بررسي دقت و واسنجي روش هاي رگرسيوني چاو و استوكاستيك (بررسي موردي: منطقه البرز مركزي)

Accuracy Assessment and Calibration of Chowʹs Regression and Stochastic Methods (Case Study: Central Elbuirz Region)

Peak flow estimation is very important and challenging issue in hydrologic and socio- planning. Flood distribution methods are reliable methods for estimating hydrological events, but their application requires computer programs. Two well known equations for estimating annual peak flood are Chowʹs regression method and stochastic method. These methods are simplified distribution methods which have not already used in Iran. In this research assessment of accuracy of these methods on basis of comparing their estimation with the best fitted distributions results was considered. Then for optimization of these methods, we defined the best coefficients in 20, 25, 30, 35 and 40 years return period using the Solver tools. In this program a cell determines a target that would has the highest value, and the coefficients of methods were placed in the changeable cells. In this research the Mean Absolute Error (MAE) was the target cell and determination coefficients of the methods were calculated on basis of its reduction. The results showed that the methods don not have significant difference with the best fitted distribution and the determined coefficients have no affect in reduction of MAE in return periods higher than 100 year.