|UM SPH Home > Faculty & Research > Environmental Health Sciences > Jolliet Lab - Impact and Risk Modeling > Research > USEtox: UNEP/SETAC toxicity model fo rthe comparative assessment of chemicals|
Jolliet Lab - Impact and Risk Modeling (iMod)
USEtox: UNEP/SETAC toxicity model fo rthe comparative assessment of chemicals
About this Project
USEtox is a multimedia fate and exposure model developed to consistently compare the environmental impacts of chemicals on human health and ecosystems. Within the SETAC-UNEP Life Cycle Initiative, an international comparison of six models first enabled us to identify the crucial fate, exposure and effect issues that made models differ. A parsimonious (as simple as possible but as complex as needed) and transparent consensus model was then created within a matrix framework, containing only the most influential model elements. Covering more than 2000 substances, USEtox provides human toxicity and aquatic ecotoxicity characterization factors – a comparative measure of the impact per kg emitted substance into (urban) air, freshwater and agricultural soils. The present study applies the model to analyze the major factors affecting both human toxicity and ecotoxicity impacts, with illustrations for key chemicals. The fate part relates emission flows to masses in the environment (e.g. transport to water and residence time in water) and is common to both human- and eco-toxoxicity assessment, with typically 5 orders of magnitude variations between chemicals. For human toxicity, exposure factors multiply by fate to yield the intake fraction – the fraction of the emission that is taken in by the human population. Intake fractions typically range from 10-2 for POPs and metals that bioconcentrate in the food chain down to 10-8 for the inhalation of shortly degraded VOCs. Effect measures for both human- and exo-toxicity are based on the best estimate of the hazard concentration that generates 50% of response for humans and 50% of species affected or disappeared for ecosystems. As a whole the characterization factors vary by 12 orders of magnitude for human health and 10 for ecotoxicity. A regression analysis shows that for an emission to water, there is a highly significant (P<0.0001) but low correlation (R2=0.14) between human- and eco-tox characterization factors, mainly linked to the common fate part of the analysis.