Identifying forest harvesting practices: Clear-cutting and thinning in diverse tree species using dense Landsat time series
Bibliographic record
Abstract
Forest monitoring plays a critical role in achieving sustainable forest management practices. The ability to identify ongoing harvesting activities is crucial for developing targeted strategies to maintain forest health. Traditional monitoring methods, which rely on field inventories, are often expensive and time-consuming. Remote sensing offers an interesting alternative, leveraging dense time series of satellite imagery and various algorithms for disturbance detection. This study presents and assesses a novel methodology for identifying forest harvesting practices (clear-cutting and thinning) using Continuous Change Detection and Classification (CCDC) algorithm, available in Google Earth Engine. The methodology comprises two steps. In the first step, performed at the pixel level, the CCDC algorithm was used to detect changes in the vegetation cover by considering Landsat 8 spectral bands, vegetation indices, and different combinations thereof. In the second step, two optimal thresholds were determined to identify forest harvesting practices based on the proportion of pixels flagged as change. This study was conducted in forest stands consisting of different conifer and broadleaf species. Accuracy was assessed using an independent set of photo-interpreted samples. The results indicated that the short-wave infrared 2 was the best individual band for forest harvesting practices identification, with an average F-score of 0.77 ± 0.06, overperforming vegetation indices. The combination of all spectral bands was the most effective to identify both clear-cuts and thinning (F-score = 0.85 ± 0.05). This combination was used to evaluate the accuracy of this approach for identifying harvesting practices over different tree species. Poplar ( Populus sp.) had the highest identification rate (F-score = 0.99 ± 0.02), while black pine ( Pinus nigra J.F. Arnold) stands had the lowest F-score (0.74 ± 0.05). These results highlight the ability to accurately identify forest harvesting practices even in heterogeneous forests with a high diversity of tree species using dense time series of Landsat imagery. • Identification of thinning and clear-cutting practices using Landsat time series. • Spatial approach to discriminate thinning, clear-cutting and non-harvested stands. • Spectral bands performed better than vegetation indices. • Combination of all Landsat bands outperformed other options. • Poplars followed by Aleppo pine yielded the best results.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| 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) | 0.000 | 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".