Eighth Symposium on the Urban Environment
The Energy Committee

J24.4

Two-way coupling approach between WRF NWP and EULAG LES models for urban area transport and dispersion modeling

Andrzej A. Wyszogrodzki, NCAR, Boulder, CO; and F. Chen, S. G. Miao, and J. Michalakes

In order to investigate small scale transport and dispersion (T&D) within urban areas as well as to improve urban area parameterizations within mesoscale models, we couple LES-based urban-scale model (EULAG) with the mesoscale numerical weather prediction- Weather Research and Forecasting (WRF) system.

The models are coupled using two-way downscale-upscale transfer approach. The mesoscale WRF conditions are downscaled and used as initial and lateral boundary conditions for EULAG T&D simulations. EULAG is resolving the building structures explicitly using immersed boundary (IB)approach to account for urban aerodynamic features as channeling, enhanced vertical mixing, downwash and street level flow. These explicitly-resolved urban-scale flow structures are then aggregated and feedback to WRF to improve mesoscale forecast over and downstream urban areas. The synchronization between the models, grid transformation and its continuous transfer is controlled by computationally efficient software architecture: Model Computing Environmental Library (MCEL).

We demonstrate potential benefits of two-way coupling and evaluate the coupled system in the contest of Oklahoma City Joint Urban 2003 (JU2003) experiment. The mesoscale calculations are performed at 500 meter grid spacing while the building resolving simulation uses the high-resolution of a few meters. We focus on the accuracy of both the mesoscale and LES wind filed statistics and the LES T&D of a SF6 tracer gas in the convective daytime (IOP6) and nighttime (IOP8) scenarios.

wrf recording  Recorded presentation

Joint Session 24, Modeling Tools for Urban and Complex Terrain Environments
Thursday, 15 January 2009, 3:30 PM-5:00 PM, Room 124A

Previous paper  Next paper

Browse or search entire meeting

AMS Home Page