Sensitivity of reservoir storage and outflow to climate change in a water-limited river basin

Abstract:

During the past several decades, numerous reservoirs have been built across the world for a variety of purposes such as flood control, irrigation, municipal water supplies, and hydropower. Consequently, streamflow timing and magnitude are altered significantly by reservoir operations. In addition, the hydrological cycle can be modified substantially by a changing climate. Therefore, a distributed hydrological model which has an embedded reservoir component is essential for representing these effects in future water management planning strategies. In this study, a multi-purpose reservoir module was integrated into the Distributed Hydrology Soil Vegetation Model (DHSVM). The DHSVM model was selected because of its high spatial and temporal resolution and because of its explicit representation of the physical processes. Prescribed operating rules, which are designed to reduce flood risk and enhance water supply reliability, were adopted in this module. The integrated model was tested over a water-limited basin (i.e. the central Brazos River Basin, Texas). Both the calibration and validation results suggest that the model performed robustly at daily, weekly, and monthly levels. Subsequently, the effect of climate sensitivity on reservoir storage and outflow was assessed by perturbing precipitation within a range from -30% to 30% and temperature from -2 °C to 2 °C. Results suggest that both variables are more sensitive to precipitation than temperature. However, there are more uncertainties associated with future precipitation than temperature. It was also found that the sensitivities vary significantly by season. Enabled with the new reservoir component, the DHSVM model provides a platform for projecting future water availability estimations under flow regulation, climate change, and land cover/land use changes. We expect this integrated model to be beneficial for sustainable water resources management.