Relationships between the distribution and stable isotopic composition of living benthic...

Relationships between the distribution and stable isotopiccomposition of living benthic foraminifera and cold methaneseep biogeochemistry in Monterey Bay, California

Anthony E. RathburnGeology Program, Indiana State University, Science Building 159, Terre Haute, Indiana 47809, USA([email protected])

Scripps Institution of Oceanography, IOD-0218, 9500 Gilman Drive, La Jolla, California 92093, USA

M. Elena PerezGeology Program, Indiana State University, Science Building 159, Terre Haute, Indiana 47809, USA

Invertebrate Paleontology, Natural History Museum of Los Angeles County, 900 Exposition Boulevard, Los Angeles,California 90007, USA

Jonathan B. Martin and Shelley A. DayDepartment of Geological Sciences, University of Florida, Gainesville, Florida 32611-2120, USA

Chris Mahn, Joris Gieskes, and Wiebke ZiebisScripps Institution of Oceanography, IOD-0218, 9500 Gilman Drive, La Jolla, California 92093, USA

David Williams and Amanda BahlsGeology Program, Indiana State University, Science Building 159, Terre Haute, Indiana 47809, USA

[1] As part of an ongoing effort to explore the use of foraminifera as a means to assess modern and

ancient methane release, we compared ambient pore water chemistry with the distribution and stable

isotopic composition of living (rose Bengal stained) foraminifera in MBARI ROV Ventana tube cores

taken from modern seepage areas (about 1000 m water depth) in Monterey Bay, California. Benthic

foraminiferal isotopic differences between sites clearly indicate that methane-influenced pore waters

affect foraminiferal distributions and carbonate isotope geochemistry. Carbon isotope signatures of living

benthic foraminifera did not conform to the very negative (�30 to �48%), methane-influenced carbon

isotope values of the pore waters they live in. Instead, the influence of methane seep pore waters was

reflected in the greater range and carbon isotopic variability of living seep foraminifera compared with

published d13C values of foraminifera living in nonseep habitats. It is not clear what relative influences

biological, ecological, and physical factors have on the carbon isotopic signatures observed in seep

foraminifera. Substantial carbon isotope differences can exist between individuals of the same seep

species. For instance, d13C values of living Globobulimina pacifica varied by as much as 2.9% between

seeps within 8 km of each other, whereas d13C values of living Uvigerina peregrina varied by as much

as 1.95% within the same seep. Provided there is no diagenetic alteration of the test carbonate, isotopic

results of individual seep foraminifera support the hypothesis that foraminifera can be used to assess past

and present methane seepage.

Components: 13,012 words, 11 figures, 3 tables.

G3G3GeochemistryGeophysics

Geosystems

Published by AGU and the Geochemical Society

AN ELECTRONIC JOURNAL OF THE EARTH SCIENCES

GeochemistryGeophysics

Geosystems

Article

Volume 4, Number 12

20 December 2003

1106, doi:10.1029/2003GC000595

ISSN: 1525-2027

Copyright 2003 by the American Geophysical Union 1 of 28

Keywords: methane seeps; stable isotopes; benthic foraminifera; paleoclimate; benthic ecology; pore water geochemistry.

Index Terms: 4870 Oceanography: Biological and Chemical: Stable isotopes; 4267 Oceanography: General: Paleoceano-

graphy; 4804 Oceanography: Biological and Chemical: Benthic processes/benthos.

Received 19 June 2003; Revised 8 October 2003; Accepted 27 October 2003; Published 20 December 2003.

Rathburn, A. E., M. E. Perez, J. B. Martin, S. A. Day, C. Mahn, J. Gieskes, W. Ziebis, D. Williams, and A. Bahls,

Relationships between the distribution and stable isotopic composition of living benthic foraminifera and cold methane seep

biogeochemistry in Monterey Bay, California, Geochem. Geophys. Geosyst., 4(12), 1106, doi:10.1029/2003GC000595, 2003.

1. Introduction

[2] An understanding of the relationships between

the ecology of microfauna, their habitat, and the

geochemistry of their tests/shells remains the key

to interpreting paleoceanographic conditions from

microfossils. As new light is shed on the ecological

factors that influence species’ distributions and

skeletal/shell geochemistries, paleoceanographers

are better able to model and interpret microfossil

data. Determining the relationships between

climate change, microbiota, and oceanographic

conditions is among the most intriguing pursuits

of geosystems research. The biogeochemical con-

sequences of periodic releases of methane from

marine resources have recently received attention

as potentially important factors in environmental

and biotic changes observed in the geologic record

[e.g., Thomas et al., 2000]. Climate-related release

of methane from marine deposits, known as clath-

rates, has been interpreted from several geologic

records, including those from the Quaternary

[Wefer et al., 1994; Kennett et al., 2000; Smith et

al., 2001; Keigwin, 2002], the Late Paleocene [e.g.,

Dickens et al., 1995, 1997], the Early Cretaceous

[Jahren et al., 2001] and the Jurassic [Hesselbo et

al., 2000]. In order to understand the relationships

between climate, biotic changes, and methane

release, it is crucial to determine the extent and

timing of natural release of greenhouse gases

through geologic time. Previous studies have used

the carbon isotope signals of fossil benthic forami-

nifera (sarcodine protists) to assess the timing of

ancient methane seepage [e.g., Wefer et al., 1994;

Kennett et al., 2000], but modern analogs are

poorly known [Rathburn et al., 2000; Torres et

al., 2003]. In order to gain insight into the use

of benthic foraminifera as indicators of methane

release, we examined living and fossil foraminifera

from known methane seeps in Monterey Bay. Here

we present comparisons of living (rose Bengal

stained) benthic foraminifera, their stable isotopic

signals, and the biogeochemistry of seep pore

waters.

2. Background

[3] Cold seeps actively venting methane are most

easily recognized by the presence of aggregations

of symbiont-bearing clams or vestimentiferan

(tube) worms [e.g., Sibuet and Olu, 1998]. Ancient

seeps typically have been recognized in the geo-

logic record by the presence of clam or worm tube

macrofossils endemic to seeps. However, because

large, epifaunal metazoans do not typically aggre-

gate in areas of diffuse or very rapid methane

seepage [Sibuet and Olu, 1998], a more reliable,

quantitative measure is needed to assess changes in

methane seepage through time.

[4] On the basis of fossil data, Wefer et al. [1994]

suggested that the isotopic geochemistry of calcar-

eous benthic foraminifera might provide a quanti-

tative means to ascertain the history of methane

release for a given region. The ultimate source of

clathrate methane is buried organic material that

has been altered by microbial or thermal processes.

Most methane samples from clathrates have d13Cvalues between �40% and �94%, with the ma-

jority having values more negative than �60%[Kvenvolden, 1993]. Carbon isotope values of

benthic foraminifera are directly or indirectly in-

fluenced by ambient conditions (the bottom waters

or pore waters they are exposed to), biological

factors, and ecological preferences [e.g., McCorkle

et al., 1997]. Some studies of fossil material

GeochemistryGeophysicsGeosystems G3G3

rathburn et al.: benthic foraminifera composition 10.1029/2003GC000595

2 of 28

suggested that, as a result of methane influence,

benthic foraminiferal carbonate secreted in seepage

areas will have more negative d13C values com-

pared to those of nonseep areas [Wefer et al., 1994;

Kennett et al., 1996, 2003]. In support of the idea

that foraminifera can be used to assess ancient

methane seepage, recent molecular studies of an-

cient Santa Barbara Basin sediments indicate that

methanotrophic activity increased dramatically

during episodes of methane release inferred from

fossil foraminifera [Hinrichs et al., 2003]. Never-

theless, this idea has not received universal accep-

tance. A study of modern foraminifera in the Santa

Barbara Basin suggests that some of these carbon

isotopic excursions can be explained by changes in

carbon oxidation rate and diagenesis [Stott et al.,

2002].

[5] The few previous studies of modern methane

seep foraminifera demonstrated that this group

of organisms may be abundant at seeps [e.g.,

Rathburn et al., 2000]. Although some taxa have

been suggested to be more tolerant of potentially

toxic conditions found in seep habitats [Akimoto et

al., 1994; Bernhard et al., 2001], to date, no

endemic foraminiferal species have been found at

methane seeps. Carbon isotope values reported

from previous studies of modern methane seep

foraminifera show wide variability within some,

but not all, taxa [Sen Gupta et al., 1997; Rathburn

et al., 2000]. Contrary to expectations based on

pore water chemistry, any given species that has a

broad depth range within nonseep sediments typi-

cally has low variability of d13C values regardless

of the sediment depth where it is found [Rathburn

et al., 1996; McCorkle et al., 1997]. Comparisons

of foraminifera living in seep pore waters having

extremely negative d13C values is an ideal means to

examine the effects of pore waters on the distribu-

tion and d13C signatures of paleoceanographically

important species.

3. Study Area

[6] Monterey Bay Research Institution ROV

Ventana was used to collect samples from clam

beds and bacterial mats located in two areas within

Monterey Bay (Figures 1 and 2; Table 1). Seep

appearance and activity were very different be-

tween these two areas. In the region known as

‘‘Extrovert Cliffs’’ (36�46.60N, 122�05.10W), at

960 m water depth, the Site we call ‘‘1780’’ had

distinctive seeps that were visually characterized

by highly defined concentric rings of different

subenvironments (Figure 2). In these seeps, a bed

of Calyptogena clams encircled a yellow bacterial

mat, which, in turn, partly encircled a dark gray

bacterial mat (Figure 2). The entire structure was

about 2 meters in diameter, and this ring was one of

several known in the area. At Site 1781 in the area

known as ‘‘Clam Flats’’, (36�44.70N, 122�16.60W),

at 1009 m water depth, a small (about 1 m

diameter) clam bed and a separate, probably inde-

pendent, bacterial mat (<1 m diameter) were sam-

pled. This methane seep area has been described by

Barry et al. [1996], and foraminifera from this area

were characterized by Bernhard et al. [2001].

4. Methods

4.1. Foraminiferal Processing

[7] Tube cores of seep and near-seep environments

were collected on June 22 and 23, 2000 during

ROV Ventana Dives 1780 and 1781 respectively.

Three adjacent (within 20 cm of each other) tube

cores (7-cm diameter) were collected from each

seep subenvironment (bacterial mats and clam

beds). One tube was analyzed for pore water

geochemistry, one for bacteria and biogeochemis-

try, and one for oxygen and foraminiferal analyses.

Figure 1. Bathymetric chart of Monterey Bay and thelocation of the Clam Flats and Extrovert Cliffs seepsites.


rathburn et al.: benthic foraminifera composition 10.1029/2003GC000595rathburn et al.: benthic foraminifera composition 10.1029/2003GC000595

3 of 28

Figure 2. MBARI Ventana photo of ‘‘seep ring’’ at site 1780 near Extrovert Cliffs (image provided courtesy ofMBARI, all rights reserved).

