Science of the Total Environment 487 (2014) 557–564
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
Isotopic evidence for anthropogenic impacts on aquatic food web
dynamics and mercury cycling in a subtropical wetland ecosystem in
the US
Yang Wang a,⁎, Binhe Gu b, Ming-Kuo Lee c, Shijun Jiang d,⁎, Yingfeng Xu a
a
Department of Earth, Ocean & Atmospheric Science, Florida State University and National High Magnetic Field Laboratory, Tallahassee, FL 32306–4100, USA
South Florida Water Management District, West Palm Beach, FL 33406, USA
c
Department of Geology and Geography, Auburn University, Auburn, AL 36839, USA
d
Institute of Hydrobiology/Laboratory of Eutrophication and Red Tide Prevention of Guangdong Higher Education Institutes, Jinan University, Guangzhou, Guangdong 510632, China
b
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• δ13C, Δ14C, and δ15N of fishes from a
subtropical wetland ecosystem were
analyzed.
• Data revealed impacts of land-use
change on food chains and Hg bioaccumulation.
• In reference wetlands, fishes relied on
modern primary production.
• In impacted wetlands, old peat was a
significant C source for fishes.
• Data suggest a shorten food chain and
less Hg bioaccumulation in impacted
areas.
a r t i c l e
i n f o
Article history:
Received 20 December 2013
Received in revised form 12 April 2014
Accepted 16 April 2014
Available online xxxx
Editor: Mae Mae Sexauer Gustin
Keywords:
Stable isotopes
Radiocarbon
Fish
Mercury
Food web
Bioaccumulation
Everglades
a b s t r a c t
Quantifying and predicting the food web consequences of anthropogenic changes is difficult using traditional
methods (based on gut content analysis) because natural food webs are variable and complex. Here, stable
and radioactive carbon isotopes are used, in conjunction with nitrogen isotopes and mercury (Hg) concentration
data, to document the effects of land-use change on food webs and Hg bioaccumulation in the Everglades – a
subtropical wetland ecosystem in the US. Isotopic signatures of largemouth bass and sunfish in reference
(relatively pristine) wetlands indicate reliance on the food supply of modern primary production within the wetland. In contrast, both fish in areas impacted by agricultural runoff had radiocarbon ages as old as 540 years B.P.,
and larger isotopic variability than counterparts in reference wetlands, reflecting differences in the food web
between impacted and reference wetlands. Consistent with this difference, particulate and dissolved organic
matter in impacted areas had old radiocarbon ages (N600 years B.P.), indicating that old carbon derived from historic peat deposits in the Everglades Agricultural Area was passed along the food chain to consumers. Significant
radiocarbon deficiencies in largemouth bass and sunfish, relative to mosquitofish, in impacted areas most likely
indicate a reduced dependence on small fish. Furthermore, largemouth bass and sunfish from impacted areas had
much lower Hg contents than those from reference wetlands. Taken together, these data suggest a shift toward
lower trophic levels and a possible reduction in mercury methylation in impacted wetlands. Our study provides
⁎ Corresponding authors. Tel.: +1 850 644 1121; fax: +1 850 644 0827.
E-mail addresses: [email protected] (Y. Wang), [email protected] (S. Jiang).
http://dx.doi.org/10.1016/j.scitotenv.2014.04.060
0048-9697/© 2014 Elsevier B.V. All rights reserved.
558
Y. Wang et al. / Science of the Total Environment 487 (2014) 557–564
clear evidence that hydrological modification and land-use change in the Everglades have changed the system
from one driven primarily by in-situ productivity to one that is partially dependent on allochthonous carbon
input from peat soils in the agricultural area and altered the Hg biogeochemical cycle in the wetlands. The results
have implications for the restoration and management of wetland ecosystems.
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
Human activities have significantly altered Earth’s atmosphere and
hydrosphere, posing substantial risks to ecosystem health and biodiversity (IPCC, 2007; Vitousek et al., 1996; Zaret and Paine, 1973). The Everglades in south Florida (U.S.A.) is a large, complex subtropical wetland
ecosystem that has experienced unprecedented anthropogenic modification over the last century. More than half of the original Everglades
wetland has been drained for agricultural and urban use, resulting
in significant habitat degradation and deterioration of water quality
(Childers et al., 2003; McCormick et al., 2001). Although the effects of
hydrological modification and nutrient loading on plant communities
have been well documented (e.g., Jickells, 1998; McCormick et al.,
2001), little is known about how these changes have affected the energy
base that supports the aquatic food web in this unique, subtropical wetland ecosystem. Furthermore, increased global atmospheric mercury
(Hg) deposition due to human activities (e.g., fossil fuel consumption,
and incineration of municipal and medical wastes) in recent decades
has resulted in high levels of Hg in fish, and other organisms in the
Everglades (Arfstrom et al., 2000; Renner, 2001; Rood et al., 1995).
This led the Florida State Department of Health to issue a series of health
warnings regarding consumption of specific species of fish caught in the
Everglades (Lambou et al., 1991). After deposition, a component of the
inorganic Hg is converted to methylmercury (MeHg) by anaerobic
bacteria in aquatic systems (Renner, 2001). Although surface waters
contain low concentrations of MeHg, after entering the food chain
MeHg biomagnifies to toxic levels in organisms occupying higher trophic positions and poses health risks to human consumers (Cabana
and Rasmussen, 1994; Cleckner et al., 1998; Henry and Bigham, 1995;
Rumbold and Fink, 2006; Rumbold et al., 2008; Stober et al., 1995).
Thus, food web structure plays a critical role in controlling Hg concentration in fish. Elucidating the complex structures of food webs can
help to understand processes and factors controlling Hg contamination
of aquatic ecosystems.
Isotopic analysis of fish tissue provides an integrated measure of diet
assimilated over time. Consumers are generally enriched in the heavy
nitrogen (N) isotope, 15 N, by ~3–5‰ relative to their diet (Minagawa
and Wada, 1984; Post, 2002; Schoeninger and Deniro, 1984). This stepwise N isotopic increase through the food web has been used to examine trophic positions of organisms (e.g., Chen et al., 2011; Stapp et al.,
1999; Vander Zanden et al., 1999), and to estimate Hg bioaccumulation
rates in fish in northern temperate lakes (Cabana and Rasmussen,
1994). The stable carbon (C) isotopic composition of a consumer, on
the other hand, is very similar to its food, with only a slight enrichment
of ~1‰ or less relative to its prey, and primarily records the C isotopic
variation at the base of the food chain (e.g., DeNiro and Epstein, 1978;
Michener and Schell, 1994; Post, 2002). Thus, C isotopes in aquatic consumers contain valuable information about their energy (or C) sources.
