Abstract:
Prediction of annual average precipitation over a regin is one of important issues in watersherd management andnatural resources planning. Geostatistics is one of the useful methods for precipitation prediction which varies based on type of the variable. This study was conducted in Karoon and Dez Basins located in southwertern of Iran. The data set including 20 years of annual precipitation of stations were used. At first normality and homogeneity of data are
examined. Then variographic analysis using three techniques including kriging, co-krigining and Weighted Moving Average (WMA) were applied for predicting annual precipitation. Finally compaction of the results using statistical techniques showed that kinging method has the highest accuracy and provides more accentuate results.
Machine summary:
"(2004) studied the capability of geostatistical techniques for estimating spaial distribution of annual rainfall in arid and semi-arid regions of south eastern Iran which showed TPSS method using elevation as a variable is the most relevant method of rainfall estimation while kinging and WMA methods having second power placed in next orders.
(2001) estimated regional average precipitation in southern Spain using Theissen and geostatistical methods and concluded that the results obtained by kriging has the highest correlation with the data.
Prodhorome and Reed (1999) used geostatistical techniques for estimating extreme rainfall in mountainous regions of Scotland that showed kriging and modified residual kriging due to simplicity are more applicable methods.
Diodato and Ceuarelli (2005) were studied spatial variation of monthly and annual precipitation in southern Italy which showed that weighted moving average has the highest error while linear regression and common co- kriging methods produce the lowest errors."