2026/07/01 by Susan Wilson · 1 voice
Computer Science · Environmental Science · #Bayesian Modeling and Causal Inference #Mercury impact and mitigation studies #Oil Spill Detection and Mitigation
paper · doi:10.1093/inteam/vjag064
openalex publication_date 2026/07/01 · openalex created_date 2026/07/07 · openalex updated_date 2026/07/07
DOI: 10.1093/inteam/vjae011 Effective ecosystem management requires the integration of diverse knowledge sources and multiple stakeholder endpoints for fully informed decision making. Jermilova et al. (2025) have developed a Bayesian network relative risk model (BN-RRM) capable of bringing together such data to inform risk mitigation strategies and they demonstrate its application for the globally recognised toxic pollutant, mercury (Hg). The study team, for the first time, have integrated multiple Hg datasets and modeling studies into a single model framework for a large circumpolar watershed, the Mackenzie River Basin (MRB) in Canada, and use the model to assess environmental risk from Hg exposure at community relevant endpoints. The BN-RRM is visually explicit and effectively communicates the current knowledge of environmental Hg. The BN-RRM was subsequently used to demonstrate the efficacy of two risk mitigation strategies: the Minamata Convention on Mercury aimed at reducing atmospheric Hg emissions, and government guidance to limit consumption of larger fish. This work provides a flexible new tool for policymakers to quantitatively track the success of both local risk mitigation efforts and global measures, enabling effective evaluation. It also empowers stakeholders and raises awareness to support environmental improvement. Best Paper Award winner Una Jermilova.