Application of measurements and modeling tools for managing air quality at a hyperlocal scale in Indian cities

In Low- and Middle-income countries like India, alternate techniques like low-cost sensors and modeling play a pivotal role in air pollution management due to the limitations posed by expensive regulatory monitors, which often restrict comprehensive data collection. These cost-effective solutions facilitate wider deployment, enabling a denser network of sensors to gather real-time data at various locations. Additionally, modeling techniques allow simulation and prediction of pollution levels to achieve better spatio-temporal variations, thus significantly enhancing our understanding of air quality and aiding in formulating targeted mitigation strategies for more effective pollution management. The study employed an integrated approach using stationary sensors, emissions inventory, and dispersion modeling techniques. Five low-cost sensors were positioned within a 3km x 3km area at Sion in Mumbai, with a regulatory monitor as the center of the study area. Simultaneously, a detailed emissions inventory was conducted, utilizing primary data collection to accurately estimate emissions at a fine resolution of 300m x 300m. AERMOD served as the modeling tool, enabling the identification of source-wise and month-wise dispersion patterns. Integrating sensor data and modeling outputs were used to understand the spatial distribution within the 3km x 3km zone. Subsequently, validation of the model was performed by comparison with the regulatory monitor and stationary sensor data. Further, the model was used to estimate the potential percentage reduction in pollution levels from proposed interventions. This comprehensive methodology ensured a holistic approach to air quality management, aiding in both the assessment and prediction of pollution hotspots and the efficacy of potential mitigation measures. The findings revealed that, within the study area of 3km x 3km, the highest contributions of PM2.5 were observed from dust resuspension (53.26%), followed by vehicular exhaust emissions (32.24%), Construction (9.87%), residential (2.31%) and others. Similarly, NOx emissions were majorly from vehicular exhaust emissions (94.72%), followed by DG sets (3.09%). Thus, the mitigation options formulated were focused on the reduction of road dust emissions, tailpipe emissions and construction activities. The model predicted results have shown a percentage reduction in PM2.5 ranging between 5% to 30%. This structured framework can assist air quality managers in identifying the most fitting and suitable policy interventions for addressing air pollution hotspots.