However, using stable isotope analyses alone to unravel the complex
structure of food webs can sometimes be problematic due to the complexity of ecosystem processes and the natural isotopic variability at
the base of the food chain (Cabana and Rasmussen, 1996; Stapp et al.,
1999; Vander Zanden et al., 1999; Vander Zanden and Rasmussen,
1999, 2001). Radioactive C isotope (14C), with a half-life of 5730 years,
can provide additional insights into C sources and food web dynamics
by allowing differentiation of recent photosynthate (originated from
modern plant material) from old soil/sedimentary organic matter
(Caraco et al., 2010; Cherrier et al., 1999; Schell, 1983).
In this study, we determined the radiocarbon, and stable C and N
isotopic signatures of fishes from both relatively pristine (reference)
and impacted (eutrophic) wetlands in the Everglades. The objective
was to test the hypothesis that human activities have not only impacted
the water quality and plant communities in the Everglades ecosystem
(e.g., Jickells, 1998; McCormick et al., 2001; Gu et al., 2006), but also
affected the food web dynamics in the wetlands.
2. Materials and methods
Our study area was in the Everglades region (~ 25-26°N, ~ 80.681°W) of Florida. The Everglades region encompasses most of the
southern Florida peninsula and represents the largest subtropical wetland ecosystem in the US. The hydrological regime of the Everglades
has been drastically altered over the last century. Prior to settlement,
the hydrology of this region was controlled by seasonal cycles in rainfall
causing sheet flow from Lake Okeechobee through a vast expanse
of wetlands southward to the Florida Bay, creating the Swampy
Everglades – a huge area of freshwater marshes known as the “River
of Grass”. To encourage settlement and provide agricultural lands,
drainage and reclamation projects were instituted at the turn of the
last century to control water flow in this area. The once expansive freshwater marshes are now dissected by drainage canals, levees, and water
control structures into many sub-basins (Fig. 1), including Everglades
Agricultural Area (EAA) (which is now completely drained), Water
Conservation Areas (WCA1, WCA2 and WCA3), Storm Water Treatment
Areas (STAs), and Everglades National Park (ENP). Water Conservation
Areas (WCAs) are diked marshes. These make up the largest remnants
of the original Everglades wetland ecosystem outside the ENP, and are
used not only to store water, but also as buffers between the EAA and
the more pristine ENP to the south. Storm Treatment Areas are reconstructed wetlands adjacent to the EAA, receiving direct runoff from
the agricultural land (Gu et al., 2006). They were established by the
South Florida Water Management District (SWFMD) in the 1990s to
treat runoff water from the EAA before the water is released into the
WCAs to help restore the remnant Everglades. The primary objective
of the STAs is to significantly reduce the amount of phosphorus (P)
in water entering the WCAs to below the requirement (10 ppb)
established by the Everglades Forever Act.
Fish samples analyzed in this study were provided by the SFWMD
and Florida Fish and Wildlife Conservation Commission (FWCC)
which routinely collects fishes from the Everglades area to monitor
the Hg contaminant levels. Our fish samples include largemouth
bass (Micropterus salmoides), sunfish (bluegill: Lepomis macrochirus),
and eastern mosquitofish (Gambusia holbrooki), and were collected
from STA2, STA5, STA6, WCA3, a canal near WCA3, and ENP (Fig. 1;
Supplementary Table). The STAs represent impacted wetlands that
receive direct runoff from the EAA, whereas the ENP and WCA3 are
relatively pristine (Childers et al., 2003; Gu et al., 2006) and serve as
reference wetlands in this study.
Mosquitofish are widespread in the Everglades, have short life spans,
and are typically less than 40 mm in length. They forage on periphyton
and some zooplankton (Browder et al., 1994; Cleckner et al., 1998).
Sunfish are thought to have an average life span of 4–7 years in the
wild. The diet of adult bluegill sunfish consists of aquatic invertebrates
and other small fish. Largemouth bass are the top predator fish in
the Everglades. Their diet includes various small fishes (e.g., bluegill),
crayfish, frogs, baby alligators, and snails (Gu and Howard, 2013).
Y. Wang et al. / Science of the Total Environment 487 (2014) 557–564
559
Fig. 1. Map showing the study localities STA2 (Storm Water Treatment Area 2), STA5 (Storm Water Treatment Area 5), STA6 (Storm Water Treatment Area 6), WCA3 (Water Conservation
Area 3) and Everglades National Park (ENP) in the Florida Everglades area.
Total mercury (THg) data for fish (mg/kg in tissue, wet weight) were
obtained from the DBHYDRO database available at the SFWMD website
(DBHYDRO, 1990). Total Hg concentration was determined for composite samples of mosquitofish (with each sample consisting of ≥100 individual fish), individual whole sunfish, and fillets of largemouth bass by
the SFWMD and the Florida Department of Environmental Protection
(FDEP) (Gabriel et al., 2008). Both agencies are certified by the Florida
Department of Health under the National Environmental Laboratory
Accreditation Program (NELAC). The analyses were performed on wet
tissues. SFWMD used USEPA Draft Method 1631 (Mercury in Water by
Oxidation, Purge and Trap, and Cold Vapor Atomic Fluorescence
Spectrometry) for THg detection in fish tissue, and FDEP used USEPA
Method 245.6 (Mercury in Tissues by Cold Vapor Atomic Absorption
Spectrometry). Both methods apply performance-based standards and
appropriate levels of QA/QC as required by NELAC. The method detection limit was 0.002 mg/kg. SFWMD archives all fish sampled for at
least1 year. We obtained our samples from this collection. Our fish
samples were collected by SFWMD in October and December of 2006
and March of 2007 (Supplementary Table), and were analyzed for stable
isotopes in 2008 and for radiocarbon in 2011.
For isotope analysis, fish samples were freeze-dried and ground into
powder. Carbon and N isotopic ratios were then analyzed using a Carlo
Erba Elemental Analyzer interfaced to a Finnigan MAT Delta Plus XP stable isotope ratio mass spectrometer (IRMS) at Florida State University.
The results are reported in the standard δ notation:
h
i
δ ¼ Rsample =Rstandard −1 1000;
ð1Þ
where, R = 13C/12C or 15 N/14 N; the standard is the international standard V-PDB for δ13C and air N2 for δ15N values. The precision of the C and
N isotope analysis was ±0.2‰ (1σ) or better on the basis of repeated
analysis of 4 different laboratory standards. The radiocarbon activity of
fish tissue was determined at the National Ocean Sciences Accelerator
Mass Spectrometry Facility (NOSAMS) in Woods Hole. The radiocarbon
data are reported as age in years before present (yr. B.P.) and Δ14C
values following the convention (Stuiver, 1980; Stuiver and Polach,
1977), where Δ14C is defined as:
14
Δ C ¼ ½ASN =Aabs –1 1000:
ð2Þ
ASN is the specific activity of a sample, which is proportional to the 14C/C
ratio, normalized to δ13C = −25‰, and Aabs is the absolute 14C activity
(i.e., 14C/C ratio) in the NBS oxalic acid standard (Stuiver, 1980; Stuiver
and Polach, 1977). The fraction of different C sources in fish diet was
estimated using a two-component mass-balance equation:
X ¼ ðδfish –δ1 Þ=ðδ2 –δ1 Þ:
ð3Þ
X is the fraction of source 2 in diet, δfish is the δ13C or Δ14C value of fish,
δ1 is the δ13C or Δ14C value of source 1, and δ2 is the δ13C or Δ14C value
of source 2. Statistical analyses were performed using the statistical
module associated with KaleidaGraph 4.5 software.
