Effects of DEM resolution on the WEPP runoff and erosion predictions: A case study of forest areas in northern Idaho.

摘要

DEMs (digital elevation models) can be used in a GIS (geographic information system) to represent topography and extract terrain features. DEMs vary in resolution and accuracy by the production method. The most widely used DEMs are the publicly accessible USGS (U.S. Geological Survey) NED (National Elevation Dataset) DEMs at 30-m and 10-m resolutions. There are other sources of DEMs with different resolutions and qualities, such as LIDAR (LIght Detection and Ranging) DEMs and SRTM (Shuttle Radar Topography Mission) DEMs. DEMs with different resolutions and sources can generate varied topographic and hydrologic features, which may in turn affect the runoff and sediment yield predictions in soil erosion models, such as the WEPP (Water Erosion Prediction Project) model.;This research project studies the effects of DEM resolutions and sources on (1) deriving topographic and hydrologic attributes, and (2) predicting watershed hydrology and water erosion using WEPP v2005. For two small forest watersheds located on Moscow Mountain in northern Idaho, six DEMs were prepared: LIDAR 30-m, 10-m, and 4-m DEMs, NED 30-m, 10-m DEMs, and SRTM 30-m DEM. These DEMs were used to calculate topographic and hydrologic parameters that served as inputs in WEPP. The model results were then compared with the runoff and sediment yield data observed at the watershed outlets.;This study has found that DEMs with different resolutions and sources can generate varied watershed shapes and structures, extract different hillslope and channel lengths and gradients, and produce significantly different sediment yield predictions in WEPP. In general, as DEM resolution became finer, its accuracy was higher, the landscape was more precisely and accurately represented, and the sediment yield estimates approached closer to the observed values. Conversely, as DEM resolution became coarser, its accuracy was lower, and the sediment yield estimates departed from the observed values. The study has also found that LIDAR DEMs are potentially very useful tools for soil erosion modeling.