Mapping the Spatiotemporal Variability of Particulate Matter Pollution in Delhi: Insights from Land Use Regression Modelling

This study investigates the spatiotemporal dynamics of pollutant concentrations in Delhi through the utilization of land use regression models. Analysis of data for year 2019 from 38 monitoring stations reveal elevated PM₁₀ and PM_2.5 levels, peaking in winter ([PM₁₀: 306.90 ± 53.76 μg/m³], [PM2.5: 185.52 ± 31.59 μg/m³]) and dropping in monsoon ([PM₁₀: 107.77 ± 31.19 μg/m³], [PM_2.5: 40.86 ± μg/m³]), surpassing national standards ([PM₁₀: 60 μg/m³], [PM_2.5: 40 μg/m³]). Spatial distribution analysis indicates higher concentrations in the north and northwest regions, attributed to dense habitation, industrial zones, and vehicular traffic. Analyzing particulate pollutants data for year alongside urban land use/cover features and socioeconomic variables, the study reveals a robust relationship between particulate concentrations and urban attributes, explaining 37–58% of PM_2.5 and 38–62% of PM₁₀ concentration variations. The models demonstrate good accuracy, with low RMSE values (PM_2.5: 9.55, PM₁₀: 27.49), underscoring the impact of urban landscape and surface conditions on air quality distribution. Understanding this link offers insights for better urban planning strategies that integrate air quality considerations, crucial for effective policy frameworks addressing pollution in urban environments.