Macrogeographical variability in the great call of Hylobates agilis: assessing the applicability of vocal analysis in studies of fine‐scale taxonomy of gibbons

Abstract

Vocal characteristics have been used extensively to distinguish different taxonomic units of gibbons (family Hylobatidae). The agile gibbon (Hylobates agilis) has a disjunct distribution range in the Southeast Asian archipelago (remnants of the former Sunda landmass), and populations on different islands are currently recognized as distinct subspecies or even species. We recorded great calls from female agile gibbons from two populations on Sumatra and two populations on Borneo and examined the vocal variability on four levels: within-individuals, between-individuals, between-populations and between-islands. The primary objective was to evaluate the effect of geographical isolation on variability in song pattern and to test whether proposed island-specific song characteristics exist, reflecting evolutionary divergence between Sumatran and Bornean agile gibbons. One hundred great calls were recorded from 20 females and analyzed for 18 spectral and temporal acoustic parameters. Principal component analysis followed by a nested ANOVA on components revealed a complex pattern of song variability not likely to reflect taxonomic or evolutionary relationship. We found no evidence that Sumatran and Bornean agile gibbons have evolved different vocal characteristics, refuting a distinction between them based on vocal characteristics. A high level of plasticity was found in great calls from the same individual, and generally the inferred pattern of variability suggested that ecological or social factors may confound any genetically based island dialects.

Understanding sources of variance and correlation among features of Bornean gibbon (Hylobates muelleri) female calls

“…A previous study on agile gibbons from populations on Sumatra and Borneo showed considerable within-individual variability, particularly in the spectral parameters (Heller et al, 2010), which is consistent with our finding of substantial intra-individual variation in note maximum frequency. We collected recordings from unhabituated gibbons at variable recording distances, and recording distance can have an effect on the frequency estimates (Kroodsma, 2017;Zollinger et al, 2012).…”

“…The choice of features to use in analyses of animal vocalizations is highly subjective, and can range from many features (58 features; Oyakawa et al, 2007) to relatively few (12 features; Haimoff and Gittins, 1985). Generally, the inclusion of a higher number of features does not provide more meaningful information regarding variation in gibbon calls (Heller et al, 2010). To estimate call features in the present study we created spectrograms using the RAVEN PRO 1.5 sound analysis software (Cornell Lab of Ornithology Bioacoustics Research Program, Ithaca, New York) with a 512-point (11.6 ms) Hann window (3 dB bandwidth ¼ 124 Hz), with 75% overlap, and a 1024-point discrete Fourier transform, yielding time and frequency measurement precision of 2.9 ms and 43.1 Hz.…”

“…In non-human primates, evidence for geographic variation in vocalizations has been found in tarsiers (Tarsius tarsier; Burton and Nietsch, 2010), pygmy marmosets (Cebuella pygmaea; de la Torre and Snowdon, 2009), Thomas langurs (Presbytis thomasi; Wich et al, 2008), gibbons (Hylobates spp. ; Dallmann and Geissmann, 2009;Heller et al, 2010), orangutans (Pongo spp. ; Delgado, 2007), and chimpanzees (Pan troglodytes; Mitani et al, 1999).…”

An analysis of white‐handed gibbon male song reveals speech‐like phrases

“…Together these findings suggest that, if gibbons are able to recognize the song of other males, then processing both individual notes and the higher‐order coda properties would facilitate greater accuracy in such recognition. Similarly, distant members of established pairs could potentially be identifiable by distinct codas, since males typically produce them only in response to their own mate's great call, and the great call is also discriminable between individuals in multiple gibbon species (Clink et al, ; Dallmann and Geissmann, 2009; Haimoff and Gittins, 1985; Heller, Sander, Wang, Usman, and Dabelsteen, ), including white‐handed gibbons (Terleph et al, ). It is not known if or to what degree gibbons attend to these cues, but playback studies would be an effective means of testing such questions.…”

Application of a semi-automated vocal fingerprinting approach to monitor Bornean gibbon females in an experimentally fragmented landscape in Sabah, Malaysia

“…individual identity) based on extracted call features (Lee 2010). One of the most commonly used methods for classification of primate vocalizations is linear discriminant function analysis (DFA) (Delgado 2007;Wich et al 2008;Heller et al 2010;Santorelli et al 2013). Linear DFA is a supervised, multivariate technique that tests whether different classes of objects (e.g.…”