Table 1. Location and Physical Characteristics of Cores Used in This Studya

LocationBottom Water

T/O2 Dive Site Description Core Use

Extrovert Cliffs(36�46.60N, 122�05.10W,960 m)

6�C17.8 mM 1780

Clam bed TC30 Foraminifera,O2 and H2S

TC79 Pore waterGray bacterial mat TC34 Pore water

TC67 Foraminifera,O2 and H2S

Yellow bacterial mat TC31 Pore waterTC29 O2 and H2S

Reference TC38 Pore waterTC71 O2 and H2S

Clam Flats(36�44.70N, 122�16.60W, 1009 m)

4�C15.6 mM 1781

Clam bed TC31 Foraminifera,O2 and H2S

TC80 Pore waterBacterial mat TC28 Pore water

TC30 O2 and H2SReference TC38 Pore water

TC71 O2 and H2STC72 Pore water

aTemperature and oxygen used in the table were measured by the ROV.



4 of 28

After shipboard oxygen microprobe analyses, the

top 0–1 cm of tubes designated for foraminiferal

processing were subsampled, then 0.5-cm sections

were subsampled from 1 cm down to 3 cm, and

1-cm sections down to at least 10 cm. Each slice of

sediment was then preserved in 4% buffered form-

aldehyde (diluting 37% formaldehyde solution by a

factor of about 10 using filtered seawater). Note that

the sampling techniques, preservation and process-

ing is the same procedure (but reported as ‘‘4%

buffered formalin’’) of McCorkle et al. [1990,

1997]; Corliss [1991]; Corliss and Emerson

[1990]; Corliss and van Weering [1993]; Rathburn

and Corliss [1994]; Silva et al. [1996]; and

Rathburn et al. [1996, 2000, 2001]. In the labora-

tory, 65 ml of rose Bengal solution (1 g/L of 4%

buffered formaldehyde) was added to each pre-

served sediment sample and allowed to stain for

at least one week before processing for foraminif-

eral analyses. Samples were then wet-sieved using

63 mm and 150 mm mesh-opening sieves (for more

detailed description of volumetric procedures, see

Rathburn and Corliss, 1994). The 63–150 mmfraction was stored for future analyses. Some

>150 mm fraction samples were split into manage-

able volumes for examination using a modified

OTTO microsplitter. Specimens from the >150 mmfraction of the upper 3 or 5 cm intervals were

identified to the species level and counted. Infaunal

densities of each sediment interval were plotted at

the deepest depth of the sample interval (e.g.,

species densities of the 0 to 1-cm interval were

plotted at 1 cm on graphs). To facilitate compar-

isons with previous work, our sampling intervals,

processing procedures, and graphing methods

correspond to those used in previous studies of

modern foraminiferal ecology and isotopic signa-

tures [e.g., McCorkle et al., 1997; Rathburn et al.,

2000].

4.2. Oxygen and Hydrogen SulfideMicrogradients

[8] Immediately after retrieval of sediment cores,

oxygen and hydrogen sulfide microprofiles were

measured onboard ship prior to sectioning of the

sediment cores and further analyses. Vertical oxy-

gen distribution in intact cores was measured by

Clark-type microelectrodes provided with a built-in

reference and a guard cathode [Jørgensen and

Revsbech, 1989; Revsbech, 1989]. The electrodes

were purchased from UNISENSE, Denmark and

had a sensing tip of 15–20 mm, a stirring sensitivity

of <2% and a 90% response time �1 s. Electrode

currents had a linear response to 0% and 100% air

saturation of O2. Linear calibration was done at

20�C in 100% saturated seawater (35%) and

nitrogen purged seawater with 0% oxygen satura-

tion. The electrodes were attached to a microma-

nipulator, signals were amplified and transformed

to mV by a picoammeter (Unisense PA 2000) and

data were collected directly on a computer. Mea-

surements were performed in vertical increments of

typically 200 mm. The position of the microsensor

tip was observed by using a dissecting microscope.

[9] The H2S microsensor used for the measure-

ments was a miniaturized amperometric sensor

with an internal reference and a guard anode

[Jeroschewsky et al., 1996]. The sensor was

connected to a high-sensitivity picoammeter

(Unisense PA 2000) and the anode was polar-

ized against the internal reference (polarization

voltage +0.085V). Calibration was performed

after the sensor signal had stabilized during

pre-polarization of 24 hours. A stock solution

of S2– (i.e., 100 mM) was prepared from

dissolving Na2S in N2-flushed 0.1 M NaOH in

a closed container. The final concentration of

stock solution was determined by standard anal-

ysis [Cline, 1969]. Oxygen was removed from a

calibration buffer (100 mM phosphate buffer,

pH 7) by vigorously bubbling with N2 before

aliquots were transferred to gas-proof containers

with rubber stoppers. Calibration points were

prepared by injecting suitable amounts of the

S2– stock solution into the calibration vials with

a micro-syringe.

4.3. Stable Isotope Analyses

[10] Before mass spectrometer analysis for car-

bon and oxygen isotopic values, stained (cyto-

plasm-containing; indicating living or recently

living) individuals from the >150 mm fraction

were each cleaned using an ultrasonic bath, exam-

ined microscopically, then stripped of organic



5 of 28

matter by soaking in 15% hydrogen peroxide for

20 minutes and rinsed in methanol. Foraminifera

were reacted at 73�C with anhydrous phosphoric

acid in a Kiel III device connected to a

Finnigan MAT 252 isotope ratio mass spectrom-

eter. Approximately 20 mg of test material was

reacted, which corresponds to roughly one to

six tests per analysis (see Table 2). With large

specimens, particularly of U. peregrina and

G. pacifica, half of the tests were analyzed to

reduce the volume of CO2. Approximately 62%

of the analyses were made using one test (or

less). Data are reported relative to the PDB

standard and precision was measured with an

internal standard of Carrera marble calibrated

with NBS-19. Precision of replicate analyses

of the Carrera marble are ±0.04% for d18Oand ±0.08% for d13C.

[11] Pore water was separated from the sediment

by centrifugation. Measured chemical analyses of

the pore waters include major and minor elements,

metals, and d13CDIC. The elemental and metal data

will be reported elsewhere; only the d13CDIC of the

pore water are included in this paper (Table 3).

The d13CDIC values were measured using tech-

niques described by Graber and Aharon [1991].

For this technique, approximately 3 to 5 ml of pore

water were stored in pre-evacuated containers

(vacutainers) and poisoned with HgCl. In the

laboratory, gas was extracted from the vacutainers

by injection of �100 ml of 100% H3PO4. The CO2

that evolved from the acidified solution was

extracted into a vacuum line, cryogenically cleaned

of contaminating gas phases, and stored in flame-

sealed, 5-mm diameter glass tubes. Standardization

of the method was achieved with KHCO3, with a

isotopic composition of �23.91% when measured

as a solid. The KHCO3 was dissolved into two

solutions with concentrations of 400 and 750 mg/gKHCO3, which was then extracted following the

procedures for the pore waters. A standard was

extracted following every fifth sample and the

average value measured was �23.37 ± 0.20%(1s). Sample data have been corrected for the

difference between the dissolved and solid isotope

ratios of the KHCO3. All gas was measured auto-

matically with a VG Prism II mass spectrometer

following introduction to the instrument by an

automated cracker system.

4.4. Diagenesis

[12] Diagenesis refers to post-depositional altera-

tions that, in this case, might change the isotopic

signature of foraminiferal carbonate. Upward

advection of methane-enriched fluids in seeps can

produce bicarbonate, causing the precipitation of

authigenic carbonates. These carbonates have very

negative d13C values (�48% to �52% in the Clam

Flats area [Stakes et al., 1999]), and any contam-

ination or overgrowth on biogenic carbonate would

alter isotopic signatures toward more negative d13Cvalues [e.g., Torres et al., 2003]. Since specimens

are cleaned of debris before analyses, and since

overgrowths of authigenic carbonate are not likely

on the skeletal carbonate of living individuals

(hence the term ‘‘test’’ rather than ‘‘shell’’), isotope

values of living foraminiferal carbonate are not

affected by inorganic carbonate or diagenesis.

4.5. Rose Bengal Limitations

[13] In order to distinguish living individuals from

dead foraminifera, it is necessary to use a vital stain

that will highlight tests containing living proto-

plasm. Rose Bengal is a protein stain that remains

the vital stain of choice for most community-scale

studies attempting to distinguish assemblages that

were living or recently living at the time of

collection (see Gooday and Rathburn [1999] and

Bernhard [2000] for discussions). However, since

protoplasm can remain in the test after death of the

individual, rose Bengal staining has limitations as

an indicator of living foraminifera [Bernhard,

2000]. Nevertheless, as noted by Gooday and

Rathburn [1999] and others, the fact that seasonal

patterns in foraminiferal assemblages are discern-

able, even in oxygen-poor settings, suggests that

protoplasm decays relatively quickly in deep-water

environments. Using a conservative approach

when distinguishing between stained and unstained

individuals seems to yield reasonable and adequate

resolution of deep-sea foraminiferal populations

[Lutze and Altenbach, 1991]. Use of a wide array

of vital techniques (ATP assay, ultrastructural

analyses and rose Bengal staining) by Bernhard



6 of 28

Table 2. Stable Isotope Values of Living (Stained) Foraminifera

Sample Site Sediment Depth, cm d13C, %, PDB d18O, %, PDBSpecies

(Number of Specimens Run)

Clam bed 0–1 �0.85 2.83 Uvigerina peregrina (1)Dive 1780 �0.54 3.01 Uvigerina peregrina (1)Tube core 30 �0.32 3.19 Uvigerina peregrina (1)

�0.71 3.06 Uvigerina peregrina (1)�0.92 3.28 Globobulimina pacifica (1)�1.36 3.28 Globobulimina pacifica (1)�0.74 3.30 Globobulimina pacifica (1)�1.46 3.28 Globobulimina pacifica (1)�0.79 3.09 Epistominella pacifica (4)�0.73 3.10 Epistominella pacifica (3)�0.82 3.26 Bulimina mexicana (3)�0.43 3.01 Uvigerina peregrina (1)�0.76 3.16 Uvigerina peregrina (1)

1–1.5 �0.57 3.25 Uvigerina peregrina (1)�1.29 3.45 Globobulimina pacifica (1)�1.84 3.33 Globobulimina pacifica (2)�1.43 3.25 Globobulimina pacifica (1)�0.91 3.49 Globobulimina pacifica (1)�0.78 3.30 Epistominella pacifica (3)�0.53 3.11 Epistominella pacifica (4)�0.70 3.26 Bulimina mexicana (2)

1.5–2 �0.56 3.09 Uvigerina peregrina (1)�0.51 3.03 Uvigerina peregrina (1)�0.49 3.07 Uvigerina peregrina (1)�0.50 3.22 Epistominella pacifica (3)�0.72 3.23 Epistominella pacifica (4)�2.14 3.39 Globobulimina pacifica (2)�2.17 3.35 Globobulimina pacifica (2)�0.72 3.29 Bulimina mexicana (2)

