Evaluation of Rainfall Patterns Using Modern Spatial Statistical Methods in the Khorasan Razavi Province

Rainfall is the most variable climatic parameter. These changes occur both in space and time scale which identify the climate of the region. The aim of this study is to evaluate the spatial autocorrelation of precipitation in the Khorasan Razavi province. For this purpose، precipitation data were obtained from12 synoptic stations in Khorasan Razavi. To acquire the variations in seasonal precipitationin Khorasan Razavi، modern statistical methods were applied such as Spatial Autocorrelation، Global Moran''s Index، Local Moran''s I index، Hotspots and cluster analysis (Clusters/Outliers). These methods were applied using Grads، MATLAB and ArcGIS Software Packages. The results showed that rainfall in the Khorasan Razavi province has a high clustering pattern. According to the local Moran index and Hot Spots، rainfall has positive spatial autocorrelation patterns (clusters of high rainfall value) in the central، northwest and northern parts. This pattern of rainfall، at confidence levels of 90%، 95% and 99%، covers totally 25% of the province areas. The southern regions have negative spatial autocorrelation (clusters of low rainfall value) which totally cover an area equal to18 percent of the province. In the other areas of the province which cover 57 percent of the entire area،no significant pattern has been observed. This indicates heterogeneity of rainfall over the province. The present study shows that altitude which is one of the local factors is the most important factor which influences the distribution of rainfall pattern over the province.