Breathing easy

Breathing easy

Researchers at York University are developing better air quality forecasts they hope will help Canadians avoid some potentially fatal health problems
November 1, 2005

What’s the favourite topic of conversation for Canadians? The weather.

As a nation that’s used to weather extremes of all kinds, our national conversation has always had a rich and varied vocabulary. These days, the vocabulary is expanding and the conversation is increasingly including another atmospheric factor: air quality.

From Vancouver to St. John’s, “smog days,” “particulate matter,” and “ground-level ozone” are now household terms. And with good reason. These pollutants are damaging the health of millions of Canadians and resulting in the premature deaths of thousands every year.

Traditionally, weather forecasts haven’t been able to provide Canadians with the valuable air quality information they need. That’s about to change. Soon, Canadians will be able to talk about weather and air quality in a single breath as weather forecasts and daily air-quality advisories are combined into new, sophisticated environmental forecasts. State-of-the-art computational air-quality models and forecasting techniques presently being developed by researchers at York University will soon be core components of these future environmental forecasts.

“What we’re doing is building our new air quality models into existing weather forecasts to create coupled models,” says John McConnell, an atmospheric scientist at York University’s Department of Earth and Space Science and Engineering.

The ability of researchers to forecast air quality currently lags behind that for weather forecasting. For example, while Environment Canada’s weather predictions range to five-day forecasts, Air Quality Index is only a daily forecast that’s largely built on immediate observations rather than predictive powers.

Future air quality forecasts will be able to literally build on decades of experience that Canadian scientists have in developing weather models. Both types of forecasts are based on sophisticated numeric models of atmospheric and chemical processes. These models are run on powerful computers to simulate future conditions. The research being done by McConnell’s research group depends on a new IBM supercomputer, similar to the federal government’s most powerful civilian computer, which is used by Meteorological Service of Canada (MSC) weather forecasters in Dorval, Quebec. The air quality model they’re developing, in close collaboration with MSC researchers, is similarly a “super” version of current forecasting techniques.

“Air quality is not simply a question of acid rain, or ozone, or particulate matter considered separately,” says McConnell. “All are important and, to varying degrees, they interact. So it’s necessary to have comprehensive air quality models that can tackle all of the problems simultaneously within the same modelling environment.”

As well as including a wider range of chemicals, the new air quality model is global in scope. It’s based on the MSC’s Global Environmental Multiscale (GEM) model, the one used in weather forecasting. “One of the limitations we’re facing right now within our current air quality forecasting framework is that there’s a significant intercontinental inflow of pollutants that aren’t taken into account,” says Richard Menard, a research scientist with MSC’s Air Quality Research Branch, and McConnell’s key collaborator on the air quality modeling work.

The new global model is also looking to the sky for help. It’s the first in Canada that has been developed to include satellite data and will incorporate air quality observations from Canada’s recently upgraded network of ground-based pollution sensors as part of the Canadian Sun-Photometer Network.

Within years, this unique combination of inputs will help provide a new world of environmental forecasts so that Canadians are better prepared to protect themselves from, and help reduce, air pollution.


Call it Canada’s summer of bad air.  The summer of 2005 saw record high levels of air pollution in cities across the country. Toronto, the nation’s largest city, experienced a record number of smog days.  The smog not only left its toll on the city’s residents, but the toxic haze extended to cover a larger area of Ontario—and affected a greater number of people—than ever before.

“We’re only beginning to understand how air pollution affects  our health,” says John McConnell, whose York University team is leading the way in developing sophisticated computer models to understand and predict air quality.

In June 2005, the Ontario Medical Association estimated that 17,000 Ontarians will be hospitalized in 2005 as a result of poisons in the air.  Worse still, 5,800 people will die prematurely, and poor air quality will cost about $ 1 billion in lost wages and health care costs.

In light of the mounting human and financial toll, it has become even more important to be able to accurately predict air quality so that Canadians can take protective and preventative measures.  “The new air quality models we’re developing will greatly increase the predictability of the atmospheric chemistry,” says Richard Menard of the Meteorological Service of Canada (MSC).

The York/MSC model will also provide epidemiologists with unparalleled information on the distribution of various-sized particles in the air.  Thirty years ago, McConnell says, poor air quality was associated with large, visible, air-borne particles such as soot from coal furnaces.  Today, the air is increasingly full of invisible killers.  Nano-sized diesel exhaust particles, for example, are coming under scrutiny by health researchers as the cause of respiratory problems. 

“An air quality modelling tool with detailed emission modelling will be invaluable in assessing the impact and benefit of emission reductions in a more objective and targeted manner than is presently possible,” says McConnell.


McConnell’s research is highly collaborative, reflecting the complexity involved in air quality forecasting. In the model development phase, he works closely with the Meteorological Service of Canada, a world leader in the development of computational weather, climate, and environmental models.

The success of air quality forecasts depends on the quality of the data inputs. To ensure the best possible outcome, McConnell’s group is working with the University of Toronto’s James Drummond, for satellite-based information on atmospheric pollutants, and the University of Sherbrooke’s Norm O’Neill, a leader in ground-based atmospheric sensing.

Learn More

John McConnell’s research is part of Canada’s extensive Multiscale Air Quality Modelling Network (MAQNet).

Learn more about Environment Canada's Canadian Meteorological Centre.

For an overview of atmospheric research in Canada, visit the Canadian Foundation for Climate and Atmospheric Sciences (CFCAS).

Learn about Canada’s first major instrument to measure pollution of the Earth’s atmosphere from space and NASA: TERRA.

Find out about Canada’s Atmospheric Chemistry Experiment (ACE) project.

As part of an international research network, the Canadian Sun-Photometer Network (AEROCAN) provides information on aerosols in the atmosphere.

Visit Environment Canada’s Clean Air Online.