Inferring fault structures and overburden depth in 3D from geophysical data using machine learning algorithms – A case study on the Fenelon gold deposit, Quebec, Canada
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Abstract We apply a machine learning approach to automatically infer two key attributes – the location of fault or shear zone structures and the thickness of the overburden – in an 18 km 2 study area within and surrounding the Archean Fenelon gold deposit in Quebec, Canada. Our approach involves the inversion of carefully curated borehole lithological and structural observations truncated at 480 m below the surface, combined with magnetic and Light Detection and Ranging survey data. We take a computationally low‐cost approach in which no underlying model for geological consistency is imposed. We investigated three contrasting approaches: (1) an inferred fault model, in which the borehole observations represent a direct evaluation of the presence of fault or shear zones; (2) an inferred overburden model, using borehole observations on the overburden‐bedrock contact; (3) a model with three classes – overburden, faulted bedrock and unfaulted bedrock, which combines aspects of (1) and (2). In every case, we applied all 32 standard machine learning algorithms. We found that Bagged Trees, fine K ‐nearest neighbours and weighted K ‐nearest neighbour were the most successful, producing similar accuracy, sensitivity and specificity metrics. The Bagged Trees algorithm predicted fault locations with approximately 80% accuracy, 70% sensitivity and 73% specificity. Overburden thickness was predicted with 99% accuracy, 77% sensitivity and 93% specificity. Qualitatively, fault location predictions compared well to independently construct geological interpretations. Similar methods might be applicable in other areas with good borehole coverage, providing that criteria used in borehole logging are closely followed in devising classifications for the machine learning training set and might be usefully supplemented with a variety of geophysical survey data types.
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.
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 | 0.000 | 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 | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it