Machine Learning in Large and Small Earthquakes: from Rapid Large Earthquake Characterization to Slow Fault Zone Processes
Bibliographic record
Abstract
This dissertation summarizes the work of integrating machine-learning and traditional seismic analysis techniques into large and small earthquake problems. Earthquake early warning for large magnitude earthquakes is one of the most challenging problems in seismology. Here I develop an algorithm, called M-LARGE, that harnesses machine-learning, rupture simulations, and GNSS data to rapidly predict magnitude without saturation issue with an accuracy of 99%, outperforming other similar methods. I then show how M-LARGE can predict finite fault parameters and their evolution when rupture unfolds for fast and accurate ground motion forecasting.This dissertation will demonstrate how machine-learning can be used as a data mining tool to detect small magnitude seismicity buried in noisy waveforms. I will show its application to detect LFEs, a special class of small earthquakes typically occur down-dip of the seismogenic zone. The model detects more than five times the number of events than the original catalog in Vancouver Island and can apply to unseen stations, which provides a more flexible way to refine the temporal resolution of subduction zone processes.\nFinally, I will show how do small and slow earthquakes link to large and fast events and their implication on earthquake hazard assessment. With jointly inverted GNSS, strong motion, and tsunami data of the 2018 M7.1 Hawaii earthquake, I find that fast slip ruptures into the area previously hosts slow slip. The result is further validated by rupture simulations, where we find that the effective stress can be a factor that exerts a dominant control on the rupture extent. This reinforces the idea that an individual section of fault can host a variety of distinct slip behaviors, and slow slip should be considered as rupture extent for a more accurate hazard assessment.\nThis dissertation includes previously published and unpublished co-authored material.
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 | 0.002 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| 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 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".