The aim of this study was selecting the most suitable interpolation method to estimate the spatial variation of maximum daily rainfall at the Atashgah basin, Ardabil. For this purpose, data of 9 rainfall gauge stations of Ardabil Province Water Regional Company were used. Homogeneity and independence of data were tested. Universal and ordinary Kriging and Cokriging interpolation methods were evaluated by circular, spherical, exponential and Gaussian regression models and Inverse Distance Weighting (IDW), Radial Basis Function (RBF), General and local estimators using Arc GIS software. To compare and evaluate the mentioned methods, statistical parameters of Mean, RMS, ASE, MS and RMSS were used. In each method, the lowest rank was devoted to the lowest statistical error and the highest rank was devoted to the highest statistical error, and then sum of the ranks was used to compare the interpolation methods. Results showed that, in the case of 2 years return period, ordinary Cokriging by circular regression model and for 5, 10, 25, 50 and 100 years return periods, ordinary Cokriging by Gaussian regression model, with the lowest ranks, were the most suitable interpolation methods in estimation of spatial variation of maximum daily rainfall. The weighted average of the maximum daily rainfall of Atashgah basin for the return periods of 2, 5, 10, 25, 50 and 100 years were determined equal to 37. 5, 48. 19, 49. 41, 53. 32, 54. 14 and 57. 69 mm, respectively.