8.1
Numerical modeling from mesoscale to urban scale to building scale
Michael J. Brown, LANL, Los Alamos, NM; and M. Leach, J. Reisner, D. E. Stevens, S. Smith, S. T. Chin, S. Chan, and R. L. Lee
For plumes originating in urban areas, the larger-scale transport may depend on the near-source dispersal patterns within the city. In addition, the near-source dispersal behavior within the urban canopy may depend on the larger-scale flow features. In this presentation, we show meteorological and tracer dispersion simulations from a hierarchy of three prognostic models that were centered around Salt Lake City, Utah. The mesoscale COAMPS, the urban-scale HIGRAD, and the building-scale FEM3MP models were used to perform the simulations in a nested configuration. The multi-scale modeling approach allows us to explicitly capture mesoscale flow features over the large domain as well the effects of individual buildings in the smaller area of interest. The COAMPS model utilized a 36, 24, and 4 km resolution nested grid mesh and captured regional meteorological effects. Wind, temperature, and humidity profiles computed by COAMPS were used to drive the HIGRAD urban-scale simulation. This simulation was performed over a 100 km domain with a variable horizontal grid resolution of 10 to 500 meters. HIGRAD meteorological profiles accounting for terrain effects and resolving building clusters were then used to drive the FEM3MP model at grid resolutions on the order of meters. Here, the flow field around individual buildings was explicitly resolved. We will discuss an early morning and an afternoon simulation case and evaluate the effects of stability on transport and dispersion on the urban and building scales. We also will study whether or not the dispersion over the mesoscale is sensitive to what happens on the building scale. This work represents ongoing work to better understand transport and dispersion within cities and the interaction of mesoscale and urban scale flow dynamics.
Session 8, Local- Meso-scale modelling
Wednesday, 16 August 2000, 1:45 PM-2:45 PM
Next paper