Skip to main content
ARS Home » Southeast Area » Oxford, Mississippi » National Sedimentation Laboratory » Watershed Physical Processes Research » Research » Publications at this Location » Publication #193822

Title: Rain infiltration into swelling/shrinking/cracking soils

Author
item Romkens, Mathias
item PRASAD, S.N. - UNIV. OF MISSISSIPPI

Submitted to: Agricultural Water Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/19/2006
Publication Date: 9/18/2006
Citation: Romkens, M.J., Prasad, S. 2006. Rain infiltration into swelling/shrinking/cracking soils. Agricultural Water Management. 86:196-205.

Interpretive Summary: Describing the hydrologic response of watersheds during storm events is of tremendous importance for water management planning, water control, and agriculture. The most difficult component to estimate is the amount of water that enters the soil or that is stored in the surface drainage system. Soil is a highly complex porous medium, and infiltration rates vary in time and place depending on the local soil properties and soil profile conditions. Of particular challenge is the problem of estimating rain infiltration into swelling, shrinking, and cracking soils. Depending on the prevalent cracking condition, large quantities of rainwater may be absorbed into the cracks during periods of drought, while during relatively wet conditions, when cracks are small or closed altogether, rain infiltration is small and most rainwater becomes runoff. This article attempts to quantify the infiltration into a highly swelling/cracking soil. To that end, an infiltration model was developed that represents a two-component process consisting of Darcy flow in the soil matrix and Hortonian flow of excess rain along the vertical crack surfaces. It was shown that ponding time can be described as a function of the crack morphology (spacing, width, and depth), rainfall intensity, and soild characteristics. For the special case, when crack closure occurs when the wetting front tip on the crack surfaces reach the bottom of the crack, an exact solution for field ponding was obtained. Also, an expression for the cumulative infiltration was obtained as a function of rainfall intensity, field ponding time, and soil parameters. The derived equations allow for improved predictions of runoff initiation and volume on swelling/cracking soils.

Technical Abstract: Rain infiltration in swelling/cracking soil is of great interest to engineers and scientists, from both the hydrologic and environmental perspectives. Depending on the degree of soil wetness and crack size, large quantities of rain may disappear into the soil profile and arrive at the groundwater before runoff starts, while under similar rainfall conditions on relatively wet soils following crack closure most rain becomes surface runoff. Making quantitative estimates of runoff from swelling/shrinking/cracking soils depends on our ability to predict the swelling dynamics of these soils during a storm event in relation to the crack conditions and rainfall regime. In addition, infiltration into the soil surface between the cracks is often limited by severe surface sealing due to the destruction of the soil surface structure by impacting raindrops. This paper describes a model for rain infiltration into swelling/shrinking/cracking/soils. The simplifying assumptions are: (1) No vertical infiltration takes place due to a sealed surface condition. Rain water moves laterally over the soil surface to the cracks, where it uniformly flows along the vertical walls of the cracks. (2) The geometry is represented by a prismatic column structure with cracks between the columns. Water imbibes horizontally and the interaction horizontal infiltration from the adjacent vertical surfaces of the columns is not considered. The analysis uses the wetting front solution of the Richards equation for horizontal diffusion into the soil matrix. The crack volume is determined from bulk density measurements, from which ponding time estimates can be made. Closed form expressions are derived for the cumulative infiltration. The analytical results are compared with experimental results obtained for a Mississippi Delta clay soil. For the case, where excess rain water has reached the bottom of the crack at the moment crack closure occurs, an exact solution is obtained for incipient ponding as a function of crack morphology, rainfall intensity and a soil parameter. Runoff is computed as the difference between the cumulative rainfall and cumulative infiltration.