560
Y. Wang et al. / Science of the Total Environment 487 (2014) 557–564
3. Results and discussion
Table 1
Isotopic differences among fishes in the Everglades wetland ecosystems.1
3.1. Stable C and N isotope compositions of fishes
Plants can be divided into two major groups, C4 and C3, based on
their photosynthetic pathways (Deines, 1980). C3 plants (using the C3
pathway) discriminate against 13CO2 during photosynthesis to a greater
extent than do C4 plants (using the C4 pathway). As a result, C3 plants
have δ13C values ranging from − 22 to − 34‰, with an average of
−27‰, whereas C4 plants have δ13C values between −9 to −17‰, averaging −13‰ (Deines, 1980). The Everglades wetlands are dominated
by C3 plants, while the primary crop in the EAA is sugarcane, which is a
C4 plant (Stern et al., 2007; Wang et al., 2002). Thus, the potential food
sources in the Everglades area include modern wetland plants (C3), peat
deposits (C3), and sugarcane (C4). Sugarcane is significantly enriched in
13
C compared to wetland plants and peat (Fig. 2). The δ13C values of
largemouth bass and sunfish from reference wetlands (i.e., ENP and
WCA3) range from − 27.3 to − 31.1‰, which indicate a food web
based entirely on C3 plants, consistent with the C3-plant-dominance of
the wetland environment (Fig. 2). Largemouth bass and sunfish from
the STAs, however, have a wider range of δ13C values of − 19.2 to
− 31.5‰ (Fig. 2), indicating that sugarcane-derived C, resulting from
erosion of soil in the EAA, has contributed a variable amount to the C
source at the base of the food web in these wetlands. Thus, our δ13C
data reveal that sugarcane-derived C originated from the EAA was
passed along the food chain to consumers in the impacted wetlands,
but not in the reference wetlands (Fig. 2). Largemouth bass appeared
to be enriched in 13C relative to sunfish in a given system except STA2
(Table 1, Fig. 2), likely reflecting different food preferences of these
fishes and/or a slight 13C-enrichment (i.e., ~ 1‰ or less per trophic
level) along food chain (e.g., Michener and Schell, 1994; Post, 2002).
Previous studies (Stern et al., 2007; Wang et al., 2002) showed
that wetland plants in the Everglades area had an average δ13C value
of − 27 ± 2‰, and sugarcane from the EAA has a δ13C value of
− 11 ± 1‰. Assuming that these average values represent the end-
LMB (ENP)
Sunfish (ENP)
LMB (WCA3)
Sunfish (WCA3)
Sunfish (canal near WCA3)
LMB (STA2)
Sunfish (STA2)
Peat and wetland plants
LMB (STA5)
Sunfish (STA5)
20
LMB (STA6)
Sunfish (STA6)
Mosquitofish (STA5/6)
15
N
16
12
ENP
WCA3
8
Increasing sugarcane-derived C
-32
-30
-28
-26
-24
-22
-20
-18
-16
13
C
ENP
WCA3
STA2
STA5
STA6
STA5- STA6
STAs
Δδ13C (L-S)
Δδ15N (L-S)
Δδ13C (LS-M)
Δδ15N (LS-M)
1.9**(8)
1.3**(5)
0.1(6)
3(4)
2.2(6)
2.5(10)
1.6(16)
−0.6(8)
2.4**(5)
3.1*(6)
1.5(4)
1.8*(6)
1.7(10)
2.2**(16)
1.3(13)
3.6(19)
1.8(13)
2.0(19)
1
Comparison of mean Δδ13C and Δδ15N (trophic fractionation) values for taxon and
habitat. Numbers in brackets indicate the number of samples. Significance based on
Student t-test (unpaired data with equal variance). One asterisk indicates mean difference
significant at p b 0.05. Two asterisks indicate difference significant at p b 0.01.
L-S = Largemouth bass vs. Sunfish; LS-M = (Largemouth bass and sunfish) vs.
Mosquitofish. ENP = Everglades National Park; WCA3 = Water Conservation Area 3;
STAs = Storm Water Treatment Areas.
member δ13C values for pure C3 and C4 food source, respectively, in
the area, the amount of sugarcane-derived C in fish diets was estimated
using a simple two-component mixing model (Eq. (3)) to range from ~5
to 49% in STA5 (n = 6), 0 to 38% in STA6 (n = 7), and 0% in the STA2
(n = 6). The highest δ13C value of −19.2‰ was found in a largemouth
bass from the STA5 and indicates that sugarcane-derived C made up
~ 49% of food consumed by this individual. Differences in δ13C among
systems most likely reflect variations in local food sources as different
STAs consist of different number of cells, have different plant communities and water residence times, and receive runoff from different parts of
the EAA. Although sugarcane was the main crop in the EAA, there are
vegetable crops growing in the area. The apparent lack of sugarcanederived C in fish diets in STA2 could be due to the small sample size
since not all individuals in the STAs consumed sugarcane-derived C
as indicated by our limited data. It is also possible that the inflow
into STA2 contained little or no sugarcane C because it drained areas
where vegetable crops were grown.
δ15N values of largemouth bass and sunfish varied from 8.5‰ to
13.5‰ in the reference wetlands (ENP and WCA 3), and from 10.0‰
to 15.6‰ in the STAs (Fig. 2). Nitrogen isotopic variation generally reflects the trophic positions in the food chain as consumers become
enriched in 15 N relative to their food by 3–5‰ (Cabana and
Rasmussen, 1994; DeNiro and Epstein, 1981; Post, 2002). Largemouth
bass and sunfish from the reference wetlands were enriched in 15 N
by ~ 4–10‰ relative to the basal consumers (i.e., Seminole ramshorn
snail Planorbella duryi: δ15N = 2.7 ± 1.0‰; scud Hyalella azteca:
δ15N = 5.0 ± 1.6‰) from the same areas (Williams and Trexler,
2006). However, no similar data were available for the STAs for comparison. Statistical analyses of our limited data show that largemouth bass
are generally enriched in 15 N (by ~2–3‰) relative to sunfish in a given
system except ENP (Table 1), suggesting that largemouth bass occupy
a higher trophic position (i.e., ~ 1 trophic level higher) than sunfish.
