The Cryosphere
The Cryosphere
TC
Articles | Volume 20, issue 2
Articles | Volume 20, issue 2
https://doi.org/10.5194/tc-20-905-2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/tc-20-905-2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
Articles | Volume 20, issue 2
Research article
 | 
03 Feb 2026
Research article |  | 03 Feb 2026

Enhanced neural network classification for Arctic summer sea ice

Anne Braakmann-Folgmann, Jack C. Landy, Geoffrey Dawson, and Robert Ricker

Viewed

Total article views: 6,275 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
4,963 1,144 168 6,275 285 316
  • HTML: 4,963
  • PDF: 1,144
  • XML: 168
  • Total: 6,275
  • BibTeX: 285
  • EndNote: 316
Views and downloads (calculated since 18 Jul 2025)
Cumulative views and downloads (calculated since 18 Jul 2025)

Viewed (geographical distribution)

Total article views: 6,275 (including HTML, PDF, and XML) Thereof 6,147 with geography defined and 128 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Saved (final revised paper)

Latest update: 16 Sep 2026
Download
Short summary
To calculate sea ice thickness from altimetry, returns from ice and leads need to be differentiated. During summer, melt ponds complicate this task, as they resemble leads. In this study, we improve a previously suggested neural network classifier by expanding the training dataset fivefold, tuning the network architecture and introducing an additional class for thinned floes. We show that this increases the accuracy from 77 ± 5 % to 84 ± 2 % and that more leads are found.
Read more
Share