Assessing interactions of multiple stressors when data are limited: A Bayesian belief network applied to coral reefs

Bayesian belief networks are finding increasing application in adaptive ecosystem management where data are limited and uncertainty is high. The combined effect of multiple stressors is one area where considerable uncertainty exists. Our study area, the Great Barrier Reef is simultaneously data-rich - concerning the physical and biological environment - and data-poor - concerning the effects of interacting stressors. We used a formal expert-elicitation process to obtain estimates of outcomes associated with a variety of scenarios that combined stressors both within and outside the control of local managers. There was much stronger consensus about certain stressor effects - such as between temperature anomalies and bleaching - than others, such as the relationship between water quality and coral cover. In general, the expert outlook for the Great Barrier Reef is pessimistic, with the potential for climate change effects potentially to overshadow the effects of local management actions.