The study of the diversity of animal signals on within- and among-species levels is the key to uncover mechanisms that shape the evolution of communication systems. However, the methods used to quantify acoustic diversity (like repertoire size) lack to grasp several aspects of acoustic diversity. Here, we propose a new framework for the study of animal communication, in which we decompose the acoustic diversity with the methodological toolbox from community ecology.

We explore how different diversity estimates reflecting different aspects of acoustic diversity can be applied to characterise the complexity of acoustic signals. We propose that this approach can be used in a wide range of animal taxa to derive further insights about the function and evolution of communication systems besides the traditional methods.

To illustrate the use of our approach in a case study, we used the song of collared flycatcher (Ficedula albicollis) as a model system. Based on three frequency and time variables, we calculated three diversity indies (FRic, FEve, and FDiv) to characterise the distribution of song elements (syllables) in the acoustic parameter space. We aimed to uncover the interrelations of diversity indices, reveal the degree of among-individual consistency, and investigate their relationships with certain aspects of individual quality.

We found that the chosen diversity indices were largely independent from each other and showed different consistency patterns that were timescale dependent indicating different signalling potential of individual-specific attributes. We also found that FEve strongly related to the age of males. Our case study showed that decomposing the diversity into different components can reveal additional biologically meaningful aspects of birdsong.

The dataset contains the acoustic measurements of syllables from 20 collared flycatchers (Ficedula albicollis) and the diversity indices for all individuals calculated on the min-max standardized values of the duration, mean frequency, and frequency band of the syllables. Syllables from 20 songs were included from each individual. For detailed data handling, see the methods in the related paper.

db_syllable_measurements.csv file contains the measurements of the syllables from 20 individual, where the columns are:

-recID -> individual ID
-dur -> duration of the syllable
-fmean -> mean frequency of the syllable
-fmin -> minimum frequency of the syllable
-fmax -> maximum frequency of the syllable
-fband -> frequency band of the syllable (fmax-fmin)

db_diversity_indices.csv file contains the diversity indices for the 20 individual for which we provided the acoustic measurement, where the columns are:

-recID -> individual ID
-FRic -> Functional Richness
-FEve -> Functional Evenness
-FDiv -> Functional Divergence

We used the modified R script of multidimFD.R by Sebastien Villeger to calculate these diversity indices.
The original script is available at http://villeger.sebastien.free.fr/Rscripts.html.
The modified script included in this data package doesn't standardize the FRic between individuals with the highest value.