Preprints
https://doi.org/10.5194/tc-2021-396
https://doi.org/10.5194/tc-2021-396

  12 Jan 2022

12 Jan 2022

Review status: this preprint is currently under review for the journal TC.

Predicting ocean-induced ice-shelf melt rates using a machine learning image segmentation approach

Sebastian Harry Reid Rosier1,2, Christopher Y. S. Bull1, and G. Hilmar Gudmundsson1 Sebastian Harry Reid Rosier et al.
  • 1Department of Geography and Environmental Sciences, Northumbria University, Newcastle Upon Tyne, UK
  • 2WSL Institute for Snow and Avalanche Research SLF, Davos, Switzerland

Abstract. Through their role in buttressing upstream ice flow, Antarctic ice shelves play an important part in regulating future sea level change. Reduction in ice-shelf buttressing caused by increased ocean-induced melt along their undersides is now understood to be one of the key drivers of ice loss from the Antarctic Ice Sheet. However, despite the importance of this forcing mechanism most ice-sheet simulations currently rely on simple melt-parametrisations of this ocean-driven process, since a fully coupled ice-ocean modelling framework is prohibitively computationally expensive. Here, we provide an alternative approach that is able to capture the greatly improved physical description of this process provided by large-scale ocean-circulation models over currently employed melt-parameterisations but with trivial computational expense. We introduce a new approach that brings together deep learning and physical modelling to develop a deep neural network framework, MELTNET, that can emulate ocean model predictions of sub-ice shelf melt rates. We train MELTNET on synthetic geometries, using the NEMO ocean model as a ground-truth in lieu of observations to provide melt rates both for training and to evaluate the performance of the trained network. We show that MELTNET can accurately predict melt rates for a wide range of complex synthetic geometries and outperforms more traditional parameterisations for > 95 % of geometries tested. Furthermore, we find MELTNET's melt rate estimates show sensitivity to established physical relationships such as a changes in thermal forcing and ice shelf slope. This study demonstrates the potential for a deep learning framework to calculate melt rates with almost no computational expense, that could in the future be used in conjunction with an ice sheet model to provide predictions for large-scale ice sheet models.

Sebastian Harry Reid Rosier et al.

Status: open (until 09 Mar 2022)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Sebastian Harry Reid Rosier et al.

Sebastian Harry Reid Rosier et al.

Viewed

Total article views: 253 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
179 70 4 253 2 2
  • HTML: 179
  • PDF: 70
  • XML: 4
  • Total: 253
  • BibTeX: 2
  • EndNote: 2
Views and downloads (calculated since 12 Jan 2022)
Cumulative views and downloads (calculated since 12 Jan 2022)

Viewed (geographical distribution)

Total article views: 249 (including HTML, PDF, and XML) Thereof 249 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Discussed

Latest update: 16 Jan 2022
Download
Short summary
Future ice loss from Antarctica could raise sea levels by several metres, and key to this is the rate at which the ocean melts the ice sheet from below. Existing methods for modelling this process are either too time-consuming or very simplified. We present a new approach, using machine learning to mimic the melt rates calculated by a full ocean model but in a fraction of the time. This could replace to existing methods, providing accurate and efficient melt rate for use in an ice sheet model.