Prediction of regional fluxes of momentum, heat and water vapor using hydrologic, weather and climate models is subject to substantial errors associated with our limited ability to account for the combined effects of land surface variability and atmospheric instability. A novel combination of wind tunnel experiments and large-eddy simulations (LES) coupled with remotely-sensed surface conditions is proposed to advance our understanding of the complex coupling between land surface heterogeneity and the atmospheric boundary layer. First, a framework will be created to evaluate the performance of the new generation LES (including recent advances in tuning-free dynamic subgrid-scale models) over heterogeneous surfaces. This will be achieved through wind tunnel experiments as well as field measurements obtained during the Southern Great Plains 1997 (SGP-97) Hydrology Experiment. Special emphasis will be placed on studying the ability of the recently developed tuning-free dynamic subgrid-scale parameterizations to consistently adjust to the flow inhomogeneity and anisotropy associated with changes in surface roughness and temperature. We will also use synthetic stochastic surrogate fields to study the effect of spatial organization of the surface properties (aerodynamic roughness, temperature and moisture) on the magnitude and distribution of the turbulent fluxes. Furthermore, our results will be used to assess the limits of conventional similarity theory and surface heterogeneity parameterizations used in operational forecast models. Physically-based modifications of similarity theory will be proposed to include the effects of the spatial autocorrelation and cross-correlation structure of surface properties.
The proposed research is expected to improve our ability to model land-atmosphere exchange processes and, as a result, the accuracy of hydrologic, weather and climate models. Although substantial improvements of large-scale numerical models have been achieved over the last several years, a major challenge still remains to account for the effects of heterogeneity of land-surface properties (temperature, roughness, moisture) at scales smaller than the resolution of the numerical models. We anticipate that our research will propose sub-grid parameterizations that are computationally efficient and physically attractive such that they can be used in operational forecast models. In addition, a new web-based educational tool will be developed to take advantage of the educational potential of visualizations/animations of high-resolution simulations of turbulent transport in the atmospheric boundary layer. The new tool will be used in undergraduate and graduate courses as well as outreach activities for high school students. Furthermore, two graduate students will receive training and will also participate in the elaboration and implementation of the educational tool.
