Near real-time deforestation detection in Malaysia and Indonesia using change vector analysis with three sensors
Bibliographic record
Abstract
Malaysia and Indonesia have been affected by deforestation caused in great part by the proliferation of oil palm plantations. To survey this loss of forest, several studies have monitored these southeast Asian nations with satellite remote sensing alert systems. The methods used have shown potential for this approach, but they are limited by imagery with coarse spatial resolution, low revisit times, and cloud cover. The objective of this research is to improve near real-time operational deforestation detection by combining three sensors: Sentinel-1, Sentinel-2 and Landsat-8. We used Change Vector Analysis to detect changes between non-affected forest and images under analysis. The results were validated using 166 plots of undisturbed forest and confirmed deforestation events throughout Sabah Malaysian State, and from 70 points from drone pictures in Sumatra, Indonesia. Sentinel-2 and Landsat-8 yielded sufficient results in terms of accuracy (less than 11% of commission and omission error). Sentinel-1 had lower accuracy (14% of commission error and 28% of omission error), probably resulting from geometric distortions and speckle noise. During the high cloud-cover season optical sensors took about twice the time to detect deforestation compared to Sentinel-1 which was not affected by cloud cover. By combining the three sensors, we detected deforestations about 8 days after forest clearing events. Deforestations were only detectable during approximately the first 100 days, before bare soils were often coved by legume crop. Our results indicate that near real-time deforestation detection can reveal most events, but the number of false detections could be improved using a multiple event detection process.
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) | 0.000 | 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 teacher head, 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".