Influence of data selection on aquatic ecotoxicity characterization factors for selected persistent and mobile substances

Abstract

Abstract Purpose This paper addresses the lack of ecotoxicity characterization factors (CFs) for persistent and mobile (PM) chemicals in life cycle assessment. The specific aims are (1) to provide CFs for 64 selected chemicals and benchmark them against the USEtox database, (2) to propose an ecotoxicity data harmonization strategy, and (3) to analyze the influence of ecotoxicity data sources and data harmonization strategies. Methods Sixty-four per- and polyfluoroalkyl substances (PFAS), triazines, and triazoles were selected. An ecotoxicity data harmonization strategy was developed to ensure data consistency. A screening strategy for including transformation products was also developed. Existing CFs in the USEtox organic substances database (version 2.01) were identified, and new CFs were calculated based on available data sources. The USEtox model (version 2.13) was used for calculations, and the collected ecotoxicity data from different sources were varied to test their influence. Results and discussion The ecotoxicity data harmonization strategy excluded a considerable amount of incompatible data. To the list of 64 chemicals, 3 transformation products were added. Out of the 67 chemicals, experimental ecotoxicity data were available for 47, leading to the use of in silico tools and average values to fill data gaps for 20 chemicals. CFs for 67 chemicals could thus be provided. Comparing CFs calculated based on experimental and estimated ecotoxicity data reveals that the estimated data leads to considerably different CF values. Conclusions The paper provides strategies for ecotoxicity data harmonization and the inclusion of transformation products. The limited availability of experimental ecotoxicity data and differing results of the applied estimation methods highlight the need for further development of estimation methods. In addition, further development in including transformation products is recommended, which is particularly relevant for PM chemicals.