Skip to main content
ARS Home » Midwest Area » West Lafayette, Indiana » National Soil Erosion Research Laboratory » Research » Publications at this Location » Publication #140274

Title: Representative hillslope methods for applying the WEPP model with DEMs and GIS

Author
item COCHRANE, T - SANTA CRUZ, BOLIVIA
item Flanagan, Dennis

Submitted to: Transactions of the ASAE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/27/2003
Publication Date: 8/31/2003
Citation: Cochrane, T.A., Flanagan, D.C. Representative hillslope methods for applying the WEPP model with DEMs and GIS. Transactions of the American Society of Agricultural Engineers. 2003. v. 46(4). p. 1041-1049.

Interpretive Summary: Predicting how runoff and soil erosion will occur is very important when trying to protect our nation's soil and water resources. It is just too expensive and impractical to even consider taking field measurements of sediment loss in every possible location. The alternative is to use computer simulation tools that can mimic the effects of the natural processes of rainfall, runoff, and soil detachment. A critical part of erosion prediction is accurate estimation of slope length and steepness, since the rates of runoff and soil detachment are directly related to slope characteristics. This paper describes different techniques and mathematical equations to estimate slope gradients and slope lengths, when digital elevation data is available. These techniques allow for rapid estimation of input values to erosion prediction computer models facilitating automatic simulation of small watersheds with "representative" slope shapes. In this study, we found that good results were obtainable with several of the procedures. This research impacts conservation agency personnel (Natural Resources Conservation Service, Forest Service, etc.), other research scientists, and farmers who may desire to apply the Water Erosion Prediction Project (WEPP) model, or other erosion prediction software to particular fields or small watersheds to assess the risk of soil erosion.

Technical Abstract: Modeling procedures called the Hillslope methods were developed that use geographical information systems (GIS) and digital elevation models (DEMs) to assess water erosion in small watersheds with the Water Erosion Prediction Project (WEPP) model. The Hillslope methods are automated procedures to develop hillslope and channel topographic characteristics from DEMs for use in the WEPP model. Each hillslope is modeled as a rectangular area with a representative slope profile that drains to the top or sides of a single channel. Three methods of creating a representative slope profile from DEMs were developed and tested: linear average, exponential transformed average, and weighted average. Additionally, two methods to determine the representative hillslope profile length, called the Calcleng and Chanleng methods, were evaluated. The Calcleng method calculates a representative length of hillslope based on the weighted lengths of all flowpaths in a hillslope as identified through a DEM. The Chanleng method sets hillslope width equal to adjacent channel length, then computes a hillslope length from hillslope area divided by width. Actual DEMs from six research watersheds were used to test these methods. The results from the application of these methods were compared to each other and to measured sediment data. Results show that the three methods for determining the representative slopes of the profiles were not significantly different from each other. There were also no significant differences between the Calcleng and the Chanleng methods for sediment yields and runoff from the six watersheds. These results help in facilitating a GIS interface for the application of WEPP to watersheds. For automated watershed simulations with WEPP, we recommend use of the weighted average method for determination of representative slope profile gradient values, in combination with the Chanleng method to determine representative profile slope lengths for hillslopes laterally adjacent to channel segments.