Lagrangian modeling of dispersion using LES fields: importance of the Prairie Grass and CONDORS experiments

We report on the use of field observations from the "classic" Prairie Grass and CONDORS experiments to test a Lagrangian particle dispersion model (LPDM). The LPDM is driven by velocity fields from large-eddy simulations (LESs), which have been made for both the convective (CBL) and stable (SBL) boundary layers. In the LPDM, passive particles from a source are tracked or followed for tens of thousands of particle trajectories with the mean concentration obtained from the local particle number density, i.e., the probability density function of particle position. The LES for the CBL used a 5 km X 5 km X 2 km domain with a 1-km boundary layer height, whereas the SBL domain was 0.4 km X 0.4 km X 04 km, with an SBL depth of 200 m, and a grid size of 2 m. Although we discuss both boundary layers, the focus will be on the SBL.

The Prairie Grass experiment provided a challenging test for the LPDM and LES since the source was at the surface or in the first vertical grid above the surface. In this grid, the "resolved" LES motion was suppressed due to the requirement of a zero vertical velocity at the surface, and thus the subgrid-scale or "unresolved" motion was a considerable fraction of the total turbulent kinetic energy. Nevertheless, LPDM predictions of surface concentrations agreed quite well with the Prairie Grass data for both boundary layer types. In addition, the LPDM simulations for the SBL exhibited interesting and realistic features such as the effect of wind direction shear in enhancing the lateral spread and producing a "tilted" plume, which was revealed by three-dimensional perspective views. The vertical dispersion agreed well with Taylor's (1921) statistical theory, and the lateral dispersion also showed agreement with this theory at early times, while exhibiting the enhanced spread at late times. These and other aspects of the dispersion and results will be discussed.