20080868 Ver 2_Science Panel Meeting_20101027

October 27, 2010 Science Panel Meeting
0945 —1000
Welcome and Introductions (Walker/All)
1000-1100
Current status of and updates to the monitoring plan (PCS / CZR)
1100-1200
Biomass Size Spectra presentation (Kimmel)
1200 —1230
Lunch
1230 —1600
Field Sites to include
Pit overview
Reclamation sites — Whitehurst Creek, Capping, finished sites
Bonnerton hardwood area
1600 —1630
Moving forward discussion (All)

Regulatory Division
Action ID. 200110096
Potash Corporation of Saskatchewan Phosphate Division, Aurora Mine
Attn: Mr. Steve Beckel
1530 NC Highway 306 S
Aurora, NC 27806
Dear Mr. Beckel:
Please reference the Department of the Army permit issued Potash Corporation of
Saskatchewan (PCS) Phosphate Division authorizing the discharge of dredged and fill
material into 3,927 acres of waters of the US adjacent to the Pamlico River and several of
its tributaries located north of the town of Aurora, Beaufort County, North Carolina.
These impacts were authorized as part of a continuation of mining activities at the
existing Aurora Mine facility.
As a condition of this authorization PCS is required to carry out chemical and biological
monitoring to assess the effects of the reduction in headwater wetlands on the utilization
of Porters Creek, Tooley Creek, Jacobs Creek, Drinkwater Creek, and Jacks Creek as
nursery areas by resident fish and appropriate invertebrate species. To this end, PCS
submitted a draft plan of study for agency review and comment in December 2009. The
plan, titled "Draft Plan of Study to Monitor Potential Effects of Reduction in Headwater
Wetlands on the Downstream Aquatic Functions and Utilization of Tributaries of South
Creek, Porter Creek, and Durham Creek Beaufort County North Carolina" included
proposed methods for testing the six (6) questions identified in Special Condition "S" of
the permit. Those questions are:
1. Has mining altered the amount or timing of water flows within the creeks?
2. Has mining altered the geomorphic or vegetative character of the creeks?
3. Has mining altered the forage base of the creeks?
4. Has mining altered the use of the creeks by managed fish?
5. Has mining increased contaminate levels within creek sediments to levels that
could impact fish or invertebrates?
6. Has mining altered overall water quality within creeks?

To aid in this effort, PCS worked with the Corps, the North Carolina Division of Water
Quality (NCDWQ) and other interested parties to establish an independent
multidisciplinary panel of researchers qualified in the subject matter to be examined. The
major purpose of this panel is to review reports for proper experimental design and
validity of data analysis and conclusions, identify potential trends, and comment on
potential areas of further study or changes to existing protocols. This panel was invited
to review the proposed sampling procedures and monitoring reports from previous work
and provide feedback to the Corps, NCDWQ, the NC Division of Coastal Management
(NCDCM) and PCS. On June 16, 2010, the panel met with representatives from the
Corps, NCDWQ, NCDCM and PCS to discuss the plan further and provide feedback.
A general suggestion offered by members of the panel was to separate each tributary into
a "stream" section and a "creek" section. Stream sections would be those segments
usually experiencing unidirectional flow from upstream runoff and groundwater input.
Creek sections would be those segments usually experiencing bidirectional flow as a
result of the overriding influence of wind and/or lunar tides. All agreed that this
distinction could be useful in experimental design and result interpretation. Therefore,
for purposes of this monitoring program, these terms will be adopted as described. The
term tributary will be used when referring to the entire system of any given named water
course.
The panel also provided input on the proposed reference creeks. It was suggested that
some preliminary sampling be done in the proposed reference creeks to determine how
similar they actually are to the study creeks. The group also suggested that Long Creek,
Short Creek, Little Creek and PA2 be considered for use as reference creeks. While these
creeks would not be appropriately defined as "unaltered" systems, they could serve as
potential controls. The goal of much of the monitoring efforts is to determine what, if
any changes in the measured parameters can be attributed to the actions of the permittee.
Since inflow from South Creek and the Pamlico River influence the conditions within the
study tributaries, monitoring within these other tributaries of South Creek would provide
valid data on changes that can be attributed to alterations in the conditions of the larger
waters. Additionally, due to the historic impacts to the drainage basins of these creeks
the monitoring may yield some evidence as to the predictable long-term conditions of the
study creeks.
The Panel provided input on specific details of the monitoring plan as well. In regard to
water table/water level monitoring, members of the panel suggested that data should, at
least periodically, be collected at a greater frequency (approx. 15 min. intervals) within
and surrounding the stream reaches in order to capture episodic pulse events. The panel
agreed that the value of monitoring water table/water level within the creek reaches of
these tributaries would be limited due to bidirectional flow and the overwhelming
influence of wind tides.
In regard to monitoring of the geomorphic or vegetative character of the tributaries, panel
members agreed that the proposed parameters would be telling but suggested some
modification in the methods and/or frequency of data collection. The group suggested

that due to the low energy and stability of these tributaries, it is very unlikely that rapid
changes in channel morphology will occur. Therefore, parameters such as stream/creek
position, length, width and sinuosity could be measured at a frequency greater than 1
year. The group further suggested that changes in the cross-sectional dimensions of the
creek reaches as a result of changes in flow would not be expected and any change
observed would not likely be discernable from predicted sea -level rise. Therefore, cross
sectional surveying should be limited to the stream reaches. The group discussed the
benefits of annual vegetation surveys along the tributary corridors and decided that less
frequent surveying would likely be appropriate.
The group discussed monitoring fish and invertebrate utilization within the subject
tributaries as an indicator of forage base. Members of the panel suggested concentrating
benthic invertebrate collection efforts during winter and early spring when species
richness and diversity are greatest. The group suggested that these efforts should be
adequate to capture all benthic invertebrate species thereby eliminating the need for
specific efforts to capture bivalves. The group also discussed sampling of more motile
and pelagic species. Members of the panel suggested that any sampling regime should be
designed to target the variety of species expected to occur within these systems. This
could require adjustment of sampling gear and timing to reflect ontogenetic shifts in
creek usage. Proper experimental design here again should be adequate to capture all
expected species including those managed under the Magnuson -Stevens Fishery
Conservation Management Act. The group discussed Biomass Size Spectra (BSS) as a
method of assessing fisheries utilization of the area. Members of the panel opined that
BSS may be a useful tool but more information is needed on the method and the level of
sampling required to produce meaningful results. Further discussion of BSS is planned.
Finally, the group discussed water quality sampling. Members of the panel felt it
appropriate to add water quality sampling stations at the lower end of the Creek reaches
near the confluence with the receiving waters. Members of the panel also suggested
adding measurement of dissolved (DOC) and total (TOC) organic carbon, using
measurement of total suspended solids (TSS) and color as indicators of turbidity, and
adopting more sophisticated techniques for measuring light attenuation. The group
discussed the fact that in these tributaries, particularly those areas most influenced by the
water quality of the receiving waters (creek reaches), no change would be predicted in
some of these parameters (e.g. Salinity). Therefore, monitoring of these parameters may
be continued on a less frequent basis or discontinued altogether once data from the first
several years can be analyzed for potential trending.
After review of the Draft monitoring plan and consideration of the above
referenced discussion, the Corps has determined that to the extent practicable and
appropriate, procedures and monitoring locations established under the 1998
monitoring plan should be continued. Additionally, the monitoring requirements
under special condition "S" of the 2009 permit are refined as follows:
1) Has mining altered the amount or timing of water flows within the
tributaries? Data collection may include:

i) Continuous water level recorders to measure/model average and event
driven (pulse) flow within the stream reaches.
ii) Rain gauges to measure local water input.
iii) Groundwater wells to measure input to the tributaries.
2) Has mining altered the geomorphic or vegetative character of the
tributaries? Data collection may include:
i) Periodic (3-5yr) aerial photography/lidar to determine tributary
position, length, width and sinuosity.
ii) Conduct baseline and periodic (3-5yr) cross sectional surveys of each
tributary at established locations within the stream reaches. Frequency
of the surveys may vary based on relative change observed (>stability
= < frequency).
iii) Periodic (3-5yr) sediment characterization including total sediment
organic matter and porosity.
iv) Periodic (3-5yr) surveys of live vegetation and organic matter
accumulation along tributaries. Frequency of the surveys may vary
based on relative change observed.
3) Has mining altered the forage base (fish and invertebrate utilization) or
use by managed fisheries species within tributaries? Data collection may
include:
i) Sampling of benthic invertebrate populations by core or grab
collection. Sampling efforts should be concentrated during winter and
early spring.
ii) Periodic sampling for pelagic species such as grass shrimp, blue crabs,
and small forage fish. Sampling gears would be chosen to reflect
ontogenetic shifts in creek usage
iii) Biomass size spectra may be used as an approach to addressing this
question.
4) Has mining increased contaminant levels within tributary sediments to
levels that could impact fish or invertebrates? Data collection may
include:
i) Sediment and water column sampling annually or at prescribed
intervals for metals, including cadmium, mercury, silver, copper, and
arsenic.
ii) If elevated levels are detected, the availability and uptake by
appropriate aquatic species (e.g., Rangia sp., blue crabs) should be
measured using appropriate bioassay techniques.
5) Has mining altered overall water quality within creeks? Water quality
parameters analyzed will include: Salinity, Temperature, Dissolved
Oxygen, pH, Secchi depth/light attenuation, Turbidity (TSS, color),

Chlorophyll a, Dissolved orthophosphate phosphorus, Total dissolved
phosphorus, Particulate phosphorus, Nitrate nitrogen, Ammonia nitrogen,
particulate nitrogen, and Dissolved Kjeldahl nitrogen, DOC/TOC
Once all necessary modification and/or amendments to the monitoring plan are
complete, you must submit a final version to the Corps and the NC Division of
Water Quality. The Corps will consider requiring PCS to implement any future
suggestions to improve and/or modify data collection to better address these
questions. Although the Corps may choose not to require PCS to implement
suggestions to study further hypotheses or other areas, PCS may elect to
implement these voluntarily.
Thank you for your attention to this matter. Please do not hesitate to contact me
should you have any questions. I can be reached at telephone # [phone] or
by e-mail at william.t.walkerDa
Copies Furnished:
Ms. Becky Fox
U.S. Environmental Protection Agency - Region 4
1307 Firefly Road
Whittier, NC 28789
Mr. Mike Wicker
U.S. Fish and Wildlife Service
Fish and Wildlife Enhancement
Post Office Box 33726
Raleigh, North Carolina 27636-3726
Mr. Ron Sechler
National Marine Fisheries, NOAA
Habitat Conservation Division
Pivers Island
Beaufort, North Carolina 28516
Mr. John Dorney

Division of Water Quality
North Carolina Department of
Environment and Natural Resources
1621 Mail Service Center
Raleigh, North Carolina 27699-1621
Mr. David Moye
Division of Coastal Management
North Carolina Department of Environment
and Natural Resources
943 Washington Square Mall
Washington, North Carolina 27889
Mr. Richard Peed
Division of Land Resources
North Carolina Department of Environment
and Natural Resources
943 Washington Square Mall
Washington, North Carolina 27889
BCF:
CESAW-RG/Jolly
CESAW-OC/Lamson
CESAW-RG/Walker