2–2.5 �0.47 3.20 Uvigerina peregrina (1)�0.50 3.01 Uvigerina peregrina (1)�0.73 3.14 Epistominella pacifica (3)�0.63 3.17 Epistominella pacifica (2)�0.59 3.15 Epistominella pacifica (2)�1.98 3.33 Globobulimina pacifica (2)�1.15 3.38 Globobulimina pacifica (1)�0.92 3.34 Globobulimina pacifica (1)�0.41 3.38 Globobulimina pacifica (1)�0.80 3.30 Bulimina mexicana (3)0.04 2.80 Planulina sp. (1)

2.5–3 �0.57 3.23 Uvigerina peregrina (1)�0.77 3.23 Uvigerina peregrina (1)�0.27 3.11 Uvigerina peregrina (1)�0.53 3.15 Epistominella pacifica (2)�0.63 3.29 Epistominella pacifica (2)�0.60 3.21 Epistominella pacifica (2)�1.03 2.29 Epistominella pacifica (1)�1.20 3.33 Globobulimina pacifica (1)�1.68 3.41 Globobulimina pacifica (1)�2.23 3.27 Globobulimina pacifica (2)�0.70 3.29 Bulimina mexicana (3)0.09 2.95 Planulina sp. (1)�0.90 3.29 Globobulimina affinis (1)

3–4 �0.59 3.15 Epistominella smithi (2)�0.76 3.19 Epistominella smithi (4)�0.85 3.28 Bulimina mexicana (4)�1.06 3.28 Globobulimina pacifica (1)�0.91 3.29 Globobulimina pacifica (2)�1.72 3.25 Globobulimina pacifica (3)�1.22 3.27 Globobulimina pacifica (2)



7 of 28

Table 2. (continued)



�0.33 2.97 Uvigerina peregrina (1)�0.78 2.87 Uvigerina peregrina (1)�0.36 3.26 Uvigerina peregrina (1)�0.59 2.98 Uvigerina peregrina (1)

4–5 �0.58 3.06 Epistominella smithi (3)�0.43 3.16 Epistominella smithi (2)�0.70 3.20 Epistominella smithi (3)�0.48 3.16 Epistominella smithi (2)�0.76 3.07 Epistominella smithi (4)�0.77 3.06 Epistominella smithi (6)�0.92 3.36 Bulimina mexicana (3)�0.81 3.20 Bulimina mexicana (4)�0.89 3.28 Bulimina mexicana (3)�0.93 3.30 Globobulimina pacifica (2)�1.81 3.23 Globobulimina pacifica (1)�1.34 3.26 Globobulimina pacifica (2)�1.47 3.33 Globobulimina pacifica (3)�0.59 2.59 Uvigerina peregrina (1)�0.04 3.11 Uvigerina peregrina (1)�0.45 2.91 Uvigerina peregrina (1)�0.43 3.14 Uvigerina peregrina (1)�0.36 3.40 Uvigerina peregrina (1)�0.37 3.15 Uvigerina peregrina (1)�0.36 3.10 Uvigerina peregrina (1)�0.49 2.66 Uvigerina peregrina (1)

Bacterial mat 0–1 �0.72 3.31 Epistominella smithi (4)Dive 1780 �1.22 3.34 Globobulimina pacifica (3)Tube core 67 1–1.5 �0.61 3.13 Uvigerina peregrina (3)

1.5–2 �0.80 3.16 Uvigerina peregrina (1)�1.12 3.19 Globobulimina pacifica (2)�1.50 3.25 Globobulimina pacifica (2)

2–2.5 �0.59 3.21 Epistominella pacifica (4)�1.04 3.14 Uvigerina peregrina (1)

2.5–3 �0.62 3.19 Epistominella pacifica (2)Clam bed 0–1 �0.70 3.15 Gyroidinoides altiformis (2)Dive 1781 �1.36 3.09 Uvigerina peregrina (2)Tube core 31 �1.18 4.04 Uvigerina peregrina (2)

�1.03 3.80 Bulimina mexicana (5)1–1.5 �0.89 3.11 Uvigerina peregrina (1)

�0.62 3.21 Uvigerina peregrina (1)�0.43 3.12 Uvigerina peregrina (1)�0.69 3.12 Uvigerina peregrina (1)�3.87 3.42 Globobulimina pacifica (1)�0.22 2.63 Planulina sp. (1)

1.5–2 �0.55 3.16 Uvigerina peregrina (0.5)�0.49 3.28 Uvigerina peregrina (1)�0.74 3.21 Uvigerina peregrina (1)�0.22 3.25 Uvigerina peregrina (1)�0.21 3.21 Uvigerina peregrina (0.5)�0.57 3.15 Uvigerina peregrina (1)�0.97 3.10 Uvigerina peregrina (0.5)�0.83 3.14 Gyroidinoides altiformis (2)�0.66 3.11 Gyroidinoides altiformis (1)

2–2.5 �2.05 3.04 Uvigerina peregrina (1)�1.13 3.17 Uvigerina peregrina (1)�0.81 3.21 Uvigerina peregrina (0.3)�1.14 3.07 Uvigerina peregrina (0.3)�0.94 3.12 Uvigerina peregrina (0.3)�0.10 3.43 Uvigerina peregrina (0.5)�0.92 3.57 Epistominella smithi (3)



8 of 28

et al. [2001] previously confirmed that foraminif-

era are alive in Monterey Bay seep habitats.

Microscopic examination of specimens counted

as living were in pristine condition, and no authi-

genic carbonate was observed on them. We recog-

nize that some cytoplasm-containing individuals in

our study may have been recently alive rather than

alive at the time of collection. However, we are

confident that our use of a conservative approach

that is consistent with our assessments in previous

studies yielded a reasonably accurate assessment of

the living foraminiferal population at seeps. Speci-

mens chosen for isotopic analyses were very care-

fully scrutinized and selected from the pool of

living (stained) individuals. Since we are using

rose Bengal staining as a proxy for the living

population, we refer to stained individuals in the

text as ‘‘living’’.

5. Results

5.1. Pore Water Geochemistry

[14] In addition to visual differences in seep ap-

pearance and activity in the two sites visited in

Monterey Bay, interstitial water geochemistry also




�3.49 3.30 Globobulimina pacifica (1)�0.92 3.15 Epistominella pacifica (2)

2.5–3 �1.82 3.18 Uvigerina peregrina (1)�1.12 3.15 Uvigerina peregrina (1)�0.25 3.21 Uvigerina peregrina (1)�0.75 3.25 Uvigerina peregrina (0.5)�1.39 3.27 Uvigerina peregrina (0.5)�0.60 3.11 Epistominella smithi 5�0.66 3.11 Gyroidinoides altiformis (2)�0.45 3.13 Gyroidinoides altiformis (1)�0.99 4.56 Epistominella pacifica (1)�1.02 4.63 Epistominella pacifica (1)�1.01 4.66 Epistominella pacifica (1)0.06 2.60 Planulina sp. (0.5)0.45 2.86 Planulina sp. (1)0.07 2.60 Planulina sp. (0.5)

3–4 �1.17 3.15 Uvigerina peregrina (1)�0.83 3.22 Uvigerina peregrina (1)�1.82 3.09 Uvigerina peregrina (1)�1.30 3.14 Uvigerina peregrina (0.5)�0.72 3.16 Uvigerina peregrina (0.5)�1.04 3.12 Uvigerina peregrina (0.5)�0.85 3.06 Gyroidinoides altiformis (1)�1.06 3.30 Bulimina mexicana (4)

4–5 �0.63 3.22 Epistominella smithi (4)�0.67 3.12 Gyroidinoides neosoldani (2)�0.57 3.28 Gyroidinoides neosoldani (1)�0.49 3.40 Gyroidinoides neosoldani (1)�5.36 4.52 Globobulimina barbata (2)�2.77 4.47 Globobulimina barbata (2)�4.56 3.20 Globobulimina pacifica (2)�0.95 3.14 Uvigerina peregrina (2)�1.53 3.14 Uvigerina peregrina (2)�0.78 3.07 Uvigerina peregrina (2)�0.78 3.28 Uvigerina peregrina (1)�1.05 3.23 Uvigerina peregrina (1)�0.86 3.23 Uvigerina peregrina (1)�0.51 3.13 Uvigerina peregrina (1)�1.17 3.22 Bulimina mexicana (6)�2.76 4.40 Buliminella tenuata (3)



9 of 28

indicates that geochemical conditions between

Extrovert Cliffs and Clam Flats sites are distinctly

different. At Site 1780 (Figure 2), the bacterial mat

and clam bed ‘‘rings’’ observed at the sediment-

water interface within the ‘‘seep ring’’ have differ-

ent, but associated, geochemical characteristics.

Interstitial water geochemistry profiles from Extro-

vert Cliffs are presented in Figures 3a and 3b.

Concentrations of chloride, calcium, magnesium,

and lithium are nearly constant to within 1 cm from

the sediment water interface in the bacterial mat core

(TC 34), indicating that fluids must flow vigorously

upward in a very restricted area underneath the gray

bacterial mat (Figure 3b). Elevated concentrations

of chloride, calcium, and lithium suggest a deep

origin of these fluids, though no a priori estimate can

bemade of the depth of origin. Geochemical profiles

of the yellow bacterial mat core (TC 31) indicate

little or no advective flow of fluids, but concentra-

tion gradients suggest diffusive flow involving

fluids of a similar composition as those in the gray

bacterial mat. Data from the clam bed core (TC 79)

and an apparently ‘‘nonseep’’ environment (refer-

ence) core outside the clam bed (TC 38) show that

concentrations are quite similar to those of seawater.

Profiles of sulfate and alkalinity (Figure 3a) show

strong similarities to those of chloride, lithium and

calcium. Dissolved inorganic carbon d13C ratios

show decreases typically observed in organic car-

bon-rich sediments (>�23%).

[15] On the basis of geochemical profiles from the

clam bed (TC 80) and bacterial mat (TC 28) at

Site 1781, upward flow of fluids is less important

in this site, when compared with 1780 (Figures 3c

Table 3. Carbon Isotope Values of Seep Pore Waters

Dive Site 1780Habitat and Core

SedimentDepth,cm

d13C,%, PDB

Gray Bact. Mat 0Tube Core 34 0.5 �12.8

2 �13.34 �15.16 �168 �18.710 �2112 �21.2

Clam Bed 0Tube Core 79 0.5 �4.8

2 �4.14 �4.46 �5.78 �6.510 �7.512 �8.6

Yellow Bact. Mat. 0 �3.7Tube Core 31 0.5 �7.6

2 �15.24 �18.66 �21.18 �22.810 �22.712 �22

Reference Core 0Tube Core 38 0.5 �4.6

2 �3.64 �4.46 �4.58 �4.510 �4.712 �4.91416 �6.118 �6.1

Clam Bed 0Tube Core 80 0.5 �36.5

2 �45.746810 �41.5121416 �42.4

Reference Core 0Tube Core 72 0.5

2 �8.24 �5.86 �7.111 �11.216 �13.221 �11

Reference Core 0 �3.9Tube Core 38 0.5 �5.5

2 �6.84 �7610


Dive Site 1780Habitat and Core

SedimentDepth,cm

d13C,%, PDB

1214 �21.6

Bact. Mat 0Tube Core 28 0.5 �13.8

24 �58.26 �56.8810 �49.512 �47.614 �47.1



10 of 28

Figure 3. Graphs of pore water dissolved sulfate, hydrogen sulfide, alkalinity, and the d13C of dissolved carbondioxide are shown for two seep sites Monterey Bay (Site 1780 - Extrovert Cliffs; Site 1781 - Clam Flats). Numbersrepresent core numbers sampled in different habitats. For 1780: 34 = gray bacterial mat; 31 = yellow bacterial mat;79 = clam bed; 38 = reference habitat 3 meters from clam bed. For 1781: 28 = bacterial mat; 72 = nonseep habitat4 meters away from bacterial mat; 80 = clam bed; 38 = reference habitat 5 meters from clam bed.



