The reference crop evapotranspiration (ETo) mainly subjects to the influence of environmental conditions, such as water dispersion of regional evaporation surface and the dynamic effect of turbulent air above it. It is closely related to the me teorological and geographical conditions and possesses a feature of regional variabl e. According to the requirement of study on regional spatial variability and opti mum isograms of water requirement under complicated climate and geographic envir onmental conditions in Inner Mongolia Autonomous Region, this paper presents an approach to the application of the Kriging method. Geostatistics, an optimum and unbi asedestimation method, was applied to estimation of regional information under irregular sampling conditions in a large area. Based on the 30 years clima tological data of 135 meteorological stations in an area of 1.18 million km2 , the irregular sampling data were regularized. After that, the appropriate extension was combined with the classical estimation method of optimum interpolation under cont rolling of estimation error. Both the estimation of regional information and the est imation error σ2k are given in this paper. This will provide an e ffective method for the drawing of isograms of irregular sampling grids in a large region