Further Investigations of Classes of Compounds Associated with Sediment Toxicity at Regional Monitoring Program River Stations Bryn M. Phillips, Brian S. Anderson, and John W. Hunt University of California Institute of Marine Sciences, Santa Cruz, CA Introduction Methods   Results   Discussion and Conclusions   References

Introduction

Since the San Francisco Regional Monitoring Program (RMP) sampling began in the winter of 1993, three stations have exhibited consistent toxicity to bivalves and intermittent toxicity to amphipods. Significant toxicity to bivalves has been detected in all but one of the sediment elutriate samples from the Grizzly Bay, Sacramento River, and San Joaquin River stations. As part of a RMP Special Study, Phase I toxicity identification evaluations (TIEs) were conducted in August 1996 to better characterize potential causes of toxicity. Abbreviated TIEs were also conducted on August 1997 river samples to characterize chemicals responsible for toxicity to bivalve embryos exposed at the sediment-water interface (SWI). TIE results and measurements of trace metals in sediment elutriates indicated trace metals were a potential cause of toxicity in sediment elutriates from Grizzly Bay and San Joaquin River. Phase I TIE manipulations suggested an organic chemical might be the source of toxicity in Sacramento River sediment.

The three stations in question are essentially freshwater stations, although there is some tidal influence in Grizzly Bay. Because RMP samples have been tested with marine/estuarine species (i.e., bivalves), sediment elutriates are prepared by mixing the sediments with water at the test salinity of 28. It is not clear what effect elution of freshwater sediment with higher saline water has on chemical bioavailability or sediment toxicity. Part of the previous investigation included sediment elutriate toxicity tests with the freshwater cladoceran, Ceriodaphnia dubia. No adverse acute effects of the river sample elutriates were observed using Ceriodaphnia.

Tests conducted on samples from the three stations prior to this portion of the study are summarized in Table 4.6. As part of continuing research into the causes of toxicity at these stations, additional Phase I and Phase II TIE manipulations were conducted in April 1998 using the bivalve larval development test (Mytilus galloprovincialis). Based on the results of the previous tests, the current TIEs emphasized treatments that would mitigate toxicity of divalent metals. Additional manipulations included a combined EDTA/C18 column treatment, sodium thiosulfate treatment, and a cation exchange column treatment. Trace metals were also measured in unfiltered elutriate samples.

Investigations into metal toxicity also include an ongoing study of cupric ion concentrations in overlying water from SWI exposures from the three sites. Copper concentrations in sediment elutriates are within the range toxic to bivalves, and sample pH suggested ionic concentrations might be elevated in these samples. Sediment-water interface exposures were conducted simultaneously with the TIEs. Free copper ion concentrations were measured in overlying water from these exposures by determining copper complexation. The analytical technique employed uses flow injection analysis with chemiluminescent detection of a reaction between a copper-binding ligand and titrated copper (Zamzow, 1997). These analyses have so far produced cupric ion concentrations for two of the samples. Additional analyses will be conducted on spiked seawater samples in order to create a cupric ion dose-response curve for Mytilus larval development. Using the dose-response information, we will be able to determine if free copper ions were present at toxic concentrations in these samples.

Methods

Sample Preparation

All toxicity testing and sample manipulations were conducted at the Marine Pollution Studies Laboratory at Granite Canyon (MPSL). Elutriate solutions were prepared by adding 200 grams of sediment to 800 mL of Granite Canyon seawater in each of 4 clean 1-liter borosilicate glass jars with Teflon^®-lined lids (1:4 volume to volume ratio; U.S. EPA/ACOE, 1991). These mixtures were shaken vigorously for 10 seconds, then allowed to settle for 24 hours (Tetra Tech, 1986). The resulting supernatant was siphoned off for use in toxicity testing, TIE manipulations, and chemical analyses.

Trace metals were measured in unfiltered elutriate samples by Mark Stephenson and Jon Goetzl at the Department of Fish and Game Trace Metals Analytical Facility in Moss Landing. The analysis method was Inductively Coupled Plasma Mass Spectrometry (U.S. EPA method 1638).

