Delineating giant Antarctic Icebergs with Deep Learning

• Main Authors: Braakmann-Folgmann, A., Shepherd, A., Hogg, D., Redmond, E.
• Format: Conference Object
• Language: English
• Published: 2023
• Subjects: Antarctic; Antarc*; Antarctica; Iceberg*; Sea ice
• Online Access: https://gfzpublic.gfz-potsdam.de/pubman/item/item_5019231

Icebergs account for half of all ice loss from Antarctica. Their melting affects the surrounding ocean properties through the intrusion of cold, fresh meltwater and the release of terrigenious nutrients. This in turn influences the local ocean circulation, sea ice formation and biological production. To locate and quantify the fresh water flux from Antarctic icebergs, we need to track them and monitor changes in their area and thickness. While the locations of large icebergs are tracked operationally by manual inspection, delineation of iceberg extent requires detailed analysis – either also manually or through automated segmentation of high resolution satellite imagery. In this study, we propose a U-net approach to automatically segment giant icebergs in nearly 200 Sentinel-1 images. It is the first study to apply a deep learning algorithm to iceberg segmentation. Furthermore, most previous studies to detect icebergs have focused on smaller bergs. In contrast, we aim to segment selected giant icebergs with the goal to automate the calculation of fresh water input. We compare the performance of our neural network to two standard segmentation algorithms. Only on the largest icebergs, these perform better, as U-net tends to miss parts. In contrast, U-net is more robust to busy backgrounds like sea ice. It is also better at ignoring small patches of nearby coast or other icebergs. Dark icebergs remain a problem for all techniques. Overall, U-net outperforms the other two techniques, achieving an F1-score of 0.84 and an absolute median deviation in iceberg area of 4.1 %.