| There is an increased demand for regional estimates of surface energy fluxes and surface soil moisture in many scientific fields. Point measurements, despite being accurate, are limited in spatial extent depending on the heterogeneity of surfaces that they represent. Boundary layer models also have limitations in spatial domain due to the requirement of spatially distributed input data. Remote sensing with airborne and satellite-based sensors provides the required spatially distributed data for boundary layer models.; Many boundary layer models have been developed over the years to simulate boundary layer processes. The soil-vegetation-atmosphere-transfer (SVAT) model that is used in this study is one of them. Modifications were made to the SVAT model by including the “top down-bottom up” approach of solving the turbulence closure problem and the Ball-Berry method of estimating stomatal resistances. Also, the SVAT model was reorganized to improve its accessibility and understandability.; The effect of spatial variability of slope and aspect (terrain) on the coupling of the SVAT model with remotely sensed data was tested using a study site that has ranges in slope of 0–20 degrees. Results indicated that assuming flat surfaces in the formulation of the SVAT model leads to the failure of the “triangle method” in coupling the model with remotely sensed data. Inclusion of slope and aspect into the model resulted in a good fit between the model outputs and remotely sensed data. Multiple linear regression between the geophysical variables and surface parameters resulted in r 2 values ranging from 0.82 to 0.99.; Sensitivity of the SVAT model to variations in vegetation types was assessed using Aspen, Sagebrush, and Idaho fescue. Comparisons between calculated and derived geophysical variables for the three vegetation types were made using graphs, maps, and statistical methods. The statistical methods used were the mean absolute difference (MAD), the maximum absolute difference (MXAD), the root mean squared difference (RMSD), and the percent mean difference (PMD). On average, differences that result from using parameters of the different vegetation types in the SVAT model are within measurement uncertainties of the variables. However, under certain combinations of surface parameters, the differences become significant. |
