Three phones, three different AQI numbers — CPCB vs IQAir vs WAQI methodology

Why Your Phone’s AQI Disagrees With Your Monitor: CPCB vs IQAir vs WAQI Methodology

Open three air-quality apps on the same morning in Delhi and you’ll see three different numbers — sometimes wildly different. AQI India’s app reads 285. IQAir reads 174. The CPCB Sameer app reads 272. WAQI (the open-data site at aqicn.org) reads 180. Your indoor monitor shows 22. None of them are “wrong.” They are using different AQI scales calibrated to different country regulations and different breakpoints for the same underlying PM2.5 measurement. This page explains the methodology differences and which number to trust for what purpose.

Key numbers

0–500 — Indian CPCB AQI scale (capped at 500, with anything higher grouped at “severe”)
0–500 — US EPA AQI scale (also capped at 500 in standard reporting)
Indian “Good” = 0–50 AQI, US “Good” = 0–50 AQI — same number, different concentrations
Indian “Severe” = 401–500, US “Hazardous” = 301–500
Indian PM2.5 24-hour standard = 60 µg/m³; US standard = 35 µg/m³; WHO = 15 µg/m³

How AQI is calculated

Both the Indian (CPCB) and US (EPA) AQI use the same general formula:

AQI = ((AQI_high - AQI_low) / (Conc_high - Conc_low)) × (Conc_measured - Conc_low) + AQI_low

This linearly maps a measured pollutant concentration to an AQI value, using lookup-table “breakpoints” that define the boundaries of AQI categories.

The formula is identical. The breakpoints are different.

The breakpoint difference

For PM2.5 (the dominant pollutant for Indian air-quality discussion), the breakpoints differ:

Indian CPCB breakpoints (PM2.5, 24-hour avg): - AQI 0–50 (“Good”): 0–30 µg/m³ - AQI 51–100 (“Satisfactory”): 31–60 µg/m³ - AQI 101–200 (“Moderate”): 61–90 µg/m³ - AQI 201–300 (“Poor”): 91–120 µg/m³ - AQI 301–400 (“Very Poor”): 121–250 µg/m³ - AQI 401–500 (“Severe”): >250 µg/m³

US EPA breakpoints (PM2.5, 24-hour avg): - AQI 0–50 (“Good”): 0–12 µg/m³ - AQI 51–100 (“Moderate”): 12.1–35.4 µg/m³ - AQI 101–150 (“Unhealthy for Sensitive Groups”): 35.5–55.4 µg/m³ - AQI 151–200 (“Unhealthy”): 55.5–150.4 µg/m³ - AQI 201–300 (“Very Unhealthy”): 150.5–250.4 µg/m³ - AQI 301–500 (“Hazardous”): >250.4 µg/m³

For the same measured PM2.5, the US scale produces a higher AQI in the lower ranges (because the breakpoints are tighter) and a lower AQI in the upper ranges (because the US scale doesn’t go above 500).

Worked example

A morning measurement of PM2.5 = 100 µg/m³:

Indian CPCB AQI: falls in 91–120 range → AQI 201–300 (“Poor”), calculated to approximately 244
US AQI / IQAir: falls in 55.5–150.4 range → AQI 151–200 (“Unhealthy”), calculated to approximately 174

Same pollution. Two different numbers (244 vs. 174) and two different category labels (“Poor” vs. “Unhealthy”). Neither is wrong. Both are correctly computing AQI per their respective country’s standards.

For very high concentrations the divergence grows:

PM2.5 = 1,100 µg/m³ (recorded in Delhi on the worst Diwali days):
Indian CPCB: AQI ~1,054 (with the >500 calculation extended)
US/IQAir: AQI ~2,043

The IQAir calculator’s larger number reflects the US scale’s more linear extrapolation; the CPCB number reflects the Indian scale’s tighter breakpoints at extreme concentrations.

Why the scales are different

Two reasons:

1. The breakpoints reflect each country’s air-quality standards. India’s National Ambient Air Quality Standard (NAAQS) for PM2.5 24-hour is 60 µg/m³. The CPCB AQI is built around this — concentrations near 60 µg/m³ correspond to the “Satisfactory–Moderate” boundary.

