Representing reservoir storages in the earth system models using storage-reliability-yield relationships

Earth System Models provide a unified framework for understanding and quantifying the coupled interactions between ocean, atmosphere and land surface. But, continued alteration of water cycle by humans, particularly with man-made storage structures, necessitates the representation of reservoir storages in ESMs. The primary challenge in quantifying the reservoir storages in the Land Surface Models (LSMs) arise from the complex operational characteristics of multi-reservoir system. Previous studies in reservoir analyses have used storage-reliability-yield (SRY) relationships for generalizing the reservoir characteristics. However, most of these relationships are typically available at annual time scale. In this study, we extend the SRY relationships to daily time scale for characterizing the multipurpose characteristics of the reservoir systems. Towards this, we consider two river basins, one from the Southeast and another from the Southwest, having contrasting climate (humid vs arid) and runoff (snowmelt vs rainfall-runoff) regimes. SRY relationships are incorporated into a watershed model for estimating the streamflow at gauges with controlled flows. The performance of the calibrated watershed model will be validated in simulating the streamflow for the selected stations. SRY relationships will also be used for evaluating the flow alteration induced by the reservoirs and for quantifying the drought impacts. Comparison of the watershed model with and without SRY relationships will give information on the utility of incorporating storage explicitly in the watershed model. Potential for extending SRY relationships in LSMs also will be discussed.