Characterising the Spatial and Temporal Variability of PM2.5 Air Pollution in Lagos Using Satellite Aerosol Products and Ground Measurements

Air quality in Lagos is severely degraded by vehicular emissions, industrial activities, open burning, and Saharan dust intrusion, yet the spatial and temporal structure of PM2.5 pollution remains poorly characterised. This study comprehensively characterises PM2.5 variability across Lagos metropolis using a combination of satellite aerosol optical depth products, chemical transport model outputs, and a network of low-cost PM2.5 ground sensors. MODIS MAIAC AOD and Sentinel-5P TROPOMI aerosol products are calibrated against data from 18 low-cost PurpleAir sensors deployed across Lagos over a 24-month monitoring period. A spatiotemporal land use regression model incorporating road density, population, land use variables, and satellite AOD predictors is developed to produce monthly PM2.5 concentration maps at 100-metre resolution. Seasonal decomposition reveals distinct pollution regimes including a harmattan-dominated winter peak, a wet season minimum, and a persistent local emission signal throughout the year. Spatial disparity analysis identifies the highest chronic exposures in densely populated low-income areas of Mushin, Agege, and Ajegunle. Keywords: PM2.5, air pollution, Lagos, aerosol optical depth, satellite monitoring