Aerosol pollution and economic growth in Thailand with a machine learning analysis of the environmental Kuznets Curve

Abstract Aerosol pollution is a chronic air pollution problem that occurs every year in Thailand. It is mostly caused by human activities and climate conditions, and has severe negative impacts on both human health and the economy. This study investigates the relationship between aerosol pollution and various economic and environmental factors using the Environmental Kuznets Curve (EKC) framework. Additionally, it forecasts aerosol pollution through 2050 by applying four machine learning approaches and comparing their accuracy and predictive performance. The results indicate that the Random Forest (RF) model provided the most accurate estimation across the train, validation, and test periods. The study found an inverted U-shape relationship between aerosol pollution and economic growth, consistent with an EKC pattern. This suggests that as Thailand’s economy expands, aerosol pollution is projected to decline at an average annual rate of − 0.0569% to − 0.0636% from 2000 to 2050, except for the Southern region (R4), which shows an increasing rate. Furthermore, environmental features, specifically regional characteristics (R), Normalized Difference Vegetation Index (NDVI), rainfall, and wind speed, strongly influence aerosol pollution levels. Among economic features, population density was found to positively influence aerosol pollution. The study’s findings indicate that population density and GPP per capita remain primary drivers of aerosol pollution in high-activity regions. Consequently, although not directly analyzed in our models, we propose introducing green finance instruments as a strategic mechanism to decouple economic activity from environmental degradation.