MétaCan
Menu
Back to cohort
Record W4415108044 · doi:10.1175/aies-d-24-0107.1

Adjustment of Decadal Ocean Carbon Sink Predictions Using Deep Learning

2025· article· en· W4415108044 on OpenAlexaffabout
Reinel Sospedra‐Alfonso, Parsa Gooya, Johannes Exenberger

Bibliographic record

VenueArtificial Intelligence for the Earth Systems · 2025
Typearticle
Languageen
FieldEnvironmental Science
TopicAtmospheric and Environmental Gas Dynamics
Canadian institutionsEnvironment and Climate Change Canada
Fundersnot available
KeywordsMean squared errorArtificial neural networkAnomaly (physics)Correlation coefficientCarbon sinkForecast skillClimate modelFlexibility (engineering)

Abstract

fetched live from OpenAlex

Abstract We implement an encoder–decoder artificial neural network (ANN) to postprocess decadal predictions of gridded air–sea carbon dioxide (CO 2 ) flux produced with an Earth system model (ESM) having prescribed CO 2 emissions. Decadal predictions are initialized by constraining the ESM with observational data and drift toward the model unconstrained climatology. The resulting biased forecasts require adjustments to make the predictions usable. By leveraging the flexibility of the ANN to learn the complex nonlinear relationship between raw forecasts and verifying data, and its ability to learn from nonlocal spatial errors, we show that the ANN-based adjustment outperforms standard bias and linear trend corrections for the first 3 years in the forecast, both in terms of spatial and temporal accuracy. The squared root of the globally integrated mean square error is approximately 0.5 PgC yr −1 for the first year of the forecast, which is about half the value from the alternative methods, and remains well below 1.0 PgC yr −1 for the first 5 years. The globally averaged anomaly correlation coefficient is above 0.5 and remains above 0.3 for the first 4 years of the forecast, whereas the alternative corrections score about 0.3 or less at all lead years. The methodology is tested with emission-driven decadal predictions produced with the Canadian Centre for Climate Modelling and Analysis forecasting system, which contribute to the Global Carbon Budget annual update. Significance Statement Deep learning has been used for postprocessing numerical weather predictions, but applications to longer-range forecasts are less common. Forecast postprocessing is a necessary step toward making predictions usable, since climate models are imperfect representations of the climate system and the forecast initial conditions are often erroneous. A deep learning (DL)-based model is implemented to adjust near-term predictions of ocean carbon sinks produced with Fifth Generation Canadian ESM (CanESM5), which couples a physics-based climate model with an interactive carbon cycle. These predictions contribute to the Global Carbon Budget annual update that tracks the distribution of global carbon emissions and sinks to inform progress toward the goals of the Paris Agreement. The DL-based postprocessing model outperforms standard correction methods for several years in the forecast, leading to more accurate near-term predictions of the ocean carbon sink.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.584
Threshold uncertainty score0.453

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.022
GPT teacher head0.255
Teacher spread0.232 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes2
Has abstractyes

Explore more

Same venueArtificial Intelligence for the Earth SystemsSame topicAtmospheric and Environmental Gas DynamicsFrench-language works237,207