article · FUDMA Journal of Sciences
Accurate mapping of fine particulate matter, or PM2.5, across Nigeria is constrained by scarce ground monitoring networks and the limitations of satellite and reanalysis data. To overcome this, a multi-source data fusion framework combines low-cost sensor data from Purple Air and Clarity networks with satellite aerosol measurements, trace gas observations, weather variables, and MERRA-2 reanalysis data. Using a Random Forest machine learning model evaluated through spatial cross-validation, the approach markedly reduces prediction errors compared to standalone MERRA-2 estimates. The model achieves lower root mean square errors, lower mean absolute errors, and improved agreement with ground measurements across monitoring stations. Feature analysis reveals that relative humidity is the most critical variable driving PM2.5 levels, followed by reanalysis estimates, ozone, nitrogen dioxide, and aerosol optical depth, providing an improved framework for estimating surface air pollution across the country.
Air pollution poses severe health risks, but tracking dangerous fine particles across Nigeria has been hindered by sparse monitoring infrastructure and inaccurate regional models. Combining readily available satellite data and low-cost local sensors through machine learning provides a more reliable method to track ground-level pollution. This helps environmental authorities and researchers better understand air quality variations even in areas lacking expensive reference monitors.
The framework demonstrates an applied, tested methodology that could enable environmental monitoring agencies, municipal planners, and public health organisations to generate higher-resolution air pollution maps. By relying on low-cost sensors and freely accessible satellite and reanalysis feeds, the tool offers an affordable monitoring pipeline. While successfully validated across regional stations, real-world deployment would require operational software engineering to translate the machine learning workflow into a continuous, real-time monitoring service.
AI-generated from the published abstract. Always read the original work before citing.
Accurate estimation of surface PM2.5 across Nigeria remains challenging because satellite AOD and global reanalysis products alone cannot adequately represent near-surface particulate concentrations owing to complex aerosol-meteorology interactions, regional transport, and limited ground observations for calibration. This study addresses these limitations by developing a multi-source data fusion framework based on the Random Forest (RF) algorithm that integrates low-cost sensor measurements, satellite observations, atmospheric composition, meteorological variables, and reanalysis data to improve PM2.5 estimation. Ground-based observations from Purple Air and Clarity sensors were combined with satellite-derived aerosol optical depth, atmospheric trace gases, meteorological variables, and MERRA-2 reanalysis products. The RF model was trained and evaluated using a spatial cross-validation (leave location out) framework and assessed using RMSE, MAE, coefficient of determination (R²), Index of Agreement (IOA), and correlation coefficient. The RF model consistently outperformed MERRA-2 across all monitoring stations, yielding substantially lower prediction errors (RMSE: 14-35 µg m⁻³ versus 30-126 µg m⁻³; MAE: 9-31 µg m⁻³ versus 18-89 µg m⁻³) and stronger agreement with observations (IOA up to 0.68). Whereas MERRA-2 produced large negative R² values at several locations, the RF model achieved improved predictive performance, including a positive R² of 0.29 in Lagos and higher correlation coefficients across most stations. Feature importance analysis identified relative humidity as the dominant predictor, followed by MERRA-2 PM2.5, O3, NO2, and AOD, highlighting the combined influence of aerosol hygroscopic growth, atmospheric chemistry, and regional transport. These findings demonstrate that multi-source machine learning data fusion substantially improves surface PM2.5 estimation over Nigeria with scalable framework...
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.33003/fjs-2026-1013-5627
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
