Citation:

Raimondo, Sandy, K. Eisenreich, M. Elias, K. Fay, M. Jankowski, K. Koehrn, A. Latier, AND C. Lilavois. Application of Interspecies Correlation Estimation (ICE) models in EPA’s regulatory frameworks. SETAC North America 43rd Annual Meeting, Pittsburgh, PA, November 13 - 17, 2022.

Impact/Purpose:

This presentation will demonstrate the application of ORD's Web-based Interspecies Correlation Estimation (Web-ICE; www3.epa.gov/webice) throughout EPA's chemical evaluation processes. This presentation is developed in collaboration with the Office of Pollution Prevention and Toxic Substances, Office of Water, and Region 10 to demonstrate how various EPA regulators apply ICE models under the Clean Water Act, Endangered Species Act, and Toxic Substance Control Act. 

Description:

A significant challenge in effects assessments conducted for chemical evaluations and deriving protective ecological values is the limited amount of data available for evaluating a chemical’s toxicity to the diversity of taxa found in the environment. Species sensitivity data are often limited to standard surrogate species, but may be absent entirely for some chemicals. Often, chemical evaluations either cannot be completed or are based solely on the measured or modeled (e.g., quantitative structure activity relationships) sensitivity of just a few species. In these cases, hazard thresholds are typically calculated using a single, most sensitive value, an approach that is heavily reliant on the amount of data available. Interspecies Correlation Estimation (ICE) models are log-linear regressions of the acute toxicity tested in two species. ICE models are developed from a standardized database of diverse species sensitivity and represent the relationship of inherent sensitivity of the two species. In application, the sensitivity of a surrogate species (represented by the x-axis) is entered into the ICE model to estimate a value of the predicted species, genus, or family (represented by the y-axis). There are 4000 ICE models available on the US EPA’s Web-ICE application (www3.epa.gov/webice) that predict to aquatic vertebrates and invertebrates, algae, terrestrial mammals and birds. From this application, a few surrogate species can estimate toxicity to over a hundred diverse taxa. As the international ecological risk assessment community of practice moves away from whole animal testing and towards New Approach Methodologies, ICE models have great potential to increase toxicity data available for a diversity of taxa that can inform chemical evaluations. This taxonomic expansion allows for probabilistic approaches of calculating hazard thresholds and data-driven ways to consider uncertainty. In this presentation, we demonstrate the application of ICE models within the US Environmental Protection Agency’s regulatory framework and highlight examples supporting the Toxic Substances Control Act, the Clean Water Act, and the Endangered Species Act. 

URLs/Downloads:

Record Details: