Applicability of machine learning in modeling of atmospheric particle pollution in Bangladesh

Shahriar, Shihab Ahmad; Kayes, Imrul; Hasan, Kamrul; Salam, Mohammed Abdus; Chowdhury, Shawan

Atmospheric particle pollution causes acute and chronic health effects. Predicting the concentrations of PM(2.5) and PM(10), therefore, is a prerequisite to avoid the consequences and mitigate the complications. This research utilized the machine learning (ML) models such as linear-support vector machine (L-SVM), medium Gaussian-support vector machine (M-SVM), Gaussian process regression (GPR), artificial neural network (ANN), random forest regression (RFR), and a time series model namely PROPHET. Atmospheric NO(X), SO(2), CO, and O(3), along with meteorological variables from Dhaka, Chattogram, Rajshahi, and Sylhet for the period of 2013 to 2019, were utilized as exploratory variables. Results showed that the overall performance of GPR performed better particularly for Dhaka in predicting the concentration of both PM(2.5) and PM(10) while ANN performed best in case of Chattogram and Sylhet for predicting PM(2.5). However, in terms of predicting PM(10), M-SVM and RFR were selected respectively. Therefore, this study recommends utilizing “ensemble learning” models by combining several best models to advance application of ML in predicting pollutants’ concentration in Bangladesh. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s11869-020-00878-8) contains supplementary material, which is available to authorized users.