The Hunt for Missing Tornadoes: Using Satellite Imagery to Detect and Document Historical Tornado Damage in Canadian Forests
Bibliographic record
Abstract
Abstract Historical tornado events from 1982 to 2020 were documented within Canada’s forested regions using high-resolution satellite imagery. Tornado forest disturbances were identified using a three-step process: 1) detecting, 2) assessing, and 3) dating each event. A grid of 120 km × 120 km boxes was created covering Canada (excluding the extreme north). Of the 484 boxes, 367 were manually searched. Once a long, narrow region of tree damage was detected, it was first cross-referenced with known tornado databases to ensure it was a unique event. Once events were classified as either tornadic or downburst, the coordinates of the start, worst damage, and end locations were documented, as well as the direction of motion, damage indicators, degree of damage, estimated maximum wind speed, and F/EF-scale rating. In total, 231 previously unknown tornadoes were identified. In Ontario, 103 events were discovered, followed by 98 in Quebec, 9 in Manitoba, 6 in Saskatchewan, 9 in Alberta, 5 in British Columbia, and 1 in New Brunswick. The largest number of discovered tornadoes occurred in 2015, and the largest number of strong F2 tornadoes occurred in 2005. Most of the discovered tornadoes occurred in July for both F/EF1 and F/EF2 ratings. Most tornado tracks had widths between 200 and 400 m, and more than 50% of the tornadoes had a pathlength of less than 10 km. Of all the events that were discovered, 125 events could be fully dated, 19 were dated only by month, 41 were dated only by year, and 46 remained undated.
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) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.007 | 0.006 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 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".