Toxicity Identification Evaluations

Phase I TIE manipulations followed methods described by U.S. EPA (1996). A brief description of the treatments follows. Filtration (0.45 mm) removed contaminants associated with particles. Sample aeration was used to assess volatile constituents such as sulfide. Two different concentrations of EDTA were used to assess toxicity due to divalent cations. C18 solid-phase extraction columns were used to remove non-polar organic compounds. The C18 column was then eluted with methanol, and the eluate was added back to clean dilution water to determine if C18-bound organics were toxic. A combination C18 column/EDTA treatment was used to remove mixtures of organic and metal contaminants. A cation exchange column was used to remove metal contaminants that were then eluted with acid and added back to clean dilution water for confirmation testing. All column samples were pre-filtered (0.45 mm) so particulate-associated contaminants did not interfere with the interpretation of the results. Graduated pH adjustments (7.9, 8.1, and 8.4) were used to assess toxicity of ionic constituents such as ammonia. The addition of piperonyl butoxide (PBO) was used to test for the presence of metabolically activated pesticides such as diazinon.

Each manipulation was conducted on five concentrations of sediment elutriate from each station and a control. Controls consisted of Granite Canyon seawater and served as blanks for TIE treatments. TIE results were compared using analysis of variance between treatments within each elutriate concentration. Treatments were considered significantly different from the baseline treatment at p < 0.05.

Sediment-Water Interface Exposures (after Anderson et al., 1996)

Intact un-homogenized sediment cores were sampled directly from a modified van Veen grab sampler during routine sediment sampling for the RMP. Cores were brought back to the laboratory on ice, prepared for testing by slowly adding 300mL of overlying seawater, and equilibrated overnight under gentle aeration. Before test initiation, 25-mm mesh screen tubes were inserted into the core tubes containing the sediment, so that the screen was positioned about 1cm above the sediment. Approximately 200 mussel embryos were pipetted into the screen tubes and exposed for 48 hours. Tests were terminated by removing the screen tube and rinsing larvae into vials that were fixed with 5% formalin. All normally developed larvae were counted in each test container to determine the percentage of embryos that developed into live normal larvae. This value was determined by dividing the observed number of normal D-shaped prodisoconch larvae at the end of the test by the mean number of live embryos inoculated at the beginning of the test. Sediment-water interface exposures were conducted concurrently with Phase I TIE manipulations. Water samples for cupric ion analysis were collected from overlying water of additional replicate cores.

Results

Elutriate Chemistry

As of this writing, bulk phase sediment chemistry had not yet been analyzed on the 1998 RMP samples. A survey of chemistry from 1996 indicates that there were exceedances of effects range low (ERL; Long et al., 1995) values for arsenic, chromium, copper, and mercury, but no exceedances of effects range median (ERM) values at any of the River sites, with the exception of nickel. Nickel concentrations exceeded the ERM on every sampling occasion. It should be noted that there is low confidence in the current nickel guideline (Long et al., 1995). There were no exceedances of either ERL or ERM values for PAHs, PCBs, or pesticides. Although analysis of selected metals in unfiltered 1998 elutriates showed concentrations well below the effect limits for silver, cadmium, and zinc, the total copper concentrations exceeded the EC₅₀ value of 7.13 (Table 4.7, MPSL unpublished data).

TIE Results

TIE treatments were conducted on six concentrations of sediment elutriate. Because significant toxic responses and toxicity mitigation generally occur within one or two concentrations of elutriate, data are represented graphically only for concentrations where treatments mitigated toxicity (Figures 4.23 , 4.24 & 4.25). Results from the cation column eluate are not presented because this treatment had 0% survival due to over-acidification.

Unionized ammonia concentrations were below the effects threshold, but some pH levels were outside the acceptable range. Initial baseline pH values for Grizzly Bay and San Joaquin River were below the tolerance threshold for Mytilus. However, pH could not have been the only cause of toxicity because other treatments with higher pH values had similar toxic responses. Baseline concentrations of hydrogen sulfide were above the effect limits for Mytilus (0.0053 mg/L, Knezovich et al., 1997), but there was no mitigation of toxicity in the aeration or graduated pH manipulations, which would be expected if sulfide were the sole cause of toxicity.

In all samples the column eluate treatment showed significantly greater normal larval development relative to the baseline treatment indicating that non-polar organic chemicals were not eluted from the C18 column.

