The composition
and quality of estuarine sediments can affect both the structure of the biotic
assemblage as well as the bioavailability of certain contaminants to local
biota. Sediments are generally composed of a combination of sand, silt and
clay. The composition of the benthic community can vary depending on how sandy
or how muddy (silts and clays combined) the sediments are. Also, contaminants
tend to adsorb to silt and clay particles so muddy sediments are more likely to
have higher contaminant concentrations than sandy sediments.
The average percentage of the silt/clay fraction at open water sites was 22% silt/clay compared to a mean of 30% silt/clay at tidal creek sites (see figure below, data tables). This difference was statistically significant (p < 0.015); however, there was considerable variability in the percent of silt/clay observed among the stations sampled in both habitats (from < 3% to > 95%; data online).
Approximately 6% of the sediments in open water habitat sampled in 2001 – 2002 were composed predominantly of silt/clay (> 80% silt/clay), while 14% of tidal creek habitats were predominantly silt and clay (see figure below, data tables). Values for mean silt/clay fraction and percent of the state’s total habitat representing each sediment type were similar between the two survey periods (1999-2000 and 2001-2002; Van Dolah et al., 2002a).
The average percent of sand versus silt/clay at open water and tidal creek sites sampled in 2001-2002 and estimates of the proportion of the state's coastal habitat that is primarily composed of the silt/clay fraction (>80% silt/clay), mixed (20-80% silt/clay), or sandy (<20% silt/clay) sediments.
The following sediment characteristics, contaminants, and toxicity were
evaluated in the SCECAP Program:
TOC:
Total organic carbon (TOC) provides a measure of how much organic material occurs in sediments. Hyland et al. (2000) found that extreme concentrations of TOC can have adverse effects on benthic communities. TOC levels below 0.5 mg/g (0.05%) and above 30 mg/g (3.0%) were related to decreased benthic abundance and biomass.
The TOC of sediments in tidal creeks ranged from 0.1 to 5.7% with a mean of 1.3% (Data Tables). Sediments in open water habitats contained lower concentrations of TOC with a mean of 0.9% and a range of 0.0 to 7.8% (see figure below). The difference between total organic carbon content in tidal creeks and open water sites was statistically significant (p < 0.004). Decomposing salt marsh plants and upland runoff are the primary sources of organic carbon. Open water sites are generally farther away from these sources resulting in lower TOC concentrations than tidal creek habitats.
Approximately 15% of the tidal creek habitats had sediment TOC levels that were above 3%, with no tidal creek habitat below 0.05%. About 6% of the open water habitats in the SCECAP survey had TOC levels that were less than 0.05%. Approximately 9% of the area of open water habitat was above 3% (figure below, data tables).
The National Coastal Assessment Program (USEPA, in review) has used TOC concentrations of above 2% and above 5% to indicate fair or poor sediment quality, respectively. Using these values, 4% of the tidal creek habitat and 2% of the open water habitat had TOC concentrations equal to or above the 5% threshold indicating poor conditions. Another 20% and 10% of tidal creek and open water respectively were in the fair category (2-5% TOC concentrations).
Average percent total organic carbon (TOC) concentration in sediments at open water and tidal creek sites sampled in 2001-2002 and estimates of the proportion of the state's coastal habitat having TOC levels (< 0.05 or > 3%), which may cause stress in benthic communities.
Total ammonia as nitrogen (TAN) in sediment porewater is another source of potential toxicity in sediments. The effects of TAN on marine biota are highly variable depending on the species considered (Sims and Moore, 1995; Moore et al., 1997). A value of 14 mg/L and 30 mg/L of TAN were used to indicate potential toxicity to seed clams (Ringwood and Keppler, 1998) and amphipods, respectively.
In the 2001-2002 survey, TAN levels were similar between open water sites (3.04 mg/L) and tidal creek sites (3.08 mg/L), and generally well below levels considered to be toxic (see figure below; data tables). Only 2% of both the open water and tidal creek habitats had TAN concentrations > 14 mg/L and none of the sites sampled in 2001-2002 had pore water TAN concentrations > 30 mg/L (data online). These values are similar to the 1999-2000 survey (Van Dolah et al., 2002a), indicating that there was no detectable change between the two survey periods.
Average total ammonia nitrogen (TAN) concentration in sediment pore water at open water and tidal creek sites sampled in 2001-2002 and estimates of the proportion of the state's coastal habitat having TAN concentrations that have a high probability of causing stress in benthic communities (>30mg/L, red), moderate probability of causing stress (>14mg/L & 30mg/L, yellow), or low probability of causing stress (<14mg/L, green).
