Seasonal distribution of AOT and its relationship with air pollutants in central Bangladesh using remote sensing and machine learning tools

Abstract

This study aims to map the seasonal distribution of AOT from 2002 to 2022 and to explore the internal relationship between AOT and ten air pollutants using remote sensing and machine learning tools. The results show that the concentrations of AOT were higher in December-January-February and March-April-May seasons. AOT was a bit less in June-July-August and September-October-November. This study also revealed that the AOT was correlated positively with PM2.5, CH4, NO, and BC while correlated negatively with CO, HCHO, SO2, APR, and NOx. Machine learning results were found be good classifiers to predict AOT.

Keywords

• AOT
• Bangladesh
• Air pollution
• Machine learning
• Remote sensing