Grizzly Bay (BF21) TIE

Several treatments significantly reduced toxicity of the 25% concentration of this elutriate sample (Figure 4.23). Filtration, both EDTA treatments, the C18 column with and without EDTA, and the cation column treatments were all significantly different from the baseline treatment. Samples that were passed through the column treatments were all filtered. The C18 treatments were not significantly different from the filtration treatment, indicating that the pre-filtration step probably caused the reduction in toxicity in these treatments. The C18 column can also remove metal chelates that are relatively non-polar (U.S. EPA, 1991). The pre-filtered cation column treatment was significantly different from the filtration treatment indicating that it had further reduced toxicity beyond the filtration step. Reduction of toxicity by the cation column as well as the two EDTA treatments, suggests that divalent cations contributed to the toxicity in this sample.

Sacramento River (BG20) TIE

Toxicity was significantly reduced in the 25% concentration by the filtration treatment, both EDTA treatments, the sodium thiosulfate treatment, the C18 column, and the cation column (Figure 4.24a). Removal of toxicity with the EDTA treatments and the cation column suggest that divalent cations might be a cause of toxicity. Removal of toxicity with the sodium thiosulfate treatment suggests removal of an oxidant or metal. Sodium thiosulfate is a strong chelator of copper, cadmium, mercury, and silver chlorides (Hockett and Mount, 1996). Removal of toxicity by the filtration treatment, along with the pre-filtration steps of the column treatments suggest that contaminants might also be particle-bound, but when the 50% elutriate concentration is examined (Figure 4.24b), toxicity was significantly mitigated by both C18 column treatments and not the filtration treatment. Although the C18 column removed some toxicity, no compounds were eluted off the column in toxic concentrations.

San Joaquin River (BG30) TIE

The C18 column treatment and the cation column treatment significantly mitigated toxicity (Figure 4.25a). Although the filtration treatment and the EDTA treatments removed some toxicity, the differences were not statistically significant. The combined C18 column/EDTA treatment did not remove toxicity. The pre-filtration step of the column treatments might be a factor in contaminant removal, but the additional removal of toxicity by the cation column in the 50% elutriate concentration suggests divalent cations as a source of toxicity (Figure 4.25b). Partial reduction of toxicity by EDTA supports this hypothesis.

Sediment-Water Interface (SWI) Tests

In addition to sediment elutriate toxicity tests, mussel embryos were exposed to intact sediment cores collected from the River stations. In these SWI exposures, embryos are exposed in screen tubes on top of the sediment in order to investigate the toxicity of fluxed chemicals to an epibenthic organism. Previous SWI exposures at the River stations have detected significant toxicity to mussel embryos (Table 4.6). Toxicity of the sediment overlying water was reduced in these experiments with the addition of EDTA to the overlying water, indicating divalent metals were responsible for the observed abnormal development. Additional SWI exposures were conducted in the current study to confirm that fluxed chemicals were toxic to mussel embryos. Sediment overlying water in cores from the River stations again demonstrated significant toxicity in the April 1998 sampling period.

Chemical analyses of sediment elutriates have indicated that copper concentrations are within the range toxic to mussel embryos at these stations. In the current experiments, sediment overlying water from SWI cores was sampled to measure cupric ion concentrations. Cupric ion concentrations in 2 of 3 river samples were successfully determined using flow injection analysis coupled with chemiluminescence detection. These concentrations will be compared to results of laboratory dose-response experiments designed to determine the concentration of cupric ion toxic to mussel embryo-larval development. These experiments will allow us to determine if cupric ion activity in sediment overlying water exceeds the toxicity threshold for Mytilus embryos.

Discussion and Conclusions

Sediment elutriates and sediment-water interface exposures from the three River stations were all significantly toxic. Toxicity identification evaluation (TIE) treatments designed to mitigate metals toxicity reduced toxicity in all three samples. C18 solid-phase extraction also reduced toxicity in the Sacramento River sample, indicating non-polar organics may have contributed to toxicity. Although some toxicity was mitigated by the C18 column in the Sacramento River TIE, past bulk phase chemistry data for RMP Sacramento River sediment samples show low levels of measured organic contaminants. The pH value of Grizzly Bay and San Joaquin River elutriate samples were low enough to cause the observed toxicity at these sites, but manipulations of sample pH would have mitigated toxicity if pH was the only factor contributing to sample toxicity. Toxicity was also observed in SWI exposures where overlying water pH was within tolerance limits.

