (4) pristane and
phytane at significant concentrations; (5) mature biomarkers (e.g., -hopanes); and (6) significant
concentrations of PAH and sulfur. Exposed petroleum and petroleums in unconsolidated surface
sediments are microbially degraded and leached, whereas interior samples are essentially
unaltered.
Hydrothermal systems operating in unsedimented rift areas (e.g., East Pacific Rise at 13°N and 21°
N, Mid-Atlantic Ridge at 26°N) generate trace amounts of petroleum. Low amounts of petroleum are
generated by pyrolysis of suspended and dissolved biogenic organic detritus (including bacteria and
algae) entrained during the turbulent cooling of the vents and in the talus for both sedimented and
bare rock systems. However, this type of petroleum is overwhelmed in sedimented systems by the
large quantity of petroleum generated from the sedimentary organic matter. In addition, low-level
maturation is observed in the surrounding area at vent sites, owing probably to warming of ambient
detritus.
The preliminary laboratory simulation (330°C hydrous pyrolysis) of sediment alteration indicates that
the products are intermediate in terms of maturity between source sediment and hydrothermal
petroleum. The duration of heating was too brief to convert all precursors to their end products. The
major simulation products are the same as in the hydrothermal oils. However, the biomarkers are
not as mature and heavy PAH have not yet formed. The unaltered sediment contains immature
lipids and no PAH, except for some perylene. Thus, hydrothermal petroleums at the seabed in
active rift zones under a sedimentary cover are accumulated mixtures of low- to high-temperature
hydrous pyrolysis products.
Acknowledgments
I thank the Deep Sea Drilling Project for access to DSDP samples, and the crews and pilots of the
D.S.V. Alvin, R. V. Lulu, and R. V. Atlantis II for their skillful recoveries of hydrothermal samples.
Samples, data, and assistance were provided by C. A. Allen, J. Baross, M. Brault, A. Grey, P. D.

Jenden, O. E. Kawka, K. A. Kvenvolden, R. N. Leif, P. F. Lonsdale, A. Lorre, M. A. Mazurek, R. P.

Philp, P. A. Rona, E. Ruth, G. Stormberg, E. Suess, and G. M. Wang. The hydrous pyrolysis
experimentation was carried out in collaboration with the Idaho National Engineering Laboratory,
EG&G Idaho, Inc., Idaho Falls, Idaho. Funding from the Division of Ocean Sciences, National
Science Foundation (Grants OCE81-18897, OCE-8312036, OCE-8512832, and OCE-8601316) is
gratefully acknowledged.
References
Ballard, R. D., J. Francheteau, T. Juteau, C. Rangan, and W. Normark. 1981. East Pacific
Rise at 21°N: The volcanic, tectonic and hydrothermal processes of the central axis. Earth
Planet. Sci. Lett. 55:1-10.
Bazylinski, D. A., J. W. Farrington, and H. W. Jannasch. 1988. Hydrocarbons in surface

sediments from a Guaymas Basin hydrothermal vent site. Org. Geochem. 12:547-558.
Blumer, M. 1975. Curtisite, idrialite and pendletonite, polycyclic aromatic hydrocarbon

minerals: their composition and origin. Chem. Geol. 16:245-256.
Blumer, M. 1976. Polycyclic aromatic compounds in nature. Scientific American 234(3):34-

45.
Brault, M. and B. R. T. Simoneit. 1988. Steroid and terpenoid distributions in Bransfield Strait
sediments: hydrothermally enhanced diagenetic transformations. In Advances in Organic
Geochemistry 1987. Org. Geochem. 13:697-705.
Brault, M. and B. R. T. Simoneit. 1989. Trace petroliferous organic matter associated with
hydrothermal minerals from the Mid-Atlantic Ridge at the Trans-Atlantic Geotraverse 26°N
Site. J. Geophys. Res. 94:9791-9798.
Brault, M., B. R. T. Simoneit, and A. Saliot. 1989. Trace petroliferous organic matter
associated with massive hydrothermal sulfides from the East Pacific Rise at 13°N and 21°N.
Oceanologica Acta 12:405-415.
Brault, M., B. R. T. Simoneit, J. C. Marty, and A. Saliot. 1985. Les hydrocarbures dans le
systeme hydrothermal de la ride Est-Pacifique, a 13°N. C.R. Acad. Sci. Paris 301, (II):807-
812.
Brault, M., B. R. T. Simoneit, J. C. Marty, and A. Saliot. 1988. Hydrocarbons in waters and
particulate material from hydrothermal environments at the East Pacific Rise, 13°N. Org.
Geochem. 12:209-219.
Cooper, J. E. and E. E. Bray. 1963. A postulated role of fatty acids in petroleum formation.
Curray, J. R., D. G. Moore, et al. 1982. Initial Reports. DSDP, 64, parts 1 and 2. Washington,
D.C.: U.S. Govt. Printing Office.

Didyk, B. M. and B. R. T. Simoneit. 1989. Hydrothermal oil of Guaymas Basin and

implications for petroleum formation mechanisms. Nature 342:65-69.
Didyk, B. M., B. R. T. Simoneit, S. C. Brassell, and G. Eglinton. 1978. Organic geochemical

indicators of paleoenvironmental conditions of sedimentation. Nature 272:216-222.
Durand, B. 1988. Understanding of HC migration in sedimentary basins (present state of

knowledge). In L. Mattavelli and L. Novelli, eds., Advances in Organic Geochemistry 1987.
Einsele, G. 1985. Basaltic sill-sediment complexes in young spreading centers: genesis and
significance. Geology 13:249-252.
Einsele, G., J. Gieskes, J. Curray, D. Moore, E. Aguayo, M. P. Aubry, D. J. Fornari, J. C.

Guerrero, M. Kastner, K. Kelts, M. Lyle, Y. Matoba, A. Molina-Cruz, J. Niemitz, J. Rueda, A.
Saunders, H. Schrader, B. R. T. Simoneit, and V. Vacquier. 1980. Intrusion of basaltic sills
into highly porous sediments and resulting hydrothermal activity. Nature 283:441-445.
Ensminger, A., A. van Dorsselaer, C. Spyckerelle, P. Albrecht, and G. Ourisson. 1974.

Pentacyclic triterpanes of the hopane type as ubiquitous geochemical markers: Origin and
significance. In B. Tissot and F. Bienner, eds., Advances in Organic Geochemistry 1973, pp.
245-260, Paris: Technip.
Geissman, T. A., K. Y. Sim, and J. Murdoch. 1967. Organic minerals. Picene and chrysene
as constituents of the mineral curtisite (idrialite). Experientia 23:793-794.
Germanov, A. I. and L. A. Bannikova. 1972. Alteration of organic matter of sedimentary rocks

during hydrothermal sulfide concentration. Dokl. Akad. Nauk SSSR 203:1180-1182 (in
Russian).
Hartmann, M. 1980. Atlantis II Deep geothermal brine system. Hydrographic situation in

1977 and changes since 1965. Deep Sea Res. 27:161-171.
Hartman, M. 1985. Atlantis II Deep geothermal brine system. Chemical processes between
hydrothermal brines and Red Sea deep water. Mar. Geol. 64:157-177.
Hékinian, R., M. Fevrier, F. Avedik, P. Cambon, J. L. Charlou, H. D. Needham, J. Raillard, J.

Boulegue, L. Merlivat, A. Moinet, S. Manganini, and J. Lange. 1983. East Pacific Rise near
13°N: Geology of new hydrothermal fields. Science 219:1321-1324.
Hunt, J. M. 1979. Petroleum Geochemistry and Geology. San Francisco: W. H. Freeman.
Ishiwatari, R. and K. Fukushima. 1979. Generation of unsaturated and aromatic

hydrocarbons by thermal alteration of young kerogen. Geochim. Cosmochim. Acta 43:1343-

1349.
Jenden, P. D., B. R. T. Simoneit, and R. P. Philp. 1982. Hydrothermal effects on

protokerogen of unconsolidated sediments from Guaymas Basin, Gulf of California:
Elemental compositions, stable carbon isotope ratios and electronspin resonance spectra. In
J. R. Curray, D. G. Moore, et al., eds., Initial Reports. DSDP 64:905-912. Washington, D.C.:
Johns, R. B., ed. 1986. Biological Markers in the Sedimentary Record, Methods in
Geochemistry and Geophysics 24. Amsterdam: Elsevier.
Kartsev, A. A., N. B. Vassoevich, A. A. Geodekian, S. G. Neruchev and V. A. Sokolov. 1971.

The principal stage in formation of petroleum. Proc. 8th World Petrol. Congr. 2:3-11.
Kastner, M. 1982. Evidence for two distinct hydrothermal systems in the Guaymas Basin. In
J. R. Curray, D. G. Moore, et al., eds., Initial Reports. DSDP 64:1143-1157. Washington, D.
Kawka, O. E. and B. R. T. Simoneit. 1987. Survey of hydrothermally generated petroleums

from the Guaymas Basin spreading center. Org. Geochem. 11:311-328.
Kawka, O. E. and B. R. T. Simoneit. 1990. Polycyclic aromatic hydrocarbons in hydrothermal

petroleums from the Guaymas Basin spreading center. In B. R. T. Simoneit, ed., Organic
Matter Alteration in Hydrothermal Systems-Petroleum Generation, Migration and
Biogeochemistry. Appl. Geochem. 5:17-27.
Kvenvolden, K. A. and B. R. T. Simoneit. 1990. Hydrothermally derived petroleum: examples

from Guaymas Basin, Gulf of California and Escanaba Trough, Northeast Pacific Ocean.
AAPG Bull. 74:223-237.
Kvenvolden, K. A., J. B. Rapp, F. D. Hostettler, J. L. Morton, J. D. King, and G. E. Claypool.

1986. Petroleum associated with polymetallic sulfide in sediment from Gorda Ridge. Science
234:1231-1234.

experimentation. Phil. Trans. Roy. Soc. Lond. A315:123-134.
Lewan, M. D., M. Bjorøy, and D. L. Dolcater. 1986. Effects of thermal maturation on steroid
hydrocarbons as determined by hydrous pyrolysis of Phosphoria Retort Shale. Geochim.
Lonsdale, P. 1985. A transform continental margin rich in hydrocarbons, Gulf of California.

AAPG Bull. 69:1160-1180.
Lonsdale, P. and K. Becker. 1985. Hydrothermal plumes, hot springs, and conductive heat
flow in the Southern Trough of Guaymas Basin. Earth Planet. Sci. Lett. 73:211-225.

McManus, D. A., R. E. Burns, et al. 1970. Initial Reports of the Deep Sea Drilling Project, 5.
Merewether, R., M. S. Olsson, and P. Lonsdale. 1985. Acoustically detected hydrocarbon

plumes rising from 2-km depths in Guaymas Basin, Gulf of California. J. Geophys. Res.
90:3075-3085.
Price, L. C. 1976. Aqueous solubility of petroleum as applied to its origin and primary
migration. AAPG Bull. 60:213-244.
Price, L. C. 1982. Organic geochemistry of core samples from an ultra-deep hot well (300°C,
7 km). Chem. Geol. 37:215-228.
Price, L. C., J. S. Clayton, and L. L. Rumen. 1981. Organic geochemistry of the 9.6 km
Bertha Rogers No. 1 well, Oklahoma. Org. Geochem. 3:59-77.
Price, L. C., L. M. Wenger, T. Ging, and C. W. Blount. 1983. Solubility of crude oil in
methane as a function of pressure and temperature. Org. Geochem. 4:201-221.
Rona, P. A. 1988. Hydrothermal mineralization at oceanic ridges. In T. J. Barrett and J. L.

Jambor, eds., Seafloor Hydrothermal Mineralization. Min. Assoc. Canada. Can. Min. 26:431-
465.
Rona, P. A., G. Thompson, M. J. Mottl, J. A. Karson, W. J. Jenkins, D. Graham, M. Mallette,

K. von Damm, and J. M. Edmond. 1984. Hydrothermal activity at the Trans-Atlantic
Geotraverse hydrothermal field, Mid-Atlantic Ridge Crest at 26°N. J. Geophys. Res.
89:11365-11377.
Rubinstein, I., O. Sieskind, and P. Albrecht. 1975. Rearranged sterenes in a shale:

occurrence and simulated formation. J. Chem. Soc., Perk. Trans. 1:1833-1876.
Scott, L. T. 1982. Thermal rearrangements of aromatic compounds. Acc. Chem. Res. 15:52-
58.
Simoneit, B. R. T. 1978. The organic chemistry of marine sediments. In J. P. Riley and R.

Chester, eds., Chemical Oceanography 7:233-311. London, Academic.
Simoneit, B. R. T. 1981. Utility of molecular markers and stable isotope compositions in the
evaluation of sources and diagenesis of organic matter in the geosphere. In A. Prashnowsky,
ed., The Impact of the Treibs' Porphyrin Concept on the Modern Organic Geochemistry, pp.
133-158, Würzburg: Bayerische Julius Maximilian Universität.
Simoneit, B. R. T. 1982a. The composition, sources and transport of organic matter to

marine sediments-the organic geochemical approach. In J. A. J. Thompson and W. D.
Jamieson, eds., Proceedings of the Symposium on Marine Chemistry into the Eighties, pp.
82-112, Natl. Res. Council of Canada.

Simoneit, B. R. T. 1982b. Shipboard organic geochemistry and safety monitoring, Leg. 64,
Gulf of California. In J. R. Curray, D. G. Moore, et al., eds., Initial Reports. DSDP 64:723-
728. Washington, D.C.: U.S. Govt. Printing Office.
Simoneit, B. R. T. 1983. Organic matter maturation and petroleum genesis: geothermal

versus hydrothermal. In The Role of Heat in the Development of Energy and Mineral
Resources in the Northern Basin and Range Province, pp. 215-241. Davis, Calif.: Geotherm.
Res. Council, Special Report no. 13.
Simoneit, B. R. T. 1984a. Hydrothermal effects on organic matter-high versus low

temperature components. In Advances in Organic Geochemistry 1983. Org. Geochem.
6:857-864.
Simoneit, B. R. T. 1984b. Effects of hydrothermal activity on sedimentary organic matter:

Guaymas Basin, Gulf of California-petroleum genesis and protokerogen degradation. In P. A.
Rona, K. Boström, L. Laubier, and K. L. Smith, Jr., eds., Hydrothermal Processes at Seafloor
Spreading Centers, pp. 453-474. New York: NATO-ARI Series, Plenum.
Simoneit, B. R. T. 1985. Hydrothermal petroleum: genesis, migration and deposition in

Guaymas Basin, Gulf of California. Can. J. Earth Sci. 22:1919-1929.
Simoneit, B. R. T. 1988. Petroleum generation in submarine hydrothermal systems: An

update, Can. Mineralogist 26:827-840.
Simoneit, B. R. T., ed. 1990. Organic matter alteration in hydrothermal systems-petroleum

generation, migration and biogeochemistry. Appl. Geochem. 5:1-248.
Simoneit, B. R. T. and O. E. Kawka. 1987. Hydrothermal petroleum from diatomites in the

Gulf of California. In J. Brooks and A. J. Fleet, eds., Marine Petroleum Source Rocks, pp.
217-228, Geol. Soc. Lond., spec. publ. no. 26.
Simoneit, B. R. T. and P. F. Lonsdale. 1982. Hydrothermal petroleum in mineralized mounds

at the seabed of Guaymas Basin. Nature 295:198-202.
Simoneit, B. R. T. and R. P. Philp. 1982. Organic geochemistry of lipids and kerogen and the
effects of basalt intrusions on unconsolidated oceanic sediments: Sites 477, 478 and 481,
Guaymas Basin, Gulf of California. In J. R. Curray, D. G. Moore, et al., eds., Initial Reports.
DSDP 64:883-904. Washington, D.C.: U.S. Govt. Printing Office.
Simoneit, B. R. T., O. E. Kawka, and M. Brault. 1988. Origin of gases and condensates in
the Guaymas Basin hydrothermal system, (Gulf of California). In M. Schoell, ed. Proc. Symp.
Origins of Methane in the Earth. Chem. Geol. 71:169-182.
Simoneit, B. R. T., S. Brenner, K. E. Peters, and I. R. Kaplan. 1981. Thermal alteration of

Cretaceous black shale by basaltic intrusions in the eastern Atlantic. II: Effects on bitumen
and kerogen. Geochim. Cosmochim. Acta 45:1581-1602.

Simoneit, B. R. T., J. O. Grimalt, J. M. Hayes, and H. Hartman. 1987. Low temperature

hydrothermal maturation of organic matter in sediments from the Atlantis II Deep, Red Sea.
Simoneit, B. R. T., P. F. Lonsdale, J. M. Edmond, and W. C. Shanks, III. 1990. Deep-water

hydrocarbon seeps in Guaymas Basin, Gulf of California. In B. R. T. Simoneit, ed., Organic
Matter Alteration in Hydrothermal Systems-Petroleum Generation, Migration and
Biogeochemistry. Appl. Geochem., 5:41-49.
Simoneit, B. R. T., R. P. Philp, P. D. Jenden, and E. M. Galimov. 1984. Organic

geochemistry of Deep Sea Drilling Project sediments from the Gulf of California
hydrothermal effects on unconsolidated diatom ooze. Org. Geochem. 7:173-205.
Simoneit, B. R. T., M. A. Mazurek, S. Brenner, P. T. Crisp and I. R. Kaplan. 1979. Organic

geochemistry of recent sediments from Guaymas Basin, Gulf of California. Deep Sea Res.
26A:879-891.
Spiess, F. N., K. C. Macdonald, T. Atwater, E. Ballard, A. Carranza, D. Cordoba, C. Cox, V.

M. DiazGarcia, J. Francheteau, J. Guerrero, J. Hawkins, R. Haymon, R. Hessler, T. Juteau,
M. Kastner, R. Larson, B. Luyendyk, J. D. Macdougall, S. Miller, W. Normark, J. Orcutt, and
C. Rangin. 1980. East Pacific Rise; hot springs and geophysical experiments. Science
207:1421-1433.
Suess, E., B. R. T. Simoneit, G. Wefer, M. J. Whiticar, M. Fisk, M. von Breymann, M. W.

Han, R. Wittstock, C. Laban, D. Kadko, and Z. Top. 1990. Hydrothermalism in the Bransfield
Strait, Antarctica. Geologische Rundschau, in preparation.
Thompson, G., S. E. Humphris, B. Schroeder, M. Sulanowska, and P. A. Rona. 1988.

Hydrothermal mineralization on the Mid-Atlantic Ridge. Can. Min. 26:697-711.
Tissot, B. P. and D. H. Welte. 1984. Petroleum Formation and Occurrence: A New Approach
to Oil and Gas Exploration. Berlin: Springer Verlag.
van de Meent, D., S. C. Brown, R. P. Philp, and B. R. T. Simoneit. 1980. Pyrolysis-high

resolution gas chromatography and pyrolysis gas chromatography-mass spectrometry of
kerogens and kerogen precursors. Geochim. Cosmochim. Acta 44:999-1013.
Vassoevich, N. B., A. M. Akramkhodzhaev, and A. A. Geodekyan. 1974. Principal zone of oil

formation. In B. Tissot and F. Bienner, eds., Advances in Organic Geochemistry 1973, pp.
309-314. Paris: Technip.
Von Damm, K. L., J. M. Edmond, C. I. Measures, and B. Grant. 1985a. Chemistry of

submarine hydrothermal solutions at Guaymas Basin, Gulf of California. Geochim.
Von Damm, K. L., J. M. Edmond, B. Grant, C. I. Measures, B. Walden, and R. F. Weiss.

1985b. Chemistry of submarine hydrothermal solutions at 21°N, East Pacific Rise. Geochim.

Whiticar, M. J., E. Suess, and H. Wehner. 1985. Thermogenic hydrocarbons in the surface
sediments of the Bransfield Strait, Antarctic Peninsula. Nature 314:87-90.
*CPI=carbon preference index: for hydrocarbons it is expressed as a summation of the odd

carbon number homologs over a defined range, divided by a summation of the even carbon
number homologs over the same range (Cooper and Bray 1963; Simoneit 1978).


17. Stable Carbon Isotope Changes During the Maturation of

Organic Matter
Natural and artificial thermal-induced changes in the isotopic composition of organic carbon were
determined for a suite of kerogen samples from the Cape Verde Rise, Deep Sea Drilling Project Leg
41, Site 368. This site is characterized by intrusions of hot diabase sills. The isotopic composition of
the organic carbon is thermally controlled. With increasing maturities, the 13C-OC values become
heavier owing to a loss of lighter carbon in the form of methane. This is supported by pyrolysis
simulation studies, which suggest a decrease in the amount of carbon-12-enriched methane that
can be generated with increasing kerogen maturities, as inferred from the H/C ratios.
Cretaceous black shales from the Cape Verde Rise, Deep Sea Drilling Project (DSDP) Leg 41, Site
368, were analyzed in this study to determine thermal effects on the 13C of organic carbon (OC).
Changes in the maturation of the organic carbon were determined by comparing the 13C of the
organic carbon and the carbon residue resulting from pyrolysis experiments at 600°C.
Several investigators have shown there is little or no isotopic alteration of the organic matter during
the diagenetic and catagenetic stages; therefore, the values of the total organic carbon reflect the
source of organic matter (Redding et al. 1980; Rigby, Batts, and Smith 1981; Yeh and Epstein
1981; Schoell 1984). There is some discrepancy about whether thermal processes alter the isotopic
composition when the kerogen reaches the metamorphic stage. Barker and Friedman (1969) found
an enrichment of carbon-13 in low-organic-carbon metamorphosed rocks relative to high-organic-
carbon nonmetamorphosed rocks. They attributed these differences either to loss of isotopically
light organic compounds in metamorphosed rocks or to the presence of two types of organic matter
with different isotopic signatures and degradation characteristics in nonmetamorphosed rocks.
Baker and Claypool (1970), in a study of incipient metamorphism on organic matter, found
differences between metamorphosed and nonmetamorphosed rocks only when the isotopic
composition of the sources differ. McKirdy and Powell (1974) and Hoefs and Frey (1976) found an
enrichment in 13C in metamorphosed vs. nonmetamorphosed samples up to 10 . Hoefs and Frey
(1976) suggested that the 13C enrichment found in the metamorphosed rocks was due to kinetic
isotope effects during methane formation, equilibrium exchange reactions between calcite and
graphite, or preferential aqueous oxidation of 12C in graphite to CO2. Laboratory pyrolysis
experiments by Chung and Sackett (1979) on four shale samples found only a small enrichment in
the residual fraction relative to the kerogen composition. Similar results were reported by Peters,
Rohrback, and Kaplan (1981) for sapropelic and humic sediments and by Simoneit, Peters, and
Kaplan (1981) for Cretaceous black shales altered by diabase intrusions (DSDP Site 41-368). Their
results show a maximum 13C enrichment of 2 in the residual fraction.
However, two studies have shown kerogen becomes more negative with increasing thermal stress.
Jenden, Simoneit, and Philip (1982) studied protokerogens in the Guaymas Basin, Gulf of California

(DSDP Sites 477, 478, and 481) that had been thermally altered by dolerite sills. Protokerogen, as
defined by Stuermer, Peters, and Kaplan (1978), is the acid resistant, insoluble fraction of immature
organic matter that contains an appreciable concentration of biopolymers. Protokerogens at Sites
477 and 481 were 1.5 more negative relative to the original carbon material. Peters, Rohrback,
and Kaplan (1981) found residues from sapropelic and humic sediments to be initially more
negative and to become systematically heavier with increasing pyrolysis temperature. They
proposed that this trend was a result of initial loss of isotopically heavy functional groups, such as
carboxyls, followed by loss of light methane, leaving the residual carbon enriched in carbon-13.
The work reported in this paper was undertaken to provide information on the effect of
metamorphism on the 13C of organic carbon in kerogens from the Cape Verde Rise shales. The
Cape Verde Rise is located at 17°30.4'N, 21°21.2'W with a water depth of 3366 m (see figure 17.1).
This site is characterized by organic-rich Cretaceous black shales that have been intruded by hot
diabase sills during the Miocene (see figure 17.2). A general description of the samples is
presented in the site reports of Lancelot et al. (1977). The organic geochemistry of these samples is
described in papers by Baker et al. (1977); Deroo et al. (1977); Doose et al. (1977); Dow (1977);
Erdman and Schorno (1977); Hunt and Whelan (1977); Kendrick, Hood, and Castano (1977); and
Kvenholden (1977).
Kerogen samples were isolated by Chevron Oil Research Company using the standard procedures
(see Peters et al. 1983). The samples analyzed include subsections from cores 58, 59, 60, and 62
from Site 368 shown in figure 17.2. They have a narrow range of maturities as indicated by the
atomic H/C ratios and vitrinite reflectances (table 17.1). Samples with atomic H/C ratios lower than
0.5 and vitrinite reflectances higher than 2% are classified as metamorphic (Tissot and Welte 1984).
These two parameters negatively correlate very well with each other (r= -0.94). The reflectance
values shown in table 17.1 are somewhat higher than those reported by Peters, Rohrback, and
Kaplan (1978) and by Simoneit, Peters, and Kaplan (1981) but fall within the range of those
reported by Dow (1977).
Thermal maturation of the kerogens was simulated by pyrolysis experiments in the laboratory.
Aliquots of samples were sealed under vacuum in Pyrex tubes while heating to about 150°C. A time
series analysis was conducted on sample 31, the most immature sample, to determine the pyrolysis
time necessary for "exhaustive" methane formation. This sample was pyrolyzed for a period from 24
to 288 hours at 600°C. The products were determined by gas chromatography (GC) and the
methane isotopic composition was measured. The GC results are shown in figure 17.3. The
amounts of ethane and ethylene/acetylene rapidly decrease after 48 hours of pyrolysis time. These
hydrocarbons disappear within 120 hours, apparently degrading to methane and carbon. Based on
these results, the pyrolysis time was set for 120 hours.
The organic carbon was combusted by a circulation technique similar to that described by Craig
(1953). Prior to combustion, samples were weighed, wrapped in aluminum foil to prevent any loss of
sample during the introduction of material into the combustion system, and placed in precombusted
ceramic boats. The residual organic carbon fraction (fraction remaining after gas generation) was
combusted in this same manner. The standard deviation for the results with this method is 0.2 .

