Abstract
Unmanned aerial vehicle (UAV) photogrammetry has become an important tool for generating multi-temporal, high-resolution ortho-images and digital elevation models (DEMs) to study glaciers and their dynamics. In polar regions, the roughness of the terrain, strong katabatic winds, unreliability of compass readings, and inaccessibility due to lack of infrastructure pose unique challenges for UAV surveying. To overcome these issues, we developed an open-source, low-cost, high-endurance, fixed-wing UAV equipped with GPS post-processed kinematic for the monitoring of ice dynamics and calving activity at several remote tidewater glaciers located in Greenland. Our custom-built UAV is capable of flying for up to 3 hours or 180 km and is able to produce high spatial resolution (0.25-0.5 m per pixel), accurately geo-referenced (1-2 pixels) ortho-images and DEMs. We used our UAV to perform repeat surveys of six calving glacier termini in north-west in July 2017 and of Eqip Sermia glacier, west Greenland, in July 2018. The endurance of our UAV enabled us to map the termini of up to four tidewater glaciers in one flight and to infer the displacement and calving activity of Eqip Sermia at short (105 minute) timescale. Our study sheds light on the potential of long-range UAVs for continuously monitoring marine-terminating glaciers, enabling short-term processes such as the tidal effects on the ice dynamics, short-lived speed-up events, and the ice fracturing responsible for calving to be investigated at unprecedented resolution.
Highlights
Recent developments in unmanned aerial vehicle (UAV) technology offer a broad range of solutions for acquisition of spatial data
In July 2017 at Inglefield Bredning, camera locations were retrieved from the camera trigger events recorded in the log file of the autopilot. This initial method has several drawbacks: (i) the on-board GPS used for Unmanned aerial vehicle (UAV) navigation relies on standard positioning which has an absolute accuracy of up to 5 m horizontally and up to 30 m vertically, (ii) there is a delay between the time the camera is triggered and the time the image is effectively taken causing a discrepancy of up to several meters between the recorded trigger location and the true image location, (iii) the number of images collected after one flight is usually lower than the number of camera triggers recorded by the autopilot due to occasional missing images (∼0.5% of images were missing in 2017)
We present the outcomes of UAV flights operated during the two field campaigns (Table 2) to demonstrate the potential of our UAV for glaciological studies in polar environments, and especially for large, inaccessible regions of rapid ice flow, such as tidewater glaciers
Summary
Recent developments in unmanned aerial vehicle (UAV) technology offer a broad range of solutions for acquisition of spatial data. For these reasons, existing commercial platforms are not always suitable, emphasizing the need for customized UAVs which can be modified and tuned to tackle the above-mentioned challenges In response to these challenges, Ryan et al (2015) built a dedicated fixed-wing UAV platform based on an open-source autopilot to perform structure-from-motion multi-view stereo (SfM-MVS) photogrammetrical mapping of calving glaciers in Greenland. Chudley et al (2019) added an on-board differential carrier-phase GNSS receiver to this platform which enabled the acquisition of georeferenced photogrammetrical products with high accuracy (in the range of one pixel size) without the use of GCPs. In this paper, we detail an open-source, low-cost, fixed-wing UAV designed for monitoring glacial processes similar to the one presented in Ryan et al (2015), Jouvet et al (2017), Chudley et al (2019), and Jouvet et al (2019), but optimized for flight endurance to survey remote areas. We present some key fields of application demonstrating the potential of our UAV platform: ortho-images, DEMs, result accuracy, ice flow velocity fields, and observation of calving events
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call