Multi-criteria evaluation for parameter uncertainty assessment and ensemble runoff forecasting in a snow-dominated basin

Abstract

Abstract The increase in global temperatures undesirably affects the ever-growing world population and reveals the significance of hydrology science. Hydrological models might estimate spatial and temporal variability in hydrological components at the basin scale, which is critical for efficient water resource management. Satellite data sets with enhanced snow mapping with high spatial and temporal resolutions have been developed. The potential of these satellite data sets is evaluated in this study for multi-criteria evaluation of a conceptual hydrological model to improve model performance and reduce uncertainty. The upstream part of the transboundary Coruh River is selected for this study because snowmelt contributes a significant portion of the streamflow feeding major reservoirs during the spring and early summer months. The region’s snow cover dynamic has been analyzed using a combination of two satellite products. Hydrologic modeling is performed using the HBV model for the 2003–2015 water years (01 Oct–30 Sep). The Monte Carlo method is used for multi-criteria optimization exploiting satellite snow cover data besides runoff data. The sensitivity and uncertainty analysis on the model parameters indicate that multi-criteria calibration effectively reduces the uncertainty of the parameters and increases the model performance. Moreover, ensemble runoff forecasts are generated with several best model parameters using 1-day and 2-day lead time numerical weather prediction data for the snowmelt period (March–June) of the 2015 water year. The results indicate that the use of multiple remote sensing products in combination better represents the snow-covered area for the region. Additionally, including these data sets into hydrological models enhances the representation of hydrological components while reducing runoff prediction uncertainty.