Skip to main content
ARS Home » Southeast Area » Oxford, Mississippi » National Sedimentation Laboratory » Watershed Physical Processes Research » Research » Publications at this Location » Publication #235243
Title: Numerically predicting seepage gradient forces and erosion sensitivity to soil hydraulic properties

Author
item FOX, G.A. - Oklahoma State University
item HERREN, D.M. - Oklahoma State University
item Wilson, Glenn
item Langendoen, Eddy
item FOX, A.K. - Oklahoma State University
item CHU-AGOR, M.L. - Oklahoma State University

Submitted to: Journal of Hydrology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/7/2010
Publication Date: 8/11/2010
Citation: Fox, G., Herren, D., Wilson, G.V., Langendoen, E.J., Fox, A., Chu-Agor, M. 2010. Numerically predicting seepage gradient forces and erosion sensitivity to soil hydraulic properties. Journal of Hydrology. 389:354-362.

Interpretive Summary: Streambank failures result in loss of land and increased sediment and nutrients in streams. Computer models that predict subsurface flow have been combined with models of bank stability to predict bank failure. Understanding the degree of characterization of the soil properties necessary to predict when seepage contributes to bank failure is needed. This research examined how sensitive predictions of seepage were to soil properties and the effect of soil layers. A groundwater flow model was used to predict velocity (v) of seepage from a streambank. Different amounts of soil data were used, from knowledge of only soil texture to knowing the actual percent sand (%S), silt (%Si) and clay (%C); to also knowing the saturated hydraulic conductivity (Ks); and bulk density (BD) of the streambank. The baseline condition for comparison was taken to be the center of the soil textural class with known %S, %Si, %C, BD and Ks. Results indicated that a complete particle size analysis may not be necessary to adequately model seepage. The Ks was the most important property for predicting seepage velocity, although measuring %C may be an inexpensive alternative for sandy soils. A layered bank was simulated with several combinations of soil types. Results indicated that a reduction in Ks by only a factor of 10 from the top layer to the layer below may be necessary to induce seepage. This research suggests that Ks and/or BD measurements are necessary to indicate a layer that will cause seepage, and Ks and/or %C measurements are necessary to adequately predict v. If a field survey is performed to locate streambanks susceptible to failure, simply identifying soil layers by soil type may not be adequate.

Technical Abstract: Streambank failures result in loss of land, increased stream sediment loads, and increased nutrient loads if nutrient levels are high. Variably saturated flow models integrated with bank stability models are being used to predict bank failure; however, understanding of the soil characterization necessary to predict when seepage contributes to bank failure is needed. This research examined sensitivity of seepage predictions to soil properties and soil layering in order to provide guidance on when seepage mechanisms need to be considered in streambank stability analyses. A two-dimensional groundwater flow code, SEEP/W, was used to model a hypothetical streambank. Varying levels of soil data were used, from textural class only to percent sand (%S), silt (%Si) and clay (%C); %S only; saturated hydraulic conductivity (Ks); and bulk density (BD). A pedotransfer function (PTF), ROSETTA, was used to estimate hydraulic parameters for SEEP/W based on the known soil properties for each treatment. The baseline condition was taken to be the centroid of each textural class with known %S, %Si, %C, BD and Ks. Error in average seepage velocity (v) due to limited soil data was analyzed. Results indicated that a complete particle size analysis may not be necessary to adequately model seepage. The Ks was the most effective parameter at reducing error in v, although measuring %C may be an inexpensive alternative for sandy soils. A layered bank was simulated with several combinations of soil types. Results indicated that only one order of magnitude difference in Ks, (equivalent to a resistance ratio (Rr) of 10) may be necessary to induce perching and seepage from the top layer. If a field reconnaissance is performed to locate streambanks susceptible to perching, simply identifying soil layers by soil type may be insufficient. This research suggests that Ks and/or BD measurements are necessary to indicate the presence of a restrictive layer, and Ks and/or %C measurements are necessary to adequately predict v.