Influence of environmental and climatic factors on AOD concentration: A model-based analysis

Aerosol optical depth (AOD) plays a vital role in air pollution studies by providing key insights into aerosol concentration, their radiative effects, and impacts on atmospheric air quality. This study examines the impact of several meteorological and climatic parameters (predictors) on AOD dynamics in the two most urbanized cities of Bangladesh: Dhaka & Chittagong. For this purpose, ten predictors with 60 months of data (2019–2023) were taken. Multiple Linear Regression (MLR), Generalized Additive Model (GAM), and Gradient Boosting Machine (GBM) were used to evaluate the complex relationships between the predictors and their influence on AOD variability. MLR captured AOD variability well in Chittagong (R 2 = 0.45) but performed poorly in Dhaka (R 2 = 0.19), where VIF analysis highlighted temperature at 2 m (Tem_2) and relative humidity as the dominant predictors. GAM analysis of Dhaka exhibited a robust correlation with AOD and precipitation, having R 2 of 0.86, while most of the parameters were statistically insignificant ( P > 0.05). Inversely, Chittagong showed strong significance with each predictor except for O 3 and Land Surface Temperature (LST). Dewpoint temperature (DPT) emerged as a strong predictor in Dhaka, while Surface Albedo dominated in both cities. GBM analysis revealed Tem_2 as the dominant factor in Chittagong (∼70% contribution), while RH is the main influencer in Dhaka, contributing over 20%. Study models perform differently in capturing AOD variability, reflecting linear versus nonlinear sensitivities, yet all emphasize moisture, surface, and thermal controls. This study provides city-specific insights into AOD variability and predictor controls, informing targeted urban air quality management strategies. • AOD dynamics analyzed in Dhaka and Chittagong from 2019 to 2023. • GAM shows strong AOD link with precipitation in Dhaka (R 2 = 0.86). • Surface albedo is the top AOD predictor in Chittagong (PI = 64.32%). • GBM reveals stronger greenhouse gas interactions with AOD patterns. • Study supports city-specific air pollution strategies for policymakers.