Managing human activity and marine mammals: A biologically based, relativistic risk assessment framework

Abstract

Presented here is a broadly applicable, transparent, repeatable analytical framework for assessing relative risk of anthropogenic disturbances on marine vertebrates, with the emphasis on the sound generating aspects of the activity. The objectives are to provide managers and action-proponents tools with which to objectively evaluate drivers of potential biological risk, to identify data gaps that limit assessment, and to identify actionable measures to reduce risk. Current regulatory assessments of how human activities (particularly those that produce sound) influence the likelihood of marine mammal behavioral responses and potential injury, rely principally on generalized characterizations of exposure and effect using simple, threshold-based criteria. While this is relatively straightforward in regulatory applications, this approach fails to adequately address realistic site and seasonal scenarios, other potential stressors, and scalable outcome probabilities. The risk assessment presented here is primarily based on a common and broad understanding of the spatial-temporal-spectral intersections of animals and anthropogenic activities, and specific examples of its application to hypothetical offshore wind farms are given. The resulting species- and activity-specific framework parses risk into two discrete factors: a population’s innate ‘vulnerability’ (potential degree of susceptibility to disturbance) and an ‘exposure index’ (magnitude-duration severity resulting from exposure to an activity). The classic intersection of these factors and their multi-dimensional components provides a relativistic risk assessment process for realistic evaluation of specified activity contexts, sites, and schedules, convolved with species-specific seasonal presence, behavioral-ecological context, and natural history. This process is inherently scalable, allowing a relativistic means of assessing potential disturbance scenarios, tunable to animal distribution, region, context, and degrees of spatial-temporal-spectral resolution.