A multi-objective optimization-based framework for extending reservoir service life in a changing world

Abstract

Reservoirs are an essential component of water resources systems, improving water supply reliability and security. With long-term climate and water demand changes, the performance of reservoirs may deteriorate, necessitating interventions such as infrastructure upgrades and/or water demand management. These interventions often require substantial financial investment, have long implementation time or impede economic development. Therefore, a more cost-effective and less disruptive alternative is needed to extend the service life of existing reservoir systems. Adapting reservoir operation policies to cater for future changes is such an alternative. By adapting reservoir operation policies to future water availability and demand changes, the capacity of existing systems can be fully utilized to avoid or postpone more expensive and disruptive interventions. However, there is a lack of an integrated framework for estimating the potential of service life extension by re-optimizing operation policies to cater to uncertain changes in future conditions. In this study, we develop a multi-objective optimization-based framework to allow operation policies to adapt to uncertain future changes in water availability and demand by optimizing operation policies for a range of plausible future conditions, thus enabling the service life of existing reservoir systems to be maximized albeit uncertainty in future changes. The developed framework is demonstrated using a reservoir system in Northern Australia. Results show that the service life of the reservoir can be extended by up to 40 years considering uncertain future water availability and demand. In addition, significant changes in operation policies and system responses with time are identified, providing valuable information for developing long-term reservoir planning strategies.