Seismic Exploration Solutions for Deep-Targeting Metallic Mineral Deposits : From high-fold 2D to sparse 3D, and deep-learning workflows
Bibliographic record
Abstract
Mineral exploration has in recent years moved its focus to greater depths than ever before, particularly in brown fields. Exploring new deposits at depth, if economical, would not only expand the life of mine but also provide minimal environmental impacts. It allows the existing mining infrastructures to be used for a longer period. Exploration at depth, however, is challenging and requires a multidisciplinary team and methods, and innovative thinking for generating new targets and effective exploration expenditure. The application of seismic methods for mineral exploration has increasingly been conducted over the past 20 years because they provide high-resolution subsurface images, and retain good resolution with depth as compared with other geophysical methods. Nevertheless, and despite challenges in hardrock settings, only limited attention has been given to seismic interpretations, often performed subjectively. With the growing application of machine-learning solutions, hardrock seismic data can benefit these for improved interpretations and target generations. This thesis showcases different workflows developed for deep-targeting metallic mineral deposits, starting from high-fold 2D, through sparse 3D reflection imaging and the implementation of deep-learning algorithms for diffraction pattern recognitions. Three different deposits were studied from Sweden and Canada. The Blötberget iron-oxide mineralization in central Sweden was first targeted in 2D, followed-up, a sparse 3D dataset was acquired enabling to image the mineralization both laterally and with depth, providing good knowledge on subsurface structures controlling the geometry of the deposits. In Canada, Halfmile Lake and Matagami mining sites were studied due to the accessibility to 3D seismic datasets, which contained diffraction signals as deposit responses. Deeplearning algorithms were utilized for the proof-of-concept and at the same time helped to generate new potential targets from other diffraction signals that were not obvious to an interpreter’s eye due to their incomplete tails originated outside of the seismic volume. The studies in this thesis show the effectiveness of seismic methods for mineral exploration at depth, especially in 3D, as they provide, among others, structural interpretation for future mineplanning purposes. Deep-learning solutions provide improved results for diffraction delineation and denoising and have great potential for hardrock seismics.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".