Abstract

Arctic beaded streams provide unique ecosystem functions and serve as important tundra habitats. Their unique ‘beads-on-a-string’ morphology is thought to form from thermokarst erosion, and they are densely represented in permafrost-ridden landscapes. Despite their ubiquity in high latitude regions, beaded stream formation and occurrence is not well studied, and beaded streams are not globally mapped. Access to these streams is challenging in their remote, dynamic environment, and up until recently, monitoring these streams through satellite imagery was difficult given their relatively small size with channel widths of a few meters. The availability of high-resolution imagery from Planet data now makes it possible to detect and map these streams over large areas. Here we observe and predict the location of beaded stream catchments throughout the pan-Arctic domain by combining the location of known beaded streams with recent advances in computer vision and high-resolution (3 m) satellite imagery. Specifically, we use the location of known existing beaded streams to classify potential river catchments as beaded or non-beaded, then download high resolution imagery across those regions, and use the latest You-Only-Look-Once (YOLO) object detection algorithm to identify beaded streams throughout the pan-Arctic, estimating 138,500 ± 43,700 beaded catchments globally, occurring in an estimated one third of all pan-Arctic catchments. In the largest dataset of beaded streams to date (Arp et al., 2015), only 375 catchments that contain beaded streams were identified, thus our estimate significantly expands our current understanding of the location and prevalence of Arctic beaded streams.

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

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.