The isotopic composition of carbon is expressed using the del notation:
Results are expressed as parts per thousand difference from a standard material. The 13C values
were determined in a triple collector Mat 250 Finigan Isotope Ratio Mass Spectrometer (IRMS). All
samples were measured relative to a working standard and reported relative to PDB ( 13C = -24.8;
vs. PDB).
The composition of hydrocarbons gases was determined in a Hewlett Packard 5710A gas
chromatograph (GC) equipped with a flame ionization detector. The column used was 3 mm i.d. x
1.5 m o.d. stainless steel columns packed with Porapak Q, mesh size 80-100 and programmed
from 80-150°C at 16°/minute.
Organic Carbon Content
The organic carbon (OC) content (reported in table 17.2) is high and variable, ranging from 22.3 to
74.9%. The large range in OC values is a result of depositional variations rather than a thermal
effect due to the intrusives (Dow 1977; Simoneit, Peters, and Kaplan 1981).
Shown in figure 17.4 and table 17.2 is the change in the kerogen 13COC values with depth. figure
17.4 includes 13COC values from Simoneit, Peters, and Kaplan (1981). The 13COC values range
from -27.5 to -20.9 . With increasing proximity to the sill, the 13C-OC values become more carbon-
13 enriched; therefore, the 13C-OC values appear to be thermally controlled rather than source
controlled. The heavier values with increased metagenesis may be due to preferential loss of
carbon-12 due to a kinetic isotope effect (Hoefs and Frey, 1976; Simoneit, Peters, and Kaplan
1981), or alternately, the thermally labile fraction of the kerogen may be composed of lighter carbon
(Simoneit, Peters, and Kaplan 1981).
As expected, no correlation was found between percent OC and 13C-OC since the organic carbon
content is a factor of the depositional environment, and the 13C-OC values appear to be the result
of the high thermal stress.
In general, the isotopic composition of the organic carbon appears to be thermally controlled;
therefore, no conclusions can be reached about its source based on its isotopic composition. Data
by Baker, Palmer, and Huang (1977); Dow (1977): and Erdman and Schorno (1977) all point to a
marine origin for the Cape Verde Rise kerogens.
Formation of Methane
The carbon mole ratio (CMR) is defined as the mole ratio of methane (total produced by

"exhaustive" pyrolysis) to kerogen carbon and is a measure of the amount of kerogen carbon that
can be used in methane generation. figure 17.5 shows a sharp decrease in the conversion of
kerogen carbon to methane, with increasing maturity, from 20.7% with a H/C ratio of 1.21, to 0.6%
when the H/C ratio is 0.23. Better correlations were found between the H/C ratios and CMR (r =
+0.85) than between vitrinite reflectance and CMR (r = -0.70). Peters et al. (1983) found that over
short periods of time and high temperatures, vitrinite reflectance responds faster than the
generation and alteration of bitumens. Rapid heating of organic matter results in changes that are
not directly comparable to those of slow burial metamorphism (Peters et al. 1983). H/C ratios are
dependent on the changes in the kerogens due to loss of hydrocarbons during maturation and may
be a better maturity indicator for these data sets than vitrinite reflectance. Although the high
correlation between these two parameters indicates a close relationship between them, H/C ratios
might be more sensitive to the rapid changes due to contact metamorphism.
A trend toward increased stability of the kerogen can be inferred from the decrease in the CMR.
The CMR is related to the carbon-carbon bonds present in the kerogen. Aromatic-type carbon is
more resistant to thermal degradation than aliphatic-type carbon is (Conkright, Sackett, and Peters
1986). In the metagenic stage, less than 1% of the kerogen carbon is reactive. Based on the
hydrogen content data presented in table 17.1, the hydrogen mole ratio (HMR = CH4 - H), or the
amount of hydrogen that can be converted to methane, can be computed. The HMR values range
from 0.678 for the least mature sample to 0.080 for the most mature one. The maturity of the
organic matter in most of these samples, as indicated by the low atomic H/C ratios, demonstrates
that the methane precursor carbon pool becomes depleted with proximity to the sill as less
hydrogen is available for methane formation owing to increased aromaticity of the kerogen
molecule. This process is described by the carbon and hydrogen mole ratio parameters.
Formation and Isotopic Composition of the Organic Residue
The organic residue resulting from the pyrolysis experiments was combusted and analyzed for all
the previous samples. The data are presented in figure 17.6. The isotopic composition of the
residue was generally heavier with proximity to the sill, ranging from -27.6 to -20.5 .
Generally, the isotopic composition of the residue was heavier than the parent carbon, a finding that
points to a loss of light carbon in the organic carbon fraction. This difference decreases with
increasing maturity. Since a decrease in the CMR is accompanied by a decrease in the difference
between 13C residue and 13C-OC, it is likely that the shift of 13C-OC toward heavier values with
increasing maturity is due to a loss of isotopically lighter methane. The organic residue data for the
Cape Verde Rise kerogens support field and laboratory data that indicate the residual fraction
becomes heavier with time (Chung and Sackett 1979; Peters, Rohrback and Kaplan 1981;
Simoneit, Peters, and Kaplan 1981). The CMR data support the premise that this shift is due to loss
of isotopically light carbon in the form of methane.
The Cape Verde Rise Cretaceous black shales are thermally altered, Type II kerogens, high in
organic carbon content. The organic carbon content is source controlled, but its isotopic
composition appears to be thermally controlled. With proximity to the sill, the 13C-OC values are
heavier owing to a loss of lighter carbon in the form of methane. This is shown from pyrolysis
simulations by a decrease in the carbon mole ratio, with increasing maturities, associated with a
decrease in the difference between the 13C-OC and the 13C-pyrolysis residue. The amount of
reactive carbon available decreases sharply in the highly altered samples. The amount of reactive

carbon in the metamorphosed samples is between 1 and 2%. The amount of reactive hydrogen is
around 8%.
The CMR is a useful parameter and indicative of the amount of "reactive" kerogen carbon available
and can be used as such for these samples. In addition, a hydrogen mole ratio can be calculated to
measure the amount of "reactive" hydrogen available for methane formation. These two
parameters, CMR and HMR, decrease with increasing maturities.
Acknowledgements
The authors would like to thank Dr. K. E. Peters for providing the samples. Funding for this research
was provided by Chevron Oil Research Field Company and the Patricia Harris Minority Fellowship.
References
Baker, D. R. and G. E. Claypool. 1970. Effects of incipient metamorphism on organic matter

in mudrock. AAPG Bull. 54:456-468.
Baker, E. W., S. E. Palmer, and W. I. Huang. 1977. Intermediate and late diagenetic
tetrapyrrole pigments, Leg 41: Cape Verde Rise and Basin. In Y. Lancelot, E. Siebold, et al.,
eds. Initial Reports: DSDP, 41:825-838. Washington, D.C.: U.S. Govt. Printing Office.
Baker, E. W., W. Y. Huang, J. G. Rankin, J. R. Castano, J. R. Guinn, and A. N. Fuex. 1977.

Electron paramagnetic resonance study of thermal alteration of kerogen in deep-sea
sediments by basaltic sill intrusion. In Y. Lancelot, E. Seibold, et al., eds., Initial Reports:
DSDP, 41:839-847. Washington, D.C.: U.S. Govt. Printing Office.
Barker, F. and I. Friedman. 1969. Carbon isotopes in Pelites of the Preecambrian

Uncompahgre Formation, Needle Mountains, Colorado. Geol. Soc. Am. Bull. 80:1403-1408.
Chung, H. M. and W. M. Sackett. 1979. Use of stable carbon isotope compositions of

pyrolytically derived methane as maturity indices for carbonaceous materials. Geochim.
Conkright, M. E., W. M. Sackett, and K. E. Peters. 1986. Application of the pyrolysis-carbon

isotope method for determining the maturity of kerogen in the Bakken shale. Advances in
Organic Geochemistry 1985. Org. Geochem. 10:1113-1117.
Craig, H. 1953. The geochemistry of the stable carbon isotopes. Geochim. Cosmochim. Acta
3:53-92.
Deroo, G., J. P. Herbin, J. Roucache, B. Tissot, P. Albrecht, and J. Scheffle. 1977. Organic
geochemistry of some Cretaceous black shales from Sites 367 and 368, Leg 41, Eastern
North Atlantic. In Y. Lancelot, E. Seibold, et al., eds., Initial Reports: DSDP, 41:865-873.
Doose, P. R., M. W. Sandstrom, R. Z. Jodele, and I. R. Kaplan. 1977. Interstitial gas analysis

of sediment samples from Site 368 and Hole 369A. In Y. Lancelot, E. Seibold, et al., eds.,
Initial Reports: DSDP, 41:861-864. Washington, D.C.: U.S. Govt. Printing Office.
Dow, W. G. 1977. Contact metamorphism of kerogen in sediments from leg 41: Cape Verde
Rise and Basin. In Y. Lancelot, E. Seibold, et al., eds., Initial Reports: DSDP, 41:815-816.
Erdman, J. G. and K. S. Schorno. 1977. Geochemistry of carbon: Deep Sea Drilling Project
Leg 41. In Y. Lancelot, E. Seibold, et al., eds., Initial Reports: DSDP, 41:849-854.
Hoefs, J. and M. Frey. 1976. The isotopic composition of carbonaceous matter in a

metamorphic profile from the Swiss Alps. Geochim. Cosmochim. Acta 40:945-951.
Hunt, J. M. and J. K. Whelan. 1977. Light hydrocarbons at site 367-370, Leg 41. In Y.
Lancelot, E. Seibold, et al., eds., Initial Reports: DSDP, 41:859-860. Washington, D.C.: U.S.
Jenden, P. D., B. R. T. Simoneit, and R. P. Philip. 1982. Hydrothermal effects of

protokerogens of unconsolidated sediments from Guaymas Basin, Gulf of California:
Elemental compositions, stable carbon isotope ratios and electron spin resonance spectra.
In Y. Lancelot, E. Seibold, et al., eds., Initial Reports: DSDP, 41:905-912. Washington, D.C.:
Kendrick, J. W., A. Hood, and J. R. Castano. 1977. Petroleum-generating potential of

sediments from Leg 41, Deep Sea Drilling Project. In Y. Lancelot, E. Seibold, et al., eds.,
Initial Reports: DSDP, 41:817-820. Washington, D.C.: U.S. Govt. Printing Office.
Kvenvolden, K. A. 1977. Organic geochemistry Leg 41, introduction and summary. In Y.

Lancelot, E. Seibold, et al. eds., Initial Reports: DSDP, 41:815-816. Washington, D.C.: U.S.
Lancelot, Y., E. Siebold, et al. 1977 Site 368: Cape Verde Rise. Initial Reports of Deep Sea
Drilling Project (DSDP), 41:233-326. Washington, D.C.: U.S. Govt. Printing Office.
McKirdy, D. M. and T. G. Powell. 1974. Metamorphic alteration of carbon isotopic

composition in ancient sedimentary organic matter. New evidence from Australia and South
Africa. Geology 2:591-595.
Peters, K. E., B. G. Rohrback, and I. R. Kaplan. 1978. Vitrinite reflectance-temperature

determination for intruded Cretaceous black shale in the eastern Atlantic. In D. F. Oltz, ed.,
Symposium on Geochemistry: low Temperature Metamorphism of Kerogen and Clay
Minerals. SEPM Pac. Sec., pp. 53-58.
Peters, K. E., B. G. Rohrback, and I. R. Kaplan. 1981. Geochemistry of artificially heated

humic and sapropelic sediments-I: Protokerogen. AAPG Bull. 65:688-705.

Peters, K. E., J. K. Whelan, J. M. Hunt, and M. E. Tarafa. 1983. Programmed pyrolysis of

organic matter from thermally altered Cretaceous black shales. AAPG Bull. 67(11):2137-
2146.
Redding, C. E., M. Schoell, J. C. Monin, and B. Durand. 1980. Hydrogen and carbon isotopic
composition of coals and kerogens. In A. G. Douglas and J. R. Maxwell, eds., Advances in
Organic Geochemistry 1979. New York: Pergamon Press.
Rigby, D., B. D. Batts, and J. W. Smith. 1981. The effect of maturation on the isotopic
composition of fossil fuels. Geochim. Cosmochim. Acta 3:29-36.
Schoell, M. 1984. Stable isotopes in petroleum research. In J. Brooks and D. Welte, eds.,
Advances in Petroleum Geochemistry, 1:215-246. New York: Pergamon Press.
Simoneit, B. R. T., K. E. Peters, and I. R. Kaplan. 1981. Thermal alteration of Cretaceous

black shale by diabase intrusions in the Eastern Atlantic. II. Effects on bitumen and kerogen.
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 spin
resonance spectra. Geochim. Cosmochim. Acta 42:989-997.
Tissot, B. and D. H. Welte. 1984. Petroleum Formation and Occurrence. New York: Springer
Verlag.
Yeh, H. W. and S. Epstein. 1981. Hydrogen and carbon isotopes of petroleum and related
organic matter. Geochim. Cosmochim. Acta 45:753-762.


18. Source and Biomarker Composition Characteristics of

Chinese Nonmarine Crude Oils
This paper presents an overview of current thinking about sources, ages, and thermal histories of
Chinese nonmarine crude oils.
Terrestrial source rocks, deposited under nonmarine conditions in continental environments, have
been divided into four major groups: Group A, sedimentary formations of big lake basins in the
interior of the Chinese plate; Group B, lacustrine clastic formations in fault-bounded basins; Group
C, salt lake evaporite-clastic formations in fault-bounded basins; and Group D, lagoonal-lacustrine
volcano-clastic formations of intermontane basins.
A number of representative crude oil samples were collected from the following four types of source
rocks: (1) lacustrine fresh-brackish water mudstones; (2) saline to hypersaline mudstones; (3)
lacustrine dolomitic mudstones; (4) lagoonal tufferous mudstones. The samples were analyzed by
gas chromatography (GC) and GC-mass spectrometry (GC-MS) to reveal the characteristic
distributions of biological markers in their aliphatic fractions. The distribution of n-alkanes,
isoprenoids, steranes, and hopanoids as well as other biological marker compounds are discussed
briefly. Some immature to marginally mature oil samples sourced from hypersaline and saline
mudstones are described in terms of biological marker parameters of their aliphatic fractions.
Immature or marginally mature crude oils occurring at shallow depths in the Jianghan basin contain
high amounts of sulfur and organic sulfur compounds; especially abundant are thiophene and
thiolane homologues with a remarkable even-carbon number predominance.
Most oil fields found to date in China are in nonmarine sedimentary basins ranging in age from
Carboniferous to Paleogene. Major source rock types deposited in different environments in China
have been classified and described in terms of their geological and geochemical setting (Powell
1986; Fu et al. 1986; Fu and Sheng 1989). In addition, exploitable immature oils discovered in
China appear to be restricted to certain subenvironments, such as lacustrine saline and hypersaline
environments.
In recent times, the information from biological marker distributions has been successfully used for
the differentiation and assessment of depositional environments (Brassell and Eglinton 1986),
especially for the characterization and distribution of general ancient marine and nonmarine
petroleum source rocks (Moldowan, Seifert, and Gallegos 1985; Brassell, Eglinton, and Howell
1987), or for definition of detailed subenvironments, such as lacustrine freshwater and hypersaline,
or marine carbonate and deltaic sediments (Palacas, Anders, and King 1984; Brassell, Eglinton,
and Fu 1986; Fu et al. 1986; Philp and Gilbert 1986; Mello et al. 1988).
This paper presents an overview on the current thinking about sources, ages, and thermal histories
of Chinese nonmarine crude oils. Molecular organic geochemical data are employed for a number

of representative crude oil samples in an attempt to characterize the major types of Chinese
nonmarine oils, including immature oils, and their relationship to the depositional environments of
their associated source rocks.
Sources, Ages, and Thermal Histories of Chinese Nonmarine Crude Oils
Type of Source Organic Matter
In view of the differing characteristics of oil- and gas-bearing basins in China, four major groups of
terrestrial source rocks were recognized (Fu, Sheng, and Liu et al., 1989):
Group A-sedimentary formations of large lake basins in the interior of the Chinese plate.
A typical example is the Songliao Basin, where the most productive source rocks are of Cretaceous
age (figures 18.1and 18.2).
Group B-lacustrine clastic formations in fault-bounded basins.
Almost all important source rocks of Cenozoic oil- and gas-bearing basins in Eastern China belong
to this group or, sometimes, to Group C. Key examples of this group are the Shengli and Liaohe oil
fields, as well as Jizhong and Subei basins.
Group C-salt lake evaporite-clastic formation in fault-bounded basins.
The typical examples of this type are the Jianghan and Biyang basins. Both are deposited in a
saline to hypersaline lake environment.
Group D-lagoonal-lacustrine, volcano-clastic formations in intermontane basins.
This is known only in Western China (e.g., the Carboniferous-Permian strata of the Zhungeer Basin
in figure 18.1). This group of source rocks is composed of basic tuff, tuffaceous mudstone, and
mudstone, of which the mineral composition is quite complicated, including montmorillonite, illite,
chlorite, and mixed-layer clays, as well as small amounts of analcime, albite, plagioclase, quartz,
dolomite, ankerite, calcite, and siderite (Ming Yushun, unpublished data). As reported in our
previous papers (Fu and Sheng 1986; Fu, Sheng, and Liu 1989), Group A source rocks are mainly
composed of Type I and Type IIa kerogen. The organic matter distribution is controlled by the
sedimentary facies in the large lake basin. Type I kerogen was generally developed within the deep
lake facies. Group B source rocks are much more complicated. Most contain Type II kerogen, with
some Type III and Type I kerogens. The main reasons for this complicated situation are the
changes in the environments related to the fault movements and the distance of the sag from the
adjacent area of uplift. For instance, the source rock of the sag that is farther away from the uplift
area is better developed, being less influenced by clastic and higher plant input. Kerogens of the
Group C source rocks are normally Type II, and those of Group D source rocks are possibly Type II
and Type I.
Ages of Source Rocks

The distribution of the oil fields is evidently controlled by the development and distribution of the
source rock series. It has been shown that the distribution of Chinese continental oil-generating
formations of different ages was controlled by tectonic and sedimentary cycles. The following five oil-
generating periods were determined to be most important (figure 18.2): the late Triassic, early-
middle Jurassic, early Cretaceous, early Tertiary, and the late Tertiary. There are three main types
of source beds classified with respect to specific geological ages according to their basic
characteristics: (1) the freshwater-brackish water lacustrine facies (T3 and K1, figure 18.2) related
to the semihumid climate; (2) the freshwater lacustrine facies (late T3 and J1+2) related to the
humid climate; and (3) various saline lacustrine facies (T) related to the semiarid climate (Huang,
Shang, and Li 1984).
Thermal Histories of Source Organic Matter
The main factors that affect thermal evolution of organic matter are temperature and time. Owing to
high sedimentation rates, high geothermal gradients, and minimal breaks in sediment deposition
during formation of Eastern China fault-bounded basins, both the oil-generating threshold (OGT)
time (from the end of deposition of source rocks to the beginning of generation of oil) and the oil-
generating period (the whole period of the oil window) are very short. The sedimentation rate of
Paleogene fault-bounded lake basins, including the salt lake basin in Eastern China, is quite high,
up to 0.2-0.3 mm/year. The average geothermal gradient of six major oil- and gas-bearing basins is
about 3.7°C/100 m, with the highest being 4.78°C/100 m (Lower Liaohe basin), and the lowest 2.7°
C/100 m (Beijing-Tianjin region). The average geothermal gradient is 4.0°C/100 m in the Songliao
basin and 4.2°C/100 m in the Biyang basin (figure 18.3 after Huang, Shang, and Li 1984). The
geothermal gradient varies greatly in salt lake basins; i.e., in the Jianghan basin, the average
gradient is 3.1°C/100 m, but it varies from 2.3°C/100 m (main region with numerous salt beds) up to
3.7°C/100 m (area with no salt beds). The depth of OGT in fault-bounded lake basins, including the
salt lake basin, is about 1700-2700 m, and the OGT temperature 90°-95°C. Both OGT time and oil-
generating period are as short as 40 to 50 million years and 30 to 45 million years respectively. In
China, the basin with the shortest OGT time discovered to date is the Mangya depression in
Chaidamu (Q+N), which had sedimentation rates up to 0.5 mm/year. OGT values are 3300 m, 126°
C (figure 18.3) and 6.5 million years. There is a slight difference between Mesozoic and Cenozoic
fault-subsided basins, such as the Songliao basin with OGT values of only 1300 m, 65°C (figure
18.3) and twenty-five million years (Huang, Shang, and Li 1984; Fu and Sheng 1986).
Sedimentary formations of some Mesozoic intermountainous basins in Western China have lower
paleotemperature, such as Permian-Jurassic sedimentary formations with geothermal gradients of
only 2.2°C/100 m (Fu and Sheng 1986).
Biological Marker Characteristics of Chinese Nonmarine Oils
Sampling and Experimental
The crude oils analyzed in this study are listed in table 18.1 and shown in figure 18.1. The source
type assessment for the oils is based on certain geological assumptions supported by geochemical
correlations or other published data. The oils selected for this study were generally the least mature
and least biodegraded available, in order to minimize the effects of maturation and biodegradation.

