Publications : 2025

Bare JL, Fender CL, Torres C, Panko J. Assessing environmental monitoring data quality and reliability in TSCA risk evaluations using 4-tert-octylphenol as a case study. Abstract 7.05.P-Mo-116, SETAC North America 46th Annual Meeting, Portland, OR, November 2025.

Abstract

Environmental and human health risks are determined for existing chemicals undergoing an EPA TSCA risk evaluation. One component of this risk assessment is evaluating environmental concentrations with via monitoring datasets, gray and peer-reviewed literature, and modeling. Here, we spatially and temporally assessed data available in EPA’s Water Quality Portal (WQP) for use in TSCA risk evaluations using 4-tert-octylphenol (4-tOP) as a case study. Data from 2012 through 2022 (the last year of available data) were evaluated based on the Draft TSCA Systematic Review Protocol and prior risk evaluations. The groundwater, surface water, and sediment datasets in WQP were moderate to large in size (i.e., >600 samples) but were highly censored, with detection frequencies less than 10%. The soil dataset was very limited (approximately a dozen samples), but it had the highest detection frequency, approximately 33%. Across all media, two of the maximum values were equal to the detection limit for the sample. Overall, the highly censored datasets limit fate and transport analyses based on 4-tOP chemical properties and meaningful spatial or temporal comparisons. As with most monitoring databases, complete details for a quality and reliability assessment may not be readily available or may require thorough review of database parameters, particularly for a database such as the WQP, which combines several reporting organizations. Although each TSCA existing chemical risk evaluation is meant to be fit-for-purpose and rely on a weight-of-evidence approach that includes more than WQP data, a standard approach is needed for determining environmental concentrations, particularly when deciding whether to (1) perform a qualitative or quantitative assessment or (2) rely on measured or facility-specific modeled data. Finally, when characterizing environmental concentrations, a standard approach is needed for handling non-detects in highly censored datasets, because the handling method may impact EPA’s determination of unreasonable risk.