Evaluation of selected CMIP6 models' simulation on Arctic sea ice
Bibliographic record
Abstract
To evaluate whether CMIP6 models provide good simulation in Arctic sea-ice extent, thickness, and motion, selected 6 CMIP6 models are EC-Earth3, ACCESS-CM2, BCC-CSM2-MR, GFDL-ESM4, MPI-ESM1-2-HR, NORESM2-LM. For CMIP6 models and observations, seasonal cycle and the annual variation from 1979-2014 of sea-ice extent were studied, for sea-ice thickness and sea-ice motion, the Arctic is separated into three regions, geographical distribution, inter-annual variation from 1979-2014, seasonal cycle, and trend were studied. Then student t-test is used to evaluate whether the model output has a significant difference from observation, to select the best model(s). For sea-ice extent, EC-Earth3 is overestimating sea-ice extent, especially in winter, BCC-CSM2-MR model underestimates sea-ice extent, ACCESS-CM2, MPI-ESM1-2-HR, NorESM2-LM models perform the best. For sea-ice thickness, BCC-CSM2-MR underestimates sea-ice thickness, EC-Earth3, ACCESS-CM2, and NORESM2-LM models are overestimating sea-ice thickness. GFDL-ESM4 and MPI-ESM1-2-HR have the best performance at sea-ice thickness simulation. For sea-ice motion, the MPI-ESM1-2-HR model overestimates sea-ice drifting speed all year round, ACCESS-CM2 model tends to overestimate sea-ice drifting speed in summer for region1 and region2, in region3 ACCESS-CM2 model mostly overestimate sea-ice motion except winter months. NorESM2-LM model has the best performance overall, and ACCESS-CM2 has the second-best simulation for region1 and region2. EC-Earth3 also has a satisfactory simulation for sea-ice motion. Models and observation also agree on common results for sea-ice properties: Maximum sea-ice extent occurs in March, and minimum sea-ice extent occurs in September. There's a decreasing trend of sea-ice extent. The Central Arctic and Canadian Archipelago always have the thickest sea ice, followed by the East Siberian Sea, Laptev Sea, and Chukchi Sea, Beaufort Sea. East Greenland Sea, Barents Sea, Buffin Bay, and the Kara Sea always have the thinnest sea ice. There's a decreasing trend for sea-ice thickness according to models, sea-ice is thicker in the Chukchi Sea and the Beaufort Sea than in Laptev and East Siberian seas. Winter sea-ice thickness is higher than in summer, and sea-ice thickness has a more rapid decreasing rate in summer than in winter. Laptev and the East Siberian Sea have the most rapidly sea-ice thinning process. Sea-ice thickness has seasonal cycle that maximum usually occurs in May, and minimum sea-ice thickness happens in October. For sea-ice motion, there's an increasing trend of sea-ice motion, and summer sea-ice motion has faster sea-ice motion than winter, Chukchi Sea, and the Beaufort Sea has faster sea-ice motion than Laptev and the East Siberian Sea. Corresponding with the comparatively faster-thinning in the Laptev and the East Siberian Seas simulated by models, there's also a faster increasing rate in the Laptev and the East Siberian Sea.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.007 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".