Applying Weight-of-Evidence in Retrospective Ecological Risk Assessment When Quantitative Data Are Limited

Abstract

Retrospective ecological risk assessment attempts to identify likely causal agents to explain adverse effects that have occurred in ecological targets. It can never be decisive since it is post hoc and usually based on limited evidence that is rarely very quantitative. It can, nevertheless, be made more transparent, systematic, and logical, and less subjective. Based on human health epidemiological criteria we develop an approach that moves from systematic consideration of seven basic questions to assigning a likelihood of involvement of putative agents. The questions are: 1. Is there evidence that the target is or has been exposed to the agent? 2. Is there evidence for correlation between adverse effects in the target and exposure to the agent either in time or in space? 3. Do the measured or predicted environmental concentrations exceed quality criteria for water, sediment or body burden? 4. Have the results from controlled experiments in the field or laboratory led to the same effect? 5. Has removal of the agent led to amelioration of effects in the target? 6. Is there an effect in the target known to be specifically caused by exposure to the agent? 7. Does the proposed causal relationship make sense logically and scientifically? We identify 15 common scenarios of answers to the questions and illustrate the approach by reference to three real-world case studies (decline in benthos in a tropical marine bay, decline in fisheries in a temperate sea, decline in marine mollusc populations). The primary challenge in retrospective risk assessment is to make best use of the available evidence to develop rational management strategies and/or guide additional analyses to gain further evidence about likely agents as causes of observed harm.