عنوان مقاله [English]
نویسندگان [English]چکیده [English]
Background and objectives: Since the problem of predicting and runoff estimating play a key role in integrated water resources management, therefore hydrological modeling especially continuous rainfall-runoff modeling may be most important part of water resource planning which is released from reservoir dams. Thus continuous daily hydrological models are useful tools for estimating runoff from rainfall. These models are able to estimate the runoff in ungagged basin. The purpose of this paper is to provide a continuous simulation model for Hydrologic forecasting so that investigate dominancy or dormancy of the processes.
Materials and methods: In this study rainfall-runoff processes involved in modified SCS-CN model calibrated in Leaf River Watershed located in US and Qarasou subbasin located in west of Iran through PSO optimization algorithm developed in MATLAB programming language with 9000 simulation numbers. Nash-Sutcliffe Efficiency (NSE) is used as objective function and the decision variables (14 parameters) within the specified range are randomly initialized. Optimum parameters were extracted through PSO. This model is calibrated and validated with two periods 1957-1961 and 1953 for Leaf River Watershed and two periods 1381-1384 and 1387 for Qarasou subbasin respectively.
Results: Model parameters were calibrated and Validation for two case studies. Comparison of the observed and simulated runoff carried out based on three performance criteria: Nash-Sutcliffe (NSE) and Kling-Gupta Efficiency (KGE) and Root Mean Square Error (RMSE). Proposed model performed these three statistics respectively for leaf River Watershed 0.81,0.87,1.40 as calibration period and 0.83, 0.86, 2.53 as validation period. Reasonable values for these criteria is also attained in Qarasou subbasin but due to more reliable data, better results is expected in Leaf River watershed. A result comparison of the SCS-CN model with Hymod as a simple conceptual model, both with the same inputs revealed latter model can simulate hydrology behavior of Leaf River Watershed and Qareso River Watershed slightly better. This may be originated due to fewer model complexities and thus less parameter uncertainty of Hydmod. In spite of this superior skill in runoff simulation of Hymod, special capabilities of modified SCS-CN model which calculate hydrological components (baseflow, percolation, throughflow, surface runoff and initial abstraction) may prove usefulness and efficiency of this new model easily.
Conclusion: modified SCS-CN model as a conceptual model calculates daily runoff using rainfall and potential evapotranspiration dataset. This model may be used to assess annual hydrologic components as well as total runoff values. Based on the results, the dominancy of the infiltration, evaporation and surface runoff processes were approved in Leaf River Watershed. These three processes but in reverse order is ranked in Qarasou subbasin as main hydrological components.