This is consistent with the ecology and dietary habits of these fishes
(Gu and Howard, 2013). In the ENP, although the δ13C values of
largemouth bass were significantly higher than those of sunfish, there
was no significant difference in δ15N values between largemouth bass
and sunfish (Table 1). This suggests similar trophic positions for the
2 fish species in the ENP. Another possible explanation is that sunfish
in the ENP, just like their counterparts in the STAs, are also at a lower
trophic level but the base of their food chain has a higher δ15N value
compared to largemouth bass. More data are needed to verify these
hypotheses.
3.2. Radiocarbon contents of fishes
Fig. 2. Comparison of δ15N with δ13C values of fishes from reference (ENP & WCA3)
and impacted (STAs) wetlands in the Everglades. Solid and open symbols represent
largemouth bass (LMB) and sunfish, respectively, and each symbol represents one sample.
Different colors represent fishes from different wetlands. Also shown is the δ13C range for
wetland plants and peat in the Everglades and the effect of incorporation of sugarcane
(δ13C = −11‰).
Radiocarbon is a very useful tracer for C in aquatic ecosystems
(Cherrier et al., 1999; Stern et al., 2007; Wang et al., 2002). Nuclear
weapons testing in the 1950s and early 1960s injected large amounts
of radiocarbon into the atmosphere, which elevated the radiocarbon
Y. Wang et al. / Science of the Total Environment 487 (2014) 557–564
content of the atmosphere by several orders of magnitude (Manning
et al., 1990). This pulse of “bomb” radiocarbon has been declining
steadily towards the pre-bomb level since 1963 and has tagged all
organic matter produced by photosynthesis over the last decades
(Manning et al., 1990; Wang et al., 1997). Positive Δ14C values indicate
significant amounts of “bomb” radiocarbon and therefore a “young”
(post-bomb) C source, whereas negative Δ14C values indicate an old
C source such as the historic peat deposits in the EAA (Wang et al.,
2002). Largemouth bass and sunfish from the STAs contrast markedly
with their counterparts from the WCA3 and ENP in radiocarbon (14C)
content (Fig. 3). Fish Δ14C values ranged from −22 ± 17‰ (n = 6) in
STA6, −27 ± 7‰ (n = 4) in STA5, to −52 ± 16‰ (n = 6) in STA2, corresponding to radiocarbon ages of 136 ± 117 yr. B.P., 161 ± 59 yr. B.P.,
and 373 ± 132 yr. B.P., respectively. Fishes from STA2 had the lowest
mean Δ14C value or the oldest average radiocarbon age among all fishes
analyzed from the area. Previous studies have found significant 14C depletion in dissolved organic carbon (DOC) and particulate organic carbon (POC) in impacted areas adjacent to the EAA due to input of old C
from the historic peat deposits in the agricultural land (Wang et al.,
2002; Stern et al., 2007). This old C was mostly decomposed during
their residence in the STAs and only a small fraction made it farther
down stream into the relatively pristine areas of the marsh (Wang
et al., 2002; Stern et al., 2007). The old radiocarbon ages of modern fishes from the STAs indicate that historic peat C is transferred beyond the
microbial level to higher organisms and resident fishes are partly
dependent on allochthonous C derived from historic peat deposits as
their energy source. In stark contrast, all fishes from the reference wetlands (WCA3 and ENP) had positive Δ14C values of 37‰ to 81‰, averaging 59 ± 17‰ (n = 12, all means reported ± 1σ, or N modern),
indicating a diet based on modern C (after 1950) derived from postbomb primary production within the wetland (e.g., Kalish, 1993).
Land-use change can have significant impacts on the rates of C turnover in soils (e.g., Wang et al., 1999; Wang and Hsieh, 2002). Old organic
matter originating from deeper soil horizons that have been tilled to
surface becomes susceptible to erosion and degradation in an aerobic,
surface soil environment. Studies have shown that the STAs receive
large erosional inputs of peat C from the EAA with DOC and POC radiocarbon ages as old as ~ 2000 yr. B.P. (Stern et al., 2007; Wang et al.,
2002). Assuming that the average peat soil Δ14C value is − 212‰, the
same as the measured Δ14C of POC in the agricultural runoff (Stern
Reference
wetlands
Impacted wetlands
80
60
Largemouth Bass
Sunfish
Mosquitofish
14
0
0
100
-20
-40
300
Radiocarbon age (years B. P.)
20
">Modern"
C
40
-60
500
-80
STA2
STA5
STA6
Canal
near WCA3
ENP
WCA3
Location
Fig. 3. Δ14C values of fishes from reference (ENP & WCA3) and impacted (STAs) wetlands
in the Everglades. Vertical bars indicate one standard deviation from the means.
561
et al., 2007), and the Δ14C value of modern C is 45‰ (Caraco et al.,
2010), the peat C content of these fishes estimated using a simple
mass balance relationship (Eq. (3)), with the exception of mosquitofish,
decreased from 31 ± 8% in STAs to 0-3% in the WCA3 and ENP. The results indicate that little or no peat C is available to the invertebrate (prey
organisms) population consumed by the fish in reference (i.e., relatively
pristine) areas of the Everglades. It is important to note that the historic
peat deposits in the EAA were accumulated in previous wetland environment over several thousands of years before being drained and converted to agricultural land and thus the Δ14C value of eroded peat C is
unlikely a constant. Limited radiocarbon data from the Everglades
showed that DOC in agricultural runoff had lower Δ14C values in the
wet season than in the dry period (Wang et al., 2002). If the eroded
peat C had lower Δ14C (i.e., were older) than assumed, our calculations
would overestimate the amount of peat C in fish. If the eroded peat C
were younger (i.e., had higher Δ14C) than assumed, our calculations
would underestimate the amount of peat C in diet. Unfortunately, the
uncertainty in these estimates is difficult to quantify due to the lack of
sufficient radiocarbon data on DOC and POC in the area. Intra-species
Δ14C variation may reflect seasonal and interannual variations in 14C
content of local DOC and POC (Stern et al., 2007; Wang et al., 2002),
and variation in age of the individuals.
Mosquitofish from the STAs yielded Δ14C values of −3.3 to 12.9‰,
significantly higher than those of largemouth bass and sunfish from
the same area (Fig. 3). Their higher Δ14C values indicate a higher percentage of modern primary production supporting their life cycles as
compared to largemouth bass and sunfish in the same wetland.
Mosquitofish are known to have high productive potential (Pyke,
2008), and are abundant in the Everglades (Abbey-Lee et al., 2013).
They consume periphyton as well as some zooplankton (Browder
et al., 1994; Cleckner et al., 1998). Like other modern plants, periphyton
samples from the Everglades yielded positive Δ14C values reflecting
the 14C content of contemporary atmosphere (Stern et al., 2007).
