In the summer of 1999, the Meteorological Service of Canada (MSC) initiated an Air Quality Forecasting program for Eastern Canada, using the Canadian Hemispheric and Regional Ozone NOx System (CHRONOS) as the chemical transport model. The Air Quality Prediction Program was then extended to cover all of Canada in 2001. The meteorological driver used by the chemical transport model is a regional version of the Global Environmental Multi-scale model (GEM), MSC’s operational weather forecast model, and, since emissions from the United States of America have a significant impact on air quality in Canada, the model domain covers most of Canada and the United States. The 2001 prediction program used gridded emissions based on the 1990 emissions inventory. Work is currently underway to update this emission inventory. Tropospheric ozone levels obtained as output from CHRONOS are used as important guidance in the issuance of air quality forecasts in the various regions and provinces of Canada (Maps of predicted ozone concentration generated by the CHRONOS system can be viewed at http://www.msc-smc.ec.gc.ca/aq_smog/chronos_e.cfm). Since these model outputs are used in real-time by meteorologists in the regional forecast offices across Canada, immediate feedback can be provided on the performance of the model, especially during smog episodes. Furthermore, a systematic model performance evaluation is done at the end of each summer by comparing the model predictions with observations collected by ground-level ozone measurement networks.

For the summer 2002 prediction program an experimental particulate matter (PM) component was added to the model. The PM component consisted of bulk PM2.5 and PM10, made up of primary components, secondary sulphate, secondary organic aerosol, and condensed water.

While continent-wide maps of the tropospheric ozone component are available on the web for public access, the PM components are being evaluated internally by Environment Canada during the summer 2002 before making them publicly available.

The presentation will provide further details of the prediction program, the model, and the evaluation of its performance.