Bats are an important component of biodiversity in Southeast Asia, and are key indicators of habitat quality. Acoustic analysis of echolocation calls not only allows the identification of bat species that are difficult to capture, but also allows for rapid and standardised ways to survey and monitor bats over large areas. However keys based on call parameters must also account for geographic variation in call parameters, and consider any effects of morphology and sex on call frequency. Presence-only modelling can predict likely geographic locations of specific taxa, and can used to refine decision making so that species unlikely to occur in a specific region can be omitted from more localised acoustic libraries. Here we develop an acoustic library for the echolocation calls of rhinolophoid bats in Thailand, and use presence-only modelling (Maxent) to explore how species with similar calls in a library developed at the national level can sometimes be largely allopatric, and hence identifiable, once geographic range is predicted. Presence-only modelling can also be used to explore whether species with similar calls adjust call frequency in likely areas of sympatry. We analysed calls from fourteen species of rhinolophid and twelve hipposiderid species from Thailand. Calls from a further three rhinolophid and one hipposiderid species are also described but not analysed statistically because of small sample sizes. Even without considering geographic variation, 69% of rhinolophid (14 species with a minimum of five individuals/ species) and 91% of hipposiderid calls (12 species) could be classified successfully to species using two call parameters (frequency of most energy (FMAXE) and duration) in a discriminant function analysis. Most of the discrimination was achieved because species often utilised different frequency bands. Morphology can also affect call frequency both across and within species. In both rhinolophids and hipposiderid families there was a negative relationship between FMAXE and forearm length. Within species, FMAXE was negatively related to forearm length in Rhinolophus microglobosus, R. pusillus and R. thomasi, and positively related to forearm length in R. affinis and R. pearsonii. Male R. pusillus called at higher frequencies than females, although there was no sexual size dimorphism in forearm length. Call frequency was negatively related to relative humidity in R. pusillus, suggesting that bats called at lower frequencies in humid environments to counter increases in atmospheric attenuation. Presence-only modelling was used to show that some species with similar call frequencies (e.g., R. lepidus and R. microglobosus; R. yunanensis and R. trifoliatus are predicted to occur largely in allopatry, and so could be identified reliably in particular parts of the country. Presenceonly modeling can assist in predicting areas of overlap between species with similar call frequencies. Other species (e.g., R. malayanus, R. coelophyllus) overlap in frequency when data from all of Thailand are combined, but seem to avoid call overlap when syntopic. Hence acoustic identification can be improved by taking into account local distribution patterns and patterns of species coexistence. The creation of call libraries at a local scale would have extensive potential as a resource to monitor changes in species distributions through time.
You have requested a machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Neither BioOne nor the owners and publishers of the content make, and they explicitly disclaim, any express or implied representations or warranties of any kind, including, without limitation, representations and warranties as to the functionality of the translation feature or the accuracy or completeness of the translations.
Translations are not retained in our system. Your use of this feature and the translations is subject to all use restrictions contained in the Terms and Conditions of Use of the BioOne website.