Runoff Prediction in Ungauged Watersheds Using Regionalization of HBV Model Parameters

Daily streamflow information has a major role for solving various hydrologic problems, but the continuous information of daily streamflow is not available in most cases. One of the approaches to estimate daily streamflow at an ungauged location is to transfer rainfall–runoff model parameters calibrated at a gauged (donor) watershed to an ungauged (receiver) watershed of interest. This method is based on selecting a similar gauged watershed to ungauged watershed. The study examines the possibility of simulating time series of daily streamflow for ungauged watersheds based on hydrological similarity. The data of 14 gauged watersheds (with drainage area between 16 to 804 square Kilometers) with 11-year time series of daily streamflow data located in central Alborz region, Iran, are used for this research. The HBV model is applied to simulate daily streamflow with parameters transferred from gauged watersheds counterparts. This model has been used widely in previous similar studies and was selected for this research due to its simplicity, input data availability and adaptability to different climates. Hydrological similarity is defined based on three regionalization methods including drainage area, spatial proximity and physical similarity. The obtained results showed that transferring HBV model parameters based on spatial proximity method produces better runoff simulation compared to the other two methods and physical similarity is preferred to drainage area. Also, the results indicate that the HBV model has a high ability for regionalization of daily streamflow in the study area and the uncertainty related to model parameters could be neglected by using the best calibrated parameters set of model. According to the aforementioned brilliant results of applying HBV model predictions in this study, we suggest this model in future similar study.