Acoustic Entropy Index vs. Acoustic Diversity Index

[Danny Zurc]

Hello.

My name is Danny Zurc, and I’m doing my master’s thesis on bioacoustics of Colombian bats. I come from a systematics background so I am new to seewave and R in general. I’m interested in computing bioacoustics diversity indices on my sound files. However I am confused on how exactly are the Total Entropy calculated by seewave (function H) and the Acoustic Diversity Index (function acoustic_diversity) different. From what I gather, they’re both based on Shannon’s entropy but they’re clearly giving me very different values. Could someone here explain?

Thank you so much.

[jeromeinsect]

Hello,

I actually understand the confusion between all these indices.

The index H was first introduced by Sueur et al (2008). It is made of two sub-indices, namely Ht and Hf. Ht computes the Shannon evenness of the temporal signal (on the Hilbert amplitude envelope) and Hf computes the Shannon evenness of the mean spectrum obtained after computing a STFT (ie a spectrogram).

Later Pijanowki et al (2011), Villanueva-Rivera et al (2011), Pekin et al (2012) derived an index named H' or ADI based on Hf. If I am not mistaken (Luis might confirm) this index mainly differs in that it is not scaled between 0 and 1 as Hf is (Shannon diversity versus Shannon evenness), in the control of the frequency binning obtained with the STFT and also in the use of an amplitude threshold as detailed here and there. 

Jerome

[Luis J. Villanueva]

Danny and list,

First, to clarify. The ADI was not based on Hf.

Second, the main difference is that H is measuring entropy in both time and frequency space, while ADI is measuring it in frequency space only. We were analyzing insanely long files (~15 min), and we noticed that the variation in time was very small, most of the differences between sites and time of the day were in the frequencies occupied. So, we ended up making bins of 1000Hz and counting everything above a threshold, to avoid counting the faint sounds and noise. 

I hope this helps. Feel free to contact me if you need more info.

Luis J. Villanueva
http://research.CoquiPR.com

--
You received this message because you are subscribed to the Google Groups "seewave" group.
To unsubscribe from this group and stop receiving emails from it, send an email to seewave+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Danny Zurc]

Hi Jerome/Luis,

Thank you for taking the time to explain this. I understand better now what the conceptual differences between these two indices are. 

Best,

Danny