Tracing the haze: satellite-based assessment of stubble burning and air quality in Delhi
摘要
New Delhi, the capital city of India, routinely records hazardous fine-particle concentrations during the post-monsoon season, yet the quantitative link between regional crop-residue burning and episodic haze remains contested. This study integrates multi-sensor satellite products with atmospheric trajectory modelling to attribute the late-October–early-November aerosol enhancement over the capital during 2020–2024. Columnar aerosol optical depth (AOD) at 550 nm was extracted from MODIS Terra–Aqua (10 km); active-fire detections were taken from VIIRS S-NPP (375 m); harvest dynamics were approximated from MODIS NDVI (250 m); and 120 h forward air-mass trajectories at 500–1000 m a.g.l. were generated with NOAA-HYSPLIT driven by GDAS 1° fields. Seasonal-trend decomposition and Theil–Sen statistics revealed a consistent AOD surge of 0.35 ± 0.06 above pre-monsoon levels (p < 0.05). Punjab contributed 90% of regional fire counts in 2020 but only 40% in 2024, whereas Haryana showed a marginal decline. Daily fire counts within Punjab–Haryana explained 78% of Delhi AOD variance during October–November (r = 0.78, p ≪ 0.01). NDVI differencing confirmed harvest-related vegetation loss across > 90% of cropland pixels in week 43 each year. Cluster analysis indicated that 60% of trajectories originating overactive-burn zones intersected Delhi within 36 h, increasing the probability of AOD > 1.2 by a factor of five. These convergent lines of evidence identify stubble combustion as the primary driver of Delhi’s recurring autumn haze. Accelerated deployment of in-situ straw incorporation, baler-mulcher systems, and regional burning-ban enforcement, supported by real-time satellite surveillance, is recommended to achieve National Clean Air Programme particulate-matter targets and to safeguard regional public health. Economic co-benefits are anticipated through fuel savings, improved soil organic carbon, and rural air-quality gains across the Indo-Gangetic Plain.