the gradient and the potential of these features to affect recruitment patterns and potentially confound
interpretation of benthic infaunal data. A principal assumption of the correlative approach is that
the primary factor affecting differences in benthic community structure and/or function along the
gradient is the level of chemical contamination and that all other factors have minimal to no influence
on the benthic community. One way to account for differences associated with physical location
might be a hybrid of the correlative and experimental/manipulative approaches where representative
sediments from each of the sites along the gradient are placed in containers at selected locations
along the gradient. In this way, factors associated with a given location would presumably act
equally upon all sediment types placed at the location, and effects related to contaminant loading
could be discriminated.
Another important consideration is selecting a field site and measuring end points that are relevant
to the lab bioassay being validated. The ideal would be to have the test organism be an important
if not critical component of the benthic community at the selected field locations. While this may
not be possible or even necessary, it is important that the end point/test organism being evaluated
at least have relevance to the individual field location. For example, one probably would not want
to validate a laboratory test with a test animal that is typically found in fine-grain material at a site
where coarse material predominates.
Finally, the potential for physical alterations of the site via wave action during such events as storms
or boat traffic should be considered as part of the site selection process in both correlative and
experimental studies. The purpose of the field validation effort is to establish whether or not
contaminant-induced responses in the laboratory are reflective of contaminant-induced effects in
the field. Having a major storm mobilize or bury all the contaminated sediments at a test location
obviously affects the ability of the study to validate the laboratory test.
In the field, a variety of physical and chemical processes act upon both sediment-associated
contaminants and organisms to constantly alter and affect exposure in a very dynamic manner.
However, in the laboratory the sample is intentionally kept under very controlled conditions to
minimize changes in these physical and chemical processes on exposure. In addition, animals in
the field are subject to competition, predation, and a host of other "naturally occurring stressors"
while these stressors are minimized or eliminated in laboratory-based assays. It has also been argued
by Swartz (Swartz et al. 1986) and others that acute laboratory-based toxicity tests with single
organisms cannot account for degraded benthic communities at lower levels of contamination
because changes in benthic community structure at these lower levels of contamination are driven
by longer term, sublethal effects. Chronic effects on growth or reproduction may result in shifts in
benthic community structure and function. Consequently, it is possible that chronic tests may better
predict potential changes in field-exposed communities.
Chronic sublethal measures of sediment toxicity are required under the existing regulatory frame-
work for the evaluation of dredged material. Because these tests are more costly and complex to
conduct, it is important to ensure that the responses in these chronic sublethal tests correlate with
effects measured in the field prior to their regulatory implementation. The application of a properly
designed field validation effort offers the best opportunity for establishing confidence in these