Delhi Air Health archive
Watch Delhi's air change hour by hour.
The site now queries its own hourly archive first. Explore the Delhi monitoring network, jump to any campus-linked monitor, or inspect an individual station without waiting for a complete 24-hour AQI window.
Stored rows
2,495
Monitor feeds
69
Owned archive · live explorer
Delhi monitoring-network median
Median of available Delhi monitor snapshots in each archived hour. This is a network summary, not an official city AQI.
PM2.5 average
65.8 µg/m³
Across 19 archived hours with a value
Peak PM2.5
73 µg/m³
8 Sept, 1:30 am
Lowest PM2.5
57 µg/m³
7 Sept, 3:30 pm
Archive coverage
19/24 h
PM2.5 19 h · PM10 19 h
PM2.5
Fine-particle archive
Latest 63 µg/m³
PM10
Coarse + fine particle archive
Avg 99.1 µg/m³ · peak 106 µg/m³
Archive began
6 Sept, 4:00 pm
Older hours are not back-filled or invented.
Newest archived observation
8 Sept, 1:00 pm
The collector runs hourly; source timestamps can lag the capture itself.
Network context
44 Delhi feeds in window
Average 43.3 monitors contributed to an available hourly median.
Method note: these are successive archived CPCB OGD avg_value snapshots grouped by source observation hour. Delhi Air Health does not assume that an archived snapshot is a validated one-hour mean, does not interpolate missing hours, and does not label the network median as an official CPCB AQI.
Long-term context · 2016–2025
Is Delhi actually getting cleaner?
The answer is more useful than “yes” or “no.” Delhi's long-run particulate concentrations are lower than they were in 2016–2018, but the improvement is uneven, strongly seasonal, and still leaves annual particulate levels far above India's ambient standards.
PM2.5 · 2016 → 2025
135 → 97
µg/m³ · -28% change
PM10 · 2016 → 2025
290 → 198
µg/m³ · -32% change
2025 vs Indian annual standard
2.4×
PM2.5 · 97 vs 40 µg/m³
Lower-AQI days · 2018 → 2024
+50
additional days with AQI ≤ 200
Delhi Air Health verdict
There is evidence of long-term improvement, but not of a solved problem. 2025 recorded the lowest annual PM2.5 and PM10 of the 2018–2025 period apart from the COVID-affected 2020, yet PM2.5 remained roughly 2.4 times and PM10 roughly 3.3 times their Indian annual standards.
Particulate trend · CPCB data reproduced by CAQM
Annual PM2.5 and PM10, 2016–2025
Dashed reference lines show India's annual NAAQS: PM2.5 40 µg/m³ and PM10 60 µg/m³.
2020 is highlighted because lockdown-related reductions in activity make it a poor structural baseline. 2026 is intentionally excluded until the year is complete.
What the line does show
The baseline has moved down.
PM2.5 fell from 135 µg/m³ in 2016 to 97 in 2025; PM10 fell from 290 to 198. That is a meaningful long-run change even though individual years move up and down.
What the line does not prove
It cannot isolate policy from weather.
Rainfall, wind, mixing height, regional transport, fire episodes and changes in the monitoring network can all alter annual concentrations. The trend is an outcome measure—not a causal attribution model.
Annual average AQI
Improvement is not monotonic.
CAQM's comparable 2018–2024 series shows annual average AQI falling from 225 in 2018 to 209 in 2024—but 2024 was not better than every recent year on this metric.
Days with AQI ≤ 200
The distribution of days can improve even when the annual average barely moves.
Delhi had 159 Good-to-Moderate days in 2018 and 209 in 2024. The COVID-affected 2020 is shown separately rather than treated as a normal benchmark.
Seasonality
Delhi does not have one air-pollution season.
A simple mean of CAQM's published monthly average AQIs from 2018–2024 makes the seasonal pattern obvious: monsoon months are much cleaner on average; October brings a sharp transition; November through January are the hardest months.
This monthly curve is a Delhi Air Health derivation from CAQM's monthly table, not an official new metric. It is meant to explain seasonality, not predict a specific day.
How to read the claim
“Delhi is improving” and “Delhi's air is still unhealthy” can both be true.
Long-run direction
Annual PM2.5 and PM10 are lower than a decade ago. That is evidence of improvement in the outcome we care about.
Remaining burden
2025 particulate levels still substantially exceeded India's annual standards. A falling line does not mean a low-risk endpoint has been reached.
Causal uncertainty
This page does not assign the improvement to one policy. Weather, regional emissions and structural changes all interact.
Primary evidence used on this page
Auditable, finite, and deliberately boring.
Next research layer
Weather-normalised trends
A future version should pair pollution with wind, rainfall, temperature, boundary-layer conditions and fire activity. That would let us ask a harder question: did emissions fall, or did the atmosphere simply disperse them more effectively?
Read the methodology →