Skip to main content
Dryad logo

A framework for the quantification of soundscape diversity using Hill numbers

Citation

Luypaert, Thomas (2022), A framework for the quantification of soundscape diversity using Hill numbers , Dryad, Dataset, https://doi.org/10.5061/dryad.1vhhmgqtq

Abstract

This is the data underlying the case study and supplementary material described in Luypaert et al. (2021): A framework for the quantification of soundscape diversity using Hill numbers. 

Methods

Acoustic data were collected at the Balbina Hydroelectric Reservoir (BHR) in Brazilian Amazonia. Surveys were conducted between July and December 2015, sampling 151 plots on 74 islands and 4 continuous forest sites on the mainland along a known gradient in taxonomic species richness. At each plot, an acoustic recorder was deployed at 1.5 m height and set to record the soundscape for 1-min every 5-min for 5-9 consecutive days at a sampling rate of 44.1 kHz using the ARBIMON Touch application. From these plots, a subset of sites was taken for which previously published data on taxonomic richness of sound-producing species was available. Moreover, several plots were omitted due to insufficient recording quality, persistent noise or positioning in riparian habitat. Ultimately, 65 plots located on 31 islands and three continuous forest sites were retained. If multiple plots were present on the same island/forest site, data was averaged across plots, resulting in a final sample size of 35 sites (32 islands and 3 continuous forest sites). For more detailed information on data collection, see Bueno et al. (2020): Sampling design may obscure species-area relationships in landscape-scale field studies

Usage Notes

What data can be found on this repository?

  • The raw sound files for all 151 plots collected at the Balbina Hydroelectric Reservoir in Brazilian Amazonia. See "Bueno et al. (2020): Sampling design may obscure species-area relationships in landscape-scale field studies" for a detailed description of the data collection protocol. 

  • The processed spectral CVR-index calculated using the QUT AnalysisPrograms.exe software. 

  • The RMarkdown files for the case study presented in the main manuscript

  • The RMarkdown files for the additional analyses presented in the supplementary materials

  • The full '.RData' file associated with each '.Rmd' file

Calculating the spectral CVR-index values yourself

The spectral index CVR-files were calculated using the AnalysisPrograms.exe software of the QUT Ecoacoustics group in Brisbane, Australia (https://ap.qut.ecoacoustics.info/tutorials/01-usingap/practical?tabs=windows), and can be downloaded from this repository.

The functions used in the analysis of this manuscript are currently under development and will be published as the 'soundscapeR' R-package. For the moment, the most recent version of the functions can be found on the repository of this manuscript on GitHub (https://github.com/ThomasLuypaert/SoundHill). 

If you wish to calculate the spectral CVR-index values yourself, follow these steps:

  1. Download the QUT AnalysisPrograms.exe software from here ()
  2. Load the 'soundscapeR' functions from GitHub (https://github.com/ThomasLuypaert/SoundHill/blob/main/soundscapeR_functions.R). Note that you might need to install some of the packages on which these functions depend. 
  3. Open up the 'Case_study.Rmd file. In section 2.1, you can find the code to calculate the indices (commented out). Activate the code by selecting it and hitting CTRL-C. Fill out the required paths (location of sound files on your device, and location of the AnalysisPrograms.exe software on your device). Run the code. Be mindful that this will take a while. 

Funding