More Accurately Estimating Aboveground Biomass in Tropical Forests With Complex Forest Structures and Regions of High‐Aboveground Biomass
Bibliographic record
Abstract
Abstract Accurately estimating aboveground biomass (AGB) in tropical forests is vital for managing the threats posed by deforestation, degradation, and climate change. However, challenges persist in accurately estimating AGB in high AGB regions. This study aims to accurately estimate the AGB of regions with high AGB by using spatial statistical analyses based on AGB estimates made by machine‐learning fusion of multisource data. We hypothesize that incorporating dominant auxiliary factors in the analysis increases the estimation accuracy. This study focuses on tropical forests located in Longyan, Fujian Province, China, covering an area of 19,028 km 2 . Multisource data are used, including airborne laser scanning, the Shuttle Radar Topography Mission digital elevation model, the Landsat Operational Land Imager, and the National Forest Inventory. Based on GeogDetector's spatial covariance matrix and the spatial similarity principle, we identify key auxiliary factors (dominant tree species, canopy closure, and herbaceous cover) and investigated how auxiliary variables can improve estimation accuracy. Empirical Bayesian kriging regression prediction introduces the main auxiliary factors to refine AGB estimates. These refinements significantly enhance the accuracy of AGB estimates, particularly for high AGB, resulting in a 0.1 increase in R 2 , a 7.0% reduction in root mean square error, a 13.5% reduction in mean square error, and a 6.6% reduction in mean absolute error when compared with the AGB estimates obtained by using machine learning to fuse multisource data. Thus, incorporating spatial statistical analysis into the integration of multisource data and machine learning for AGB estimation can enhance the accuracy of high‐AGB estimates in intricate forest structures, resulting in precise AGB maps.
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 | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| 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".