The record · Delhi air
September 2026
9 days of readings from CPCB's Delhi network, 139 hourly observations in all. Every day carries the hours actually observed, so a partial day cannot be read as a complete one.
Day by day
The highest and lowest worst-monitor reading recorded on each day, the monitor that produced the peak, and how many of the city's stations were over 100 at the worst hour.
| Date | Peak AQI | Band | Monitor at the peak | Lowest | Stations over 100 | Hours observed |
|---|---|---|---|---|---|---|
| 1 Sep 2026 | 238 | Poor | Wazirpur, Delhi - DPCC (PM10) | 203 | 33 | 21 |
| 2 Sep 2026 | 195 | Moderately Polluted | Wazirpur, Delhi - DPCC (PM10) | 133 | 21 | 21 |
| 3 Sep 2026 | 204 | Poor | Aya Nagar, Delhi - IITM (PM10) | 129 | 7 | 21 |
| 4 Sep 2026 | 156 | Moderately Polluted | Aya Nagar, Delhi - IITM (PM10) | 118 | 8 | 21 |
| 5 Sep 2026 | 122 | Moderately Polluted | Jawaharlal Nehru Stadium, Delhi - DPCC (CO) | 96 | 3 | 5 |
| 6 Sep 2026 | 161 | Moderately Polluted | NSIT Dwarka, Delhi - CPCB (OZONE) | 90 | 2 | 16 |
| 7 Sep 2026 | 178 | Moderately Polluted | Dwarka-Sector 8, Delhi - DPCC (PM10) | 118 | 22 | 16 |
| 8 Sep 2026 | 213 | Poor | Wazirpur, Delhi - DPCC (PM10) | 188 | 28 | 16 |
| 9 Sep 2026 | 218 | Poor | Bawana, Delhi - DPCC (PM2.5) | 216 | 36 | 2 |
When the source served it again
The feed was re-served for 24 observation times this month. None of them changed the city’s headline reading.
Each observation time is read when it appears and re-read on later passes. Where a later pass returns something different for a time already stored, the earlier snapshot is kept beside the new one rather than overwritten — so a figure cited from this record can still be found at the address it was cited from. What differs is usually further down than the headline: a station reporting late, or one dropping out of the set.
How to read this table
Four things the columns are, and are not.
The reading is the worst reporting monitor, named — not a city average. CPCB publishes a population-weighted city figure as well, and the two answer different questions. This record keeps the worst because it is what somebody in that part of the city was breathing.
An AQI is already a rolling figure. CPCB's sub-indices are computed over averaging periods set per pollutant — 24 hours for the particulates. So an hourly observation describes a window ending at that hour, not that minute, and the day's peak is the highest such window.
Hours observed is the honesty column. A day with four hours in it is four hours of record. Nothing here is interpolated and no missing hour is filled.
100 is the top of CPCB's ‘Satisfactory’ band, which corresponds to the 24-hour standard for the governing pollutant. It is the line this table marks, and what the index actually measures explains the rest.
Cite this month
Reuse freely. The grant is ours to give; CPCB's own terms travel with the underlying readings.
Swechha (2026). Delhi air quality record, September 2026.
Readings from the Central Pollution Control Board network.
https://swechha.in/record/air/2026/09
Licensed CC BY 4.0. Full terms, method and limitations.
Next
The live reading, the rest of the record, and what the numbers mean.
