Richie Carmichael, QCI, University of Idaho
Idaho Office of Species Conservation
Bureau of Reclamation
University of Idaho
USFS RockyM Mountain Research Station
Generation of seamless, high resolution, watershed scale digital surface models (DSMs) is rapidly becoming an essential tool for describing and modeling characteristics and processes important to aquatic habitat suitability, sediment transport, riparian vegetation mapping, characterization of topographic change processes, flood analysis and prediction, numerical flow simulation, and pre- and post-restoration efforts (Benjankar et al. 2016; Leskens et al. 2015; Lyon et al. 2015; Mandleburger et al. 2011; McKean et al. 2008, 2009, 2014; Montealegre et al. 2015; Wyrick et al. 2015). Green waveform bathymetric LiDAR has emerged as the leading technology in high-resolution terrain and bathymetric mapping.
Topobathymetric light detection and ranging (LiDAR) data were collected along the Lemhi River using Experimental Advanced Airborne Research LiDAR-B (EARRL-B). EARRL-B is the latest generation of EARRL LiDAR systems developed by NASA and the USGS (McKean et al. 2008, 2014). Data were recorded over three days of flight, focusing on in-channel and floodplain habitat.
To validate the data, a three-person crew conducted high-resolution RTK-dGPS surveys in-channel and on the floodplain, and collected points on top of the highway paralleling the river and along the river thalweg. They filtered and analyzed the LiDAR point cloud for vertical accuracy by comparing point-by-point elevations of ground classified LiDAR points to those of two high-resolution ground surveys within two morphologically distinct control reaches. Elevation values of the LiDAR point cloud along the road surface paralleling the study reach were also analyzed using the same point-to-point method. While LiDAR was able to penetrate heavily vegetated banks, data from these areas provided more noise than in-channel or floodplain areas.
Findings and Uses
LiDAR and EARRL-B are able to resolve stream bathymetry at the 10’s of cm scale; however, are limited at the micro-habitat (e.g., cobble) scale (Hilldate et al. 2007).
LAS Tools can be used to filter massive datasets where water surface, vegetation, and other noise returns exist.
Prospective users of bathymetric LiDAR data now have a framework for processing and filtering point cloud data. This framework may also help improve ground determination in raw point clouds from previous studies.