Abstract: The midstream oasis of the Heihe River basin is located in an arid region of severe water shortage in northwest China, where groundwater is crucial to food security and ecology health. Understanding of the spatio-temporal variations of local groundwater level in this area can provide useful information for the sustainable development and utilization of groundwater. This study has focused on an application of linear regression and geo-statistics methods to analysis of the spatial distribution of groundwater level. We compared three methods for spatial interpolation of groundwater level: the ordinary Kriging method, the Yamamoto correction method of Kriging smoothing effect, and the universal Kriging method. Results show a linear correlation of the groundwater level with space coordinates for the period of 1985-2010 when the vegetation cover was relatively stable. In each month, the groundwater level of this oasis area also followed a significant spatial correlation. The Yamamoto correction method can minimize the smoothing effects in Kriging method, although it has an overall accuracy slightly lower than those of the other two methods.