The Canadian urban dispersion modeling (CUDM) system: Results from applications over Vancouver and Toronto

The Canadian Urban Dispersion Modeling (CUDM) system was designed to respond effectively to the possibility of a terrorist or accidental release of a chemical, biological, radiological or nuclear (CBRN) agent in a large and densely populated urban area. During such an event, rapid decisions need to be made concerning the transport, dispersion, deposition and fate of the CBRN agent and its concomitant effects on the exposed population. The system was developed over the past five years, through funding from the CBRN Research and Technology Initiative (CRTI) program. The system is also applied to planning scenarios as well as forensic assessment.

It is a multiscale system that can be used to predict the mean flow and turbulence in the urban environment and the dispersion of contaminants. The first component of the system involves a cascade of meteorological models from the operational regional model to an urbanized mesoscale model. The second component includes urbanSTREAM, a building-aware computational fluid dynamics (CFD) model which resolves the highly disturbed flow down to the street or building scale. The third component is the urban dispersion model. A Lagrangian stochastic (LS) model for urban dispersion referred to as urbanLS has been implemented in the integrative multiscale urban modeling system.

The different components of the prototype were validated primarily using the Joint Urban 2003 Oklahoma City campaign during the first phase of the project. The second phase is focused on the seamless integration of the different modeling components in order to produce an operational prototype.

This prototype was executed in test mode during the Vancouver 2010 Winter Olympics and Paralympics. Following the Olympics period, upgrades to the modeling system were included in preparation for support during the G8/G20 summit meetings in Toronto in June 2010. The presentation will highlight the results associated with these two test periods. Results from an initial sensitivity study will be shown as well as the assessment of the benefits from using a high resolution meteorological model to drive the CFD model.