Assessing the spatial-temporal relationships between land use land cover changes and PM_2.5 concentrations across 38 Indonesian Provinces (2000-2020)

¹ The University of Osaka, Graduate School of Engineering, Division of Sustainable Energy and Environmental Engineering, 2-1 Yamadaoka, Suita, Osaka, Japan
² Kalimantan Institute of Technology, School of Sustainable Development, Department of Environmental Engineering, Balikpapan 76127, East Kalimantan, Indonesia

^* Corresponding author: chandra.survani@,ea.see.eng.osaka-u.ac.ip

Abstract

PM2.5 poses significant health risks and has been a global concern, yet systematic evidence linking land use and land cover changes (LULCC) to air quality outcomes across different provinces in Indonesia remains absent. This study addresses this gap by examining spatial-temporal correlations between LULCC and PM2.5 concentrations across all 38 provinces of Indonesia in 2000 to 2020, employing satellite-derived PM2.5 data validated against 15 ground-based stations with R²=0.63, Global 30- meter Land Cover (GLC) data and Global Fire Emissions Database Version 5 (GFED5) biomass burning data. Months with major biomass burning episodes were systematically excluded from Pearson correlation analysis. Results revealed substantial national transformations including forest decline from 83.1% to 77.6% and cropland expansion of 9.1% to 13.1%. However, after excluding biomass burning periods, LULCC exhibited predominantly weak correlations with PM2.5 concentrations across most provinces, while biomass burning episodes contributed 2.6-7.4 times higher PM2.5 concentrations. These findings demonstrate that episodic biomass burning rather than gradual land cover changes, dominates PM2.5 pollution in Indonesia, requiring regionally tailored air quality management strategies.