Presentation Information
[U08-01]Using Machine Learning Models to Track Legacy and Emerging Contaminants
*Michael S Bank1 (1.Institute of Marine Research, Bergen, Norway)
Keywords:
Legacy & Emerging contaminants,Machine Learning,Modeling,Science-Policy Interface,Complex Systems
Using Machine Learning Models to Track Legacy and Emerging Contaminants
Global environmental pollution from legacy and emerging contaminants continues to be a critical issue of concern and represents a grand planetary health challenge that requires a strengthening of the science-policy interface. Advances in computing capabilities have allowed scientists to address larger and more complex datasets that consider a wider scope of variables and potential drivers and outcomes of pollution. Here using machine learning models and stable isotope tracer biogeochemistry analyses I present and synthesize pollution case studies from both land and the ocean and that consider legacy and emerging contaminants in a wide array of sampling matrices across several ecosystem compartments and types. The role of machine learning based contaminant ensemble models are discussed in the context of planetary health assessments and in support of the relevant sustainable development goals and global chemical governance frameworks (UNEP-ISP-CPW, and Basel/Rotterdam/Stockholm/Minamata Conventions).
Global environmental pollution from legacy and emerging contaminants continues to be a critical issue of concern and represents a grand planetary health challenge that requires a strengthening of the science-policy interface. Advances in computing capabilities have allowed scientists to address larger and more complex datasets that consider a wider scope of variables and potential drivers and outcomes of pollution. Here using machine learning models and stable isotope tracer biogeochemistry analyses I present and synthesize pollution case studies from both land and the ocean and that consider legacy and emerging contaminants in a wide array of sampling matrices across several ecosystem compartments and types. The role of machine learning based contaminant ensemble models are discussed in the context of planetary health assessments and in support of the relevant sustainable development goals and global chemical governance frameworks (UNEP-ISP-CPW, and Basel/Rotterdam/Stockholm/Minamata Conventions).