Ahead of winter, system to track emissions reactivated
As Delhi braces for its annual onslaught of winter pollution, the city’s only operational pollution-tracking model -- the Decision Support System (DSS) -- has been reactivated to estimate the daily contribution of various pollution sources to Delhi’s air. But officials admit that the system, run by the Indian Institute of Tropical Meteorology (IITM) in Pune, still operates on an outdated 2021 emissions inventory, raising questions over the accuracy of its forecasts.
DSS is currently the only system providing estimates of how much Delhi’s air pollution comes from local sources such as transport, dust and industries, and from external regions such as crop-burning in neighbouring states. This follows the Delhi government’s decision last winter to halt its own real-time source apportionment study, leaving a crucial data gap in the city’s anti-pollution planning.
According to DSS first readings on Sunday, stubble burning has not yet contributed to Delhi’s PM2.5 levels this season, even though farm fires have begun in the northern plains. The model shows transport emissions as the single largest source, accounting for 18% of the city’s particulate matter load on Sunday. The next biggest contributor was pollution from Gautam Buddha Nagar (14%), followed by Bulandshahr (9%).
“DSS is operational again and the estimated contribution will now be updated daily. However, as mentioned on the website, results will not be entirely accurate as the system is still running on the old emissions inventory,” said an IITM official aware of the development. Even the DSS portal carries a disclaimer noting that “for more accurate estimations, the latest emission fields are required.”
Officials said a new emissions inventory is under preparation and will be incorporated as soon as it is ready. “This will make the forecasts more reliable and region-specific,” the official added.
The system’s limitations have long been flagged. In November 2023, HT had reported that Delhi’s ability to identify real-time pollution sources had weakened because of non-functional or outdated models. Since 2020, at least three major studies meant to guide real-time interventions have either been discontinued or rendered obsolete.
The Delhi government’s 2021 real-time study, initially awarded to Washington University, was scrapped before its findings were made public, citing “unsatisfactory results.” A fresh contract was given to IIT Kanpur, which ran the study until late 2023. When its two-year term ended, the system was discontinued, and no replacement has since been finalised.
To be sure, DSS itself came under the scanner of the Commission for Air Quality Management (CAQM) last year. On December 3, the commission suspended its operations, citing inaccuracies, and asked IITM to improve its model. Operations resumed six days later, but CAQM clarified that DSS would not be used for pollution-related policy decisions until its accuracy improved.
Despite its flaws, researchers stressed that DSS the only available tool – rendering it important. A recent Council on Energy, Environment and Water (CEEW) study found that forecasts by DSS and IITM’s Air Quality Early Warning System (EWS) correctly predicted “very poor and above” air quality on over 80% of days between October 2023 and February 2024. The study, however, concluded that updated emission inventories were crucial to enhance the model’s precision. “An updated inventory will help us better understand what pollutes Delhi’s air — and in what quantities,” the study noted.
Sunil Dahiya, founder of the think-tank Envirocatalysts, said decisions could still be guided by DSS data but stressed the urgency of modernising it. “The current model is functional and can still inform action on the ground. But without updated emissions data, some inaccuracies will persist,” he said.
