During the past few years, we have developed and demonstrated the utility of a building scale CFD (Computational Fluid Dynamics) model for simulating flow and dispersion of chemical/biological agents released in the urban environment. The results from such a model can be highly useful for emergency planning of special events, vulnerability analyses, post-event assessments, and development of mitigation strategies. However, due to its high demand on computer resources and long turnaround time, it is not feasible to use the present model for emergency response situations. To meet such needs, we are developing a simplified CFD version suitable for operational use on the computational platforms of the National Atmospheric Release Advisory Center. In this new approach, targeted buildings can be explicitly treated with fine grid resolution but non-targeted buildings are modeled as drag elements (or virtual buildings) with coarser grid resolution. This approach reduces significantly the number of grid points required in a dispersion simulation because less precision is needed for non-targeted buildings. Preliminary test of the new approach for a hypothetical release in the downtown area of Salt Lake City indicates that the virtual building results are reasonably similar to the conventional CFD results but with more than an order of magnitude savings in computational cost.

Flow and dispersion in urban areas sometimes involve highly variable winds, with turbulent fluctuations of the same order as the mean velocity. To account for such effects, we have also developed a capability for imposing time-dependent boundary conditions (or forcing) in the flow and dispersion simulations. Such a capability has been tested and the predicted results compared with field data from IOP-7 of the Urban 2000 experiments. Drastic improvements were observed in the dispersion results from the simulation using time-dependent forcing over an analogous simulation using steady-state forcing with the mean wind conditions.

The above new capabilities in our CFD model will be further evaluated using field data from the Urban 2000 experiments and findings will be reported at the conference.

Acknowledgement. This work was performed under the auspices of the U.S. Department of Energy by the University of California, Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48.

* Corresponding author address: Stevens T. Chan, Lawrence Livermore National Laboratory, P.O. Box 808, L-103, Livermore, CA 94551, e-mail: schan@llnl.gov