11 of 28

and 3d). Only chloride concentrations are slightly

elevated in Cores 28 and 80, but much less than was

observed in the gray bacterial mat core (TC 34) at

Site 1780. Profiles of calcium and magnesium are

typical for many seep sites, with decreases in

calcium and very small decreases in magnesium

(Figure 3d). In addition, lithium profiles indicate

decreases in concentration especially in the cores

that show sulfate reduction.

[16] Of special interest in this study are the changes

in sulfate, sulfide, alkalinity, and d13C of dissolved

inorganic carbon (DIC) (summarized in Figures 3a

and 3c). In the Extrovert Cliffs area the sulfate

concentrations in the gray bacterial mat core

(TC 34) showed a sharp decrease with depth

similar to that observed in other constituents. In

the yellow bacterial mat core (TC 31), sulfate

decreases linearly with depth, again similar to

gradients in the constituents shown in Figure 3b.

On the other hand, changes in sulfate are minimal

in the clam bed core (TC 79) as well as in the

reference core (TC 38). In contrast, in the Clam

Flats clam bed core (TC 80) sulfate concentrations

rapidly decrease in the upper 4–5 cm associated

with the production of large amounts of sulfide

(Figure 3c), indicating strong sulfate reduction in

the upper 4 cm. This phenomenon is typical for a

case in which advection of methane gas occurs

close to the surface of the sediments.

[17] Increases in alkalinity are typically associated

with the sulfate reduction process and the subse-

quent generation of sulfide (Figures 3a and 3c).

Depending on the source of organic carbon, a

distinct signature in the d13CDIC is expected. In

the Extrovert Cliffs area d13C values do not drop

below �23%, whereas in the Clam Flats area

values of d13C below 4 cm are typically between

�45 and �58%. An important question arises

from the Site 1780 d13C data. If indeed the origins

of the fluids in the seep ring at Site 1780 are from

greater depths, then higher values of d13C would be

expected in accordance with previous observations

[e.g., Doose, 1980], in which d13C values will

become heavier with depth. On the other hand it

is possible that the sources of organic carbon,

which support the sulfate reduction processes, are

entirely different; mostly marine organic carbon in

the Extrovert Cliffs area [Doose and Kaplan, 1981;

Martin et al., 1997], but mostly methane gas in the

Clam Flats area [Martin et al., 1997]. Dissolved

ammonium data (not shown) from Site 1780 (cores

TC31 and TC34) show values as high as 4–5

millimolar, supporting a deeper origin of the

1780 fluids and the involvement of alteration of

marine organic matter. In contrast, at the Clam

Flats Site (1781) ammonium concentrations are

generally well below 500 micromolar. Site 1781

cores show very rapid decreases within the upper 1

to 2 centimeters of the cores, much larger in

magnitude when compared to cores of Site 1780.

5.2. Oxygen and Hydrogen SulfideMicroprofiles

5.2.1. Site 1780

[18] Microprofiles measured in four intact sediment

cores that were retrieved from the central gray

bacterial mat, the yellow bacterial mat, the sur-

rounding ring of densely accumulated clams and

from an apparently nonseep environment nearby

showed distinctive differences in oxygen penetra-

tion and sulfide concentrations (Figures 4a and 4b

and Figures 5a and 5b). Oxygen was already

consumed at the sediment-water interface in the

gray bacterial mat and penetrated only 1-mm deep

into the sediment that was covered by a yellow

bacterial mat. Whereas oxygen penetrated 2 mm

into the sediment inhabited by clams and more than

4 mm deep into the sediment of the reference core,

where bioturbation of irregular sea urchins results

in the intrusion of oxygen-rich water to even

deeper sediment layers (>20 mm) (Figure 5b). In

the gray bacterial mat hydrogen sulfide reached the

sediment-water interface and increased rapidly

with sediment depth, reaching an extremely high

maximum of 17 mM at only 7-mm sediment depth.

The profile also suggests a flux of sulfide from the

sediment to the overlying water. In the yellow

bacterial mat hydrogen sulfide was detectable at

1 cm sediment depth and increased linearly with

depth, reaching a value of 7 mM at 6-cm sediment

depth (Figure 4a). Both profiles indicate high

sulfide production or could imply an upward

transport of sulfide-rich pore water from deeper

sediment layers. In contrast, hydrogen sulfide con-



12 of 28

centrations were low in the clam bed sediments and

not detectable in the reference core (Figure 4b).

The hydrogen sulfide profile in the clam bed

showed a peak between 25 and 35 mm sediment

depth with a maximum value of 0.2 mM sug-

gesting very local sulfide production by sulfate

reduction.

5.2.2. Site 1781

[19] Oxygen and sulfide microprofiles measured in

cores retrieved from the bacterial mat showed

differences to the profiles measured in cores from

the clam bed patch and the reference environment

at site 1781 (Figures 4c and 4d and Figures 5c

and 5d). Oxygen was consumed within the upper

4 mm of a conspicuously dense, 6-mm thick

bacterial mat that covered the sediment surface of

the bacterial mat core (Figure 5c). Oxygen pene-

tration was similar in the clam bed and the refer-

ence core, reaching a depth of 3 mm (Figure 5d). A

high-resolution hydrogen sulfide profile measured

through the bacterial mat showed a local peak in

sulfide concentration just above the sediment-water

interface. Interestingly, both sulfide and oxygen

were present in the microbial mat providing the

sulfide-oxidizing bacteria simultaneously with

the required electron acceptor and donor. Below

the peak at the sediment interface sulfide was not

detectable down to 20 mm sediment depth where it

increased with increasing sediment depth reaching

1.5 mM at 50 mm (Figure 4d). In the clam bed

sulfide concentrations were <0.2 mM in the upper

20 mm and could not be followed to deeper depths

due to the breaking of electrodes. Sulfide was not

detectable in the reference core down to a depth of

40 mm (Figure 4d).

5.3. Living (Stained) Foraminifera

[20] Monterey Bay seep foraminiferal densities

(1382/50 cc in TC 31 of Site 1781; 703/50 cc in

TC 30 of 1780; 282/50 cc in TC 67 of Site 1780)

are within the range of those found in previous

studies of nonseep environments along the Pacific

margin at similar depths (see discussions of

Rathburn et al. [2000] and Bernhard et al.

[2001]). We found no endemic calcareous species

of benthic foraminifera at Monterey Bay seeps;

calcareous foraminiferal species from these seeps

are all found elsewhere in nonseep environments.

These results are consistent with those of previ-

ously studied seeps including the northern Califor-

nian Margin [Rathburn et al., 2000], the Gulf

of Mexico [Sen Gupta et al., 1997], Japanese

seeps [Akimoto et al., 1994], and Monterey Bay

[Bernhard et al., 2001].

[21] Vertical distribution patterns of foraminifera

varied between seep environments. Total forami-

niferal densities were higher in clam beds than in

the bacterial mat, with maximum densities in both

habitats occurring below 1 cm, typically between

2–4 cm (Figures 6a, 6b, and 7). Distributions of

some benthic foraminiferal species from seep clam

beds show appreciable infaunal densities within

the sediments from 1.5 cm to 4 cm (Figures 6a,

6b, and 7). In the Clam Flats clam bed (TC31)

several species, including U. peregrina, increased

in density in the 1.5 to 2 cm interval, with notable

densities down to at least 4 cm (Figure 7). In the

clam ring clam bed (core TC30) Bulimina

mexicana and Epistominella pacifica increased

appreciably in density at the 1.5–2 cm interval,

and had relatively high densities from 3 to 5 cm

(Figure 6a). Globobulimina pacifica reached a

maximum density in the 2.5 to 3 cm interval, while

U. peregrina densities very gradually increased to

4 cm (Figure 6a). On the basis of nonseep distri-

bution patterns, it is very unusual that U. peregrina

had a density maximum deeper than G. pacifica

in the same core [e.g., McCorkle et al., 1990;

Rathburn and Corliss, 1994]. In the clam ring gray

bacterial mat (TC 67), total densities were lower

than in the clam beds (Figures 6a and 6b). Bacterial

mat densities of U. peregrina and B. mexicana

remained relatively stable down to 2.5 cm, while

G. pacifica had a maximum density at 2–2.5 cm,

and E. pacifica had a density maximum at 2.5–

3 cm (Figure 6b).

5.4. Stable Isotopes of Living Foraminifera

[22] Isotopic signatures of living (rose Bengal

stained) foraminifera from seep environments

show a wide range of variability within and

between seeps (Figures 6a–10). Note that these



13 of 28

Figure 4. Hydrogen-sulfide microprofiles measured immediately after retrieval in cores that were taken in thedifferent zones at site 1780 - Extrovert Cliffs (A and B) and 1781 - Clam Flats (C and D). (A): Comparing profilesmeasured in the 1780 bacterial mat cores, core # 67 (gray bacterial mat), core # 29 (yellow bacterial mat).(B): Comparing profiles measured in 1780 cores # 30 (clam bed) and # 71 (reference environment 3 meters from theclam bed). (C): High-resolution profile through the 1781 bacterial mat covering core # 30, showing a peak in sulfideconcentration just above the sediment surface. (D): Comparing 1781 profiles measured in core # 30 (bacterial mat),core # 31 (clam-bed patch), and core # 71 (reference environment 5 meters from the clam bed). Note the differentscales.



14 of 28

isotope values are plotted as d13C values, not Del

d13C values (Del d13C = the difference between d13Cvalues bottom-water DIC and those of biogenic

carbonate). We chose not to present Del d13C values

as plotted by McCorkle et al. [1997] and Rathburn

et al. [1996] because the bottom water values near

the seeps may be affected by methane seepage.

[23] At Dive Site 1780 (Extrovert Cliffs), carbon

isotope values of conspecifics were similar for both

clam beds and bacterial mats (Figures 6a, 6b, and 8),

and showed less variability compared to those from

Dive Site 1781 (Figures 8 and 10). At the Clam

Flats site (1781) d13C values of foraminifera are

more variable (Figures 7, 8, and 9) and have, on

average, lower d13C values than conspecifics in

nonseep environments (Figure 10). Figure 6a

shows distributions and carbon isotope values of

abundant taxa at Dive Site 1780 (Extrovert Cliffs).

