Delhi's air pollution data gap: What do missing readings, conflicting AQIs reveal?

Was Delhi's AQI 373 or 1,747 on Diwali? While the government’s official reading remained far lower, Delhi Pollution Control Committee's (DPCC) monitoring data showed hourly PM2.5 concentrations peaking at 4,985 µg/m³ in Jahangirpuri.

The striking difference has raised questions about how the capital's pollution is measured, recorded and reported.

The differences include exceptionally high hourly PM2.5 concentrations, subsequent revisions to some readings and missing observations from several stations during the period of severe pollution around Diwali.

At Jahangirpuri, PM2.5 concentrations reportedly reached 4,985 micrograms per cubic metre (µg/m³) at 6 p.m. and 8 p.m. on Diwali. The figures were subsequently changed in the DPCC data, which showed a peak of 776 µg/m³ at midnight.

The changes have raised questions about how the readings were processed and why such substantial revisions were made. The discrepancy is particularly significant because several other monitoring stations also recorded exceptionally high hourly pollution levels during the same period.

Mundka's PM2.5 concentration rose from 245 µg/m³ at 7 p.m. on 8 November to 3,744 µg/m³ at 2 a.m. the following day. JNU recorded 3,023 µg/m³ at 4 a.m., compared with 118 µg/m³ at 7 p.m. on Sunday. NSUT West Campus recorded 3,008 µg/m³, while Mandir Marg registered 2,908 µg/m³.

Other stations, including Nehru Nagar, Wazirpur, Talkatora Garden, the Commonwealth Games Sports Complex, Patparganj and Ashok Vihar, also recorded hourly PM2.5 concentrations above 2,500 µg/m³.

The discrepancies extend beyond individual station readings. On Monday evening, the DPCC released 125 hourly observations that had been missing from the Central Pollution Control Board's (CPCB) records. All 125 readings were reportedly above 1,000 µg/m³.

Mundka recorded the highest of these readings at 3,744 µg/m³. JNU registered 3,023 µg/m³ at 4 a.m., while NSUT West Campus recorded 3,008 µg/m³ at 5 a.m.. Eleven monitoring stations reportedly recorded PM2.5 concentrations above 2,000 µg/m³.

The missing data was part of a wider disruption in the monitoring network. Thirty-six of Delhi's 59 monitoring stations stopped reporting overnight between Sunday and Monday. The number of stations not reporting rose from one at 7 p.m. on Sunday to 36 by 1 a.m. on Monday.

The Centre for Research on Energy and Clean Air (CREA) counted 227 missing hourly observations across the 59 stations between midnight and 5 a.m.

The DPCC has attributed the interruptions, at least partly, to a technical limitation in the monitoring equipment. Its member Anil Gupta told NDTV that stations stopped providing readings when pollution levels reached 999 µg/m³, describing it as a technical glitch.

The DPCC's 2019 technical specifications reportedly prescribe a measuring range of 0–1,000 µg/m³ for continuous particulate matter analysers. However, this does not fully explain why stations stopped reporting at different pollution levels. For instance, the last recorded reading at RK Puram was 924 µg/m³.

The sequence raises questions about whether the missing observations resulted from equipment limitations, data transmission failures or the validation procedures followed by the two pollution control agencies. It also remains important to establish how the delayed readings were recovered and validated before being published, and whether their absence affected the calculation of Delhi's average AQI.

Why different platforms report different AQI figures

The controversy has also been accompanied by widely varying AQI figures for Delhi. On Diwali, some platforms reported an AQI of 1,747, while the CPCB's figure was 373.

This difference partly reflects the use of separate AQI methodologies rather than necessarily indicating a discrepancy in the underlying pollution concentration.

The CPCB follows India's National Air Quality Index (NAQI), which has an upper limit of 500. Several international platforms use the United States Environmental Protection Agency's methodology, which has no equivalent fixed ceiling.

The systems also use different pollutant breakpoints and averaging periods. India's system generally relies on 24-hour averages for several pollutants, while the US methodology uses different averaging periods, including one-hour and eight-hour measurements for some pollutants. Consequently, the same pollution concentration can produce substantially different AQI scores.

The data sources can also differ. The CPCB uses readings from its continuous ambient air quality monitoring stations, while some private platforms combine government data with readings from privately operated, low-cost sensors. Differences in calibration and data quality can influence the final figures.

The CPCB maintains that its national AQI is the official measure for public dissemination and that scores generated through different methodologies should not be directly compared.

However, these methodological differences do not explain missing observations in government monitoring records or the subsequent revision of station-level PM2.5 readings. Those issues concern the integrity and transparency of the underlying data itself.

What the discrepancies mean for pollution monitoring

The scale of the pollution was also reflected in estimates by environmental experts. Sunil Dahiya, founder of Envirocatalyst, estimated that PM2.5 concentrations averaged between 450 and 550 µg/m³ through Diwali night. He said even a few hours of exposure at such levels could trigger or worsen health problems.

The immediate concern is therefore not simply which AQI figure is correct. It is whether the monitoring network is consistently recording pollution levels during periods when accurate information is most critical.

The agencies need to clarify why stations stopped reporting, how missing readings were handled, why some published figures were subsequently revised and whether the changes affected the official AQI calculations.

A transparent account of the data collection, transmission and validation process would help establish the extent of the discrepancies. It would also allow the public to distinguish between differences arising from AQI calculation methods and those resulting from gaps or changes in the underlying measurements.

Delhi's pollution monitoring system is meant to provide a reliable account of the air people breathe. That requires not only publishing an AQI figure but also ensuring that the data behind it is complete, consistent and open to scrutiny.