Detecting forest and linear woody feature change between 1954 and 2019 in Southeastern Canadian agroecosystems for regional biodiversity assessment
Bibliographic record
Abstract
Effectively mapping and detecting changes in forests and linear woody features (LWFs) is crucial for assessing their impact on biodiversity and ecosystem services. This study investigated this capability using heterogeneous, high-resolution aerial imagery, in terms of spectral and spatial properties. Mitigating the influence of these factors, arising from differences in sensor specifications and acquisition conditions, is essential for robust detection and analysis of temporal change across historical image datasets. The deep learning model developed here successfully mapped forests and LWFs between 1954 and 2019 using just a single image band, enabling reliable change estimation. Assessment at the pixel scale showed forest mapping achieved an accuracy of 90%, while LWF accuracy was lower at 69%, primarily due to their narrow widths and boundary errors in both the reference and predicted results. For LWFs an object-based assessment was undertaken to reduce the effect of precise delineation achieving a higher accuracy of 77%. As a final assessment, comparison of area within 200 by 200 m extents showed good agreement, with a mean absolute error of 1.3% for LWFs. For forests this was 2.7%. In terms of change detection, the accuracy was greater than 81% for both forests and LWFs. Change analysis indicated an 8.5% net increase in forests since 1954, along with a small net loss of less than 1% in LWFs. LWF loss was mainly attributed to forest gains. In areas without significant forest gain, LWFs slightly increased. These changes are generally seen as beneficial for biodiversity and ecosystem services in the region. However, other factors such as urban development and larger agricultural field sizes need to be considered in future studies.
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".