Accurate representation of surface characteristics in a mesoscale model is critically important to the production of meteorological inputs used for air quality modeling. In the Weather Research and Forecasting (WRF) model, various model inputs are required in the land surface model (LSM) to describe the momentum, heat and moisture exchange between the surface and atmosphere. The high resolution satellite-based land use and land cover (LULC) data from the University of Texas Center for Space Research (UT-CSR) and Texas Forest Service (TFS) is incorporated into the WRF model. Two land use categories, residential and residential tree, are introduced in addition to the urban category to represent Houston metropolitan area. The vegetation fraction, a critical factor determining the evapotranspiration processes, is compiled based on the satellite-based LULC data. The WRF model is run with Noah LSM to simulate the weather condition over the Southeastern Texas for a summer month of 2006. Spatial analysis of the sensitivity simulations reveals distinguishable effects of the input changes such as the distribution and magnitudes of daily temperature, the planetary boundary layer height and local flow patterns. Model-derived fluxes are compared with the measurements at the Houston Coastal Center (HCC) site which is operated by the University of Houston and located about 15 miles away from the Gulf of Mexico. Through these analyses, we intend to investigate LSM responses to changes in land surface inputs, as well as the appropriate values of the parameters corresponding to the residential and residential tree categories.