However, some immature and marginally mature oils were used for understanding the
characteristics of Chinese nonmarine immature oils.
All oil samples were separated by standard chromatographic methods to yield aliphatic and
aromatic fractions, which were analyzed by gas chromatography (GC) and computerized
chromatography-mass spectrometry (C-GC-MS), according to previously reported methods
(Brassell, Eglinton, and Fu 1986; Fu et al. 1986). Biological marker compounds were identified by
reference to standard compounds, literature spectra, or spectral interpretation.
According to the nature of the sedimentary basin and the type of source rock (Fu et al. 1986; Fu,
Sheng, and Liu 1989; Fu and Sheng 1989), Chinese nonmarine crude oils may be divided into the
following major types:
Crude Oils Sourced from Lacustrine Fresh-BrackishWater Mudstones
This type of oil is sourced from lacustrine fresh-brackish water mudstones and is very important in
China and also widely distributed in different continental sedimentary basins ranging in age from
Jurassic to Tertiary. The typical samples of this group are those found in the Daqing oil field with
dominant subenvironments in lacustrine brackish and fresh water.
The GC results reveal some common features, such as n-alkanes with slight odd over even carbon
number predominance, a high-molecular-weight pre-dominance, and pristane-to-phytane ratios of 1
to 2.5 (figure 18.4).
The specific biological markers found by GC-MS are mainly steranes and terpanes. Liaohe oils
have a very low concentration of steranes. Regular steranes show a predominance of C29 steranes
over the C27 and C28. Steranes assigned by mass chromatography in Daqing oils include regular
steranes (C21, C22, and C27-29 compounds), as well as a small amount of 4-methylsteranes and
diasteranes.
The terpanes, as determined by the M/Z 191 in Daqing oils, show a series of tricyclic (C16-C29)
and tetracyclic hydrocarbons in very low concentrations. The abundance of hopanes is much higher
than that of steranes, with the C30 hopane dominant in the typical C27 to C35 range (occasionally
up to C37 with C28 absent).
Crude Oils Sourced from Saline to Hypersaline Mudstones
This type of oil is considered to be sourced from saline to hypersaline mudstones and is produced
from Tertiary fault-bounded basins in Eastern China, of which the typical sample is the Jianghan oil
field (Fu and Sheng 1986).
These oils, in most cases, have very high concentrations of phytane and a relatively high
abundance of gammacerane and occasionally show even carbon predominance on n-alkanes (OEP
value < 1, figure .18.5). The C27 regular steranes are dominant over C29 compounds. There are

also C28-C30 4-methylsteranes in lower concentration possibly derived from dinoflagellates, as well
as pregnane and homopregnane. The M/Z 191 traces, in some cases, reveal an unusual
distribution pattern of homohopanes and high abundance of gammacerane (figure 18.6). These
remarkable features resemble those of their source rocks (Brassell et al. 1988; Fu, Sheng, and Liu
1989).
Crude Oils Sourced from Lacustrine Dolomitic Mudstones
This type of oil has been discovered and pooled very recently from a restricted area in the Biyang
oil field. From geological and geochemical evidence it is thought that these oils are sourced from a
particular source type-lacustrine dolomitic mudstone (Fu and Sheng 1986).
Like the previous type, this oil also shows a very high proportion of phytane and carotanes and
slight odd carbon number predominance of normal alkanes in its GC traces of aliphatic. However,
the terpane distribution in M/Z 191 mass chromatograms is quite different from that of the previous
type. In particular, the chromatograms show a series of tricyclic and tetracyclic terpanes,
Gammacerane and very low concentrations of homohopanes (except C31 homohopane)
(figure .18.7). This unusual distribution pattern of homohopanes may be caused by specific bacterial
input and/or higher plant input (Philp and Gilbert 1986) or by a diagenetic environmental effect.
Crude Oils Sourced from Lagoonal Tufferous Mudstones
This type of oil has been shown to be generated from a particular source rock type, e.g.,
Carboniferous-Permian lagoonal tufferous mudstone, which is thought to be the main source of
crude oils produced in the Klamayi oil field, Xingjiang province.
The n-alkanes of this group reveal a distribution pattern of possible algal or bacterial origin: low-
molecular-weight predominance with slight odd over even carbon dominance.
Further evidence of bacterial or algal input to the oils comes from their typical distribution feature of
isoprenoid homologues. The GC traces of aliphatic fractions show abundant acyclic isoprenoid
components, including higher molecular weight compounds up to C40 (lycopane). Another
remarkable feature is the occurrence of a very high concentration of carotanes, including - and -
carotanes (figure 18.8; Fu and Sheng 1989).
Immature and Marginal Mature Crude Oils
Immature to marginally mature oils have been discovered in China and are likely to be restricted to
those sourced from lacustrine saline to hypersaline mudstones and dolomitic mudstones (Shi et al.
1982; Fu et al. 1986; Zhou 1987). They are pooled mainly in reservoirs of Tertiary fault-bounded
basins in Eastern China.
Typical examples of biological marker parameters of the aliphatic fractions of the oils from the
Jianghan oil field and the Shengli oil field, including immature or marginally mature oils, are given in
table .18.2. Immature and marginally mature oils always contain a high amount of sulfur. An
increase in burial depth or maturity causes the content of sulfur to decrease markedly (figure 18.9).

Meanwhile, changes in their biomarker compositions occur not only in their aliphatic fractions but
also in their aromatic fractions. For instance, the least mature oils occurring at shallow depths have
abundant thiophene and thiolane homologues with even carbon number predominance (value of
0.66; figure 18.10; Fu and Sheng 1989), whereas the aromatic fractions of less mature oils (e.g.,
marginal mature oils) contain smaller amounts of sulfur and sulfur compounds.
Several preliminary conclusions can be drawn concerning the sources, ages, thermal histories, and
biological marker compositional characteristics of Chinese nonmarine crude oils.
1. In terms of the type and nature of the oil- and gas-bearing basins, four major groups of terrestrial
source rocks have been identified: Group A, sedimentary formations of large lake basins in the
interior of the Chinese plate; Group B, lacustrine clastic formations in fault-bounded basins; Group
C, salt lake, evaporite-clastic formations in fault-bounded basins; and Group D, lagoonal-lacustrine,
volcano-clastic formations in intermontane basins.
2. The distribution of Chinese nonmarine oil-generating formations of different ages was controlled
by tectonic cycles and sedimentary cycles, resulting in five oil-generating periods: the late Triassic,
the early-middle Jurassic, the early Cretaceous, the early Tertiary, and the late Tertiary.
3. The type and abundance of organic matter is related to the source rock types and to the
sedimentary facies in lacustrine basins, particularly the distance of a depression from the area of
uplift and the occurrence of the source rocks within a particular sag. The main factors controlling the
thermal evolution of the organic matter are the palaeogeothermal gradient and nonconformities.
4. The main biomarker composition of the aliphatic fraction of the four major crude oil types has
been discussed briefly. Each has its specific biomarker characteristics as follows: (a) crude oils
sourced from lacustrine fresh-brackish water mudstones are poor in gammacerane but rich in 4-
methyl-steranes; (b) crude oils sourced from saline to hypersaline mudstones have an extremely
high amount of phytane and series of sulfur compounds, with high ratios of phytane/pristane and
gammacerane/C30 hopane, and sometimes high even carbon predominance of n-alkanes; (c)
crude oils sourced from lacustrine dolomitic mudstones contain series of tricyclic and tetracyclic
terpanes, gammacerane, and very low concentrations of homohopanes; and (d) crude oils sourced
from lagoonal tufferous mudstones contain abundant isoprenoid alkanes and carotanes ( -carotane
and -carotane).
5. Immature to marginally mature oils discovered in China so far are likely to be restricted to those
sourced from lacustrine saline to hypersaline mudstones and dolomitic mudstones. Those oils
contain high amount of sulfur and organic sulfur compounds, especially abundant thiophene and
thiolane homologues with a remarkable even carbon number predominance.
Acknowledgments
We wish to thank Professor G. Eglinton, Dr. S. C. Brassell, Mrs. A. P. Gowar, Mr. Xu Jiayou, Mr.
Peng Ping'an, and Dr. E. W. Della for help and advice during the analytical work and in the
preparation of this paper. Many thanks are due to Drs. J. K. Whelan and J. W. Farrington for
reviewing and clarifying the manuscript. We are also grateful to the Chinese Academy of Sciences
for research funds (project on the theory of generation of Chinese oil and gas, 1988-1990).

References
Brassell, S. C. and G. Eglinton. 1986. Molecular geochemical indicators in sediments. In M.

Sohn, ed., Organic Marine Geochemistry, pp. 10-32. Washington D.C.: ACS.
Brassell, S. C., G. Eglinton, and Jiamo Fu. 1986. Biological marker compounds as indicators
of the depositional history of Maoming oil shale. In D. Leythaeuser and J. Rullkötter eds.,
Advances in Organic Geochemistry 1985. Org. Geochem. 10:927-942, Oxford: Pergamon
Press.
Brassell, S. C., G. Eglinton, and V. J. Howell. 1987. Palaeoenvironmental assessment of

marine organic-rich sediments using molecular organic geochemistry. In J. Brooks and A. J.
Fleet, eds., Marine Petroleum Source Rocks, pp. 79-98. London: Blackwell.
Brassell, S. C., Guoying Sheng, Jiamo Fu, and G. Eglinton. 1988. Biological markers in
lacustrine Chinese oil shales. In K. Kelts, A. Fleet, and M. Talbot, eds., Lacustrine Petroleum
Source Rocks pp. 299-308. London: Blackwell.
Fu, Jiamo and Gouying Sheng. 1986. Organic geochemical characteristics of major-type
sedimentary formations of oil- and gas-bearing basins in China. In Guangzhi Tu, ed.,
Advances in Science of China-Earth Sciences 1: 251-286. Science Press: Beijing, China:
New York: Wiley.
Fu, Jiamo and Gouying Sheng. 1989. Biological marker composition of typical source rocks
and related crude oils of terrestrial origin in the People's Republic of China: a review. Appl.
Geochem. 4:13-32.
Fu, Jiamo, Guoying Sheng, and Dehan Liu. 1989. Organic geochemical characteristics of
major types of terrestrial petroleum source rocks in China. In K. Kelts, A. Fleet, and M.
Talbot, eds., Lacustrine Petroleum Source Rocks, pp. 279-289. London: Blackwell.
Fu, Jiamo, Gouyang Sheng, Pingan Peng, S. C. Brassell, G. Eglinton, and Jigang Jiang.
1986. Peculiarities of salt lake sediments as potential source rocks in China. In D.
Leythaeuser and J. Rullkötter, eds., Advances in Organic Geochemistry, pp. 119-127,
Huang, Difang, Huiyun Shang, and Jinchao Li, 1984. Advances in theoretical research on
continental oil generation in China. Beijing Petroleum Symposium, Beijing, September.
Mello, M. R., P. C. Gaglianone, S. C. Brassell, and J. R. Maxwell. 1988. Geochemical and

biological marker assessment of depositional environments using Brazilian offshore oils.
Mar. Petrol. Geol. 5:205-223.
Moldowan, J. M., W. K. Seifert, and E. J. Gallegos. 1985. Relationship between petroleum

composition and depositional environment of petroleum source rocks. AAPG Bull. 69:1255-
1268.

Palacas, J. G., D. E. Anders, and J. D. King. 1984. South Florida Basin-A prime example of
carbonate source rock potential of carbonate rocks. In J. G. Palacas, ed., AAPG Studies in
Geology, 18:71-96.
Philp, R. P. and T. D. Gilbert. 1986. Biomarker distribution in Australian oils predominantly

derived from terrigenous source material. In D. Leythaeuser and J. Rullkötter, eds.,
Advances in Organic Geochemistry, pp. 73-84. Oxford: Pergamon Press.
Powell, T. G. 1986. Petroleum geochemistry and depositional setting of lacustrine source

rock. Mar. Pet. Geol. 3:200-219.
Shi, Jiyang, A. S. Mackenzie, R. Alexander, G. Eglinton, A. P. Gowar, G. A. Wolf, and J. R.

Maxwell. 1982. A biological marker investigation of petroleums and shales from the Shengli
oilfield, People's Republic of China. Chem. Geol. 35:1-31.
Zhou, Guangjia. 1987. Immature crude oils from non-marine fault-bounded basin in China. In
Chinese Assoc. of Petroleum, eds., Continental Petroleum Generation and Organic
Geochemistry, pp. 27-37. Beijing: Petroleum Press.


19. Volatile Organic Compounds Associated with Oil Seepage

Along the Northern Continental Slope of the Gulf of Mexico
Volatile organic compounds (VOC) were quantified and identified at three petroleum seepage sites
on the northern Gulf of Mexico continental slope. Concentrations of individual VOC ranged from
below detection limit (0.01 ng/g) to greater than 500 ng/g with total concentrations as high as 7600
ng/g. Gas chromatographic patterns from these sites display a complex mixture of hydrocarbons, up
to 400 individual VOC, and an unresolved complex mixture. This is in contrast to nonseep sites
where relatively simple VOC mixtures are found (fewer than forty compounds) with individual
concentrations less than 1.0 ng/g and a total concentration of less than 100 ng/g. VOC
concentrations in sediments are dependent upon the amount and type of organic matter delivered
to the sediment and the subsequent degradation processes occurring in the sedimentary column.
The origin of volatile organic compounds (VOC) in shallow marine sediments has been an enigma.
Though there is generally neither sufficient temperature nor pressure for diagenetic production of
these compounds, VOC have been detected in trace amounts (sub ng/g dry sediment) in marine
sediments from a number of Deep Sea Drilling Project (DSDP) sites (McIver 1974; Hunt 1975; Hunt
and Whelan 1978; Whelan and Sato 1980; Brooks et al. 1982), gravity cores (Whelan, Hunt, and
Berman 1980; Whelan and Hunt 1983; McDonald 1988), and fluvial and estuarine sediments
(Whelan, Blanchette, and Hunt 1983; McDonald 1988). Volatile organic compounds are often
associated with biologically produced methane and have been postulated to be produced in situ by
either low-temperature chemical diagenesis or biological processes (Whelan, Hunt, and Berman
1980; Whelan and Hunt 1983). Several authors (Whelan and Hunt 1983; Whelan, Simoneit, and
Tarafa 1988) have documented VOC patterns in sediments heated by basaltic intrusions associated
with active ocean spreading centers. Studies of VOC in marine sediments deep below the sea floor
(>1000 m) have concluded that VOC at depth are thermally cracked (> 50°C) from organic matter
(Huc and Hunt 1980; Hunt, Huc, and Whelan 1980a; Hunt, Miller, and Whelan 1980; Thompson
1983; Jasper, Whelan, and Hunt 1984; Schafer and Leythauser 1986). However, the association
between VOC and the upward migration of petroleum has received little attention.
This study was undertaken to aid in defining and understanding the processes controlling the
distribution of VOC in marine sediments. Sediments from areas known to be contaminated with oil
seepage were analyzed to determine the composition of VOC associated with hydrocarbon
seepage. Results are compared and contrasted with VOC data from areas affected by few or no
thermogenic hydrocarbons.
Geologic Setting
The continental slope of the Gulf of Mexico has been found to be a site of active liquid as well as
gaseous petroleum seepage (Anderson et al. 1983; Brooks et al. 1986; Kennicutt, Brooks, and
Denoux 1988). The northern continental slope offshore Louisiana is a bathymetrically complex

region underlain by extensive salt diapirs and ridges that are interconnected at shallow depths
(Ewing and Antoine 1966; Antoine and Bryant 1969; Martin and Bouma 1978). The geologic
features of the area include intraslope basins with thick sediment accumulations, graben structures
over shallow diapirs, and growth and normal faults (Bouma 1983). The Mississippi Canyon is a
major submarine channel near the Mississippi Fan. Sediments were collected at three locations on
the Gulf of Mexico northern continental slope (figure 19.1).
Methods
Sea-bottom sediment samples were collected with a 1000 kg, 6 m piston corer. The sediments were
placed in precombusted glass jars, poisoned with sodium azide, and stored inverted at -4°C until
analysis. The headspace was replaced with purified nitrogen to retard organic matter oxidation after
collection.
Quantitative determination of VOC content was accomplished with a closed-loop stripping system
(CLS) (figure 19.2). Wet sediment (10 to 100 grams) was added to the CLS unit slurried in organic
free water. Sufficient water was added to bring the total volume of the sediment/water slurry to
approximately 800 mL. The headspace was purged with a purified helium stream to remove
laboratory air, the internal standards (C5, C8, C10, C12 chlorinated alkanes) were added, and the
CLS unit was sealed and heated to 45°C with an external heater. The VOC were stripped from the
sediment for one hundred eighty minutes at a purge rate of 500 mL/min. The charcoal trap was
maintained at 27°C. The purged compounds were removed from the charcoal trap by elution with
carbon disulfide. The VOC were quantified by a Hewlett-Packard 5790A gas chromatograph
equipped with a fused silica capillary column operated in splitless mode. The gas chromatograph
was linked to an HP-1000 main frame computer via LAS 3357 software package. Individual peaks
were quantified based on the flame ionization response of authentic standards (Alltech
Corporation). Unknown peaks were calculated by an average n-alkane response factor. Tentative
identification of sample components was based on retention time and mass fragment patterns. Gas
chromatography/mass spectrometry was conducted with a Hewlett-Packard 5970 mass selective
detector (MSD) coupled to a Hewlett Packard 5890 GC (RTE/RPN software package). Sample
components were identified after background subtraction. Spectral interpretations were aided by the
periodical literature, computer searches on the HP-1000 computer, authentic standards, and the
NIH-EPA library.
Results
Green Canyon 234
Station 1 was located near Green Canyon block 234, an area of active petroleum seepage and
thermogenic hydrate occurrence (Brooks et al. 1986; Cox 1986). The sediment throughout the core
was visibly stained with a yellowish-brown oil and had a strong H2S odor. TOC values ranged from
3.0 to 3.8%. Total VOC concentrations ranged from 135 to 270 ng/g (wet weight) (table 19.1; figure
19.3). The gas chromatograms (GC/FID) contained few resolved peaks (figure 19.4). The
concentration of the identifiable compounds was generally less than 2.0 ng/g. Individual n-alkane
concentrations were generally less than 1.0 ng/g with the total n-alkanes about 8 ng/g. Individual
one-ring aromatic compounds were present at concentrations generally less than 0.4 ng/g and total
one-ring aromatics ranged from 5 to 10 ng/g. The higher molecular weight alkylated benzenes were

more abundant in the upper two sections, whereas benzene, toluene, and ethylbenzenes were
more abundant in the lower section. Naphthalenes were randomly distributed with depth both in
composition and concentration and were dominated by naphthalene and methyl naphthalenes. The
cyclic and branched alkane compositions were variable with depth but relatively uniform in
concentration. Several organosulfur compounds were detected at this station, including
benzothiophenes and dibenzothiophenes. The organosulfur compounds were detected by flame
photometric methods and confirmed by GC/MS. Similar compounds were identified at the other two
locations in this study (Green Canyon block 96/97 and Mississippi Canyon block 988). Several
compounds tentatively identified included alkenes (normal, branched, and cyclic hexenes and
heptenes), furans (methyl and ethyl furans), and aldehydes (C2, C3, and C4 aldehydes). These
compounds were detected at concentrations of less than 0.01 ng/g.
Green Canyon 96/97
A core from station 2 was sampled eight times over the 5.6 m length. This core was also located in
an area of petroleum seepage (Anderson et al. 1983; Cox 1986). The sediment was generally
brownish/grey with varying amounts of carbonate rubble present. Oil staining was present
throughout the core, and a strong H2S odor was apparent. TOC values ranged from 1.1 to 3.4%.
Total VOC concentrations ranged from 701 to 7634 ng/g with both parameters maximizing at 2.0 m.
(table 19.2; figure 19.5). There was a significant correlation between total VOC and TOC (R2 =
0.68). Typical gas chromatograms are displayed in figure 19.6. The n-alkane concentrations were
variable with depth, though a similar compound distribution was present at all depths. Higher
molecular weight n-alkanes (> n-C10) were more abundant than lower molecular weight n-alkanes
(n-C10). One-ring aromatic compounds decreased in total concentration with depth and varied in
composition. The two-ring aromatic compounds were the most abundant compounds in all but two
sections, where one-ring aromatic compounds and n-alkanes dominated. The depth distribution of
the two-ring aromatic compounds maximized at 1.3 m depth. This maximum was primarily due to
increased amounts of dimethyl and trimethyl naphthalenes. The cyclic and branched alkanes were
also highly variable in concentration and compound composition with depth. The functionalized
compounds detected at this site were similar to those tentatively identified at Station 1 and were
again present at concentrations less than 0.01 ng/g.
Mississippi Canyon 988
Station 3 was sampled eight times from 0.05 to 3.6 m subbottom (water depth 1200 m). The
sediment from this site was grey-brown and had no H2S odor or visible oil staining. TOC decreased
with depth from 1.3 to 0.4% at 0.8 m and then remained relatively constant. Total VOC ranged from
330 to 2043 ng/g (table 19.3; figure 19.7). and significantly correlated with TOC (R2 = 0.73). More
than four hundred resolvable VOC were present as well as an unresolved complex mixture. Typical
gas chromatograms are displayed in figure 19.8. The n-alkanes covaried with total VOC and were
the most concentrated group of compounds at all but two depths, where one-ring aromatic
compounds were more concentrated. One-ring aromatic compounds had variable compound
distributions but displayed a concentration profile similar to that of the n-alkanes. The two-ring
aromatic compounds were generally dominated by naphthalene and the dimethylnaphthalenes and
had a depth profile similar to that of the n-alkanes. The cyclic and branched compounds reached a
concentration maximum between 2 and 3 m. Numerous organosulfur compounds were detected at
this site, including thiophenes, benzothiophenes, and dibenzothiophenes. Functionalized organic
compounds similar to those detected at Station 1 were also present at Station 3.

