THE SHAPE OF SOUND: ELLIPTIC FOURIER DESCRIPTORS (EFD) DISCRIMINATE THE ECHOLOCATION CALLS OF MYOTIS BATS ( M. DAUBENTONII, M. NATTERERI AND M. MYSTACINUS ) (original) (raw)
Related papers
Identifying Bats From Time-Expanded Recordings of Search Calls: Comparing Classification Methods
Journal Information, 2005
Recording ultrasonic echolocation calls of bats using bat-detectors is often used for wide-scale monitoring in studies on bat management and conservation. In Europe, the most important legal instrument for bat conservation is the Habitat Directive (43/92/EEC), which defines various levels of species (and habitat) protection for different bat species and/or genera. Thus for most management needs, the usefulness of bat-monitoring techniques depends on the possibility to determine to species/genus. We compared the discrimination performances of 4 statistical methods applied to identify bat species from their ultrasonic echolocation calls. In 3 different areas of northern Italy, we made recordings of 20 species of bat (60% of those occurring in Italy), 17 from the family Vespertilionidae and 3 from Rhinolophidae. Calls of bats identified to species level from morphological and genetic characters were time-expanded and recorded on release. We measured 7 variables from each call, and we developed classification models through both conventional tests (multiple discriminant analysis and cluster analysis) that were based on a classical statistical approach, and through 2 nonconventional classifiers (classification and regression trees, and neural networks) that relied on generalization and fuzzy reasoning. We compared the performance of the 4 techniques using the percentage of cases classified correctly in 5 classification trials at various taxonomic levels that were characterized by an increasingly difficult identification task: (1) family level (Rhinolophidae vs. Vespertilionidae), (2) species level within genus Rhinolophus, (3) genus level within Vespertilionidae, (4) species level within genus Myotis, and (5) all species. Multiple discriminant function analysis (DFA) correctly classified marginally more cases than artificial neural networks (ANN; 74–100% against 64–100%), especially at the species level (trial 4, species of genus Myotis; trial 5, all species). Both these techniques performed better than cluster analysis or classification and regression trees, the latter reaching only 56 and 41% in Myotis species and all species trials. Artificial neural networks do not yet seem to offer a major advantage over conventional multivariate methods (e.g., DFA) for identifying bat species from their ultrasonic echolocation calls.
The Journal of Basic and Applied Zoology, 2018
Background: Despite their abundance and ecological importance, bats are under significant threat worldwide. There is little information about their distribution, roosting, and habitat requirement for most species, making assessing which species is threatened or in need of special conservation measures difficult. The knowledge gap may partly be due to limitations of the old methods of studying bats which mainly involved capture/or observational techniques. Material and methods: In order to evaluate the potential of identifying insectivorous bats by their echolocation calls in the Sahelian zone of northern Cameroon, 65 bats belonging to five species were captured using standard mist netting: Mops condylurus, Chaerephon major, Mops niveiventer, Scotophilus dinganii, and Scotophilus leucogaster. The bats were identified by using morphometric measurements. An Anabat SD1 detector was later used to record echolocation calls of each individual bat in flight after it was hand-released. The sonogram of each individual bat was analyzed using Analook and categorized into two call types (frequency modulation and frequency modulation/ quasi constant frequency) in order to develop a library of bat reference calls that could be used for a qualitative acoustic survey and species identification. Discriminant function analysis (DFA) was applied to search phase calls of the 65 individual bats in order to evaluate the potential for classifying calls into five species groups. Seven parameters calculated from each search phase call were used to classify calls. Results: Bats where place into two groups according to the structure of calls: FM bats (Mops condylurus, Chaerephon major, Mops niveiventer) and FM/QCF bats (Scotophilus dinganii and Scotophilus leucogaster). The DFA resulted in a correct overall classification of 69.7%. Conclusion: This preliminary study showed that DFA of call parameters is a feasible method that can be used to identify insectivorous bats in the region by their echolocation calls.