Except for the deep infaunal, G. pacifica, isotope

values of foraminiferal species from this habitat

have d13C ranges similar to those of nonseep

habitats. In the Gray bacterial mat habitat at

Site 1780 (Figure 6b), few data are available, and

Figure 5. Oxygen microprofiles measured immediately after retrieval in cores that were taken in the different zonesof site 1780 - Extrovert Cliffs (A and B), (A): Comparing oxygen profiles measured in 1780 core # 67 (gray bacterialmat), core # 29 (yellow bacterial mat), core # 30 (clam bed), and core # 71 (reference environment 3 meters from theclam bed). (B): The core from the reference environment was bioturbated by irregular sea urchins resulting in theintrusion of oxygen to deeper sediment layers, as evidenced by a separate penetration with an oxygen probe. Oxygenmicroprofiles measured immediately after retrieval in cores that were taken at site 1781 - Clam Flats (C and D).(C): Oxygen microprofiles measured through the 6-mm thick bacterial mat that covered core # 30. (D): Comparingmicroprofiles measured in cores # 31 (clam bed) and # 71 (reference environment 5 meters from the clam bed).



15 of 28

d13C values of the three plotted species are within

the range of conspecifics in the clam bed habitat. At

site 1781, isotopic ranges for G. pacifica and

U. peregrina in the clam bed habitat (Figure 7) are

much greater than those of the same species at

Site 1780. The few data available for B. mexicana,

G. altiformis, and E. pacifica suggest relatively low

isotopic variability. Figure 8 summarizes the carbon

isotopic data for Sites 1780 and 1781. In each

habitat, there are differences in isotopic composi-

tions between species. As in nonseep habitats, for

each seep habitat, the deep infaunal G. pacifica

has the lowest d13C values, compared to the

shallow infaunals (U. peregrina, B. mexicana, and

G. altiformis) and transitional taxa (E. pacifica),

with epifaunals (Planulina sp.) having the highest

value. On average, d13C values for species with

sufficient data (G. pacifica and U. peregrina) are

greater than those of published nonseep data

(Figure 9). Figure 10 shows the differences in

ranges and variability of foraminifera living in

different seep environments. The range and vari-

ability of d13C values in a species is most pro-

nounced in G. pacifica, a deep infaunal species,

with a range of �0.4 to �4.5% (Figure 10;

Table 2). In previous studies of isotopic values of

living foraminifera, d13C values as low as

�3.77% have been reported for deep infaunal

Globobulimina species [McCorkle et al., 1997].

The very low values for some G. pacifica speci-

mens at 1781 extend the range of reported values

for Globobulimina, although the d13C values of

many specimens are within the range reported for

this species.

Figure 6a. Vertical sediment distribution patterns (densities) and stable carbon isotope values of selectedforaminifera from the clam bed habitat of Extrovert Cliffs, Site 1780 (core TC 30). Note the change in horizontalscales and note also that densities are plotted at the deepest depth of the sampled interval (Example: densities of the 0to 1-cm interval are plotted at 1-cm) to facilitate comparisons with previous work [e.g., Corliss and Emerson, 1990;Rathburn et al., 2000]. Closed symbols represent number of specimens/50cc; open symbols represent d13C values.



16 of 28

[24] Oxygen isotope values of a given living seep

species do not vary appreciably; typically values are

within 0.3% [e.g., McCorkle et al., 1990, 1997;

Corliss et al., 2002]. This is to be expected if pore

water d18O values and temperatures in seeps are

similar to those of overlying bottom waters. Authi-

genic carbonate d18O values in this area tend to range

from 2–5% (J. Gieskes, unpublished data, 2003).

6. Discussion

6.1. Living Foraminifera

[25] As a result of the ephemeral and ubiquitous

nature of seepage in the area, it is unclear whether

total population densities of foraminifera are in-

fluenced by seep habitats. On the basis of avail-

able data, it appears that methane seeps do not

appreciably enhance or reduce total foraminiferal

densities in productive bathyal environments [see

Rathburn et al., 2000; Bernhard et al., 2001]. It is

not unusual, and perhaps typical in some nonseep

environments, that living deep-sea foraminifera

are found well below the sediment depth of

oxygenated pore waters (compare foraminiferal

distributions in Figures 6a, 6b, and 7 with oxygen

profiles in Figure 5), prompting some researchers

to suggest that at least some species are faculta-

tive anaerobes [Bernhard, 1996]. Oxygen may not

be limiting for species that are able to live deep

within the sediments [e.g., Fontanier et al., 2002],

and these infaunal taxa may track and feed

selectively on bacteria or bacterial degradation

byproducts [Jorissen et al., 1998]. In regions

Figure 6b. Vertical sediment distribution patterns (densities) and stable carbon isotope values of selectedforaminifera from the gray bacterial mat habitat of Extrovert Cliffs, Site 1780 (core TC 67). Only the top 3 cm werepicked for TC 67. Note the change in scales and note also that densities are plotted at the deepest depth of the sampledinterval (Example: densities of the 0 to 1-cm interval are plotted at 1-cm) for ease of comparisons with previous work[e.g., Corliss and Emerson, 1990; Rathburn et al., 2000]. Closed symbols represent number of specimens/50cc; opensymbols represent d13C values.



17 of 28

Figure 7. Vertical sediment distribution patterns (densities) and stable carbon isotope values of selectedforaminifera from the clam bed habitat of Clam Flats, Site 1781 (core TC 31). Note the change in horizontal scalesand note also that densities are plotted at the deepest depth of the sampled interval (Example: densities of the 0 to1-cm interval are plotted at 1-cm) to facilitate comparisons with previous work. Closed symbols represent number ofspecimens/50 cc; open symbols represent d13C values.



18 of 28

and water depths where food availability is more

limiting, seeps may act more like oases of pro-

ductivity, as seems to be the case with infaunal

macrofauna [Levin et al., 2000, 2003]. Calcareous

foraminiferal assemblages living in seeps located

in low productivity areas have not yet been

studied.

[26] On the basis of currently available data,

foraminiferal assemblages at seeps are not unique.

Although some taxa appear more adapted to

conditions at seeps, it is not yet possible to distin-

guish seep assemblages from nonseep assemb-

lages. Additional seep/nonseep comparisons and

information about sulfide and methane tolerances

of species are needed in order to evaluate any

distinguishing differences between seep and non-

seep assemblages.

[27] Benthic foraminifera are mobile, and can ad-

just distributions within the sediments in response

to environmental, and perhaps, biological cues [see

Gooday and Rathburn, 1999]. As a result, forami-

niferal distributions within sediment profiles are

thought to reflect species-specific ecological

preferences and responses to ambient conditions.

Under otherwise uniform conditions, foraminiferal

species distributions are expected to be predictable

Figure 8. Carbon isotope values of abundant foraminifera for clam bed habitats at sites 1780 and 1781 (top row),and the gray bacterial mat of site 1780 (bottom row). Isotope values are plotted versus sediment depth for each core.Note the change in horizontal scales. Densities are plotted at the deepest depth of the sampled interval (Example:densities of the 0 to 1-cm interval are plotted at 1-cm) to facilitate comparisons with previous work.



19 of 28

primarily on the basis of species ecology, geo-

chemical gradients and food availability [e.g.,

Fontanier et al., 2002]. In nonseep environments,

density (abundance per 50 cc) profiles of total

foraminiferal populations typically, though not

always, reach a maximum within 1 cm of

the sediment-water interface [e.g., Corliss and

Emerson, 1990; Jorissen et al., 1995; Rathburn

et al., 1996]. Variable, and unusual species distri-

bution patterns within clam bed sediments off

northern California led Rathburn et al. [2000] to

suggest that foraminifera were responding to var-

iations in the unusual pore water environments of

these methane seepage areas. The influence of seep

pore water conditions in Monterey Bay seeps may

also explain infaunal density maxima of abundant

species at 2–4 cm within the 1781 clam bed core

and downcore increases in the densities of some

species in the 1780 clam bed core (Figures 6a

and 7). The downcore distributions of several

seep species are quite different from those of

conspecifics observed in previous studies of

nonseep habitats, with broader and deeper depth

ranges of some seep species, such as U. peregrina.

The reasons for these differences in vertical distri-

bution patterns are unknown, but may result from

deeper, more wide-spread microbial food sources

at seeps compared to nonseep habitats. Some taxa

may have higher tolerances for the potentially

stressful conditions at seeps. For example, Akimoto

et al. [1994] suggested that Rutherfordoides

cornuta prefers habitats with high methane gas

content, whereas the distribution of Bulimina

striata at seeps is affected by hydrogen sulfide

gas content. The presence of living foraminifera

exposed to H2S concentrations exceeding 16 mM

in this study (1781 clam bed; Figures 3, 4, and 7)

indicates a high tolerance for sulfide in some spe-

cies, including B. mexicana, E. smithi, G. pacifica,

Globobulimina barbata, Gyroidinoides altiformis,

and U. peregrina. Down-core increases in these

species (unusual, on the basis of published nonseep

distributions [e.g., McCorkle et al., 1997]) suggest

that these taxa are able to thrive in sulfide-rich

Figure 9. Box and whisker plots of carbon isotope data for Globobulimina pacifica and Uvigerina peregrina fromMonterey Bay seeps (1781 = Clam Flats and 1780 = Extrovert Cliffs) and published nonseep habitats (data fromseveral cores reported in McCorkle et al. [1997]). The horizontal line in each box represents the median; the upperand lower lines of the box are the quartiles. The bars outside the box extend to the extreme values. The numbersunderneath the boxes indicate the number of isotope values considered in the statistical analysis, not the number oftests analyzed (in Monterey Bay samples, each value usually represents 1 foraminiferan (see Table 2); in McCorkle etal. [1997], 2–4 specimens/run).



20 of 28

(and perhaps methane-rich) environments. Since

these species occur in nonseep environments, it is

unclear whether high sulfide tolerance is a charac-

teristic of all populations or prevalent just in

populations exposed to high sulfide concentrations,

such as those in seeps and organic-rich, oxygen-

poor environments (see discussion by Bernhard et

al. [2001]).

[28] Vertical distribution patterns (density profiles)

of foraminiferal species do not necessarily coincide

with variations in pore water chemistry or test

carbon isotope excursions. In other words, maxi-

mum densities of infaunal species’ populations do

not necessarily occur at sediment depths where

variations in pore water geochemical profiles are

evident. Likewise, depth intervals showing maxi-

mum variability of foraminiferal carbon isotope

values do not necessarily occur at the same depth

interval where maximum or minimum species

densities occur (Figures 6a, 6b, and 7). Neverthe-

less, it appears that both the distribution patterns

and d13C values (discussed below) of calcareous

benthic foraminifera are influenced by methane

seepage.

