Derivation and characterization of environmental hazard concentrations for chemical prioritization: a case study in...

Ongoing anthropogenic activities and analytical advancements yield continuously expanding lists of environmental contaminants. This represents a challenge to environmental managers, who must prioritize chemicals for management actions (e.g., restriction, regulation, remediation) but are often hindered by resource limitations. To help facilitate prioritization efforts, this study presents several strategies for deriving environmental hazard concentrations using publicly accessible data and open-source computational tools. Using a Great Lakes tributaries aquatic monitoring dataset as a case study, environmental hazard concentrations were obtained or derived for 334 organic chemicals. These concentrations were based on (1) current water quality guidelines; (2) apical screening values; (3) apical and (4) nonapical effect concentrations from the ECOTOXicology Knowledgebase; (5) in vitro effect concentrations from the ToxCast database; (6) cytotoxic burst concentrations collated from the Comptox Dashboard; (7) "estimated screening values" derived from modeled or estimated data and available from various regulatory and nonregulatory agencies; (8) pharmaceutical potency estimates from the MaPPFAST database; and (9) quantitative structure-activity relationship (QSAR)-derived acute toxicity estimates. Environmental fate data included aquatic half-lives and bioconcentration factors collated from the Comptox Dashboard or estimated using QSARs. To identify patterns that could be used for characterization, availability of ecotoxicological concentrations and environmental fate data were evaluated. Furthermore, exceedances of hazard concentrations were evaluated and compared across diverse ecotoxicological data types. Altogether, by providing detailed methodology and practical examples generated with real monitoring data, this study demonstrated that these hazard concentration derivation strategies can be efficiently and effectively used with large, complex datasets and identified critical considerations for future prioritization efforts.