Assessing Seasonal Changes of Spatial Complexity in Riverscapes using Drone-Based Laser Scanning
Hession, W. Cully
MetadataShow full item record
Light detection and ranging (lidar) is a form of remote sensing using laser pulses to measure distances. Recent advancement in lidar technology has made units small enough to mount on drones, which makes high-quality data more accessible. Recent studies have utilized drone-based photogrammetry to measure characteristics of streams and rivers, as well as their associated riparian areas. These areas have been referred to as riverscapes. The physical characteristics of riverscapes are traditionally difficult to measure due to ever-changing characteristics across space and time. Drone-based laser scanning (DLS), is uniquely positioned to measure changing physical characteristics as it allows for increased temporal (daily, monthly, seasonal flights) and spatial (more than 400 pts/m2 at 30-m flight elevation) resolutions. It has more upfront costs compared to photogrammetry, as a DLS system (large drone and lidar) is vastly more expensive than a small drone with a digital camera payload. However, lidar can penetrate through vegetation, allowing for high-quality ground data, as well as vegetation points, which is a limitation of photogrammetry. One use of this ground and vegetation data is to analyze small changes of the topography to estimate complexity (an important habitat variable), as well as obstructions to flow such as vegetation. These obstructions to flow result in increased roughness, which is an important metric in biological studies and hydraulic modeling. In previous studies, estimating roughness was limited to visual observations or back-calculating from flow measurements, which can be time consuming and does not produce continuous spatial data. Using DLS-derived ground and vegetation, we will monitor small changes in vegetation and topography over the course of the stream both longitudinally, laterally, and through time. We will test various methods of computing roughness from detailed lidar point clouds to determine roughness. Some possibilities estimating roughness and complexity include the standard deviation of the elevation change, the variation between maximum and minimum elevations in a pixel, slope variability, surface roughness factors, and others. These values can be compared to a calibrated 2D hydraulic flood modeling (HEC-RAS), DLS has the potential to change the way we map and understand spatial complexity and habitat characteristics of riverscapes.