Ultrasonic detectors are widely used to survey bats in ecological studies. To evaluate efficacy of acoustic identification, we compiled a library of search phase calls from across the eastern United States using the Anabat system. The call library included 1,846 call sequences of 12 species recorded from 14 states. We determined accuracy rates using 3 parametric and 4 nonparametric classification functions for acoustic identification. The 2 most flexible classification functions also were the most accurate: neural networks (overall classification accuracy ¼ 0.94) and mixture discriminant analysis incorporating an adaptive regression model (overall classification accuracy ¼ 0.93). Flexible nonparametric methods offer substantial benefits when discriminating among closely related species and may preclude the need to group species with similar calls. We demonstrate that quantitative methods provide an effective technique to acoustically identify bats in the eastern United States with known accuracy rates. ß 2011 The Wildlife Society.
A continental-scale tool for acoustic identification of European bats
Journal of Applied Ecology, 2012
1. Acoustic methods are used increasingly to survey and monitor bat populations. However, the use of acoustic methods at continental scales can be hampered by the lack of standardized and objective methods to identify all species recorded. This makes comparable continent-wide monitoring difficult, impeding progress towards developing biodiversity indicators, transboundary conservation programmes and monitoring species distribution changes. 2. Here we developed a continental-scale classifier for acoustic identification of bats, which can be used throughout Europe to ensure objective, consistent and comparable species identifications. We selected 1350 full-spectrum reference calls from a set of 15 858 calls of 34 European species, from EchoBank, a global echolocation call library. We assessed 24 call parameters to evaluate how well they distinguish between species and used the 12 most useful to train a hierarchy of ensembles of artificial neural networks to distinguish the echolocation calls of these bat species. 3. Calls are first classified to one of five call-type groups, with a median accuracy of 97Á6%. The median species-level classification accuracy is 83Á7%, providing robust classification for most European species, and an estimate of classification error for each species. 4. These classifiers were packaged into an online tool, iBatsID, which is freely available, enabling anyone to classify European calls in an objective and consistent way, allowing standardized acoustic identification across the continent. 5. Synthesis and applications. iBatsID is the first freely available and easily accessible continental-scale bat call classifier, providing the basis for standardized, continental acoustic bat monitoring in Europe. This method can provide key information to managers and conservation planners on distribution changes and changes in bat species activity through time.
Field identification of the cryptic vespertilionid bats, Myotis lucifugus and M. yumanensis
Acta Chiropterologica, 2007
Recent advances in molecular techniques have provided new tools for confirming species identities, however they can be expensive and results are not immediately available. Myotis lucificugus and M. yumanensis are morphologically cryptic species of bats sympatric in western North America that can be difficult to distinguish in the field. We evaluated a set of models that used morphological and echolocation call characters obtained in the field to predict species identity as determined by DNA analysis. We constructed models using data from 98 M. lucifugus and 100 M. yumanensis captured throughout the Pacific Northwest from which we had obtained high-quality, time-expansion recordings of their echolocation calls. The best model for distinguishing the species combined forearm length and characteristic frequency of echolocation calls and was able to identify 92% of M. lucifugus and 91% of M. yumanensis individuals, with ≥ 95% confidence. We evaluated the applicability of our model by testing it on additional datasets. Our model correctly classified 83% of M. lucifugus (n = 30) and 93% of M. yumanensis (n = 29) individuals captured in north-central Oregon, whose echolocation calls were recorded using a zero-crossings echolocation detection system. It also correctly classified 86% of M. lucifugus (n = 22) and 85% of M. yumanensis (n = 26) individuals, captured throughout our study area, for which only poor-quality time-expansion recordings of echolocation calls were obtained. Combining morphometrics with echolocation call characteristics may be a useful approach for distinguishing among pairs of cryptic species of bats in other areas.
