Clean air is a basic part of a decent life in a city: it makes going outside pleasant as much as safe. The smoky smell of a barbeque might be nice before a summertime dinner, but less so when it coats your city in noxious fumes. Bad air makes for more trips to the hospital, and higher death rates.
Across Canada, recent years have seen a resurgence in poor air quality, and growing wildfires are the biggest culprit. 2025 was the second worst wildfire season on record, just two years after the record-shattering 2023 wildfire season. Human-made climate change is to thank – it makes for more extreme fire weather conditions, hotter and windier summers, and more erratic rainfall.
Below, we've plotted out the number of days that various cities across Canada were exposed to harmful levels of pollution, from 2005 to 2025, based on Canada's Air Quality Health Index (AQHI). The AQHI ranges from 1 to 11, where at 'High' and 'Very High' levels at-risk populations are recommended to reduce or avoid outdoor activities.
The AQHI is a combined metric based on three different pollutants:
- PM2.5: Tiny pollution particles from things like wildfire smoke or traffic that can get deep into your lungs.
- NO2: A gas mainly from vehicles or industry that irritates the lungs.
- O3: Formed when other pollutants mix with sunlight, it can make breathing harder.
We specifically analyze two versions of the AQHI:
- AQHI: Computed using the relative amounts of harmful pollutants in the air over any 3-hour period.
- AQHI+: An extension of AQHI sensitive to extreme events.
For years, deliberate attempts to improve air quality had been working, especially along Ontario's industrial corridor. The new wave of bad air is sweeping across cities in the Prairies, with increasing severity. But it isn't stopping there: slowly but surely, it's making a comeback in Eastern Canada as well.
When we see poor air quality, where does it come from? In our analysis below, PM2.5 is increasingly responsible. This is not surprising, as it's the pollutant released by wildfires and the last three years have been the worst three wildfire seasons on record.
For a time, it seemed that poor air quality was set to become a thing of the past in Canadian cities – a health issue that had received successful treatment. And yet, nowadays, it looks like it's never been worse. In a time of nation-building projects, what will be done to make the air in our cities clean?
Data & Methodology
We obtained hourly pollutant data for PM2.5, NO2, and O3 from the National Air Pollution Surveillance (NAPS) program for 2005-2023, and from OpenAQ for 2024-2025 (through October 31). For cities with multiple sensors, we selected those offering the most consistent temporal coverage and proximity to the city center.
Air quality was assessed using the national Air Quality Health Index (AQHI), developed by Stieb et al. (2008). The AQHI is an exponential formula combining the relative contributions of three-hour averages of PM2.5, NO2, and O3. We also computed AQHI+, a Government of Ontario-specific extension that increases sensitivity to short-term pollution spikes and outliers. Our full implementation of both measures is available here.
Because AQHI is derived from additive pollutant-specific components, we can estimate a single pollutant's percent contribution by recomputing the index while omitting the other two. This forms the basis of the relative contribution analysis presented in the final figure, with the corresponding code available here.
We restrict results to years where at least 50% of pollutant data is available. Due to variation in seasonal trends, we opted against any interpolation of missing data. We then count the number of days exceeding specified thresholds based on daily maximum values to measure exposure, and aggregate these counts into five-year intervals to assess long-term rankings.
All code and processed data are publicly available in the accompanying GitHub repository.