Mosquitofish have short life spans, typically less than 6 months. As
hydro-environmental conditions change, their diets change to take advantage of the food available (Browder et al., 1994; Cleckner et al.,
1998), which would explain their Δ14C variability. If other small fish
are also characterized by high Δ14C values as observed in mosquitofish,
the large 14C difference between mosquitofish and the larger fish in the
STAs (Fig. 3) would suggest that largemouth bass and sunfish in the impacted areas have become less dependent on small fish or otherwise
their Δ14C values would be similar to the small fish. That is, input of agricultural runoff into the wetland may have shortened the aquatic food
chain. Although this inference about a shorten food-chain in impacted
areas is supported by the fish Hg data as discussed in the next section,
more complete sampling and analysis of the food web, including other
small fish and primary consumers, are needed to elucidate the possible
links in the food chain, and to test this hypothesis of a shortened food
chain in impacted wetlands.
Taken together, the Δ14C, δ13C, and δ15N values of fishes from both
impacted and reference areas of the Everglades indicate a major change
in food web dynamics in the wetlands due to land-use change. Fishes in
the relatively pristine wetlands of the Everglades rely on the food supply
of modern primary production within the wetlands. In contrast, fishes
in the impacted areas depend in part on the trophic transfer of allochthonous C (derived from historic peat deposits in the drained agricultural land and crop residues) through the food web and attain only partial
dependence on primary production within the wetland. Furthermore,
Δ14C data suggest that largemouth bass and sunfish in impacted areas
may have become less dependent on small fish, implying a shortened
food chain in impacted wetlands.
An earlier study showed that erosional inputs of peat C were an
important food source for resident fish and ducks in the Arctic environment where primary production is limited to ~ 5 months of the year
(Schell, 1983). A more recent study documented the importance of
highly aged organic matter (N2000 yrs B.P.) derived from erosion of
562
Y. Wang et al. / Science of the Total Environment 487 (2014) 557–564
source materials exceeding the isotope fractionation in the food chain
(Fig. 2). For example, the δ13C value of sugarcane is −11.4‰, which is
more than 10‰ higher than that of the pristine wetland vegetation
(Wang et al., 2002), while nitrification and denitrification processes
are known to produce large nitrogen isotope fractionations of N 10‰
(e.g., Durka et al., 1994). Thus, changes in the allochthonous input of C
and nutrients and/or the N cycle could easily produce isotopic variability
greater than the fractionation through the food web, rendering δ15N and
δ13C less useful or even useless as indicators of trophic positions of organisms in the food chain. Comparison of the δ15N and δ13C values
with the length of the fish revealed weak or no significant relationship
between fish length and the δ15N and δ13C values in the STAs (Fig. 5b,
c). In the reference wetlands, however, our limited data show a positive
correlation between δ13C and length of fish but an inconsistent relationship (i.e., negative correlation in the ENP and positive correlation in the
WCA3) between δ15N and fish length (Fig. 5b, c). This indicates that the
δ15N and δ13C values of these fishes primarily reflect the N and C isotopic
variability of source materials or variation in consumption of different
preys with different δ13C/δ15N values (cf. Abbey-Lee et al., 2013). That
is, higher δ15N and δ13C values of these fishes do not necessarily indicate
higher trophic positions but rather reflect system-specific biogeochemical processes influencing the C and N cycling and the baseline δ15N and
δ13C values. In such systems where isotopic variability at the base of the
food chain exceeds the isotopic fractionation due to the trophic level
effect, using stable C and N isotopes alone to determine trophic positions of organisms may yield ambiguous results.
Another possible explanation for the lower THg in fishes from the
STAs are lower MeHg concentrations at the base of the food chain than
in the reference wetlands (Gilmour et al., 1998). Previous studies have reported lower MeHg in surficial sediments, water and biota from eutrophic
sites compared to more pristine areas in the WCA2 and WCA3 with a few
exceptions (Cleckner et al., 1998; Gilmour et al., 1998; Hurley et al., 1998;
Miles and Fink, 1998). The lower production of MeHg could result from
soil and/or ancient sediments as a significant food subsidy to zooplankton in a temperate riverine ecosystem (Caraco et al., 2010). In comparison, the subtropical Everglades in south Florida has a much higher
primary production than the Arctic and temperate environments, and
yet our data show that resident fish had become partially dependent
on peat C as their energy/food source in areas impacted by agricultural
runoff. This demonstrates that conversion of wetlands to agricultural
lands remobilizes the old C previously sequestered in the wetlands
to aquatic systems (Wang et al., 2002), and this old C is bioavailable,
providing an additional food source to aquatic organisms.
3.3. Comparison of isotopic signatures with fish Hg contents
Fishes in the reference wetlands (ENP and WCA3) had a narrow
range of δ13C and δ15N values (δ13C: − 27.3 to − 31.1‰; δ15N: 8.5–
13.5‰), high Δ14C values and variably high Hg contents (Fig. 4). In comparison, fishes in the impacted wetlands (STAs) had a larger range of
δ13C (from −19.2 to −31.5‰) and δ15N values (from 9.5‰ to 15.6‰),
but consistently low Hg and 14C contents (Fig. 4).
MeHg bioaccumulates in fish and other organisms with top predators having the highest Hg levels (Korhonen et al., 1995). As a fish
moves up the trophic ladder and/or grows larger, the Hg concentration
increases (Fig. 5a). Thus, the much lower THg in fishes from the STAs, in
comparison to their counterparts in reference wetlands, likely indicates
a shortened food chain for fishes in the STAs, which is consistent with
the radiocarbon data (as discussed in the previous section). In other
words, both largemouth bass and sunfish in the STAs have become
less dependent on small fish and primary production within the wetlands than their counterparts in reference wetlands, but more dependent on allochthonous, peat/soil-derived organic C characterized by
low Δ14C and more variable δ13C and δ15N values. The larger interand intra-species δ13C and δ15N variation for fish in the STAs relative
to the reference wetlands primarily reflect the isotopic variability of
2
2
c
LMB-ENP
LMB-WCA3
LMB-STAs
Sunfish-ENP
Sunfish-WCA3
Sunfish-WCA3 canal
Sunfish-STAs
Mosquitofish-STAs
1.5
1
1.5
1
0.5
0.5
0
-32
0
-30
-28
-26
-24
-22
-20
-18
-16 0
100
13
200
300
400
500
Length (mm)
2
2
b
d
1.5
1.5
1
1
0.5
0.5
Total Hg (mg/Kg)
Total Hg (mg/Kg)
Total Hg (mg/Kg)
Total Hg (mg/Kg)
a
0
0
8
10
12
14
15
16
18
-50
0
50
100
14
Fig. 4. Comparison of (a) δ13C, (b) δ15N, (c) length, and (d) Δ14C values with total mercury contents of largemouth bass (LMB), sunfish and mosquitofish from reference (ENP & WCA3) and
impacted (STAs) wetlands in the Everglades. Solid symbols represent fishes from reference wetlands and open symbols impacted wetlands, with each symbol representing one sample.