Limnol. Oceanogr., 32(6), 1987, 1195-1213
® 1987, by the American Society of Limnology and Oceanography, Inc.
Sulfur, carbon, and nitrogen isotopes used to trace
organic matter flow in the salt -marsh estuaries of
Sapelo Island, Georgia'
Bruce J. Peterson
The Ecosystems Center, Marine Biological Laboratory, Woods Hole, Massachusetts 02543
Robert W. Howarth
Section of Ecology and Systematics, Corson Hall, Cornell University, Ithaca, New York 14853
Abstract
The stable isotopes of sulfur, nitrogen, and carbon were used to trace organic matter flow in salt
marshes and estuarine waters at Sapelo Island, Georgia. Organic matter inputs from terrestrial
sources as detrital input either from forests adjacent to the marshes or from rivers were not
detectable by their isotopic signatures in estuarine consumers. The results suggest that there are
two major sources of organic matter for the fauna of the marshes and estuarine waters of Sapelo
Island: Spartina and algae. The long-standing debate about the relative importance of Spartina
detritus and algae in supporting marsh and estuarine secondary production appears from this
analysis to be a draw; both sources are important and their relative importance is determined by
feeding mode, size, location, and trophic position of the marsh and estuarine consumers.
For 30 yr one of the dominant themes in
estuarine research has been the role of Spar-
tina alterniflora in providing detrital food
to marsh and estuarine consumers. The
concept of "outwelling" of organic matter
from marshes to estuarine and nearshore
marine habitats has been tested in various
locations and with several approaches
(Odum and de la Cruz 1967; Odum 1980;
Nixon 1980). One of the most telling chal-
lenges to the outwelling hypothesis was pre-
sented by Haines (1977). Her studies of the
stable carbon isotopic distribution in the
Sapelo Island marshes of Georgia showed
that seston in the tidal Duplin River and
filter feeders along with other consumers in
the tidal creeks had ratios of 13C: 12C that
were more characteristic of plankton than
of Spartina. The work of Haines and others
at the Sapelo Island marshes has caused a
revision in the way we think about marshes.
Now there is a realization that phytoplank-
ton and benthic diatoms are perhaps equally
as important as Spartina in fueling second-
ary production in the Georgia estuaries
(Pomeroy et al. 1981).
Determining the fate of the carbon fixed
by primary producers in marshes and es-
tuaries is one of the most important con-
' This research was supported by a grant from the
NSF Ecosystems Studies Program DEB 81-04701.
siderations in understanding how these eco-
systems function. However, determining
how this carbon flows in estuaries is partic-
ularly difficult because of the problems en-
countered in quantifying the total net pri-
mary production of estuarine producers, in
measuring the dispersal of this carbon by
the tides, and in determining the consump-
tion of this carbon by bacteria, fungi, and
other consumers both on the surface of the
marsh itself and in the estuary as a whole.
The ultimate fate of the detritus produced
during the senescence and decomposition
of S. alterniflora has been a key question in
salt marsh ecology since the early studies of
Odum and de la Cruz (1967) and Teal (1962)
in the Georgia estuaries.
Since one of the most powerful ap-
proaches for testing the outwelling hypoth-
esis has been the use of carbon isotopic
tracers, it is worthwhile to explore the pos-
sibility that this approach may lead to er-
roneous conclusions. Spartina, plankton,
and upland plants have quite different mean
613C values of about —13, —21, and—2717oo,
respectively, because the classes of plants
either have different sources of CO2 (air vs.
seawater) or different carbon isotopic frac-
tionations (C-3 vs. C-4 photosynthetic
pathways). One reason that the carbon iso-
topic approach may be misleading is that a
mixture of carbon from Spartina and up -
1195

1196
Peterson and Howarth
land plants would very likely have a carbon
isotopic ratio similar to that found for
plankton (Peterson et al. 1980). There are,
in fact, large inputs of allochthonous carbon
from rivers into the Georgia estuaries,
though little is known of the fate of this
material (Haines 1977; Hopkinson and
Hoffman 1984).
The chemoautotrophic bacteria which use
sulfide as an energy source to fix CO2 may
also complicate the carbon isotope story.
These bacteria use the energy provided by
the anaerobic belowground decomposition
of Spartina to fix CO2 in marsh sediments
or at the sediment surface and their pro-
duction in marshes may be high (Howarth
1984). The energy is derived indirectly from
Spartina but the carbon isotopic ratio of the
bacteria may be similar to that of plankton
(Peterson et al. 1980, 1986).
One means of resolving these ambiguities
of the carbon isotopic tracer approach is to
use additional stable isotopic tracers. Sulfur
is a second isotopic tracer of the important
organic matter sources in estuaries and is
independent of the carbon isotopic distri-
bution. The sulfur isotopic approach is based
on the fact that the sulfur available to and
taken up by upland plants, estuarine phy-
toplankton, and Spartina has different iso-
topic ratios. Upland_plants use sulfate avail-
able from precipitation (634S = + 2 to + 8°Ioo),
estuarine phytoplankton use sea -salt sulfate
(634S = +21°loo), and Spartina takes up sul-
fide from sediments for a major fraction of
its total sulfur uptake (Carlson and Forrest
1982). Sulfide produced via sulfate reduc-
tion in sediments averages 5017oo lighter than
sulfate in pore waters in a Massachusetts
marsh (Peterson et al. 1986). The dual tracer
approach using C and S isotopes can be ex-
tended to include the isotopes of nitrogen,
'IN and 14N. Atmospheric nitrogen has a
6'SN value of zero, and upland vegetation
tends to have 6'SN values close to 0°Ioo,
whereas Spartina has 61IN values of around
+60/oo, and nearshore plankton have values
of around +6 to + 1011oo (Wada 1980; Macko
et al. 1984; Altabet and McCarthy 1985). A
multiple tracer approach with isotopes of C,
S, and N can help identify the important
organic matter sources in estuaries.
Most of our work in developing the sulfur
isotopic information and the multiple tracer
approach for marshes has been at Sippe-
wissett marsh, Massachusetts (Peterson et
al. 1985, .1986). In the present study we
tested the multiple stable isotopic tracer ap-
proach in a larger and more complex south-
ern estuary for comparison. We chose the
Sapelo Island marshes because there was a
large body of synthesized ecological infor-
mation on them (Pomeroy and Wiegert
1981) and because E. Sherr (formerly
Haines) and coworkers had previously stud-
ied the carbon and nitrogen isotopic distri-
butions in the biota of the Sapelo marshes
(Haines 1976, 1977; Haines and Montague
1979; Mariotti et al. 1983). We used these
sources of information to guide our sam-
pling.
The objective of this study was to assess
the major sources of energy and organic
matter for macroconsumers in the Sapelo
Island marsh and estuary with the multiple
isotopic tracer approach. In this paper we
first describe the fractionation of the sulfur
isotopes which occurs during sulfate reduc-
tion in salt marsh sediments at Sapelo. We
then present the isotopic compositions of
Spartina, upland plants, and plankton. Then
the results .for May and September sam-
plings of the marsh fauna are presented for
carbon, nitrogen, and sulfur isotopes. We
use dual -isotopic plots .to _show the rela-
tionships between the isotopic composition
of the marsh consumers and the primary
producers. In conclusion, we compare our
results with similar work at Sippewissett salt
marsh.
We thank H. Garritt, S. Merkel, R. Mari-
no, and S. Beard for assistance in collecting
and preparing samples. We are grateful to
the scientists at Sapelo Island who provided
assistance, including D. Kinsey, E. Sherr, C.
Hopkinson, A. Chalmers, W. Wiebe, R.
Wiegert, R. Kneib. We thank P. Mankie-
wicz for advice on sample preparation. B.
Fry, G. Rau, and P. Parker provided sug-
gestions which improved the manuscript.
Sampling times and locations
We collected samples at the Sapelo
marshes on field trips in May 1982 and Sep-
tember 1983. We sampled mostly at the
Kenan Field marsh site on the Duplin Riv-

Stable isotopes in a salt marsh
er, a site deep in the marsh and equally
distant from Doboy Sound and, the head of
the Duplin. The Duplin is a tidal river with
no substantial freshwater surface runoff.
Additional samples were collected at the
airport marsh on the Duplin, at Lab Creek
near the Sapelo Island Marine Laboratory,
at Dean Creek near the bridge between the
laboratory and the barrier beach, on the out-
er barrier beach, and at Bighole Creek which
is a tidal creek opening to the ocean through
the barrier beach on the eastern shore of
Sapelo Island. Upland plant samples were
collected near the marine laboratory and
along a road in the interior of Sapelo Island.
On both sampling trips we collected a va-
riety of samples including seawater, sedi-
ments, S. alterniflora, seston and plankton,
upland plants, and marsh macroconsumers.
In September 1983 we took large cores at
Kenan Field from a short Spartina site and
from a tall Spartina creekbank site for pore -
water sulfate, pore -water sulfide, and pyrite
634S determinations.
Methods and notations
The ratios ofheavy to light stable isotopes
are expressed in the 6 notation which in-
dicates the depletion (—) or the enrichment
(+) of the heavy isotope compared to the
lighter isotope relative to a standard ac-
cording to the formula:
6X (%o) _ [(Rw pie/Rstandard) — 1 ] x 103
where X is 13C, 345, or 15N and R is 13C: 12C,
34S: 325, or 15N : 14N; standard is Peedee Be-
lemnite for C, Canyon Diablo triolites for
S, and air for N.
Pore -water and sediment samples were
taken from stands of short and tall S. al-
terniflora at Kenan Field during the Sep-
tember 1983 sampling trip. The 634S sam-
ples for sulfides and sulfates in pore water
were collected using a 15-cm-diam PVC core
tube and squeezing 5 -cm sediment sections
anoxically in a large Reeburgh (1967) core
press. Sulfide 634S samples were collected
from acidified pore water (H3PO4) with a
bubbling gas -stripping apparatus and
trapped anoxically as cadmium sulfide in
cadmium acetate solution (15%). Samples
for pyrite were collected as cadmium sulfide
precipitates following a chromium reduc-
1197
tion acid distillation (Zhabina and Volkov
1978). Pore -water samples for sulfate were
taken from the anoxic stripping vessel after
sulfide had been removed, and sulfate was
precipitated as barium sulfate in barium
chloride solution (10%). Dried (70°C) and
weighed cadmium sulfide and barium sul-
fate precipitates were sent to Krueger En-
terprises (Cambridge, Mass.) for mass spec-
trometry analyses of 634 S. We used
subsamples of sulfate precipitated from one
batch of seawater as a blind control between
different batches of samples and found dif-
ferences of 0.2-0.30/o0.
Plant tissue samples for 6345, 613C, and
PIN were initially washed free of extra-
neous mud and debris. The tissues were
oven -dried at 70°C and then ground in a
Wiley mill (1 -mm mesh). Samples were
washed in four 1-h rinses ofdeionized water
(4: 1 ratio, water to sample). Samples were
redried at 70°C.
Animal tissue samples for 614S,613 C, and
615N were prepared in a manner preventing
gut and bone contamination, usually by fi-
leting meaty tissues. Animal tissue was dried
at 70°C then ground with mortar and pestle.
Samples were washed in four 1-h rinses of
deionized water (4: 1 ratio, water to sam-
ple). Samples were redried at 70°C. A sub -
sample (> 3 dry g) was selected for 634S anal-
ysis. The remaining sample (>0.1 g for 613C
and 615N) was acid -washed in 10% HCl for
1.5 h to remove carbonate contaminants
and again washed in deionized water.
The plant and animal tissue determina-
tions were performed on pooled samples
from many organisms of the same species
at each sampling site. The primary reason
for pooling was that the sulfur isotopic de-
terminations required a large amount of tis-
sue, usually > 1 g dry weight. This pooling
probably masks variability that might be
detected if individual organisms were ana-
lyzed (Montague et al. 1981). The difference
between duplicate determinations on the
same homogenized sample was usually
<0.20/oo for 613C, 615N, and 634S.
Chloride was measured with a Buchler-
Cotlove chloridometer (Cotlove et al. 1958)
and also with a Dionex ion chromatograph
by anion exchange and conductivity detec-
tion. Sulfate was measured on a Dionex ion

