We describe important feedbacks between model and experimental data which improved the quality of the project. For example, preliminary numerical results were used to identify key regions of the flow and arrays of anemometers were placed in these areas. New insights gained from the field program, such as the importance of the gaps in a line of trees close to the building, or more accurate estimates of the drag caused by the trees, were fed back into the second round of modeling. Gaps in the treeline, for example, proved crucial in obtaining accurate turning of the modeled wind vectors around the southeast corner of the building.
Various forms of turbulence closure are discussed and compared. In particular, several Reynolds Averaged Navier-Stokes (RANS) solutions utilizing different turbulence models and Large-Eddy Simulations (LES) are compared. We discuss our efforts to assess the accuracy of transient and high frequency turbulence structures present in LES results. Several sonic anemometers were placed on "high frequency mode" for a two hour period and this data is compared to our Large-Eddy Simulations.
This work was performed under the auspices of the U. S. Department of Energy by the University of California, Lawrence Livermore National Laboratory, under contract No. W-7405-ENG-48