Delhi’s air pollution forecasting system continues to underestimate fine particulate matter (PM2.5) concentrations in the city by as much as 30–35 per cent, or almost a third of the actual number, according to climate policy think tank Council on Energy, Environment and Water (CEEW).
“If the actual PM2.5 level is 100 micrograms per cubic metre, the system reports it as around 65 micrograms per cubic metre. So, it is underpredicting PM2.5 levels,” Mohammad Rafiuddin, programme lead at CEEW, told PTI.
PM2.5 refers to fine particulate matter measuring 2.5 micrometres or smaller in diameter. These tiny particles, comprising dust, soot, smoke, and liquid droplets, can penetrate deep into the lungs and bloodstream, causing respiratory and cardiovascular diseases, and even lung cancer.
Rafiuddin said a major reason for the erroneous readings is the outdated emission inventories used by the system, adding that inventories have not been updated since 2016 at the national level, and not since 2021 for Delhi.
An emissions inventory is a database that records the amount of air pollutants released by each source over a year or other time period, according to the US Environmental Protection Agency. Governments use these inventories to identify major pollutant sources and guide regulatory actions. The effect on air quality of potential regulatory actions can be predicted by applying estimated emissions reductions to emissions inventory data in air quality models.
Also Read
Forecasting gap due to outdated emission inventories
The system in question — Air Quality Early Warning System (AQEWS) — was launched in 2018 after a series of smog episodes and dust storms. It is jointly operated by the Indian Institute of Tropical Meteorology (IITM) and the India Meteorological Department (IMD). The system provides pollution forecasts three to ten days in advance.
Updated inventory will aid decision-making
Rafiuddin highlighted the need for updated inventories to tackle Delhi’s pollution problem.
“There should be a framework in place to update the emission inventory every two to three years. The immediate focus should be on upgrading the EI for the Delhi NCR region to improve Delhi’s forecasts,” Rafiuddin said in a study titled ‘How well can Delhi predict air quality? Insights for India’s decision support system’.
He added that such updates would improve the prediction capability and further augment the system’s ability to support informed air quality decision-making.
AQEWS shows accuracy in forecasting severe air quality
The study also notes that while Delhi’s AQEWS underpredicts PM2.5 levels, it successfully forecasts days with ‘very poor and above’ air quality with over 80 per cent accuracy during the past two winters. The system correctly forecast 83 out of 92 such episodes (Air Quality Index above 300) in winter 2023–24, and 54 out of 58 in 2024–25.
The system has also improved in predicting ‘severe’ pollution days (AQI above 400). While it flagged only one out of 15 such days in 2023–24, this increased to five out of 14 the following winter.

)