The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

Changes in the Acoustic Diversity Index

The Acoustic Diversity Index (ADI), first used by us in Pijanowski et al. 2011 and then explored in Villanueva-Rivera et al. 2011, divided the spectrogram of an audio file in 10 bins of 1kHz each, up to 10 kHz. Then, the proportion of signals above a threshold in each bin was calculated. These proportions were used to calculate the ADI.

When revising the code, we found that there was an error. The ADI was supposed to apply the Shannon's Diversity Index to the values of the bins, but the code had a different equation. It was the average of the sum of the logs:

Score1=0
for (i in 1:length(Freq)) {
    #Add a small amount to avoid log(0)
    Score1=Score1 + (Score[i] * log(Score[i]+0.0000001))
}

#Average
Score_left=(-(Score1))/length(Freq)

It seems that this draft code was erroneously kept in the script and was not changed to properly calculate the index or used a function of a package that did this. The acoustic_diversity() function now uses the diversity() function from the package vegan to calculate the Shannon's Diversity Index for the bins. However, to maintain backwards compatibility, you can obtain the result using the old code by setting the shannon parameter to FALSE.

library(soundecology)
data(tropicalsound)

#Using Shannon's Diversity Index:
acoustic_diversity(tropicalsound)

## 
##  This is a mono file.
## 
##  Calculating index. Please wait... 
## 
##   Acoustic Diversity Index: 1.727905

#Using the original code:
acoustic_diversity(tropicalsound, shannon = FALSE)

## 
##  This is a mono file.
## 
##  Calculating index. Please wait... 
## 
##   Acoustic Diversity Index: 0.189338

The results from both the original equation and the Shannon's Index from the sound recordings used in Pijanowski et al. 2011 are highly correlated (p < 2.2e-16), with values between 0.71 and 0.95 in each site. The shape of the figures did not change, only the magnitude of the values, so our conclusions were not affected.

References

Pijanowski, B.C., L. J. Villanueva-Rivera, S Dumyahn, A. Farina, B. Krause, B. Napoletano, S. Gage and N. Pieretti. 2011. Soundscape Ecology: The Science of Sound in Landscapes. BioScience 61: 203-216. doi:10.1525/bio.2011.61.3.6.

Villanueva-Rivera, L. J., B. C. Pijanowski, J. Doucette, and B. Pekin. 2011. A primer of acoustic analysis for landscape ecologists. Landscape Ecology 26: 1233-1246. doi:10.1007/s10980-011-9636-9.

Vignette “Changes in the Acoustic Diversity Index” by LJ Villanueva-Rivera

Version 1.0 (24 October 2013)

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.