|Martz, Lawrence - UNIV OF SASKATCHEWAN|
|Syed, Kamran - USDA,ARS,TUCSON|
Submitted to: American Society of Civil Engineers Hydraulic Conference Proceedings
Publication Type: Proceedings
Publication Acceptance Date: September 1, 1999
Publication Date: September 1, 1999
Interpretive Summary: Computer program simulations are often used to estimate changes in surface runoff and soil erosion as a function of alternative agricultural practices. The use of computer programs requires input information such as weather, land use, soil conditions and land surface characteristics. The measurement of land surface characteristics by survey or derivation from maps is time and resource consuming. This can be automated by using a digital representation of the land surface topography and a computer program. The automated procedure is particularly effective for large watersheds that contain numerous subcatchments for which the traditional evaluation is not practical. Algorithms and procedures for a computer program to automatically extract representative land surface slope and length values from the digital land surface topography are presented in this paper. The algorithms and procedures are applied to a watershed situation and the results are compared to map-derived values. Differences between automated and manual methods are found and explained by differences in the underlying assumptions of the methods. Automated methods that are based on similar assumptions than the manual method produced comparable results. From a practical point of view, it was concluded that all methods produced equally valid representative land surface slope and length values, and the user must select the method that is most consistent with and suitable for their specific application. For example, the method based on travel distance and flow path produces length estimates that are more suitable for runoff, erosion and contaminant transport, whereas the method based on local slope produces slope estimates that are more suitable for vertical energy and water vapor fluxes.
Technical Abstract: In distributed watershed modeling it is often necessary to identify representative values for subcatchment characteristics, such as length and slope. The distributed nature of such characteristics within a subcatchment precludes the direct measurement of a representative value from the land surface topography. In this study several Data-Reduction (DR) models are presented to automatically calculate representative subcatchment length and slope values from Digital Elevation Models. The representative values calculated by the DR models are compared to each other and to traditional, map-derived measurements. Significant differences in results are found between models. The differences are not errors, but represent valid, yet different, conceptualizations and interpretations of length and slope by the DR models. Once the differences in models are accounted for, the results are consistent with one another, and no one method, including the traditional, map-based method, i found to be superior. The selection of the appropriate model must be based on the intended subsequent application of the representative values. For example, local terrain-slope values are believed to provide a better slope approximation for vertical energy and water vapor flux applications, whereas flow path based length and slope values are believed to provide a more suitable value for runoff, erosion and contaminant transport applications.