Discussion
High levels of VOC were detected at three petroleum and gas seep sites along the northern
continental margin of the Gulf of Mexico. At all sites, a complex mixture of VOC (>five hundred
compounds) and a large unresolved complex mixture (UCM) were present. The gas
chromatographically resolved compounds included homologous series of n-alkanes, cyclic and
branched alkanes, and alkylated aromatics (one and two ring), suggesting that the VOC were
directly derived from petroleum. Several functionalized organic compounds were also detected in
these sediments. These compounds have previously been attributed to bacteria (Whelan, Hunt, and
Berman 1980; Whelan and Hunt 1983.)
Green Canyon 234 and Green Canyon 96/97
The distribution of VOC at these two Green Canyon sites has been heavily influenced by the input
of petroleum and its subsequent degradation. The major source of the VOC at these sites is the
upward migration of petroleum, as shown by the oil staining throughout both cores. Anderson et al.
(1983) and Brooks et al. (1986) reported high concentrations of petroleum in sediments from this
region. At both stations the VOC have been severely altered, Green Canyon 234 exhibiting a near
total absence of all GC-resolved compounds. A GC pattern of a sediment extract or a whole crude
oil, which displays a large baseline rise (UCM) and a total absence of n-alkanes, is typical of
biodegraded petroleum (Jordan and Payne 1980; Romeu 1986; Kennicutt et al. 1987). The
complete degradation of resolved hydrocarbons is apparent in the GC traces from the Green
Canyon 234 core. The concentrations of the resolved VOC ranged from only 135 to 240 ng/g, even
though the TOC was greater than 3% and there was a large UCM in the traces. By assuming a
response factor for the compounds in the unresolved hump to be equal to the average for the n-
alkanes hexane through hexadecane, the concentration of the UCM would be approximately 10,000
ng/g.
In contrast to the core from Green Canyon 234, the core from Green Canyon 96/97 was dominated
by naphthalene and alkylated naphthalenes. A homologous series of naphthalenes is consistent
with a petroleum hydrocarbon source. The presence of naphthalene compounds at this site differs
from the results obtained by Anderson et al. (1983). They found a lack of individual aromatic
hydrocarbons, which they attributed to microbial degradation and water washing of the surficial
sediments. They did, however, show the presence of large amounts of the aromatic fraction relative
to the saturate and polar fraction in the extractable hydrocarbons on a weight percent basis. This
difference in GC patterns is most likely explained by the heterogeneous distribution of hydrocarbons
and degradative organisms. The core in this present study apparently sampled a more recent
seepage that was less microbially altered than in the previous report.
Mississippi Canyon 988
Sediment from this station contained more than 400 resolved VOC. Upward migration of petroleum
accounts for the majority of the VOC in this core. Brooks et al. (1986) and Cox (1986) have
suggested that petroleum fluids, gas and oil, have migrated into this region. Cox (1986) determined
that the isotopic composition of the methane associated with gas hydrates at this site had a value of
-48.2 ppt (relative to PDB) and suggested that the methane originated from a thermal source. The
hexane-extractable organic material (C15+) displayed a complete pattern of hydrocarbons,

including a UCM. Kennicutt et al. (1987) suggest that this type of hydrocarbon pattern in surficial
marine sediments is due to petroleum hydrocarbons. Extractable petroleum hydrocarbons are found
throughout the core and exhibit no significant variation in concentration with depth. Deposition of
organic matter from the water column and active microbial populations may account for the
presence of the functionalized compounds and some of the low-molecular-weight organosulfur
compounds such as the thiophenes and alkylated sulfides. Functionalized compounds may also be
produced as byproducts during the microbial oxidation of hydrocarbons. Organosulfur compounds
such as the alkylated sulfides may also be a product of other processes, including postdepositional
alteration of organic and inorganic sulfur-containing compounds. Benzothiophenes and
dibenzothiophenes in surficial sediments have been directly correlated to reservoired petroleum
(Lacerda, Kennicutt, and Brooks 1987)
One anomalous feature of this site is the low concentration of VOC in the 0.05 m section. The
concentration of total VOC increased from 400 ng/g at 0.05 m to 1500 ng/g at 1.0 m. The reduced
amount of VOC is probably the result of several processes, including: (1) loss to the overlying water
column by dissolution or diffusion, (2) changes in the sediment texture such as increased sand-
sized particles, or (3) microbial alteration of the compounds. VOC loss to the overlying water
column by dissolution would preferentially remove light aromatics and branched and cyclic alkanes.
A second process causing lower VOC concentrations could be the increased amount of carbonate
rubble, approximately 18% by weight, versus 11% for the remainder of the core. Two observations
suggest that microbial degradation of both the VOC and extractable organic matter is occurring: (1)
enhancement of the isoprenoids in the shallow section over the deeper section, suggesting
preferential bacterial degradation of the n-alkanes, and (2) the presence of functionalized organic
compounds. Functionalized organic compounds containing oxygen or sulfur are thought to be
microbially sourced, suggesting an active bacterial population (Whelan, Hunt, and Berman 1980).
The distribution of VOC in sediments is controlled by the magnitude of the input of petroleum as
balanced against its subsequent microbial degradation.
Functionalized Organic Compounds
Functionalized organic compounds detected at these sites included aldehydes (C2, C3, and C4
aldehydes), furans (methyl and ethyl furans), and alkenes (normal, branched, and cyclic hexenes
and heptenes). These functionalized compounds are probably indicative of an active microbial
population in these sediments as well as of secondary detrital material settling from the water
column. Whelan and Hunt (1983) and Whelan, Blanchette, and Hunt (1983) describe similar
functionalized compounds in the Peru upwelling region and in anoxic sediments from the
Pettaquamscutt River (Rhode Island, USA). They attributed these compounds to destructive and
nondestructive transformation via various processes occurring in the sediment. These processes
include microbial degradation, chemical destruction and alteration, and physical losses. The most
widespread postdepositional process is microbial degradation. Hydrocarbons may be altered by
microorganisms to form ketones, aldehydes, organic acids, alcohols, phenols, and low-molecular-
weight organosulfur compounds (Hunt 1979). The rate at which VOC are utilized depends upon the
type of hydrocarbon, environmental conditions, and the species of microorganism present.
Functionalized organic compounds can result from the degradation of petroleum as well as from
conversion of biological debris that has settled from the overlying water column or from benthic
biota.
Comparison with Nonpetroleum Seep Site

The magnitude of the influence of petroleum on the VOC concentrations and distributions is readily
apparent when the data are compared with VOC distributions in sediments that have not been
impacted by oil (McDonald 1988). Previous studies have analyzed sediments from an area along
the northern Gulf of Mexico continental slope that are believed to be relatively unimpacted by
petroleum seepage (table 19.4). Total organic carbon from this region averaged 0.75% and total
VOC were less than 100 ng/g. Gas chromatograms from this region are simple, usually containing
less than fifty resolved compounds (figure 19.9). Simple mixtures (< fifty compounds) with low
individual concentrations suggest a predominantly biological source for the VOC. Individual
compound concentrations in pristine areas are generally less than 1 ng/g. Pentadecane, an
indicator of marine phytoplankton (Sauer 1978), is one of the compounds most often detected at
these sites.
Concentrations of VOC at three petroleum seep sites along the northern continental margin of the
Gulf of Mexico ranged from below detection limit (<0.01 ng/g) to greater than 500 ng/g, with total
VOC concentrations as high as 7600 ng/g (table 19.5). The major source of VOC to these sites is
migration from more deeply buried sources. GC patterns from these stations show a complex
mixture of VOC with up to 400 compounds resolved by gas chromatography. Thermogenic
hydrocarbons are recognized by the presence of homologous series of n-alkanes, cyclic and
branched alkanes, and alkylated aromatics (one and two ring). In addition to the migrated VOC,
debris from planktonic and benthic fauna and flora may account for the functionalized VOC detected
at these sites. These biological inputs were indicated by the presence of aldehydes, ketones, and
furans. Once the VOC are deposited in surficial sediments they undergo degradation and
transformation. Destruction or transformation occurs by two pathways, microbial alteration and low-
temperature diagenesis. Microbial degradation was a dominant removal process on the northern
Gulf of Mexico continental slope.
References
Anderson, R. K., R. S. Scalan, P. L. Parker, and E. W. Behrens, 1983. Seep oil and gas in
Gulf of Mexico Sediments. Science 222:619-622.
Antoine, J. W. and W. R. Bryant. 1969. Distributions of salt and salt structures in the Gulf of
Mexico. AAPG Bull. 53:2543-2550.
Bouma, A. H. 1983. Intraslope basins in the northwest Gulf of Mexico: A key to ancient
submarine canyons and fans. AAPG spec. publ. no. 32, pp. 567-581.
Brooks, J. M., L. A. Bernard, D. A. Wiesenburg, M. C. II Kennicutt, and K. A. Kvenvolden.

1982. Molecular and isotopic compositions of hydrocarbons at Site 533, Leg 76. In R. E.
Sheridan and F. M. Gradstein et al., eds., Initial Reports DSDP, 76:377-389. Washington D.
Brooks, J. M., H. B. Cox, W. R. Bryant, M. C. II Kennicutt, R. G. Mann, and T. J. McDonald.

1986. Association of gas hydrates and oil seepage in the Gulf of Mexico. Org. Geochem.
10:221-234.
Cox, H. B. 1986. Gas hydrates in the Gulf of Mexico. M. S. thesis, Texas A & M Univ.

Ewing, M. and J. W. Antoine. 1966. New seismic data concerning sediments and diapiric
structures in Sigsbee Deep and upper continental slope, Gulf of Mexico. AAPG Bull. 50:479-
504.
Huc, A. Y. and J. M. Hunt. 1980. Generation and migration of hydrocarbons in offshore

South Texas Gulf coast sediments. Geochim. Cosmochim. Acta 44:1081-1089.
Hunt, J. M. 1975. Origin of gasoline range alkanes in the deep sea. Nature 254:411-413.
Hunt, J. M. 1979. Petroleum geochemistry and geology. San Francisco: W. H. Freeman.
Hunt, J. M. and J. K. Whelan. 1978. Dissolved gases in Black Sea sediments. In D. A. Ross
and Y. P. Neprochnov, et al., eds., Initial Reports: DSDP, 42:661-665. Washington D.C.: U.
Hunt, J. M., A. Y. Huc, and J. K. Whelan. 1980. Generation of light hydrocarbons in

sedimentary rocks. Nature 288:688-690.
Hunt, J. M., R. J. Miller, and J. K. Whelan. 1980. Genesis of petroleum hydrocarbons in

marine sediments. Science 209:403-404.
Jasper, J. P., J. K. Whelan, and J. M. Hunt. 1984. Migration of C1 to C8 volatile organic

compounds in sediments from the DSDP Leg 75, Hole 530A, Walvis Ridge, In W. W.
Hayland, J. C. Sibuet, et al., eds., Initial Reports: DSDP, 75:1001-1008. Washington D.C.: U.
Jordan, R. E. and J. R. Payne. 1980. Fate and Weathering of Petroleum Spills. Ann Arbor,
Mich.: Ann Arbor Science Publishers.
Kennicutt II, M. C., J. M. Brooks, and G. J. Denoux. 1988. Leakage of deep, reservoired
petroleum to the near surface on the Gulf of Mexico continental slope. Mar. Chem. 24:39-59.
Kennicutt II, M. C., G. J. Denoux, J. M. Brooks, and W. A. Sandberg. 1987. Hydrocarbons in

Mississippi fan and intraslope basin sediments. Geochim. Cosmochim. Acta 51:1457-1466.
Lacerda, C. P., M. C. Kennicutt II, and J. M. Brooks. 1987. The distribution of

dibenzothiophenes in Gulf of Mexico sediments. Appl. Geochem. 2:297-304.
Martin, R. G. and A. H. Bouma. 1978. Physiography of the Gulf of Mexico. In A. H. Bouma,

G. T. Moore, and J. M. Coleman, eds., Framework, Facies, and Oil Trapping Characteristics
of the Upper Continental Margin. pp. 3-19. Tulsa: AAPG Studies in Geology, no. 7.
McDonald, T. J. 1988. Volatile organic compounds in Gulf of Mexico Sediments. Ph.D.

dissertation, Texas A & M Univ.
McIver, R. D. 1974. Analysis of ten Leg 22 cores for organic carbon and gasoline range

hydrocarbons. In C. von der Borch and C. Christopher et al., eds., Initial Reports: DSDP
10:67. Washington, D.C.: U.S. Govt. Printing Office.
Romeu, A. A. 1986. Biodegradation of Kuwait crude oil in the presence and absence of the
Dispersant Corexit 9527. Ph.D. dissertation, Texan A&M Univ., College Station.
Sauer, T. C. 1978. Volatile liquid hydrocarbons in the marine environment. Ph.D.

dissertation, Texas A&M Univ., College Station.
Schafer, R. G. and D. Leythauser. 1986. Low molecular weight hydrocarbons in sediments of

DSDP Leg 89, Sites 568, East Marine Basin, and Site 586, Ontong-Java Plateau. In R.
Moberly and S. O. Schlanger et al., eds., Initial Reports: DSDP, 89:577. Washington, D.C.: U.
Thompson, K. F. M. 1983. Classification and thermal history of petroleum based on light

Whelan, J. K. 1983. C1-C8 organic compounds from the Peru upwelling region. Org.
Geochem. 5:13-28.
Whelan, J. K. and J. M. Hunt. 1983. C1-C8 hydrocarbons in Leg 64 sediments, Gulf of

California. In J. Curray and D. G. Moore et al., eds., Initial Reports: DSDP 64:1348-1354.
Whelan, J. K. and C. D. Sato. 1980. C1-C5 hydrocarbons from core gas pockets, DSDP Leg
56 and 57: Japan Trench transects. In M. Langseth and H. Okada et al., eds., Initial Reports:
DSDP 56:1335-1347 Washington, D.C.: U.S. Govt. Printing Office.
Whelan, J. K., M. A. Blanchette, and J. M. Hunt. 1983. Volatile C1-C7 organic compounds in
an anoxic sediment core from the Pettaquamscutt River (Rhode Island, U.S.A.). Org.
Geochem. 5:29-33.
Whelan, J. K., J. M. Hunt, and J. Berman. 1980. Volatile C1-C7 organic compounds in
surface sediments from Walvis Bay. Geochim. Cosmochim. Acta 44:1767-1785.
Whelan, J. K., B. R. T. Simoneit, and M. E. Tarafa. 1988. C1-C8 hydrocarbons in sediments

from Guaymas Basin, Gulf of California-Comparison to Peru Margin, Japan Trench and
California Borderlands. Org. Geochem. 12:171-194.

Organic Matter: Productivity, Accumulation, and Preservation in Recent and Ancient Sediments,
by Jean K. Whelan and John W. Farrington
20. Maturity and Facies-Controlled Composition of

the Organic Matter of Selected Oil Shales
A set of oil shales from different localities in Europe and the Middle East that represent lacustrine as
well as marine depositional conditions were examined in order to study the transformation of organic
matter (OM) during diagenesis and catagenesis. The OM originates in nearly all cases from the
incorporation of the remnants of floral and faunal organisms into the carbonatic sediments; however, its
degree of transformation is different for both facies types. Whereas in marine oil shales nonstructured
OM ("prebitumen") prevails, the main organic constituent of lacustrine shales was found to be
laminated algal material ("lamalginite"). Another difference is the higher total sulfur content of the
marine samples in contrast to the lacustrine ones, which reflects the difference in sulfate content and
sulfate reduction in both environments. The preservation or degradation of the OM might also be
influenced by the latter process.
The influence of the thermal history during subsidence on the transformation of the OM is reflected by
the occurrence of unsaturated and , -configurated triterpenoids in the least mature shales. With
increasing maturity, the , -compounds (i.e., hopanes) become predominant. Organic-rich sediments
from two Eocene lakes in Germany differ in the composition of the OM with respect to the visual
characterization and in the degree of maturity. However, in both sample sets a group of unusual
terpenoid compounds was detected. Based on the tentative assignment of a degraded arborane/
filicane structure, a land plant source is postulated.
Besides the molecular composition of the extractable OM, the yield of shale oil is very much maturity
dependent. The relation between total organic carbon (TOC), vitrinite reflectance, and yield of Fischer
pyrolysis is given for the Lower Saxony Posidonia Shale.
Oil shales have been encountered on all continents since the Proterozoic. The name is misleading for
in most cases the organic-rich sediments contain neither extractable nor free oil in major quantities nor
are they pure shales. The oil is released only on destructive distillation and the mineral matrix mainly
consists of a mixture of clay and carbonate minerals with cherts, phosphorites, and heavy minerals as
minor compounds.
The shaly appearance results from the foliated arrangement of the organic matter, a result of a
rhythmic sedimentation without bioturbation. An elevated primary production of organic matter and/or
stratified water column (lack of circulation) is necessary to preserve the organic matter under stagnant
conditions. The oxic/anoxic boundary may exist below the sediment-water-interface or above; the
position of the boundary influences the petrographic composition of the kerogen.
The depositional controls on the character of source rocks and oil shales and on the corresponding
crude oil composition have been given by Powell (1987). Hutton et al. (1980) and Cook, Hutton, and
Sherwood (1981) have made important contributions to the classification of oil shales, relating the

occurrence of certain chemical compounds to the organic input and its preservation. The relationship
between metal enrichment and the organic content of sediments, within the geological frame, has been
discussed by Eugster (1985).
The data given in this paper were collected over the years and partly published elsewhere (Hollerbach,
Hufnagel, and Wehner 1977; Wehner and Hufnagel 1986; Köster, Wehner, and Hufnagel 1989) or
presented in internal reports. Therefore, in table 20.1 and figure 20.1 for some localities, only a
selection of new samples and data is presented.
Analytical Methods
Total organic carbon (TOC) and total sulfur contents were determined in a LECO CS-344 analyzer. The
extractable organic matter (OM), obtained through Soxhlet extraction with dichloromethane, was
separated by column chromatography on silica gel and aluminium oxide into the fractions of saturated
hydrocarbons, aromatic hydrocarbons, and heterocompounds by use of n-heptane, dichloromethane,
and dichloromethane/methanol (2:1). The elemental sulfur was removed from the extract, prior to the
liquid chromatograph (LC) procedure, by boiling it with activated copper granulate. The saturated
hydrocarbons were analyzed on a Hewlett Packard 5985 A gas chromatograph-mass spectrometer
(GC-MS) and an INCOS 50 MS (Finnigan) coupled to a Hewlett Packard 5890 GC to obtain the alkane
distribution as well as the biomarker fingerprints. Normally, the selected ion monitoring mode was
applied; for special compound identification full data acquisition was necessary. A CDS pyroprobe
coupled to a Packard 439 GC or to the HP GC-MS-computer system was used to do pyrolysis-GC and
pyrolysis-GCMS. All gas chromatographs were equipped with 30 m fused silica capillary columns
coated with SE-54. A more detailed description of the methods applied is given by Teschner and
Wehner (1985). The optical investigations of the polished rock samples followed the procedure
described by Hufnagel et al. (1981).
Samples and Locations
A short characterization of the samples is given in table 20.2. The shales from the lacustrine
environment were deposited in small crater lakes of volcanic origin (Pula, Gerce/Hungary; Eckfelder
Maar/western border of Germany) or of meteoritic origin (Nördlinger Ries/southern Germany), or in a
lake in a tectonically induced depression (Messel). They are all of Tertiary age and the degree of
maturity is low.
The marine environment is represented by a set of oil shales from Schröfeln and Seefeld, northern
Calcareous Alps (near Innsbruck). The oil shales were deposited in restricted lagoons and special
basins on the Upper Triassic "Hauptdolomit" platform.
Lower Saxony (northern Germany). The Posidonia shale of Toarcian age (Lower Jurassic) is a
widespread epicontinental shelf sediment that occurs in a different maturity stage in northwestern
Germany, owing to a Cretaceous intrusion (Massif of Bramsche).
Jurf ed Darawish/Jordan. The oil shales of Upper Maastrichtian age are intercalated between
phosphorites on the bottom and organic-lean limestones on top. As in the "Hauptdolomit," special
basins within a carbonate platform were filled with anoxic sediments.
Results

Microscopical and Chemical Investigations
The Messel oil shale has been very well investigated and has become of growing interest to organic
geochemists since the work of Welte (1965). For further documentation of literature, see Heil et al.
(1987). A comparison of the Eocene sediments of Messel and those from the Eckfelder Maar, which up
to now have not yet reached the paleontological importance of the Messel deposit, seems to be
worthwhile.
The organic petrography of samples from both localities is summarized in table 20.3. In this respect
there is a close relationship between both deposits, the differences occurring in the abundance of
certain macerals. The ratio of liptinite/huminite for the Messel shales is about 1.5, for the samples from
the Eckfelder Maar, 0.7. However, since only four surface samples from the Eckfelder Maar could be
obtained so far, these results should be taken as preliminary.
The chemical composition of the soluble organic matter in samples from both localities is similar, but
some remarkable differences are quite obvious. The saturated hydrocarbons, whose content is
somewhat higher in the Eckfelder Maar samples, are dominated by long chain n-alkanes showing a
strong odd-even predominance. In the short-chain range, n-C17 is the dominating compound, and the
pristane/phytane ratio is greater than 1 (figure 20.1). Within the series of triterpenoids, the pentacyclic
hopenes are the major compounds in the Messel shales, followed by the , -hopanes. In the Eckfelder
Maar samples, however, the , -hopanes are more abundant than the hopenes (figure 20.2). The
steroids, represented mainly by sterenes, are of minor importance in both sample sets.
Another feature common to both deposits is the occurrence of a pair of triterpenoid compounds with a
base peak at m/z 161 in the mass spectra. These compounds are identified as triterpenoid
stereoisomers on the basis of spectral interpretation. The structure has been tentatively related to an
arborane or filicane skeleton from which the A-ring has been totally removed and replaced by an
unsaturation between the previous bridging atoms (figure 20.3).
The main organic constituent of the Miocene/Pliocene crater fillings of Nördlingen and Gerce/Pula is
lamellar alginite (= partly denominated as lamalginite), the remnant of epipelic mats that covered the
surface of the sediment and consisted of algae and/or other microbial organisms (Hollerbach,
Hufnagel, and Wehner 1977). The algal-microbial mats indicate a suboxic or anoxic environment with
the oxic/anoxic boundary oscillating above and below the living, photosynthesizing complex
assemblage of organisms. Some huminites, dispersed liptinites (pollen), and prebitumen (bituminite)
are also present.
As in the case of the Messel shale, unsaturated terpenoids and steroids are the major compounds of
the biomarker families; the , -hopanes are less abundant. Land-plant-derived material is indicated by
a marked odd-even predominance of the long-chain n-alkanes in the case of the Hungarian samples,
whereas in the Nördlingen oil shales this feature is much less pronounced (figure 20.1).
The odd-even predominance of the long-chain n-alkanes is normally attributed to a higher land plant
input. However, there are samples, e.g., from Jordan, where no higher plant remains-spores, pollens,
etc.-can be identified under the microscope. In these cases an origin of these molecules from algae
should be considered, as is discussed by Hundrieser (1985) and Powell (1987).

The marine environment is represented by Mesozoic oil shales from Austria, Germany, and Jordan.
Under the microscope a major organic constituent, if not the dominating one, is "prebitumen" (Hufnagel
1984). The term prebitumen was chosen to demonstrate that this important organic material is a
precursor or source material from which in nature, at suitable conditions, petroleum is generated. In the
retort, shale oil is produced when the shales are exposed to temperatures between 300° and 500°C.
The solubility in organic solvents is relatively low to medium; in most cases less than 10% of the OM is
soluble. This nonstructured flaky to granular organic material, also called amorphous lipid kerogenous
material (Dow 1977), bituminous groundmass (Hufnagel et al. 1981) or bituminite (Teichmüller 1971),
originated during the sedimentation or in the very early phases of the diagenesis by mainly microbial
activity from suitable initial floral and faunal material of lipidic composition. In the marine oil shales,
prebitumen is at least partly attributed to "marine snow," flocculent amorphous macroscopic
components of the nearshore epipelagic sea water that are inhabited by various kinds of
microorganisms (Silver, Shanks, and Trent 1978, Prezelin and Allredge 1983). In oxygen-containing
water, while sinking downward, the marine snow is normally decomposed by the microbial activity. The
mechanism of the degradation process changes when the particles reach the anoxic bottom water.
Finally, together with other mineral or biological components, they are deposited at the sea bed.
Common to all the marine oil shales investigated is the high content of total sulfur. Despite the fact that
part of the sulfur may occur in the form of heavy metal sulfides (e.g., pyrite) a positive correlation
between total sulfur and TOC is observed, indicating the uptake of sulfur into the kerogen structure.
The samples from the marine and lacustrine environment can be differentiated according to their sulfur
content, following the suggestion by Berner and Raiswell (1984) and Gautier et al. (1984), who offered
a tentative guide to discriminate between freshwater and marine argillaceous sediments. The amount
of sulfur, which corresponds to a TOC value of zero (intercept on the S-axis in figure .20.4) is
inorganically bound.
Because seawater is normally enriched in sulfate as compared with lake water, the bacterial reduction
of the sulfates leads to a greater supply of elemental and sulfide sulfur in a marine environment at low
Eh values (Orr 1977). Therefore sulfur-rich kerogens form preferentially under marine conditions in the
absence of iron oxides (carbonate deposition). Hypersalinity, on the other hand, may inhibit both sulfate
reduction and methanogenesis (Barnard and Cooper 1981).
An example of an immature high-sulfur kerogen is represented by the oil shales from Jordan, which
were investigated by Sinninghe Damsté et al. (1989): thiolanes, alkylthiolanes, alkyl thiophenes, and
alkylthianes were identified. Sulfur is postulated to be incorporated into precursors with double bonds at
an early stage of diagenesis.
A series of alkylthiophenes and benzothiophenes (figure 20.5) could be identified in the Austrian oil
shales from Seefeld and Schröfeln by pyrolysis-GC-MS (Köster, Wehner, and Hufnagel 1989).
However, these compounds do not occur in stratigraphically equivalent but high mature samples
(percent Rr > 1.5).
There might be two reasons for this observation. First, thermal instability of part of the sulfur
compounds causes their destruction during maturation. Second, during early maturation of the
kerogen, sulfur compounds are enriched in the mobile fraction, leading to the generation of sulfur-rich
crude oil. For example, shale oils from the Jordanian oil shales have sulfur contents up to 7%
(Hufnagel 1984).
The fingerprints of the biomarkers of the lacustrine samples and the marine Jordanian oil shales, which
are of equivalent maturity, are different because of the occurrence of some sulfur-containing

