The Long and Short of Modeling Water Resources Management: A new module for Earth system models fills the gap across scales
Earth system models track all forms of water and its movement through atmosphere, land and ocean systems. However, the significant effect of reservoirs on the amount of freshwater evaporated into the atmosphere and flowing to the oceans has been neglected. Existing water management models take either an expansive global view that can miss important details or narrow in on specific water reservoirs and lose critical trends across regions and time periods. Filling the gap between the two approaches, scientists at Pacific Northwest National Laboratory designed a water management model that can provide input to an Earth system model by drawing its water data at the regional level.
The study team chose Washington State’s Columbia River Regulation System as an example because of the area’s many dams and reservoirs for generating electricity, irrigation, managing water flow, and serving as municipal water sources. The scientists first developed their new model apart from any Earth system model to ensure it was accurate on its own. It combines generic release and storage targets adjusted for each reservoir. They then coupled it to a river transport model and added irrigation demand and other types of water use measured by the U.S. Geological Survey. The team validated the model’s realistic representation of reservoir operations by showing that measurements taken in the field closely matched the predictions made by their model.
Their results showed promise for accurately simulating the regulated flow of water and withdrawal activities at the regional scale. The resulting model can be used to determine the impacts of water use and water management on regional and global climate and improve understanding of the mutual constraints between energy and water production and use.
It's much more than a drop in the bucket. Globally, as much as 15 percent of freshwater is withdrawn for drinking, growing food, producing power and other human activities. Failing to accurately depict these impacts on the water cycle affects the results from Earth system models, which are used to predict climate change and inform resource management and policy decisions. The team’s approach simulates reservoir operations such as flood control, irrigation and water supply using release and storage information for each individual reservoir with a focus on long-term water flow and demand.
“Water integrates many processes in both the natural and human components of the Earth system,” said Dr. Nathalie Voisin, the PNNL hydrologist who led the team. “Earth system models must accurately represent all branches of the hydrologic cycle—atmosphere, land, ocean, and human systems—including water and energy management, and socioeconomics.”
Dams and reservoirs regulate streamflow and water storage to meet water demand. By redistributing water resources in space and time, impoundments have important effects on the hydrologic cycle at regional and global scales. Representing these processes is essential for hydrologic predictions, earth system modeling, and assessing the impacts of climate change on water resources. Traditionally, impoundment is represented at regional and local scales using agency-specific reservoir operating rules and constraints with optimization that requires numerous time-forward simulations. Emerging large-scale research reservoir models use generic operating rules that are more flexible for coupling with Earth system models. DOE scientists at Pacific Northwest National Laboratory adopted the latter approach and developed a water management model coupled with a river transport model to represent the effects of reservoir operations in an Earth system model. They investigated the uncertainties of the reservoir model from different implementations of the generic operating rules to understand their effects not only on regulated flow but also on reservoir storage and the fraction of demand that is met. The model has been evaluated over the complex multi-objective Columbia River Regulation System in Northwestern United States as an example. Although the management model was able to reproduce the regulated flow at regional scale, realistic simulation of subregional reservoir storage requires development of new generic operating rules that combine both storage and release targets for multi-purpose reservoirs. The new model produces a more realistic representation of the reservoir operations and flow storage distribution across the region. Future implementation of the water management model in an Earth system model will provide a meaningful translation from global water cycle changes to local water resources impacts and vice versa.
This study was part of the Integrated Earth System Modeling (iESM) a collaborative project supported by U.S. Department of Energy Office of Science Earth System Model Development program. The Platform for Regional Integrated Modeling and Analysis (PRIMA) initiative, funded by PNNL’s Laboratory Directed Research and Development program, supported the development of databases used in reservoir modeling and some analyses of the results.