Use the data
Take it and
use it.
Everything Swechha publishes as a reading is open. Quote it, republish it, build on it, argue with it. This page is the licence, the attribution, the method and the part most datasets leave out — what each one does not cover.
The licence
One grant, stated once, and the limit of what it is ours to give.
Everything on this site that is a reading, a table or a record is licensed CC BY 4.0. You may copy it, redistribute it, transform it and use it commercially, provided you give attribution.
The grant covers our compilation, our derived figures and our pages. It does not and cannot cover the upstream sources — CPCB, IMD, the Forest Survey of India, NASA, ECMWF and the others keep their own terms, which is why every figure on this site names the source it came from. If you are republishing an upstream figure at scale, go to that source's terms as well.
Photographs are not covered. They are Swechha's own or licensed to Swechha, and they are not part of this grant.
How to cite it
Three shapes, depending on what you are quoting.
A live reading
Swechha, Delhi air quality. Reading from the Central Pollution Control Board
network, observed [date and hour IST]. https://swechha.in/now/air
A month of the record
Swechha (2026). Delhi air quality record, [month year].
Readings from the Central Pollution Control Board network.
https://swechha.in/record/air/2026/09
An explanation
Swechha. “What is PM2.5?” https://swechha.in/learn/pm25
Quote the kind with the number. If a figure is modelled, say so; if it is counted or measured, say so. A modelled national estimate presented as a measurement is the commonest way an honest number becomes a false claim, and the difference is on the page it came from.
Quote the observation time, not the time you read it. Every reading here carries the source's own stamp for when the air, the water or the imagery was actually observed.
Where each number comes from
Source, cadence and the limit it is judged against — the same register the record itself renders.
| Subject | Source | Cadence | Judged against | Where |
|---|---|---|---|---|
| Delhi air | CPCB CAAQMS, mirrored on data.gov.in | Hourly | PM2.5 24-hour: 60 µg/m³ (NAAQS 2009) | Live · Record |
| The Yamuna | CPCB National Water Quality Monitoring Programme | Annual | DO > 5.0 mg/L, BOD < 3.0 mg/L (PWQC 1986) | Live |
| Heat | ERA5 reanalysis, fourteen stations | Seasonal | IMD heatwave criteria — a definition, not a standard | Live |
| Forest loss | FSI ISFR and Hansen / GFW | Annual | No limit — two definitions that disagree | Live |
| Fire | NASA FIRMS, VIIRS S-NPP | Annual sample | No limit published anywhere | Live |
| Extreme rain | ERA5, with NCRB for the deaths | Seasonal | IMD day categories — severity, not permission | Live |
What it does not cover
Stated per subject, because a dataset used past its edges produces a confident wrong answer.
- Air
- A station reading describes a point, not a neighbourhood, and the index cannot be converted back into a concentration. Indoor air, where most of the day is spent, is not measured at all.
- The Yamuna
- One year of grab samples per station, published as a minimum and a maximum rather than a mean. Metals, pesticides and pharmaceuticals are not among the five parameters.
- Heat
- Days meeting IMD’s temperature criteria at single grid points — related to, but not the same as, an IMD declaration, which needs two or more stations in a subdivision.
- Forest
- Two sources that disagree by definition. Neither measures forest quality, biodiversity or carbon, and neither can attribute a change to a cause.
- Fire
- Detections, not fires and not forest fires. A fixed ten-day window each year is a sample and can miss a peak; it is comparable across years precisely because it does not chase one.
- Rain
- A 24-hour category is blind to intensity within the day, and there is no legal rainfall threshold for anything to be over.
Nothing here is interpolated, and no gap is filled. An hour with no observation is absent; a station that stopped reporting is missing, not clean; a year the survey was not run has no figure. If you need a complete grid, this is the wrong dataset and that is a property of the measurement, not of the archive.
Next
The archive, the explanations, and a person to ask.
Ask about the data
Straight to a person
This goes to Vimlendu Jha, Swechha’s Executive Director.
Or write to vimlendu@swechha.in.