compounds in the saturated hydrocarbon fraction of the Jordanian samples. The greater part of the
Jordanian biomarkers consists, however, of unsaturated compounds, accompanied by some , -
hopanes (Wehner and Hufnagel 1986), as was found in the Messel samples.
The dominance of the , -hopane series in the shales from Austria (Köster, Wehner, and Hufnagel
1989) and northwest Germany is attributed to the higher maturity level rather than to source
differences.
Influence of Diagenesis and Catagenesis
As described above, an increasing maturity can be followed on a molecular level by looking at the
hopanoid structures. The least mature sediments from Hungary, southern Germany, Jordan, and
Messel, which are all in the range of 0.3% vitrinite reflectance, mainly contain unsaturated compounds.
Nevertheless, even at this immature stage, within the carbonatic (60% CaCO3) sulfur-rich shales from
Jordan, a generation and short-distance migration of mobile compounds had occurred, as can be
deduced from the bitumen-filled foraminifera in the lower section of the Chalk Marl Unit/Jurf ed
Darawish (Wehner and Hufnagel 1986).
The shales from the Ries crater, low in carbonate content ( S6%) but relatively high in sulfur (1.2%),
also generated oil at this immature stage. During the drilling operations, an oil film was observed on the
drilling mud (Hollerbach, Hufnagel, and Wehner 1977).
Migrated bitumen was not observed in the immature samples low in sulfur from Hungary, Messel, and
Eckfelder Maar. Therefore, sulfur incorporated into the kerogen in the form of aliphatic sulfide and
disulfide groups, as was discussed by Orr (1986) for the Monterey kerogens, might be the cause for the
generation of traces of migratable oil in the Jordanian and Ries samples at an early maturation stage.
Besides the molecular composition of the soluble OM, the yield of shale oil is very much dependent on
the maturity of the shales. For a given basin this relation can be used to estimate the recoverable shale
oil quantity. In the Lower Saxony Basin the maturity of sections at the base of the Jurassic was
measured by Koch and Arnemann (1975). The TOC content, the Fischer pyrolysis yield, and the
vitrinite reflectance of the Posidonia shale samples were determined by the Geological Survey of Lower
Saxony (Ludwig 1977). Therefore, based on figure 20.6, the quality of additional samples, whose TOC
content can be analyzed much faster than their pyrolysis yield, can easily be classified. The regression
lines represent different localities in the Lower Saxony Basin, starting in the eastern part (near
Brunswick) with immature shales. Approaching a Cretaceous intrusion in the western part of the basin,
the maturity of the shales increases, causing a drastic reduction of the shale oil quantity. Interestingly,
within the maturation zone representing the main stage of oil generation (0.6-0.9% Rr), the shale oil
yield is reduced by about 50% in comparison with samples from the immature zone of the Lower
Saxony Basin. In figure 20.6 the data points for samples less mature than the Posidonia shales indicate
equivalent or even higher shale oil quantities at a given TOC content. So, during the whole maturity
range considered (0.3 < Rr < 1.8), the quantity of recoverable shale oil steadily decreases.
This trend is not paralleled by the amount of the extractable OM. table .20.4 presents mean values on
the bulk composition of the Posidonia shale's OM at different maturity levels. In the immature stage,
where the shale oil production is at its maximum, the yield of extractable OM is rather low. It reaches its
peak in the "main stage of oil generation," where the shale oil quantity is already reduced. The TOC
content is diminished by more than 30% within the maturity range considered. These net results mirror
a complex process of generation, expulsion, and cracking, whose quantification was further developed

by Leythaeuser et al. (1988), who also investigated a set of Posidonia shales.
Within the sample series considered here, the lacustrine sediments often contain remains of lamellar
algae as main organic constituents, indicating at least some oxygen supply at the sediment surface
where the algal-bacterial mats grow.
In the marine samples, prebitumen is the major compound in most cases. It is a nonstructured
degradation product of phytoplankton and zooplankton. The high sulfur content of marine kerogens
points to an intensive sulfate reduction and hence to a strong bacterial reworking of the organic matter,
which is used as an energy source. It is this sulfur incorporation into the kerogen that may lead to a
generation of migratable compounds at a very early stage of maturity. At the lower maturity level, the
bulk composition of the organic matter of sediments from the lacustrine and marine environment is
quite similar, consisting mostly of heterocompounds and asphaltenes and, to a lesser extent, of
hydrocarbons. The structural configuration of the hopanoids is controlled more strongly by maturity
than by facies changes, which does not exclude the occurrence of some rare compounds such as
degraded arboranes in some locations. In this case, a higher land plant contribution into the lakes of
Messel and Eckfeld is indicated.
The maturation of the organic matter causes a steady decrease in the yield of shale oil, recoverable on
retorting, whereas the amount of extractable OM passes a maximum. The reactive part of the OM
(soluble + pyrolysable OM) is partly transferred to the inert and partly to the soluble material, leading to
an increase of the extractable matter at a maturity level that represents the peak oil generation. At this
stage, a loss of mobile compounds also occurs through migration. Therefore, that part of the soluble
OM remaining in the sediments is the result of a dynamic process of generation and migration
(Leythaeuser, Schaefer, and Radke 1988). At higher maturity levels the reactive part of the OM is
further reduced through expulsion and cracking.
References
Barnard, P. C. and B. C. Cooper. 1981. In L. V. Illing, G. D. Hobson, eds., Petroleum Geology

of the Continental Shelf of North-West Europe, pp. 169-175. London: Institute of Petroleum.
Bayer. Geolog. Landesamt. 1977. Ergebnisse der Ries-Forschungsbohrung 1973: Struktur des
Kraters und Entwicklung des Kratersees. Geologica Bavarica, vol. 75. München.
Berner, R. A. and R. Raiswell. 1984. C/S method for distinguishing fresh water from marine
sedimentary rocks. Geology 12:365-368.
Bray, E. E. and E. D. Evans. 1961. Distributions of n-paraffins as a clue to recognition of

source beds. Geochim. Cosmochim. Acta 22(1):2-15.
Bruckner-Wein, A., J. Vetö, E. Dudich. 1983. A geochemical study of the sedimentation of the
Oligocene anoxic Tard Clay/Hungary. Annual Report of Hungarian Geological Survey, pp. 271-
301.
Cook, A. C., A. C. Hutton, and N. R. Sherwood. 1981. Classification of oil shales. Bull. Centres
Rech. Explor.-Prod. Elf Aquitaine 5(2):353-381.

Dow, W. C. 1977. Kerogen studies and geological interpretations. J. Geochem. Explor. 79:99.
Amsterdam.
Eugster, H. P. 1985. Oil shales, evaporites and ore deposits. Geochim. Cosmochim. Acta 49
(3):619-635.
von Gaertner, H. R., H. Kroepelin, H.-H. Schmitz, H. Fesser, K. Mädler, H. Jacob, and K.
Hoffmann. 1968. Zur Kenntnis des nordwestdeutschen Posidonienschiefers. Beih. geol. Jb. 58.
Hannover, Germany.
Gautier, D. L., J. L. Clayton, J. S. Leventhal, and N. J. Reddin. 1984. Origin and source rock
potential of the Sharon Springs member of the Pierre Shale, Colorado and Kansas. In J.
Woodward, F. F. Meissner, and J. L. Clayton, eds., Hydrocarbon Source Rocks of the Greater
Rocky Mountain Region, pp. 369-385. Golden, Colo.: Rocky Mtn. Assoc. Geol.
Heil, R., v. W. Königswald, H. G. Lippmann, D. Graner, C. Heunisch. 1987. Fossilien der

Messel-Formation. Hessisches Landesmuseum in Darmstadt, Germany.
Hollerbach, A., H. Hufnagel, H. Wehner. 1977. Organisch-geochemische und -petrologische

Untersuchungen an den See-Sedimenten aus der Forschungsbohrung Nördlingen 1973.
Geologica Bavarica 75:139-153.
Hufnagel, H. 1984. Die Ölschiefer Jordaniens. Geol. Jb. A 75:295-311. Hannover, Germany.
Hufnagel, H., K. Kuckelkorn, H. Wehner, G. Hildebrand. 1981. Interpretation des Bohrprofils

Vorderriss 1 aufgrund organo-geochemischer und geophysikalischer Untersuchungen.
Geologica Bavarica 81:123-143.
Hundrieser, A.-J. 1985. Organisch-chemische Zusammensetzung von Bohrkernen und

Fossilien aus der Ölschiefergrube Messel. Ph.D. thesis, Univ. Hannover, Germany.
Hutton, A. C., A. J. Kantsler, A. C. Cook, and D. M. Mckirdy. 1980. Organic matter in oil shales.
APEA J. 20(1):44-67.
Jambor, A. and G. Solti. 1975. Geological conditions of the Upper Pannonian oil shale deposits
recovered in the Balaton Highland and at Kemenshat. Acta Miner. Petr. Szeged 22:9-28.
Koch, J. and H. Arnemann, 1975. Die Inkohlung in Gesteinen des Rhät und Lias im südlichen
Nordwestdeutschland. Geol. Jb. A 29:45-55. Hannover, Germany.
Köster, J. 1988. Organische Geochemie und Organo-Petrologie kerogenreicher und

bituminöser Einschaltungen im Hauptdolomit (Trias, Nor) der nördlichen Kalkalpen. Thesis,
Tech. Univ. of Clausthal, Germany.
Köster, J., H. Wehner, H. Hufnagel. 1989. Organic geochemistry and organic petrology of
organic rich sediments within the "Hauptdolomit"-Formation (Triassic, Norian) of the Northern
Calcareous Alps. In L. Mattavelli and L. Novelli eds., Advances in Organic Geochemistry 1987,
part 1. Org. Geochem. 13:1-3, 377-386. Oxford: Pergamon Press.

Leythaeuser, D., R. G. Schaefer, M. Radke. 1988. Geochemical effects of primary migration of

petroleum in Kimmeridge source rocks from Braefield area, North Sea. I: Gross composition of
C15+-soluble organic matter and molecular composition of C15+-saturated hydrocarbons.
Leythaeuser, D., R. Littke, M. Radke, and R. G. Schaefer. 1988. Geochemical effects of

petroleum migration and expulsion from Toarcian source rocks in the Hils syncline area, NW-
Germany. In L. Mattavelli and L. Novelli, eds. Advances in Organic Geochemistry 1987. Org.
Geochem. 13:1-3, 489-502.
Ludwig, G. 1977. Untersuchung einiger Ölschiefer-Vorkommen in Niedersachsen. Final report,

Niedersä hsisches Landesamt für Bodenforschung, no. 078649. Hannover, Germany.
Negendank, J. F. W., J. Linden and G. Irion. 1982. Ein eozänes Maar bei Eckfeld nordöstlich
Manderscheid (SW-Eifel). Mainzer geowiss. Mitt. 11:157-172. Mainz, Germany.
Orr, W. L. 1977. Sulfur in heavy oils, oil sands and oil shales. In O. P. Strausz and E. M. Lown,
eds., Oil Sand and Oil Shale Chemistry, pp. 223-244, N.Y.: Verlag Chemie.
Orr, W. L. 1986. Kerogen/asphaltene/sulfur relationships in sulfur-rich Monterey oils. In D.

Leythaeuser and J. Rullkötter, eds. Advances in Organic Geochemistry 1985. Org. Geochem.
10:499-516.
Powell, T. G. 1987. Depositional controls on source rock character and crude oil composition.
12th WPC, 2: 31-42, New York: Wiley.
Prezelin, B. B., A. L.Allredge. 1983. Primary production of marine snow during and after an
upwelling event. Limnol. Oceanogr. 6:1156-1167.
Silver, M. W., A. L. Shanks, J. D. Trent. 1978. Marine snow: Microplankton habita and source
of small-scale patchiness in pelagic populations. Science 201:371-373.
Sinninghe Damsté, J. S., W. I. C. Rijpstra, A. C. Kock-Van Dalen, J. W. De Leeuw, P. A.

Schenck. 1989. Quenching of labile functionalised lipids by inorganic sulphur species:
evidence for the formation of sedimentary organic sulphur compounds at the early stages of
diagenesis. Geochim. Cosmochim. Acta 53(6):1343-1356.
Teichmüller, M. 1971. Anwendung kohlenpetrographischer Methoden bei der Erdöl-

Erdgasprospektion. Erdöl und Kohle 24:69-76. Hamburg, Germany.
Teschner, M. and H. Wehner. 1985. Chromatographic investigations on biodegraded crude

oils. Chromatographia 20:407-416.
Wehner, H. and H. Hufnagel. 1986. Some characteristics of the inorganic and organic
composition of oil shales from Jordan. Mitt. Geol. Paläont. Inst. Univ. Hamburg 60:381-395.
Welte, D. H. 1965. Kohlenwasserstoffgenese in Sedimentgesteinen: Untersuchungen über den

thermischen Abbau von Kerogenen unter besonderer Berü ksichtigung der n-Paraffin-Bildung.
Geol. Rdsch. 55:131-144.


21. Organic Matter Response to Change of Depositional

Environment in Kimmeridgian Shales, Dorset, U.K.
A 30-m-thick section of the Kimmeridge clay outcropping on the south coast of Dorset (U.K.) has
been studied. In this interval, lithofacies are highly variable. Geochemical data and optical
examination of thin sections show a wide range of variations in the content and the composition of
the sedimentary organic matter. This fluctuation is discussed in terms of depositional environments
including changes in the planktonic production, in detrital input, and in the level of oxygenation of
the bottom water. In the investigated section, a relationship seems to exist between the chemical
properties of the organic matter and the occurrence and types of biogenic sedimentary structures.
This suggests that the extent of biological degradation, which in turn is related to fluctuation in the
oxygenation of benthic environments is an important factor controlling the organic status in the
sediments.
Organic facies are considered to be basically the result of the combined influence of biomass
productivity, biochemical alteration, depositional processes, and sediment accumulation rate
prevailing during the time of deposition (Muller and Suess 1979; Demaison and Moore 1980; Ibach
1982; Pratt 1984; Stein 1986; Huc 1988a, b).
The relative importance of these controlling factors determines the regional and stratigraphic
occurrence and quality of organic-bearing formations. Field evidence shows the vertical
heterogeneity of sediment and organic matter within source rocks during prolonged periods of
organic-rich mudrock deposition (Pratt 1984; Huc, Irwin, and Schoell 1985, Stow and Atkin 1987).
These heterogeneities have to be carefully considered when source rocks are investigated for oil
exploration purposes (i.e., evaluation of petroleum potential, expulsion behavior, selection of
samples for oil-source rock correlation studies).
In this respect the Kimmeridge clay formation, outcropping on the south coast of Dorset (U.K.),
which exhibits rapidly alternating beds of organic-poor mudstones, organic-rich shales, highly
organic-rich oil shales, and carbonate-rich layers, is an interesting example of extreme
heterogeneity within the same source rock (Gallois 1976; Tyson, Wilson, and Downie 1979; Morris
1980; Cox and Gallois 1981; Williams 1986; Myers and Wignall 1987; Oschmann 1988).
In this paper the fluctuations of organic facies occurring through a selected section of the
Kimmeridge clay in Dorset are discussed. Discussion is based on the data related to organic matter
richness and properties and on examination of thin sections. This study has been conducted on
material collected during two field expeditions, which have been possible thanks to the help of the
British Geological Survey.
Material

In northwestern Europe, the upper Jurassic Kimmeridge clay formation is a thick organic-bearing
mudrock sequence, which is commonly recognized as the main source rock for both North Sea and
southern England oils (Glennie 1984).
The Kimmeridge clay, outcropping as a cliff on the south coast of Dorset, belongs to the Wessex
Basin, which is located south of the Brabant High, which in turn acts as a boundary separating this
basin from more northerly basins: East Midlands Shelf, Cleveland Basin, Sole Pit Trough, and the
southern North Sea Basin. In the Kimmeridge Bay area, the total thickness of the Kimmeridge clay
formation is circa 500 m (Gallois 1978). A broad anticline exposed a portion of it, which includes the
middle part of the Euxodus ammonite zone to the top of the Kimmeridge formation. The sediments
consist of alternating decimetric to metric beds of mudrocks and shales, which range from organic-
poor to very organic-rich shales. A few of the very rich organic beds have been locally mined to be
processed as oil shales during the last century. Calcareous and dolomitic horizons occur at intervals
within the mudstone-shale-dominated sequences (Cox and Gallois 1981; Oschmann 1988). These
alternating beds are practically unaffected by tectonic features and can be easily traced along the
cliff.
For the purpose of this study a 30-m-thick section overlapping the Wheatleyensis and the
Hudlestoni ammonite zones has been investigated. This section, which is topped by the Short Joint
Coal, a coccolithic limestone sandwiched by two oil shale beds, and based by the Grey Ledge, a
prominent dolomitic cementstone band, includes the well known Blackstone oil shale horizon and
the Rope Lake Head, another prominent dolomitic cementstone (figure 21.1).
Samples have been collected along the section at 20-cm intervals and, in several portions, at 5-cm
intervals. As far as possible, special care has been taken to recover unweathered samples.
Organic Geochemistry
Throughout the studied interval, the total organic carbon content (TOC), measured by a LECO
apparatus, is high with an average of 11.7% TOC (for the 20-cm-spaced samples). The organic
content of the different lithofacies varies dramatically throughout the sequence (standard deviation:
9.6), from 40-60% TOC in the Blackstone bed, a finely laminated oil shale with calcareous and
pyritic concretions (plate 21.1a), down to 1-3% TOC in pale gray mudstones and in the Rope Lake
Head and Grey Ledge dolomitic cementstones, which also display decimetric burrowing features
(plate 21.1g).
Rock-Eval pyrolysis reveals substantial differences in the properties of the organic matter. The
hydrogen index, from the series of 20-cm-spaced samples, exhibits an average of 608 mgHC/
gTOC, a value usually found for Type II organic matter. However the hydrogen index ranges from
200 up to 850 (standard deviation: 118). Rock-Eval pyrolysis shows that the level of organic
maturity is low (Tmax = 415°-420°C).
When plotted on a graph displaying hydrogen index versus total organic carbon content, most of the
samples are distributed along a clear trend (figure 21.2). This plot includes additional points
corresponding to a more detailed sampling (1-cm spacing) within the Blackstone bed. On this graph
the very organic-rich samples (50-60% TOC) have a high hydrogen index (750 to 850). As the
organic content of the samples decreases down to circa 20% TOC, the hydrogen index stays high

but tends to decrease slowly from 750-850 down to 650-750; beyond this point the hydrogen index
drops in correspondence with the decrease in TOC. Alteration of the pyrolysis signal of the organic-
poor samples due to mineral matrix effect (Espitalie, Madec, and Tissot 1980) cannot be inferred to
explain this decrease of hydrogen index since the same relationship occurs when kerogens
(isolated from the sediment by acid digestion) are pyrolyzed instead of whole sediments (figure
21.2). This trend is thus related to a real variation in the composition of the organic matter as a
function of organic concentration in the sediment.
Variation in the composition of the organic matter is associated with a significant change in the
isotopic signature of the kerogen. The 13C value of the kerogen varies widely between -22 and -
28 . The organic-rich oil shales have kerogen with less negative 13C values than adjacent
organic-poor beds. Moreover, the enrichment in 13C is concurrent with increasing organic content
in the sediment (figure 21.3). In contrast with the variability in the kerogen composition, it seems
that, regardless of the sediment lithology and organic content, the gas chromatograms of the
saturated hydrocarbons in the extractable bitumen have very similar patterns (figure 21.4). The
chromatograms are dominated by C15-C17 n-alkanes and are characterized by the importance of a
high-molecular-weight mode related to polycyclic hydrocarbons. As previously reported by Mann
and Myers (1989), GC-MS data indicate that the latter mainly include hopanes, steranes, and
methylsteranes. The distribution of these biomarkers exhibits no significant change with the organic
richness of the samples (figure 21.5). The presence of 17 -21 hopanes series confirms the low
level of thermal maturity of the organic matter in the investigated section.
Optical Examination
Samples within the range 2.5-48.2% total organic carbon content were selected for optical
examination. Thin sections parallel and perpendicular to the sedimentary stratification have been
prepared and were observed in reflected and transmitted light. Organic concentrates have been
obtained by palynological-type methods. In thin sections perpendicular to the stratification, the
samples exhibiting the highest organic content appear to be mainly composed of elongated brown-
orange bodies (100 m) well organized along stratigraphic planes, which are interbedded by mineral
layers (plate 21.1b). In thin sections parallel to the stratification, these bodies display more or less
discoidal shapes. Judged from their shape, size, and transparency, they are presumed to be of
algal origin, but no taxonomic relation can be inferred. These bodies are destroyed during the
process of acidic digestion and cannot be observed as such in the organic concentrates; however,
they probably contribute significantly in the amorphous organic matter, which is the dominant phase
in these preparations. In thin sections the shape and boundaries of these bodies tend to become
poorly defined as total organic carbon content of the samples decreases (plate 21.1b, c, d, e).
Moreover, a progressive change in the composition of the algal remains is suggested by the
observation of the sequence of thin sections. Besides the brown-orange bodies, which have been
previously described, two other types of organic entities have been tentatively identified as algal
remains: one is reddish in transmitted light and occurs in all the samples, and the other is perfectly
transparent in transmitted light, can be observed only under ultraviolet excitation, and exhibits a
yellow-green fluorescence color (Bertrand et al. 1989). Compared with the brown-orange bodies,
the relative proportion of these two algal populations appears to increase as TOC decreases; in the
lowest TOC samples, they are the dominant algal constituents.
Woody fragments are recognized in thin sections and organic concentrates. According to optical

estimation the woody particles, which often appear to be pyritized, compose a subordinate fraction
of the organic concentrates. Typically the woody particles account for 1 to 10%, whereas
amorphous material accounts for 90 to 95% of the bulk organic matter. In the lowest TOC samples,
the woody debris becomes very small in size. Thus, recognizable terrestrial plant debris appears to
have been a small component of the organic matter deposited within the sedimentary interval
investigated.
Sediment fabrics observed in thin sections reveal a wide range of patterns. The most organic-rich
samples (more than 40% TOC) show a well-defined alternating organic and mineral horizontal
arrangement (plate 21.1b). Within the TOC range 40-25%, in spite of an increase in the relative
contribution of the mineral phase, the previously described sedimentary arrangement is relatively
well preserved, suggesting that the main process responsible for the decrease of the TOC is a
progressive dilution of organic inputs by a relative increase of the mineral input (plate 21.1c, d).
The first evidence of biogenic sedimentary structures appears in samples containing 20-25% TOC;
they occur as occasional submillimetric microburrows, parallel to the stratification. These
microburrows are filled by a micritic phase associated with grains of pyrite, which appeared as black
particles in transmitted light (plate 21.1e). As the TOC of the samples decreases, the frequency of
the bioturbation features increases and the size of burrows increases. In a 13.1% TOC sample, only
burrows less than 100 m wide occur, whereas in a 6.3% TOC sample, the burrows less than 100
m wide become more frequent, together with a second type of burrows 0.2-1 mm wide (plate
21.1f). In samples with TOC content less than 3% TOC, the latter type becomes dominant. Bigger
bioturbation features can even be directly observed in the field in the organic-poor Rope Lake Head
cementstone bed (2-3% TOC), in which they appear as oblique, mainly Zoophycos-type, burrows,
centimetric in diameter and decimetric in length (plate 21.1g). Besides this increase in size and
frequency, the burrows become subparallel to the stratification, and clearly cut across stratification
as the TOC of the samples decreases. Additionally the quantity of pyrite tends to increase in the
burrows and become a significant filling material, together with recognizable remains of organic
debris (plate 21.1e, f, h).
Discussion
As far as organic matter is concerned, the dramatic fluctuation of organic concentration is an

obvious feature of the investigated organic-rich sedimentary interval (figure 21.1). Within 30 m of
section, beds containing up to 60% TOC (the carbon content of the related kerogen being circa 65-
70%, we are dealing with quasi-pure organic sediment) alternate with beds in which organic carbon
accounts only for 1 to 3% of the sediment, whereas other beds display the whole range of
intermediate organic contents. The second salient feature is that the composition of the sedimentary
organic matter is also variable, and this composition apparently accompanies the change in organic
concentration. Possible explanations for these observations might be rationalized in terms of
primary biological input (i.e., modification of the planktonic communities associated with a variation
of productivity or change in the relative input of continental organic matter diluting the
autochthonous organic material) or in terms of subsequent alteration by biogenic reworking within
the water column and the surface sediment. Variations in the environmental conditions (i.e., water
fertility and availability of specific nutrients) are known to promote significant changes in the thriving
of specific planktonic forms and to result in variations in the chemical properties of the resulting
sedimentary organic matter (Calvert and Price 1983; Pelet 1983; Sellner, Hendrikson, and Ochoa
1983). Actual changes in the composition of the planktonic remains, which is apparently the main

organic constituent in these sediments, have been suggested by the study of thin sections.
However, this probable variation in the planktonic population does not seem to be reflected in the
hydrocarbon composition of the extractable bitumen as exemplified in figures figure 21.4 and figure
21.3. Moreover, even if a change in planktonic species cannot be disregarded as a controlling factor
of the sedimentary organic matter properties, its influence is probably insufficient to independently
account for the observed important decrease of the hydrogen index values for samples with less
than 20% TOC.
The change in the hydrogen index as a function of decreasing organic content can also be
explained by a progressive mixing of a given organic-rich sediment displaying a high hydrogen
index with an organic-poor detrital sediment with a low hydrogen index. Such a situation is
consistent with isotopic data, which can be interpreted in terms of an increasing influence of more
continently derived organic matter, often characterized by more negative 13C (Sackett and
Thompson 1963; Silverman 1967; Degens and Stoffers 1976), in the organic-poor samples. In order
to test this hypothesis we have constructed a numerical model simulating the influence of such a
mixing on the hydrogen index values. In this model we have considered the mixing of a sediment A,
characterized by a given organic carbon content (TOCA) and a given hydrogen index (HIA), and of
a sediment B, characterized by a given organic carbon content (TOCB) and a given hydrogen index
(HIB), HI and TOC being respectively the hydrogen index and the total organic carbon content of
the resulting sediment. HI can be computed as a function of TOC according to the following
equation (see appendix 21.1):
This mixing model has been applied according to the following criteria.
The occurrence of beds of sediments composed of quasi-pure organic matter in the studied
sedimentary interval (Blackstone and associated levels) provides a well-defined end member for the
simulation; they are taken as sediment A (TOCA = 60%, HIA = 820). This sediment is considered to
be progressively diluted by a sediment B containing an organic matter displaying a hydrogen index
equivalent to the lowest values obtained in our set of samples (HIB = 200). The organic content of
this sediment B is unknown. However, since it is supposed to be of detrital origin, a relatively low
content can be inferred. Accordingly, the simulation has been performed using three different values
for TOCB: 1%, 2%, 3%, values that can be considered as reasonably representative of the organic
content of sediments deposited in dominantly detrital situations (Debyser et al. 1978; Huc et al.
1986; Bustin 1988).
The curves HI versus TOC representative of these three mixing scenarios are shown in figure 21.2,
where they can be compared with field data. The general pattern of these curves demonstrates that
mixing phenomena can affect the hydrogen index values according to a trend similar to the one
exhibited by our set of actual samples, with a dramatic decrease in HI with the lowest TOC values.
However, if the fit can be considered as good in the organic-rich range (TOC>25%), discrepancies
tend to occur in the samples with TOC<25%. Moreover, the best fit is obtained for TOCB = 3%,
which is probably an unrealistic value in our situation: on the basis of palynological assemblage
studies (Partington 1983, in Mann and Myers 1989), arid to semiarid climate has been inferred to
prevail during the biostratigraphic zones including the considered sedimentary interval (end of

