The U.S. Environmental Protection Agency (EPA) defines emerging contaminants as “chemicals that have not previously been detected in water, or that are being detected at significantly different levels than expected.” These potential pollutants include pharmaceuticals, microplastics, and endocrine disrupting chemicals caused by industrial land use and agricultural runoff. Researchers and government agencies warn that these chemicals may pose adverse health and ecological effects.

Only a fraction of these contaminants have been extensively evaluated, but our project aims to address this. Our study will explore how new data and metadata standards can be used to harmonize diverse environmental health information. Integrating a variety of data types in this way could help other researchers investigate drinking water contaminants and their associated impact on human health. To extract and integrate these data types, we will deploy artificial intelligence (AI) techniques like natural language processing (NLP) and machine learning. We also plan to build an environmental exposure knowledge graph, and engage with users to evaluate the impact of our project.