Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{193036,
        author = {Rojins S Martin and Aljo Joseph and Ashik Shaji and Renny Thomas and Jintu Ann John},
        title = {Satellite-Driven Machine Learning Framework for Estimating Surface-Level PM2.5 Concentrations},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {9},
        pages = {4006-4012},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=193036},
        abstract = {Fine particulate matter (PM2.5) is one of the most critical air pollutants due to its ability to penetrate deep into the respiratory system, posing severe risks to human health. Reliable estimation of surface-level PM2.5 remains challenging, particularly in regions with sparse ground-based monitoring infrastructure. This study proposes a satellite-driven machine learning framework for estimating surface-level PM2.5 concentrations by integrating satellite radiance data, meteorological variables, and temporal dependency features. Key atmospheric parameters—including precipitation, boundary layer height, relative humidity, wind speed, and air temperature—are combined with lag-based PM2.5 features and principal component representations to enhance predictive robustness. A Random Forest regression model is employed to effectively capture complex non- linear relationships between the input features and PM2.5 con- centrations. The proposed model demonstrates strong predictive performance, achieving a coefficient of determination (R2) of 0.91 and a root mean square error (RMSE) of 0.165 on the test dataset. The results highlight the potential of satellite-assisted machine learning approaches as a cost-effective and scalable solution for air quality assessment, particularly in areas lacking dense monitoring networks.},
        keywords = {PM2.5 prediction, air pollution modeling, satellite radiance, meteorological features, Random Forest},
        month = {February},
        }