Skip to main content
ARS Home » Southeast Area » Oxford, Mississippi » National Sedimentation Laboratory » Watershed Physical Processes Research » Research » Publications at this Location » Publication #239525

Title: Numerical modeling of the impact of riparian soil water dyanmics on channel width adjustment

Author
item Langendoen, Eddy
item Lowrance, Robert

Submitted to: Congress of International Association for Hydraulic Research Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 7/1/2009
Publication Date: 8/9/2009
Citation: Langendoen, E.J., Lowrance, R.R. 2009. Numerical modeling of the impact of riparian soil water dyanmics on channel width adjustment. In: Proceedings of the 33rd IAHR Congress, August 9-14, 2009, Vancouver, Canada. pp. 3011-3018 (CD-ROM).

Interpretive Summary: Fine-grained sediment (clays and silts) in stream systems is a leading cause of stream impairment. Sediments eroded from the banks of disturbed streams are a significant contributor of fine-grained sediment. Studies at the U.S. Department of Agriculture-Agricultural Research Service National Sedimentation Laboratory have shown that the amount and rate of streambank erosion is related to differing amounts of water within the streambanks. Therefore, tools that assess streambank erosion need to properly account for the amount of water and its movement in a bank. Three models with different levels of sophistication were evaluated for their capability to accurately simulate the movement of groundwater in streambanks. The most complex model, which can simulate the movement of groundwater in both the saturated and unsaturated zone, showed that it can accurately simulate both the vertical and lateral movement of groundwater. The model with intermediate complexity, which can simulate the movement of groundwater in the saturated zone and the location of the groundwater surface, was capable of simulating averaged groundwater distributions, but underpredicted the rate at which the groundwater moved within a streambank. The simplest model was able to predict the correct final groundwater distributions, but the rate at which groundwater moves was poorly predicted. The model evaluation further showed that erosion caused by groundwater moving out of the bank into the stream may be more important to the stability of the streambank than the distribution of groundwater within the bank. Therefore, streambank erosion assessment tools or models that include the effects of groundwater should minimally use a groundwater model with intermediate complexity that appropriately simulates the elevation of the groundwater surface and the rate at which the groundwater moves out of the streambank.

Technical Abstract: Occurrence of streambank failure is closely related to redistribution of soil water that affects soil shear strength and may lead to seepage-induced erosion. Pore-water pressure in a streambank is affected, among others, by infiltrating rainfall, streambank-material texture, riparian vegetation, and interactions between surface water and groundwater. The hydrologic and biogeochemical components of the riparian ecosystem model REMM were integrated into the channel evolution model CONCEPTS to produce a model capable of simulating bank pore-water pressure distributions. Testing of the model showed that the subsurface flow component of REMM may not adequately simulate pore-water pressure dynamics at depth and near the groundwater table for streambank stability computations. Improved subsurface flow formulations that can be used to evaluate the effects of pore-water pressures on streambank stability at different temporal scales are presented and tested. To resolve the movement of wetting and drying fronts and their impact on bank stability temporally (i.e., time scales ranging from minutes to hours), a model based on the Richards equations describing unsaturated flow is needed. However, if streambank stability assessment uses time intervals of one day or larger, over which the pore-water pressure distribution may have reached a quasi-equilibrium state, the use of simpler models, such as those based on the Boussinesq equation, can satisfactorily predict subsurface flow and pore-water pressure. Further, seepage-induced erosion may be more important to the stability of a streambank than the redistribution of soil water.