Guidelines for appropriate use of BirdNET scores and other detector outputs

Journal of Ornithology, Journal Year: 2024, Volume and Issue: 165(3), P. 777 - 782

Published: Feb. 14, 2024

Language: Английский

---

Global birdsong embeddings enable superior transfer learning for bioacoustic classification

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: Dec. 18, 2023

Automated bioacoustic analysis aids understanding and protection of both marine terrestrial animals their habitats across extensive spatiotemporal scales, typically involves analyzing vast collections acoustic data. With the advent deep learning models, classification important signals from these datasets has markedly improved. These models power critical data analyses for research decision-making in biodiversity monitoring, animal behaviour studies, natural resource management. However, are often data-hungry require a significant amount labeled training to perform well. While sufficient is available certain taxonomic groups (e.g., common bird species), many classes (such as rare endangered species, non-bird taxa, call-type) lack enough train robust model scratch. This study investigates utility feature embeddings extracted audio identify other than ones were originally trained on. We evaluate on diverse datasets, including different calls dialect types, bat calls, mammals amphibians calls. The vocalization consistently allowed higher quality general datasets. results this indicate that high-quality large-scale classifiers can be harnessed few-shot transfer learning, enabling new limited quantity Our findings reveal potential efficient novel tasks, even scenarios where few samples.

Language: Английский

---

Assessing the potential of BirdNET to infer European bird communities from large-scale ecoacoustic data

Ecological Indicators, Journal Year: 2024, Volume and Issue: 164, P. 112146 - 112146

Published: May 20, 2024

Passive acoustic monitoring has become increasingly popular as a practical and cost-effective way of obtaining highly reliable data in ecological research projects. Increased ease collecting these means that, currently, the main bottleneck ecoacoustic projects is often time required for manual analysis passively collected recordings. In this study we evaluate potential current limitations BirdNET-Analyzer v2.4, most advanced generic deep learning algorithm bird recognition to date, tool assess community composition through automated large-scale data. To end, 3 datasets comprising total 629 environmental soundscapes 194 different sites spread across 19° latitude span Europe. We analyze using both BirdNET listening by local expert birders, then compare results obtained two methods performance at level each single vocalization entire recording sequences (1, 5 or 10 min). Since provides confidence score identification, minimum thresholds can be used filter out identifications with low scores, thus retaining only ones. The volume did not allow us estimate species-specific taxa, so instead evaluated global selected optimized when consistently applied all species. Our analyses reveal that if sufficiently high threshold used. However, inevitable trade-off between precision recall does obtain satisfactory metrics same time. found F1-scores remain moderate (<0.5) studied, extended duration seem currently necessary provide minimally comprehensive picture target community. estimate, however, usage species- context-specific would substantially improve benchmarks study. conclude judicious use AI-based provided represent powerful method assist assessment data, especially duration.

Language: Английский

---

A scalable and transferable approach to combining emerging conservation technologies to identify biodiversity change after large disturbances

Journal of Applied Ecology, Journal Year: 2024, Volume and Issue: 61(4), P. 797 - 808

Published: Jan. 31, 2024

Abstract Ecological disturbances are becoming more extensive and intensive globally, a trend exemplified by ‘megafires’ industrial deforestation, which cause widespread losses of forest cover. Yet the hypothesis that contemporary environmental affecting biodiversity has been difficult to test directly. The novel combination landscape‐scale passive acoustic monitoring, new machine learning algorithm, BirdNET improved Bayesian model‐fitting engines enables cohesive, community‐level before‐after, control‐impact studies disturbances. We conducted such study 2020 megafire in Sierra Nevada, USA. used bespoke dynamic multi‐species occupancy modelling approach, enabled us account for imperfect detection, misclassifications, share information among species. There was no difference colonization between burned unburned forest. In contrast, probability site extinction forest, 0.36, significantly higher than 0.12. Of 67 species our study, 6 (9%) displayed positive response fire, while 28 (41%) significant response. observed 12% decrease avian 1 year post‐fire, substantial shift community composition. However, this ecosystem, many display time‐dependent responses fire unobservable after just year. Synthesis applications . have shown three emerging conservation technologies, animal sound identification algorithms, advances statistical tools, can provide previously unattainable about ecological change. Critically, approach is transferrable scalable, as workflow agnostic or ecosystem each component either freely available (all relevant software) relatively inexpensive (recording hardware). Environmental change unfolding rapidly, but analytical techniques may help understanding and—thus interventions—keep pace.

