A model for flow and dispersion around buildings and its validation using laboratory measurements

Numerical modeling of airflow and pollutant dispersion around buildings is a challenging task due to the geometrical variations of buildings and the extremely complex flow created by such surface-mounted obstacles. The airflow around buildings inevitably involves impingement and separation regions, a multiple votex system with building wakes, and the jetting effects in street canyons. The interference from adjacent buildings further complicates the flow and dispersion patterns. Accurate simulations of such building-scale transport phenomena requires not only appropriate physics submodels but also significant computing resources. We have developed an efficient, high resolution CFD model for the purpose of simulating chemical and biological releases in an urban area. Our primary goal is to support incident response and preparedness in the areas of emergency response planning, vulnerability analysis, and the development of mitigation techniques.

Our model is based on solving the three-dimensional, time-dependent Navier-Stokes equations. The basic numerical algorithm is composed of an innovative finite element approach to accurately represent complex building shapes and a fully implicit projection method for efficient time integration. In order to model the turbulence processes accurately, we have implemented a nonlinear (cubic) eddy viscosity and a large eddy simulation turbulence submodels. In addition, our model physics includes UV radiation decay, aerosol effects, surface energy budget, and tree canopy parameterization. Our model has been developed to run on both the serial and massively parallel computer platforms.

We have performed model validations using, among others, the tow-tank experimental data obtained from flow and dispersion past a cubical building and also similar data around a 2-D array of buildings in a wind tunnel. Despite the relatively simple geometry of the buildings, these experiments exhibited the essential features usually observed in full-scale building flows. Therefore laboratory experiments can provide very useful data sets for validating our model.

In this paper, we will describe briefly the salient features of our model, present results from a model-data comparison using the above data sets, and discuss the performance of the turbulence submodels that have been tested.

This work was performed under the auspices of the U.S. Department of Energy by University of California Lawrence Livermore National Laboratory under contract number W-7405-ENG-48.