Multi-sensor remote sensing approach for oil palm mapping and stand age detection using 38-year landsat and sentinel time series data in the google earth engine
Bibliographic record
Abstract
Rapid oil-palm expansion in Southeast Asia demands accurate, up-to-date information on plantation extent and stand age, yet single-sensor approaches often misclassify mixed tropical vegetation and cannot reconstruct planting histories that underlie yield forecasts and sustainability metrics. We integrated the full 1987–2024 Landsat surface-reflectance archive (38 years) with 2024 Sentinel-2 imagery to build a cloud-native monitoring pipeline in Google Earth Engine (GEE). A Random-Forest (RF) model, trained on stratified field samples, delineated plantation extent from six Sentinel-2 bands, five vegetation/surface indices, three Sentinel-1-derived back-scatter layers, and two terrain variables. Age was retrieved inside the verified extent mask by locating the most recent year in which smoothed NDVI dropped below its 20th percentile while the Bare-Soil Index (BSI) exceeded its 80th percentile, thereby flagging the latest planting or replanting events in the Landsat record. Model skill was assessed with a 30% hold-out set using overall accuracy, κ, precision, and recall; age estimates were validated against 234 ground plots with MAE and RMSE. The multi-sensor classifier reached an overall accuracy of 90.5% and Kappa = 0.81, with balanced error rates (oil palm: 88.6% precision, 92.7% recall; non-oil palm: 92.5% precision, 88.4% recall). Mapped plantations covered 561 km², or 60.9% of the study district. Age retrieval achieved an RMSE of 4.0 yr, revealing distinct replanting events that correspond to known market cycles and providing spatially explicit age strata relevant to yield and carbon-stock modelling. Combining the long Landsat record with high-resolution Sentinel data markedly outperforms single-sensor methods for both extent mapping and age estimation. The workflow relies solely on free, globally available imagery and scalable cloud computing, making it immediately transferable to government and NGO monitoring programmes. By delivering accurate maps of plantation area and stand age, the approach fills a critical information gap for sustainable palm-oil certification, carbon accounting, and land-use policy across the humid tropics.
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 | 0.001 | 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".