Improving interpretation of sea-level projections through a machine-learning-based local explanation approach
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 interpretabi...
Published in: | The Cryosphere |
---|---|
Main Authors: | , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
2022
|
Subjects: | |
Online Access: | https://hdl.handle.net/11250/3090671 https://doi.org/10.5194/tc-16-4637-2022 |
id |
ftnorce:oai:norceresearch.brage.unit.no:11250/3090671 |
---|---|
record_format |
openpolar |
spelling |
ftnorce:oai:norceresearch.brage.unit.no:11250/3090671 2023-10-25T01:39:04+02:00 Improving interpretation of sea-level projections through a machine-learning-based local explanation approach Rohmer, Jeremy Thiéblemont, Rémi Le Cozannet, G. Goelzer, Heiko Durand, Gaël 2022 application/pdf https://hdl.handle.net/11250/3090671 https://doi.org/10.5194/tc-16-4637-2022 eng eng Norges forskningsråd: 324639 Sigma2: NS9560K Sigma2: NN8006K EC/H2020/869304 Sigma2: NS5011K Sigma2: NN8085K Sigma2: NS9252K Sigma2: NS8006K Sigma2: NS8085K The Cryosphere. 2022, 16 (11), 4637-4657. urn:issn:1994-0416 https://hdl.handle.net/11250/3090671 https://doi.org/10.5194/tc-16-4637-2022 cristin:2069309 Navngivelse 4.0 Internasjonal http://creativecommons.org/licenses/by/4.0/deed.no © Author(s) 2022 The Cryosphere 16 11 4637-4657 Havnivå Sea level VDP::Geofag: 450 VDP::Geosciences: 450 Peer reviewed Journal article 2022 ftnorce https://doi.org/10.5194/tc-16-4637-2022 2023-09-27T22:49:50Z 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 known as “SHAP” (SHapley Additive exPlanations). 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) numerical implementation, (2) initial conditions, (3) modelling of ice-sheet processes, and (4) environmental forcing. 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. publishedVersion Article in Journal/Newspaper Greenland Ice Sheet The Cryosphere NORCE vitenarkiv (Norwegian Research Centre) Greenland The Cryosphere 16 11 4637 4657 |
institution |
Open Polar |
collection |
NORCE vitenarkiv (Norwegian Research Centre) |
op_collection_id |
ftnorce |
language |
English |
topic |
Havnivå Sea level VDP::Geofag: 450 VDP::Geosciences: 450 |
spellingShingle |
Havnivå Sea level VDP::Geofag: 450 VDP::Geosciences: 450 Rohmer, Jeremy Thiéblemont, Rémi Le Cozannet, G. Goelzer, Heiko Durand, Gaël Improving interpretation of sea-level projections through a machine-learning-based local explanation approach |
topic_facet |
Havnivå Sea level VDP::Geofag: 450 VDP::Geosciences: 450 |
description |
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 known as “SHAP” (SHapley Additive exPlanations). 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) numerical implementation, (2) initial conditions, (3) modelling of ice-sheet processes, and (4) environmental forcing. 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. publishedVersion |
format |
Article in Journal/Newspaper |
author |
Rohmer, Jeremy Thiéblemont, Rémi Le Cozannet, G. Goelzer, Heiko Durand, Gaël |
author_facet |
Rohmer, Jeremy Thiéblemont, Rémi Le Cozannet, G. Goelzer, Heiko Durand, Gaël |
author_sort |
Rohmer, Jeremy |
title |
Improving interpretation of sea-level projections through a machine-learning-based local explanation approach |
title_short |
Improving interpretation of sea-level projections through a machine-learning-based local explanation approach |
title_full |
Improving interpretation of sea-level projections through a machine-learning-based local explanation approach |
title_fullStr |
Improving interpretation of sea-level projections through a machine-learning-based local explanation approach |
title_full_unstemmed |
Improving interpretation of sea-level projections through a machine-learning-based local explanation approach |
title_sort |
improving interpretation of sea-level projections through a machine-learning-based local explanation approach |
publishDate |
2022 |
url |
https://hdl.handle.net/11250/3090671 https://doi.org/10.5194/tc-16-4637-2022 |
geographic |
Greenland |
geographic_facet |
Greenland |
genre |
Greenland Ice Sheet The Cryosphere |
genre_facet |
Greenland Ice Sheet The Cryosphere |
op_source |
The Cryosphere 16 11 4637-4657 |
op_relation |
Norges forskningsråd: 324639 Sigma2: NS9560K Sigma2: NN8006K EC/H2020/869304 Sigma2: NS5011K Sigma2: NN8085K Sigma2: NS9252K Sigma2: NS8006K Sigma2: NS8085K The Cryosphere. 2022, 16 (11), 4637-4657. urn:issn:1994-0416 https://hdl.handle.net/11250/3090671 https://doi.org/10.5194/tc-16-4637-2022 cristin:2069309 |
op_rights |
Navngivelse 4.0 Internasjonal http://creativecommons.org/licenses/by/4.0/deed.no © Author(s) 2022 |
op_doi |
https://doi.org/10.5194/tc-16-4637-2022 |
container_title |
The Cryosphere |
container_volume |
16 |
container_issue |
11 |
container_start_page |
4637 |
op_container_end_page |
4657 |
_version_ |
1780734244096573440 |