recognition and regionalization of daily precipitation frequency distribution in iran

in this research, the daily precipitation data of 1437 synoptic, climatic and rain gauging stations of iran in the time interval 1961- 2004 (equal to 15705 days) was utilized. using kriging geostatistics method, the data were interpolated on the pixels of 15 *15km2, so as an array to the dimension of 15705 * 7187 was formed. to identify the fittest distribution of the days with over 5.0mm precipitation, in each pixel kolmogorov-smirnov fitness test was performed. from among 7187 pixels with a surface area of 15 * 15km which covers the entire surface area of iran, the precipitation empirical function of rainy days in 6861 pixels equivalent to 5.95% of the country's surface area follows the theoretical two-parameter gamma distribution, and the empirical function of 326 pixels equivalent to 5.4% of the country's surface area which is concentrated on south-east and zagros heights follows the exponential function. parametric mapping of mean, standard deviation, skewness and kurtosis of the fittest distribution (probability model) indicates that mean long-term daily rainfall below 10 mm is a confirmation to dry climate of 95% of the country's surface area. fluctuation and amount of mean precipitation, and its standard deviation, skewness and kurtosis in central and eastern iran is minimum relative to other regions of the country, whereas in other regions of iran such uniformity is not observed in the parameters. in major part of zagros and southern coast of iran the expected precipitation is 5 to 10 mm, but mean standard deviation of precipitation in southern coasts has increased so as the rain has become less reliable. at the eastern feet of zagros and from beginning of zagros heights towards inner part of iran, the expected precipitation quantity is greater than that over large part of zagros. however, the precipitations in these regions with many slight precipitations are trusted the least.