Extending Canadian forest disturbance history maps prior to 1985
Bibliographic record
Abstract
Abstract An accurate depiction of wildfire, harvesting, and insect outbreak disturbances is essential for sustainable ecosystem management of forests in Canada. Even though the advent of temporally consistent 30‐m resolution Landsat data has enabled the detailed mapping of forest disturbances in Canada from 1985 onward, the disturbance record prior to 1985 remains sparse. This study aimed to extend the existing pre‐1985 disturbance history record by mapping wildfire, harvest, and insect outbreaks in Canadian forests between 1965 and 1984. Our geospatial data processing methodology relied on multilayer perceptrons (MLP) trained on spectral recovery signatures to map and age these disturbances. Our model detected approximately 4.8, 7.3, and 3.8 million ha of burnt, harvested, and insect‐ravaged forest areas, respectively, that were absent from national and provincial disturbance databases and forest inventories. Results were validated using both internal and external validation datasets. Our disturbance detection methodology was highly effective, with an internal validation kappa score of 0.91 and an external score of 0.81. The fire and harvest age disturbance MLPs, whose predictions can also be used as a proxy of forest stand age, performed adequately on the internal (fire R 2 = 0.675; root mean squared error [RMSE] = 4.42; harvest R 2 = 0.723; RMSE = 3.17) and external validation datasets (fire R 2 = 0.242; RMSE = 4.69; harvest R 2 = 0.257; RMSE = 5.46), outperforming existing forest age disturbance products. Finally, we relied on several open data products, such as provincial forest inventories, to correct our disturbance type and year prediction whenever these more reliable, but incomplete, data sources were available. Specific years were not assigned to insect outbreaks due to the lack of dependable training and validation data. We also illustrate how extending the existing forest disturbance record by 20 years may provide a more in‐depth understanding of landscape‐disturbance dynamics with a case study of the 2023 Canadian wildfire season.
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 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 | 0.000 |
| Insufficient payload (model declined to judge) | 0.013 | 0.034 |
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; both teacher heads agree on what is shown here.
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".