Air
What is source apportionment?
Source apportionment is the exercise of deciding what share of measured pollution came from which source. It is modelled, never measured — no instrument can tell you that a particle came from a truck. Two government-commissioned studies of Delhi put the transport share at 6–29 per cent and 17–28 per cent respectively, and that gap is the honest state of the knowledge rather than a fault in either one.
Two ways to answer the same question
Apportionment runs in one of two directions, and they can produce different answers from the same city.
Receptor modelling starts at the monitor. It takes the chemical fingerprint of collected particles — the elements, ions and carbon fractions present — and works backwards to the mixture of sources that could have produced it. It is grounded in a real sample. Its weakness is that sources with similar fingerprints are hard to separate, and it cannot see a source that was not represented in the profiles used.
Dispersion modelling starts at the source. It takes an emissions inventory — how much each sector releases — and simulates transport and chemistry forward to the monitor. It can allocate secondary particulates back to the gases that formed them, which receptor modelling struggles with. Its weakness is that it is only as good as the inventory, and inventories in India carry large uncertainties for exactly the informal sources that matter.
Delhi has been studied both ways. IIT Kanpur's 2016 study used receptor modelling. TERI and ARAI's 2018 study did both and published the dispersion result, because that is the one that hands secondary particles back to the sectors responsible for them.
- Receptor model
- Works backwards from what was collected at the monitor to a plausible mix of sources.
- Dispersion model
- Works forwards from an emissions inventory through weather and chemistry to a predicted concentration.
- Emissions inventory
- An estimate of how much each sector emits. Not a measurement, and the largest single uncertainty in any dispersion result.
- Secondary particulate
- Formed in the air from gases. Its source is the gas emitter, which is often nowhere near the monitor.
What each one actually found
Both were commissioned by government, both are public, and they do not agree.
TERI-ARAI, transport share of PM2.5
17 – 28 %
TERI-ARAI 2018, dispersion model, base year 2016
IIT Kanpur, vehicle share of PM2.5
6 – 29 %
IIT Kanpur 2016, receptor model, across sites and seasons
Studies compared on this site
2
Both government-commissioned, both published in full
There is no published limit here to break, and that is the point of the band. Apportionment produces shares, not breaches. A share cannot be compared with a standard, and a study that reports a range should never be quoted as a point.
How to read an apportionment number
Four questions, and a claim that cannot answer them is not usable.
What method? Receptor and dispersion results are not interchangeable and should not be averaged.
What base year? An inventory from 2016 describes a fleet and an industrial base that has since changed. It is a snapshot, not a standing fact.
What season, and where? Delhi's mix changes completely between a November inversion and a July monsoon, and between a roadside site and a background one. A single annual percentage conceals both.
Is it PM2.5 or PM10? The coarse fraction is dominated by dust and the fine fraction by combustion, so the same source has two very different shares depending on which is being apportioned.
What this number cannot tell you
- It cannot be measured. Every apportionment figure is a model output, however carefully it is grounded in samples.
- It cannot settle blame. A sector's share of concentration is not the same as its share of the harm, or of the cost of abating it.
- It cannot be updated by re-reading the monitors. A new answer needs a new study.
- It cannot be averaged across studies. Two ranges produced by different methods do not combine into a better range.
Where this comes from
Every figure above is read out of one of these, at the address printed beside it.
- Source Apportionment of PM2.5 & PM10 of Delhi NCR TERI and ARAI, for the Department of Heavy Industry, Government of India · Published August 2018, monitoring April 2016 – February 2017
- PRANA — National Clean Air Programme portal Central Pollution Control Board · Where the commissioned city studies are hosted
Reuse freely — CC BY 4.0. The grant covers this page. Each source above keeps its own terms, which is why every one of them is named.
Next
The reading this explains, the explanations next to it, and the one place to stand in it.