Using the hybrid gain algorithm to sample data assimilation uncertainty
Bibliographic record
Abstract
At the Canadian Meteorological Centre (CMC), an ensemble variational (EnVar) data assimilation system is used for the global deterministic prediction system and an ensemble Kalman filter (EnKF) is used for the global ensemble prediction system. These two systems are co‐developed and co‐evolving at the CMC and in this study we explore how to maximize the impact of having two algorithms. Following earlier work at the European Centre for Medium‐Range Weather Forecasts (ECMWF), we perform experiments with a pure EnKF and an EnKF system that is recentered on the EnVar solution, as well as with a hybrid gain configuration, in which the EnKF system is recentered on the mean of the EnKF and EnVar analyses. Encouraged by the results of the hybrid gain algorithm, we modify it to leave half of the members unchanged and to recenter the other half on the EnVar analysis. With this multi‐analysis approach, we sample the different design decisions made for the EnKF and EnVar and see corresponding improvements, notably for the stratospheric analysis. An evaluation using humidity‐sensitive radiance channels shows more mixed results of the hybrid gain and multi‐analysis approaches. An investigation of the spread–skill relation showed that the background ensembles were overdispersive for humidity and this issue was resolved by a reduction of the additive error for humidity. This highlights the fact that diagnostic information from two analysis systems can be used to identify where those systems have room for improvement. Finally, for various aspects of data assimilation systems, we weigh the benefits of algorithmic diversity against the corresponding additional development cost.
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 imitationNot 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.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
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.
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 teacher head, 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".