Most ecosystems and crops experience water stress in arid and semiarid areas of the Inner Mongolia grassland, Northern China. Yet the lack of long-term in situ monitoring data hinders the managerial capacity of changing water vapor environment, which is tied with sustaining the grassland in the Inner Mongolia. Environmental remote sensing monitoring and modeling may provide synergistic means of observing changes in thermodynamic balance during drought onset at the grassland surface, providing reliable projections accounting for variations and correlations of water vapor and heat fluxes. It is the aim of this paper to present a series of estimates of latent heat, sensible heat, and net radiation using an innovative first-principle, physics-based model (GEOMOD: GEO-model estimated the land surface heat with MODis data) with the aid of integrated satellite remote sensing and in situ eddy covariance data. Based on the energy balance principle and aerodynamics diffusion theory, the GEOMOD model is featured with MODIS (Moderate Resolution Imaging Spectroradiometer) data with 250 m spatial resolution to collectively reflect the spatial heterogeneity of surface properties, supplement missing data with the neighborhood values across both spatial and temporal domains, estimate the surface roughness height and zero-plane displacement with dynamic look-up table, and implement a fast iterative algorithm to calculate sensible heat. Its analytical framework is designed against overreliance on local micro-meteorological parameters. Practical implementation was assessed in the study area, the Xilin Gol River Basin, a typical grassland environment, Northern China. With 179 days of MODIS data in support of modeling, coincident ground-based observations between 2000 and 2006 were selected for model calibration. The findings indicate that GEOMOD performs reasonably well in modeling the land surface heat exchange process, as demonstrated by a case study of Inner Mongolia.