A climatology of trapped lee waves over Britain and Ireland obtained using deep learning on high‐resolution model output
Bibliographic record
Abstract
ABSTRACT This article presents a climatology of trapped lee waves over Britain and Ireland obtained through deep learning. Several deep‐learning models trained to diagnose lee‐wave occurrence, amplitude, wavelength, and orientation are applied to a 31‐year high‐resolution hindcast dataset covering 1982–2012, from UK Climate Projections (UKCP18) data, driven by ERA‐Interim reanalysis data. Building on previous work to examine lee‐wave characteristics over Britain and Ireland, this study applies a new technique to a much larger dataset than has been used in the past. There is little diurnal variability observed in the occurrence and characteristics of lee waves. Spatially, most lee waves occur over hilly regions, such as the Scottish Highlands, the Lake District and the Pennines in England, and North Wales. Seasonally, lee waves occur more in the winter months than in the summer. The link between synoptic weather patterns and lee waves is quantified, with more lee waves produced and a higher likelihood of higher amplitude waves under patterns with faster synoptic wind speeds, such as the positive phase of the North Atlantic Oscillation (NAO+). The mean orientation of waves is broadly in line with the synoptic wind direction, though with a large spread in some cases. High horizontal wind speeds aloft are a necessary but not sufficient indicator of high‐amplitude lee waves. When other meteorological variables are used to predict the prevalence of lee waves using a random forest, the Scorer parameter is the most important for predicting the generation of lee waves alongside horizontal wind speed: there is less importance placed on the stability. This climatology provides a novel data‐driven insight into the formation and propagation of lee waves over Britain and Ireland.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 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".