Examining The Impact of The Transportation, Manufacturing, and Energy Sectors on Air Quality In Jakarta Using Spatial Regression

Abstract This study employs Geographically Weighted Regression (GWR) to analyze the spatial distribution of air pollutants NO 2 , SO 2 , and CO in Jakarta and its surrounding areas, focusing on variations between dry and wet months in 2023. The analysis utilizes pollution data from Sentinel-5P images, processed using Google Earth Engine and ArcGIS/QGIS software. The study area, encompassing Jakarta and a 100 km radius, includes industrial and energy sector data to understand pollution source contributions. The GWR models explored three scenarios with different predictor variables: network density, number of intersections, and industrial proportion. The results reveal significant spatial heterogeneity in pollutant concentrations, with higher emissions during dry months. Scenario 1, which includes all predictor variables, shows the highest LocalR 2 values in highly industrialized zones. Scenario 2, excluding the energy sector variable, demonstrates broader model applicability, while Scenario 3, with only transportation-related variables, offers the widest coverage but reduced specificity. These findings provide critical insights for policymakers to formulate targeted strategies for air quality management, aiming to mitigate the adverse effects of air pollution on the population.