1198 Peterson and Howarth
SULFATE (mM liter-) SULFIDE (mM liter-) PYRITE (mM g DW -1)
7
10
x
CL
20
W
a
30
40
b aeSO4 (%o) b34S2 (%01
+20 +40 +60 -t0 +10 +30
T—T
L
1
1
1
O •
1
0.0 0.4 0.8
•
1
1
1
•
b Fe34SZ(%. )
-20 0
•
1
1
1
1
1
•
•
• SHORT o TALL A DUPLIN RIVER SEAWATER
Fig. 1. Pore -water profiles of sulfate, sulfide, 634S042-, and &14S2- . Also shown are the pyrite concentrations
in the scdiments and the sulfur isotopic composition of pyrite.
chromatograph and also by indirect titra-
tion (Howarth 1978).
Results and discussion
Sulfur in sediments and pore water—We
selected both a tall Spartina creekbank site
and a short Spartina site at Kenan Field to
measure the concentrations of sulfur and the
sulfur 6345 values in sediments and pore
waters (Fig. 1). The chloride profiles (not
shown) were quite similar at the two sites
with only a slight (< 11760) increase in chlo-
rinity near the surface at the short site, prob-
ably due to less effective tidal flushing of
salts accumulated by evapotranspiration.
The two sites had dramatically different
concentrations of pore -water sulfate and
sulfide (Fig. 1). The tall Spartina creekbank
site had high sulfate concentrations which
decreased only slightly with depth and very
low sulfide concentrations which _increased
only slightly below 25 cm. The -short Spar-
tina site had much lower sulfate concentra-
tions and much higher pore -water sulfide
levels than the creekbank site. The concen-
trations of pyrite were similar at these two
particular sites, but in other studies we have
found that pyrite concentrations at Sapelo
are usually lower at creekbank sites (Ho-
warth unpubl.). The sulfate deficit is a mea-
sure of the net depletion of sulfate due to
sulfate reduction relative to the chloride
content of the pore water (Howarth and
Giblin 1983). Both sites had increasing sul-
fate deficits with depth, but the short Spar-
tina site had deficits of up to 90% of the
available sulfate while the tall Spartina site
had deficits of only about 20%.
The isotopic ratios for sulfate, sulfide, and
pyrite were also quite different at the creek-
O
0 10
20
0
2
4
u
to
x
a
20
1
�
i
W
Q
•
1
��
30
40
7
10
x
CL
20
W
a
30
40
b aeSO4 (%o) b34S2 (%01
+20 +40 +60 -t0 +10 +30
T—T
L
1
1
1
O •
1
0.0 0.4 0.8
•
1
1
1
•
b Fe34SZ(%. )
-20 0
•
1
1
1
1
1
•
•
• SHORT o TALL A DUPLIN RIVER SEAWATER
Fig. 1. Pore -water profiles of sulfate, sulfide, 634S042-, and &14S2- . Also shown are the pyrite concentrations
in the scdiments and the sulfur isotopic composition of pyrite.
chromatograph and also by indirect titra-
tion (Howarth 1978).
Results and discussion
Sulfur in sediments and pore water—We
selected both a tall Spartina creekbank site
and a short Spartina site at Kenan Field to
measure the concentrations of sulfur and the
sulfur 6345 values in sediments and pore
waters (Fig. 1). The chloride profiles (not
shown) were quite similar at the two sites
with only a slight (< 11760) increase in chlo-
rinity near the surface at the short site, prob-
ably due to less effective tidal flushing of
salts accumulated by evapotranspiration.
The two sites had dramatically different
concentrations of pore -water sulfate and
sulfide (Fig. 1). The tall Spartina creekbank
site had high sulfate concentrations which
decreased only slightly with depth and very
low sulfide concentrations which _increased
only slightly below 25 cm. The -short Spar-
tina site had much lower sulfate concentra-
tions and much higher pore -water sulfide
levels than the creekbank site. The concen-
trations of pyrite were similar at these two
particular sites, but in other studies we have
found that pyrite concentrations at Sapelo
are usually lower at creekbank sites (Ho-
warth unpubl.). The sulfate deficit is a mea-
sure of the net depletion of sulfate due to
sulfate reduction relative to the chloride
content of the pore water (Howarth and
Giblin 1983). Both sites had increasing sul-
fate deficits with depth, but the short Spar-
tina site had deficits of up to 90% of the
available sulfate while the tall Spartina site
had deficits of only about 20%.
The isotopic ratios for sulfate, sulfide, and
pyrite were also quite different at the creek-

Stable isotopes in a salt marsh
1199
w
a
o �
J _
LU_ c=n +50 $ 34SO4 = -1.60 ( SO4 deficit mM) + 23.8
cr r=-0.98 n=5
Z
LU
+40
v
M O
o
+30 - o
� �� 634S2 - - 2.50 ( SO4 deficit mM) - 20.1
o N +20 % r=-0.98 /n=7
M
+10
C-10 TALL SHORT
C-10 SULFATE 0 •
SULFIDE o •
0 PYRITE 0 •
o
W
W _
n -20 S Fe4S= -0.89 (SO4 deficit mM) - 21 0 0
M
r=-0.88 n=6
-30
-20 -15 -10 -5 0
SULFATE DEFICIT ( mM liter-' )
Fig. 2. The relationship between the 634S values for sulfide, sulfate, and pyritic sulfur and the sulfate deficit.
bank and interior marsh sites. At the creek -
bank site the 634S value for pore -water sul-
fate near the sediment surface was similar
to the values measured in Duplin River sea-
water, but the values increased to + 300/co at
depth (Fig. 1). The sulfide concentrations at
this site were so low that we could only
obtain a single sample which had a 634S val-
ue of -11%o at 30 em. At the short Spartina
site, both the sulfide and sulfate 634S values
were higher than those from the creekbank
site at all depths. The pyrite 634S values were
also higher at the short Spartina site than
at the creekbank site (Fig. 1). At each sam-
pling depth at both sites the 634S values for
sulfide were much lower than the values for
sulfate. There was a very strong relationship
between the sulfate deficit and the 634S val-
ues for sulfate, sulfide, and pyritic sulfur
(Fig. 2). As the size of the pore -water sulfate
pool decreases due to sulfate reduction in
excess of sulfide oxidation, the isotopic ratio
of the sulfate remaining increases. This in-
crease is the result of isotopic fractionation
by sulfate -reducing bacteria during sulfate
reduction. The other consequence of this
discrimination against 34S is that the sulfide
and pyrite produced are depleted in 34S rel-
ative to sulfate. The sulfide 634S values are
30-400/o0 lower than 634S values for sulfate,
and the pyrite 634S values are 40-500lo0 low-
er. The difference in the 34S content of pyrite
and sulfide may be because much of the
pyrite was produced at times of the year
when the degree of sulfate depletion in the
pore water was less. The smaller sulfide pool
probably reflects the 634 ratio of more re-
cently reduced sulfur.
Since the process of sulfate reduction is
important in belowground decomposition
in salt marshes (Howarth and Giblin 1983;
Howarth 1984), we thought that it might be
possible to measure a deficit or excess of
sulfate relative to chloride in Duplin River
seawater (Fig. 3). Either of these conditions
would affect the 634S value of sulfate in the
estuary and therefore could cause the 634S
value of plankton to depart from the value

1200 Peterson and Howarth
w �` 0.145-
UJ
.145W0.135
N0.125- 0.125 0
J
2 I I I t I II 1 1 I A
U
25
19
°-'
24
--SULFATE
18
+ 23
23
ao
17
0,
E
22
•--- ;_O�
�`�� .
16
0
o
�`.�
...
+ 20
21
CHLORIDE
15
w
6 4 2 0 12
10 8 6 4 2 0 Tidal
0
DUPLIN RIVER creek
20
14
0
vii
19
13
v
15�
. . Q
12
w �` 0.145-
UJ
.145W0.135
N0.125- 0.125 0
J
2 I I I t I II 1 1 I A
U
Fig. 3. -Sulfate, chloride, the sulfate: chloride ratio, and 614SO4' - (%o) values for seawater along a transect
from 4.8 km offshore, into Doboy Sound, and up the Duplin River. Tidal creek water was collected at low tide
in a small creek draining the Kenan Field marshes.
expected in seawater. We measured con-
centrations of chloride and sulfate along a
transect and found higher values offshore
and lower values in the Duplin River (Fig.
3). The decrease in sulfate was paralleled by
the decline in .chlorinity; both are due to
nearshore inputs of freshwater from coastal
rivers and groundwater. The ratio of sulfate
to chloride and the isotopic compositic n of
the sulfate did not change significantly along
the transect from offshore to the upper Du-
plin River. Planktonic algae throughout the
Duplin River thus must use sea -salt sulfate
with an isotopic 334S value of +2117o0. How-
ever, a water sample from a small marsh
creek at low tide as water drained from the
Kcnan Field marshes had a lower -sulfate-
to-chloride ratio and a higher 634S value than
Duplin River waters (Fig. 3). This decrease
is what one would expect if sulfate -depleted
and 34S -enriched pore water were draining
into the creeks -as -the tide ebbs.
Isotopic ratios in organic matter produc-
ers --The carbon isotopic ratios in upland
plants, Spartina, and phytoplankton are dif-
ferent (Table 1). Spartina is a C-4 plant, and
photosynthesis in these plants exhibits much
less discrimination against 13C than occurs
during photosynthesis in C-3 plants. Up-
land oaks and pines on Sapelo Island have
613C values of around—290Ioo whereas Spar-
tina averages about —13%0. Both values ex-
+24-
0
8
+ 23
ao
+ 22
It
ro
+ 21
0 0
0
+ 20
6 4 2 0 12
10 8 6 4 2 0 Tidal
OFFSHORE --UP
DUPLIN RIVER creek
SAMPLING
STATIONS (km)
Fig. 3. -Sulfate, chloride, the sulfate: chloride ratio, and 614SO4' - (%o) values for seawater along a transect
from 4.8 km offshore, into Doboy Sound, and up the Duplin River. Tidal creek water was collected at low tide
in a small creek draining the Kenan Field marshes.
expected in seawater. We measured con-
centrations of chloride and sulfate along a
transect and found higher values offshore
and lower values in the Duplin River (Fig.
3). The decrease in sulfate was paralleled by
the decline in .chlorinity; both are due to
nearshore inputs of freshwater from coastal
rivers and groundwater. The ratio of sulfate
to chloride and the isotopic compositic n of
the sulfate did not change significantly along
the transect from offshore to the upper Du-
plin River. Planktonic algae throughout the
Duplin River thus must use sea -salt sulfate
with an isotopic 334S value of +2117o0. How-
ever, a water sample from a small marsh
creek at low tide as water drained from the
Kcnan Field marshes had a lower -sulfate-
to-chloride ratio and a higher 634S value than
Duplin River waters (Fig. 3). This decrease
is what one would expect if sulfate -depleted
and 34S -enriched pore water were draining
into the creeks -as -the tide ebbs.
Isotopic ratios in organic matter produc-
ers --The carbon isotopic ratios in upland
plants, Spartina, and phytoplankton are dif-
ferent (Table 1). Spartina is a C-4 plant, and
photosynthesis in these plants exhibits much
less discrimination against 13C than occurs
during photosynthesis in C-3 plants. Up-
land oaks and pines on Sapelo Island have
613C values of around—290Ioo whereas Spar-
tina averages about —13%0. Both values ex-

