Abstract
Maintaining a supply of freshwater is not only necessary for basic human survival, but water supports a variety of anthropogenic activities like industry, agriculture and recreation. Traditional management of water is done by controlling the storage volume inside a reservoir requiring a robust and adaptive operational policy that balances the hydroclimatic variability of the environment, defining how much water is available at one time, with human demands across a range of time scales. The technical research presents a method of multireservoir operation optimization, combining a multi-objective evolutionary algorithm (MOEA) with streamflow simulations as an adaptable operating policy. This work aims to demonstrate how a reservoir can be controlled to meet competing objectives under a variety of potential future demand scenarios. Tightly coupled to this idea is the science, technology and society (STS) research, which uses Actor-Network Theory as a framework to analyze the impact that social factors have on this system, and establish a methodology to further optimize the system. Climate change is largely human driven, so together the technical and STS work outline an adaptive and robust approach to better serve users and increase reservoir efficiency.
Water withdrawals from a reservoir are considered pareto-optimal, increasing the benefits from one activity decreases the performance ability of another. Using the Columbia River Basin in the U.S. as the study region the technical report outlines the development of a MOEA simulation, optimizing environmental flows, economic benefits, flood protection and power generation. Physical parameters, like streamflow, temperature and wind speed were incorporated into the California and West Coast Power Systems (CAPOW) model while abstract social needs were quantified as changes to daily energy demand or meeting the threshold value of an optimization goal, as determined through literature review. The model was run under nine different future scenarios defined by combining one of three climate pathway projections determining physical parameters with one of three socioeconomic pathways determining social trends and energy demand, calculated from the Coupled Model Intercomparison Project (CMIP5).
Through CAPOW the decade of 2050-2059 was simulated, detailing both the daily price per MWh, and the amount of MWh generated across different energy sources like fossil fuels and hydropower. Relative changes in both price and energy source between the three climate scenarios were minimal, while social scenario exhibited strong variations reinforcing that human activities have enormous influence on basin behavior. Withdrawals are expected to increase regardless of scenarios, with the greatest increase occurring under continued fossil fuel development. A streamflow simulation of the basin demonstrated the viability of these potential operating policies comparing total volume against a historical baseline, with no policy exceeding the storage volume.
While physical parameters were fairly easy to quantify as this particular basin is well studied, abstract social needs are more difficult to define and subject to change with changing socioeconomic contexts. The reservoir system is bound by a complex network of actors spanning several states, companies, animal and plant populations, governments and individuals. An optimal policy for the whole basin may not be optimal for users at the local level, so the STS research provides a socially guided optimization pathway.
An Actor-Network analysis through found that governments and power companies have the greatest direct influence on basin operations. Individuals can directly influence these actors by altering their own behavior, as governments are obligated to meet demands like drinking water and flood protection while companies want to meet energy demand to continue profiting. More indirectly is the promotion of sustainable policies and investment in renewable energies at the local, state and federal levels which individuals can advocate or vote for that additionally alter the climatic context of the basin. Done repeatedly this process mirrors a MOEA where the social, environmental and economic context informs an operating policy which creates a new context that can be used to optimize the policy further. Regional authorities like Bonneville Power Administration have utilized similar analysis to expand their local wind generation capacity, and as climate and markets change, this adaptive framework highlights where people can direct that change to better meet their own needs.
Reservoirs are large-scale, capital intensive projects with long life spans. There is a clear and present need for this infrastructure, much of which was built using simplified or incomplete information, to be made more efficient. While its difficult for individuals to alter the basin directly, the social dependence of climate change means that they can alter the context around themselves to better meet their own demands.
