Biography: 

Red palm weevil is an ecologically, economically and socially important pest of date palms, causing millions of dollars in yield losses and tree removal costs. It affects the livelihoods, cultural heritage and food security of farming communities. Substantial progress has been made in understanding RPW physiology, spread, and impact on palm trees. However, there is a pressing need to develop mechanistic, process-based models of RPW that link environmental drivers, spatial–temporal spread, and quantitative damage and yield loss.

Early detection and risk estimation have been recognised as key factors in implementing RPW control strategies, but the challenges are exacerbated by unpredictable weather patterns and a mixture of natural and man-mediated dispersal routes. There remains a gap in providing real- and near-time predictions utilising the multi-modal data, accounting for uncertainties, and presented in a user-friendly way. To address the gap, we propose to develop a novel modelling framework based on models we constructed for other systems, including Fall Armyworm, Emerald Ash Borer and bark beetles. The framework will combine a statistical analysis using Machine Learning and statistical approaches with a mechanistic model involving dispersal kernels. Data from real-time RPW counts based on trapping and tree monitoring will be linked with long-term environmental records, as well as near-time weather predictions. The goal will be to construct risk maps highlighting areas of high suitability for the RPW occupancy in real- or near-time. Such maps and alerts will subsequently be linked to a user-friendly web- or mobile-based decision support tool.