Training and usage of detection dogs to better understand bumble bee nesting habitat: Challenges and opportunities
Bibliographic record
Abstract
Bumble bees are among the most imperiled pollinators. However, habitat use, especially nest site selection, remains relatively unknown. Methods to locate nests are invaluable to better understand habitat requirements and monitor wild populations. Building on prior study findings, we report constraints and possibilities observed while training detection dogs to locate bumble bee nests. Three conservation detection dogs were initially trained to three species of bumble bee nest material, first within glass jars concealed in a row of cinder blocks, then placed in the open or partially hidden for area searches. The next intended training step was to expose the dogs to natural nests located by community science volunteers. However, significant effort (> 250 hrs), yielded only two confirmed, natural nests suitable for dog training purposes. Although the dogs did not progress past the formative training stage valuable insight was gained. Maximum observed detection distance for bumble bee nest material during initial controlled training was 15 m, which decreased significantly (< 1 m) once training progressed to buried samples and natural nests. Three main considerations around future training and usage of detection dogs were identified. First, dogs might benefit from transitional training via exposures to known natural nests, regardless of species. However, it may be too difficult for people to find natural nests for this, and prior work demonstrated the ability of dogs to generalize and find natural nests after testing to artificially-buried nest material. Second, confirming a dog's nest find, via resident bee presence, is nuanced. Third, future study design and objectives must harness strengths, and reflect limitations of detection dog surveys and search strategies, as extensively discussed in this paper. Prospective studies involving detection dogs for locating bumble bee nests would benefit from considering the drawbacks and opportunities discussed and can mitigate limitations through incorporating these considerations in their study design.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".