Simulating Interactions among River Water, Groundwater, and Land Surfaces by Coupling Different Models
New coupled model, CP v1.0, will improve understanding of water cycling and complex Earth system dynamics.
Many Land System Models (LSMs) do not consider lateral transport of water, leaving out a cross-sectional view and understanding of the coupling of surface water with groundwater along rivers, streams, and other water bodies. And yet, detailed observational studies and their accompanying model simulations suggest that the lateral flow of water in the subsurface along the continuum of river water and groundwater saturates the pore space in the soils and sediments. There is a need to advance large-scale LSMs so that they capture the variable gradient of water within soils because this is critical for understanding and modeling energy and water budgets, as well as biogeochemical cycling in the terrestrial surface and subsurface systems. This work enables this modeling advance by coupling a widely used, massively parallel multiphysics reactive transport code with the Community Land Model version 4.5 to create a coupled model called CP v1.0.
The open-source coupled model developed in this study, CP v1.0, can be used to improve the mechanistic understanding of ecosystem functioning and biogeochemical cycling along river corridors and their functions in watersheds. The associated dataset from a well-characterized river shoreline site can also be used as a benchmark for testing other integrated models.
The research community increasingly recognizes that rivers, despite their relatively small imprints on the landscape, play important roles in watershed functioning through their connections with groundwater aquifers and riparian zones. The Columbia River, a 1,243 mile stretch of water, served as an ideal test case for long-term observations, as well as simulations using a coupled three-dimensional (3D) surface and subsurface land model.
The interactions between groundwater and river water are important because they influence the volume of water in soils, from simply moist to fully saturated. This volume determines the rates of biogenic gas emissions due to soil evaporation, plant transpiration, and respiration of carbon dioxide from plants and soils, which are poised to vent into the atmosphere. These same interactions also enhance the reactive transport process that alters water chemistry and the downstream transport of materials and energy.
However, past simulations of these processes and their impacts haven’t always mirrored the reality of field observations, in part because such models do not take into account the lateral flow of water and transport of constituents in the subsurface.
During a five-year monitoring of groundwater wells along the Columbia River shoreline, a team of researchers from the Pacific Northwest National Laboratory (PNNL), Lawrence Berkeley National Laboratory, and Sandia National Laboratory recognized the value of observing the layers within the subsurface rather than just what happens above ground. They used two open-source codes, PFLOTRAN and CLM4.5, to compare simulations to observations. They then coupled the two models to create CP v1.0. The coupled-model approach allowed the research team to estimate moisture availability, for example, particularly during changes in the river stages, and to validate the new model using data from the shoreline site.
Researcher Maoyi Huang from PNNL noted one surprise during the study: spatial resolution matters. The influence of river-aquifer interactions can be “seen” in shallow groundwater using coarser-resolution simulations, but it is important to refine the model resolution along river corridors that were part of this study. The difference, she explained, is that southeastern Washington state is situated in an arid climate zone so the team had to use finer resolutions in their study in order to capture the processes at the surface and in the subsurface within the narrow riparian zone.
A coupled model like the one used in this study can also be applied to larger modeling problems, such as simulating the impact of a drought on watershed functioning by explicitly considering the role of river-aquifer-land interactions. Using models that do not consider lateral flow and transport can be misleading. For example, models without this 3D view results often erroneously show parched plants, one signature that a drought is underway. But a model incorporating this view shows that plants are still, in fact, getting water from the soil.
Pacific Northwest National Laboratory
Lawrence Berkeley National Laboratory
This research was supported by the Subsurface Biogeochemical Research program (SBR) of the Office of Biological and Environmental Research (BER), within the U.S. Department of Energy (DOE) Office of Science. This contribution originates from the SBR Scientific Focus Area (SFA) at Pacific Northwest National Laboratory.
Bisht, G., Huang, M., Zhou, T., Chen, X., Dai, H., Hammond, G. E., Riley, W. J., Downs, J. L., Liu, Y., and Zachara, J. M. “Coupling a three-dimensional subsurface flow and transport model with a land surface model to simulate stream-aquifer-land interactions (CP v1.0).” Geoscientific Model Devopment 10, 4539–62 (2017). [DOI:10.5194/gmd-10-4539-2017].