AirNow vs PurpleAir: Regulatory Monitors vs Citizen Sensors

Last updated May 20, 2026 · 7 min read

Open two air-quality apps in the same neighborhood and you may see meaningfully different numbers. One reads AQI 95; the other reads AQI 140. Both are honest; they're using different data sources with different strengths and weaknesses. This guide explains the AirNow regulatory monitor network, the PurpleAir citizen sensor network, why they disagree, and which is right when they do.

What AirNow is

AirNow is the U.S. government's official air-quality data service, operated by the EPA in partnership with NOAA, NASA, the National Park Service, and tribal, state, and local air-quality agencies. The data comes from roughly 2,000 regulatory-grade monitoring stations across the country.

The key word is regulatory. AirNow monitors use Federal Reference Methods (FRM) or Federal Equivalent Methods (FEM) — specific EPA-certified instrument designs that cost tens of thousands of dollars apiece, are calibrated and audited on a strict schedule, and are sited according to EPA siting rules to represent population exposure. The data they produce is what the EPA uses to enforce the National Ambient Air Quality Standards.

For a deeper dive on the AirNow data pipeline, see our How AirNow Data Works guide.

What PurpleAir is

PurpleAir is a Utah-based company that sells consumer-grade air-quality sensors for $200–300 each and publishes the data from the network of installed sensors on a public map. As of 2026, there are roughly 30,000 PurpleAir sensors deployed worldwide, with U.S. coverage substantially denser than the AirNow regulatory network — particularly in residential neighborhoods, school campuses, and citizen-science deployments.

The sensors use optical (light-scattering) particle counters — a different and substantially cheaper measurement principle than the regulatory instruments. A laser shines through a sample of air; particles scatter the light; the scattering pattern is processed to estimate particle count and size. The technology is good enough to detect smoke; it's not the same as gravimetric filter weighing.

Why they disagree

Three physical reasons:

1. Different measurement physics

Light-scattering sensors are sensitive to particle size, shape, refractive index, and humidity. They tend to overread during smoke events because smoke particles scatter light efficiently per microgram. They tend to underread or behave erratically for fog (which is droplets, not aerosol particulate), dust storms (large particles), and unusual conditions like fireworks smoke.

2. Different siting

AirNow monitors are sited per EPA rules to represent population exposure — typically away from immediate roadways, at appropriate heights, with airflow that's representative of the broader area. PurpleAirs are sited wherever someone bought one and bolted it to their fence. A PurpleAir on a porch behind a barbecue grill is going to read very differently than one on a school rooftop.

3. Different sampling cadence

AirNow regulatory data goes through quality-assurance review and is published with a small delay (typically 15–45 minutes). PurpleAir publishes near-instantly with no QA layer. So PurpleAir can show transient spikes (from a passing diesel truck, from someone using a leaf blower) that AirNow's smoothed data doesn't capture.

The EPA correction

The disagreement was significant enough that during the 2020 wildfire season the EPA developed a U.S.-wide correction factor for PurpleAir data when used alongside regulatory monitors. The correction adjusts for the systematic over-read during smoke events and produces a value much closer to what a regulatory monitor at the same location would read.

The EPA's Fire and Smoke Map ingests PurpleAir data, applies the correction, and displays it alongside regulatory monitors. The resulting blended product gives you:

• The accuracy and authority of regulatory data where AirNow monitors exist.
• The geographic density of PurpleAir's network where the regulatory network is sparse.
• A consistent correction applied so the numbers across the network are comparable.

This is the most useful integrated data product during active fire events.

When to trust which

What Smog Report uses

Smog Report (the iOS app) pulls from the EPA AirNow API, which means it uses regulatory-monitor data — the authoritative version. If you live in a neighborhood with a closer PurpleAir than the nearest AirNow station and you want microclimate awareness, the EPA Fire and Smoke Map web view is the right supplemental tool during fire events. For routine year-round AQI checking, AirNow data is what you want.

Caveats about PurpleAir as an absolute measure

A few things worth knowing if you rely on PurpleAir data:

• Many apps display uncorrected PurpleAir. If an app shows raw PurpleAir AQI in a smoke event, it's likely 30–50 AQI points higher than corresponding regulatory readings.
• Two-channel sensors check themselves. Each PurpleAir has two independent sensors and reports both. If they diverge significantly, one is malfunctioning. The PurpleAir map flags this.
• The PurpleAir map's color is not standard EPA AQI. By default it uses the LRAPA conversion (Lane Regional Air Protection Agency, Oregon) — a wildfire-tuned correction. Toggle to the U.S. EPA conversion for AirNow-comparable readings.

Other low-cost sensor networks

• AirGradient — Thailand-based, open-source hardware, also light-scattering. Smaller network, growing.
• IQAir AirVisual — consumer/commercial sensors, integrated with the IQAir app. Their data feeds the IQAir global ranking.
• Awair Element — primarily indoor, includes CO₂ and VOC sensing alongside PM2.5.
• Clarity Movement — commercial deployments for cities, not consumer-facing.

All of these have similar trade-offs to PurpleAir: cheaper, denser, but less accurate than regulatory monitors per individual reading.

Primary sources: EPA AirNow · EPA Air Sensor Toolbox · EPA AirNow Fire and Smoke Map