Analysis of Spatial Veriation of Precipitation: A Case Study of Western Iran

One of the challenges in hydrologic and hydraulic design is to estimate missing or ungauged data specially when spatial distribution of data is important. Estimates can be made by spatial interpolation of data available at the other sites. In case of precipitation data there are several interpolation techniques for estimating the spatial variations of data such as simple regression, thiessen polygon, second order polynomial, kriging and cokriging . This paper presents the results of a study in the application of multivatiate geostatical algorithms in forms of kriging method with varying local means. And kriging with an external drift for estimating spatial variations of precipitation in the western part of Iran. In order to increase the accuracy of interpolation, Digital Elvation Model (DEM) data and precipitation gradienr is also considered. The techniques are applied using annual and monthly precipitation data measured at 54 climatic stations in the study area. Cross validation is used to compare the accuracy of the techniques. The results have shown that the kriging estimation variance depends on the framework of the gauging network and it can be used for the optimal design of the rain gauge network. Based on the results of this study, new locations for establishing rain gauges and also recommendation for reducing the number of rain gauges in the region could be mode.

کلیدواژه ها:

precipitation ، kriging ، Spatial variation ، Rain gauge network ، Cross validation