reliability-aware pm2.5 mapping in africa via satellite-reanalysis fusion and sparse monitors

Africa's sparse ground-based monitoring limits exposure assessment for fine particulate matter (pm2.5). this work presents a pan-African PM2.5 mapping pipeline that fuses public ground observations with satellite and reanalysis covariates and produces reliability-aware uncertainty for decision support. Using 2,068,901 quality-controlled PM2.5 records from 404 monitoring locations across 29 african countries (2016-2025), the model integrates aerosol optical thickness, satellite NO₂, planetary boundary layer height, meteorology, and population density. Under leakage-resistant 5-fold location-grouped spatial cross-validation, lightgbm achieves rmse 30.83 +/- 5.07 ug/m3 and r^2 0.134 +/- 0.023 with stronger aqi-style classification balance, while XGBoost yields slightly better regression accuracy. split-conformal prediction targeting 90% marginal coverage reveals strong regional heterogeneity, with severe degradation in east africa consistent with covariate shift. The release includes deterministic reliability flags, monitor prioritization, and out-of-fold Shap analyses to communicate when and why predictions should not be trusted.