Contaminants:
Sediments collected for SCECAP were examined for a wide range
of contaminants including 15 metals (thallium was added during the 2001 sampling
year), 25 polycyclic aromatic hydrocarbons (PAHs), 30 polychlorinated biphenyls
(PCBs), and 23 pesticides. For many of these contaminants,
Long et al. (1995) published bioeffects guidelines
that reflect the concentration of a contaminant that resulted in adverse
bioeffects in 10% of the studies examined (defined as Effects Range-Low or ER-L)
and concentrations that resulted in adverse effects in 50% of the studies
(defined as Effects Range-Median or ER-M).
Eight of the randomly selected open water sites in 2001 and six in 2002 had one or more contaminant concentrations above ER-L values. Nine tidal creek sites in 2001 and eleven in 2002 had one or more contaminant concentrations above ER-L values (Data Tables). Many of the ER-L exceedances in the tidal creeks were due to high levels of arsenic. Arsenic concentrations are naturally elevated in South Carolina estuarine sediments (Scott et al., 1994; 2000; Sanger et al., 1999a) and therefore the values observed are probably not related to anthropogenic stress. Other metal contaminants that exceeded ER-L values include nickel, chromium, mercury, lead, copper, cadmium, and zinc. A few PCBs, PAHs, and pesticides also exceeded their respective ER-L values. In most cases, the stations with ER-L exceedences were located in urbanized estuaries such as Charleston Harbor and Winyah Bay, reflecting the increased loadings of contaminants in these areas. Only one of the randomly selected sites sampled in 2001-2002 by the SCECAP program had contaminant concentrations that exceeded ER-M values. This station (RO026010) was located in Winyah Bay and had zinc levels of 628 µg/g (ER-M value for zinc is 410 µg/g). The contaminant concentrations found in the randomly located stations sampled during the 2001-2002 survey are similar to those found in the 1999-2000 survey.
Among the seven non-random stations in 2001-2002, two stations had contaminant levels that exceeded their respective ER-M values, in addition to having seven to eight ER-L exceedances. At station NO01098 in the Ashley River, ER-M values were exceeded for copper and zinc. Six other metals and two PAH analytes exceeded ER-L concentrations at this site (Data Tables), which is located adjacent to the Columbia Nitrogen Plant and the Koppers Plant. Both of these plants are EPA Superfund (CERCLA) sites. At Station NT01599 (Brickyard Creek in the Ashley River), Total DDT levels of 49.4 ng/g slightly exceeded the ER-M value for Total DDT of 46.1 ng/g. This station also had ER-L exceedances for seven metals and one other pesticide (Data Tables). This station is in a tidal creek that drains a heavily industrialized area of the Charleston peninsula.
While individual contaminants were elevated at some sites, a better assessment of overall contaminant exposure may be derived from the combined concentrations of all contaminants present at a site relative to bioeffects guidelines. Dividing the measured concentration of 24 contaminants by their respective ER-M values, and taking the average of all 24 values creates a combined value. The ERM-Quotient (ERM-Q) has been evaluated by Hyland et al. (1999) at more than 230 estuarine sites throughout the southeast, and provides a method for predicting stress in benthic invertebrate communities. ERM-Q values < 0.02 represent a low risk of observing degraded benthic communities, values > 0.02 and < 0.058 represent a moderate risk, and values > 0.058 represent a high risk of observing degraded benthic communities.
The mean ERM-Q among open water stations was 0.016 with a range of 0.001 to 0.122 (see figure below; data tables). The mean ERM-Q among tidal creek stations was 0.016 with a range of 0.001 to 0.046. Mean ERM-Q between habitat types was not significantly different. Using the criteria developed by Hyland et al. (1999), 21 of the tidal creek stations sampled (9 in 2001 and 12 in 2002) had ERM-Q values indicative of a moderate risk to benthic assemblages while the remainder had ERM-Q values indicative of a healthy benthos. Thirteen open water stations had ERM-Q values representing a moderate risk to benthos (6 in 2001 and 7 in 2002). Additionally, two stations sampled in 2002 had ERM-Q values indicative of high risk to benthic health (ERM-Q > 0.058). These stations were located in the Cooper River across from the old Navy Base (RO026090) and in the Ashley River, just below the Koppers Superfund site (RO026030) (Data Tables).
