Evaluation of an integrated WRF-Chem framework for air quality forecasting and source attribution in Jaipur, India

This study presents the design and evaluation of a high-resolution Air Quality Early Warning and Decision Support System (AQEWS–DSS) for Jaipur, India, developed to provide short-term (1–5 day) forecasts and real-time source attribution of particulate matter (PM₂.₅). The system employs the WRF-Chem model configured at 2 km resolution with daily data assimilation of satellite aerosol optical depth (AOD) and surface observations from RSPCB (Rajasthan State Pollution Control Board) stations. Model evaluation during October 2024–January 2025 demonstrates robust forecast skill, with Root Mean Square Error (RMSE) values of 26.2 µg/m³ for PM₂.₅ and 48.8 µg/m³ for PM₁₀, Index of Agreement (IOA) of 0.78 and 0.75, and Performance Index (PI) scores of 85.1 and 82.5, for PM_2.5 and PM₁₀ respectively. Over 90% of model predictions fall within a factor of two of observations, and diurnal patterns are also well captured. Although peak events such as Diwali are slightly underestimated (by ~ 35%), the model effectively reproduces seasonal transitions and day-to-day variability. DSS-based source apportionment reveals that Jaipur city contributes 27–28% of its PM₂.₅ burden, with the remainder dominated by external sources. In October, dust transport from the Thar Desert accounts for ~ 28% of PM₂.₅, while in November, biomass burning and regional transport become more influential. These results emphasize that Jaipur’s air quality is shaped by both persistent local emissions and seasonally varying regional sources. The findings support the need for integrated regional control strategies and operational use of high-resolution DSS tools to guide air quality management across northern India.