Stable isotopes in a salt marsh
1201
Table 1. Carbon, sulfur, and nitrogen isotopic ratios (foo) and the number of samples (1) for producers in
the Sapelo Island marshes. Plankton values are from the literature.
Type of sample 611C 6"s b"N
Upland C-3 plants -29.3±1.4(4) +1.8±1.0(2) +0.4±0.9(4)
Spartina alterniJlora -12.9±0.5(10) +0.9±5.2(9) +6.0±2.1(10)
Plankton -21.3±1.1(56)* +18.8±0.6(4)t +8.6±1.0(4)$
* Gearing et al. 1984.
t Hartmann and Nielsen 1969; Kaplan et al. 1963; Kaplan and Rittenberg 1964.
t Rau 1982.
hibit small deviations. Phytoplankton take
up inorganic carbon from seawater and have
6130 values near -217o0. We did not use
seston samples from the Duplin River to
obtain the isotopic ratios for plankton given
in Table 1 because of the strong possibility
that marsh, riverine, and upland sources of
organic matter contribute to the suspended
particulate organic matter in the estuary.
Instead we used data either from estuaries
that are strongly dominated by phytoplank-
ton production or from more oceanic sta-
tions. Nonetheless, Haines (1976, 1977) has
shown that seston in the Duplin River does
have 613C values similar to the values ex-
pected for phytoplankton (Table 1).
The 6345 values for upland plants and for
Spartina at Sapelo Island are similar, but
both are much lower than the values of + 18
to +200/oo characteristic of plankton which
draw on seawater sulfate with a 634S value
of about +210/o0 (Table 1). Upland plants
take up sulfate originating either from
weathering of the geologic substrate or from
sulfate inputs in precipitation. The 634S val-
ues for Spartina from the Sapelo marshes
averaged +0.90/oo and there was no obvious
pattern of variation with site or with sam-
pling date in spite of the rather wide range
of values from -6.3 to +8.50/o0. This vari-
ability is probably related to site and sea-
sonal differences in sulfate depletion and
sulfide concentration in the root zone where
the plants are growing. The variability is of
about the same magnitude as found for
Spartina in a Cape Cod marsh (Peterson et
al. 1985).
The 615N values for upland plants are
lower than the values for Spartina and for
plankton and are close to the value for at-
mospheric nitrogen gas which is assigned a
standard value of zero (Table 1). The mean
value of +6%o found for Sapelo Island Spar-
tina is slightly but not significantly higher
than the value of +40/oo for Spartina from a
Cape Cod marsh (Peterson et al. 1985).
Plankton often have higher 615N values than
either upland plants or Spartina. For the
sake of consistency, we have used literature
values for oceanic plankton free from marsh
and upland detritus here also. However, five
seston samples taken at our transect stations
in the Duplin gave a mean 615N value of
8.0±1.4 which is not significantly different
from the literature value we have chosen
and also similar to values for fine particulate
organic matter in the northwest Gulf of
Mexico (Macko et al. 1984).
The ability to trace and to discriminate
organic matter flows from sources such as
plankton, Spartina, or upland vegetation in
the estuary depends on both the mean iso-
topic difference between any two organic
matter sources and the variation in isotopic
ratio values for each source (Table 2). For
example, it is relatively easy to discriminate
between organic carbon derived from up-
land C-3 plants and from Spartina carbon
because the mean 613C values differ by more
than 161/co and the standard deviations are
very small. On the other hand, the 634S val-
ues of upland plants and Spartina are so
similar that the sulfur isotopic ratio cannot
distinguish between them. However, the si-
multaneous use of more than one isotope
allows adequate discrimination among these
three potential organic matter sources be-
cause the sulfur isotopic values can readily
distinguish upland vegetation from plank-
ton while the carbon isotopic values easily
separate upland vegetation from Spartina.
The signal-to-noise ratios for the 615N val-
ues are not as favorable as the ratios for
carbon and sulfur (Table 2). Therefore it will
be difficult to recognize patterns of organic
matter flow with the PIN data alone.

1202
Peterson and .Howarth
Table 2. Signal-to-noise ratio for discriminating the dominant food sources in studies of estuarine food webs
with 613C, 634S, and VN tracers. The signal is the separation or difference of the mean S values. The noise is
the sum of the standard deviations of the two mean values used in each comparison. The signal-to-noise ratio
is the difference between means divided by the sum of their standard deviations. The data used to derive this
table are from Table 1.
Signal range Noise (SUI + S132)
Isotope Sample comparison (I vs. 2) (%0) (%D) Signal: noise
13C Plankton-Spartina
8.4
1.6
5.2
Upland-Spartina
16.4
1.9
8.6
Upland -plankton
8.0
2.5
3.2
34S Plankton-Spartina
17.9
5.8
3.1
Upland-Spartina
0.9
6.2
0.1
Upland -plankton
17.0
1.6
10:6
15N Plankton-Spartina
2.6
3.1
0.8
Upland-Spartina
5.6
3.0
1.9
Upland -plankton
8.2
1.9
4.3
Isotopic fractionation in trophic trans-
fers -The use of stable isotopic ratios as
tracers of organic matter flow only works if
animal assimilation of food entails either
very little isotopic fractionation or very pre-
dictable fractionation. Others have noted
that the 613C values in consumer organisms
are either similar to the values in food
(Haines and Montague 1979) or higher by
about 101co (Fry et al. 1984; Fry and Sherr
1984; McConnaughey and McRoy 1979;
Rau et al. 1983). The nitrogen 615N values
tend to be considerably higher in consumers
than in their food, and the difference can be
as much as 2-40/oo per trophic transfer (Rau
1-981; Minagawa and Wada 1984). Since the
mean difference in 615N values between
Spartina and plankton is <3%0, even a sin-
gle trophic transfer is of significance and
trophic level will be one key determinant of
consumer 6t5N values in estuaries.
Table 3 summarizes our information from
feeding experiments with gypsy moth larvae
and with brook trout (Salvelinus fontinalis).
The 613C values of trout were higher than
their food, but values for gypsy moth larvae
were lower. The 615N values increased by
>40/oo in the trout but went up by only 1.61/o0
in gypsy moth larvae. The 6315 values in-
creased in both the trout and the moth lar-
vae by about 1-1.5°/00. Table 3 includes a
set of values for the leaf hopper (Orcheli-
mum fidicinium) collected on short Spar-
tina at the Sapelo Island airport marsh in
September. It is encouraging to see -that the
leaf hopper which feeds on Spartina is very
close to the mean 613C, 615N, and 634S values
for Spartina and is well within the vari-
ability found in Spartina for each isotope.
In this paper we do not make adjustments
in marsh faunal isotopic ratios to correct for
assimilatory and metabolic fractionation by
animals. We report only raw uncorrected
data. We do, however, recognize that the
nitrogen 6 values are usually higher in con-
sumers than in their food. The range of615N
values :found in the marsh fauna should give
a rough indication of the number of trophic
levels in the food web. The trophic transfer
shifts for the other isotopes are relatively
small compared to the large isotopic differ-
ences between the important food sources
but they may not be insignificant when con-
sidering a multiple -link food chain. For ex-
ample, several trophic links involving suc-
cessive + 1%o shifts in 613C could affect
mixing model calculations for foods such as
plankton and Spartina which are separated
by roughly 80/o0. The larger isotopic differ-
ence between plankton and Spartina for sul-
fur isotopes represents a significant advan-
tage in this regard. Trophic transfers cannot
affect those comparisons as much as the car-
bon or nitrogen comparisons.
Consumers in the Sapelo marshes and es-
tuaries- We sampled in both May 1982 and
September 1983 because we wanted to in-
clude an estimate of seasonal variability in
the isotopic composition of the marsh con-
sumers. We collected as nearly as was pos-
sible from the same sites in both surveys
because our studies at Sippewissett marsh
have shown a marked effect of location on
the C, N, and S 6 values for filter feeders

Stable isotopes in a salt marsh
1203
Table 3. Trophic transfer shifts in 613C, 615N, and 634S values during assimilation of food by insects and fish.
Food Consumer Isotope shill
Consumer
Isotope
M4
Small Salvelinus fontinalis
611C
-21.7
-19.7
+2.0
Large S. fontinalis
-21.7
-20.9
+0.8
Porthetria dispar larvae
-24.5
-25.9
-1.4
Orchelimum fidicinium (Sapelo)
-12.9
-13.2
-0.3
Small S. fontinalis
615N
+8.5
+12.9
+4.4
Large S. fontinalis
+8.5
+13.2
+4.7
P. dispar larvae
+3.4
+5.0
+1.6
O. fidicinium (Sapelo)
+6.0
+7.1
+1.1
Small S. fontinalis
634S
+8.2
+9.6
+1.4
Large S. fontinalis
+8.2
+9.4
+1.2
P. dispar larvae
+6.3
+7.7
+1.4
O. frdicinium (Sapelo)
+0.9
-0.2
-1.1
(Peterson et al. 1985). The scientific and
common names of the organisms collected,
their feeding modes, and the sampling lo-
cations are given in Table 4. We have also
given the 6 values from the two sampling
trips for each producer and consumer. The
May and September values for animals for
each isotope are plotted in Figs. 4, 5, and
6. The feeding modes are not, unfortu-
nately, a fully satisfactory way of charac-
terizing the diet of marsh consumers. Most
of these animals are more or less omnivo-
rous. In fact, it would probably be difficult
if not impossible for most consumers to
completely separate benthic algae or phy-
toplankton from Spartina detritus when
feeding.
The general impression one gets from Ta-
ble 4 is that with the exception of the leaf
hopper all of the consumers fall between the
613C and 634S values for the potential foods
Spartina and phytoplankton. Feeding mode
is not a good predictor of the 634 and 613C
values; animals from each feeding mode
overlap. The 615N values tend to fall within
the range we have found for Spartina and
for plankton, but about a third of the con-
sumers have 61 -IN values of + 10%o or great-
er, higher than expected for either Spartina
or plankton. One reason is that consumers
fractionate nitrogen by from +I to + 50loo
per trophic transfer (Table 3). The larger
and higher trophic level consumers do tend
to have the highest 615N values. A second
potential reason for 615N shifts is that during
decomposition the 615N values of detrital
organic matter may change due to microbial
immobilization of nitrogen (Macko and
Zieman 1983).
Carbon isotopic ratios in consumers -The
May and September 613C values of marsh
fauna are plotted in Fig. 4 with values span-
ning the range from - 22 to -12%o. Al-
though upland plants have 613C values as
low as - 30%o, we collected no marsh con-
sumers that approach that value. The marsh
fauna fall in between the 6130 values char-
acteristic of plankton (-21%o) and Spartina
(-13%0). For most animals the seasonal dif-
ferences are fairly small, but there are large
differences between the May and September
values for several of the consumers. Litto-
rina irrorata from Kenan Field had a some-
what lower 613C value in May (-17.70/oo)
than in September (-14.60/co). Similarly,
Crassostrea virginica had a lower value in
May (-21.40/oo) than in September
(-19.1111oo). These shifts may indicate di-
etary changes over the course of the year,
changes in the isotopic composition of foods,
or perhaps metabolic shifts with production
of gametes. There are pronounced isotopic
differences between different species. For
example, the ribbed mussel and the oyster
are isotopically more similar to plankton
than to Spartina, whereas the leaf hopper
and L. irrorata are isotopically similar to
Spartina.
The most outstanding conclusion from
this figure is that most of the marsh and
estuarine macrofauna are not isotopically
similar to either plankton or to Spartina.
The 613C values fall most frequently in the
range of -18 to -15%0, which may indicate

