¹Department of Life Sciences and Chemistry, Roskilde University Universitetsvej 1, PO Box 260,DK-4000 Roskilde, Denmark
²Department of Animal and Plant Science, The University of Sheffield, Sheffield S10 2TN, U.K

Abstract

  Sediments are complex systems, and contaminants interact with them in a multitude of ways that influence exposure and effects on sediment-dwelling biota. Likewise a variety of non-chemical agents may act on benthic communities; these may dominate, contribute to or mask any effects from chemical contaminants. The first problem is to recognize that adverse effects have occurred. Once effects have been convincingly identified it is then necessary to assess likely causal agents retrospectively. Here we present a series of questions, elaborated from human health epidemiology, that can help to structure retrospective sediment assessments. We propose a method that aims to guide interpretation of various combinations of answers to the questions so that conclusions about the likelihood that identified agents have caused the observed effects in sediment systems can be consistently drawn. We demonstrate the approach by applying it to two published case studies. The first deals with estuarine communities in the northern Gulf of Mexico exposed to a wide variety of contaminants originating from a range of sources. The second involves a freshwater stream community impacted by road runoff. Although simpler than other weight of evidence approaches, we believe that the method provides a systematic, explicitly documented and consistent approach that may be particularly effective for defining priorities in situations where the evidence is limited and/or at least partly qualitative.