Large Eddy Simulation of Dispersion in Urban Areas

Yu-Heng Tseng, Elie Bou-Zeid, Marc Parlange & Charles Meneveau

Dispersion of air pollution-causing contaminants, such as bio-aerosols or particles from brown-field sites, is affected critically by wind that transports pollutants from the emitter to other locations. Computer simulations of air movement and pollutant transport in the urban environments are especially challenging due to the complex ground topology typically found in cities. A building cluster, consisting of a group of buildings of roughly comparable size, is expected to provide the most complicated flow patterns because the flow interference among buildings needs to be taken into account. In particular, it is impossible to resolve all flow features in the atmospheric flow over an urban canopy. Typically in such applications, fairly coarse grids must be used where much of the physics is unresolved and a subgrid-scale (SGS) model is expected to play a crucial role.

`````

Figure 1(a) Baltimore down-town skyline, (b) Simulation domain representing downtown Baltimore and prevailing wind-direction from the west. Shaded region represents the water in Inner Harbor.

In this research we apply the JHU-LES code that includes the immersed boundary method to simulate bluff bodies in the flow and the Scale-Dependent Lagrangian Dynamic Model developed at JHU (see Bou-Zeid et al. 2005) for simulating subgrid turbulence. This model avoids the prescription of arbitrary model parameters and, instead, uses the resolved-scale turbulence to "learn" about turbulence from the simulation itself. The physical model parameter (relative strength of the eddy-viscosity) is thus not imposed by the user but computed self-consistently. In the JHU-developed version, this dynamic model allows for dependence on grid-scale (i.e. deviations from strict scale-invarance, see Porte-Agel et al. 2000), and the learning process occurs via time averaging along fluid trajectory, i.e. it is Lagrangian (see Meneveau et al. 1996). To validate the LES code in a geometry reminiscent of urban canopies, we simulate the flow over square cylinder, and over a cluster of wall-mounted cubes for which there is is experimental data available. Good results are obtained (Tseng et al. 2005, in preparation).

To study the flow and pollutant transport around realistic buildings, we further extend the model to simulate the flow around downtown Baltimore, MD. The city consists of several skyscrapers and buildings (Figure 1). Figure 2 shows the computational domain. Since the wind flows eastward frequently from the annual statistics, we force the turbulent inflow from the west. We also parameterized the effects of water in Inner Harbor based on different (lower) roughness (Bou-Zeid et al., 2004). The pollutants are transported downstream and detoured due to the building structures (Animations 1 and 2). Strong turbulent mixing is found and the plume is trapped within the building cluster. The pollutants accumulate locally among the buildings. This LES tool can be used to compute probabilities of extreme events such as the probability of a concentration locally to exceed a critical threshold. This model can also be used to ask questions on the “design of cities” concerning location of industrial facilities, potential risks associated with brown fields and construction sites, associated human health impacts.

Charles Meneveau, Department of Mechanical Engineering, Johns Hopkins University, 3400 N. Charles Street, Baltimore MD 21218, USA, Phone: 1-410-516-7802, Fax: 1-(410) 516-7254, email: meneveau@jhu.edu