Automated identification of individual birds by song enables multi-year recapture from passive acoustic monitoring data
Bibliographic record
Abstract
Abstract Autonomous sensors and machine learning are transforming ecology by enabling large-scale observation of organisms and ecosystems. However, sensor data collected by camera traps, acoustic recorders, and satellites are typically used only to produce detections of unidentified individuals. Tracking individuals over time to study movement, survival, and behavior continues to require invasive and high-effort capture and marking techniques. Here, we introduce an automated, general approach that identifies individual animals from passive acoustic recordings based on individually distinctive vocalizations. Unlike previous approaches, ours can identify individuals in passive acoustic recordings without previously labeled examples of their vocalizations. We apply our approach to a model songbird species (Ovenbird, Seiurus aurocapilla ), estimating abundance and annual survival across 126 locations and four years. Our approach identifies individuals with 96% accuracy. We find high Ovenbird apparent annual survival (0.70) and acoustic recapture probability (0.89) across 405 individuals. Our approach can readily be applied to other species with individually distinctive vocalizations using open-source Python implementations. Automated individual identification will broadly unlock the ability to passively recapture individual animals at the massive scale of autonomous sensing, supporting the study of population trajectories and informing proactive ecosystem management to prevent biodiversity loss.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
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".