Characterizing the Seasonal Particulate Matter (PM 2.5 , PM 10 ) Concentrations in Nine Indian Cities, Using Exploratory Seasonal Analysis (ESA), Supervised Statistical Methods and Contextual Appraisal
Elevated ambient particulate matter levels have become a critical urban sustainability concern in India, endangering health and well‐being (UN SDG 3) of a vast section of the population. We present an exploratory seasonal analysis (ESA) of the ambient PM 2.5 and PM 10 concentrations, for nine Indian cities, using daily archival records from the Central Control Room for Air Quality Management (CAQM) for 2023. The ESA results revealed distinct seasonality in both PMs, varying as post‐monsoon (most polluted) > winter > summer > monsoon (cleanest). The National Capital Region (NCR: Delhi, Gurugram, Noida) appeared the most vulnerable to PM pollution, with Delhi as the most polluted among the nine cities, while Varanasi, the cleanest. Jaipur and Ahmedabad revealed concerns over windblown dust in summer, elevating the PM 10 risks. Seasonal PM 2.5 /PM 10 ratios indicated that besides fuel combustion in general, there is need for more location‐specific air quality control mechanisms, addressing local sources. For each city, about 9%–10% of the days appeared as “outlier”, representing sudden high spikes in PM levels. Both PMs exceeded the WHO health safety thresholds for about 90%–95% of days, even during monsoon, while for 99–100% of the days in winter and post‐monsoon times indicating year‐round public health concerns. Hierarchical Clustering (HCA) offered a spatio‐temporal (city‐season) PM 2.5 ‐vulnerability ranking scheme comprising of four statistically distinguishable clusters: Cluster 1 (148.67 ± 17.94 µg.m −3 ) > Cluster 3 (79.469 ± 1.51 µg.m −3 ) > Cluster 4 (59.90 ± 6.31 µg.m −3 ) > Cluster 2 (32.64 ± 7.24 µg.m −3 ). Cluster 1 represented the NCR cities for post‐monsoon and winter, while cluster 2 included all nine cities for monsoon and summer. Overall, HCA outputs could help setting up a joint action plan, bringing together various city municipal bodies to address the PM pollution. Applying Multiple Linear Regression (MLR) to the HCA clusters individually, revealed significant ( p < 0.05) statistical associations between meteorological parameters and PM2.5 concentrations in HCA clusters 1 (R 2 : 0.76) and 3 (R 2 : 0.67), while significantly lower for clusters 4 (R 2 : 0.52), and negligible for Cluster 2 (R 2 : 0.39). In Cluster 1 and 3, relative humidity and rainfall (monsoon season) exhibited strong inverse statistical relationships with PM 2.5 levels, consistent with potential scavenging and wet deposition processes to reduce the PM 2.5 risks, which was aggravated during post‐monsoon and winter. The latter occurred due to atmospheric stagnation (low wind speed, low solar radiance, cooler temperatures) that trapped the PM 2.5 near the ground. Meteorological influences, however, were minimal for HCA cluster 2 (R 2 : 0.39), indicating importances of other factors (e.g., built‐up types and geometry; emission sources types, vegetation type and density) that need to be researched further. In the final sections, we reflect on a future vision about reimagining the urban transport sector, aligning policies with UN SDGs 7 (clean energy), 9 (infrastructure, and innovation) 16 (institution building) and 17 (forging multi‐sectoral partnerships) to address the particulate pollution crises.