Total mercury content in fish is presented as mg/Kg on a wet weight basis.
Y. Wang et al. / Science of the Total Environment 487 (2014) 557–564
3
ENP : THg = 0.64xlength/100 - 0.59, r2=0.93, p < 0.001
WCA3 THg = 0.28xlength/100 - 0.36, r2=0.98, p =0.001
WCA3:
STAs : THg = 0.09xlength/100 - 0.091, r2=0.72, p <0.001
WCA3 canal
Mosquitofish-STAs
a
Total Hg (mg/Kg)
2.5
2
1.5
1
0.5
0
0
100
200
300
400
500
Length (mm)
15
22
20
18
δ15 N (% )
2
N-15 (ENP)
ENP:
N = - 0.34xlength/100 + 13.25, r = 0.36, p = 0.1
15
2
N-15 (WCA3)
WCA3:
N = 0.70xlength/100 + 7.76, r =0.94, p =0.03
15
2
N-15
(STAs)
N = 1.02xlength/100 + 10.92, r =0.18, p = 0.1
STAs:
N-15 (WCA3
WCA3
canal canal)
Mosquitofish - STAs
b
16
14
12
10
8
0
100
200
300
400
500
Length (mm)
-10
13
2
C-13
ENP:(ENP)
C = 1.07xlength/100 - 31.9, r = 0.89, p < 0.001
13
2
C-13
(WCA3)
C = 0.41xlength/100 - 29.31, r =0.91, p =0.01
WCA3:
13
2
C-13
(STAs)
STAs:
C = 0.57xlength/100 - 27.95, r =0.01, p = 0.7
C-13 (Canal)
WCA3
canal
Mosquitofish -STAs
c
δ15 C (% )
-15
-20
-25
-30
0
100
200
300
400
500
Length (mm)
Fig. 5. Comparison of fish length with total mercury content (a), δ15N (b), and δ13C (c) in
the Everglades. Solid symbols represent fishes from reference wetlands and open
symbols are for fishes from impacted wetlands. Each symbol represents one sample.
Each mosquitofish sample was a composite of 100 or more individuals (n = 100-250)
and the average length was assumed to be ~25 mm as mosquitofish in the area typically
range from 9 to 39 mm in length (http://www.epa.gov/region4/sesd/reports/
epa904r98002/app-d.pdf). Also shown are the linear regression equations and associated
correlation coefficients and p values for each system.
elevated microbial sulfate-reduction rates in the STAs as hydrogen sulfide
produced by sulfate-reducing bacteria is known to inhibit Hg methylation
(Gilmour et al., 1998; Rumbold and Fink, 2006). Hydrogen sulfide may
also react with dissolved metals to form insoluble metal sulfides; such reaction would decrease the availability of dissolved Hg for methylation by
sequestering it in insoluble HgS (Kongchum et al., 2006) or Fe-sulfides
(Huerta-Diaz and Morse, 1992; Saunders et al., 2008; Sklar and Browder,
1998). Other studies, however, suggest that some forms of HgS (such as
HgS nanoparticles and dissolved neutral complex HgSo) are bioavailable
to Hg-methylating bacteria (Benoit et al., 2001; Graham et al., 2012;
Zhang et al., 2012). Thus, it is difficult to relate sulfate-reduction rates to
either lower or higher THg in fish. Moreover, widespread drainage and
water loss through surface flow diversion in the Everglades since 1880
might have lowered the water table and increased salinity of many wetlands (Sklar and Browder, 1998). Compeau and Bartha (1987) showed
that Hg methylation may be inhibited in high-salinity environments
563
where sulfate reducing bacteria compete with methanogenic bacteria
for limited organic matter produced by fermentation organisms. High salinity is also likely to decrease Hg methylation rate due to the formation of
charged Hg complexes (e.g., HgCl2−/HgCl−) that are less bioavailable by
the sulfate-reducing bacteria (Barkay et al., 1997). However, this is unlikely the case in the studied wetlands because they are freshwater wetlands
with high amounts of DOC (e.g., Wang et al., 2002). It has also been postulated that periphyton communities support an active microbial sulfur
cycle and Hg methylation (Cleckner et al., 1999). Thus, the reduction in
periphyton communities in the STAs may further limit Hg methylation,
resulting in limited biomagnification of Hg in resident fishes. Limited
data available for the Everglades show that MeHg concentrations
in both surficial sediments and water in the relatively pristine WCA3
(i.e., ~5 ng/g dry weight sediment and ~0.5 ng/L water) were N16 times
higher (~50 times higher in surficial sediments and ~17 times higher in
water) than those in the eutrophic areas (b0.1 ng/g dry weight sediment
and ~0.03 ng/L water) (Cleckner et al., 1998; Gilmour et al., 1998; Hurley
et al., 1998). In comparison, THg in both sunfish and largemouth bass
from the reference wetlands were 1 to 8 times higher than their similar
size counterparts in the STAs (Fig. 4c). This cannot solely be explained
by the difference in MeHg at the base of the food chain assuming no
change in the length of the food chain. Thus, a shorten food chain, as
suggested by our radiocarbon data, could explain the low THg in fishes
in the impacted wetlands.
Because MeHg bioaccumulates in organisms, the amount of THg in
fish should increase not only with trophic level but also with increasing
length of the fish (Korhonen et al., 1995). As expected, our data show
that fish THg content is strongly and positively correlated with fish
length within each ecosystem but this THg-length relationship varies
significantly among the three ecosystems examined (Fig. 5a). The bioaccumulation of THg in fish is highest in the relatively pristine ENP
(~ 0.64 mg/kg rise in THg per 100 mm increase in length) and lowest
(more than 7 times lower than in the ENP) in the eutrophic STAs
(Fig. 5a), reflecting differences in the food web structure and biogeochemical processes controlling baseline MeHg concentrations or
bioavailability among these ecosystems. Excess nutrients have many
adverse effects on the aquatic ecosystem (e.g., McCormick et al.,
2001). Thus, the seemingly positive influence of nutrient loading on
Hg in fish in the Everglades wetlands should not be considered as a rational for relaxing nutrient control measures.
4. Conclusions
The isotopic and THg signatures of fishes reveal substantial differences in the food web and Hg cycling between reference and impacted
areas in the Florida Everglades wetland ecosystem. In reference
(relatively pristine) wetlands, largemouth bass and sunfish had modern
radiocarbon ages and a small range of δ13C and δ15N values, indicating
dependence on the food supply of modern primary production within
the wetland. In impacted (eutrophic) areas, largemouth bass and sunfish had old radiocarbon ages and a wider range of δ13C and δ15N values
while mosquitofish had modern radiocarbon signatures. These trends
indicate that allochthonous organic matter derived from historic peat
deposits and crop residues in the Everglades Agricultural Area has become a significant energy/food source for resident fishes. This change
in food web from one driven primarily by in situ productivity to one
that is partially dependent on allochthonous input of peat C and crop
residues, as evidenced by the isotope data, demonstrates the impacts
of human activities on the Everglades food chains. Fish radiocarbon
data also suggest that largemouth bass and sunfish may have become
less dependent on small fish in impacted areas based on the older radiocarbon ages than mosquitofish in the same wetlands. This implies a shift
toward lower trophic levels (i.e., a shortened food chain), consistent
with lower THg levels in fish, in the eutrophic areas. These results
show that hydrological modification and land-use change can alter the
native food web structure and the C and Hg cycles in aquatic ecosystems.
