Estimating the landscape-scale abundance of an arboreal folivore using thermal imaging drones and binomial N-mixture modelling
Bibliographic record
Abstract
Estimating the abundance of wildlife populations at a landscape-scale is vital for conservation, but is often hampered by survey costs, data processing and imperfect detection. In this study, we developed a framework that combines a protocol for validating nocturnal thermal drone detections in real-time with N-mixture modelling to estimate the landscape-scale abundance of arboreal folivores. As a case study, we estimated the abundance of koalas ( Phascolarctos cinereus ) across seven reserves (673 km 2 ) in New South Wales, Australia. We conducted thermal drone surveys of 208, 25-ha sites stratified across vegetation type and fire history, on average, three times over consecutive nights (range 1–12 repeats), between 18:00–04:00 h (May to September). All koala detections were validated by field personnel or in real-time with drones equipped with a thermal camera and searchlight. Koalas were detected on 245 occasions. We fitted N-mixture models to validated repeat count data to quantify the effect of site and observation variables on abundance and detectability. Using our top set of competing models, we estimated that 4357 koalas (95 % CI = 2319–8307) occupy the seven reserves, with a mean detection probability of 0.22 (95 % CI = 0.15–0.31) over all survey occasions. We found detection probability decreased with increases in relative humidity and temperature. Koala abundance was negatively associated with fire severity, elevation, tree height and soil clay content, and positively associated with available water content, forest cover and soil organic carbon. Our framework, which combines real-time field validated drone data while accounting for imperfect detection, improves the accuracy of abundance estimates for arboreal folivores across large-scales.
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 | 0.006 | 0.017 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".