1204
Peterson and Howarth
SON -+h••.
N
O"t.7
OOOo000
O� OOH
N QOM
(V v1 �O ni -� I
++I+++
O
I
O, M It
+++
�O 00 O -
M 1 00 t`
10 I t-- M
++++
+ ++
+ ++
M
vl -N N
O V:
Vi O� O�
OC en
NMO fn
en r-
+
+ + +
++
+
+ + + +
+
N et -. 1O O\
d' t- N
00 %O co M
M �O %0
v) v1 M
r� -: vi o o
t`
t� 00 06
Oo of G rr
%6
Ci
++I++ I
+
+++
++ f+
+++
+ If+
N
MON007Nh
SON DD
M V'DD
NvI
4-�N- OM
vi z Id
00 1 r-: 'i
M V �O%O
0;0+:�
+ +I + I + +
+ +
+ + +
+ + + +
+ + +
+
00 in O r rn
N
In 00
-+ �D n
t` V: IO
I -O t- 00
nNoviOiO+
M
a; 06
rr- tnr
%0 t 00v
vi t ID 10
+I N NN
I I I I
--� 7 %O t- O 1* t- N - t- ON 00 N 1.0 vt �O n
MONNOMb 1
INN v1%O Orr [� V7�1v1
I I I I 1 I I I I�� I I I I I I I I
O
00
N O r-
O� ' N oO
+ + +'
ro
iZ v
q u
°grow w"�. 'W>;;
U� N� U �.o UC3U "S
o p a 'J`iR: 4 �' a as
r. 3 3'a 03
^V o j 3 dA u� cqiz
n U X+
cd
ro �0G4 y �T�aai S" ayi3� ��tix uy r;�i3
oo v•
Sz
It
O O �. V 'v 'SOH Z p q�, q—Ct3 t3
C Q rn
Z '•� oy �g> Hrij y
it3 QO oo O •vC $OV
•4oy iC ?�
,
�r� tin Ol�j V x O VC�U� Q a 4 ��ti,a Ors rs:C�O o. eQoC�Qo

Stable isotopes in a salt marsh 1205
• Orchellmum fidic/n/um
O • Littorino irrorata ( Kenon) (Airport)
• O Mupi/ cepholus ( Large; Bighole Cr., Borrow
O Busycon canallculatum ( Outer Beach) Pit
•O Fundulus heteroclitus ( Small; Kenan )
O Fundulus heteroctitus ( Large; Duplin)
O Choetodipterus fober ( Dean Cr.)
• O Cofl/nectes sopldus ( Lab Cr. )
• O Po/oemonstes pap/o (Kenon Cr., Lab Cr.)
O Brevoortie tyranmus ( Dean Cr.)
O • Uco pupnex ( Kenon)
• Ocypode quadrate ( Outer Beach)
• O Mup/l cepholus ( Small; Borrow Pit)
O Anpulllo rostrata ( Adult; Dean Cr.)
• Peneous setlferus ( Adult; Outer Beach)
• O l/yonosse obsolete ( Kenon)
O Uco papilater ( Dean Cr.)
• Mug# cepholus ( Small; Duplin)
• Crossostreo vlrpin/ce ( Bighole Cr.)
O • Crassostrea v/rp/n/co ( Lob Cr.)
O • Geukenslo demisso ( Kenon)
-22 4 -21 -20 -19 -18 -17 -16 -15 -14 # -12
PLANKTON 13C DEPLETED s- 6 13C M.) — t3C ENRICHED Sportillo
Fig. 4. The 613C values for consumers in the salt marshes at Sapelo Island in May (o) and September (•).
that a mixture of plankton and Spartina de-
tritus is used or that other foods such as
benthic diatoms (613C = —16.7%, Table 4)
are important. The sulfur -oxidizing bacteria
may provide another potentially important
food resource (Howarth 1984). We have no-
ticed abundant colorless sulfur bacteria on
the marsh surface but have not been able
to obtain sufficiently large and clean sam-
ples for isotopic analysis. However, we ex-
pect that their 613C values would be similar
to plankton or even more depleted in 13C
than plankton (Peterson et al. 1980, 1986).
Nitrogen isotopes in consumers—The 615N
• Ocypode quadrate ( Outer Beach)
O Choetodipterus fober ( Dean Cr.)
O Brevoortie tyronnas ( Dean Cr.)
O• Mupil cepholus ( Large; Borrow Pit a Bighole Cr.)
O • Fundulus heterocfftus ( Smolt; Kenon)
O • Callinectes sopidus ( Lob Cr.)
• Balydlelle chrysure ( Duplin)
• Penoeus setiferus ( Outer Beach)
O Fundulus heteroefltus ( Large; Duplin)
O anyulllo rostrata ( Dean Cr.)
O • Poloernonetes pupio ( Lab Cr., Kenan)
O Busycon conaficulotum ( Outer Beach)
O • C/OSSOstlee vlrglnlco ( Lob Cr.)
• Crossostreo vlrglnico ( Bighole Cr.)
O • Llttorina irrorote ( Kenon)
O • llyonessa obsoleto (Kenon)
O • Goalmnslo demisso ( Kenon)
O • Mug# cepholus ( Small; Borrow Pit)
• Mugil cepholus ( Small; Duplin)
• Orchellmant fldlcfnlum ( Airport)
O • Uco pupnax ( Kenon)
O Uco puyllofor ( Dean Cr.)
+4 +5 +7 +8 +9 +10 +11 +12 +13
Spartina PLANKTON
t5N DEPLETED ----- b15N ('leo) --- 15N ENRICHED
Fig. 5. As Fig. 4, but for 611N values.

1206
Peterson and Howarth
O Busycon conalrculafum ( Outer Beach)
• Crossoslreo vlrpinice (Blghole Cr.)
• O Uca pupnox ( Kenon )
• O Crossostreo virpinice ( Lob Cr.1
• O L/Morino irrorafo (Kenn)
O Uco puyllalor ( Dean Cr.)
•Ocypode quodrolo ( Outer Beach)
O Brevoorlla /yronnus ( 10 cm, Dean Cr.)
• O Coll/necfes sopidus ( Lob Cr.)
0 • Mug# cephalus ( 25 cm, borrow Pit; 33 cm Bighole Cr.)
O -0 Mug# cephalus ( 3-4 cm, 7-10 cm, Duplin)
0 Choefodipforus Faber t Dean Cr.)
• O Geukens/a demisso ( Kenon)
• 0 Poloemenefes pqyio (Kenon, Lab "Cr. )
• 0 Ilyanesso obsole/o ( Kenon)
O Anguilla roslravo ( Dean Cr.)
• Peamus sedferus l Outer Beach)
0 Fundulus he/erocl/lus ( 7=9 cm, Duplin)
0 • Fundulus he/eroclilus ( 4 - 6 cm, Duplin )
i O Muyll cephalus (-Borrow Pit)
• Orchel/mum fididolum ( Airport)
O t +5 +10 +15
Sportino UPLAND $ 34_S(%a) _,._34S ENRICHED PLANKTON
PLANTS
Fig. 6. As Fig. 4, but for 634S values.
values in consumers (Fig. 5) are centered
around the range expected for phytoplank-
ton (+7 to +11°/00). Very few of the fauna
have 615N values as low as the mean value
for Spartina (+60/bo), but the ratios in both
Spartina and in plankton vary widely (Ta-
ble 1), and animal metabolism tends to shift
consumer tissue values in the positive di-
rection (Table 3; Minagawa and Wada
1984). In fact a single trophic transfer might
shift the values upward by the +2.50/oo dif-
ference between Spartina and plankton. The
most surprising result in Fig. 5 is that the
615N values for animals collected in May
are consistently lower than the values for
animals collected in September. It is not
clear to us why such a shift should occur.
Spartina also shifted seasonally from +4 to
+70/o0, but the upland plants did not. We do
not know if this apparent seasonal differ-
ence in 615N values reflects a difference in
animal foods, a change in animal metabo-
lism, a change in the marsh nitrogen cycle,
or some unknown artifact. The carbon iso-
topic or the sulfur isotopic values do not
suggest a possible answer since neither of
these isotopes exhibited a consistent shift in
values between May and September. If the
difference is real, this shift in 615N values
may indicate an important seasonal change
in the marsh nitrogen cycle.
The fiddler crabs, leaf hopper, and small
mullet have the lowest 615N values whereas
the ghost crab, spade fish, and menhaden
have the highest. In general the higher
trophic level consumers—predators and
omnivores—have higher PIN values from
+9 to + 130/co, while the deposit and deposit -
suspension feeders have lower values from
+4 to +9170.
Sulfur isotopes in consumers—The sulfur
isotopes provide a third tracer of organic
matter flow from primary producers to con-
sumers in marshes (Peterson et al. 1985).
The upland plants and Spartina have low
6345 values while plankton has a high value
of + 18 to +20%o. The consumer 634S values
range from 0 to + 1696 and most of the
marsh and estuarine consumers have 6345
values in the range from +6 to + 140loo, which
are between the values for Spartina and
plankton (Fig. 6). This pattern is very sim-
ilar to the situation for the consumer 613C
values shown in Fig. 4.
The herbivore, Orchelirnum fidicinlum, is
isotopically similar to Spartina and is known
to feed directly on Spartina (Smalley 1960).
It provides confidence in the fidelity of the
sulfur isotopic ratio during trophic transfer.
Small mullet from the borrow pits also have
relatively low 634S values, whereas mullet
of similar size from the Duplin River have

