Approaches to assess and manage the risks of excessive contamination of food crops by toxic trace elements in agroecosystem soils exposed to diffuse contaminations
Abstract
Diffuse contamination of cultivated soils by toxic trace elements (TE) is increasingly becoming a serious concern when it comes to minimizing the impact on human health. Health authorities are multiplying the alerts demonstrating the risk of an excessive exposure of humans to toxic TEs (e.g for cadmium (Cd) and Nickel (Ni)). Therefore, regulation limits for TE in food products are becoming more strict, as illustrated for Cd (EC1323/2021).
This presentation will show how both experimental work and mechanistic + statistical modelling can be complementary to assess and to manage TE accumulation in edible parts of crops. The presentation will concentrate on Cd because of its high toxicity and because of the current issues for human health but the approach aims at being generic. We will first focus on the understanding and the prediction of TE phytoavailability in agricultural soils exposed to diffuse contamination, with a particular interest on TE sorption onto the soil solid phase and complexation in the soil solution. Then, we will concentrate on the ecophysiology of TE allocation to plant organs in order search for traits that can be used in plant breeding for selecting genotypes that less accumulate toxic TEs. Finally, we will present tools that we developed to predict the risk for a food crop to not comply with the European regulatory limit for Cd content. The tools rely on the knowledge of dominant mechanisms governing crop contamination by soil TEs and they use machine learning methods to predict the risk from few soil and plant input variables. The models demonstrate good predictive performances and they can be used for individual risk assessment or for deriving soil maps of risks. For durum wheat, the model is currently freely available as a web service (https://ispa.bordeaux.inra.fr/services/blesur/) and can readily be used by the agricultural sector.
Domains
Environmental SciencesOrigin | Files produced by the author(s) |
---|