Emission Inventories: How We Count Air Pollution
Before we can say who is polluting our air, we have to count every source. Here is the simple equation behind an emission inventory, and why Delhi’s inventories do not agree.
By Saurabh Dhakad, PhD scholar, Atmospheric Chemistry and Climate Change Lab, School of Environmental Sciences, Jawaharlal Nehru University · · 3 min read
On 6 November, Delhi’s average AQI touched 382 and the third stage of GRAP came into force across the NCR. Every winter the same question follows: who is making this smog? Vehicles, dust, factories or smoke from the fields? Before anyone can answer, someone has to count. That counting is called an emission inventory. This post explains how one is built, and why inventories of the same city disagree.
What is an emission inventory?
An emission inventory is a record of how much of each pollutant each source releases into the air, for a given area and a given period, usually a year. It is the starting point for models, forecasts and clean air plans.
We almost never measure every chimney and exhaust pipe. A city like Delhi has millions of them. So we estimate.
The one equation behind it
The US EPA writes the basic method as a short equation:
E = A × EF × (1 − ER/100)
E is the emission. A is the activity, such as tonnes of coal burnt, kilometres driven or bricks fired. EF is the emission factor, a typical value that links the pollutant released to that activity, for example grams of PM per kilometre. ER is the efficiency of any control device, in per cent.
Take round numbers just to see how it works. Say a boiler burns 100 tonnes of coal a day, the factor is 10 kg of particles per tonne, and a filter removes 90% of them. Then the boiler emits 100 × 10 × 0.1 = 100 kg of particles a day. Now repeat this for every type of vehicle, factory, kitchen and kiln, and add it all up.
From a list to a map
A single total for the city is not enough for a model. The emissions are spread onto a grid using maps of roads, industries and where people live. The global EDGAR inventory, for example, gives air pollutants such as SO2, NOx, NH3, black carbon and PM2.5 on a 0.1° × 0.1° grid, roughly 10 km wide, split into 26 human-made sectors. City inventories go finer, often to 1 km cells. A chemical transport model then mixes these emissions with weather data to predict the air we breathe.
Why studies disagree
Delhi has several inventories, and their numbers do not match. Transport’s share of the PM2.5 inventory was about 20% in the IIT Kanpur study (2015), 39% in TERI-ARAI (2018) and 41% in SAFAR-IITM (2018). For PM10, road dust ranged from about 36% to 66% across five studies. The reasons are ordinary:
- different study areas, with or without the towns around Delhi;
- different lists of sources and different survey seasons;
- different estimates of activity, such as how many old trucks are still on the road;
- different emission factors, many of them borrowed from other countries.
The uncertainty can be large. One Asian inventory put the uncertainty in household PM2.5 emissions at up to 259%. This is why inventories are checked against measurements: receptor models that read the chemical fingerprint of particles on filters, and satellite data. Faster daily tools now exist too. A CSE analysis of the Decision Support System run by IITM Pune found that vehicles gave about 51–53% of the PM2.5 from Delhi’s own local sources during 10–20 November in 2023, 2024 and 2025.
What this means for us
An inventory is a bit like a household budget for the air. If the budget is wrong, money and effort go to the wrong place. Emission factors measured on Indian vehicles, stoves and kilns, regular updates instead of one-time studies, and open data that others can check would all make our inventories stronger. And when we read that one source causes a certain per cent of our pollution, it helps to ask which pollutant, which area, which season and which method.
References and further readings
- US EPA. Basic information of air emissions factors and quantification (AP-42).
- Crippa, M. et al. (2018). Gridded emissions of air pollutants for the period 1970–2012 within EDGAR v4.3.2. Earth System Science Data, 10, 1987–2013.
- EMEP/EEA air pollutant emission inventory guidebook. European Environment Agency.
- CEEW. What’s polluting Delhi’s air? Understanding uncertainties in emissions inventories.
- Shukla, S. (22 Dec 2025). From episodic smog to structural pollution: why vehicles remain Delhi’s biggest challenge. Down To Earth.
- Business Today (10 Nov 2026). North India gasps: Delhi air improves slightly; NCR, UP remain in toxic grip.