Wheatleyensis, beginning of Hudlestoni ammonite zones). Such a climate is unlikely to have

supported a high terrestrial biomass productivity and consequently to have resulted in the discharge
of significant amounts of organic-rich detrital sediment. Evidence of the low contribution of the
terrestrial biomass in the samples is provided by the low content of woody particles observed in thin
sections. This is further confirmed by the low concentration of high-molecular-weight, odd-
predominant n-alkanes, often associated with organic matter derived from land plant biomass, and
which does not show any noticeable increase in the low TOC samples (figure 21.4).
An alternative explanation to account for the variation in quantity and quality of the organic matter
throughout the investigated sedimentary section is related to environmental changes. The
occurrence, frequency, and size of burrows penetrating the sea bottom sediments are considered to
be strongly dependent on the dissolved oxygen content of the water bathing the surface sediment
(Rhoads and Morse 1971; Byers 1977; Savrda, Bottjer, and Gorsline 1984). According to this
concept, anaerobic environments are characterized by well-laminated sediments, whereas most
sediments sitting within dysaerobic and aerobic conditions are completely bioturbated and
homogenized. Intermediate, partially bioturbated sediment fabrics are believed to result from short-
term migration of the dysaerobic-anaerobic boundary in the water column. Accordingly, the
occurrence of macrobioturbated beds (clearly visible in the Rope Lake Head cementstone)
alternating with microbioturbated intervals and finely laminated oil shales is evidence of fluctuation
in the concentration of oxygen in bottom water, probably produced by the temporal changes in the
position of the dysaerobic-anaerobic boundary in the water column as a response to variations of
primary productivity and/or water circulation. During periods of lowering of the dysaerobic-anaerobic
boundary, eventually intersecting the surface sediments, the alteration of organic matter is likely to
be significantly enhanced, owing to transit through a more oxygenated water column. Additionally,
burrowing activity of bottom-dwelling fauna, which is dependent on oxygen, results in a mixing of
the upper layer of sediment and greatly increases the time of exposure of organic matter to
decomposition processes (Demaison and Moore 1980).
Besides a decrease in the quantity of organic matter ultimately buried in the sediment, it has been
suggested that this increased biogenic reworking leads to a decrease of the hydrogen index of the
remaining organic matter (Pratt 1984; Hollander 1989). In this respect, the abrupt drop in the
hydrogen index, which is observed around 20-25% TOC in figure 21.2 and which is apparently
concomitant with the first occurrence of microburrows in the sediments, and the subsequent
progressive decrease of organic content and hydrogen index with the increase in frequency and
size of the burrows are probably significant.
It is thus suggested that, in the Dorset series, biogenic reworking is probably an important factor
controlling the destruction of organic matter (decrease in TOC) and the alteration of its chemical
properties (decrease of HI).
The change observed in the 13C of the kerogen has to be considered within the scope of this
hypothesis. Various papers have shown that changing 13C is not necessarily related to change in
the type of organic input (terrestrial versus aquatic precursors) (Stiller 1977). This change can be
rationalized also in terms of differential degradation of the organic matter and isotopic composition
of the dissolved inorganic carbon in the water, which may fluctuate as a result of carbon cycling in
the water column.
Biological reworking has been proposed as an important process causing a fractionation toward
lighter isotopic values in the residual organic matter in the water column (Jeffrey et al. 1983) or

during early diagenesis (Hatcher et al. 1983). Autotrophic organisms rely for photosynthesis on the
assimilation of dissolved CO2 in surface water. The CO2 fixation by primary producers results in an
isotopic fractionation responsible for an enrichment of 12C in photosynthesized biomass as
compared with dissolved CO2 (Park and Epstein 1960; Deuser, Degens, and Guillard 1968;
Galimov 1980; Hayes 1988). Additionally the carbon isotopic composition of the primary biomass
will be influenced by the 13C value of the available CO2. As a consequence of degradation of
organic matter in the water column, 12C-enriched CO2 can be regenerated back into the surface
water, thereby decreasing the 13C value of the CO2 in surface water reservoir. The organic matter,
which is subsequently produced from the assimilation of this CO2, and a portion of which is
eventually buried in the sediment, should have more negative 13C value. In this respect the
structure of the water mass, by controlling the extent of biological degradation of organic matter and
the water circulation regime, is likely to regulate the isotopic composition of the CO2 in surface
water reservoir (Kusspert 1982; Schoell 1984; Hollander 1989). In summary, the temporal rise of
the dysaerobic-anaerobic boundary in the water column, by increasing the organic preservation and
by limiting the reutilization of organically derived 12 C-enriched CO2, can tentatively explain the
observed more positive 13C values of the kerogens in association with high TOC contents in the
studied section.
The variation in the organic matter properties in a sedimentary section of an outcrop of the
Kimmeridgian shales is tentatively explained in terms of change in depositional environment.
The sediments containing between 25-60% total organic carbon were deposited in strictly anaerobic
conditions, resulting in a well-laminated sediment fabric. In these sediments, decrease in TOC,
associated with a faint decrease in hydrogen index, can be explained in terms of dilution by detrital
sediments exhibiting low content of terrestrial-derived organic matter and/or in terms of change in
the planktonic production (productivity and communities composition).
On the other hand, the sediments containing less than 25% total organic carbon exhibit low to
moderate bioturbation features and were probably deposited in more oxygenated conditions. In
these sediments the decrease of organic content is associated with an important change in the
chemical properties of the organic matter. The relationship that occurs between this change and the
increase in size and frequency of biogenic burrows, which are controlled by the oxygen level of the
benthic environment, suggests an increasing influence of biological degradation related to
environmental variations.
Appendix 21.1
Construction of the numerical model: We consider a given weight of autochthonous sediment A

(WA), characterized by a total organic carbon TOCA, a petroleum potential S2 A (mg HC/g of
sediment), provided by Rock-Eval pyrolysis, and a hydrogen index HIA (mg HC/g TOC), which is
progressively mixed by an increasing weight of sediment B (WB), characterized by a total organic
carbon TOCB, a residual petroleum potential S2 B and a hydrogen index HIB. The resulting
sediment will be characterized by a total organic carbon TOC, a residual petroleum potential S2, a
hydrogen index HI and a weight WT = WA + WB.

Acknowledgments
We would like to acknowledge the British Geological Survey and more specially Dr. R. W. Gallois

for stimulating discussions from which the idea of the study originated and for helping us obtain the
permission to work in the protected Kimmeridge Bay area; Dr. B. M. Cox, who introduced us to the
field stratigraphy; and Dr. I. Penn for helpful advice. We would like to thank J. P. Herbin and F.
Kenig for constructive discussions and J. Dufau, J. Fernandez, and S. Harsono, who participated
actively in the field expeditions. Thanks are given to Professor R. Letolle (University of Paris VI),
who provided us with the isotopic data.
References
Bertrand, P., E. Lallier-Verges, L. Martinez, B. Pradier, P. Tremblay, A. Y. Huc, R.

Jouhannel, J. P. Tricart. 1989. Spatial relationships between organic matter and mineral
groundmass in the microstructure of the Dorset formation organic-rich rocks (Great Britain).
In B. Durand and F. Behar, eds., Advances in Organic Chemistry, pp. 661-676. Oxford:
Pergamon Press.
Bustin, R. M. 1988. Sedimentology and characteristics of dispersed organic matter in

Tertiary Niger delta: origin of source rocks in a deltaic environment. AAPG Bull. 68:1179-
1192.
Byers, C. W. 1977. Biofacies patterns in euxinic basins: a general model. In H. E. Cook and
P. Enos, eds., Deep-Water Carbonate Environments. SEPM spec. pub. 25:5-17.
Calvert, S. E. and N. B. Price. 1983. Geochemistry of Namibian shelf sediments. In E. Suess

and J. Thiede, eds., Coastal Upwelling, Its Sediment Record, part A, pp. 337-376. New York:
Plenum.
Cox, B. M. and R. W. Gallois. 1981. The stratigraphy of the Kimmeridge clay of the Dorset
type area and its correlation with other Kimmeridge sequences. Rep. Inst. Geol. Sci. London
80:4.
Debyser, Y., F. Gadel, C. Leblond, and M. J. Martinez. 1978. Etude des composes
humiques, des kerogenes et de la fraction hydrolysable. In Geochimie Organique des
sediments marins profonds, Orgon II. Atantique Nord-Est, Bresil, pp. 339-354. Paris: Ed.
CNRS.
Degens, E. T. and P. Stoffers. 1976. Biogeochemistry of stable isotopes. In G. Eglington and

M. Murphy, eds., Organic Geochemistry, pp. 304-329. Berlin: Springer Verlag.
Demaison, G. T. and G. T. Moore. 1980. Anoxic environment and oil source bed genesis.
AAPG Bull. 64:1179-1209.
Deuser, W. G., E. T. Degens, and R. R. L. Guillard. 1968. Carbon isotope relationships

between plankton and sea water. Geochim. Cosmochim. Acta 32:657-660.
Espitalie, J., M. Madec, and B. Tissot. 1980. Role of mineral matrix in kerogen pyrolysis;
influence on petroleum generation and migration. AAPG Bull. 64:59-66.

Galimov, E. M. 1980. C13/C12 in kerogens. In B. Durand, ed., Kerogen: Insoluble Organic

Matter from Sedimentary Rocks, pp. 271-299. Paris: Technip.
Gallois, R. W. 1976. Coccolith blooms in the Kimmeridge clay and the origin of North Sea oil.
Nature 259:473-475.
Gallois, R. W. 1978. A pilot study of oil shale occurrences in the Kimmeridge clay. Rep. Inst.
Geol. Sci. n. 78/13.
Glennie, K. W. 1984. Introduction to the Petroleum Geology of the North Sea. Oxford:
Blackwell.
Hatcher, P. G., E. C. Spiker, N. M. Szeverenyi, and G. E. Maciel. 1983. Selective

preservation and origin of petroleum-forming aquatic kerogen. Nature. 305:498-501.
Hayes, J. M. 1988. Mechanism of "preservation" of organic matter. GSA annual conference,

GSA abstract, A295.
Hollander, D. J. 1989. Carbon and nitrogen isotopic cycling and organic geochemistry of
eutrophic lake Greifen: implications for preservation and accumulation of ancient organic
carbon-rich sediments. Ph.D. thesis, Univ. Zurich. Diss. ETH no. 8916.
Huc, A. Y. 1988a. Sedimentology of organic matter. In: F. H. Frimmel and R. F. Christman,

eds., Humic Substances and Their Role in the Environment, pp. 215-243. Dahlem
Konferenzen. Chichester: Wiley.
Huc, A. Y. 1988b. Aspects of depositional processes of organic matter in sedimentary

basins. In L. Mattaveli and L. Novelli, eds., Advances in Organic Geochemistry 1987. Org.
Geochem. 13:263-272.
Huc, A. Y., H. Irwin, and M. Schoell, 1985. Organic matter quality changes in an upper
Jurassic shale sequence from the Viking Graben. In: B. M. Thomas et al., eds., Petroleum
Geochemistry in the Exploration of the Norwegian Shelf, pp. 179-183. London: Graham and
Trotman.
Huc, A. Y., B. Durand, J. Roucachet, M. Vandenbroucke, and J. L. Pittion. 1986. Comparison

of three series of organic matter of continental origin. In D. Leythauser and J. Rullkötter,
eds., Advances in Organic Geochemistry 1985, Part 1, Petroleum Geochemistry. Org.
Geochem. 10:65-72.
Ibach, L. E. J. 1982. Relationship between sedimentation rate and total organic content in
ancient marine sediments. AAPG Bull. 66:170-188.
Jeffrey, A. W. A., R. C. Pflaum, J. M. Brooks, and W. M. Sackett. 1983. Vertical trends in

particulate organic carbon 13C:12C ratios in the upper water column. Deep Sea Res. 30:971-
983.

Kusspert, W. 1982. Environmental changes during oil shale deposition as deduced from
isotope ratios. In G. Einsele and A. Seilacher, eds., Cyclic and Event Stratification, pp. 482-
501. Berlin: Springer Verlag.
Mann, A. L. and K. J. Myers. 1989. The effect of climate on the geochemistry of the
Kimmeridge clay formation, pp. 139-142. ACS preprints, Dallas meeting, April 9-14, 1989.
Morris, K. A. 1980. Comparison of major sequences of organic-rich mud deposition in the

British Jurassic. J. Geol. Soc. London 137:157-170.
Muller, P. J. and E. Suess. 1979. Productivity, sedimentation rate, and sedimentary organic
content in the oceans. 1. Organic carbon preservation. Deep Sea Res. 26A:1347-1362.
Myers, K. J. and P. B. Wignall. 1987. Understanding Jurassic organic-rich mudrocks. New

concepts using gamma-ray spectrometry and paleoecology: examples from the Kimmeridge
clay of Dorset and the Jet rock of Yorkshire. In J. K. Legget and G. G. Zuffa, eds., Marine
Clastic Sedimentology, pp. 172-189. London: Graham and Trotman.
Oschmann, W. 1988. Kimmeridge clay sedimentation, a new cycling model. Palaeoclimatol.

Palaeoecol. 65:217-251.
Park, R. and S. Epstein. 1960. Carbon isotope fractionation during photosynthesis. Geochim.
Geochim. Acta 21:110-126.
Partington, M. A. 1983. Unpublished PhD thesis, University of Aberdeen. In A. L. Mann and

K. J. Myers, The effect of climate on the geochemistry of the Kimmeridge clay formation, pp.
139-142. ACS preprints, Dallas meeting, April 9-14, 1989.
Pelet, R. 1983. Preservation and alteration of present-day sedimentary organic matter. In M.

Bjoroy, ed., Advances in Organic Geochemistry 1981, pp. 241-250. Chichester: Wiley.
Pratt, L. M. 1984. Influence of paleoenvironment factors on the preservation of organic

matter in Middle Cretaceous Greenhorn Formation, Pueblo, Colorado. AAPG Bull. 68:1146-
1159.
Rhoads, D. C. and J. W. Morse. 1971. Evolutionary and ecologic significance of oxygen-

deficient marine basins. Lethaia 4:414-428.
Sackett, W. M. and R. R. Thompson. 1963. Isotopic organic carbon composition of recent

continental derived clastic sediments of eastern gulf coast. AAPG Bull. 47:525-531.
Savrda, C. E., D. J. Bottjer, and D. S. Gorsline. 1984. Development of a comprehensive,

oxygen-deficient marine biofacies model: evidence from Santa Monica, San Pedro, and
Santa Barbara basins, California continental borderland. AAPG Bull. 68:1179-1192.
Schoell, M. 1984. Recent advances in petroleum isotope geochemistry. Org. Geochem.

6:645-663.

Sellner, K. G., P. Hendrikson, and N. Ochoa. 1983. Relationship between the chemical
composition of particulate organic matter and phytoplankton distributions in recently
upwelled waters off Peru. In: E. Suess and J. Thiede, eds., Coastal Upwelling, Its Sediment
Record, Part A, pp. 273-288. New York: Plenum.
Silverman, S. R. 1967. Carbon isotopic evidence for the role of lipids in petroleum. J. Am. Oil
Chem. Soc. 44:691-695.
Stein, R. 1986. Organic carbon and sedimentation rate. Further evidence for anoxic deep-
water conditions in the Cenomanian/Turonian Atlantic ocean. Mar. Geol. 72:199-209.
Stiller, M. 1977. Origin of sedimentation in Lake Kinneret traced by their isotopic

composition, In: H. L. Golterman and W. Junka, eds., Interactions Between Sediments and
Freshwater, pp. 54-64.
Stow, D. V. A. and B. P. Atkin. 1987. Sediment facies and geochemistry of the Upper
Jurassic mudrocks in the Central North Sea area. In: J. Brooks and K. Glennie, eds.,
Petroleum Geology of the North West Europe, pp. 797-808. London: Graham and Trotman.
Tyson, R. V., R. C. L. Wilson, C. Downie. 1979. A stratified water column environmental

model for the type Kimmeridge clay. Nature 277:377-380.
Williams, P. F. V. 1986. Petroleum geochemistry of the Kimmeridge Clay of onshore

southern and eastern England. Mar. Petrol. Geol 3:258-281.


22. The Distribution and Generation of Hydrocarbons in

Carbonate Source Rocks.
Most common types of sedimentary rocks can be described as mixtures of shales, limestones, and
siliceous rocks. Oil generation and expulsion in most sedimentary basins, which usually contain
various rock types controlled by the clay-carbonate-silica mineral balance, can be understood better
by evaluating each type of source rock. In this paper, we principally focus our attention on the early
generation of hydrocarbons from carbonate-rich rocks. Recent and ancient carbonates usually
contain smaller amounts of amorphous Type I or II kerogen than of clay-rich sediments. Laboratory
simulations show that the rate of melanoidin formation is relatively high in the presence of clay
minerals, whereas reactions in association with CaCO3 show extremely low rates of melanoidin
formation. On the other hand, Gehman (1962) shows that the average absolute amounts and
distribution of hydrocarbons of ancient limestones and shales are nearly identical, increasing from
recent to ancient sediments in both cases. In recent sediments, however, the average absolute
amounts of hydrocarbons in clay-rich sediments are significantly less than in carbonates-rich
sediments. This leads to the interpretation that carbonate sediments contain a proportionately
higher hydrocarbon content than clay-rich sediments do, even during initial deposition. The increase
of hydrocarbons in carbonates with greater burial is not significant in comparison with clay-rich
sediments. We suggest that the pattern of generation and expulsion of hydrocarbons is not the
same in carbonate and shale source rocks.
The sources of hydrocarbons during early stages of generation in carbonates are attributed to
characteristic organic constituents, that is, sapropelic organic matter predominating in
proteinaceous and solvent-soluble organic matter, which did not contribute to the formation of
"protokerogen." In contrast, in the presence of clay minerals those compounds would have
contributed to easy kerogen formation. The protokerogen formed in the depositional environment of
carbonates abound in compounds with sulfur-carbon bonds. The early generation of hydrocarbons
from carbonate sediments is also attributed to this fact. Most of the organic matter left after the early
generation of hydrocarbons is decomposed in oxygenated environments. This results in a lower
TOM(TOC) content and higher bitumen-hydrocarbons/TOM(TOC) in carbonate rocks than in clay-
rich sediments (shales).
Most common types of sedimentary rocks in the sedimentary sequence are perhaps best described
in terms of the end member concept (Krumbein and Sloss 1963), as indicated in figure 22.1. The
quartz, clay, carbonate, and chert vertices represent sandstone, shale, and limestones, including
evaporite and siliceous sediments formed chemically, respectively.
If we consider the sedimentary rocks broadly, as above, such sediment types can be used not only
to recognize their depositional environment but also to discriminate their significance during
hydrocarbon generation and expulsion as either source rocks or cap rocks. In the present study,
however, sandstone is excluded because of its insignificance as a petroleum source rock.

