Shaw/Abstract The objective of this research is large-eddy simulation (LES) of the atmospheric surface layer in which the lowest portion of the three-dimensional computational domain represents a plant copy such as a forest. The vegetation is represented by a foliage element are density, a drag coefficient, and a sensible heat and water vapor source or sink at each grid point within the canopy. In the LES, scales of motion resolved by the grid network are computed directly in time-dependent form, while sub-grid-scale processes are parameterized. Our plan is to use the simulations to study the nature of turbulence in the vicinity of plant canopies, and to apply the technique to more realistic (such as inhomogeneous) situations. We will perform in-depth studies of coherent structures created within the computed flow fields and follow them from inception to dissipation (an impossible task in the field). Structures will initially be detected using wavelet and other related techniques, complemented by computer animation. It is our aim to obtain as complete a picture as possible of coherent structures developing at the interface between the lower atmosphere and a plant canopy. The method of proper orthogonal decomposition will also be applied to computed LES fields. A complete thermal energy balance of the vegetation and of the underlying soil will be incorporated into the model to more correctly introduce soil evaporation, transpiration, and heat storage in both soil and biomass. The influence of aeroelastic properties of the vegetation will also be introduced and a study made of the phenomenon of honami. As a demonstration of our interests in this regard, we have begun an examination of the windbreak problem. All computation will be performed on the Cray Y-MP at the National Center for Atmospheric Research (NCAR).
