Article
Refine
Document Type
- Article (3) (remove)
Language
- English (3) (remove)
Has Fulltext
- yes (3)
Is part of the Bibliography
- no (3)
Keywords
- Chiroptera (3) (remove)
Institute
Publisher
- Wiley (2)
- BioMed Central (BMC) (1)
Urbanization is a major contributor to the loss of biodiversity. Its rapid progress is mostly at the expense of natural ecosystems and the species inhabiting them. While some species can adjust quickly and thrive in cities, many others cannot. To support biodiversity conservation and guide management decisions in urban areas, it is important to find robust methods to estimate the urban affinity of species (i.e. their tendency to live in urban areas) and understand how it is associated with their traits. Since previous studies mainly relied on discrete classifications of species' urban affinity, often involving inconsistent assessments or variable parameters, their results were difficult to compare. To address this issue, we developed and evaluated a set of continuous indices that quantify species' urban affinity based on publicly available occurrence data. We investigated the extent to which a species' position along the urban affinity gradient depends on the chosen index and how this choice affects inferences about the relationship between urban affinity and a set of morphological, sensory and functional traits. While these indices are applicable to a wide range of taxonomic groups, we examined their performance using a global set of 356 bat species. As bats vary in sensitivity to anthropogenic disturbances, they provide an interesting case study. We found that different types of indices resulted in different rankings of species on the urban affinity spectrum, but this had little effect on the association of traits with urban affinity. Our results suggest that bat species predisposed to urban life are characterized by low echolocation call frequencies, relatively long call durations, small body size and flexibility in the selection of the roost type. We conclude that simple indices are appropriate and practical, and propose to apply them to more taxa to improve our understanding of how urbanization favours or filters species with particular traits.
BatNet: a deep learning-based tool for automated bat species identification from camera trap images
(2023)
Automated monitoring technologies can increase the efficiency of ecological data collection and support data-driven conservation. Camera traps coupled with infrared light barriers can be used to monitor temperate-zone bat assemblages at underground hibernacula, where thousands of individuals of multiple species can aggregate in winter. However, the broad-scale adoption of such photo-monitoring techniques is limited by the time-consuming bottleneck of manual image processing. Here, we present BatNet, an open-source, deep learning-based tool for automated identification of 13 European bat species from camera trap images. BatNet includes a user-friendly graphical interface, where it can be retrained to identify new bat species or to create site-specific models to improve detection accuracy at new sites. Model accuracy was evaluated on images from both trained and untrained sites, and in an ecological context, where community- and species-level metrics (species diversity, relative abundance, and species-level activity patterns) were compared between human experts and BatNet. At trained sites, model performance was high across all species (F1-score: 0.98–1). At untrained sites, overall classification accuracy remained high (96.7–98.2%), when camera placement was comparable to the training images (<3 m from the entrance; <45° angle relative to the opening). For atypical camera placements (>3 m or >45° angle), retraining the detector model with 500 site-specific annotations achieved an accuracy of over 95% at all sites. In the ecological case study, all investigated metrics were nearly identical between human experts and BatNet. Finally, we exemplify the ability to retrain BatNet to identify a new bat species, achieving an F1-score of 0.99 while maintaining high classification accuracy for all original species. BatNet can be implemented directly to scale up the deployment of camera traps in Europe and enhance bat population monitoring. Moreover, the pretrained model can serve as a baseline for transfer learning to automatize the image-based identification of bat species worldwide.
Background
Hibernation allows species to conserve energy and thereby bridge unfavorable environmental conditions. At the same time, hibernation imposes substantial ecological and physiological costs. Understanding how hibernation timing differs within and between species can provide insights into the underlying drivers of this trade-off. However, this requires individualized long-term data that are often unavailable. Here, we used automatic monitoring techniques and a reproducible analysis pipeline to assess the individualized hibernation phenology of two sympatric bat species. Our study is based on data of more than 1100 RFID-tagged Daubenton’s bats (Myotis daubentonii) and Natterer’s bats (Myotis nattereri) collected over seven years at a hibernaculum in Germany. We used linear mixed models to analyze species-, sex- and age-specific differences in entrance, emergence and duration of the longest continuous period spent in the hibernaculum.
Results
Overall, Daubenton’s bats entered the hibernaculum earlier and emerged later than Natterer’s bats, resulting in a nearly twice as long hibernation duration. In both species, adult females entered earlier and emerged from hibernation later than adult males. Hibernation duration was shorter for juveniles than adults with the exception of adult male Natterer’s bats whose hibernation duration was shortest of all classes. Finally, hibernation timing differed among years, but yearly variations in entrance and emergence timing were not equally shifted in both species.
Conclusions
Our results suggest that even in sympatric species, and across sex and age classes, hibernation timing may be differentially affected by environmental conditions. This highlights the necessity of using individualized information when studying the impact of changing environments on hibernation phenology.