ContrailBench: evaluating the performance of contrail models
Bibliographic record
Abstract
The development of effective contrail warming mitigation strategies requires the ability to accurately model contrail formation. This is a challenging problem for a number of reasons, including high uncertainty in the humidity data which is a key component of such models. Observational datasets can be used to constrain and improve contrail formation models. Here we present an analysis of existing contrail models on the task of predicting whether a contrail will be observed in a collection of observational datasets (collectively termed 'ContrailBench'). The observational datasets include one based on Ref [1] using automated contrail detections from the GOES-16 satellite and an automated contrail attribution algorithm, another based on Ref [2] which detects contrails using GOES-16 and uses knowledge of the altitude from LIDAR measurements to attribute them to flights, and a third dataset based on Global Meteor Network ground-based camera imagery [3] with automated contrail detection and high-confidence attribution to the flights that formed them. Different downstream applications require different properties from contrail models, so we evaluate the contrail models based on their performance in both ‘high-recall’ mode (which prioritizes identifying all the flights which make contrails) as well as in ‘high-precision’ mode (which prioritizes minimizing the number of flights incorrectly predicted as forming a contrail). We find that models using raw ERA5 weather reanalysis data perform poorly on all metrics, but the use of machine learning to correct the weather data can lead to improvement. [1] A. Sarna et al, “Benchmarking and improving algorithms for attributing satellite-observed contrails to flights”, https://doi.org/10.5194/egusphere-2024-3664[2] V. Meijer, thesis, “Satellite-based Analysis and Forecast Evaluation of Aviation Contrails”[3] D. Vida et al, “The Global Meteor Network – Methodology and first results” https://doi.org/10.1093/mnras/stab2008
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 | 0.016 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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".