: an R package for modelling and mapping risk of crop pest development based on their thermal biology
Source: PubMed Central Open Access, NCBI / U.S. National Library of Medicine
Under ongoing global warming and recent crop pest invasions, there is an urgent need to fill the gap between experimental research on pest thermal biology and applied forecasting tools that inform pest-management policies. The R packageprovides an accessible, open-source workflow that enables researchers, particularly those working in experimental pest biology, to forecast pest risk based on collected data on the thermal response of pest development rates. Built upon recent advances in open-source software development in the R environment,offers an accessible pipeline that spans from fitting performance curves for development rates data to generating broad-scale pest risk maps. Starting with user-provided temperature-dependent life-history dataset, the package fits nonlinear regression models and visualizes their thermal performance curves, with Akaike Information Criterion (AIC) scores and bootstrapped uncertainty ribbons. This allows users to select the most appropriate model based on biological, ecological and statistical criteria. The package then calculates thermal boundaries around the optimal region of the selected curve. These boundaries are used to extract climatic data for a user-defined spatial region or country and to map the number of months per year with optimal temperatures for pest development. The output consists of static or interactive raster maps that provide valuable insights for pest risk based on the known thermal biology of the target pest. This workfl
Abstract
Under ongoing global warming and recent crop pest invasions, there is an urgent need to fill the gap between experimental research on pest thermal biology and applied forecasting tools that inform pest-management policies. The R packageprovides an accessible, open-source workflow that enables researchers, particularly those working in experimental pest biology, to forecast pest risk based on collected data on the thermal response of pest development rates. Built upon recent advances in open-source software development in the R environment,offers an accessible pipeline that spans from fitting performance curves for development rates data to generating broad-scale pest risk maps. Starting with user-provided temperature-dependent life-history dataset, the package fits nonlinear regression models and visualizes their thermal performance curves, with Akaike Information Criterion (AIC) scores and bootstrapped uncertainty ribbons. This allows users to select the most appropriate model based on biological, ecological and statistical criteria. The package then calculates thermal boundaries around the optimal region of the selected curve. These boundaries are used to extract climatic data for a user-defined spatial region or country and to map the number of months per year with optimal temperatures for pest development. The output consists of static or interactive raster maps that provide valuable insights for pest risk based on the known thermal biology of the target pest. This workflow contributes to making pest forecasts open, reproducible and accessible to the scientific community, while also providing relevant information for policy-making institutions and plant-protection organizations involved in crop-pest management.