Previous chemical analyses of sediment elutriate samples indicated that metal concentrations were below the effect thresholds of the test organism. Samples from the current study contained concentrations of copper that were above the lowest observed effect concentration (LOEC) of 7 mg/L (MPSL, unpublished data). Other metals may have contributed to toxicity through additivity. Combinations of certain metals have been shown to be additive in their toxicity. Masnado et al. (1995) found that combinations of metals including cadmium, chromium, copper, nickel, and zinc with concentrations below National Pollutant Discharge Elimination System (NPDES) water quality permit limits were toxic to Ceriodaphnia dubia. The additive, synergistic, and antagonistic effects of metals on larval invertebrates is the subject of a current State Water Board study at MPSL.

The current TIE manipulations, combined with past experiments with Phase I TIEs and SWI exposures (Table 4.6), indicate divalent cations are the likely cause of toxicity at the three river delta sites. EDTA and the cation column treatments successfully removed toxicity to some degree in all three samples. The C18 column also removed toxicity from the Sacramento River sample. The additional study of cupric ion concentrations in sediment overlying water samples will help to confirm the role of cupric ions in river sediment toxicity. Because the influence of salinity manipulation is still unclear, additional experiments will also be conducted on freshwater elutriate, including cupric ion analysis, additional toxicity tests with freshwater organisms, and trace metal analyses of freshwater elutriates.

References

Anderson, B.S., J.W. Hunt, M.M. Hester, and B.M. Phillips. 1996. Assessment of sediment toxicity at the sediment-water interface. In G.K. Ostrander, (ed.). Techniques in Aquatic Toxicology. Lewis Publishers, Ann Arbor, MI.

Hockett, J.R. and D.R. Mount. 1996. Use of metal chelating agents to differentiate among sources of acute aquatic toxicity. Environ. Toxicol. Chem. 15:1687­1693.

Knezovich, J.P., D.J. Steichen, J.A. Jelinski, and S.L. Anderson. 1997. Sulfide tolerance of four marine species used to evaluate sediment and pore water toxicity. Bull. Environ. Contam. Toxicol. 57:450­457.

Long, E.R., D.L. MacDonald, S.L. Smith, and F.D. Calder. 1995. Incidence of adverse biological effects within ranges of chemical concentrations in marine and estuarine sediments. Env. Mgmt. 19:81­87.

Martin, M., K.E. Osborn, P. Billig, and N. Glickstein. 1981. Toxicities of ten metals to Crassostrea gigas and Mytilus edulis embryos and Cancer magister larvae. Mar. Poll. Bull. 9:305­308.

Masnado, R.G., S.W. Geis, and W.C. Sonzogni. 1995. Comparative acute toxicity of a synthetic mine effluent to Ceriodaphnia dubia, larval fathead minnow and the freshwater mussel Anodonta imbecillis. Environ. Toxicol. Chem. 14:1913­1920.

Tetra Tech, 1986. Recommended protocols for measuring selected environmental variables in Puget Sound. Prepared for the Puget Sound Estuary Program by: Tetra Tech Inc., Bellevue, WA.

U.S. EPA/ACOE. 1991. Evaluation of dredged material proposed for ocean disposal (testing manual). EPA-503/8-91/001, U.S. EPA Office of Water (WH-556F) and US Army Corps of Engineers, Washington, D.C.

U.S. EPA. 1991. Method for Aquatic Toxicity Identification Evaluations, Phase I Toxicity Characterization Procedures. U.S. EPA, ORD, EPA/600/6-91/003. Washington, DC.

U.S. EPA. 1996. Marine Toxicity Identification Evaluation (TIE), Phase I Guidance Document. U.S. EPA, ORD, EPA/600/R-95/054. Washington, DC.

Zamzow, H. 1997. Determination of copper complexation in California coastal waters using flow injection analysis. M.S. Thesis. San Francisco State University. 138p.