Challenge 4. Acquiring and deriving data for predictive modelling in environmental risk assessment
Risk ecological models to assess the impact of pollutants in scenarios of social and environmental relevance require quality input data, often quantifying the potential interaction and mobility of pollutants in environmental compartments. Key information derives from sorption data, which usually lead to the estimation of solid-liquid distribution coefficients (Kd). Internationally relevant models use single Kd values as input data, but they also increasingly require parametric models or distribution functions to better asses the risk and its associated uncertainty. Our research group has a strong expertise in sorption and desorption studies of pollutants in environmental samples. The performance of sorption and desorption experiments with a well-characterized set of soils, and the creation of datasets also including literature-gathered data, have permitted to elucidate sorption mechanisms and soil properties governing sorption for a certain number of radionuclides and organic pollutants, and also to propose Kd predictive models and derive Kd best-estimates. Moreover, the use of chemical analogues for a target pollutant, or the use of soil analogue environmental matrices, allow completing sorption data gaps to enhance the knowledge of the sorption of the target pollutant in soils or other environmental matrices. The successful results derived from this overall approach methodology encourage its application, in the next future, to some naturally occurring radionuclides lacking sorption information and relevant for human dose predictions, such as U, Ra or Po isotopes, both in the sort- and the long-term. Regarding the modelling step, it will be of interest to examine the comparability of the derived models with more complex ones such as geochemical-based models or models based on mechanistic surface complexation and ion exchange.
|
|
|