Stable isotopes in a salt marsh
much higher 634S values. It seems likely that
the diet ofthe mullet which are marsh -locked
in the borrow pits is quite different from
that of mullet in the Duplin River. The most
straightforward explanation (that the mullet
in the borrow pits use more Spartina detri-
tus while the mullet in the Duplin use more
plankton) is only weakly supported by the
carbon isotopic data in Fig. 4. The organ-
isms with relatively high 6345 values include
the welk (Busycon canaliculatum) from the
outer barrier beach area, fiddler crabs, oys-
ters, and the marsh periwinkle (L. irrorata).
It seems unlikely that the fiddler crabs and
Littorina actually are drawing most of their
organic sulfur from plankton. Probably they
are feeding on benthic flora and on epi-
phytes that also obtain most of their sulfur
from seawater sulfate.
Figure 7 summarizes the data on the dis-
tribution of isotopic values for C, N, and S
in both the spring and fall collections of
consumers. These plots exhibit no resem-
blance to the distributions expected if either
Spartina or algae were the sole important
or dominant organic matter source. The
general conclusion from the carbon and sul-
fur data is that very few of the marsh and
estuarine animals assimilate either Spartina
or plankton exclusively. The more probable
situation seems to be that the consumers
use a mixture of organic matter derived from
both plankton and Spartina. The suspen-
sion feeders tend to be more similar to
plankton, and the deposit and deposit -sus-
pension feeders are more similar to Spar-
tina. The nitrogen isotopic data seem to in-
dicate that most of the marsh consumers
are more similar to the 615N values expected
for plankton. However, since trophic trans-
fers tend to shift the consumer B15N value
upward from +I to +4%o per trophic trans-
fer (Table 3), the foods of these organisms
could well be Spartina or a mixture of Spar-
tina and plankton rather than mainly plank-
ton. Thus the nitrogen isotopic data do not
conflict with the carbon and sulfur isotopic
information.
Producer -consumer relationships—The
dual isotope approach— When there are three
or more potentially important sources of
food for animals, it is not possible to dis-
criminate between the different sources of
m
0
4-
0
Cr
W
m
Z
Z
U,
4
O
-22 t -20 -18 -16 -14 1-12
PLANKTON 6"C (%.) Spartina
1207
Sporting 615N
sporting V's (%.) PLANKTON
Fig. 7. Histograms of the S' 3C, 615N, and 6145 values
for consumers in the Sapelo Island marshes, including
all determinations from the May and September col-
lections.
organic matter through use of a single tracer
unless the isotopic ratios in consumers clus-
ter about one end -member. However, a
combination of two or more isotopes should
make it possible to identify the important
organic matter sources more unambiguous-
ly (Rau et al. 1981; Fry 1984; Peterson et
al. 1985). The relationship between the
marsh consumers and their potential foods
on a diagram of 613C vs. 6345 is shown in
Figs. 8 (May 1982) and 9 (September 1983).
The consumer organisms are numbered in
order of lowest to highest 634S values. The
locations of the potential foods are identi-
fied as rectangles indicating the mean ± 1
SD for the potentially important foods (Ta-
ble 1).
We were not able to collect sufficient ben-
thic diatoms or sulfur -oxidizing bacteria to
obtain sulfur isotopic ratios for these po-
tential food sources. Our data on sulfur ox-
idizers from a Cape Cod salt marsh indi-
cated a mean 613C value of around —200/oo
and a mean S34S value of —12%0 (Peterson
et al. 1986). None of the consumers in Figs.

1208
Peterson and Howarth
+20-
Iv
PLANKTONEe`,
w
+20
PLANKTON ,Fq,
czi +18
,
+18
�,
,
+16Z
Iz
�, 16
+16-
rr
�i ,
rn +14-
�`
Iu
v
+14-
��� 16 14
M
+12
�;
12 SII
1415,
1113` io
M
+12-
/r is IS ���
t1D
'
�/ i .911
679, ,,
i1D
' r
i i .` 11 12`
101 �.
8 +8
�;
�
�� s 34 .�
+8
6 �
a l'? ,
1`4
+4
2
w
t4
-'
2
,
0
_`
0
Sporting
w
W -2
UPLAND -
PLANTS
��
W
-2-PLANTS
UPLAND - Sp01#11
-
-_
-
o -6
N
-6
a
M
-30 -20 -10
13C DEPLETED --- S 13C (%) - 13C ENRICHED
Fig. 8. Sulfur isotopic ratios of marsh consumers
plotted as a function of 613C in relation to the mean
ratios in the potential organic matter sources during
May 1982. Key to consumers: 1. Mugil cephalus (small,
borrow pit); 2. Fundulus heteroclitus (medium, Duplin
River); 3. F. heteroclitus (large, Duplin River); 4. Mugil
cephalus (large, borrow pit); 5. Anguilla rostrata (Dean
Creek); 6. Ilyanassa obsoleta (Kenan Field); 7. Palae-
monetes pugio (Lab Creek); 8. Geukensia demissa (Ke-
nan Field); 9. Chaetodipterus faber (Dean Creek); 10.
Callinectes sapidus (Dean Creek); 11. Uca pugnax
(Kenan Field); 12. Crassostrea virginica (Lab Creek);
13. Brevoortia tyrannus (Dean Creek); 14. Uca pugi-
lator (Dean Creek); 15. Littorina irrorata (Kenan Field);
16. Busvcon canaliculatum (outer beach); 17. M. ceph-
alus (juvenile mullet, Kenan Field).
8 and 9 appear to be similar to these values,
but the isotopic composition of sulfur -oxi-
dizing bacteria in the Sapelo marshes is un-
known and may be different from Sippe-
wissett. Pyrite which might be oxidized had
634S values between -10 and -200/oo, but
the sulfides at the short Spartina site were
+ 100/oo near the surface. Depending on the
source of the reduced sulfur, sulfur bacteria
may have a wide range of 634S values and
we cannot rule out their importance on the
basis of the isotopic data obtained thus far.
Also, some bacteria may be mixotrophic,
taking up DOC as well as fixing DIC che-
mosynthetically.
The macroconsumers are distributed sys-
tematically within or close to the band of
-30 -20 -10
131: DEPLETED 13C (%.) I i3C ENRICHED
Fig. 9. As Fig. 8, but during September 1983. Key
to consumers: 1. Orcheli►num fidicinium (airport
marsh); 2. Mugil cephalus (small, borrow pit); 3. Fun-
dulus heteroclitus (small, Kenan Field); 4. Palaemo-
netes pugio (Kenan Field); S. Callinectes sapidus (Lab
Creck);-6. Littorina irrorata (Kenan Field); 7. Penaeus
setiferus (outer beach); S. Ilyanassa obsoleta (Kenan
Field); 9. Bairdiella chrysura (Duplin River); 10. Geu-
kensia demissa (Kenan Field); 11. M. cephalus (small,
Duplin River); 12. M. cephalus (large, Bighole Creek);
13. Ocypode quadrata (outer beach); 14. Uca pugnax
(Kenan Field); 15. Crassostrea virginica (Lab Creek);
16. C. virginica (Bighole Creek).
carbon and sulfur S values expected if these
organisms derive most of their nutrition
from Spartina, from phytoplankton, or most
often from a mixture of the two. This con-
clusion is supported equally by the May 1982
and September 1983 collections (Figs. 8, 9).
There is no isotopic Evidence in these sam-
ples for any important influence of organic
matter derived from the uplands. This find-
ing is consistent with the facts that the Du-
plin River is tidal but has no large input of
freshwater and that this marsh is fairly re-
mote from large rivers. For these reasons
one does not expect a large upland organic
matter input here. A couple of data points
appear to depart from the expected pattern.
For example, in Fig. 8, C. virginica (No. 12
on the figure) appears to have a lower 6345
value than one would expect for a consumer
that is similar to plankton in 613C. The op-

Stable isotopes in a salt marsh
posite situation also occurred; B. canalicu-
latum (No. 16) had a higher 6345 value than
expected for an organism with a 613C value
of—15.5°Ioo. Perhaps for these organisms the
isotopic classification of the potential foods
is incomplete. The welk (Busycon) was col-
lected on the outer barrier beach and we
thought it would be a marine rather than a
marsh organism. It does fall at the marine
plankton end of the spectrum, but both the
carbon and sulfur 6 values indicated that the
welk derives a significant proportion of its
food from organic matter sources that are
depleted in 34S and enriched. in 13C relative
to plankton. The most likely source is de-
tritus derived from the Spartina marshes,
but benthic algae or sulfur bacteria cannot
be ruled out because their 6345 values are
unknown and perhaps explain the low 6345
value. The connection would of course be
indirect because the welk is a predator.
The September collection also contained
one unexpected point. The mullet (No. 2)
from the borrow pits had a lower 613C value
than expected for an organism with a 634S
value of +2%o. Mullet from other sites (No.
11 and 12) gave values within the range
expected for organisms using organic matter
from both plankton and Spartina. We do
not understand how the mullet in the bor-
row pits could be so depleted in 6345 relative
to plankton and simultaneously depleted in
613C relative to Spartina. One explanation
is that a significant fraction of its diet is
derived from detritus from upland plants,
but it is unlikely because the borrow pits
are not surrounded by forest. We can also
not eliminate the possibility that the use of
benthic algae or sulfur bacteria might cause
this shift in isotopic values. Finally, one must
keep in mind the possibility that consumers
may derive different fractions of their ele-
mental requirements from foods with con-
trasting elemental compositions or availa-
bilities.
It is, of course, possible to plot C vs. N
and N vs. S in addition to the C vs. S dual
isotopic plot. The amount of information
gained from these other approaches is not
as much as from the C vs. S plots. Figure
10 shows the position of marsh consumers
on the 61SN vs. 613C diagram for the Sep-
tember collection. On this figure the distri-
1209
-30 -20 -10
t3C DEPLETED 13C ENRICHED
Fig. 10. Carbon and nitrogen isotopic ratios ofcon-
sumers in relation to the ratios in potential organic
matter sources in September 1983. Key to consumers:
1. Orchelimum ftdicinium (airport marsh); 2. Mugil
cephalus (small, borrow pit); 3. Fundulus heteroclitus
(small, Kenan Field); 4. Palaemonetes pugio (Kenan
Field); 5. Callinectes sapidus (Dean Creek); 6. Littorina
irrorata (Kenan Field); 7. Penaeus setiferus (outer
beach); 8. Ilyanassa obsoleta (Kenan Field); 9. Bair-
diella chrysura (Duplin River); 10. Geukensia demissa
(Kenan Field); 11. M. cephalus (small, Duplin River);
12. M. cephalus (large, Bighole Creek); 13. Ocypode
quadrata (outer beach); 14. Uca pugnax (Kenan Field);
15. Crassostrea virginica (Lab Creek); 6. C. virginica
(Bighole Creek).
bution of consumers describes a vertical
band of values intermediate between plank-
ton and Spartina on the 613C axis and rang-
ing from Spartina to greater than plankton
on the 615N axis. The distribution supports
the conclusion that the marsh consumers
derive most of their organic matter from a
mixture of Spartina and plankton. The ver-
tical extension of the 615N values well be-
yond the ratios in the foods is probably an
example of the positive 6' SN shifts that oc-
cur in metabolism of organic nitrogen.
If we look at the distribution of 61 IN val-
ues in the September collections (Figs. 5,
10), we can see that there is an easily rec-
ognized relation between trophic level of
each consumer and the 615N values. In the-
ory the assignment of trophic level based
+ 12
13
3
W
t11
12
a
U_
+10
9
c
W
+9
7
PLANKTON ,`15 4 -.
+8-
; L 16
+7
6'
; `; ro1B2 •' 1
+6
�' 114
+5-
+4-
4
+3
+3
+2
o
J
a
Wo
0
'
.'
Z
_2
UPLAND
PLANTS
1209
-30 -20 -10
t3C DEPLETED 13C ENRICHED
Fig. 10. Carbon and nitrogen isotopic ratios ofcon-
sumers in relation to the ratios in potential organic
matter sources in September 1983. Key to consumers:
1. Orchelimum ftdicinium (airport marsh); 2. Mugil
cephalus (small, borrow pit); 3. Fundulus heteroclitus
(small, Kenan Field); 4. Palaemonetes pugio (Kenan
Field); 5. Callinectes sapidus (Dean Creek); 6. Littorina
irrorata (Kenan Field); 7. Penaeus setiferus (outer
beach); 8. Ilyanassa obsoleta (Kenan Field); 9. Bair-
diella chrysura (Duplin River); 10. Geukensia demissa
(Kenan Field); 11. M. cephalus (small, Duplin River);
12. M. cephalus (large, Bighole Creek); 13. Ocypode
quadrata (outer beach); 14. Uca pugnax (Kenan Field);
15. Crassostrea virginica (Lab Creek); 6. C. virginica
(Bighole Creek).
bution of consumers describes a vertical
band of values intermediate between plank-
ton and Spartina on the 613C axis and rang-
ing from Spartina to greater than plankton
on the 615N axis. The distribution supports
the conclusion that the marsh consumers
derive most of their organic matter from a
mixture of Spartina and plankton. The ver-
tical extension of the 615N values well be-
yond the ratios in the foods is probably an
example of the positive 6' SN shifts that oc-
cur in metabolism of organic nitrogen.
If we look at the distribution of 61 IN val-
ues in the September collections (Figs. 5,
10), we can see that there is an easily rec-
ognized relation between trophic level of
each consumer and the 615N values. In the-
ory the assignment of trophic level based

