Animals can encode information within acoustic signals, particularly, bird songs can be remarkably complex and can indicate individual identity and quality. Two main sets of hypotheses attempt to explain the evolution of increased birdsong complexity across large-scale geographic ranges: (1) larger acoustic space availability, and (2) greater sexual selection intensity, both of which would favor the evolution of more complex songs at higher latitudes, more seasonal and/or species-poor environments. However, few studies have assessed patterns of song complexity for birds with broad geographic ranges. Here, we determined patterns of song variation in the blue-black grassquit (Volatinia jacarina), considering metrics of song complexity, structure and performance. This Neotropical bird occurs from Mexico to Argentina and produces a monosyllabic song. Using recordings from online databases, we calculated song metrics, such as bandwidth, song rate, number of song components, and proportion of vibratos of this signal. We found that song features varied with latitude, climate seasonality, bird species richness and hemisphere. However, contrary to theoretical predictions, complexity mostly decreased with latitude and greater seasonality, while it was positively correlated with bird species richness. Proportion of vibratos was positively correlated with latitude and seasonality, and may be a feature under sexual selection in this species. Overall, our results did not support the main hypotheses proposed as explanations for song complexity. Our findings also highlight that song complexity does not vary uniformly among songbirds and song parameters, and future studies encompassing more species should clarify patterns and drivers of song variation across broad geographic dimensions.

Data on the blue-black grassquit song complexity, structure and performance were collected from recordings available in online databases from citizen science projects and audiovisual collections, i.e., Xeno-Canto (www.xeno-canto.org), Macaulay Library at the Cornell Lab of Ornithology (www.macaulaylibrary.org), WikiAves (www.wikiaves.com.br) and Fonoteca Neotropical Jacques Vielliard (FNJV) (www2.ib.unicamp.br/fnjv). We also collected climate and bird species richness data from online databases, such as WorldClim v2 (www.worldclim.org) and BiodiversityMapping.org website. To calculate the acoustic metrics of the blue-black grassquit song, we inspected spectrograms of the recordings and selected the beginning and the end of up to three songs per recording, and we also selected uninterrupted song bouts from 15 s to 120 s. We used these selections to calculate different song metrics, using the Raven Pro 1.6 (Cornell Lab of Ornithology, www.ravensoundsoftware.com) and the R environment (i.e., warbleR package).

R codes for the statistical analyses of the manuscript: https://doi.org/10.5281/zenodo.5513930 Funding provided by: Coordenação de Aperfeiçoamento de Pessoal de Nível SuperiorCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100002322Award Number: 001Funding provided by: Coordenação de Aperfeiçoamento de Pessoal de Nível SuperiorCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100002322Award Number: 88887.469218/2019–00Funding provided by: Conselho Nacional de Desenvolvimento Científico e TecnológicoCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100003593Award Number: 471945/2013–7Funding provided by: Programa de Pós-Graduação em Ecologia*Crossref Funder Registry ID: Award Number: Funding provided by: Programa de Pós-Graduação em EcologiaCrossref Funder Registry ID: