نوع مقاله : مقاله پژوهشی
نویسندگان
1 دانشکده کشاورزی، دانشگاه بوعلیسینا
2 دانشکده کشاورزی، دانشگاه تربیت مدرس
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Estimation of regional rainfall is one of important problems of water resources studies that can be accomplished by different methods. In this research, based on GIS capability and application of some geostatistical methods such as Kriging (simple, ordinary and universal) and radial basis functions (RBF), thespatial distribution of annual rainfall characteristics of Hamedan province were evaluated. In this regard, the observed data of 11 years in 36 meteorological stations and digital elevation model of the study area were used. To evaluate the methods, the cross validation technique was applied by error estimation. The results show that rainfall of the area varies between 206.7 and 494.8 mm. They also illustrate that the simple Kriging (with exponential model variogram) is the best method for average annual rainfall interpolation based on the mean bios error, mean absolute error and root mean squared error. They were 0.002, 16.53 and 21.04, respectively. Based on the estimated errors, in the case of maximum data, simple Kriging (with circular model) was identified as the reliable estimation method. The ordinary and universal Kriging methods (with Gaussian model) were also recognized as the most precise methods for the minimum rainfall data estimation. These results can be used to local estimation of rainfall characteristics of the region.
کلیدواژهها [English]
)