Multi-Model Fire-Risk Mapping in a Fuel-Limited Tropical Forest Reserve: A Remote-Sensing Assessment of Gambari Forest Reserve, Nigeria

This deposit contains the data and analysis code supporting the study "Multi-Model Fire-Risk Mapping in a Fuel-Limited Tropical Forest Reserve: A Remote-Sensing Assessment of Gambari Forest Reserve, Nigeria." It provides the reproducibility notebook together with the input layers and generated outputs needed to reproduce every table and figure in the manuscript and its supplementary materials. Inputs include MODIS Collection 6.1 active-fire detections for Gambari Forest Reserve, Nigeria (2001–2025), the reserve boundary, ten environmental criterion rasters (vegetation, surface temperature, terrain, distance, and NASA POWER climate layers), the literature-weighted AHP risk surface, annual NASA POWER climate series, and a de-identified stakeholder questionnaire (n = 55). The Python notebook performs the fire-regime characterisation, AHP construction and validation, data-driven modelling (logistic regression, random forest, gradient boosting, and maximum entropy), pseudo-absence sensitivity analysis, spatial point-pattern and autocorrelation tests, and the climate–fire temporal models, writing all result tables (CSV) and figures (PNG/PDF) to an output directory. Running the notebook end to end regenerates every quantitative result reported in the paper. Associated article: [DOI to be added on acceptance]. Please cite both the article and this deposit.