Language: Английский

---

Frequent, heterogenous fire supports a forest owl assemblage

Ecological Applications, Journal Year: 2025, Volume and Issue: 35(1)

Published: Jan. 1, 2025

Abstract Fire shapes biodiversity in many forested ecosystems, but historical management practices and anthropogenic climate change have led to larger, more severe fires that threaten animal species where such disturbances do not occur naturally. As predators, owls can play important ecological roles biological communities, how changing fire regimes affect individual assemblages is largely unknown. Here, we examined the impact of severity, history, configuration over past 35 years on an assemblage six forest owl Sierra Nevada, California, using ecosystem‐scale passive acoustic monitoring. While negative impacts this appeared be ephemeral (1–4 duration), spotted avoided sites burned at high‐severity for up two decades after a fire. Low‐ moderate‐severity benefited small cavity‐nesting great horned owls. Most study adapted within region's natural range variation, characterized by higher proportions low‐ relatively less some may resilient wildfire than others, novel “megafires” are frequent, contiguously limit distribution reducing prevalence eliminating habitat closed‐canopy multiple decades. Management strategies restore with patches promote mosaic conditions will likely facilitate conservation predators.

Language: Английский

---

Estimating population size for California spotted owls and barred owls across the Sierra Nevada ecosystem with bioacoustics

Ecological Indicators, Journal Year: 2023, Volume and Issue: 154, P. 110851 - 110851

Published: Aug. 27, 2023

Monitoring population size at ecosystem scales is difficult for most species of conservation concern. While assessing site occupancy broad has proven feasible, rigorous tracking changes in over time not – even though it can provide a stronger basis status and conservation-decision making. Therefore, we demonstrate how relatively low-intensity, ecosystem-scale passive acoustic monitoring (PAM) be linked to local-density estimate the native California spotted owls (Strix occidentalis occidentalis) invasive barred (S. varia) across western Sierra Nevada, California. Based on PAM sampling grid with 400 ha cells (the approximate home range these species), estimated between 0.42 (SE = 0.02) 0.30 using liberal strict criteria, respectively, considering cell occupied. PAM-based estimates within local-scale density study areas (range 0.41–0.78 0.28–0.76 respectively) were strongly positively correlated local 0.08–0.31 owl/km2) this species. In contrast, ecosystem-wide was very low based (0.034, SE < 0.01), as densities studies 0–0.005 owls/km2). By scaling studies, that, depending 2,218 278) or 2,328 489) occurred Nevada 2021. Thus, while are rare subspecies, they well-distributed Nevada. Because there so few owl detections, could abundance, which reflects success prior experimental removals region. conclusion, our provides generalizable framework estimating territorial when available. that approach novel valuable insights into populations aid conservation.

Language: Английский

---

Can the Hermit Warbler ( Setophaga occidentalis ) serve as an old-forest indicator species in the Sierra Nevada?

Journal of Field Ornithology, Journal Year: 2024, Volume and Issue: 95(1)

Published: Jan. 1, 2024

Changing fire regimes in western North America have raised the possibility of widespread loss forest cover, making restoration a major priority. In one such ecosystem, Sierra Nevada California, implications management policy been evaluated primarily via their potential effects on California Spotted Owl (Strix occidentalis occidentalis). Yet owl's cryptic life history, large home range, and declining population all make it difficult to study. The Hermit Warbler (Setophaga occidentalis) may be valuable proxy species for because two similar associations with older habitat, but former could enable researchers achieve higher statistical power when studying changes key habitats. We conducted passive acoustic surveys across entire west slope between May July 2021, identified both vocalizations resulting audio using BirdNET algorithm, used single-season occupancy models examine relationship six remotely sensed variables representing attributes forests as well presence. Warblers were observed at sites which Owls present, those represented just 30.5% Warbler's total occupied range. site was positively associated mean tree diameter presence (model weight = 0.97). is more appropriate habitat beneficial than owl itself. As such, monitoring means understanding important old-forest habitat.

Language: Английский

---

HawkEars: A regional, high-performance avian acoustic classifier

Ecological Informatics, Journal Year: 2025, Volume and Issue: unknown, P. 103122 - 103122

Published: March 1, 2025

Language: Английский

---

A comparison of statistical methods for deriving occupancy estimates from machine learning outputs

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: April 27, 2025

Abstract The combination of autonomous recording units (ARUs) and machine learning enables scalable biodiversity monitoring. These data are often analysed using occupancy models, yet methods for integrating outputs with these models rarely compared. Using the Yucatán black howler monkey as a case study, we evaluated four approaches ARU into models: (i) standard verified data, false-positive (ii) presence-absence (iii) counts detections, (iv) continuous classifier scores. We assessed estimator accuracy effects decision threshold, temporal subsampling, verification strategies. found that classifier-guided listening model provided an accurate estimate minimal effort. yielded similarly estimates under specific conditions, but were sensitive to subjective choices including threshold. inability determine stable parameter priori, coupled increased computational complexity several (i.e. detection-count continuous-score models), limits practical application models. In high-performance readily detectable species, paired provides efficient approach accurately estimating occupancy.

Language: Английский

---

Advancing invasive species monitoring: A free tool for detecting invasive cane toads using continental-scale data

Ecological Informatics, Journal Year: 2025, Volume and Issue: unknown, P. 103172 - 103172

Published: April 1, 2025

Language: Английский

---