The estimated percent of the state’s tidal creek habitat that had ERM-Q values indicative of moderate risk to benthic health was 24% compared to 17% of the open water habitat. None of the state’s tidal creek habitat had a high ERM-Q, and only 3% of the state’s open water habitat had a high ERM-Q value (see figure below). These results are similar to the 1999-2000 survey. A year-by-year comparison of percent of total habitat (creek and open water habitats combined) shows some minor variation in the percentage of habitat that falls in the poor or fair categories, but no major increasing or decreasing trend in the proportion of South Carolina estuarine habitat with poor or fair contaminant levels. However, the 1999-2002 period coincided with a 4-5 year drought. Some contaminant concentrations may, in periods of normal rainfall, increase as runoff from the land increases.
Average cumulative sediment contaminant concentrations at open water and tidal creek sites sampled in 1999 - 2000 and the proportion of the state's coastal habitat having concentrations representing a low, moderate, or high risk of observing stress in benthic communities.
Toxicity:
Even if estuarine sediments have levels of contaminants shown
to cause adverse effects or mortality in laboratory exposure studies, these
contaminants may not be bioavailable to organisms living in and around the
sediments due to chemical binding properties with some sediments. Laboratory
bioassays are used as indicators of contaminant bioavailability. The three
bioassays used for the SCECAP survey provide useful evidence of probable
contaminant effects on benthic species, particularly when two or more of the
assays show toxicity.
A weight of evidence approach is used to define sediment toxicity. Positive tests in two or more of the assays indicate a high probability of toxic sediments, only one positive test indicates possible evidence of toxic sediments, and no positive tests indicates non-toxic sediments. For the 2001-2002 survey, 18% of both the tidal creek and open water habitats were considered toxic, and 35% of open water habitats and 55% of tidal creek habitats were considered possibly toxic (see figure below). When compared to the 1999-2000 survey (Van Dolah et al., 20 02a), there was a substantial increase in the area of tidal creek habitat considered toxic or possibly toxic (7% in 1999-2000 and 18% in 2001-2002). However, due to the high variability of the data, this difference is not statistically significant. Thirteen of the 25 sites (52%) sampled in 2001-2002 that had positive toxicity in both assays also had ERM-Q values > 0.02, which represents a moderate to high risk of observing stress in benthic communities. Toxicity in the sites with lower ERM-Q values may reflect toxicity from contaminants with no bioeffects guidelines, or it may represent a “false positive” test result.
Summary of sediment
bioassay results for 2001-2002 using multiple assays. Sediments are not
considered to be toxic if no significant toxicity was observed in any of the
tests (green), possibly toxic if one of the tests showed positive results
(yellow), and toxic if two or more of the tests showed positive results (red).
Integrated Assessment of
Sediment Quality:
The integrated sediment quality index combines measures of
sediment contaminant concentrations (ERM-Q) and sediment toxicity. For SCECAP,
an integrated score < 2 represents relatively poor sediment quality conditions,
scores > 2 but < 4 represent fair sediment quality conditions, and scores > 4
represent good sediment quality conditions. The results of the 2001-2002 survey
are similar to the 1999-2000 survey. For 2001-2002, none of the tidal creek
habitat had poor overall sediment quality and 40% coded as only fair in overall
quality (see figure below). In comparison, in 1999-2000, none of the tidal
creek habitat coded as poor, and 38% coded as fair in quality. For open water
habitats, 2% of the habitat was considered to have poor overall quality, and 28%
coded as having only fair sediment quality (values for 1999-2000 were 3% and
30%, respectively).
Annual comparisons, combining both habitat types, show an increasing area of habitat that was considered to be fair from 1999 to 2002, but little change in the proportion of habitat considered to be poor. The 1999 evaluation showed that none of the estuarine habitat was considered poor and 15% of the habitat was fair. The 2002 evaluation shows 3% of the estuarine habitat was considered poor and 27% was fair, an overall increase of 15% of the habitat falling into the poor or fair categories. While the current trend is statistically non-significant, as the data from the 2003 and 2004 sampling seasons becomes available, this trend can be re-evaluated.
The proportion of South Carolina's
estuarine habitat that ranks as good (green), fair (yellow) or poor (red), using
the integrated sediment quality score developed for SCECAP. This measure
of overall sediment quality incorporates the concentration of 24 contaminants
relative to known bioeffects levels, and the number of bioassays showing
toxicity.