The Application of Airborne Lidar Technology at the Stream Reach Scale: A Comparion of Lidar-Derived Stream Channel Cross-Sections and Long Profiles to Field Based Surveys

LiDAR-derived digital elevation models (DEMs) have proven a useful tool for geomorphologists at a variety of spatial scales. Despite the high-resolution of many LiDAR-derived DEMs, little focus has been placed on the use of these DEMs at the stream reach scale. This study compared two field-based stream reach surveys to those obtained from 0.5 m LiDAR-derived DEMs in a GIS for two third-order mixed bedrock-alluvial mountain streams. A thalweg long profile and ten channel cross-sections were surveyed in the field for each stream reach. The end points of each cross-section and long profile were recorded using a Trimble GeoXplorer XT differentially corrected GPS unit. The location of each endpoint was then imported into ArcMap as a point based shapefile. Using the interpolate line tool in ArcMap (ArcGIS 9.2), long profiles and channel cross-sections were created between these points using the 0.5 m LiDAR-based DEM. For each cross-section channel width, maximum depth, mean depth, and cross-sectional area were calculated. A comparison of LiDAR-derived cross-sections to field based cross-sections shows a systematic overestimation of channel width and underestimation of both maximum and mean depth for the LiDAR-based surveys. The overestimation of channel width is likely a bi-product of converting point based data to a 0.5 m grid based DEM and the smoothing that occurs when interpolating a line. The underestimations of both mean and maximum depth were all within the vertical accuracy of the LiDAR instrumentation, but may also represent data lost from post-processing or water in the channel. The LiDAR-based cross-sectional areas are within 10% of field-based cross-sectional areas. Comparison of LiDAR-derived long-profile channel slopes to those measured in the field showed no significant difference.