Automated Satellite Detection of Tornado Forest Damage in Canada
Bibliographic record
Abstract
The Northern Tornadoes Project (NTP), a division of the Canadian Severe Storms Laboratory (CSSL), has the goal of detecting every tornado that occurs in Canada. However, due to Canada’s vast area, often the only significant damage caused by a tornado occurs in remote forested regions that may go unreported. In order to detect these remote, unreported tornadoes, researchers at the NTP perform an end-of-season systematic sweep of Canada using satellite imagery. By comparing satellite imagery between different dates, large swaths of damaged trees can be detected and classified as tornado damage, among other severe storms (including downbursts).Performing systematic sweeps of satellite imagery for all forested regions of Canada (over 4 million km2) is time-consuming, currently taking a team of researchers multiple months to complete. In order to speed up this process, an automated computer vision model is utilized. However, due to the rarity and diversity of tornado tracks, existing computer vision instance segmentation architectures may fail to achieve a high enough accuracy to detect every tornado in the vast search area. Instead of directly predicting tornadoes, this study develops a forest damage detection model that utilizes a convolutional neural network to automatically compare small sections of satellite imagery between different dates and identify regions that contain significant changes to the forest. Resultingly, only areas with detected forest damage need to be manually searched for tornado tracks, substantially speeding up the required search time, while being reliable enough to ensure no tornado is misclassified. Future work will aim to utilize this model outside of Canada in the similar forested regions of Northern and Eastern Europe.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| 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 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".