|Prasad, Shyam - UNIV. OF MS - CIVIL ENGR|
|Wells, Robert - UNIV. OF MS - CIVIL ENGR.|
Submitted to: Federal Interagency Hydrologic Modeling Conference
Publication Type: Proceedings
Publication Acceptance Date: July 1, 2002
Publication Date: July 28, 2002
Citation: Romkens, M.J., Prasad, S.N., Dicarlo, D.A., Wells, R.R. 2002. Development of a predictive relationship for infiltration into swelling/cracking soils: an update. Federal Interagency Hydrologic Modeling Conference. (CD-ROM) Session 1.B. Interpretive Summary: Predicting soil loss and runoff from upland areas including cultivated bottomland remains a difficult challenge. In order to arrive at better estimates of runoff from bottomland conditions, scientifically based prediction equations must be derived for infiltration of swelling and cracking soils. This paper summarizes recent papers developing such prediction equations, including estimates of incipient ponding time. The relationship includes the most relevant factors: the rainfall intensity regime, the crack morphology (crack width, depth, and spacing) and a soil parameter, which consists of factors, that are part of the soil diffusivity function. The relationship obtained should be very helpful in providing infiltration estimates for swelling/cracking soils in soil erosion and runoff prediction models.
Technical Abstract: Predicting the hydrologic response of watersheds during a rainstorm event is of major interest to hydrologists. To a large degree, the prediction difficulties encountered are related to the problem of arriving at realistic estimates of the infiltration component. Spatial and temporal variations of the soil water regime, the dynamic nature of the soil water matrix of real world soils, and the non-linearity of the governing soil water flow (Richard's) equation make accurate predictions extremely difficult and complex. One of the areas of major concern is the prediction of rain infiltration in swelling/cracking soils commonly found in alluvial plains. This paper describes the current status of the development of a spectral series solution approach to this class of soils. The solution is based on a prismatic model in which excess rainwater is laterally absorbed through the vertical faces of the prismatic elements. During the 1998 Federal Interagency Hydrologic Modeling Conference, this model was used to describe incipient ponding time as a function of the crack morphology (depth, spacing), the rainfall intensity, and a soil parameter. This paper extends this analysis and arrives at a cumulative infiltration equation that includes also the crack width. The underlying assumptions of the model are: (1) the surface of the prismatic columns is crusted and infiltration through the surface is negligible, (2) rain water flows uniformly over the edge of the columns into the crack and is laterally absorbed through the lateral surfaces of the columns, (3) the proximity of adjacent lateral surfaces does not affect the infiltration process. The analytical results are compared with experimental observation. These observations were made in the laboratory on prepared soil beds of a Mississippi Delta swelling/cracking soil (Sharkey soil), subjected to a series of rainstorms with intensities of 20 mm h-1. Good agreements between predicted and observed results were obtained.