About CIGINT
CIGINT estimates how much air pollution you breathed on your outdoor Strava activities, and expresses the fine particle part of it as a number of cigarettes. It's a rough, fun estimate built on real public data. Here's how each number is made, and where it falls short.
The short version
For each outdoor activity with GPS, CIGINT:
- takes the place and time your activity started;
- looks up the air quality there from Environment and Climate Change Canada, averaged over the 3 hours up to the start;
- works out the Air Quality Health Index (AQHI);
- turns the fine particles you were exposed to, over the time you were out, into a cigarette equivalent.
Where the air quality comes from
Every hour, Environment and Climate Change Canada (ECCC) combines its air quality forecast model with measurements from monitoring stations across Canada and the United States. The result is its Regional Deterministic Air Quality Analysis: a map of pollution near the ground, in squares about 10 km across. It covers Canada (apart from the high Arctic), the contiguous United States (apart from the southern tip of Texas and the Florida Keys) and most of Alaska.
CIGINT reads the square nearest your start for three pollutants:
- PM2.5: fine particles from smoke, exhaust and dust, small enough to reach deep into the lungs;
- NO₂: nitrogen dioxide, mostly from traffic and burning fuel;
- O₃: ground-level ozone, formed in sunlight from other pollutants.
ECCC publishes each hour twice. A preliminary analysis appears within about an hour, before every station has reported, and the final one follows about two hours after the hour. CIGINT shows the preliminary reading straight away, marks it Preliminary, and swaps in the final one once it's out. ECCC keeps these files online for about a month, which is why CIGINT covers roughly your last 30 days.
The Air Quality Health Index
The AQHI is Canada's health-based air quality scale. Health Canada researchers built it from studies of how daily deaths rise with pollution (Stieb et al., 2008). Each pollutant's contribution is added up, so moderate levels of all three count for more than any one alone. The formula uses 3-hour averages, with NO₂ and O₃ in parts per billion and PM2.5 in µg/m³:
AQHI = (1000 ÷ 10.4) × [(e^(0.000871 × NO₂) − 1) + (e^(0.000537 × O₃) − 1) + (e^(0.000487 × PM2.5) − 1)]
It's shown as a whole number from 1 up, with anything over 10 shown as 10+, in the official risk bands:
- 1–3Low risk
- 4–6Moderate risk
- 7–10High risk
- 10+Very high risk
How close is it to the official AQHI? Official values come from monitoring stations, while CIGINT computes its own from the 10 km map. We compared the two for 216 station-hours in British Columbia during two wildfire smoke events. They were 0.29 apart on average, and in the same risk band 97% of the time. Very few of those hours were high, though, and a 10 km square smooths out local peaks, so CIGINT probably reads low in heavy smoke.
The cigarette equivalent
The cigarette number comes from a rule of thumb by Richard and Elizabeth Muller of Berkeley Earth: breathing air with 22 µg/m³ of PM2.5 for a day carries roughly the same health risk as smoking one cigarette.
They got there by comparing deaths, not the amount of smoke inhaled. In the US, smoking causes about 480,000 deaths a year from roughly 350 billion cigarettes, or about one death per 730,000 cigarettes. In China, fine particle pollution was estimated to cause 1.6 million deaths a year at an average of 52 µg/m³. That's the same toll as about 2.4 cigarettes per person per day, and 52 ÷ 2.4 ≈ 22.
CIGINT scales that to the time you were out:
cigarettes = PM2.5 × hours outside ÷ (22 × 24)
So an hour at 22 µg/m³ counts as 1/24 of a cigarette. Hours outside is the activity's elapsed time, stops included, because you're still breathing that air at a red light. The exception is activities that were mostly standing around: if elapsed time is more than 1.5 times moving time, moving time is used instead.
Please read this number loosely:
- The Mullers call it a rough estimate, meant to give a feel for the health effects, not a medical figure.
- It compares health risk, not the weight of particles. Measured by weight, the equivalent would be about ten times lower.
- It was worked out for a whole day of exposure. Applying it to an hour-long run is CIGINT's own simplification.
- It only counts PM2.5. The AQHI also covers NO₂ and ozone.
Place names
The place under each activity comes from OpenStreetMap, looked up once from the start point and then saved. Activities that start within about 100 m of each other share a name, so we ask as rarely as possible.
What CIGINT doesn't account for
- Your route. Air quality is read at the start only. A long ride that climbs out of a smoky valley gets the valley's reading.
- Timing. The reading covers the 3 hours up to your start, not the hours you were actually out.
- How hard you were breathing. Exercise makes you breathe more, and take in more pollution with it (Koehle, 2024). A sprint and a stroll count the same here, so for hard efforts the real dose is probably higher.
- Local sources. A 10 km square can't see the bus you ran behind or the campfire you passed.
- You. Health effects depend on age, fitness and health conditions. This isn't medical advice. For real decisions, check the official AQHI and forecasts.
Sources
- Muller, R. A. & Muller, E. A. (2015, updated 2023). Air pollution and cigarette equivalence. Berkeley Earth.
- Stieb, D. M., Burnett, R. T., Smith-Doiron, M., Brion, O., Shin, H. H. & Economou, V. (2008). A new multipollutant, no-threshold air quality health index based on short-term associations observed in daily time-series analyses. Journal of the Air & Waste Management Association, 58(3), 435–450. doi:10.3155/1047-3289.58.3.435
- Government of Canada. About the Air Quality Health Index.
- Environment and Climate Change Canada, Meteorological Service of Canada. Regional Deterministic Air Quality Analysis (RDAQA). MSC Open Data.
- Hasselback, P. & Taylor, E. (2010). Air Quality Health Index variation across British Columbia. BC Ministry of Healthy Living and Sport.
- Koehle, M. S. (2024). Physiological impacts of atmospheric pollution: Effects of environmental air pollution on exercise. Physiological Reports, 12(7), e16005. doi:10.14814/phy2.16005
Data licences and credits
- Air quality: Data Source: Environment and Climate Change Canada, used under the ECCC Data Servers End-use Licence.
- Place names: © OpenStreetMap contributors, available under the Open Database Licence, via Nominatim.
- Activities: Powered by Strava.
CIGINT isn't affiliated with or endorsed by Strava or Environment and Climate Change Canada. See the privacy page for what we store, or try the demo feed.