Structure from motion will revolutionize analyses of tidal wetland landscapes

Ecosystem services of tidal wetlands depend upon hydrology and vegetation, which in turn vary with elevation differences on the order of centimeters. Variability on such a fine scale is not captured in digital elevation models prepared from conventionally acquired LiDAR data products that generally have a spatial resolution of 0.5-1.0 m and vertical uncertainties up to 15 cm. Until recently, capturing critical fine scale features required laborious, hands-on field surveys that took days to collect data and time limitations usually required surveys to be restricted to selected areas of a wetland. Using Structure-from-motion (SfM) photogrammetry and a small unmanned aerial vehicle, precise three-dimensional point clouds, digital surface models (DSM) and color orthomosaics were produced for three salt marshes in Eastern Canada. Vertical and horizontal measurements from the SfM photogrammetry compared favorably to those taken with a Differential Global Positioning System DGPS). Average horizontal displacements of 1.0-2.9 cm were found across the three salt marshes with an average elevation difference of 2.7 cm (+/- 1.7 cm) in comparison to DGPS. Analysis of the relationship of elevation between points taken with the DGPS and extracted from the SfM DSM gave an R-2 of 0.99. With a ground sampling distance of 23 cm our SfM photogrammetry generated models captured variations in topography associated with geomorphic features such as creeks, ponds, channel edges, and logs not visible in the DSM prepared from LiDAR of the same sites. SfM photogrammetry enables mapping of important hydrological features, such as creeks carrying drainage from upland watersheds or connectivity of ponds on the wetland surface. The former are important for transport of contaminants or diadromous fish, and the latter is important for resident fish, water birds, and mosquito larvae. Using SfM to distinguish vegetation structure that may indicate vegetation composition will enable more informed analyses of elevational controls on plant distribution and better prediction of their fate with sea level rise. Crown Copyright (C) 2017 Published by Elsevier Inc. All rights reserved.