Accordingly, biogenic silica is adopted in place of sandstone and chert.
Of the three rock types above, the important roles of clay-rich sediments (shales) as petroleum
source rocks have been studied from various viewpoints by many researchers. Also, recent studies
of carbonates and siliceous shales reveal their characteristic roles as petroleum source rocks and
provide a generalized explanation for the origins of both immature and mature oils (Hunt 1967; Fu
Jiamo 1980; Taguchi 1982a, b, c; Jones 1984; Palacas 1984; Taguchi, Hasegawa, and Suzuki
1988).
The ultimate goal of our study is to identify characteristics among the three types of rocks from the
viewpoint that the carbonate-clay-silica mineral balance has a significant effect on oil generation
and expulsion. This paper mainly documents, however, the effectiveness of carbonate sediments as
petroleum sources in comparison with clay-rich sediments mostly from a standpoint of oil
generation. The significance of siliceous sediments as petroleum source rocks has been reported
elsewhere recently (Taguchi, Hasegawa, and Suzuki 1988).
The Controversy Over Carbonate Sediments as Possible
Source Rocks
Status of Studies on Carbonate Sediments as Possible Sources Despite the fact that 40% of
"reservoir oils" in major world oil fields or 50% of total oil in "giant oil fields" (Moody 1975) is
produced from carbonate reservoirs, only the excellent pioneer works by Hunt (1961, 1967) and
Gehman (1962) and a few other studies have been published concerning the source potential of
carbonate rocks except for relatively recent studies. Moreover, much controversy persists
concerning the effectiveness of carbonate rocks as major sources of petroleum (e.g., Palacas 1983;
Gardner and Bray 1984; Hunt and McNichol 1984; Jones 1984; Oehler 1984; Connan et al. 1985;
Baric, Maricic, and Radic 1988).
The most important aspects of carbonate sediments, as pointed out by those who propose that
carbonate rocks are important major sources of petroleum, are as follows: (1) carbonate sediments
generally have a much lower content of total organic carbon (TOC) and a higher bitumen-
hydrocarbon/TOC ratio than clay-rich sediments; (2) the organic matter in carbonates mostly
consists of oil-prone sapropelic and/or Type I or Type II kerogen; (3) the carbon preference index
(CPI) of carbonates normally has values lower than 1.0, (i.e., an even number carbon preference);
and (4) pristane/phytane ratios (pr/ph) are usually lower than 1.0, as summarized by Taguchi
(1982a, b) on the basis of various studies on carbonate rocks (Kvenvolden 1970; Dembicki et al.
1976; Tissot and Welte 1978; Hunt 1979; Connan et al. 1980; Demaison et al. 1980; Kikuchi,
Takasu, and Watanabe 1980; McKirdy and Kanstler 1980; Mori and Taguchi 1981). As a result,
most of those who advocate carbonates as possible source rocks suggest a lower wt percent of
TOC is required to produce a "good source rock" than is the case for clay-rich sediments (shales).
Low Content of TOM (TOC) and High Bitumen-Hydrocarbons/TOM (TOC)
Here, the term hydrocarbon is defined as the organic matter extracted from rock that contains only
the elements carbon and hydrogen. The bitumen is organic matter soluble in organic solvents. It
consists of hydrocarbons and nonhydrocarbons, such as resin and asphaltene. TOM (total organic

matter) is estimated by multiplying the organic carbon usually (but sometimes nitrogen content) by a
certain factor (so-called "organic conversion factor") for which values differ slightly among different
authors. The reason that terms such as "TOM(TOC) and bitumen-hydrocarbon/TOM(TOC)" are
used in this paragraph head is that problems on organic matter distribution in sediments are mostly
discussed with the two terms used interchangeably and the bitumen/TOC ratio generally correlates
with the hydrocarbon/TOC ratio.
Of the important aspects concerning carbonate sediments mentioned above, item (1) has been
debated for a long time and remained unsolved. Gehman (1962) first pointed out that ancient
carbonate rocks have a lower TOM content and a higher hydrocarbon/TOM ratio than ancient
shales do, despite the fact that the average organic contents are identical in both recent carbonate
and clay-rich sediments (figure 22.2). At the same time, he suggested that most carbonates lose
organic matter faster than shales because of exposure to oxygenated meteoric water. Degens
(1967) proposed an alternative interpretation, that carbonate-rich sediments contain primarily
organic-rich proteins that are hydrolyzed during recrystallization, whereas clay-rich sediments
contain primarily humic materials (figures 22.3 and 22.4). Hunt (1967) asked (but did not resolve)
why the residual organic matter in such carbonates generates hydrocarbons so efficiently. Jones
(1984) more recently proposed that the primary control on the bitumen/TOC is the amount of TOC
and not the percentage of carbonate. Using data of Uspenskiy and Chernysheva (1951) (table 22.1)
and a hypothesis of Bordovsky et al. (1974), he suggested that the high bitumen/TOC ratio in
carbonate rocks is due to the faster rate of kerogen decomposition in carbonates. At the same time,
he emphasized that the total amounts of bitumen and/or hydrocarbons in low-TOC rocks is much
less than that necessary to generate significant oil accumulations, in contrast to the general opinion
that carbonates do not require as much TOM as clay-rich sediments (shales) to be source rocks.
We believe that significant unsolved problems remain in the Uspenskiy and Chernysheva study. For
example, no consideration is given to the relationship between the grain size distribution and total
organic carbon content, despite the fact that the samples studied show sizable differences in
carbonate content and recognizable differences in grain size distribution.
In recent sediments, as will be shown below, changes of grain size distribution in sediment
significantly influence the correlation between TOC and bitumen/TOC. Moreover, the siliceous
Mowry shales have a clear positive correlation between bitumen and TOC (Schrayer and Zarrella
1966), contrary to results obtained by Uspenskiy and Chernysheva (1951).
A Case Study: The UMM Addalkh Oil Field, Abu Dhabi, U.A.E.
In conjunction with the problem above, organic geochemical results from a case study of the UMM
Addalkh field are summarized below (Mori and Taguchi 1981; Taguchi and Mori 1982).
Carbonate rocks belonging to the Shilaif, Mishirif, and Laffan formations (Upper Cretaceous, listed
in ascending order) were sampled from three wells (UA-5, -6, and -8 in table 22.2). Most of the
samples studied contained more than 95 wt percent carbonate (calcium carbonate and magnesium
carbonate, table 22.2) and correspond to "pure" limestone according to the classifications of
Pettijohn (1975). Only a few samples contain less than 95% carbonate (figure 22.5 and table 22.2).
The magnesium carbonate content ranges from 1.8 to 4.5 wt percent, with an average of 3.2%.
One of the purposes of this study was to examine the relationship between thermal maturation and
bitumen/TOC ratios in carbonates. Unfortunately, however, the vitrinite reflectance (percent Ro,

figure 22.6) does not delineate the oil window. The data appear to indicate the carbonates are
thermally immature. On the other hand, elemental analyses of kerogens (table 22.3 and figure 22.7)
show that most samples from the Shilaif Formation plot in the zone of Type I and/or II kerogens on
a Van Krevelen diagram, while samples from the deeper Mishirif Formation plot in the zone of Type
II and/or near Type III kerogens. The kerogen data also indicate that all samples from the Shilaif
Formation fall into the principal zone of oil generation (figure 22.7). This discrepancy observed
between the maturity obtained by vitrinite reflectance measurement (percent Ro) and elemental
analyses of kerogens may be caused by analysis problems. The "vitrinite" being measured is
probably not "true vitrinite," as suggested by Thompson-Rizer and Dembicki (1986) and Thompson-
Rizer (1987), or the discrepancy may be attributed to scatter in measurements of percent Ro of
different types of vitrinite such as vitrinite 1 and vitrinite 2, as reported by Buiskool Toxopeous
(1983), Durand et al. (1986), and Tissot, Pelet, and Ungerer (1987).
As is evident from table 22.3 and figure 22.8, the sulfur contents of kerogens isolated from the
carbonate samples are much higher than those in clay-rich sediments (shales), even when sulfide
mineral (pyrite) contamination found closely associated with kerogens is considered. This tendency
of higher sulfur concentrations in carbonates is confirmed by comparing the sulfur contents of
kerogens in carbonates with those of kerogens in a shale sample and a thin coal seam in our
sample set. The problem of high sulfur content of kerogens in carbonates as compared with those
in clay-rich sediments is discussed later.
Extractable organic matter (EOM) was obtained in a two-stage process yielding two fractions: EOM
(1) and EOM(2). EOM(1) was extracted from powdered samples smaller than 200 mesh, with
benzene-methanol solvents (90/10 v/v) at room temperature with a stoppered bottle in a mechanical
shaker. EOM(2) was obtained after EOM(1) extraction by treatment of the residue with 1N HCl.
Therefore, "EOM" contents represent the sum of EOM(1) and EOM(2). EOM(1) and EOM(2) were
separated into saturate, aromatic, and nonhydrocarbon fractions by elution chromatography on
silica gel. N-alkanes in EOM(1) and EOM(2) were separated from branched and cyclic
hydrocarbons by urea adduction and then were analyzed by gas chromatography. "Hydrocarbon
contents" (total hydrocarbon contents) represent the sum of hydrocarbons obtained from EOM(1)
and EOM(2). The residues of carbonates thoroughly removed from the original samples by
repeated treatment with 1N HCl are designated as "HCl-insoluble residues" in this study.
The HC/TOC ratio of the studied samples (table 22.2) ranges from 0.09 to 1.55 with an average of
0.38. These values are distinctly higher than the average values of 0.14 from carbonate rocks
obtained by Hunt (1972) and 0.01-0.005 from "normal" rocks (i.e., shales) by Tissot and Welte
(1984). The CPI of n-alkanes obtained from EOM(1) indicates even carbon number preference from
carbonates as reported in many studies (Welte and Waples 1973, and others), while those from
EOM(2) definitely indicate odd carbon number preference. Probably, this result is attributable to the
fact that EOM(2) is derived from organic matter closely associated with HCl-insoluble residues
consisting mostly of clay minerals as is discussed later. The pr/ph (pristane/phytane) ratios of two
samples, examined preliminarily in this study, are 0.9 and 0.4. These values are consistent with the
usual observation that pr/ph values are generally less than 1.0 in organic fractions with even carbon
number preference of n-alkanes (Welte and Waples 1973).
Figure 22.9 shows an inverse relationship of the bitumen/TOC ("extract index") with TOC consistent
with Uspenskiy and Chernysheva (1951, table 22.1). The extract index (E) is defined by the
following equation (Sato, Sasaki, and Taguchi 1972).

Accordingly, the extract index nearly corresponds to the degree of "bituminization (BA)" introduced
by Bordovskiy (1965) and Bordovskiy and Takh (1978). Textures of carbonate rocks were not
determined in our study, as was the case in the study by Uspenskiy and Chernysheva (1951). On
the other hand, the TOC of the samples studied shows a strong positive correlation with HCl-
insoluble residue content, as shown in figure 22.10 Two mudstone samples (#579 and 584; table
22.1 and figure 22.10), which show much higher organic carbon content than other samples, have
much higher HCl-insoluble residue content as well (5.70 and 5.32 wt percent, respectively).
We attribute the positive correlation between the HCl-insoluble residues and TOC to the fact that
organic matter in carbonates is easily affected by hydrolysis, while more resistant organic matter
tends to be closely associated with the HCl-insoluble residue (which consists principally of clay
minerals). However, note that the loss of organic matter by hydrolysis is different from the loss of a
large fraction of organic matter in carbonates shortly after deposition as is discussed later.
Relationship Between Particle Size Distribution and Bitumen-Hydrocarbons/

TOM(TOC) Ratio
It is well known that the texture of sediments becomes finer as organic content increases (Trask
1932; Emery and Rittenberg 1952; Emery 1960). According to Trask's pioneer work (1932), clays
contain on average twice as much organic matter as silts do, and silts have twice that of sands. At
present, the underlying relationship that controls the correlation between texture and organic
content is considered to be primarily due to the similarity between the density of clay particles and
organic debris and to the chemical and/or mechanical adsorption of organic molecules onto clay
minerals.
In connection with this point, an inverse relationship between TOM and the bitumen/TOM ratio,
which was pointed out in studies of Caspian Sea Recent sediments (Bordovskiy and Takh 1978)
and in Recent Bering Sea sediments (Bordovskiy 1965), seems to be governed by grain size
(figures 22.11 and 22.12). We have drawn each trend line for grain size in these figures according
to our interpretation based on the textual data described in their papers. As is evident from figures
22.11 and 22.12, organic content decreases as the texture of sediments becomes coarser, but
bitumen/TOM ratios increase. It appears, therefore, that the primary factor controlling the bitumen/
TOM ratio is not the amount of TOM but rather the grain size of sediments. For sediments with a
certain range of grain sizes, the bitumen/TOM ratio clearly shows a proportional relation to the
amount of TOM.
Figure 22.13 shows the relationship between extractable organic matter (bitumen) and TOC content
and also between TOC and the extract index in Neogene Tertiary rocks from Japanese oil fields
(Taguchi 1962; Sato, Sasaki, and Taguchi 1972). Bitumen and TOC are positively correlated, as
was the case in the study of Bering Sea sediments (Bordovskiy 1965; figure 22.12). In contrast to
the Bering Sea sediments, the relationship between the TOC and the extract index in Japanese
Tertiary sediments shows a slightly positive correlation. However, if we consider the particle size

distribution of the samples, the extract index in the fine-grained sandstone-siltstone group tends to
be slightly higher than in the other groups of fine-grained samples. Namely, this is a tendency
consistent with organic content decreasing and bitumen/TOC ratio increasing as the texture of
sediment becomes coarser.
The apparent difference between the two studies is only in the difference of treatment of
sedimentary textures in the studied samples. However, the increase in the bitumen/TOM ratios and/
or the exact index values as the texture of sediments becomes coarser will arise as an important
problem in regard to the migration of oil. Future research along this line will make an important
contribution to the study of the formation of oil deposits.
Relationship Between Bitumen-Hydrocarbons/TOC and TOC in Ancient

Argillaceous Rocks
figure 22.14 shows the relationships between benzene-soluble extract and TOC (upper) and
between extractability and TOC (lower) in a large number of samples from the Cretaceous siliceous
Mowry Shale, Wyoming, U.S.A. (Schrayer and Zarrella 1966). "Extractability" is defined by Schrayer
and Zarrella as the number of grams of benzene-soluble extractable organic matter per 100 g of
organic carbon. Because organic matter extracted from sediments by use of benzene consists
primarily of hydrocarbons, as reported by Ferguson (1962), "extractability" approximately
corresponds to the HC/TOC ratio. Since the EOM/TOC ratio generally correlates with the HC/TOC
ratio, extractability may be regarded geochemically as being similar to degree of bituminization
(Bordovskiy 1965) and/or the extract index (Sato, Sasaki, and Taguchi 1972). The relationships
between extract yield and the TOC and between "extractability" and TOC are approximately equal
and proportional (figure 22.14).
A positive correlation of the bitumen ratio and/or the hydrocarbon ratio to the TOC seems to be a
generalized relation for argillaceous rocks, except in special cases such as those for samples from
the zones of oil generation or oil migration.
Factor Controlling the Inverse Correlation Between TOM(TOC) and Bitumen-

Hydrocarbons/TOM(TOC)
The inverse correlation between bitumen/TOM and TOM in Recent sediments from the Bering Sea
and the Caspian Sea reported by Bordovskiy (1965) and by Bordovskiy and Takh (1978) can be
actually viewed as positive correlations when grain sizes are considered. However, studies of
ancient noncarbonate rocks (Taguchi 1962; Schrayer and Zarrella 1966; Sato, Sasaki, and Taguchi
1972) reveal that the bitumen-hydrocarbons/TOC ratio and TOC are directly proportional to each
other. Only two studies on ancient carbonate rocks (Uspenskiy and Chernysheva 1951; Taguchi
and Mori 1982) showed an inverse relationship between the bitumen/TOM(TOC) ratio and TOM
(TOC), though these studies did not indicate the grain size of the samples studied.
How, then, can we explain the inverse relationship observed in ancient carbonates between
bitumen/TOM(TOC) and the TOM(TOC)? A primary concern is that the studies above unfortunately
did not describe grain size in the studied samples. At present, we believe that although it is usually
difficult to determine grain sizes in carbonate rocks except in special cases, such as those of clastic
or coquina carbonate rocks, the confusion about sedimentary texture in carbonate rocks prevents a

clear understanding of the primary factors controlling the inverse correlation between the TOM
(TOC) and the bitumen/TOM(TOC) ratio in carbonate rocks.
Consequently, we suggest that the difference observed between the bitumen-hydrocarbons/TOM

(TOC) and the TOM(TOC) in carbonate rocks and noncarbonate rocks is basically attributable to (1)
the inherent nature of the organic matter in pure carbonates and (2) the role of clay minerals
contained in those rocks. The inverse correlation between bitumen/TOM(TOC) and the TOM(TOC)
is probably caused by the mixing of characteristic organic matter in carbonates with organic matter
associated with clay minerals and/or exotic organic debris in some cases. Our interpretation is
indirectly supported by the positive relationship between TOC and the HCl-insoluble residue, the
odd carbon number preference in EOM(2) from the Abu Dhabi carbonate rocks, and the abnormally
high bitumen/TOC and/or hydrocarbons/TOC in carbonates as compared with shales, as previously
noted, as well as by the characteristic properties of organic matter in carbonates and the poor
formation of kerogen in carbonates.
We conclude that the degree of bituminization (bitumen/TOM), extractability, and the extract index,
in general, are directly proportional to TOM(TOC), except in special cases such as those for
samples from the oil window zone or near oil migration paths and from pure carbonate sediments
contaminated by terrestrial material. Accordingly, Jones' conclusion (1984), that carbonates are
ineffective petroleum sources because the ratio of bitumen-hydrocarbons/-TOM(TOC) increases as
TOM(TOC) content decreases, needs to be reconsidered in the future.
Comparison of Organic Constituents in Carbonate andClayey Sediments
figure 22.3 shows the differences in the distribution of major organic fractions between Recent
marine carbonates and clay-rich sediments reported by Degens (1967) and Hunt (1967). We note
that amino compounds constitute 90% or more of total organic matter in Recent shell material and
in Florida Bay carbonate sediments, whereas they constitute less than 10% in clay-rich sediments.
At the same time, the "organic residue" in carbonates, which possibly corresponds to humic and
lignitic organic matter, or "protokerogen" in the modern sense, occupies only a few percent of the
total organic matter, whereas in the clay-rich sediments it occupies a large part of the organic
matter.
Kerogen Content
figure 22.15 shows the relative amounts of kerogen and bitumen in various ancient rock types,
which we based on Bergman's data (1963). Limestone has an extremely low content of kerogen, as
compared with other relatively clay-rich rocks. figure 22.16 indicates the relative amounts of
kerogen and bitumen in carbonate rocks from the Umm Addalkh Oil Field, Abu Dhabi (Taguchi and
Mori 1982). As is well known, more than 90% of total organic matter is kerogen in clay-rich rocks
(shales). Therefore, it is apparent that the relative content of kerogen in carbonate rocks is
considerably lower than in clay-rich rocks. Both recent and ancient carbonate sediments usually
contain relatively small amounts of kerogen, as compared with recent and ancient clay-rich
sediments.
figure 22.17 shows the experimental results of the formation of acid-insoluble melanoidins, which
corresponds to kerogen-like melanoidins (Taguchi and Sampei 1986), in three different reaction

mixtures: nonassociation, CaCO3, and montmorillonite suspended with glucose-glycine or glucose-

alanine systems. The rate of melanoidin formation is highest in the montmorillonite-associated
mixtures and is extremely low in CaCO3 mixtures. The experimental results are strongly consistent
with the observation (above) that carbonate sediments usually contain relatively small amounts of
kerogens in comparison with clay-rich sediments in both recent and ancient sediments.
In connection with this, Suzuki and Taguchi (1983) report that amorphous kerogen is present in the
clay-rich sedimentary rocks in close association with clay minerals. Therefore, we can estimate that
kerogens in clay-rich sediments consist of amorphous kerogen (Types I and II) most strongly
associated with clay minerals, and herbaceous and woody kerogens (Types II and III) are also
associated with clay minerals, but less strongly than amorphous kerogens (depending on
depositional environments). The kerogen in clay-rich sediments probably could survive compaction
and diagenesis, in contrast with the type of organic matter inherent to carbonate sediments, as is
discussed later.
Recently, an alternative hypothesis for the conventional mechanism of kerogen formation has been
given by several researchers (e.g., Largeau et al. 1986; Goth et al. 1988; Tegelar et al. 1989).
According to their studies, the selective preservation of insoluble and nonhydrozable
macromolecular materials derived from extant organisms is mainly responsible for the formation of
kerogen. We think that this mechanism does not conflict with our interpretation, given the difference
between the bitumen-hydrocarbons/TOM(TOC) and the TOM(TOC) in carbonate rocks and
noncarbonate rocks, and also with our observation that the relative content of kerogen in carbonate
rocks is considerably lower than in clay-rich rocks.
Sulfur Contents of Kerogen
The sulfur content of kerogen in carbonates seems to be significantly higher than in clay-rich
sediments. As seen in table 22.3, the sulfur content in kerogen in carbonates ranges approximately
from 8 to 12% with an average of 10% (normalized to the major elements C, H, O, S, N). In
contrast, Tissot and Welte (1984) report that sulfur content ranges from 2 to 3.8% in Type I
kerogen, obtained mostly from oil shales, with an average value of 2.7%; 0.2 to 4.9% in Type II
kerogen from shales and oil shales, with an average of 2.7%; 0.0 to 0.3% in Type III kerogen from
shales, with an average of 0.16%, and 0.5 to 2.5% in coal, with an average of 1.26%. Hunt (1979)
gives an average value of 5% as a representative value for kerogen. figure 22.8 shows clearly that
kerogens in carbonate rocks from the Umm Addalkh oil field are considerably richer in organic sulfur
than "typical kerogens" from argillaceous rocks. A high content of sulfur in kerogens of carbonate
rocks seems to be common.
Orr (1986) reported that Monterey Formation of California contains high-sulfur kerogens ("Type II-
S") and that Type II-S kerogens generate heavy oil at lower temperatures than classical Type II
kerogens do, because of a high abundance of weak carbon-sulfur bonds. Jones (1984), Lewan
(1985), and Tannenbaum and Aizenshtat (1985) suggest that sulfur-rich kerogens begin to generate
oil at lower maturation levels than Type II kerogens do. Sinninghe Damsté et al. (1989) and Taguchi
(1989) suggest that sulfur-rich kerogens are more likely to be formed in a nonclastic environment
such as carbonate and siliceous rocks.
Consequently, it is suggested that early generation of hydrocarbons from carbonates is caused in

part by the breaking of sulfur-carbon bonds at a lower temperature than conventionally thought

possible.
Hydrocarbon Content
figure 22.2 shows the comparison between carbonates and clay-rich sediments concerning their
average TOC and the average total hydrocarbon contents from recent and ancient sediments
(Gehman 1962). The lower TOM and higher hydrocarbon/TOM ratio in carbonates, as compared
with those in clay-rich sediments (shales), have been mostly discussed from a viewpoint of the
effectiveness of carbonates as possible petroleum sources. However, we emphasize the
importance of the difference of average absolute amounts of hydrocarbons in carbonates and
shales from a different point of view.
As stated previously, the average absolute amounts of hydrocarbons in ancient carbonates

(limestones) and clay-rich sediments (shales) are nearly identical, showing an increase from recent
to ancient in both type of sediment (figure 22.2). In recent sediments, however, the average
absolute amount of hydrocarbon in clay-rich sediments is slightly less than in carbonates. The
average absolute amount of hydrocarbons in clay-rich sediments significantly increases from recent
to ancient, whereas that in carbonate sediments only slightly increases. This shows that carbonate
sediments have a much higher hydrocarbon content than clay-rich sediments do, even at the initial
stage of deposition, and that the increase of hydrocarbons with increasing burial depth in
carbonates is not significant in comparison with clay-rich sediments.
Summary of Hydrocarbon Generation from Carbonates
We regard most hydrocarbons present in carbonate rocks as generated from carbonate sediments
early in their diagenesis. Some part of these hydrocarbons is considered to have been synthesized
by organisms. That is, most hydrocarbons in carbonate rocks have not been generated by "fossil
kerogen" during catagenesis, as is conventionally expected in shales. Accordingly, the sources of
hydrocarbons of early generation in carbonates are attributed to their characteristic organic
constituents, that is, sapropelic organic matter predominating in proteinaceous and solvent-soluble
organic matter (Degens 1967), which did not contribute to the formation of protokerogen, although
in the presence of clay minerals those compounds should have contributed to fossil kerogen
formation. As is well known, proteinaceous and compounds soluble in organic solvent readily
convert to light hydrocarbons through hydrolysis, decarboxylation, deamination, and microbial
fermentation under relatively low temperature (e.g., Thompson and Creath 1966; Philippi 1977).
On the other hand, the protokerogen formed in the depositional environment of carbonates, which
occupy a relatively small part of organic matter in carbonate sediments, abound in compounds with
sulfur-carbon bonds, as stated previously. The early generation of hydrocarbons from carbonate
sediments is also supported from this viewpoint. Most organic matter such as proteins,
carbohydrates, and lipids are probably not involved in the formation of kerogen, but most of those
compounds would have been converted to hydrocarbons directly through microbial fermentation.
This is supported by CPI values of less than 1.0 and low pr/ph ratios in carbonate sediments and
partly through recrystallization of carbonates in a relatively earlier diagenesis as suggested
previously.
Therefore, the decomposition of organic matter in carbonate sediments results in a lower TOM
content in carbonate rocks, which would mostly have occurred after the generation of hydrocarbons.