1210
Peterson and Howarth
on an organism's PIN value should take
into account the 615N value for the primary
producer at the base of each trophic pyra-
mid, but in our case this is difficult because
almost all of the 6130 values in Fig. 10 are
intermediate between the expected values
for plankton and Spartina and because 615N
values of detritus may not reflect the parent
plant material. For simplicity we could
probably assign a PIN value of about +6
or +7%o for this mixture of foods. If we
assume that each trophic transfer produces
an increase in 615N values of about +20/oo,
then we can classify consumers as detriti-
vores or herbivores with 615N values of 6
to 8.417oo and predators with 615N values of
> 8.4%o.
The detritivores and herbivores include,
in order of increasing 615N value: Uca pug-
nax, O. fidicinium, Mugil cephalus (small),
Geukensia demissa, Ilyanassa obsoleta, L.
irrorata and C. virginica. These animals are
known to feed mainly on either Spartina
alive or as detritus, benthic diatoms, plank-
ton or seston, and epiphytes (Montague et
al. 1981). The carnivores or predators in-
clude Palaemonetes pugio, Penaeus setife-
rus, Bairdiella chrysura, Callinectes sapidus,
Fundulus heteroclitus, M. cephalus (large),
and Ocypode quadrata. These organisms are
on average clearly higher in the .food web
than the ones with lower 615N values, al-
though there is a tendency for many marsh
and estuarine animals to be omnivorous,
and thus we must expect overlap in diet and
isotopic composition between the so-called
herbivores and predators. In fact, as we saw
in Fig. 7, there is no indication of any bi-
modal distribution in the 6' IN values. There
is a single broad histogram of 615N values
peaking at roughly +8°loo, which is near our
arbitrary dividing line for herbivores and
carnivores. More refined interpretation of
615N values for estuarine fauna will require
more focused work with one or several
species where both the 615N value of the
food resources at the base of the food web
and the metabolic fractionation in con-
sumers are well known.
Conclusions
The results of both our May and Septem-
ber collections of Sapelo Island marsh and
estuarine consumers support the contention
of Haines (1977) that phytoplankton are im-
portant contributors to estuarine food webs
at Sapelo. All of our isotopic data are con-
sistent with a roughly equal contribution by
Spartina and by .phytoplankton plus per-
haps benthic algae. Some organisms such as
the leaf hopper (O. ftdicinium) are isotopi-
cally very similar to Spartina, while others
such as the oyster (C. virginica) are more
similar to plankton (Figs. -8, 9). However,
most are intermediate between the values
for Spartina and plankton for 613C, 634S, and
61SN when the metabolic shift in the nitro-
gen isotopic values is taken into account.
This observation does not mean that all the
organisms assimilate the same thing, but
they must eat mixtures of varying propor-
tions. Also, organisms with similar carbon
or sulfur isotopic ratios can be quite differ-
ent in their 615N values, probably because
they feed at different trophic levels (Fig. 10).
The isotopic data also support the view
that Spartina detritus is important as a food
resource for virtually all the marsh and es-
tuarine macrofauna. However, although
Spartina dominates carbon and energy flow
in the marsh ecosystem at the level of car-
bon fixation (Pomeroy et al. 1981), Spartina
is only roughly equal to algae at the level of
assimilation by the macroconsumers. The
refractory nature of Spartina detritus re-
quires extensive processing by the microbial
community with.large respiratoryCO2 losS-
es before some small fraction is available to
the macrofauna.
The broad spectrum of C and S isotopic
values that we find in the marsh and estu-
arine macroconsumers :is in marked con-
trast to the narrow isotopic distributions
found in food webs where a single food re-
source is dominant. For example in marine
offshore food webs there is a tight clustering
of 613C values for each trophic level, al-
though a small 13C enrichment occurs with
trophic transfer (Fry and Sherr 1984). A
similar situation occurs in lakes. Only a nar-
row range of 6130 and 634S values was found
for native fishes in Fayetteville Green Lake
and Round Lake (Fry 1986). A broad iso-
topic spectrum implies a variety of food re-
sources.
We find no evidence from the multiple

Stable isotopes in a salt marsh
1211
Table 5. Comparison of mean 611C, 615N, and 6145 values for Spartina and fauna at the Sapelo Island marsh
and at Sippewissett marsh. The mean differences are compared for the fauna only, not including Spartina
alterniflora.
isotopic approach to suggest an important
input of organic matter either directly from
the neighboring uplands or from more dis-
tant rivers to the Duplin River and the ad-
jacent marshes at Sapelo. Although it may
be difficult or impossible to rule out the pos-
sibility that a portion of the dissolved or
particulate organic carbon in the water may
originate from river inputs, there is no evi-
dence that this material contributes in any
measurable way to the isotopic composition
of consumers in the marsh and estuary. This
finding is similar to our conclusion that up-
land organic matter is not an important food
resource for fauna in Great Sippewissett
Marsh on Cape Cod. Sampling nearer the
mouth of several major rivers should be
carried out to explore this issue further and
to define the realm of influence and the uti-
lization by consumers in estuaries of riv-
erborne organic detritus and dissolved or-
ganic matter.
In general there does not seem to be a
large seasonal isotopic shift, as indicated by
our comparisons of May and September
macrofaunal collections. The nitrogen iso-
tope may be an exception as the September
values for consumers were about 2%o higher
than the May values. On the other hand,
there do appear to be some important dif-
ferences in isotopic composition that are re-
lated to location. For example, M. cephalus
collected from the borrow pits was isoto-
pically more similar to Spartina than were
mullet of similar size collected from the tid-
al creeks at Kenan Field.
Our isotopic data from the Sapelo Island
marsh make an interesting comparison with
similar data from Great Sippewissett Marsh
(Table 5; Peterson et al. 1986). Spartina from
Sapelo Island has about the same 613C value
as Spartina from Sippewissett, but both the
WN and 6345 values for the Sapelo Island
Spartina are higher. The higher 6345 value
at Sapelo is perhaps related to the greater
sulfate depletion in the Sapelo sediments
(Figs. 1, 2). The fauna at Sapelo Island how-
ever have lower 6t3C values than the Sip-
pewissett animals, suggesting that sources
of organic matter other than Spartina, such
as plankton and benthic algae, are relatively
more important in the Georgia marshes than
at Sippewissett. Nitrogen isotopic ratios av-
erage about the same in the fauna from Sa-
pelo and Sippewissett. However, the 634S
values for Sapelo Island fauna average about
Isotope
Sippewissett
sapdo
Difference
S. allerniflora
613C
-13.1
-12.9
(+0.2)
11yanassa obsoleta
-11.0
-17.8
-6.8
Fundulus heteroditus
-12.5
-15.6
-3.1
Callinectes lapidus
-13.3
-16.2
-2.9
Geukensia demissa
-15.2
-19.8
-4.6
Uca pugnax
-16.2
-16.8
-0.6
-3.6±2.3
S. alterniflora
615N
+3.8
+6.0
(+2.2)
1. obsoleta
+4.6
+7.2
+2.6
F. heteroclitus
+10.6
+9.2
-1.4
C. sapidus
+9.4
+10.0
+0.6
G. demissa
+7.3
+6.5
-0.8
U. pugnax
+4.3
+4.9
+0.6
+0.3±1.5
S. alterniflora
634S
-2.4
+0.9
(+3.3)
L obsoleta
+5.6
+9.0
+3.4
F. heleroclitus
+1.0
+7.9
+6.9
C. sapidus
+3.8
+9.8
+6.0
G. demissa
+0.7
+9.8
+9.1
U. pugnax
+8.6
+ 13.0
+4.4
+6.0±2.2
isotopic approach to suggest an important
input of organic matter either directly from
the neighboring uplands or from more dis-
tant rivers to the Duplin River and the ad-
jacent marshes at Sapelo. Although it may
be difficult or impossible to rule out the pos-
sibility that a portion of the dissolved or
particulate organic carbon in the water may
originate from river inputs, there is no evi-
dence that this material contributes in any
measurable way to the isotopic composition
of consumers in the marsh and estuary. This
finding is similar to our conclusion that up-
land organic matter is not an important food
resource for fauna in Great Sippewissett
Marsh on Cape Cod. Sampling nearer the
mouth of several major rivers should be
carried out to explore this issue further and
to define the realm of influence and the uti-
lization by consumers in estuaries of riv-
erborne organic detritus and dissolved or-
ganic matter.
In general there does not seem to be a
large seasonal isotopic shift, as indicated by
our comparisons of May and September
macrofaunal collections. The nitrogen iso-
tope may be an exception as the September
values for consumers were about 2%o higher
than the May values. On the other hand,
there do appear to be some important dif-
ferences in isotopic composition that are re-
lated to location. For example, M. cephalus
collected from the borrow pits was isoto-
pically more similar to Spartina than were
mullet of similar size collected from the tid-
al creeks at Kenan Field.
Our isotopic data from the Sapelo Island
marsh make an interesting comparison with
similar data from Great Sippewissett Marsh
(Table 5; Peterson et al. 1986). Spartina from
Sapelo Island has about the same 613C value
as Spartina from Sippewissett, but both the
WN and 6345 values for the Sapelo Island
Spartina are higher. The higher 6345 value
at Sapelo is perhaps related to the greater
sulfate depletion in the Sapelo sediments
(Figs. 1, 2). The fauna at Sapelo Island how-
ever have lower 6t3C values than the Sip-
pewissett animals, suggesting that sources
of organic matter other than Spartina, such
as plankton and benthic algae, are relatively
more important in the Georgia marshes than
at Sippewissett. Nitrogen isotopic ratios av-
erage about the same in the fauna from Sa-
pelo and Sippewissett. However, the 634S
values for Sapelo Island fauna average about

