Analysis of Spatial Variation of Precipitation : Comparison of Conventional and Kriging Methods

Analysis of Spatial Variation of Precipitation: Comparison of Conventional and Kriging Methods

One of the major challenges in water resources management is estimation of missing or un-gauged data. Estimates of the unknown data in different points can be made by spatial interpolation of the observed data recorded in some special points of the study region. In case of spatial variations of precipitation data, there are many interpolation techniques changing from simple linear methods to complicated multivariate methods. Each method has its own necessities and constraints which result in different levels of accuracy and precision in the estimated values. Accordingly regarding the needed accuracy in different applications and the available time and data, different spatial analysis methods are utilized. In this paper different conventional and modern methods such as interpolation and Kriging are used for spatial analysis of the precipitation in the western part of Iran. Many different techniques have been applied to measure annual and monthly precipitation data at 38 stations for the period of 1967 to 2005 in the study area. The results are compared by estimation of the data for some stations in the region. The results showed that different methods of interpolation should be used with care considering some local characteristics that affect the climatic variables. In this case study, universal Kriging had the best performance among different interpolation techniques. Based on the best results of conventional and Kriging methods in this study, the locations of some new stations are proposed.