A key concern related to particulate air pollution is the development of an early warning system that can predict local PM_2.5 levels and excessive PM_2.5 concentration episodes using vertical meteorological factors. Machine learning (ML) algorithms, particularly those with recognition tasks, show great potential for this purpose. The objective of this study was to compare the performance of multiple linear regression (MLR) and multilayer perceptron (MLP) in predicting PM_2.5 levels. The software was trained to predict PM_2.5 levels up to 7 days in advance using data from long-term measurements of vertical meteorological factors taken at five heights above ground level (AGL)—10, 30, 50, 75, and 110 m—and PM_2.5 concentrations measured 30 m AGL. The data used were collected between 2015 and 2020 at the Microclimate and Air Pollutants Monitoring Tower station at Kasetsart University, Bangkok, Thailand. The results showed that the correlation coefficients of PM_2.5 predicted and observed using MLR and MLP were in the range of 0.69–0.86 and 0.64–0.82, respectively, for 1–3 days ahead. Both models showed satisfactory agreement with the measured data, and MLR performed better than MLP at PM_2.5 prediction. In conclusion, this study demonstrates that the proposed approach can be used as a component of an early warning system in cities, contributing to sustainable air quality management in urban areas.

In Bangkok, the megacity of Thailand, concentrations of fine particulate matter (PM_2.5) have often exceeded the National Ambient Air Quality standards. During severe smog events over Bangkok, the air quality has exhibited moderate to unhealthy atmospheric conditions, according to the air quality index of the United States. To investigate the formation of secondary organic aerosols (SOA), a field campaign to estimate secondary organic carbon (SOC) in Bangkok using the EC tracer method was conducted in January 2021, when the concentrations of PM_2.5 were high. The monthly period was classified into three pollution groups, including high pollution, high PM, and low pollution events. The study showed that the correlations between PM_2.5 and O₃ were negative during both the daytime and night-time. The OC/EC ratios varied from 4.32 to 5.43, while the moderate OC/EC values implied that fossil fuel combustion was the major carbonaceous aerosol in Bangkok. The EC tracer-estimated SOC and POC showed that SOC contributed between 32.5 and 46.4% to OC, while the highest SOC contribution occurred during the low pollution event. The heightened formation of SOA during the low pollution event was perhaps owing to the levels of oxides of nitrogen (NO_x). Since Bangkok is more likely to have a NO_x-rich photochemical reaction regime, an increase in the NO_x level tended to decrease the SOA yield ([NO_x] was 21.6 ppb, 20.8 ppb, and 17.1 ppb during the high pollution, high PM, and low pollution events, respectively). Together with the high humidity and high light intensity during the low pollution event, the SOA formation was enhanced. Even though the driving factors of SOA formation over Bangkok remain unclear, the results of this study reveal the significance and urgency of local actions to reduce NO_x and O₃ towards more habitable and sustainable urban environments.

This study investigates the vertical variation and temporal characteristics and indicates the sources of black carbon (BC) and brown carbon (BrC) within particulate matter fraction PM₁ during light (November–December 2024) and heavy (January–February 2025) haze episodes in Bangkok, Thailand, a topic where data are still limited data regarding Southeast Asian megacities. Continuous measurements were conducted at 30 and 110 m above ground level, together with particle size distribution measurement, micrometeorological observations, and backward air mass trajectory analysis. During the haze periods, the highest particle number concentrations occurred in the 0.3–0.4 µm size range, indicating dominant contributions from combustion-related emissions and secondary aerosol formation. Mean PM₁ mass concentrations during the heavy haze episodes were more than 2.5 times higher than those during light haze. BC concentrations increased substantially during heavy haze, while the BC fraction of PM₁ remained relatively constant (~10%). In contrast, the BrC fraction reached nearly 20%, reflecting an increasing influence of biomass burning emissions associated with regional transport. Combined analyses of BC/BrC relationships, wind-direction dependence, and air mass trajectories demonstrate mixed contributions from local fossil fuel combustion and long-range transport of biomass burning aerosols during severe haze events.