1212
Peterson and Howarth
+6°/co heavier than for the same species from
Sippewissett. Since Spartina from Sapelo is
only about +3%o heavier than at Sippe-
wissett, the shift in the animals may result
from .a greater dependence on plankton at
Sapelo in addition to the difference in Spar-
tina. This interpretation agrees with the
conclusion from the 613C value comparison.
It is not clear why Spartina should be
relatively more important at Sippewissett
than at Sapelo. Perhaps the productivity and
availability of plankton or benthic algae rel-
ative to Spartina is greater at Sapelo Island.
Or perhaps Spartina detritus is more avail-
able in Massachusetts since it is weighted
down by snow and ice in the winter and
kept wet thereafter on the marsh surface; in
Georgia, standing dead grasses decompose
in place. Because of the variability among
individuals and among species, these com-
parisons cannot be given great weight but
they point to a potentially fruitful means of
comparing marshes and one which could
provide an integrated measure of the dif-
ferences in how various marshes function.
References
ALTABET, M. A., AND J. J. MCCARTHY. 1985. Tem-
poral and spatial variations in the natural abun-
dance of 'IN in PON from a warm -core ring. Deep -
Sea Res. 32: 755-772.
CARLSON, P. R., JR., AND J. FORREST. 1982. Uptake
of dissolved sulfide by Spartina alterniJlora: Evi-
dence from natural sulfur isotope abundance ra-
tios. Science 216: 633-636.
COTLOVE, M. D., H. V. TRANTHAM, AND R. L. BowMAN.
1958. An instrument and method for automatic,
rapid, accurate and sensitive titration of chloride
in biologic samples. J. Lab. Clin. Med. 51: 461-
468.
FRY, B. 1984. Fish and shrimp migrations in the
northern Gulf of Mexico analyzed using stable C,
N, and S isotope ratios. Fish. Bull. 81: 789-801.
. 1986. Sources of carbon and sulfur nutrition
for consumers in the meromictic lakes of New
York State. Limnol. Oceanogr. 31: 79-88.
-, R. K. ANDERSON, L. ENTZEROTH, J. L. BIRD,
AND P. L. PARKER. 1984. 13C enrichment and
oceanic food web structure in the northwestern
Gulf of Mexico. Contrib. Mar. Sci. 27: 49-63.
, AND E. B. SHERR. 1984. 613C measurements
as indicators of carbon flow in marine and fresh-
water ecosystems. Contrib. Mar. Sci. 27: 13-47.
GEARING, J. N., P. J. GEARING, D. T. RUDNICK, A. G.
REQUEJO, AND M. J. HUTCHINS. 1984. Isotopic
variability of organic carbon in a phytoplankton-
based temperate estuary. Geochim. Cosmochim.
Acta 48: 1089-1098.
HAINES, E. B. 1976. Stable carbon isotope ratios in
the biota, soil and tidal water of a Georgia salt
marsh. Estuarine Coastal Mar. Sci. 4: 609-616.
1977. The origins of detritus in Georgia salt
marsh estuaries. Oikos 29: 254-260.
AND C. L. MONTAGUE. 1979. Food sources
of estuarine invertebrates analyzed using 13C/11C
ratios. Ecology 60: 48-56.
HARTMANN, M. VON, AND .H. NIELSEN. 1969. 634S-
Werte in rezenten Meeressedimenten and ihre
Deutung am Beispiel ciniger Sedimentprofile aus
der westlichen Ostsee. Geol. Rundsch. 58: 621-
655.
HOPKINSON, C. R., JR., AND F. A. HOFFMAN. 1984.
The estuary extended -a recipient -system study
of estuarine outwelling in Georgia, p. 313-330. In
V. S. Kennedy [ed.], The estuary as a filter. Aca-
demic.
HOWARTH, R. W. 1978. A rapid and precise method
for determining sulfate in seawater, estuarine
waters, and sediment pore waters. Limnol. Ocean-
ogr. 23: 1066-1069.
1984. The ecological significance of sulfur in
the energy dynamics of salt marsh and coastal ma-
rine sediments. Biogeochemistry 1: 5-27.
AND A. GIBLIN. 1983. Sulfate reduction in
the salt marshes at Sapelo Island, Georgia. Limnol.
Oceanogr. 28:-70-82.
KAPLAN, 1. R., K. O. EMERY, AND S. C. RITTENBERG.
1963. The distribution and isotopic abundance
of sulfur in recent marine sediments off southern
California. Geochim. Cosmochim. Acta 27: 297-
331.
AND S. C. RITTENBERG. 1964. Microbiolog-
ical fractionation of sulfur isotopes. J. Gen. Mi-
crobiol. 34: 195-212.
MCCONNAUGHEY, T., AND C. P. MCROY. 1979. Food -
web structure and the fractionation of carbon iso-
topes in the Bering Sea. Mar. Biol. 33: 257-262.
MACKO, S. A., L. ENTZEROTH, AND P. L. PARKER. 1984.
Regional differences in nitrogen and carbon iso-
topes on the continental shelf of the Gulf of Mex-
ico. Naturwissenschaften 71: 374.
AND J. ZIEMAN. 1983. Stable isotope com-
position and amino acid analysis of estuarine plant
litter undergoing decomposition [abstract]. Estu-
aries 6: 304.
MARIOITI, A., R. LETOLLE, AND E. SHERR. 1983. Dis-
tribution of stable nitrogen isotopes in a salt marsh
estuary [abstract]. Estuaries 6: 304.
MINAGAWA, M., AND E. WADA. 1984. Stcpwlse en-
richment of ISN along food chains: Further evi-
dence and the relation between 615N and animal
age.. Geochim. Cosmochim. Acta 48: 1135-1140.
MONTAGUE, C. L., S. M. BUNKER, E. B. HAINES, M. L.
PACE, AND R. L. WETZEL. 198 L Aquatic mac-
roconsumers, P. 69-85. In L. R. Pomeroy and R.
G. Wiegert [eds.], The ecology of a salt marsh.
Springer.
NIxoN, S. W. 1980. Between coastal marshes and
coastal waters -a review of twenty years of spec-
ulation and research on the role of salt marshes in
estuarine productivity and water chemistry, p. 437-
525. In E. Hamilton and K. B. MacDonald [eds.],
Estuarine and wetland processes. Plenum.

Stable isotopes in a salt marsh
ODuM, E. P. 1980. The status of three ecosystem -
level hypotheses regarding salt marsh estuaries:
Tidal subsidy, outwelling, and detritus -based food
chains, p. 485-495. In V. S. Kennedy [ed.], Es-
tuarine perspectives. Academic.
, AND A. A. DE LA CRUZ. 1967. Particulate
organic detritus in a Georgia salt marsh -estuarine
ecosystem, p. 383-388. In G. H. Laulf [ed.], Es-
tuaries. Am. Assoc. Adv. Sci. Pub]. 83.
PETERSON, B. J., R. W. HOWARTH, AND R. H. GARRITT.
1985. Multiple stable isotopes used to trace the
flow of organic matter in estuarine food webs. Sci-
ence 227: 1361-1363.
, AND . 1986. Sulfur and carbon
isotopes as tracers of salt -marsh organic matter
flow. Ecology 67: 865-874.
F. LlrscHULTZ, AND D. ASHENDORF.
1980. Salt marsh detritus: An alternative inter-
pretation of stable carbon isotope ratios and the
fate of Spartina alternfJlora. Oikos 34: 173-177.
POMEROY, L. R., AND R. G. WIEGERT (EDS.]. 1981.
The ecology of a salt marsh. Springer.
, AND OTHERS. 1981. Primary production, p.
39-68. In L. R. Pomeroy and R. G. Wiegert [eds.],
The ecology of a salt marsh. Springer.
RAu, G. H. 1981. Low 15N/10N in hydrothermal vent
animals: Ecological implications. Nature 289:484-
485.
'1982. The relationship between trophic level
and stable isotopes of carbon and nitrogen, p. 143-
1213
148. In Coastal water research project biennial
report for 1981-1982. SCCWRP Tech. Rep. South.
Calif. Coastal Water Res. Project.
A. J. MEARNS, AND D. R. YoUNo. 1981. Dif-
ferences in animal 13C, 'IN, and D abundance be-
tween a polluted and an unpolluted coastal site:
Likely indicators of sewage uptake by a marine
food web. Estuarine Coastal Shelf Sci. 13: 701-
707.
,ANDOTHERS. 1983. Animal 11C/12C correlates
with trophic level in pelagic food webs. Ecology
64: 1314-1318.
REEBURGH, W.S. 1967. An improved interstitial water
sampler. Limnol. Oceanogr. 12: 163-165.
SMALLEY, A. F. 1960. Energy flow of a saltmarsh
grasshopper population. Ecology 41: 672-677.
TEAL, J. M. 1962. Energy flow in the salt marsh eco-
system of Georgia. Ecology 43: 614-624.
WADA, E. 1980. Nitrogen isotope fractionation and
its significance in biogeochemical processes oc-
curring in marine environments, p. 375-398. In
Isotope marine chemistry. Uchida Rokakuho.
ZHABINA, N. N., AND I. I. VOLKOv. 1978. A method
of determination of various sulfur compounds in
sea sediments and rocks, p. 735-746. In W. E.
Krumbein [ed.], Environmental biogeochemistry
and gcomicrobiology. V. 3. Ann Arbor Sci.
Submitted. • 2 May 1986
Accepted: 10 July 1986

iolz-ll2Dco I � � . �,-}-a,�c� � � cpm . C�e..a_5 (�• sD mss, )
a RA
F—
t"t� 3= �\A 4kttr-4 CA-C-J�
CW' b -e- Qix/, 'k v, 4-W-o�cp
QCA (re_ Wu Q�Q
A
( o qen 0�K e SvM - C-
A
A l v,�/ 4-uw-
q�
(a-" Ca
U3 cu d�, 1Z3 d re c
-o\vA-- ,�-�q 9D?c S f P� , 2 -,LCA-..
f�A'A un e iron, +-
=C cla-,A -OVA 0
it
���5 1ol�ncted ► �3
-�Q
P4
P&k4 n, rov , -
-ry -.� -i�o