Improving interpretation of sea-level projections through a machine-learning-based local explanation approach
<p>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 int...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2022-11-01
|
| Series: | The Cryosphere |
| Online Access: | https://tc.copernicus.org/articles/16/4637/2022/tc-16-4637-2022.pdf |
| _version_ | 1798045547913805824 |
|---|---|
| author | J. Rohmer R. Thieblemont G. Le Cozannet H. Goelzer G. Durand |
| author_facet | J. Rohmer R. Thieblemont G. Le Cozannet H. Goelzer G. Durand |
| author_sort | J. Rohmer |
| collection | DOAJ |
| description | <p>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.</p> |
| first_indexed | 2024-04-11T23:22:43Z |
| format | Article |
| id | doaj.art-f55168a8b86f495b9b90ea928d0f0c04 |
| institution | Directory Open Access Journal |
| issn | 1994-0416 1994-0424 |
| language | English |
| last_indexed | 2024-04-11T23:22:43Z |
| publishDate | 2022-11-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | The Cryosphere |
| spelling | doaj.art-f55168a8b86f495b9b90ea928d0f0c042022-12-22T03:57:25ZengCopernicus PublicationsThe Cryosphere1994-04161994-04242022-11-01164637465710.5194/tc-16-4637-2022Improving interpretation of sea-level projections through a machine-learning-based local explanation approachJ. Rohmer0R. Thieblemont1G. Le Cozannet2H. Goelzer3G. Durand4BRGM, 3 av. C. Guillemin, 45060 Orléans, FranceBRGM, 3 av. C. Guillemin, 45060 Orléans, FranceBRGM, 3 av. C. Guillemin, 45060 Orléans, FranceNORCE Norwegian Research Centre, Bjerknes Centre for Climate Research, Bergen, NorwayUniv. Grenoble Alpes, CNRS, IRD, Grenoble INP, IGE, 38000 Grenoble, France<p>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.</p>https://tc.copernicus.org/articles/16/4637/2022/tc-16-4637-2022.pdf |
| spellingShingle | J. Rohmer R. Thieblemont G. Le Cozannet H. Goelzer G. Durand Improving interpretation of sea-level projections through a machine-learning-based local explanation approach The Cryosphere |
| title | 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_short | 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 |
| url | https://tc.copernicus.org/articles/16/4637/2022/tc-16-4637-2022.pdf |
| work_keys_str_mv | AT jrohmer improvinginterpretationofsealevelprojectionsthroughamachinelearningbasedlocalexplanationapproach AT rthieblemont improvinginterpretationofsealevelprojectionsthroughamachinelearningbasedlocalexplanationapproach AT glecozannet improvinginterpretationofsealevelprojectionsthroughamachinelearningbasedlocalexplanationapproach AT hgoelzer improvinginterpretationofsealevelprojectionsthroughamachinelearningbasedlocalexplanationapproach AT gdurand improvinginterpretationofsealevelprojectionsthroughamachinelearningbasedlocalexplanationapproach |