564
Y. Wang et al. / Science of the Total Environment 487 (2014) 557–564
Isotopic measurements, along with THg and MeHg analyses, of primary
producers and invertebrates as well as a larger number of fishes would
improve our understanding of the complexity and dynamics of the
food web and Hg biomagnification in the Everglades wetland ecosystem.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.scitotenv.2014.04.060.
Acknowledgements
This work received financial support from NSFC grants U1033602
and 31200357, Guangdong Higher Education Institutes Grant for
High-level Talents, and South Florida Water Management District. All
isotope analyses were performed at the Florida State University Stable
Isotope Laboratory supported by grants from the U.S. National Science
Foundation (EAR 0716235 and EAR 0517806). We thank three anonymous reviewers and Associate Editor Dr. M. S. Gustin for their helpful
comments and suggestions that improved this manuscript.
References
Abbey-Lee RN, Gaiser EE, Trexler JC. Relative roles of dispersal dynamics and competition
in determining the isotopic niche breadth of a wetland fish. Freshw Biol 2013;58:
780–92.
Arfstrom C, Macfarlane AW, Jones RD. Distributions of mercury and phosphorous in
Everglades soils from Water Conservation Area 3A, Florida, USA. Water Air Soil Pollut
2000;121:133–59.
Barkay T, Gillman M, Turner RR. Effects of dissolved organic carbon and salinity on
bioavailability of mercury. Appl Environ Microbiol 1997;63:4267–71.
Benoit JM, Gilmour CC, Mason RP. The influence of sulfide on solid phase mercury
bioavailability for methylation by pure cultures of Desulfobulbus propionicus
(1pr3). Environ Sci Technol 2001;35:127–32.
Browder JA, Gleason PJ, Swift DR. Periphyton in the Everglades: Spatial variation, environmental correlates, and ecological implications. In: Davis S, Ogden JC, editors. Everglades:
The Ecosystem and Its Restoration. Delray Beach, FL: St. Lucie Press; 1994. p. 379–419.
Cabana G, Rasmussen JB. Modeling food-chain structure and contaminant bioaccumulation using stable nitrogen isotopes. Nature 1994;372:255–7.
Cabana G, Rasmussen JB. Comparison of aquatic food chains using nitrogen isotopes. Proc
Natl Acad Sci U S A 1996;93:10844–7.
Caraco N, Bauer JE, Cole JJ, Petsch S, Raymond P. Millennial-aged organic carbon subsidies
to a modern river food web. Ecology 2010;91:2385–93.
Chen G, Wu ZH, Gu BH, Liu DY, Li X, Wang Y. Isotopic niche overlap of two planktivorous
fish in southern China. Limnology 2011;12:151–5.
Cherrier J, Bauer JE, Druffel ERM, Coffin RB, Chanton JP. Radiocarbon in marine bacteria:
Evidence for the ages of assimilated carbon. Limnol Oceanogr 1999;44:730–6.
Childers DL, Doren RF, Jones R, Noe GB, Rugge M, Scinto LJ. Decadal change in vegetation and
soil phosphorus pattern across the everglades landscape. J Environ Qual 2003;32:344–62.
Cleckner LB, Garrison PJ, Hurley JP, Olson ML, Krabbenhoft DP. Trophic transfer of methyl
mercury in the northern Florida Everglades. Biogeochemistry 1998;40:347–61.
Cleckner LB, Gilmour CC, Hurley JP, Krabbenhoft DP. Mercury methylation in periphyton
of the Florida Everglades. Limnol Oceanogr 1999;44:1815–25.
Compeau G, Bartha R. Effect of salinity on mercury methylating activity of sulfatereducing bacteria in estuarine sediments. Appl Environ Microbiol 1987;53:261–5.
DBHYDRO. DBHYDRO (Environmental Data). South Florida Water Management District;
1990 [(http://www.SFWMD.gov), last accessed on March 24, 2014].
Deines P. The isotopic composition of reduced organic carbon. In: Fritz P, Fontes JC, editors. The Terrestrial Environment, Handbook of Environmental Isotope Geochemistry.
Amsterdam: Elsevier; 1980. p. 329–406.
DeNiro MJ, Epstein S. Influence of diet on distribution of carbon isotopes in animals.
Geochim Cosmochim Acta 1978;42:495–506.
DeNiro MJ, Epstein S. Influence of diet on the distribution of nitrogen isotopes in animals.
Geochim Cosmochim Acta 1981;45:341–51.
Durka W, Schulze ED, Gebauer G, Voerkelius S. Effects of forest Decline on uptake and
leaching of deposited nitrate determined from N-15 and O-18 measurements. Nature
1994;372:765–7.
Gabriel M, Howard N, Matson F, Atkins S, Rumbold D. Appendix 5–7: Annual permit
compliance monitoring report for mercury in the STAs. South Florida Environmental
Report. West Palm Beach, FL: South Florida Water Management District; 2008.
[(http://my.sfwmd.gov), last accessed on March 24, 2014].
Gilmour CC, Riedel GS, Ederington MC, Bell JT, Benoit JM, Gill GA, et al. Methylmercury
concentrations and production rates across a trophic gradient in the northern
Everglades. Biogeochemistry 1998;40:327–45.
Graham AM, Aiken GR, Gilmour CC. Dissolved organic matter enhances microbial mercury methylation under sulfidic conditions. Environ Sci Technol 2012;46:2715–23.
Gu B, Howard N. Appendix 3–1, Attachment C: Annual permit compliance monitoring
report for Mercury in the STAs. South Florida Environmental Report. West Pam
Beach, FL: South Florida Water Management District; 2013. [(http://my.sfwmd.gov),
last accessed on March 24, 2014].
Gu BH, Chimney MJ, Newman J, Nungesser MK. Limnological characteristics of a subtropical constructed wetland in south Florida (USA). Ecol Eng 2006;27:345–60.
Henry EA, Bigham GN. Trophic state and mercury bioaccumulation in fish in the Florida
Everglades. Abstr Pap Am Chem S, 210. ; 1995. [106-GEOC].
Huerta-Diaz MA, Morse JW. Pyritization of trace metals in anoxic marine sediments.
Geochim Cosmochim Acta 1992;56:2681–702.
