Lepidopteran‐active Bt‐toxins expressed in genetically modified (GM) maize events can adversely affect non‐target lepidopteran (NTL) species when larvae consume harmful amounts of Bt‐maize pollen deposited on their host plants. The briskaR‐NTL project, as commissioned by the European Food Safety Authority, aims to: (i) further develop a spatially‐explicit, generic modelling framework to estimate risks of Bt‐maize pollen to NTLs at landscape scale; and (ii) offer a user‐friendly tool to risk managers to inform decision‐making. Using a literature map and Expert Knowledge Elicitation (EKE), the initial briskaR package was expanded to integrate: (i) a wider range of dispersal kernels for maize pollen; (ii) the variability of exposure to Bt‐maize pollen over time through ToxicoKinetic‐ToxicoDynamic ecotoxicological models; (iii) sublethal effects; and (iv) multi‐year and cumulative effects of chronic exposure. A user‐friendly and public R Shiny application was developed to run the model. Two real‐life case studies were used to test the model in contrasting environmental conditions, and identify key factors that adversely affect larvae of the Common Swallowtail (Papilio machaon). These factors include: the slope of dose‐response for Bt‐toxicity; cumulative effects; impact of an additional stressor (the microsporidian parasite Nosema); and host plant distribution. Overall, the case studies confirm the effect of landscape patterns and crop management practices at local level, suggesting a case‐by‐case approach for the assessment of risks and their possible mitigation. Several sources of uncertainty have been considered, but remaining ones as well as population processes should be considered in future (risk) assessments. While the modelling approach and EKE enabled to address the scarcity of input data by translating uncertainties into wider confidence intervals, gathering additional laboratory and field data is required to support a more robust and ecologically meaningful assessment of impacts of Bt‐maize pollen on NTL and model validation.
This publication is linked to the following EFSA Journal article: http://onlinelibrary.wiley.com/doi/10.2903/j.efsa.2021.e190301/full