Geographic Variation in Acoustic Signals in Wildlife: A Systematic Review

Aim: Our main aim is to understand the geographic variation of acoustic signals in wildlife, its spatial, temporal and taxonomic extent, the methodological approaches used in the scientific literature, and the main drivers of geographic variation. Also, specifically for birds, we aim to understand the effect of learning behaviour and vocalisation type (i.e., call vs. song) on the geographic variability of their acoustic traits. Location: Worldwide. Methods: We performed a systematic literature review and identified 306 peer-reviewed articles covering geographic variation in animal acoustic signals. Results: The studies were carried out in 93 countries, but most of them focussed on northern temperate areas and on a single country. The articles used mainly birds as study models (53.6%), followed by mammals (27.8%), anurans (10.5%), insects (7.2%), fishes (0.7%) and reptiles (0.3%). Most articles studied the existence of geographic variation among populations by comparing differences in signal features, such as frequency, duration or a combination of several signal features through ordination methods. The proportion of case studies (i.e., comparisons of different acoustic traits, with some studies including more than one comparison) finding geographic variation in acoustic signals was quite similar among all taxonomic groups (range 75.2%–82.7%) but differed between males and females (80.7% vs. 60.0%, respectively). We also identified a higher proportion of case studies finding acoustic differences among geographic regions when using as response variable multiple signal features (i.e., through ordination axes), followed by the use of acoustic signal type (i.e., categorical classification) or frequency. Besides, 69.7% of the case studies that assessed the relationship between acoustic similarity and geographic distance found a negative and significant pattern among these variables, while this was true only for 60.0% and 66.7% of the case studies assessing the relationship between acoustic similarity with genetic distance and ecological variables, respectively. Finally, when focussing on birds, there was a higher probability of finding geographic differences when using songs (84.2%) as the target vocalisation than when using calls (78.3%) and when considering as target species acoustic learners (83.4%) than when focussing on nonlearners (78.0%). Main Conclusions: Overall, our review shows that our current knowledge of geographic variation in wildlife is biased towards birds from northern temperate areas and that most monitored wildlife populations differ in their acoustic traits to some extent, which is more often related to geographic distance than to genetic distance.

E.S.G. received the grant RYC2019- 027216-I funded by MCIN/AEI/10.13039/501100011033 and by ESF Investing in your future. C.P.- G. acknowledges the support of the Ministerio de Educación y Formación Profesional through the Beatriz Galindo Fellowship (Beatriz Galindo—Convocatoria 2020)