A Comparative Performance Evaluation of Interpolation Methods for Estimating Annual and Monthly Rainfall in the Namak Lake Watershed, Iran

One of the important problems of water resources studies is estimating the spatial distribution of rainfall based on point rainfall observations. In the present study, interpolation methods including definitive mathematical methods including Inverse Distance Weighting(IDW)  and Radial Basis Functions (RBF), Global polynomials, and local polynomials, and Kriging and Cokriging geostatistical methods using Geographic Information System (GIS) to estimate monthly and annual rainfall in the Namak lake watershed were evaluated. Spatial distribution analysis was carried out in the period 1988-1999 to 2006-2007 water year. In this study, in order to compare interpolation methods in iso-rainfall of the Namak lake watershed, 110 meteorological stations into the mentioned watershed with the statistical period of 30 years were used. The results showed that overall in annual rainfall according to Root Mean Square (RMSE) index, Cokriging and Kriging geostatistical methods in rank one and two, respectively have the best rainfall estimation. Therefore, Cokriging method with RMSE of 63.68mm had the best estimation. RBF, Cokriging and Kriging methods had also in the best estimation for   monthly rainfall data. In general, geostatistical methods have been overestimated in monthly and annual rainfall estimation in the Namak lake watershed.