| Abstract: | Sediment contamination and toxicity were monitored at 14 sites in San Francisco Bay between 1991 and 1996. Sediment contamination patterns were different in the major reaches of the Bay, and at each site. Several contaminants were consistently above concentrations previously associated with toxicity at most sites. Bulk sediment bioassays using the amphipod Eohaustorius estuarius and sediment elutriate bioassays using larval bivalves (Mytilus spp., Crassostrea gigas) also indicated different patterns of sediment toxicity in space and time. Sediments were most toxic to the amphipods at Redwood Creek (90% of the tests), and were toxic in at least half the tests conducted at five other sites. Sediment elutriates severely reduced normal bivalve larval development at the San Joaquin and Sacramento Rivers in all samples, but toxicity occurred in less than a third of the tests in the Central and South Bays. Toxicity could not be statistically related to seasonal freshwater flow or rainfall in the Bay, but seasonal variation in contaminant concentrations and toxicity was observed. Amphipod toxicity was inversely and significantly related to the mean effects range-median quotient, suggesting that cumulative concentrations of several contaminants were related to toxicity. Further analysis identified suites of specific contaminants at each site that were variably related to amphipod toxicity at each site. Chlordanes, cadmium, and silver were significantly related to amphipod survival in the North Bay. Seasonal patterns in low, and high molecular weight polycyclic aromatic hydrocarbons (PAHs) were related to toxicity at Alameda, and metals and PAHs were related to toxicity at Castro Cove. Larval bivalve toxicity was associated with metals in bulk sediments, but elutriate chemistry was not measured, and relationships with toxicity could not be examined. Hypotheses about effective concentrations of several individual contaminants and mixtures of contaminants were posed. |
