PM2.5 Estimation using Machine Learning Models and Satellite Data: A Literature Review

Abstract

Most researchers are beginning to appreciate the use of remote sensing satellites to assess PM 2.5 levels and use machine learning algorithms to automate the collection, make sense of remote sensing data, and extract previously unseen data patterns. This study reviews delicate particulate matter (PM 2.5 ) predictions from satellite aerosol optical depth (AOD) and machine learning. Specifically, we review the characteristics and gap-filling methods of satellite-based AOD products, sources and components of PM 2.5 , observable AOD products, data mining, and the application of machine learning algorithms in publications of the past two years. The study also included functional considerations and recommendations in covariate selection, addressing the spatiotemporal heterogeneity of the PM 2.5 -AOD relationship, and the use of cross-validation, to aid in determining the final model. A total of 79 articles were included out of 112 retrieved records consisting of articles published in 2022 totaling 43 articles, as of 2023 (until February) totaling 19 articles, and other years totaling 18 articles. Finally, the latest method works well for monthly PM 2.5 estimates, while daily PM 2.5 and hourly PM 2.5 can also be achieved. This is due to the increased availability and computing power of large datasets and increased awareness of the potential benefits of predictors working together to achieve higher estimation accuracy. Some key findings are also presented in the conclusion section of this article. Keywords-AOD; machine learning; PM 2.5; remote sensing; pollutant Eligibility: Eligibility was determined by reading the main findings, use of data, results, and discussion. The authors considered journal articles and books published by reputable publishers as high-quality research and included them in summary. The authors used "Scimago Journal & Country Rank" to check the rankings of the included articles. Inclusion: The research then lists literature articles that correspond to the main topic.Our initial search yielded 112 articles. After passing the initial screening to eligibility assessment, this study used 61 primary and 18 other articles.

Knowledge Management System For Forest and Land Fire Mitigation in Indonesia: A Web-Based Application Development

“…This stage produces knowledge of the types of mitigation that can be done to reduce the risk of impacts from forest and land fires (karhutla), such as prevention, preparedness, response, recovery, and collaboration and co-operation (Syaufina, 2018). Remote sensing monitoring technology has been developed (Unik et al, 2023). However, some sensing results have not been understood by field officers and the community due to limited counselling from forest and land fire experts.…”

Calibration of CAMS PM2.5 data over Hungary: a machine learning approach

“…Relative humidity can also impact PM 2.5 concentrations, as higher humidity can lead to the formation of secondary aerosols, which can contribute to PM 2.5 concentrations. Boundary layer height and wind speed are also important factors, as they can affect the dispersion and transport of PM 2.5 in the atmosphere [31].…”

An Estimation of Daily PM2.5 Concentration in Thailand Using Satellite Data at 1-Kilometer Resolution

“…AOD measures the aerosol in the atmosphere and can serve as a proxy for surface PM2.5 [11]. Additionally, other factor variables, including meteorological factors, land use and cover, and time variables, are often included to improve the accuracy of the modeling [12]. These variables can explain seasonal variations and long-term trends in PM2.5 levels and indicate potential PM2.5 sources and areas of concern [13].…”