Detection and validation of forest disturbances using RADARSAT-2 data
Bibliographic record
Abstract
RADARSAT 2 SAR data was used to develop a monitoring program for Canadian forest lands with the aim to provide information on forest harvesting. A study site in British Columbia, Canada, characterized by coniferous forest, was selected. RADARSAT-2 MultiLook Fine mode, acquired from mid-June through mid-September, from 2011 to 2015 was analyzed with the aim to detect forest disturbances. Due to large data volumes and the need for efficiency, an automated end-to-end solution was implemented. The automated solution included image coregistration, temporal filtering, detection of forest disturbances, and delineation of the disturbances. To reduce the detection of false positives, a non-forest mask was developed that entailed a combination of CanVec data that delineated areas such as water bodies, roads, and urban/industrial areas and SAR-derived information such as layover and scattering from urban areas. To assess the performance of the change detection algorithm, the RADARSAT-2 changes were compared to tree-loss information from the Canadian Forest Service (CFS) and cut-block information from the BC Forest Service (BCFS). Since CFS and the BCFS information was representative of annual changes, but the RADARSAT-2 derived changes were representative of summer-only changes, there were discrepancies between the RADARSAT-2 data and the CFS/BCFS data. Notwithstanding these discrepancies, the detection performance was better than 80% for 2011/12 and 2012/13. For 2013/15, however, due to the two-year gap between data acquisition, the detection performance was 74%.
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) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| 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".