ABSTRACT
Hydrological models are considered as an important tool for planning and management of the complex hydrologic system. Despite the improvement in actual level of knowledge of the physical processes of the hydrological system and the computational advances, the models, still can provide an approximation of reality. In the field of hydrology, ensemble modeling has been applied only over last decade through cooperative initiatives. Multi-model ensemble, outperforms both single models and single-model ensemble. The effect of combining models to reduce errors may be expressed in terms of bias and variance. In ensemble modeling, imperfect predictions are combined which creates a way of managing individual limitations. The model utilizes time series of rainfall and pan evaporation data to simulate stream flow at the catchment outlet. Linear programming technique was used to develop 189 possible ensemble combinations using results of eight hydrological models.
