Methods - Soundscape Ecology: Principles, Patterns, Methods and Applications - page 241

signal digitization depth
(8 or 16 bits). To obtain M values varying from 0 to 1, M has been standardized
as M / A ( t ) max.

where A ( t ) ¼ amplitude envelope and depth ¼

9.7.2.3 Acoustic Richness (AR)

This index is the combination of the indices described in the section 9.7.2.1 and
9.7.2.2 (Sueur et al. 2008 )

ð rank HðÞ rank M ðÞ Þ

AR ¼

n 2

with 0 M 1

9.7.2.4 The Acoustic Dissimilarity Index (D)

This index measures the β diversity between two acoustic communities (Sueur et al.
2008 ). D is composed of two subindices: D t , the temporal dissimilarity index and
D f , the spectral dissimilarity index: D

¼ D t D f

for D [0,1]

∈

X

n

A 1 t

D t ¼ 0 : 5 ðÞ A 2 t ðÞ

t¼ 1

with D t ∈ [0,1]

X

N

D f ¼ 0 : 5 j S 1 f ðÞ S 2 f ðÞ j

f ¼ 1

with D f ∈ [0,1]

where A 1 ( t ) and A 2 ( t ) ¼ probability mass function of the amplitude envelope

and S 1 ( f ) and S 2 ( f ) ¼ probability mass function of the mean spectrum.

9.7.2.5 The Acoustic Complexity Index (ACI)

This index, elaborated by Farina andMorri ( 2008 ) and by Pieretti et al. ( 2011 ), is applied
to a spectral representation of the sound signals registered in audio files and produces
a direct and quick quantification of the soundscape structure (Figs. 9.16 and 9.17 ).
The computation, starting from a .wav file, is carried out by the SoundscapeMeter
(Farina et al. 2012 ), a plug-in of WaveSurfer (Sj¨lander and Beskow 2000 ).
The hypothesis on which the ACI algorithm has been created is based on the
observation that many biophonies, such as bird songs, are characterized by an