Effectiveness of Fourier transform near-infrared spectroscopy spectra for species identification of anurans fixed in formaldehyde and conserved in alcohol: A new tool for integrative taxonomy

Abstract

Integrative taxonomy takes into account multiple perspectives to delimit the units of diversity, which is fundamental for communication in ecological studies and conservation actions. Near-infrared spectroscopy is efficient for identification of species in many taxa and may be a viable alternative to the descriptions used in classical taxonomy as it is highly cost-effective, rapid, and can be non-destructive. Animal specimens are often conserved in alcohol and formaldehyde, which hinders some techniques. In this study, we tested the effectiveness of near-infrared spectroscopy in recognizing species of fixed specimens of closely related anuran species when the raw data were used and after preprocessing to reduce noise due to variation in surface structure. We used individuals of adult frogs belonging to different species, sizes, and time conserved in alcohol. The tests were divided into spectral readings of the belly and back of each sample in wet and predrying manipulations. The results indicated that the presence of superficial alcohol in individuals does not impede the use of the technique to recognize anuran species. We recognized anuran species using spectra collected on the back or belly with up to 100% correct identification in raw data or preprocessed data. These results open up numerous possibilities for taxonomic investigations using individuals of zoological collections worldwide, but this requires the creation of open-access data banks and availability of spectra of all species. © 2020 Wiley-VCH GmbH