Researchers at the University of Manchester have developed a systems modelling approach which determines how much water managers should store across complex constellations of dams. The approach could save California alone over $100 million annually.
FutureDAMS Research Director Julien Harou, Chair of Water Engineering, The University of Manchester is one of the authors of a new open access paper that proposes a better way to manage water storage in reservoir systems. ‘Estimating the Economic Value of Interannual Reservoir Storage in Water Resource Systems’, in Water Resources Research argues:
Reservoir operators face pressures on timing releases of water. Releasing too much water immediately can threaten future supplies and costs, but not releasing enough creates immediate economic hardship downstream.
This paper examines how the economic valuation of end‐of‐year carryover storage can lead to optimal amounts of carryover storage in complex large water resource systems.
The economic valuation of storage helps inform water storage decisions.
The approach was applied to California’s Central Valley water resource system, including 30 reservoirs, 22 aquifers, and 51 water demand sites which creates over $40 Billion of revenue annually in agricultural production alone. In some reservoirs, average stored water values exceeded $0.05 per cubic metre.
It is estimated that improving water management to consider the economic value of stored water could decrease California’s water scarcity costs by over $100 million annually.
The authors state that, “To our knowledge, this is the first instance of coupling an evolutionary algorithm with a hydroeconomic model, and the first approach to economically value storage in large‐scale systems where the associated optimization problem is nonconvex.”
Professor Jay Lund, Director of the Center for Watershed Sciences at the University of California described the article as “Geeky, but fun. Feast on insights and better questions arising from Table 1!”.
Read the full, open access version here.