Evaluation of WMS model in basins without statistical data in southwestern Iran using Dicken's experimental method (case study: Kuhgel Basin of Khuzestan Province)

Abstract

Floods are important in studies related to water resources utilization, dam construction, basin management and hydrological studies. Therefore, to a large extent, the accuracy of these studies and the safety of water facilities depends on the flood study methods. The aim of this study was to investigate the WMS model in determining the maximum flood discharge in three sub-basins with areas of 10.3, 1.5 and 7.3 square kilometers in the Kuhgel region located in southwest of Iran. Comparing the results of the model with the results of the Dicken's experimental method in the region showed that the maximum flood discharge and flood volume estimated by the WMS model are well consistent with the values of the Dicken's experimental method. Therefore, the relative error percentage index, root-mean-square error and Nash–Sutcliffe coefficient of the model were calculated equal to 6.12, 0.51 and 0.82, respectively. The results also revealed that the mean of model estimates was approximately 0.16% higher than Dicken's experimental method. Based on Student's t-test, the difference between the model values and Dicken's experimental method was not significant at the level of one percent. The results show that the WMS model simulates flood values with appropriate accuracy for basins with no statistical data in southwest of Iran. Due to the high sensitivity of the SCS method in the WMS model to rainfall distribution, rainfall analysis in the study area is necessary to achieve the desired results and the rainfall distribution and its time distribution should be as close as possible to the actual values in the region.