Application of multimedia models for screening assessment of long-range transport potential and overall persistence.

We propose a multimedia model-based methodology to evaluate whether a chemical substance qualifies as POP-like based on overall persistence (Pov) and potential for long-range transport (LRTP). It relies upon screening chemicals against the Pov and LRTP characteristics of selected reference chemicals with well-established environmental fates. Results indicate that chemicals of high and low concern in terms of persistence and long-range transport can be consistently identified by eight contemporary multimedia models using the proposed methodology. Model results for three hypothetical chemicals illustrate that the model-based classification of chemicals according to Pov and LRTP is not always consistent with the single-media half-life approach proposed by the UNEP Stockholm Convention and thatthe models provide additional insight into the likely long-term hazards associated with chemicals in the environment. We suggest this model-based classification method be adopted as a complement to screening against defined half-life criteria at the initial stages of tiered assessments designed to identify POP-like chemicals and to prioritize further environmental fate studies for new and existing chemicals.

Comparing estimates of persistence and long-range transport potential among multimedia models.

Overall persistence (Pov) and long-range transport potential (LRTP) of organic chemicals are environmental hazard metrics calculated with multimedia fate and transport models. Since there are several models of this type, it is important to know whether and how different model designs (model geometry, selection of compartments and processes, process descriptions) affect the results for Pov and LRTP. Using a set of 3175 hypothetical chemicals covering a broad range of partition coefficients and degradation half-lives, we systematically analyze the Pov and LRTP results obtained with nine multimedia models. We have developed several methods that make it possible to visualize the model results efficiently and to relate differences in model results to mechanistic differences between models. Rankings of the hypothetical chemicals according to Pov and LRTP are highly correlated among models and are largely determined by the chemical properties. Domains of chemical properties in which model differences lead to different results are identified, and guidance on model selection is provided for model users.