Bats’ echolocation call characteristics of cryptic Iberian Eptesicus species
European Journal of Wildlife Research, 2015
Advances in molecular methods and analyses, such as DNA sequencing and phylogenetic reconstructions, are being widely used to help clarify the taxonomic challenge posed by cryptic species. While almost morphologically indistinguishable, such species can also present other diagnostic features, including ecological and physiological characteristics. The main goal of this work was to investigate whether it is possible to distinguish two cryptic bat species in Iberia, Eptesicus serotinus and Eptesicus isabellinus, through their echolocation call characteristics. After molecular identification of species' colonies, echolocation calls were recorded during emergence from roosts for 52 individuals. A stepwise discriminant analysis was used to test if the variables measured in the recordings could significantly differentiate the two species. This analysis was able to extract one discriminant function, with the variables' peak frequency and duration of pulses identified as statistically significant. These provided a correct overall classification of approximately 78.8 %. We found that on average peak frequency is higher in the echolocation calls of E. isabellinus compared with that of E. serotinus, but overlap occurred between 23.4 and 28.8 kHz. Moreover, in our recording conditions, calls belonging to E. isabellinus tended to be shorter than those of E. serotinus. Possibly, some acoustic differences could be explained by local adaptations to different climate conditions and ecological niches experienced by each species.
Acoustic identification of Mexican bats based on taxonomic and ecological constraints on call design
1. Monitoring global biodiversity is critical for understanding responses to anthropogenic change, but biodiversity monitoring is often biased away from tropical, megadiverse areas that are experiencing more rapid environmental change. Acoustic surveys are increasingly used to monitor biodiversity change, especially for bats as they are important indicator species and most use sound to detect, localise and classify objects. However, using bat acoustic surveys for monitoring poses several challenges, particularly in megadiverse regions. Many species lack reference recordings, some species have high call similarity or differ in call detectability, and quantitative classification tools, such as machine learning algorithms, have rarely been applied to data from these areas. 2. Here, we collate a reference call library for bat species that occur in a megadiverse country, Mexico. We use 4685 search-phase calls from 1378 individual sequences of 59 bat species to create automatic species identification tools generated by machine learning algorithms (Random Forest). We evaluate the improvement in species-level classification rates gained by using hierarchical classifications, reflecting either taxonomic or ecological constraints (guilds) on call design, and examine how classification rate accuracy changes at different hierarchical levels (family, genus and guild). 3. Species-level classification of calls had a mean accuracy of 66%, and the use of hierarchies improved mean species-level classification accuracy by up to 6% (species within families 72%, species within genera 71Á2% and species within guilds 69Á1%). Classification accuracy to family, genus and guild-level was 91Á7%, 77Á8% and 82Á5%, respectively. 4. The bioacoustic identification tools we have developed are accurate for rapid biodiversity assessments in a megadiverse region and can also be used effectively to classify species at broader taxonomic or ecological levels. This flexibility increases their usefulness when there are incomplete species reference recordings and also offers the opportunity to characterise and track changes in bat community structure. Our results show that bat bioa-coustic surveys in megadiverse countries have more potential than previously thought to monitor biodiversity changes and can be used to direct further developments of bioacoustic monitoring programs in Mexico.
Barbastella, 2020
Ultrasound detectors are becoming an increasingly used tool in bat research. Still, the lack of standardization of measurement criteria in the analysis of recordings is problematic when attempting identifications. For instance, the analytical procedures depend on the technique for obtaining the spectral content of an echolocation pulse (eg. zero-crossing, Fourier analysis)-which is directly related to the detection device and software-, and they should be as consistent as possible among studies' and observers' to allow comparisons. Using full-spectrum recordings of a Myotis species, we measured the minimum frequency and maximum frequency on the spectrogram, as well as considering four thresholds on the power spectrum: at-55 and-50 decibels from 0 decibels, and at 18 and 6 decibels below peak frequency. Focusing on the minimum frequency, we found statistically significant differences of measurements obtained on the spectrogram vs. power spectrum, with higher mean values in the latter as product of pulse truncation, as well as statistically significant differences between two trained observers. The measurements relative to the peak frequency were less variable. Without proper considerations, these issues may represent confounding factors and derive data unsuitable for cross-studies comparisons and potential misidentifications. We make recommendations about measurement criteria and emphasize the importance of replication and more rigorous reports.