GSA Annual Meeting in Seattle, Washington, USA - 2017

Paper No. 354-12
Presentation Time: 9:00 AM-6:30 PM

CAN DRONES MEASURE LWD?: HIGH RESOLUTION AERIAL IMAGERY AND STRUCTURE FROM MOTION AS A METHOD FOR QUANTIFYING INSTREAM WOOD


TRUKSA, Tyler, Mathematics and Physical Science, College of Idaho, 2112 Cleveland Blvd, Caldwell, ID 83605, BOWMAN, Jake, Biology, College of Idaho, 2112 Cleveland Blvd, Caldwell, ID 83605 and GOODE, Jaime, Geosciences, Colorado State University, Campus Delivery 1482, Fort Collins, CO 80521, tyler.truksa@yotes.collegeofidaho.edu

Large wood debris (LWD) plays an important role in shaping geomorphic complexity and aquatic habitat; quantifying LWD in the field has valuable application in river and floodplain management. Advances in Unmanned Aerial Vehicles (UAVs) and Structure from Motion (SfM) software have opened new possibilities to quantify morphologic characteristics of landscapes and river systems more easily, quickly, and accurately than the previous manual survey methods. Our study focused on Clear Creek in the South Fork of the Payette drainage in central Idaho, an area affected by significant blowdown in the wake of the Pioneer Fire in 2016. We used a DJI Phantom 3 Professional to collect imagery along six 200-m reaches. These images were then processed with Agisoft Professional SfM software to create high-density point clouds and meshes to determine volumes of the LWD. These volumes were then compared with manual length and diameter field measurements, estimated with the equation for cylinders. Variables in the point cloud generation process (e.g., number of pictures, accuracy of photo alignment, and density of points and meshes) were altered to determine limitations of the method relative to traditional methods. This method has the potential to enhance estimates of LWD in rivers by improving our ability to represent more complete volumetric data (i.e. the inclusion of root systems and branches into total debris volume) which traditional methods lack.