Air-Quality Modelling over Quebec
Testing a chemical transport model against measurements across Quebec, on grids from a few kilometres down to 1 km over Montréal and Québec City.
- Category
- Air Quality
- Location
- Quebec, Canada
- Industry
- Environment & Public Sector
Overview
Before an air-quality model can be trusted for forecasts or for policy studies, it has to be tested against what was actually measured. We took part in a study, led by the University of Toronto with partners in Canada, France and Sweden, that validated the Polair3D chemical transport model over Quebec.
The study
- Nested grids: the model was run from a regional grid, through a 3 km grid, down to 1 km windows over Montréal and Québec City.
- Four months of 2018: one month in each season was simulated, and the results were compared with measurements at monitoring stations.
- Six pollutants: carbon monoxide, ozone, nitrogen dioxide, nitric oxide, sulphur dioxide and fine particles (PM2.5) were evaluated.
What the study found
- Spatial and seasonal patterns: the model reproduces them well.
- Hour-to-hour variation at a single site: is harder to capture: bias and correlation vary widely from site to site, on both the 3 km and the 1 km grid.
- Finer is not always better: the 3 km grid correlated slightly better for carbon monoxide, nitrogen dioxide and nitric oxide, and the 1 km grid for ozone, sulphur dioxide and fine particles.
- Comparable with other models: the performance was in line with other Canadian models and with Polair3D's performance in Europe.
Why it matters
An honest validation tells users what a model can and cannot do. Here it shows that the model suits mapping the seasonal and spatial patterns of air pollution, and where its day-to-day skill is limited and needs work. That is the evidence required before such a model is used for forecasts or for assessing health effects and policy.
Reference
- Yamanouchi, S., Gamage, S. M., Torbatian, S., Zalzal, J., Minet, L., Smargiassi, A., Liu, Y., Liu, L., Azargoshasbi, F., Kim, J., Kim, Y., Yazgi, D. and Hatzopoulou, M. (2024). Validation and analysis of the Polair3D v1.11 chemical transport model over Quebec. Geoscientific Model Development, 17, 3579–3597. doi:10.5194/gmd-17-3579-2024
Figure: Fig. 1 of the paper above, CC BY 4.0.