6.2. Stable Isotopes of Living ForaminiferalCarbonate

[29] There is an appreciable difference between

foraminiferal carbon isotope patterns at Site 1780

compared to 1781. Carbon isotope values of

Figure 10. Average d13C values and standard deviations for selected taxa from clam bed and bacterial mat habitats inthe study area. Note the change in scale for G. pacifica from site 1781. As in the case of nonseep environments,foraminiferal d13C values are different for different species, but for some species, the variability within and betweenspecies and nearby sites is far greater than nonseep environments. The clam bed at site 1780 does not show the widevariation in U. peregrina values observed in the other environments, while the deep infaunal taxon, G. pacifica, haslarge variation in the clam beds but not in the shallow habitats of the bacterial mat. These isotope values are plotted asd13C values, not Del d13C values as plotted by McCorkle et al. [1997] and Rathburn et al. [1996] (Del d13C = thedifference between d13C values bottom-water DIC and those of biogenic carbonate). Isotopic ranges of published d13Cvalues of U. peregrina and G. pacifica from nonseep environments are given in Figure 9. See text for discussion.



21 of 28

many conspecifics are distinctly more negative at

Site 1781 compared to 1780. For example,

U. peregrina from the 1780 clam bed had average

d13C values from sample intervals ranging from

about �0.4% to �0.6% (Figure 10; Table 2).

About 8 kilometers away in similar water depths,

living (stained) U. peregrina from the 1781 clam

bed had average d13C values of about �0.5% to

�1.3%. Similarly, living Globobulimina pacifica

from the same clam beds had average d13C values

of between about �1.1% and �1.7% at 1780 and

about �3.5% and �4.0% at 1781. At site 1781,

the intraspecific variability of d13C values in living

species is much greater than for those of published

conspecifics living in nonseep habitats. For exam-

ple, in one clam bed (Dive 1781; TC 31) carbon

isotopic values of Uvigerina peregrina ranged

from �0.10% to �2.05% for individuals living

at the same sediment depth interval (Table 2). In

other clam beds, one from about 8 kilometers

away in Monterey Bay, and two others on the

margin off Eel River, about 500 kilometers to the

north, d13C values of living U. peregrina vary with

ranges of �0.85% to �0.04% and �0.51% to

�1.53% respectively (Table 2) [Rathburn et al.,

2000]. In nonseep environments, d13C values of

U. peregrina do not typically vary by more than

0.2 to 0.4% (Figure 9) (McCorkle et al. [1997],

Corliss et al. [2002], and discussion by Rathburn

et al. [2000]). These very large differences in

isotopic values are unheard of when comparing

conspecifics from adjacent nonseep environments

[e.g., McCorkle et al., 1997; Rathburn et al., 1996;

Corliss et al., 2002]. The most likely explanation

for these differences is the negative isotopic

influence of methane.

[30] One curious similarity between seep and non-

seep foraminiferal isotope patterns is the relative

difference between infaunal taxa and those living

in shallower sediment habitats. It is not uncommon

to have a d13C value range of about 4% between

living species within a nonseep assemblage

[McCorkle et al., 1990, 1997; Rathburn et al.,

1996]. This range in d13C between species has

been attributed to a combination of differences in

depth distribution preferences between species,

pore water isotope gradients and biological factors

[McCorkle et al., 1990]. Pore water d13C values in

nonseep environments are typically more negative

(lighter) than bottom water carbon isotope values

because of the contribution of 13C-depleted CO2

from the decomposition of isotopically light (about

�23%) marine organic matter (see McCorkle et al.

[1990] for discussion). Globobulimina species

are considered ‘‘deep infaunal’’ taxa because

of their ability to live relatively deep within the

sediments [e.g., Jorissen et al., 1995]. Bulimina

and Uvigerina species are considered ‘‘shallow

infauna’’, and Cibicides and Planulina species are

considered ‘‘epifaunal’’ (surface dwelling) [e.g.,

Corliss and Emerson, 1990]. As noted by several

authors, these categories are only rough approx-

imations of typical distribution patterns, as popu-

lation dynamics can change in accordance with

alterations in environmental conditions [e.g.,

Gooday and Rathburn, 1999]. Although each spe-

cies in a nonseep environment has low isotopic

variability, species that are able to live deeper

within the sediments have more negative d13Cvalues than those that prefer shallower sediment

depths [McCorkle et al., 1990, 1997; Rathburn et

al., 1996]. Despite the influence of pore water

isotopic compositions, there is an offset between

pore water isotope values and foraminiferal d13Csignatures. This difference is attributed to biolog-

ical factors called ‘‘vital effects’’ [e.g., McCorkle et

al., 1990]. Although individuals of a given species

may occur over a relatively wide range of sediment

depths, each benthic foraminiferal species main-

tains a narrow range of isotopic signatures in

nonseep habitats [Rathburn et al., 1996; McCorkle

et al., 1997; Corliss et al., 2002]. Thus, in nonseep

habitats, foraminifera are influenced by pore water

isotope composition, but individuals are somehow

able to maintain a narrow range of isotopic values

despite pore water isotopic gradients [McCorkle et

al., 1990, 1997; Rathburn et al., 1996]. Species

occurring at seeps exhibit a great deal more isotopic

variability than nonseep species. The offset

between seep foraminiferal isotopes and pore water

values is also much greater than in nonseep

habitats (Figures 3 and 6a–10). However, the

general pattern of deep infaunal taxa having more

negative values than shallow infauna and epifauna

is observed in both seep and nonseep habitats



22 of 28

(Figure 8). This suggests that factors that promote

foraminiferal depth stratification and isotope sig-

natures in nonseep habitats are also influencing

foraminifera living in seeps.

[31] Given the spatial and temporal variability of

fluid flow through seepage area sediments, and the

resulting diversity of pore water chemistries in

these environments, it is not surprising that pop-

ulations of single-celled creatures such as forami-

nifera could be exposed to a wide array of

microenvironments within a given seep. This vari-

ety of microenvironments may be reflected in the

unusually high variability of d13C values from

living foraminiferal carbonate (Figures 6a–10).

Symbionts, bacterial consortia, species’ distribu-

tions, and pore water variability may all influence

stable isotope values in seep foraminifera. Symbi-

ont activities can affect foraminiferal isotopes [e.g.,

Erez, 1978], and may be influencing foraminiferal

carbonate isotopes in seep habitats. Although

foraminiferal symbionts are likely in seeps, more

work is needed to examine this possibility, as

endosymbionts have not yet been demonstrated in

seep foraminifera [Bernhard et al., 2001]. In

methane seeps, bacterial consortia are found in

concentric aggregations, typically concentrated near

the sulfate reduction boundary [Boetius et al., 2003].

These bacterial consortia produce isotopically neg-

ative methane, potentially altering pore water iso-

tope values in their immediate vicinity. At least

some Globobulimina species and U. peregrina are

known to ingest bacteria [Goldstein and Corliss,

1994]. Since these aggregates influence surround-

ing pore waters on micrometer scales, and since

bacteria are prey for at least some foraminifera, it is

likely that within a 0.5 cm depth interval some

individuals of foraminifera would be in close

proximity to aggregates, while others might not

be. Heterogeneity in aggregates and their influence

on local pore water d13C together with variability

in seep fluid flow could easily account for the

observed infaunal foraminiferal carbonate d13Cvariability. It is likely that variability in pore water

fluid flow and methane influence would be more

pronounced below the upper 1 cm, and we would

expect individuals living on/near the sediment-

water interface to be less affected by pore water

variability during the life of a foraminiferan. This

idea is supported by the infaunal increase in

U. peregrina d13C variability in the 1781 clam

bed and the high variability of the deep infaunal,

G. pacifica (Figure 7; Table 2).

[32] The influence, if any, of symbionts on the

isotopic geochemistry of seep foraminifera is also

not known. In nonseep sulfidic environments of the

Cariaco Basin, Virgulinella fragilis, has symbionts

and a d13C value of �6.4% [Bernhard, 2003].

Thus it is possible that the presence of symbionts

influences d13C signatures of some taxa. Although

symbionts are common in some metazoan fauna

inhabiting seeps [e.g., Levin et al., 2000], on the

basis of very limited data, endosymbionts have not

yet been found in seep foraminifera [Bernhard et

al., 2001]. Possible ectobionts were found in the

test pores of one U. peregrina specimen from

Monterey Bay seeps, and peroxisome-endoplasmic

reticulum complexes were found in some speci-

mens [Bernhard et al., 2001]. The role of perox-

isomes, if any, is not known [Bernhard, 1996;

Bernhard et al., 2003]. The presence of endobionts

in Buliminella tenuata from nonseep habitats

in Santa Barbara Basin, and the lack of such

symbionts in conspecifics from Monterey Bay

seep habitats indicates that at least some species

are able to live symbiont-free in seeps [Bernhard

et al., 2001]. It is unclear why a species with

symbiont-bearing capabilities apparently lacks

symbionts when inhabiting the potentially toxic

microhabitats of seeps. Both the lack of known

endemic foraminiferal taxa in methane seeps and

symbiont variability within a species suggest

ecological plasticity. This plasticity may also be

reflected in the variability of isotopic signatures of

the tests.

6.3. Relationship With Pore Water D13C

[33] Considering the carbon isotope values of

ambient seep pore waters, it is intriguing that

the range of living seep foraminiferal carbonate

values is not larger than it is. At Site 1780,

geochemical data indicate much lower methane

oxidation compared to Site 1781. Pore water

dissolved CO2 (DIC) d13C values at Site 1780



23 of 28

are about �18% at 4 cm (Figure 3). On the other

hand, d13CDIC values of the 1781 clam bed and

bacterial mat had very negative d13C values below

the upper 2–4 cm (�45% and �59% respectively;

Figure 3). This influence of methane on the pore

waters is reflected in the more negative averages of

living foraminiferal carbonate d13C signatures in

1781 environments compared to 1780 (Figure 10;

Table 2). However, the isotopic influence on

foraminiferal carbonate is not as substantial as

would be expected on the basis of the difference

in pore water d13C between the two sites. Although

the isotopic geochemistry of seep foraminifera is

influenced by environmental factors, the stable

isotopic effects on foraminiferal carbonate are

limited to a few per mil. While these effects

are substantial compared to nonseep effects,

foraminiferal carbonate d13C values are greatly

out of equilibrium with carbon isotope values of

ambient pore water. This isotopic disequilibrium

was also observed for two species (U. peregrina

and Cibicides wuellerstorfi) collected from Hydrate

ridge seeps [Torres et al., 2003]. Although we don’t

know for certain, it seems likely that preferred

micro-environments (sediment depths where

species densities are greatest) are where most

members of the population are growing and

generating their carbonate tests.