The US EPA NAAQS for PM2.5 24-hour is 35 µg/m³. The US AQI is built around this — concentrations near 35 µg/m³ correspond to the “Moderate–Unhealthy for Sensitive Groups” boundary.

Both scales are calibrated to make their respective national standards “meaningful.” For an Indian regulator monitoring compliance, the CPCB scale makes sense. For a US-residing visitor judging risk, the US scale (via IQAir) makes sense.

2. WHO 24-hour PM2.5 guideline is 15 µg/m³. Both national standards are more lenient than WHO. The CPCB standard is 4× the WHO limit; the US standard is ~2.3×. Both AQI scales accordingly under-state risk relative to global health science.

Which app to trust for what

For Indian regulatory and policy understanding: - CPCB Sameer app or aqi.in with Indian AQI displayed. Gives you the official Indian number.

For health-protective decision-making: - IQAir or AirVisual with US AQI. The tighter breakpoints align better with WHO-aligned health science. - The Indian “Moderate” zone (AQI 100–200, PM2.5 60–90 µg/m³) is the US “Unhealthy” zone — categorically different colour and recommended response. Believe the US scale.

For raw concentration: - All apps display µg/m³ alongside AQI. The PM2.5 µg/m³ number is the underlying physical measurement and doesn’t depend on which scale is used. Compare across apps to verify station readings.

For specific location precision: - WAQI / aqicn.org has the most monitoring stations globally including aggregated low-cost sensor data. Best for hyper-local readings; methodology is the same as IQAir (US scale).

For your own indoor air: - Your own NDIR / particle monitor. Don’t trust city-scale data for what’s in your bedroom.

What aqi0 uses internally

For aqi0 site surveys and field measurements:

Raw µg/m³ for PM2.5 and PM10 as the primary unit
Cross-referenced with WHO guidelines (15 µg/m³ 24-hour, 5 µg/m³ annual) for health context
Display AQI only when comparing to public data, with explicit note of which scale
WHO-aligned thresholds for “good indoor air” — under 10 µg/m³ PM2.5

We don’t think in CPCB AQI numbers because the breakpoints don’t map to the health science.

Why this matters in practice

Three implications:

1. A reading of “AQI 200” on different apps means different things. The category label (“Poor” or “Unhealthy”) and the implied response (close windows, mask up, restrict children’s outdoor play) varies. Read the category, not just the number.

2. Indian regulatory compliance ≠ health protection. A day reading “AQI 100” on the CPCB Sameer app corresponds to PM2.5 around 60 µg/m³ — 4× the WHO 24-hour limit. The Indian “Satisfactory” rating is not the same as “safe.”

3. Use raw µg/m³ when comparing to WHO standards. The unambiguous comparison is to WHO’s annual 5 µg/m³ guideline or the 24-hour 15 µg/m³ guideline. Both AQI scales obscure this.

FAQ

Which is the “right” AQI? Neither is wrong; they answer different questions. Indian AQI answers “how does this compare to Indian regulations.” US AQI answers “how does this compare to US regulations.” Neither perfectly answers “is this safe for my health” — that requires WHO context.

Why does IQAir often show worse readings than CPCB? For the same concentration, IQAir’s US scale typically produces a higher AQI in the lower ranges. This is methodology, not a “stricter” measurement.

Are the underlying PM2.5 readings the same across apps? Mostly yes if they’re pulling from the same monitoring stations. Different apps may average across slightly different station networks or apply different temporal smoothing.

Can I trust crowd-sourced low-cost sensor readings? Individual sensors drift. Aggregated networks (IQAir, WAQI) usually reconcile via cross-station comparisons. For your own home indoor monitor, calibrate against a known reference periodically.

Should I use the Indian or American scale? For Indian regulatory and policy purposes: Indian. For health-protective decisions: US (tighter breakpoints). For absolute health context: refer to WHO µg/m³ values, not any AQI scale.