Assessment of remotely piloted aircraft data classification of wetland vegetation communities and changes in their pattern with elevation
Bibliographic record
Abstract
Wetlands host a myriad of services both to the environment and society. While the provision of those services are partly dependent on the terrestrial plant communities that both comprise and are adjacent to a wetland, little has been done to map these communities at high spatial resolutions or measure their composition and configuration. Remotely piloted aircraft (RPA) have been used to capture very-high resolution imagery that can aid our understanding of wetland characteristics and function without harming the local environment. However, a gap remains in our understanding of the composition and configuration of vegetation communities in high-elevation wetlands. As a step toward improving our understanding, an RPA was used to collect multi-spectral data (red edge, green, blue, near infrared) and analyzed to determine the ability to identify and map the composition and configuration of wetland vegetation communities at high elevations in Alberta, Canada. These wetlands have been excluded from the Canadian Wetland Classification System mapped areas. In addition to assessing the ability for automated classification of vegetation communities in an object-based image analysis, the relative contribution each spectral band, their combination in different indices (e.g., NDVI), and a generated digital surface model had on classification accuracy was quantified. Results from a random tree classifier obtained an overall accuracy of 91.43%, total producer’s accuracy of 86.9% and total user’s accuracy of 92.7%. Of 17 inputs (12 image layers and 5 objects features) included in the classifier, the Digital Surface Model had the greatest overall importance (average of 15.9%). While comparison between the classifier and testing samples derived through manual segmentation selection yielded high accuracies, comparison of the automated classification against ground survey plots were substantially less accurate (total overall accuracy 54.5%). Analysis of the composition and configuration of the classified RPA data identified five non-correlated landscape metrics that showed statistically significant differences when wetlands were compared across a gradient in elevation. These results demonstrate that elevation can affect the pattern of wetland vegetation and that further research should be done to determine if and how these pattern changes may affect specific wetland functions. In addition to situating the aforementioned results among the broader literature, the potential for improved vegetation mapping with remotely piloted aircraft is discussed along with the need for a standard set of reporting variables for RPA data collection to facilitate comparative analyses.
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 | 0.004 |
| 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 | 0.000 |
| 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".