Even as government data suggests a sharp drop in stubble burning in Haryana — from 1,406 incidents last year to 662 this year — data derived from satellite shows that the burnt area in the state area has actually risen from 7,117 sq km in 2024 to 8,812 sq km in 2025, a new report has revealed. In Punjab, the area burnt has not reduced too significantly from 23,262 sq km to 19,757 sq km even as a huge decline was reported in the cases of farm fires — from 10,909 last year to 5,114 this year.
The International Forum for Environment, Sustainability and Technology (iFOREST), which released a report on Monday, has said this mismatch persists because more than 90% of large fires in 2024 and 2025 occurred after 3 pm, outside satellite overpass times.
This aligns with the findings of an ISRO study reported by The Indian Express on December 5, observed up to 2024, underscoring that India’s monitoring system is missing most fires and underestimating pollution contributions to Delhi.
The iFOREST, an independent nonprofit, said the current satellite-based monitoring system is structurally incapable of capturing most late-afternoon fires. MODIS and VIIRS instruments mounted on NASA satellites — on which the government’s monitoring system relies on — pass over North India only between 10:30 am and 1:30 pm.
The Indian Agriculture Research Institute (IARI) keeps a track of active fire count data.
With fires now increasingly lit after 3 pm, the system is detecting only a fraction of actual burning. In Punjab alone, over 90% of large fires in both 2024 and 2025 occurred after 3 pm, compared to just 3% in 2021. In Haryana, most large fires have been happening after 3 pm since 2019, meaning undercounting has been persisting for several years. This late-afternoon shift, the authors said, explains why burnt-area mapping shows far higher levels of burning than active fire counts.
The findings have major implications for Delhi’s pollution forecasting. The Decision Support System (DSS), run by the Indian Institute of Tropical Meteorology (IITM), uses the same fire-count data to estimate how much stubble burning contributes to Delhi’s PM2.5 levels. With active fire counts missing most late-afternoon fires, this year’s stubble-burning contribution to Delhi’s air found at around 22% may have been severely underestimated.
“These monitoring gaps have far-reaching implications,” the authors said. “Fires missed by polar-satellite sensors lead to underestimated emissions, mischaracterised aerosol and particulate-matter loads, and incorrect simulations of air-quality dynamics across India.”
Story continues below this ad
Top government sources told The Indian Express that they are aware of the limitations, which is why a ground-truthing exercise was carried out on an experimental basis during the 2025 Kharif season.
Burnt-area mapping, which uses Sentinel-2 MSI data at 10-20 m resolution, is considered a more reliable indicator of actual stubble burning. Unlike fire counts, which record only visible flames during satellite flyovers, burnt-area analysis captures the total land that has burned including small fires, brief fires, and fires under cloud cover. “Smoke depends on how much land burned, not how many fires were seen,” the authors explained. “Burnt area gives a better picture of impact.” The reductions reflected by burnt-area mapping roughly 25-35% are far smaller than those suggested by fire-count data, which show declines of over 90-95% in both states.
According to officials, while there are resolution-related limitations in pinpointing exact locations, the total burnt-area estimates are accurate. “This should still help in finding the right algorithm and provide a check on the claims,” an official said.
The study was authored by Ishaan Kochhar (Programme Lead), Manish Bilore (Programme Associate) and Meet Makwana (Research Associate), who are trained in remote-sensing technologies, with alumni backgrounds from IIT Bombay and other institutions.
Story continues below this ad
Chandra Bhushan, CEO of iFOREST, said, “Our analysis provides incontrovertible evidence that India’s current stubble-burning monitoring system is structurally misaligned with ground realities. Farmers have shifted burning to the late afternoon, while our monitoring relies on satellites that capture active fires only during a narrow time window 10:30 am to 1:30 pm. The result is a massive underestimation of fires, emissions, and their contribution to air pollution in Delhi. We urgently need to overhaul the system.”
The think tank’s multi-sensor approach covered MODIS (Terra at 10:30 am; Aqua at 1:30 pm), VIIRS (Suomi-NPP at 1:30 pm and 1:30 am), Sentinel-2 burnt-area analysis, and SEVIRI geostationary data, which captures imagery every 15 minutes between ~5:30 am and 7:30 pm. Only SEVIRI was able to detect the majority of late-afternoon fires revealing the timing shift and the magnitude of undercounting.
The report has also warned that stubble burning is increasing in other states, particularly parts of Uttar Pradesh and Madhya Pradesh, based on rising active fire counts. However, burnt-area estimated for these states has not yet been analysed, something the authors say must be prioritised.
Ishaan Kochhar, Programme Lead at iFOREST, said, “We cannot manage what we do not measure accurately. Policy decisions are currently being shaped by incomplete information. To solve the stubble-burning problem in the Indo-Gangetic Plain, the government must urgently reform the monitoring protocol to integrate burnt-area mapping and geostationary data. We also need to expand our focus beyond Punjab and Haryana to emerging hotspots in Uttar Pradesh and Madhya Pradesh.”
