Skip to main content
U.S. flag

An official website of the United States government

A new global water management module: Distributed vs. Lumped representations

Presentation Date
Tuesday, December 14, 2021 at 4:00pm
Convention Center - Poster Hall, D-F

Surface water reservoirs provide a critical human response to natural variability and short-term environmental influences such as droughts. Reservoirs also have the potential to alter the long-term co-evolution of energy-water-land systems. Understanding the future role of reservoirs in shaping the co-evolution of energy-water-land systems at a global scale requires carefully representing reservoirs, and their multiple purposes—such as irrigation, hydropower, and flood control—in global hydrologic models. The resulting dynamics emerging from global hydrologic models can then be used to inform the dynamics in global integrated human-earth system models such as the Global Change Analysis Model (GCAM). As a first step toward this goal, this study aims to develop a water management module for Xanthos, an existing distributed global hydrologic model that is part of the Global Change Intersectoral Modeling System framework, by adding local surface water extraction and reservoir operation. Based on primary purposes (i.e., irrigation, hydropower, and flood-control), we generated unique operation rules for 3683 large reservoirs globally. We then evaluated the water management module's performance in reproducing observed streamflow variability under normal and water scarcity conditions across 94 large river basins. Overall, results showed satisfactory performance in capturing the inter-and intra-annual variability in the observed monthly streamflow. We conducted an initial exploration of parameter sensitivity at knowledge-based intervals and found that runoff parameters had a more significant effect than the water management parameters on the outputs. Further, we also develop a lumped version of the water management framework, which produces the streamflow simulations at the basin outlets very similar to those from the distributed version. The lumped approach reduced computational cost by ~99.9%, enabling rapid evaluation of reservoirs' implications under various scenarios at the regional or global scales. The presented framework will eventually allow for assessing future reservoir development and management from a coupled human-natural system perspective by including virtual reservoirs within GCAM itself.

Funding Program Area(s)