07 July 2015

Evaluating Global Streamflow Simulations by a Physically-based Routing Model Coupled with the Community Land Model


Understanding a river's flow is important, especially as water uses and sources change. As reported in the Journal of Hydrometeorology, a team led by scientists at Pacific Northwest National Laboratory coupled a newly developed river-routing model with an Earth system model, and the simulated streamflow compared favorably against the observed streamflow from more than 1,600 major river stations worldwide. They also found that the added complexity in the new model adopted in Earth system models improves the model's ability to capture the variability of the observed streamflow. This new feature allows it to represent human influence on the river systems. 


In this study, a physically based routing model, the MOdel for Scale Adaptive River Transport (MOSART), was coupled with the land component of Community Earth System Model (CESM) called Community Land Model. The gridded CLM-simulated surface runoff and base flow were provided to MOSART at the end of each time step, and MOSART routed the runoff across hillslope and through tributaries and main channels of the river network. One distinct feature of MOSART compared to the previous river transport model in CLM is that MOSART explicitly simulates through both space and time the variability of flow velocity. The team showed that simulating the spatial and temporal variability of river velocity is necessary for capturing the seasonal nature of streamflow and annual maximum floods.  

The PNNL team collaborated with scientists from University of Maryland to develop a comprehensive global hydrography database at 7 different spatial resolutions, ranging from 1/16 to 2 degree resolutions. The team evaluated the simulated streamflow globally against the observations from 1,674 major river gauge stations worldwide and systematically examined possible sources of model biases, which can impact the range of answers in simulations. These included model structure complexity, atmospheric forcing and the human influences on streamflow reflected in the observed streamflow in regulated rivers.

The river-routing model has been added to CESM and the Department of Energy's new Earth system model called Accelerated Climate Modeling for Energy (ACME) to support a wide range of research.


The quality and quantity of freshwater is being increasingly affected by new demands on the system-not only increased demands for use, but changes in the climate. Drought, floods and changes in weather systems all affect freshwater supply. In many areas, rivers provide the majority of freshwater for important uses. Accurately simulating the ebb and flow of rivers is important for understanding and predicting changes in the river. The global river-routing model described in this study simulates water quantity but it forms the basis for modeling stream temperature, sediments and nutrients that can greatly affect rivers and water quality.

"Evaluating the global streamflow simulation is a major milestone before we take on the next challenges in modeling other aspects of the river system, because water quantity affects the energy and matter flowing in the rivers," said Hong-Yi Li, a PNNL hydrologist who led the study.


Streamflow is a key component of the terrestrial system. By redistributing water and the associated heat content and nutrients through the hillslope, tributary, and stream network, streamflow plays an important role in the regional and global water, energy, and biogeochemistry cycles of the Earth system. To improve modeling of streamflow in Earth System Models, Department of Energy scientists at Pacific Northwest National Laboratory (PNNL) with collaborators at NASA Goddard Space Flight Center and University of Maryland evaluated the global implementation of the Model for Scale Adaptive River Transport (MOSART) recently developed at PNNL and coupled with the Community Land Model (CLM4.0). To support global modeling using MOSART, a co­mprehensive global hydrography dataset was derived at multiple resolutions from different sources. They first evaluated the simulated runoff fields against the composite runoff from Global Runoff Data Center (GRDC). With routing of the runoff from CLM by MOSART, the simulated streamflow reproduced reasonably well the observed daily and monthly streamflow at over 1,600 world’s major river stations in terms of annual, seasonal and daily flow statistics. They evaluated the impacts of model structure complexity. Results showed the spatial and temporal variability of river velocity simulated by MOSART is necessary for capturing streamflow seasonality and annual maximum flood. Other sources of simulation biases include uncertainties in the atmospheric forcing, as revealed by simulations driven by four different climate datasets, and human influences, based on a classification framework that quantifies the impact levels of large dams on the streamflow worldwide. In addition to simulating streamflow, MOSART provides a physically based global framework for modeling stream temperature and river biogeochemistry, both currently under or not represented in Earth System Models.

Hongyi Li
Pacific Northwest National Laboratory
"Evaluating Global Streamflow Simulations by a Physically Based Routing Model Coupled with the Community Land Model." JOURNAL OF HYDROMETEOROLOGY 16, 948–971 (2015). [10.1175/JHM-D-14-0079.1].

This study was supported by the Office of Science of the U.S. Department of Energy as part of the Earth System Modeling (ESM) and Integrated Assessment Modeling (IAM) programs. Development of the datasets used in this study is partly supported by the PNNL Platform for Regional Integrated Modeling and Analysis (PRIMA) initiative. The Pacific Northwest National Laboratory is operated by Battelle for the U.S. Department of Energy under Contract DE-AC05-76RLO1830. A. Getirana is funded by the NASA Postdoctoral Program (NPP) managed by Oak Ridge Associated Universities (ORAU). H. Wu is supported by NASA’s Applied Science Program. The first author also wants to thank C. Nilsson and C. A. Reidy for providing the data of river system classification.