The loss of organic matter was caused by its occasional exposure to oxygenated environments
(Gehman 1962) and partly by hydrolyzation of organic matter during crystallization of carbonates.
The remarkable differences in stability between the hydrocarbons and the remaining
nonhydrocarbon organic matter in carbonates, except that closely associated with clay minerals, is
probably responsible for the great survival of hydrocarbons relative to nonhydrocarbon organic
matter. Our use of the term nonhydrocarbon organic matter means all organic matter exclusive of
hydrocarbons present in carbonates. Accordingly, the nonhydrocarbon organic matter surviving in
carbonate rocks is a remnant after the generation of most of the hydrocarbons and the
decomposition of other kinds of organic matter.
Most of the hydrocarbons generated during the earlier diagenetic stage (before the consolidation
stage) tend to be retained until migration paths are developed through fractures, fissures, and
solution channels, owing to early lithification of carbonates (as suggested by Hunt 1967). Moreover,
the early generation of hydrocarbons in carbonate sediments, not via fossil kerogen, should
accompany the formation of "immature oil" deposits. In this connection, Hetherington and Horan
(1960) and Bashbush et al. (1983) report that tar mats occur in Burgan Field in Kuwait and in El
Bundug Field in Abu Dhabi/Qatar. Bashbush et al. (1983) suggest that these tar mats originated in
the paleo-oil water contact, at a reservoir depth estimated to be less than 1,000 m, indicating the
formation of oil at shallower depths than expected.
The generation of hydrocarbons from carbonates does not always follow the hydrocarbon genesis
model observed in clay-rich sediments (shales). Most hydrocarbons in carbonate rocks are
generated during earlier depositional stages and are retained in the rock matrix through the
diagenesis-to-catagenesis stages because of their relative stability as compared with
nonhydrocarbon organic matter. Eventually they are provided the opportunity to migrate under
optimal geological conditions. Most of the organic matter left after the early generation of
hydrocarbons is decomposed in oxygenated environments. This results in a lower TOM(TOC)
content and higher bitumen-hydrocarbons/TOM(TOC) in carbonate rocks than in clay-rich
sediments (shales). These conditions, and the formation of cap-rocks by earlier consolidation of
carbonate-evaporite sequences, will commonly lead to the formation of "immature oil" deposits if
reservoirs exist at shallower depths.
Acknowledgment
We gratefully acknowledge the president, Takeyoshi Tanaka, Dr. Hiroshi Hosoi, and Mr. Satoshi
Tono of Japan Petroleum Development Co. Ltd., Tokyo, for financial support and permission to
publish and for having provided us with samples through the study of carbonate rocks from the
Umm Addalkh Oil Field, Abu Dhabi. We also thank Dr. Saburo Iwasa, director of Umm Addalkh
Development Co., and Messrs. Isamu Ogura, Manabu Shiobara, and Yoshihiro Shimizu of Japan
Petroleum Development Co. for practical help during this study. A special thanks is given to Dr.
Yoshihiko Fujita, director of the Technical Research Center, Teikoku Oil Co., Ltd., and Mr. Shuji
Kudo, director of Central Technical Laboratory, Japan Petroleum Exploration Co., Ltd., for their
support and encouragement during the preparation of this paper. The associate editor, Dr. J. K.
Whelan, is acknowledged for her valuable comments.
References

Baric, G., M. Maricic, and J. Radic, 1988. Geochemical characterization of organic facies in
the Dugi Otok Basin, the Adriatic Sea (Yugoslavia). Org. Geochem. 13:343-349.
Bashbush, J. L., W. K. Savage, R. B. Nagai, T. Ogimoto, J. Wakamiya, and H. Takizawa,

1983. A reservoir optimization study-El Bunduq Field, Abu Dhabi, Qatar. 3d SPE of AIME
Middle East Tech. Conf., Bahrain, preprint no. SPE 11481, pp. 1-7.
Bergman, W. 1963. Geochemistry of lipids. In I. A. Breger, ed., Org. Geochem., pp. 503-542.
Oxford: Pergamon.
Bordovskiy, O. K. 1965. Accumulation of organic matter in bottom sediments. Mar. Geol.

3:33-82.
Bordovskiy, O. K. and N. I. Takh, 1978. Organic matter in the Recent carbonate sediments of
the Caspian Sea. Oceanol. 18:673-678.
Bordovskiy, O. K. et al. 1974. Evaluation of the role of bottom fauna in the transformation of
organic matter in sediments (with specific reference to the deep-sea detritus feeders in the
Kuril-Kamchatka Trench). Oceanol. 14:128-132.
Brooks, J. 1981. Organic maturation of sedimentary organic matter and petroleum

exploration-a review. In J. Brooks, ed., Organic Maturation Studies and Fossil Fuel
Exploration, pp. 1-37. London: Academic Press.
Buiskool Toxopeus, J. M. A. 1983. Selection criteria for the use of vitrinite reflectance as a
maturity tool. In J. Brooks, ed., Petroleum Geochemistry and Exploration of Europe, pp. 295-
307. Oxford: Blackwell Scientific Publications.
Conan, J., G. Hussler, and P. Albrecht. 1980. Geochemistry of crude oils and crude oil:
Source rock correlations in four carbonate basins. Abs. GSA Annual Meeting, Atlanta
Georgia, October 1980, p. 405.
Connan, J., J. Bouroullec, D. Dessort, and P. Albrecht. 1985. The microbial input in
carbonate-anhydrite facies of a sabkha paleoenvironment from Guatemala: A molecular
approach. In D. Leythaeuser and J. Rullkötter, eds., Advances in Organic Geochemistry,
1985. Org. Geochem. 10:29-51. Oxford: Pergamon Press.
Degens, E. T. 1967. Diagenesis of organic matter. In G. Larsen and G. V. Chilinger, eds.,

Diagenesis in Sediments, pp. 343-390. Amsterdam: Elsevier.
Demaison, G., F. Bourgois, and F. Melendez. 1980. Geochemistry and petrology of Miocene
(Alcanor) carbonate source beds, Casablanca Field, Tarragona Basin, Spain. Abs. GSA
Annual Meeting, Atlanta Georgia, October 1980, pp. 411-412.
Dembicki, H., W. G. Meinschein, and D. E. Hattin. 1976. Possible ecological and

environmental significance of the predominance of even-carbon number C20-C30 n-alkanes.

Durand, B., B. Alpern, J. L. Pittion, and B. Pradier, 1986. Reflectance of vitrinite as a control
of thermal history of sediments. In J. Burrus, ed., Thermal Modeling in Sedimentary Basins,
pp. 441-474. Paris: Technip.
Emery, K. O. 1960. The Sea Off Southern California. A Modern Habitat of Petroleum, pp.
366. New York: Wiley.
Emery, K. O. and S. C. Rittenberg, 1952. Early diagenesis of California basin sediments in

relation to origin of oil. AAPG Bull. 36:735-806.
Ferguson, W. S. 1962. Analytical problems in determining hydrocarbons in sediments. AAPG

Bull. 46:1613-1620.
Fu, Jiamo. 1980. Some characteristics of the evolution of organic matter in carbonate
formations. In A. G. Douglas and J. R. Maxwell, eds. Advances in Organic Geochemistry
Gardner, W. C. and E. E. Bray, 1984. Oils and Source Rocks of Niagaran Reefs (Silurian) in
the Michigan Basin. In J. G. Palacas, ed., Petroleum Geochemistry and Source Potential of
Carbonate Rocks, pp. 33-44. AAPG Studies in Geology, no. 18. Tulsa: AAPG.
Gehman, H. M. 1962. Organic matter in limestones. Geochim. Cosmochim. Acta 26:885-

894.
Goth, K., J. W. de Leeuw, W. Puttmann, and E. W. Tegelaar. 1988. Origin of Messel Oil
Shale kerogen. Nature 336:759-761.
Hetherington, G. and A. J. Horan. 1960. Variations with the elevation of Kuwait reservoir
fluids. J. Inst. Pet. 46:109-114.
Hunt, J. M. 1961. Distribution of hydrocarbons in sedimentary rocks. Geochim. Cosmochim.

Acta 22:37-49.
Hunt, J. M. 1967. The origin of petroleum in carbonate rocks. In G. V. Chilinger, J. H. Bissell,

and R. W. Fairbridge, eds., Carbonate Rocks-Physical and Chemical Aspects, pp. 225-251.
Amsterdam: Elsevier.
Hunt, J. M. 1972. Distribution of carbon in the crust of the earth. AAPG Bull. 56:2273-2277.
Hunt, J. M. 1979. Petroleum Geochemistry and Geology, pp. 617. San Francisco: W. H.
Freeman.
Hunt, J. M. and A. P. McNichol. 1984. The Cretaceous Austin Chalk of South Texas-a
Petroleum source rock. In J. G. Palacas, ed., Petroleum Geochemistry and Source Potential
of Carbonate Rocks, pp. 117-125. AAPG Studies in Geology, no. 18. Tulsa: AAPG.

Jones, R. W. 1984. Comparison of carbonate and shale source rocks. In J. G. Palacas, ed.,
Petroleum Geochemistry and Source Rock Potential of Carbonate Rocks, pp. 163-180.
AAPG Studies in Geology, no. 18. Tulsa: AAPG.
Kikuchi, Y., Y. Takasu, and T. Watanabe, 1980. Petroleum geology of the Western Bakr
Area, the Eastern Desert, Egypt. J. Jpn. Assoc. Petrol. Technol. 46:107-119.
Krumbein, W. C. and L. L. Sloss. 1963. Stratigraphy and Sedimentation. 2d ed. San

Francisco: W. H. Freeman.
Kvenvolden, K. 1970. Evidence for transformation of normal fatty acids in sediment. In G. D.

Hobson and G. C. Speers, eds., Advances in Organic Geochemistry 1966, pp. 355-366.
Oxford: Pergamon.
Largeau, C., S. Derenne, E. Casadevall, A. Kadouri, and N. Sellier. 1986. Pyrolysis of

immature torbanite and of the resistant bioplymer (PRB A) isolated from the extant alga
Botryococcus braunii. Mechanism for the formation and structure of torbanite. In D.
Leythaeuser and J. Rullkötter, eds., Advances in Organic Geochemistry 1985. Org.
Geochem. 10:1023-1032.

experimentation. Phil. Trans. Roy. Soc. London A315:123-134.
McKirdy, D. M. and A. J. Kanstler. 1980. Hydrocarbon genesis in Cambrian carbonates of

the eastern Officer basin, South Australia. Abstracts GSA Annual Meeting, Atlanta, Georgia,
October 1980, p. 481.
Moody, J. D. 1975. Distribution and geological characteristics of giant oil fields. In A. G.

Fischer and S. Judson eds., Petroleum and Global Tectonics, pp. 307-320, Princeton, N. J.:
Princeton Univ. Press.
Mori, K. and K. Taguchi. 1981. Petroleum-geochemical study of Upper Cretaceous

carbonate rocks, Abu Dhabi offshore oil field. J. Jpn. Assoc. Petrol. Technol. 47:264.
(abstract in Japanese).
Oehler, J. H. 1984. Carbonate source rocks in the Jurassic Smackover trend of Mississippi,
Alabama, and Florida. In J. G. Palacas, ed., Petroleum Geochemistry and Source Potential
of Carbonate Rocks, pp. 63-69. AAPG Studies in Geology, no. 18. Tulsa: AAPG.
Orr, W. L. 1986. Kerogen/asphaltene/sulphur relationships in sulphur-rich Monterey oils. In

D. Leythaeuser and J. Rullkötter, eds., Advances in Organic Geochemistry 1985. Org.
Geochem. 10:499-516.
Palacas, J. G. 1980. South Florida basin, a prime example of carbonate source rocks of
petroleum. Abstracts GSA Annual Meeting, Atlanta, Georgia, October 1980, p. 495.
Palacas, J. G. 1983. Carbonate rocks as sources of petroleum: Geological and chemical

characteristics and oil-source Correlations. In Proc. 11th World Petrol. Congr. 2:31-43.
Chichester: Wiley.
Palacas, J. G. 1984. Petroleum Geochemistry and Source Rock Potential of Carbonate

Rocks, AAPG Studies in Geology. no. 18. Tulsa: AAPG.
Pettijohn, F. J. 1975. Sedimentary Rocks, 3d ed., New York: Harper & Row.
Powell, T. G. 1980. Geochemical characteristics of oils and source rocks in carbonate

regimes of N. Alberta/N.W.T. and Southern Ontario, Canada. Abstracts GSA Annual
Meeting, Atlanta, Georgia, October 1980, p. 502.
Powell, T. G. 1984. Some aspects of the hydrocarbon geochemistry of a Middle Devonian

barrier-reef complex. In J. G. Palacas ed., Petroleum Geochemistry and Source Potential of
Carbonate Rocks, pp. 45-62. AAPG Studies in Geology, no. 18, Tulsa: AAPG.
Rice, D. D. 1980. Indigenous biogenic gas in Upper Cretaceous chalks, eastern Denver
basin. Abstracts GSA Annual Meeting, Atlanta, Georgia, October 1980, p. 509.
Sato, S., K. Sasaki, and K. Taguchi. 1972. Distribution of organic carbon and extractable
organic matter in Neogene Tertiary rocks of Akita and Niigata Districts, with particular
reference to the removal of carbonate carbons in sediments. J. Geol. Soc. Japan 78:643-
651.
Schrayer, G. J. and W. M. Zarrella. 1966. Organic geochemistry of shales-II. Distribution of

extractable organic matter in the siliceous Mowry Shale of Wyoming. Geochim. Cosmochim.
Acta 30:415-434.
Sinninghe Damsté, J. S., T. I. Eglinton, J. W. de Leeuw, and P. A. Schenck. 1989. Organic

sulphur in macromolecular sedimentary organic matter I. Structure and origin of sulphur-
containing moieties in kerogen, asphaltenes and coal as revealed by flash pyrolysis.
Suzuki, N. and K. Taguchi. 1983. Characteristics and diagenesis of kerogens associated

with clay fractions of mudstone. In M. Bijorøy et al. eds., Advances in Organic Geochemistry
1981, pp. 607-612. Chichester: Wiley.
Taguchi, K. 1962. Basin architecture and its relation to the petroleum source rocks
development in the bordering Akita and Yamagata Prefectures and the adjoining areas, with
special reference to the depositional environment of petroleum source rocks in Japan. Sci.
Rep. Tohoku Univ., Series 3, 7:278-289.
Taguchi, K. 1982a. Study of carbonate source rocks (1)-Characteristics of organic matter in

carbonate sediments. J. Jpn. Assoc. Petrol. Technol. 47(1): 62-72 (in Japanese with English
abstract).
Taguchi, K. 1982b. Study of carbonate source rocks (2)-Geochemical problems on

carbonate sediments as petroleum source rocks. J. Jpn. Assoc. Petrol. Technol. 47(2): 85-
92. (in Japanese with English abstract).
Taguchi, K. 1982c. On the characteristics of carbonate sediments as petroleum source rocks-

a review. Sci. Rep. Tohoku Univ., series 3, 15:149-161.
Taguchi, K. 1989. Formation and expulsion of immature oils in carbonate and siliceous
sediments. Abstracts 28th Internat. Geol. Cong. July 1989, Washington, D. C. 3:210-211.
Taguchi, K. and K. Mori. 1982. Characteristics of organic matter in carbonate sediments-a

case study in Umm Adahlk Oil Field, Abu Dhabi, U. A. E. In Proc. Annual Meeting 1982,
Geochem. Soc. Japan (extended Abstract in Japanese), pp. 317-318.
Taguchi, K. and Y. Sampei. 1986. The formation, and clay mineral and CaCO3 association
reactions of melanoidins. In D. Leythaeuser and J. Rullkötter eds., Advances in Organic
Geochemistry 1985, Org. Geochem. 10:1081-1089. Oxford: Pergamon Press.
Taguchi, K., K. Hasegawa, and T. Suzuki. 1988. The relationship between silica minerals
and organic matter diagenesis: its implication for the origin of oil. In L. Mattavelli and L.
Novelli, eds., Advances in Organic Geochemistry 1987, Org. Geochem. 13:97-108. Oxford:
Pergamon Press.
Tannenbaum, E. and Z. Aizenshtat. 1985. Formation of immature asphalt from organic-rich

carbonate rocks-I. Geochemical correlation. Org. Geochem. 8:181-192.
Tegelaar, E. W., J. W. de Leeuw, S. Derenne, and C. Largeau. 1989. A reappraisal of

kerogen formation. Geochim. Cosmochim. Acta 53:3103-3106.
Thompson-Rizer, C. L. 1987. Some optical characteristics of solid bitumen in visual kerogen

preparations. Org. Geochem. 11:385-392.
Thompson-Rizer, C. L. and H. Jr. Dembicki. 1986. Optical characteristics of amorphous

kerogens and the hydrocarbon-generating potential of source rocks. Int. J. Coal Geol. 6:229-
249.
Tissot, B. and D. H. Welte. 1978. Petroleum Formation and Occurrence. Berlin: Springer
Verlag.
Tissot B. P. and D. H. Welte. 1984. Petroleum Formation and Occurrence. 2d ed. Berlin:
Springer Verlag.
Tissot, B. P., R. Pelet, and Ph. Ungerer. 1987. Thermal history of sedimentary basins,
maturation indices, and kinetics of oil and gas generation. AAPG Bull. 71:1445-1466.
Tissot, B., R. Pelet, B. F. Furollet, and J. L. Oudin. 1980. Recurrent appearance of source-
rock facies in Cretaceous to Eocene carbonate series of Tunisia. Abstracts GSA Annual
Meeting, Atlanta, Georgia, October 1980, p. 536.

Trask, P. D. 1932. Origin and Environment of Source Sediments of Petroleum. Houston: Gulf
Publishing.
Uspenskiy, V. A. and A. S. Chernysheva. 1951. The composition of the organic material from
Lower Silurian limestones in the region of Chudovo City. In Contributions of Geochemistry.
Israel Program for Scientific Translations. 1965, pp. 103-114.
Welte, D. H. and D. Waples. 1973. Uber die Bevorzugung geradzahliger n-alkane in

Sedimentgestein. Naturwissenschaften 60:516-517.

Whelan&Fearrington - 1992 - Materia Organica

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Whelan&Fearrington - 1992 - Materia Organica

Uploaded by

Copyright:

Available Formats

Organic Matter: Productivity, Accumulation, and Preservation in Recent and Ancient Sediments

EARTHSCAPE DATE: 9/99

Organic Matter: Productivity, Accumulation, and Preservation

Columbia University Press: New York

2. Reflections on the Career and Times of John M. Hunt

3. Modeling Petroleum Generation in Sedimentary Basins

4. Sources, Cycling, and Distribution of Water Column Particulate and

5. Organic Matter Accumulation, Remineralization, and Burialin an Anoxic Coastal

6. Organic Carbon Remineralization and Preservation in Sediments of Skan Bay,

7. Preservation of Sargassum Under Anoxic Conditions: Molecular and Isotopic

file:///F|/Usuarios/Juan%20carlos/articulos/articulos...anic%20matter/Organic%20Matter%20Productivity,%20.htm (1 de 3)17/01/2006 07:01:03 p.m.

8. Geochemical Features of Organic Matter in Sediment Cores from Lützow-Holm

9. Sedimentation and Preservation of Amino Compounds and Carbohydrates in

10. Hydrodynamic Controls of Anoxia in Shallow Lakes

11. Organic Carbon Accumulation and Preservation in Marine Sediments: How

12. Early-Stage Incorporation of Sulfur Into Protokerogens and Possible Kerogen

13. Bitumen Classification and Biomarker Correlation Studies Based on Organic

14. Resolution of Sediment Hydrocarbon Sources:Multiparameter Approaches

15. Biomarkers in Recent and Ancient Sediments:he Importance of the Diagenetic

16. Natural Hydrous Pyrolysis: Petroleum Generation in Submarine Hydrothermal

18. Source and Biomarker Composition Characteristics of Chinese Nonmarine

20. Maturity and Facies-Controlled Composition of the Organic Matter of Selected

21. Organic Matter Response to Change of Depositional Environment in

22. The Distribution and Generation of Hydrocarbons in Carbonate Source Rocks.

file:///F|/Usuarios/Juan%20carlos/articulos/articulos...anic%20matter/Organic%20Matter%20Productivity,%20.htm (2 de 3)17/01/2006 07:01:03 p.m.

file:///F|/Usuarios/Juan%20carlos/articulos/articulos...anic%20matter/Organic%20Matter%20Productivity,%20.htm (3 de 3)17/01/2006 07:01:03 p.m.

Organic Matter: Productivity, Accumulation, and Preservation in Recent and Ancient

file:///F|/Usuarios/Juan%20carlos/articulos/articulos.../organic%20matter/Organic%20Matter%20Introduction.htm (1 de 5)17/01/2006 07:00:52 p.m.

need to understand modern-day analogs of depositional environments in order to interpret quantity

file:///F|/Usuarios/Juan%20carlos/articulos/articulos.../organic%20matter/Organic%20Matter%20Introduction.htm (2 de 5)17/01/2006 07:00:52 p.m.

file:///F|/Usuarios/Juan%20carlos/articulos/articulos.../organic%20matter/Organic%20Matter%20Introduction.htm (3 de 5)17/01/2006 07:00:52 p.m.

file:///F|/Usuarios/Juan%20carlos/articulos/articulos.../organic%20matter/Organic%20Matter%20Introduction.htm (4 de 5)17/01/2006 07:00:52 p.m.

file:///F|/Usuarios/Juan%20carlos/articulos/articulos.../organic%20matter/Organic%20Matter%20Introduction.htm (5 de 5)17/01/2006 07:00:52 p.m.

Organic Matter: Productivity, Accumulation, and Preservation in Recent and Ancient

2. Reflections on the Career and Times of John M. Hunt

file:///F|/Usuarios/Juan%20carlos/articulos/articulo.../organic%20matter/Organic%20Matter%20Chapter%202.htm (1 de 11)17/01/2006 06:49:48 p.m.

file:///F|/Usuarios/Juan%20carlos/articulos/articulo.../organic%20matter/Organic%20Matter%20Chapter%202.htm (2 de 11)17/01/2006 06:49:48 p.m.

file:///F|/Usuarios/Juan%20carlos/articulos/articulo.../organic%20matter/Organic%20Matter%20Chapter%202.htm (3 de 11)17/01/2006 06:49:48 p.m.

Publications of Dr. John M. Hunt

file:///F|/Usuarios/Juan%20carlos/articulos/articulo.../organic%20matter/Organic%20Matter%20Chapter%202.htm (4 de 11)17/01/2006 06:49:48 p.m.

1949 Hunt, J. M. and T. W. Dennis. Application of certain Instrumental Methods in

Hunt, J. M. and D. S. Turner. Determination of mineral constituents of rocks by infrared

1957 Hunt, J. M. and J. P. Forsman. Analytical Techniques in the separation of organic

Forsman, J. P. and J. M. Hunt. Insoluble organic matter (kerogen) in sedimentary rocks.

Forsman, J. P. and J. M. Hunt. Insoluble organic matter in sedimentary rocks of marine

Kidwell, A. L. and J. M. Hunt. Migration of oil in Recent sedmients of Pedernales, Venezuela.

1961 Hunt, J. M. Distribution of hydrocarbons in sedimentary rocks. Geochim. Cosmochim.

file:///F|/Usuarios/Juan%20carlos/articulos/articulo.../organic%20matter/Organic%20Matter%20Chapter%202.htm (5 de 11)17/01/2006 06:49:48 p.m.

Hunt, J. M. and M. L. Dunton. Distribution of low molecular weight hydrocarbons in Recent

1966 Degens, E. T. and J. M. Hunt. Evolutionary trends in calcified tissues of Marine

Hunt, J. M. The significance of carbon isotope variations in marine sediments. In G. D.

file:///F|/Usuarios/Juan%20carlos/articulos/articulo.../organic%20matter/Organic%20Matter%20Chapter%202.htm (6 de 11)17/01/2006 06:49:48 p.m.

1972 Hunt, J. M. Distribution of carbon in crust of Earth. AAPG Bull. 56:2273-2277.

Dickey, P. A. and J. M. Hunt. Geochemical and hydrogeologic methods of prospecting for

Hunt, J. M. Organic geochemical studies: introduction and summary. In B. C. Heezen et al.,

Hunt, J. M. Hydrocarbon and kerogen studies, Initial Reports: DSDP 22:673-675.

Hunt, J. M. The petroleum problem. Oceanus 18:4-5.

file:///F|/Usuarios/Juan%20carlos/articulos/articulo.../organic%20matter/Organic%20Matter%20Chapter%202.htm (7 de 11)17/01/2006 06:49:48 p.m.

Emery, K. O. and J. M. Hunt. Summary of Black Sea investigations. In E. T. Degens and D.

Hunt, J. M. Statement on petroleum in the marine environment. Minutes of the Hearing of

Hunt, J. M. Origin of athabasca oil. AAPG Bull. 60:1112.

1977 Hunt, J. M. Distribution of carbon as hydrocarbons and asphaltic compounds in

1978 Hunt, J. M. Characterization of bitumens and coals. AAPG Bull. 62:301-303.