Articles | Volume 16, issue 8
The Cryosphere, 16, 3215–3233, 2022
The Cryosphere, 16, 3215–3233, 2022
Research article
 | Highlight paper
12 Aug 2022
Research article  | Highlight paper | 12 Aug 2022

TermPicks: a century of Greenland glacier terminus data for use in scientific and machine learning applications

Sophie Goliber et al.


Total article views: 2,894 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
2,099 760 35 2,894 62 33 29
  • HTML: 2,099
  • PDF: 760
  • XML: 35
  • Total: 2,894
  • Supplement: 62
  • BibTeX: 33
  • EndNote: 29
Views and downloads (calculated since 14 Oct 2021)
Cumulative views and downloads (calculated since 14 Oct 2021)

Viewed (geographical distribution)

Total article views: 2,894 (including HTML, PDF, and XML) Thereof 2,673 with geography defined and 221 with unknown origin.
Country # Views %
  • 1

Discussed (final revised paper)

Latest update: 07 Dec 2022
Goliber et al. present a rich new dataset for the lengths of 278 glaciers around Greenland. This dataset, "TermPicks", contains 39,060 detailed traces of the edges of these glaciers, where ice meets ocean. TermPicks spans the entire satellite record (1970s-present), with air photo coverage for some glaciers going back >100 years to 1916. TermPicks is designed for use as training data in machine learning applications, which are the future of the tedious "terminus picking" work that has largely been performed by students to date. Thus, TermPicks will facilitate a significant leap forward in Greenland glacier research by facilitating machine-learning-enabled analysis of the continual high-flux, big-data output of high-resolution imagery by our international constellation of earth-observing satellites.
Short summary
Terminus traces have been used to understand how Greenland's glaciers have changed over time; however, manual digitization is time-intensive, and a lack of coordination leads to duplication of efforts. We have compiled a dataset of over 39 000 terminus traces for 278 glaciers for scientific and machine learning applications. We also provide an overview of an updated version of the Google Earth Engine Digitization Tool (GEEDiT), which has been developed specifically for the Greenland Ice Sheet.