Comparison Study: Glacier Calving Front Delineation in Synthetic Aperture Radar Images With Deep Learning

Authors:
  • NGNora Gourmelon
    Nora Gourmelon
  • KHKonrad Heidler
    Konrad Heidler
  • ELErik Loebel
    Erik Loebel
  • DCDaniel Cheng
    Daniel Cheng
  • JKJulian Klink
    Julian Klink
  • ADAnda Dong
    Anda Dong
  • FWFei Wu
    Fei Wu
  • NMNoah Maul
    Noah Maul
  • MKMoritz Koch
    Moritz Koch
  • MDMarcel Dreier
    Marcel Dreier
  • DPDakota Pyles
    Dakota Pyles
  • TSThorsten Seehaus
    Thorsten Seehaus

Abstract
Continuous monitoring of glacier calving fronts is essential for sea level rise projections. This study benchmarks Deep Learning systems for front delineation in Synthetic Aperture Radar imagery. While Deep Learning systems exhibit errors up to 221 m, human annotators deviate by only 38 m, underscoring the need for further research.