Assessing multitemporal water‐level changes with uav‐based photogrammetry

Abstract

Unmanned aerial vehicle (UAV) images and structure-from-motion (SfM) methodology allow very high resolution surface reconstruction. This fairly low-cost and time-saving method is used for northern bog hydrology characterisation. The primary interest of this study lies in seasonally changing bog pool water levels. The study site is located in a potential conflict zone where water extraction from an underground oil shale mine threatens protected wetlands. The study involves two test areas that were surveyed three times, allowing spatio-temporal quantification of seasonal changes of the bog surface. Aerial lidar data and manual measurements were used to assess the quality of SfM surfaces and discrepancies between the different methods were analysed. Overall, there is a good fit between the methods used, validating the use of SfM methodology in a natural wetland environment. This study demonstrates that the nearby mine is influencing the seasonal water-level fluctuations of the protected wetland.

Looking to the Skies: Realising the Combined Potential of Drones and Thermal Infrared Imagery to Advance Hydrological Process Understanding in Headwaters

“…Although this commentary focuses on drones and TIR for understanding headwater dynamics, we nonetheless recognize the potential that standard (RGB) visible drone imagery holds for improving process understanding of headwater hydrology. For example, the ability to resolve catchment microtopography from highly detailed digital surface models (DSMs, derived from structure from motion photogrammetry) has the potential to revolutionize the extraction of hydrological networks at a resolution several orders of magnitude higher definition than existing LiDAR datasets acquired by national mapping agencies (e.g., Scottish Government, 2012), as well as the derivation of water level/depth with centimetric precision (e.g., Dietrich, 2017; Kohv et al., 2017). Such advances hold promise for enhanced modeling of surface flows, although it is important to consider that large increases in resolution would also necessitate bulk improvements in computing power to avoid yielding prohibitively long model runtime.…”

Lightweight Deep Neural Network Method for Water Body Extraction from High-Resolution Remote Sensing Images with Multisensors

“…Unmanned aerial vehicle remote sensing has become one of the most important methods for acquiring image data in recent years. Compared with satellite images, it has the advantages of a low cost, fast transmission speed and is not limited by geographical environment [51]. However, UAV remote sensing also has the problems of a small load, short endurance time, small image coverage and low productivity.…”

A New Beach Topography-Based Method for Shoreline Identification

“…To extract the shoreline from DEMs and orthomosaics, we identified a new semi-automatic method based on the beach profile by the SfM technique. The method is based on the principle that SfM performs poorly on uniform or reflecting surfaces like the sea [29]. The beach profiles obtained with SfM are more irregular and unrealistic on sea, becoming regular and realistic when the points are referred to the land.…”