Data and R code for Modeling environmental suitability for pathogenic Phytophthora species associated with wild olive dieback to support disease surveillance and management in Mediterranean forests

This repository contains the data, metadata, R code, and analytical outputs supporting the manuscript “Modeling environmental suitability for pathogenic Phytophthora species associated with wild olive dieback to support disease surveillance and management in Mediterranean forests”. The study evaluates the influence of spatial grain on the modeling of relative environmental suitability for pathogenic Phytophthora taxa associated with wild olive (Olea europaea var. sylvestris) dieback in central Sardinia, Italy. Following spatial thinning, 55 occurrence records were analyzed using MaxEnt models across five spatial resolutions ranging from 10 to 50 m. Predictor selection was conducted independently at each resolution, and model complexity was optimized through systematic parameter tuning and repeated spatial block cross-validation. The repository includes the occurrence dataset used for model calibration, supporting metadata, R scripts for spatial thinning, environmental predictor processing, variable selection, model tuning, performance evaluation, variable-importance analysis, and suitability-map production, together with the principal analytical outputs. Environmental datasets obtained from third-party providers are not redistributed where their licensing conditions do not permit redistribution. Their original sources, access information, and the procedures used to derive the environmental predictors are documented in the accompanying files. Model outputs should be interpreted as indicators of relative environmental suitability and surveillance priority, rather than as absolute probabilities of pathogen occurrence, disease risk, or pathogen spread. Associated manuscript: [manuscript citation or DOI to be added when available].