Atmospheric Sciences & Global Change
Gauging Water's Future
Earth system model tested to estimate watershed runoff
The American River in eastern Washington is one of the stream gage locations where data was collected. This and other data were used to improve runoff model predictions. Photo courtesy of Walter Siegmund. Enlarge Image
Results: Scientists at Pacific Northwest National Laboratory, the Chinese Academy of Sciences and Oak Ridge National Laboratory found ways to improve the capabilities of a land model within global and regional Earth system models to estimate water runoff. Accurate runoff predictions can lead to better information for managing sustainable water for personal and industrial consumption. As it was tested, the original model produced runoff variations that are not realistic when compared to observations. The team identified several methods to improve the simulations, mainly by improving how the below-surface runoff is estimated.
Why It Matters: Population growth, urban expansion, and increased development and industrialization will put pressure on planners and resource managers to meet the growing demand for water. As precipitation becomes more uneven due to climate change, this demand will be one of the largest challenges facing a thirsty world. Scientists are finding ways to simulate how fresh water moves through the complete water cycle and predict how that cycle will be affected by climate changes. Researchers in this study looked at water runoff, which is defined as the amount of precipitation falling on the land that ends up in a stream. They are working to predict when, where, and how much water will be available in streams or reservoirs. These water sources are vital to fulfill the growing demand for water.
Methods: The Community Land Model has been used to simulate land surface responses to the atmosphere including water and energy balances at the global scale. The team of researchers sought to test the model's ability to simulate runoff and streamflow at local and regional scales.
They compared model simulations of runoff as well as surface energy flux at various locations using streamflow gage measurements from the U.S. Geological Survey, and from various flux towers installed in North America. The flux towers, large vertical structures, are used to gather data on the exchanges of water and energy between the land and the atmosphere.
The team employed three methods to produce simulations more closely matching observational data. They calibrated the model, adjusting parameter values for soil properties and topographic conditions to optimize the model output with the data. They also implemented different parameterizations—alternative ways to represent complex systems within the model. Finally, they increased the resolution of the model to improve simulation.
Of these three methods, the team demonstrated that a calibrated parameter set produced a closer representation of actual site conditions. In addition, they showed it is important to represent spatial heterogeneity in land cover, vegetation, soil, and topography for better simulation of streamflow. Their research demonstrates the important constraint of soil hydrology on the surface energy budget and highlights the need to improve runoff parameterization in land and surface models.
The study was published in the Journal of Geophysical Research-Atmospheres.
What's Next? The team is working on the North American and global domains to gain insights in model performance under different climate and hydrologic regimes. They will use alternative parameterizations, as well as inverse modeling and uncertainty quantification to continue their assessment of Community Land Model's capability for simulating regional and global hydrologic cycles.
Acknowledgments: This research was supported by the U.S. Department of Energy's Investigation of the Magnitudes and Probabilities of Abrupt Climate Transitions (IMPACTS), and Improving the Representations of Human-Earth Interactions (Strengthening the Coupling between Climate and Earth System Models and Integrated Assessment Models). The research was performed by Drs. Hong-Yi Li, Maoyi Huang, Mark S. Wigmosta, Yinghai Ke, André M. Coleman and L. Ruby Leung of PNNL, Dr. Aihui Wang of the Chinese Academy of Sciences in Beijing, and Dr. Daniel M. Ricciuto of Oak Ridge National Laboratory.
Reference: Li H, M Huang, MS Wigmosta, Y Ke, AM Coleman, LR Leung, A Wang, and DM Ricciuto. 2011. "Evaluating Runoff Simulations from the Community Land Model 4.0 using Observations from Flux Towers and a Mountainous Watershed." Journal of Geophysical Research - Atmospheres 116:D24120. DOI:10.1029/2011JD016276.