Research library
Weather forecasting and climate applications: completion history
All area pages | Area definition | Scoring rubric
Created November 7, 2026. The three earlier snapshots are retrospective judgment-based estimates, not scores recorded at the time. The current assessment is unchanged. Historical confidence is very low; current confidence is low. Zero can indicate an evidence gap rather than absent capability.
Chart layout updated November 8, 2026: one graph with four dated lines. Scores and assessment dates are unchanged. Overlapping lines indicate equal values; point labels show November 7, 2026. The table below contains every date's exact values.
Scalable graph | Combined 14-page report
Percentage estimates
| Measurable outcome | Nov 7, 2025 (reconstructed) | May 7, 2026 (reconstructed) | Oct 7, 2026 (reconstructed) | Nov 7, 2026 (saved) |
|---|---|---|---|---|
| Climate application validity | 35% | 35% | 35% | 35% |
| Computation | 45% | 45% | 45% | 45% |
| Decision benefit | 15% | 15% | 15% | 15% |
| Extreme-event performance | 40% | 40% | 40% | 40% |
| Forecast skill | 40% | 40% | 45% | 45% |
| Robustness | 20% | 20% | 20% | 20% |
Reconstruction notes and component scores
Components below are execution / breadth / agenda-setting / validation / sustained autonomous improvement, each out of 20. Their sum gives the percentage. The component values are judgments, not published statistics. Source links document evidence available by the cutoff, not the numerical estimate itself. The source section lists availability dates and version caveats. No area-wide average is calculated.
Climate application validity
The atmospheric climate-statistics evidence predates all snapshots. Future-forcing and coupled-system validity remain unresolved.
| Assessment date | Component points | Evidence used |
|---|---|---|
| 2025-11-07 (reconstructed) | 10 / 5 / 5 / 10 / 5 = 35% | gcmpaper |
| 2026-05-07 (reconstructed) | 10 / 5 / 5 / 10 / 5 = 35% | gcmpaper |
| 2026-10-07 (reconstructed) | 10 / 5 / 5 / 10 / 5 = 35% | gcmpaper |
| 2026-11-07 (saved) | 10 / 5 / 5 / 10 / 5 = 35% | Original score and evidence |
Computation
Fast inference and operational forecasting were already available in 2025. No new full-pipeline energy/cost measurement justifies an extra increment.
| Assessment date | Component points | Evidence used |
|---|---|---|
| 2025-11-07 (reconstructed) | 15 / 10 / 5 / 10 / 5 = 45% | genblog, aifs25 |
| 2026-05-07 (reconstructed) | 15 / 10 / 5 / 10 / 5 = 45% | genblog, aifs25 |
| 2026-10-07 (reconstructed) | 15 / 10 / 5 / 10 / 5 = 45% | genblog, aifs25 |
| 2026-11-07 (saved) | 15 / 10 / 5 / 10 / 5 = 45% | Original score and evidence |
Decision benefit
The wind-power prediction application is available throughout, but realized operational decision savings remain unestablished.
| Assessment date | Component points | Evidence used |
|---|---|---|
| 2025-11-07 (reconstructed) | 5 / 5 / 0 / 5 / 0 = 15% | gen |
| 2026-05-07 (reconstructed) | 5 / 5 / 0 / 5 / 0 = 15% | gen |
| 2026-10-07 (reconstructed) | 5 / 5 / 0 / 5 / 0 = 15% | gen |
| 2026-11-07 (saved) | 5 / 5 / 0 / 5 / 0 = 15% | Original score and evidence |
Extreme-event performance
The same historical event evaluation supports all snapshots; no fresh prospective event series is substituted for it.
| Assessment date | Component points | Evidence used |
|---|---|---|
| 2025-11-07 (reconstructed) | 10 / 10 / 5 / 10 / 5 = 40% | gen |
| 2026-05-07 (reconstructed) | 10 / 10 / 5 / 10 / 5 = 40% | gen |
| 2026-10-07 (reconstructed) | 10 / 10 / 5 / 10 / 5 = 40% | gen |
| 2026-11-07 (saved) | 10 / 10 / 5 / 10 / 5 = 40% | Original score and evidence |
Forecast skill
Initial deployment supports strong bounded execution. May 12 version progression supports limited repeated-improvement credit only after the May 7 cutoff, not autonomous research management.
| Assessment date | Component points | Evidence used |
|---|---|---|
| 2025-11-07 (reconstructed) | 15 / 10 / 5 / 10 / 0 = 40% | gen, aifs25 |
| 2026-05-07 (reconstructed) | 15 / 10 / 5 / 10 / 0 = 40% | gen, aifs25 |
| 2026-10-07 (reconstructed) | 15 / 10 / 5 / 10 / 5 = 45% | gen, aifs |
| 2026-11-07 (saved) | 15 / 10 / 5 / 10 / 5 = 45% | Original score and evidence |
Robustness
Earlier global evaluation supports limited robustness; the May initialization issue reinforces existing limits rather than earning more credit.
| Assessment date | Component points | Evidence used |
|---|---|---|
| 2025-11-07 (reconstructed) | 5 / 5 / 5 / 5 / 0 = 20% | gen, aifs25 |
| 2026-05-07 (reconstructed) | 5 / 5 / 5 / 5 / 0 = 20% | gen, aifs25 |
| 2026-10-07 (reconstructed) | 5 / 5 / 5 / 5 / 0 = 20% | farewell |
| 2026-11-07 (saved) | 5 / 5 / 5 / 5 / 0 = 20% | Original score and evidence |
Dated sources
All sources accessed November 7, 2026. Dates below are publication or conservative availability dates, as noted.
gcmpaper | 2024-07-22 - Neural general circulation models, Nature. Hybrid atmospheric modeling; prescribed ocean boundaries and resolution differences limit comparison.
gen | 2024-12-04 - GenCast, Nature. Historical probabilistic forecasting evaluation; no claim of realized decision savings.
genblog | 2024-12-04 - DeepMind: GenCast runtime and evaluation. Per-member inference runtime; full operational pipeline cost is separate.
aifs25 | 2025-02-25 - ECMWF AIFS operational introduction. Operational deployment with continuing human-managed data and research pipelines.
farewell | 2026-05-11 - ECMWF: farewell to external AI models. Operational input changes expose model robustness issues after the May 7 cutoff.
aifs | 2026-05-12 - ECMWF AIFS version history. AIFS version 2 operation; not available at the May 7 cutoff.
Future updates
Keep these reconstructions separate from subsequently recorded observations. Append new dated assessments with changed component scores and source evidence; do not silently replace historical reasoning. The shared historical-method notes describe how these snapshots were made.