Preprints
https://doi.org/10.5194/egusphere-2022-435
https://doi.org/10.5194/egusphere-2022-435
 
 | Highlight paper
16 Jun 2022
16 Jun 2022

Improving interpretation of sea-level projections through a machine-learning-based local explanation approach

Jeremy Rohmer1, Remi Thieblemont1, Goneri Le Cozannet1, Heiko Goelzer2, and Gael Durand3 Jeremy Rohmer et al.
  • 1BRGM, 3 av. C. Guillemin, 45060, Orléans, France
  • 2NORCE Norwegian Research Centre, Bjerknes Centre for Climate Research, Bergen, Norway
  • 3IGE, University Grenoble Alpes, Grenoble, France

Abstract. Process-based projections of the sea-level contribution from land ice components are often obtained from simulations using a complex chain of numerical models. Because of their importance in supporting the decision-making process for coastal risk assessment and adaptation, improving the interpretability of these projections is of great interest. To this end, we adopt the local attribution approach developed in the machine learning community by combining the game-theoretic approach known as ‘SHAP’ (SHapley Additive exPlanation) with machine-learning regression models. We apply our methodology to a subset of the multi-model ensemble study of the future contribution of the Greenland ice sheet to sea-level, taking into account different modelling choices related to (1) the numerical implementation, (2) the initial conditions, and (3) the modelling of ice-sheet processes. This allows us to quantify the influence of particular modelling decisions, which is directly expressed in terms of sea level change contribution. This type of diagnosis can be performed on any member of the ensemble, and we show in the Greenland case how the aggregation of the local attribution analyses can help guide future model development as well as scientific interpretation, particularly with regard to spatial model resolution and to retreat parametrisation.

Journal article(s) based on this preprint

04 Nov 2022
| Highlight paper
Improving interpretation of sea-level projections through a machine-learning-based local explanation approach
Jeremy Rohmer, Remi Thieblemont, Goneri Le Cozannet, Heiko Goelzer, and Gael Durand
The Cryosphere, 16, 4637–4657, https://doi.org/10.5194/tc-16-4637-2022,https://doi.org/10.5194/tc-16-4637-2022, 2022
Short summary

Jeremy Rohmer et al.

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2022-435', Anonymous Referee #1, 27 Jul 2022
    • AC1: 'Reply on RC1', Jeremy Rohmer, 12 Sep 2022
  • RC2: 'Comment on egusphere-2022-435', Anonymous Referee #2, 29 Jul 2022
    • AC2: 'Reply on RC2', Jeremy Rohmer, 12 Sep 2022

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2022-435', Anonymous Referee #1, 27 Jul 2022
    • AC1: 'Reply on RC1', Jeremy Rohmer, 12 Sep 2022
  • RC2: 'Comment on egusphere-2022-435', Anonymous Referee #2, 29 Jul 2022
    • AC2: 'Reply on RC2', Jeremy Rohmer, 12 Sep 2022

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision
ED: Publish subject to minor revisions (review by editor) (20 Sep 2022) by Ginny Catania
AR by Jeremy Rohmer on behalf of the Authors (28 Sep 2022)  Author's response    Author's tracked changes    Manuscript
ED: Publish subject to technical corrections (30 Sep 2022) by Ginny Catania
AR by Jeremy Rohmer on behalf of the Authors (07 Oct 2022)  Author's response    Manuscript

Post-review adjustments

AA: Author's adjustment | EA: Editor approval
AA by Jeremy Rohmer on behalf of the Authors (27 Oct 2022)   Author's adjustment   Manuscript
EA: Adjustments approved (28 Oct 2022) by Ginny Catania

Journal article(s) based on this preprint

04 Nov 2022
| Highlight paper
Improving interpretation of sea-level projections through a machine-learning-based local explanation approach
Jeremy Rohmer, Remi Thieblemont, Goneri Le Cozannet, Heiko Goelzer, and Gael Durand
The Cryosphere, 16, 4637–4657, https://doi.org/10.5194/tc-16-4637-2022,https://doi.org/10.5194/tc-16-4637-2022, 2022
Short summary

Jeremy Rohmer et al.

Jeremy Rohmer et al.

Viewed

Total article views: 348 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
244 91 13 348 2 2
  • HTML: 244
  • PDF: 91
  • XML: 13
  • Total: 348
  • BibTeX: 2
  • EndNote: 2
Views and downloads (calculated since 16 Jun 2022)
Cumulative views and downloads (calculated since 16 Jun 2022)

Viewed (geographical distribution)

Total article views: 256 (including HTML, PDF, and XML) Thereof 256 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 04 Nov 2022
Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

This manuscript addresses an urgent problem: the proper quantification and attribution of uncertainties relating to sea-level rise. The authors show how a machine-learning approach may show the way towards a more rigorous treatment of these uncertainties, and how this might be used for policy making.
Short summary
To improve the interpretability of process-based projections of the sea-level contribution from land ice components, we apply the machine-learning-based ‘SHapley Additive exPlanation’ approach to a subset of the multi-model ensemble study for Greenland ice sheet. This allows us to quantify the influence of particular modelling decisions (either related to numerical implementation, initial conditions, or parametrisation of ice-sheet processes) directly in terms of sea level change contribution.