كليدواژه :
Rainfall-runoff model , Regression models , Kasilian , مدل سازي بارش - رواناب , Iran , حوزه آبخيز كسيليان , تجزيه و تحليل آماري , Factor Analysis , كشاورزي
چكيده لاتين :
Determination of hydrologic parameters is important in view point of safety, project economy and efficiency of hydraulic structures. The runoff is one of the important parameters whose measurement has different limitations and therefore the application of hydrologic models in estimation of flood hydrograph is necessary. In this research an attempt has been made to compare the applicability of different statistical analysis methods to select the best one in developing rainfall-runoff model. To achieve the study goals, 15, 11 and 12 characteristics of hyetograph, hydrograph and their time spacing were considered, respectively, for 49 storms in Kasilian watershed. The study area located in the north of Iran and comprises 66.75 Km2. The relationships were investigated using bivariate in different forms viz. Linear, logarithmic, power, exponential, cubic, quadratic, compound, inverse, grow curve and S models as well as multivariate in different forms viz. stepwise, forward and backward after conducting Factor Analysis approach. The results of rainfall-runoff modeling process showed that the bivariate regression were mostly developed in the form of cubic, linear, compound and power models. Furthermore, consideration of results of multivariate regression verified the low accuracy of factor analysis method. Among other three methods, the stepwise and backward methods have had better performance, alternatively. The results of forward method were also similar to stepwise method.
