Identification of twenty‐two bat species (Mammalia: Chiroptera) from Italy by analysis of time‐expanded recordings of echolocation calls

Abstract

Spectral and temporal features of echolocation calls emitted by 22 bat species from Italy (three rhinolophids, 18 vespertilionids and the molossid Tadarida teniotis) are described. Time-expanded recordings of calls from 950 bats of known identity were examined. Rhinolophus ferrumequinum, R. hipposideros, R. euryale and T. teniotis could be identi®ed by measuring the call frequency of highest energy (FMAXE). Quadratic discriminant function analysis with cross-validation was applied to calls from the remaining 18 species. A function based on start frequency (SF), end frequency (EF), FMAXE and duration (D) provided a correct overall classi®cation of approximately 82%. A classi®cation model at genus level that also comprised middle frequency (MF) and inter-pulse interval (IPI) reached 94% correct classi®cation. Two separate discriminant functions were devised for species emitting FM (frequency modulated) and FM/QCF calls (i.e. calls consisting of a frequency-modulated component followed by a terminal part whose frequency is almost constant) respectively. The former function included SF, EF, FMAXE and D and provided an overall classi®cation rate of 71%; the latter comprised EF, MF, D and IPI, and reached 96%. The functions may be applied to bat habitat surveys in southern Italy since they cover most of the species occurring in the area.

Variability in echolocation call design of 26 Swiss bat species: consequences, limits and options for automated field identification with a synergetic pattern recognition approach

“…PARAMETRIC APPROACHES Parametric measurements taken e.g., from oscilloscopes, powerspectra or spectrograms allow for later analysis with a variety of approaches ranging from discriminant function analysis or DFA (Zingg 1990;Vaughan et al 1997;Jones et al 2000;Russo & Jones 2002) to decision trees (Herr et al 1997) and artificial neural networks or ANN (Stocker 1998;Parsons 2001). For measurements taken from data-reduced samples high recognition rates above 90% were reported for some species, others failed completely to be separated (Herr et al 1997).…”

An Extensive Review of Methods of Identification of Bat Species through Acoustics

“…Qualitative methods involve researchers listening to calls [13], taking account of the echolocation call structure [10]. These methods require that the researcher has to get a good site(a suitable habitat) in which they can see the bats and record the echolocation calls.…”

“…These methods require that the researcher has to get a good site(a suitable habitat) in which they can see the bats and record the echolocation calls. Hence the observer must wait for the opportunity to identify a bat and identify its staying place which is called the roost [13]. The researcher must follow the bats along flight paths to roosts where bats can be captured.…”

“…Russo and Jones [13] have proposed identification of twentytwo bat species(Mammalia: Chiroptera) from Italy by analysis of time-expanded recordings of echolocation calls. They described the spectral and temporal features of echolocation calls emitted by twenty two bat species from Italy (three rhinolophids, eighteen vespertilionids and the molossid Tadarida teniotis).…”

“…Although these previous studies accurately classify many of the species on which they are trained and prove the concept and value of quantitative call identification, they have not been made publicly accessible and are restricted to a regional(often national) level (eg. Venezuela [8]; Greece; Italy [13]; Meditteranean area [23]; UK [24]; Switzerland [22];). Therefore, they cannot be used to generate comparable classifications at a continental scale [2].…”