Hurley JP, Krabbenhoft DP, Cleckner LB, Olson ML, Aiken GR, Rawlik PS. System controls
on the aqueous distribution of mercury in the northern Florida Everglades. Biogeochemistry 1998;40:293–310.
IPCC. Climate change 2007: The physical science basis. In: Solomon S, Qin D, Manning M,
Chen Z, Marquis M, Averyt KB, Tignor M, Miller HL, editors. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change. Cambridge, United Kingdom and New York, NY, USA: Cambridge University
Press; 2007. p. 996.
Jickells TD. Nutrient biogeochemistry of the coastal zone. Science 1998;281:217–22.
Kalish JM. Pre-bomb and post-bomb radiocarbon in fish otoliths. Earth Planet Sci Lett
1993;114:549–54.
Kongchum M, Devai I, Delaune RD, Jugsujinda A. Total mercury and methylmercury in
freshwater and salt marsh soils of the Mississippi river deltaic plain. Chemosphere
2006;63:1300–3.
Korhonen P, Virtanen M, Schultz T. Bioenergetic calculation of mercury accumulation in
fish. Water Air Soil Pollut 1995;80:901–4.
Lambou VW, Barkay T, Braman RS, Delfino JJ, Jansen JJ, Nimmo D, et al. Mercury Technical
Committee interim report. Tallahassee: Florida State University; 1991. p. 60.
Manning MR, Lowe DC, Melhuish WH, Sparks RJ, Wallace G, Brenninkmeijer CAM, et al. The
use of radiocarbon measurements in atmospheric studies. Radiocarbon 1990;32:37–58.
McCormick PV, O'Dell MB, Shuford RBE, Backus JG, Kennedy WC. Periphyton responses to experimental phosphorus enrichment in a subtropical wetland. Aquat Bot 2001;71:119–39.
Michener R, Schell D. Stable isotope ratios as tracers in marine aquatic food webs. In:
Lajtha K, Michener R, editors. Stable Isotopes in Ecology and Environmental Science.
London: Blackwell Scientific Publications; 1994. p. 138–57.
Miles CJ, Fink LE. Monitoring and mass budget for mercury in the Everglades Nutrient
Removal Project. Arch Environ Contam Toxicol 1998;35:549–57.
Minagawa M, Wada E. Stepwise enrichment of N-15 along food-chains – Further evidence and the relation between Delta-N-15 and animal age. Geochim Cosmochim
Acta 1984;48:1135–40.
Post DM. Using stable isotopes to estimate trophic position: Models, methods, and
assumptions. Ecology 2002;83:703–18.
Pyke GH. Plague Minnow or Mosquito fish? A review of the biology and impacts of introduced Gambusia species. Annu Rev Ecol Evol Syst 2008;39:171–91.
Renner R. Everglades mercury debate. Environ Sci Technol 2001;35:59A–60A.
Rood BE, Gottgens JF, Delfino JJ, Earle CD, Crisman TL. Mercury accumulation trends in Florida Everglades and Savannas Marsh flooded soils. Water Air Soil Pollut 1995;80:981–90.
Rumbold DG, Fink LE. Extreme spatial variability and unprecedented methylmercury concentrations within a constructed wetland. Environ Monit Assess 2006;112:115–35.
Rumbold DG, Lange TR, Axelrad DM, Atkeson TD. Ecological risk of methylmercury in
Everglades National Park, Florida, USA. Ecotoxicology 2008;17:632–41.
Saunders JA, Lee M-K, Shamsudduha M, Dhakal P, Uddin A, Chowdury MT, et al.
Geochemistry and mineralogy of arsenic in (Natural) anaerobic groundwaters. Appl
Geochem 2008;23:3205–14.
Schell DM. C-13 and C-14 abundances in Alaskan Aquatic Organisms – Delayed production from peat in Arctic food webs. Science 1983;219:1068–71.
Schoeninger MJ, Deniro MJ. Nitrogen and carbon isotopic composition of bone-collagen
from marine and terrestrial animals. Geochim Cosmochim Acta 1984;48:625–39.
Sklar FH, Browder JA. Coastal environmental impacts brought about by alterations to
freshwater flow in the Gulf of Mexico. Environ Manag 1998;22:547–62.
Stapp P, Polls GA, Pinero FS. Stable isotopes reveal strong marine and El Nino effects on
island food webs. Nature 1999;401:467–9.
Stern J, Wang Y, Gu B, Newman J. Distribution and turnover of carbon in natural and
constructed wetlands in the Florida Everglades. Appl Geochem 2007;22:1936–48.
Stober QJ, Jones RD, Scheidt DJ. Ultra-trace level mercury in the Everglades ecosystem, a
multimedia canal pilot-study. Water Air Soil Pollut 1995;80:991–1001.
Stuiver M. Workshop on C-14 data reporting. Radiocarbon 1980;22:964–6.
Stuiver M, Polach HA. Reporting of C-14 data – Discussion. Radiocarbon 1977;19:355–63.
Vander Zanden MJ, Rasmussen JB. Primary consumer delta C-13 and delta N-15 and the
trophic position of aquatic consumers. Ecology 1999;80:1395–404.
Vander Zanden MJ, Rasmussen JB. Variation in delta N-15 and delta C-13 trophic fractionation: Implications for aquatic food web studies. Limnol Oceanogr 2001;46:2061–6.
Vander Zanden MJ, Casselman JM, Rasmussen JB. Stable isotope evidence for the food web
consequences of species invasions in lakes. Nature 1999;401:464–7.
Vitousek PM, DAntonio CM, Loope LL, Westbrooks R. Biological invasions as global
environmental change. Am Sci 1996;84:468–78.
Wang Y, Hsieh YP. Uncertainties and novel prospects in the study of the soil carbon
dynamics. Chemosphere 2002;49:791–804.
Wang Y, Jahren AH, Amundson R. Potential for C-14 dating of biogenic carbonate in hackberry (Celtis) endocarps. Quat Res 1997;47:337–43.
Wang Y, Amundson R, Trumbore S. The impact of land-use change on C turnover in soils.
Glob Biogeochem Cycles 1999;13:47–57.
Wang Y, Hsieh YP, Landing WM, Choi YH, Salters V, Campbell D. Chemical and carbon
isotopic evidence for the source and fate of dissolved organic matter in the northern
Everglades. Biogeochemistry 2002;61:269–89.
Williams AJ, Trexler JC. A preliminary analysis of the correlation of food-web characteristics with hydrology and nutrient gradients in the southern Everglades. Hydrobiologia
2006;569:493–504.
Zaret TM, Paine RT. Species introduction in a tropical lake. Science 1973;182:449–55.
Zhang T, Kim B, Leyard C, Reinsch BC, Lowry GV, Deshusses MA, et al. Methylation of
mercury by bacteria exposed to dissolved, nanoparticulate, and microparticulate
mercuric sulfides. Environ Sci Technol 2012;46:6950–8.