The pros and cons of ecological risk assessment based on data from different levels of biological organization

Ecological risk assessment (ERA) is the process used to evaluate the safety of manufactured chemicals to the environment. Here we review the pros and cons of ERA across levels of biological organization, including suborganismal (e.g., biomarkers), individual, population, community, ecosystem and landscapes levels. Our review revealed that level of biological organization is often related negatively with ease at assessing cause-effect relationships, ease of high-throughput screening of large numbers of chemicals (it is especially easier for suborganismal endpoints), and uncertainty of the ERA because low levels of biological organization tend to have a large distance between their measurement (what is quantified) and assessment endpoints (what is to be protected). In contrast, level of biological organization is often related positively with sensitivity to important negative and positive feedbacks and context dependencies within biological systems, and ease at capturing recovery from adverse contaminant effects. Some endpoints did not show obvious trends across levels of biological organization, such as the use of vertebrate animals in chemical testing and ease at screening large numbers of species, and other factors lacked sufficient data across levels of biological organization, such as repeatability, variability, cost per study and cost per species of effects assessment, the latter of which might be a more defensible way to compare costs of ERAs than cost per study. To compensate for weaknesses of ERA at any particular level of biological organization, we also review mathematical modeling approaches commonly used to extrapolate effects across levels of organization. Finally, we provide recommendations for next generation ERA, submitting that if there is an ideal level of biological organization to conduct ERA, it will only emerge if ERA is approached simultaneously from the bottom of biological organization up as well as from the top down, all while employing mathematical modeling approaches where possible to enhance ERA. Because top-down ERA is unconventional, we also offer some suggestions for how it might be implemented efficaciously. We hope this review helps researchers in the field of ERA fill key information gaps and helps risk assessors identify the best levels of biological organization to conduct ERAs with differing goals.