Comparison of Kriging spatial interpolation methods of non-stationary regionalized variables

To reduce the workload of the spatial interpolation of nonstationary regionalized variables with the Kriging method,the ordinary Kriging and universal Kriging methods were used for spatial interpolation of data of average annual precipitation for 59 stations in the middle and east regions of China from 1960 to 2000.The results of interpolation of the two methods are close to each other,while the precision of the former is a little higher.As a conclusion,the ordinary Kriging method can be used for replacing the universal Kriging method for spatial interpolation of nonstationary variables if it is unnecessary to investigate the spatial variability of variables.Without need for determination of drifts,the ordinary Kriging method is simple in calculation,and the interpolation result is satisfactory.