Improving forest disturbance labels through Sentinel-1 change detection validation
Bibliographic record
Abstract
Global forest ecosystems face unprecedented challenges, such as fire, wind, drought, and insect outbreaks, resulting in rapid forest decline. Analyzing these disturbances on a large scale requires the use of remote sensing techniques, but the spatial and temporal uncertainty in forest disturbance reference data poses a significant obstacle.In this study, we validate and refine existing disturbance labels of the U.S. Forest Service Forest Health Protection [1] Dataset USDA by using a change detection algorithm [2] based on radar data from Sentinel-1. To this end, we analyze the spatio-temporal overlap of disturbed areas from Sentinel-1 with the USDA labels and further explore spatio-temporal fingerprints of remote sensing indices commonly used for disturbance detection. As the analysis of the remote sensing indices shows, this refinement of the accuracy of disturbance labels provides a more reliable basis for ecological research and land management practice. References:[1] Coleman, T. W., Graves, A. D., Heath, Z., Flowers, R. W., Hanavan, R. P., Cluck, D. R., & Ryerson, D. (2018). Accuracy of aerial detection surveys for mapping insect and disease disturbances in the United States. Forest Ecology and Management, 430, 321–336. https://doi.org/10.1016/j.foreco.2018.08.020[2] Cremer, F., Gans, F., Cortes, J. & Thiel, C. (2023). Mapping Forest Loss in Europe with Sentinel-1. In European Commission, Joint Research Centre, Soille, P., Lumnitz, S., Albani, S., Proceedings of the 2023 conference on Big Data from Space (BiDS’23) – From foresight to impact – 6-9 November 2023, Austrian Center, Vienna, Soille, P.(editor), Lumnitz, S.(editor), Albani, S.(editor), (pp. 361 - 364) Publications Office of the European Union, 2023, https://data.europa.eu/doi/10.2760/46796
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 | 0.009 | 0.018 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".