Comparison of Pilgrim and Huff methods for derivation of time distribution of rainfall in Zabol station (Iran)

Selecting the designing rainfall is a basic step in many commonly used methods for estimation of design floods. Design rain that results in design flood has some characteristics such as time distribution of rainfall in rain period. In this research for derivation of time distribution of rainfall, the data of rain gauge station of Zabol station has been used. All severe rainstorms in each station is located in 1, 2, 3, 6, 9, 12, 14 hours duration. The Pilgrim method was used based on storms separation into four quarters and estimation of rainfall percentage for each quartile and ranking. By averaging ranks in each quartile, specifying index rank the final pattern of each time duration is obtained. To statistically evaluate the obtained patterns with durations of interest, Chi square test was used after establishing a contingency table. Based on Huff method, in every time duration based on this fact that in which quarter the maximum rain has happened, the first, second, third and fourth rainstorms were defined and the related charts with 10 to 90% probability were prepared. The result of two method comparisons shows that: The Pilgrim method is an easier and more expressive method than the Huff method and if suitable data are available, regardless of simplicity it is an accurate method. Comparison of Pilgrim temporal rainfall pattern with Huff, and fifty percent rainfall distribution probability curve, shows that the result of these two methods in every time distribution are the same pattern. The best design conditions of Huff probability curves in each quartile is to consider the first, second, third and forth quartile. In this condition determined curves will be more clear and accurate.