Skip to main content
ARS Home » Research » Publications at this Location » Publication #73396
Title: NEW DIGITAL LANDSCAPE PARAMETERIZATION METHODOLOGIES

Author
item Garbrecht, Jurgen
item Starks, Patrick
item MARTZ, LAWRENCE - UNIV OF SASKATCHEWAN

Submitted to: American Water Resources Association Conference Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 9/1/1996
Publication Date: N/A
Citation: N/A

Interpretive Summary: The evaluation of the shape and form of a landscape has recently been automated by use of computer technology. The methods that were used to determine runoff over the landscape have several weaknesses that limit their application. This paper proposes improvements in the determination of the runoff over flat areas, addresses the identification of landscape features, and determines width, length, and slope measurements of the subcatchments. These proposed improvements and additions greatly speed up computer modeling of the runoff and provide more precise and reproducible determinations than traditional manual techniques.

Technical Abstract: Traditional watershed parameter extraction from raster Digital Elevation Models (DEM) is handicapped by methodological weaknesses with respect to the treatment of sinks and flat surfaces. These features are often artifacts of the DEM and are problematic for the identification of landscape drainage. New DEM pre-processing procedures to eliminate sinks and flat surfaces are presented. The pre-processed DEMs lead to more realistic and consistent drainage patterns. Additional procedures are presented that capture the topology of drainage networks from raster images and identify essential geometric characteristics of overland areas. These procedures enable the automated management of the flow routing process in the watershed and generate network and subcatchment data for rainfall-runoff modeling.