Correlative/Observational Studies

ERDC TN-DOER-C23

September 2001

their persistence, equilibrium partitioning, and kinetics" (Gentile et al. 1988). Comparison of tissue

residues from these laboratory-based bioassays with field-collected animals indicated good corre-

lation between laboratory and field exposures for both N. incisa (Figure 2a) and M. edulis

(Figure 2b). The author suggests based on these results that both tests offered reasonable predictions

of potential bioaccumulation in the field, at least with regard to PCBs.

Correlative/Observational Studies. A number of studies published by Swartz et al. have used

the correlative approach described previously to demonstrate the utility of solid-phase acute

amphipod bioassays in predicting the existing conditions of the infaunal community associated with

in-place contaminated sediments (Swartz et al. 1982, 1985, 1986, and 1994). In the 1982 study,

the authors evaluated sediment toxicity of 175 sediment samples collected from Commencement

Bay, Washington, to the amphipod, Rhepoxynius abronius. Amphipod survival (10-day) was

compared with sediment chemistry and community structure data of in-place sediments. Results

indicated a significant correlation between amphipod distribution and sediment toxicity.

In studies published in 1985 and 1986, Swartz et al. evaluated the temporal and spatial extent of

impacts from the Los Angeles County Sanitation Districts' sewage outfalls on the benthic commu-

nity of the Palos Verdes Shelf. Sampling surveys were conducted in the late spring/early summer

of 1980 and 1983. A total of eight sampling stations located at increasing distances from 1 to

15 km downcurrent of the outfall diffusers along the 60-m depth contour were sampled. In the

1980 sampling Swartz et al. found that the three stations closest to the outfall (stations 1-3) showed

statistically significant acute toxicity relative to the control (98 percent survival). There were no

differences in survival among these three stations (average survival equaled 80 percent). In the

1983 sampling survey, survival ranged from 83 to 97 percent for seven of the eight sampling stations

(station 2 was not sampled) with no differences relative to control (97 percent). The authors showed

that the reduction in acute toxicity to the amphipod R. abronius between 1980 and 1983 correlated

with a recovery in the macrobenthos and a reduction in the mass emission of biological oxygen

demand and chemical contaminants from the outfall. However, Swartz et al. were careful to caution,

"The absence of acute sediment toxicity does not demonstrate the absence of benthic degradation."

The authors noted that while some amphipods appeared in samples for the first time in the 1983

survey, phoxocephalid amphipods were no closer to the outfall than station 4 (approximately 5 km

downcurrent of the outfall diffuser). This finding suggests that effects were occurring at contami-

nant concentrations lower than would produce effects in the acute tests.

In a similar study published in 1994, Swartz et al. evaluated sediment toxicity to the amphipod

Eohaustorius estuarius, contaminant concentrations, and amphipod abundance along a contamina-

tion gradient in Richmond Harbor, upper San Francisco Bay. The predominant contaminants were

dichlorodiphenyltrichloroethane (DDT) and dieldrin, which were produced at a facility in the area

from 1946 to 1966. Nine stations showing a gradient in ΣDDT (11.6 to 77,700 g/dry kg) and dieldrin

(0.63 to 748 g/dry kg) concentrations were evaluated for amphipod toxicity and amphipod

abundance. An evaluation of the ratio of acid volatile solids to simultaneously extracted metals, and

concentrations of dieldrin, ΣPAHs, and Arochlor 1254 suggested that concentrations of these

constituents were too low to affect toxicity in E. estuarius and that the concentrations of ΣDDT

were sufficient to result in acute toxicity. Further, the authors found a correlation (R2 = 0.72, p = 0.004

for the Lauritzen Channel/Richmond Harbor sediments) between increasing carbon-normalized