MMSeaIce: a collection of techniques for improving sea ice mapping with a multi-task model

Chen, Xinwei; Patel, Muhammed; Pena Cantu, Fernando J.; Park, Jinman; Noa Turnes, Javier; Xu, Linlin; Scott, K. Andrea; Clausi, David A.

doi:https://doi.org/10.5194/tc-18-1621-2024

Articles | Volume 18, issue 4

https://doi.org/10.5194/tc-18-1621-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Special issue:

AutoICE: results of the sea ice classification challenge

https://doi.org/10.5194/tc-18-1621-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 18, issue 4

Research article

|

08 Apr 2024

Research article |

| 08 Apr 2024

MMSeaIce: a collection of techniques for improving sea ice mapping with a multi-task model

Xinwei Chen, Muhammed Patel, Fernando J. Pena Cantu, Jinman Park, Javier Noa Turnes, Linlin Xu, K. Andrea Scott, and David A. Clausi

Data sets

AI4Arctic Sea Ice Challenge Dataset J. Buus-Hinkler et al. https://doi.org/10.11583/DTU.c.6244065.v2

Short summary

This paper introduces an automated sea ice mapping pipeline utilizing a multi-task U-Net architecture. It attained the top score of 86.3 % in the AutoICE challenge. Ablation studies revealed that incorporating brightness temperature data and spatial–temporal information significantly enhanced model accuracy. Accurate sea ice mapping is vital for comprehending the Arctic environment and its global climate effects, underscoring the potential of deep learning.