Evaluating spatio-temporal dune volume changes from high resolution mobile terrestrial lidar
Ian Conery1, Nicholas Spore1, Shannon Walker2, Nicholas Cohn3, Katherine L Brodie3 and Julie C Zinnert4, (1)U.S. Army Engineer Research and Development Center, Coastal and Hydraulics Laboratory, Field Research Facility, Duck, NC, United States, (2)Virginia Commonwealth University, Biology, Richmond, VA, United States, (3)U.S. Army Engineer Research and Development Center, Coastal and Hydraulics Laboratory, Field Research Facility, Duck, United States, (4)Virginia Commonwealth University, Department of Biology, Richmond, VA, United States
Abstract:
Coastal foredunes are critical morphologic features that act to prevent flooding to infrastructure during major storm events. However, dune properties can vary considerably over short spatial scales – contributing to corresponding spatial variability in the flood protective services that they provide. This alongshore variable dune evolution is the result of spatial heterogeneity in environmental forcings (e.g., winds, waves, water levels), sediment characteristics, beach morphology, ecological properties, and coastal management actions. However, there are limited suitable field datasets which allow for the de-coupling of these competing signals. Repeat mobile terrestrial lidar (MTL) surveys, which can collect point returns at up to 1000 pts/m2, can provide data at the appropriate spatial and temporal scales to de-couple these environmental, morphologic, and ecological controls on dune dynamics.
Regional (40 km) MTL surveys of the beach and dune were conducted 18 times in the Outer Banks of NE North Carolina over the course of ~2.5 years. These high resolution datasets provide a valuable source of information on spatio-temporal dune dynamics. However, quantification of sediment volume change from these MTL datasets necessitates an accurate definition of the bare earth surface. In this work we first explore the relative accuracy of different automated filtering techniques (e.g., cloth simulation) and parameters (e.g., slope limiters) for extracting dune grass vegetation from the MTL point clouds. Morphological outputs are validated against RTK ground-truthed surfaces and remote sensing derived vegetation metrics (e.g., density and height) are compared against ecological quadrat measurements. Utilizing these optimum filtering settings, bed elevation and vegetation properties are extracted from all of the available MTL datasets to relate spatial patterns in erosion and deposition on the backshore and the dune face to local vegetation characteristics.