[34] In isotope studies of foraminifera living in

nonseep environments, differences between pore

water isotope values and those of foraminiferal

carbonate have traditionally been explained by

migration, inherent biological fractionation (vital

effects), seasonal growth, or a combination of

factors [see Rathburn et al., 1996; McCorkle et

al., 1997]. These explanations are less satisfactory

in seep environments because of the very

negative pore water isotope values, the long

migration distances required to reach microhabi-

tats with pore waters having similar values to

those of the tests, and the necessity to congregate

in the upper one millimeter (or less) to achieve

isotopic equilibrium with ambient pore waters

with subsequent dispersal deeper into sediments

to produce observed distribution patterns. In

other words, in order to achieve the observed

combination of distribution and d13C signature,

most foraminiferal species would have to grow

within a few millimeters of the sediment-water

interface and then migrate down in the sediments

(for some reason other than obtaining food for

growth). These scenarios seem unlikely. Temporal

changes in fluid flow direction have been

reported by Tryon et al. [1999] for methane

seeps on Hydrate Ridge, and the inflow of

bottom water could bring bottom water fluids

into infaunal habitats [Torres et al., 2003]. The

variability of living foraminiferal carbon isotopes

within and between methane seep environments

implies that foraminifera do not grow in selected

places and times when methane seepage is not

active, as fluid advection (and therefore methane

influence) vary with time and space in seep

habitats. It seems unlikely that all seep forami-

nifera delay growth for only the inflowing pulses,

and are not growing, but merely surviving during

outflow. This scenario also presumes downward

flow in all seeps and seep habitats, and alterations

in seep flow causing changes in pore water

isotope values to the appropriate equilibrium

value for each species in sequence within the

sediments (deeper taxa have more negative iso-

tope values). Pumping of bottom water fluids into

infaunal habitats by Calyptogena clams may also

influence the d13C values of seep foraminifera.

This is supported by the fact that clam bed

foraminifera seem to have greater d13C variability

than those living in bacterial mats. However, this

isotopic variability may also result from more

variable microhabitats in the clam beds and more

rapid fluid flow in the mats.

[35] Clearly biological factors play a role in the

isotopic signature of benthic foraminifera, but

conspecific isotopic differences between seep

sites less than 8 km apart in Monterey Bay

indicate that there also is an ambient environ-

mental effect on foraminiferal isotopes, confirm-

ing conclusions reached by previous nonseep

work [e.g., McCorkle et al., 1990; Rathburn et

al., 1996] and current culturing work [McCorkle

et al., 2002]. Rathburn et al. [1996] speculated

that food preferences were a possible explanation

for the narrow range of isotope values observed

in nonseep foraminifera distributed in sediments



24 of 28

having different pore water isotope values.

Microhabitats created by aggregations of seep

bacteria [Boetius et al., 2000] and differences

in dietary composition between foraminiferal

species [Goldstein and Corliss, 1994] may help

to explain the differences between measured pore

water d13C and the carbon isotope values of

foraminiferal carbonate in seep and nonseep

habitats. In a recent study examining carbon

isotope values of metazoan tissues, Levin and

Michener [2002] found methane-influenced dif-

ferences between the isotope signatures of seep

infauna. On the basis of these isotope signatures,

dietary differences of different taxa were inferred.

The dietary habits of seep foraminifera have

never been examined, though this information

is important for understanding foraminiferal dis-

tribution patterns and geochemical signatures.

[36] Causes for the variability of d13C signatures

within foraminiferal species associated with seeps

are unknown, but we speculate that they likely

reflect the variability of seep pore water geo-

chemistry and possibly the influence of aggrega-

tions of organic material and bacteria. There are

clearly biological and environmental influences

affecting d13C signatures of benthic foraminifera,

but it is unclear why seep foraminifera secrete

carbonate with d13C values that are extremely

different from ambient pore water d13C values.

Additional work is needed to discern how this is

possible.

6.4. Relevance to the Fossil Record andImplications for Recognition ofPaleo-seepage

[37] Some studies of fossil material suggest that, as

a result of methane influence, benthic foraminiferal

carbonate secreted in seepage areas will have more

negative d13C values compared to those of nonseep

areas [e.g., Sen Gupta and Aharon, 1994; Wefer et

al., 1994; Kennett et al., 2003]. However, the idea

that a negative d13C signature in foraminiferal

carbonate indicates methane influence has not been

universally accepted [Stott et al., 2002; Bernhard,

2003]. Available data confirm that different fora-

miniferal species are affected differently by ambi-

ent conditions (Figures 6a–10) [Rathburn et al.,

1996; McCorkle et al., 1997]. These species-

specific differences in isotopic response to ambient

conditions may account for some of the controversy.

Results from this study indicate that living methane

seep foraminifera have more variable and more

negative d13C values (on average) than those living

in nonseep environments. Although the carbon

isotope signal in Monterey Bay seep foraminifera

is not as large as expected on the basis of pore

water data, our living seep foraminiferal isotope

results are consistent with those obtained from

Santa Barbara Basin and Peruvian margin fossil

material interpreted to result from changes in

methane seepage.

[38] The significantly larger carbon isotopic vari-

ability observed within a given species is unique

to methane seep environments. Two questions

remain: (1) why nonseep foraminiferal d13C values

are typically invariable, and (2) why the ranges of

seep species are not much more negative, more

closely reflecting the isotopic composition of the

DIC. Although the isotopic signal is not uniform

throughout the assemblage or even within a given

species, it is clear that the range and variability of

d13C observed in Monterey Bay seep foraminifera

exceed those of conspecifics from nonseep envi-

ronments. As mentioned above, it is unclear why

nonseep foraminiferal d13C values are typically

invariable, nor is it clear why the ranges of seep

species d13C values are not larger. Nevertheless, it

is reasonable to attribute the d13C range and

variability observed within a seep species to prop-

erties of the seeps. Thus we believe it is possible to

recognize the extent and chronology of methane

seepage on the basis of the variability and range of

d13C values of fossil foraminifera. It is important to

note that the variability and range of isotopic

signals in seep foraminifera can only be observed

through analyses of individual tests (or analysis of

as few tests as possible). It is therefore critical that

assessments of methane seepage be based on

individual foraminiferal isotopes, and not on anal-

yses that group several individuals together for an

averaged isotopic signal. Average isotopic signals

from groups of foraminifera could dilute the meth-

ane seepage signal and mask isotopic variability.

Analyses of fossil material from Monterey Bay



25 of 28

seeps will be presented in a companion paper by

J. B. Martin et al. (Relationships between the stable

isotopic signatures of living and fossil foraminifera

in Monterey Bay, California, submitted to Geo-

chemistry, Geophysics, Geosystems, 2003).

7. Conclusions

[39] 1. On the basis of the foraminiferal isotopic

differences between sites, it is clear that there is an

effect of methane-influenced pore waters on foram-

iniferal distributions and carbonate geochemistry.

This effect, however, is much smaller than would

be predicted on the basis of ambient d13CDIC.

[40] 2. Living foraminiferal carbonate from seeps

have much greater heterogeneity in d13C values

than published d13C values of foraminifera living

in nonseep habitats.

[41] 3. Through isotopic analyses of individuals,

the range and variability of d13C values within a

given foraminiferal species can be used to assess

methane seepage. Provided there is no diagenetic

alteration of the test carbonate, it should be possi-

ble to use d13C values of individual fossil forami-

nifera to evaluate ancient methane seepage.

[42] 4. It is not clear what relative influences

biological, ecological, and physical factors have

on the carbon isotopic signatures observed in seep

foraminifera. Without additional biological and

geochemical data, proposed mechanisms remain

speculative.

[43] 5. Nevertheless, our results support the hy-

pothesis that foraminifera can be used to assess past

and present methane seepage.

Acknowledgments

[44] The authors would like to thank the MBARI Ventana

crew and staff for very professional and pleasant operations.

We also would like to thank Geoff Wheat for his advice and

assistance with logistics. We are sincerely grateful for help at

sea and/or in the laboratory from Laurie McConnico, Peter

Weber, and Jason Curtis. We appreciate thorough reviews by

Bruce Corliss and an anonymous reviewer. This project was

funded by NOAA’s National Undersea Research Program

through the West Coast and Polar Regions Undersea Research

Center at the University of Alaska Fairbanks (NOAA-NURC

grants FP004999 and UAF98-0043). Support for M.E.P

was provided by a Fellowship from the Basque Country

Government.

References

Akimoto, K., T. Tanaka, M. Hattori, and H. Hotta, Recent

benthic foraminiferal assemblages from the cold seep com-

munities-a contribution to the methane gas indicator, in

Pacific Neogene Events in Time and Space, edited by

R. Tsuchi, pp. 11–25, Univ. of Tokyo Press, Tokyo, 1994.

Barry, J. P., H. G. Greene, D. L. Orange, C. H. Baxter, B. H.

Robinson, R. E. Kochevar, J. W. Nybakken, D. L. Reed, and

C. M. McHugh, Biologic and geologic characteristics of cold

seeps in Monterey Bay, California, Deep Sea Res., Part II,

43, 1739–1762, 1996.

Bernhard, J. M., Microaerophilic and facultative anaerobic

benthic foraminifera: A review of experimental and ultra-

structural evidence, Rev. Paleobiol., 15, 261–275, 1996.

Bernhard, J. M., Distinguishing live from dead foraminifera:

Methods review and proper applications, Micropaleontology,

46, suppl. 1, 38–46, 2000.

Bernhard, J. M., Potential symbionts in bathyal foraminifera,

Science, 299, 861, 2003.

Bernhard, J. M., K. R. Buck, and J. P. Barry, Monterey Bay

cold-seep biota: Assemblages, abundance, and ultrastructure

of living foraminifera, Deep Sea Res., Part I, 48, 2233–

2249, 2001.

Bernhard, J. M., P. T. Visscher, and S. S. Bowser, Sub-milli-

meter life positions of bacteria, protists, and metazoans in

laminated sediments of the Santa Barbara Basin, Limnol.

Oceanogr., 48(2), 813–828, 2003.

Boetius, A., K. Ravenschlag, C. J. Schubert, D. Rickert,

F. Widdel, A. Gieseke, R. Amann, B. B. Jorgensen, U. Witte,

and O. Pfannkuche, A marine microbial consortium appar-

ently mediating anaerobic oxidation of methane, Nature,

407, 623–625, 2003.

Cline, J. D., Spectrophotometric determination of hydrogen

sulfide in natural waters, Limnol. Oceanogr., 14, 454–458,

1969.

Corliss, B. H., Morphology and microhabitat preferences of

benthic foraminifera from the northwest Atlantic Ocean,

Mar. Micropaleontol., 17, 195–236, 1991.

Corliss, B. H., and T. C. E. van Weering, Living (stained)

benthic foraminifera within surficial sediments of the

Skagerrak, Mar. Geol., 111, 323–335, 1993.

Corliss, B., and S. Emerson, Distribution of Rose Bengal

stained deep-sea benthic foraminifera from the Nova Scotian

continental margin and Gulf of Maine,Deep Sea Res., Part A,

3, 381–400, 1990.

Corliss, B. H., D. C. McCorkle, and D. M. Higdon, A time

series study of the carbon isotopic composition of deep-sea

benthic foraminifera, Paleoceanography, 17(3), 1036,

doi:10.1029/2001PA000664, 2002.

