Surface modelling of annual precipitation in the DongJiang river basin, China

Precipitation is the major climatic factors. However, what the limited meteorological station can provide are discrete observational data. At present, in the field of GIS, the methods which are widely used in the precipitation spatial interpolation include IDW, TIN, Kriging, Spline and so on. Based on the precipitation data from 65 sampling points in the DongJiang river basin by the China Meteorological Administration, this paper integrates the advantages of the spatial autocorrelation model and spatial differentiation model, with the advantages of the spatial autocorrelation model, classical interpolation model (IDW, Kriging, spline) and HASM are used to interpolate the annual precipitation. Interpolate result shows that HASM algorithm is much more accurate than classical interpolation algorithm. So the residual anomaly for local change is simulated by HASM algorithm. The results show that estimated annual precipitation correctly replicates real spatial distribution of precipitation qualitatively and quantitatively.