MULTISENSOR FUSION Canopy Height Modelling and Mapping (2020 and 2025)

This repository provides a reproducible Python/Google Colab workflow for GEDI-based canopy-height modelling and mapping using conventional multisensor Earth observation predictors across Southwestern Nigeria for 2020 and 2025. It integrates GEDI RH98 observations with Sentinel-1, Sentinel-2, ALOS PALSAR, terrain, soil, and climate predictors and uses XGBoost regression to develop continuous canopy-height models. The workflow includes data extraction and preprocessing, upper-tail filtering, five-fold cross-validation, out-of-fold evaluation, model training, and wall-to-wall canopy-height prediction. The same modelling framework is applied to both years to support consistent comparison of canopy-height patterns and changes between 2020 and 2025.