Lagrangian particle modeling of dispersion in planetary boundary layers with and without a plant canopy

We investigate passive scalar dispersion from point sources in neutral (NBL)

and stable (SBL) boundary layers using a Lagrangian particle model (LPM)

driven by velocity fields from large-eddy simulations (LESs). The approach

is the same as that adopted earlier for dispersion in convective and stable

boundary layers (e.g., Weil et al., 2004) with particles tracked using

velocities from the LES resolved fields and a stochastic subgrid-scale model.

In the LES, the geostrophic wind was either 5 m/s or 8 m/s, and the boundary

layer height was either 615 m (NBL) or 200 m (SBL). A plant canopy and its

mechanical drag were modeled using the approach of Patton et al. (2003).

Dispersion simulations were conducted for source heights ranging from the surface

to the upper part of the boundary layer. As a test of the LPM-LES approach,

we compared the surface concentrations from a surface release in the no-canopy

case with field data from the Prairie Grass experiments and found good agreement.

For sources within and above a canopy, the mean plume height increased more

rapidly with downstream distance than in the no-canopy case. This was caused

by the positive vertical gradient of the vertical velocity variance over a

greater height range near the surface (i.e., the roughness sublayer) than in

the no-canopy case. The positive gradient led to a greater mean upward

"drift" velocity of the particles. In addition, plumes from all release

heights showed a mean lateral deflection due to the wind direction shear,

which was stronger in the no-canopy case. These and other aspects of the

results will be discussed.