Dickens, G. R., J. R. O’Neil, D. K. Rea, and R. M. Owen,

Dissociation of oceanic methane hydrate as a cause of

the carbon isotope excursion at the end of the Paleocene,

Paleoceanography, 10, 965–971, 1995.



26 of 28

Dickens, G. R., M. M. Castillo, and J. C. G. Walker, A blast of

gas in the latest Paleocene: Simulating first-order effects of

massive dissociation of oceanic methane hydrate, Geology,

25, 259–262, 1997.

Doose, P. R., The bacterial production of methane in marine

sediments, thesis, 262 pp., Univ. of Calif. Los Angeles, Los

Angeles, 1980.

Doose, P. R., and I. R. Kaplan, Biogenetic control of gases in

marine sediments of Santa Barbara Basin, California, AAPG

Bull., 65(5), 919–920, 1981.

Erez, J., Vital effect on stable-isotope compositions seen in

foraminifera and coral skeletons, Nature, 273, 199–202,

1978.

Fontanier, C., F. J. Jorissen, L. Licari, A. Alexandre,

P. Anschutz, and P. Carbonel, Live benthic foraminiferal

faunas from the Bay of Biscay: Faunal density, composi-

tion, and microhabitats, Deep Sea Res., Part I, 49, 751–

785, 2002.

Goldstein, S. T., and B. H. Corliss, Deposit feeding in selected

deep-sea and shallow-water benthic foraminifera, Deep Sea

Res., Part I, 41, 229–241, 1994.

Gooday, A. J., and A. E. Rathburn, Temporal variability in

living deep-sea benthic foraminifera: A review, Earth Sci.

Rev., 46, 187–212, 1999.

Graber, E. R., and P. Aharon, An improved microextraction

technique for measuring dissolved inorganic carbon (DIC),

d13CDIC and d18OH2O from milliliter-size water samples,

Chem. Geol., 94(2), 137–144, 1991.

Hesselbo, S. P., D. R. Grocke, H. C. Jenkyns, C. J. Bjerrum,

P. Farrimond, H. S. M. Bell, and O. R. Green, Massive

dissociation of gas hydrate during a Jurassic oceanic anoxic

event, Nature, 406, 392–395, 2000.

Hinrichs, K., L. Hmelo, and S. Syla, Molecular fossil record of

elevated methane levels in Late Pleistocene coastal waters,

Science, 299, 1214–1217, 2003.

Jahren, A. H., N. C. Arens, G. Sarmiento, J. Guerrero, and

R. Amundson, Terrestrial record of methane hydrate disso-

ciation in the Early Cretaceous, Geology, 29(2), 159–162,

2001.

Jeroschewsky, P., C. Steuckart, and M. Kuehl, An ampero-

metric microsensor for the determination of H2S in aquatic

environments, Anal. Chem., 68, 4351–4357, 1996.

Jørgensen, B. B., and N. P. Revsbech, Oxygen uptake, bacte-

rial distribution and carbon-nitrogen-sulphur cycling in sedi-

ments from the Baltic Sea North Sea transition, Ophelia, 31,

29–49, 1989.

Jorissen, F. J., H. C. de Stigter, and J. G. V. Widmark, A

conceptual model explaining benthic foraminiferal microha-

bitats, Mar. Micropaleontol., 22, 3–15, 1995.

Jorissen, F. J., I. Wittling, J. P. Peypouquet, C. Rabouille, and

J. C. Relexans, Live benthic foraminiferal faunas off Cape

Blanc, NW Africa: Community structure and microhabitats,

Deep Sea Res., Part I, 45, 2157–2188, 1998.

Keigwin, L. D., Late Pleistocene-Holocene paleoceanography

and ventilation of the Gulf of California, J. Oceanogr., 58,

421–432, 2002.

Kennett, J. P., I. L. Hendy, and R. J. Behl, Late Quaternary

foraminiferal carbon isotopic record in Santa Barbara Basin:

Implications for rapid climate change, Eos Trans. AGU,

77(46), Fall Meet. Suppl., F294, 1996.

Kennett, J. P., K. G. Cannariato, I. L. Hendy, and R. J. Behl,

Carbon isotopic evidence for methane hydrate instability

during Quaternary interstatidials, Science, 288, 128–133,

2000.

Kennett, J. P., K. G. Cannariato, I. L. Hendy, and R. J. Behl,

Methane Hydrates in Quaternary Climate Change: The

Clathrate Gun Hypothesis, 216 pp., AGU, Washington,

D. C., 2003.

Kvenvolden, K., Gas hydrates: Geological perspective and

global change, Rev. Geophys., 31, 173–187, 1993.

Levin, L. A., and R. H. Michener, Isotopic evidence for

chemosynthesis-based nutrition of macrobenthos: The light-

ness of being at Pacific methane seeps, Limnol. Oceanogr.,

47(5), 1336–1345, 2002.

Levin, L. A., D. W. James, C. M. Martin, A. E. Rathburn, L. H.

Harris, and R. H. Michener, Do methane seeps support

distinct macrofaunal assemblages? Observations on commu-

nity structure and nutrition from the northern California

slope and shelf, Mar. Ecol. Progr. Ser., 208, 21–39, 2000.

Levin, L. A., W. Ziebis, G. Mendoza, V. Growney, C. Mahn,

J. Geiskes, M. Tryon, K. Brown, and A. Rathburn, Distribu-

tion of macrofauna and large sulfur bacteria at California

methane seeps in relation to sulfide concentration and fluid

flow, Mar. Ecol. Progr. Ser., in press, 2003.

Lutze, G. F., and A. Altenbach, Technik und Signifikanz der

Lebendfarbung benthischer Foraminiferen mit Bengalrot,

Geol. J., A128, 251–265, 1991.

Martin, J. B., D. L. Orange, T. D. Lorenson, and K. A.

Kvenvolden, Chemical and isotopic evidence of gas-influ-

enced flow at a transform plate boundary: Monterey Bay,

California, J. Geophys. Res., 102(B11), 24,903–24,915,

1997.

McCorkle, D. C., L. D. Keigwin, B. H. Corliss, and S. R.

Emerson, The influence of microhabitats on the carbon iso-

topic composition of deep-sea benthic foraminifera, Paleo-

ceanography, 5, 161–185, 1990.

McCorkle, D. C., B. H. Corliss, and C. Farnham, Vertical

distributions and isotopic compositions of live (stained)

benthic foraminifera from the North Carolina and California

continental margins, Deep Sea Res., Part I, 44, 983–1024,

1997.

McCorkle, D. C., J. M. Bernhard, C. J. Hintz, J. K. Blanks,

D. R. Ostermann, J. T. Shaw, and G. T. Chandler, The carbon

and oxygen stable isotopic composition of cultured benthic

foraminifera (Bulimina aculeata), Eos Trans. AGU, 83(47),

Fall Meet. Suppl., abstract PP51A-0273, 2002.

Rathburn, A. E., and B. H. Corliss, The ecology of deep-sea

benthic foraminifera from the Sulu Sea, Paleoceanography,

9, 87–150, 1994.

Rathburn, A. E., B. H. Corliss, K. D. Tappa, and K. C.

Lohmann, Comparisons of the ecology and stable isotopic

compositions of living (stained) deep-sea benthic foramini-

fera from the Sulu and South China Seas, Deep Sea Res.,

Part I, 43(10), 1617–1646, 1996.

Rathburn, A. E., L. A. Levin, Z. A. Held, and K. C. Lohmann,

Benthic foraminifera associated with cold methane seeps on



27 of 28

the northern Californian Margin: Ecology and stable isotopic

composition, Mar. Micropaleontol., 38, 247–266, 2000.

Rathburn, A. E., M. E. Perez, and C. Lange, Benthic-pelagic

coupling in the Southern California Bight: Relationships

between sinking organic matter, diatoms and benthic forami-

nifera, Mar. Micropaleontol., 43, 261–271, 2001.

Revsbech, N. P., An oxygen microelectrode with a guard cath-

ode, Limnol. Oceanogr., 34, 472–476, 1989.

Sen Gupta, B. K., and P. Aharon, Benthic foraminifera of

bathyal hydrocarbon vents of the Gulf of Mexico: Initial

report on communities and stable isotopes, Geo Mar. Lett.,

14, 88–96, 1994.

Sen Gupta, B. K., E. Platon, J. M. Bernhard, and P. Aharon,

Foraminiferal colonization of hydrocarbon-seep bacterial

mats and underlying sediment, Gulf of Mexico slope,

J. Foraminiferal Res., 27, 292–300, 1997.

Sibuet, M., and K. Olu, Biogeography, biodiversity and fluid

dependence of deep-sea cold-seep communities at active and

passive margins, Deep Sea Res., Part II, 45, 517–567, 1998.

Silva, K. A., B. H. Corliss, A. E. Rathburn, and R. C. Thunell,

Seasonality of living benthic foraminifera from the San Pedro

Basin, California Borderland, J. Foraminiferal Res., 26(1),

71–93, 1996.

Smith, L. M., J. P. Sachs, A. E. Jennings, D. M. Anderson, and

A. deVernal, Light delta 13C events during deglaciation of

E. Greenland continental shelf attributed to methane release

from gas hydrates, Geophys. Res. Lett., 28(11), 2217–2220,

2001.

Stakes, D. S., D. L. Orange, J. B. Paduan, K. A. Salamy, and

N. Maher, Cold-seeps and authigenic carbonate formation in

Monterey Bay, California, Mar. Geol., 159(1–4), 93–109,

1999.

Stott, L. D., T. Bunn, M. Prokopenko, C. Mahn, J. Gieskes, and

J. M. Bernhard, Does the oxidation of methane leave an iso-

topic fingerprint in the geologic record?, Geochem. Geophys.

Geosyst., 3(2), 1012, doi:10.1029/2001GC000196, 2002.

Thomas, E., J. C. Zachos, and T. J. Bralower, Deep-sea envi-

ronments on a warm Earth: Latest Paleocene - early Eocene,

in Warm Climates in Earth History, edited by B. Huber,

K. MacLeod, and S. Wing, pp. 132–160, Cambridge Univ.

Press, New York, 2000.

Torres, M. E., A. C. Mix, K. Kinports, B. Haley, G. P.

Klinkhammer, J. Manus, and M. A. de Angelis, Is methane

venting at the seafloor recorded by d13C of benthic forami-

nifera shells?, Paleoceanography, 18(3), 1062, doi:10.1029/

2002PA000824, 2003.

Tryon, M., K. Brown, M. E. Torres, A. M. Trehu, J. McManus,

and R. W. Collier, Measurements of transience and flow

cycling near episodic methane gas vents, Hydrate Ridge,

Cascadia, Geology, 27, 1075–1078, 1999.

Wefer, G., P.-M. Heinze, and W. H. Berger, Clues to ancient

methane release, Nature, 369, 282, 1994.



28 of 28

Relationships between the distribution and stable isotopic composition of living benthic...

Documents

Transcript of Relationships between the distribution and stable isotopic composition of living benthic...