A new methodology to infer the singing activity of an avian community: The Acoustic Complexity Index (ACI)

https://doi.org/10.1016/j.ecolind.2010.11.005Get rights and content

Abstract

The animal soundscape is a field of growing interest because of the implications it has for human–landscape interactions. Yet, it continues to be a difficult subject to investigate, due to the huge amount of information which it contains. In this contribution, the suitability of the Acoustic Complexity Index (ACI) is examined. It is an algorithm created to produce a direct quantification of the complex biotic songs by computing the variability of the intensities registered in audio-recordings, despite the presence of constant human-generated-noise. Twenty audio-recordings were made at equally spaced locations in a beech mountain forest in the Tuscan-Emilian Apennine National Park (Italy) between June and July 2008. The study area is characterized by the absence of recent human disturbance to forest assets but the presence of airplane routes does bring engine noise that overlaps and mixes with the natural soundscape, which resulted entirely composed by bird songs. The intensity values and frequency bin occurrences of soundscapes, the total number of bird vocalizations and the ACI were processed by using the Songscope v2.1 and Avisoft v4.40 software. The Spearman's rho calculation highlighted a significant correlation between the ACI values and the number of bird vocalizations, while the frequency bin occurrence and acoustic intensity were weaker correlated to bird singing activity because of the inclusion of all of the other geo/anthro-phonies composing the soundscape. The ACI tends to be efficient in filtering out anthrophonies (such as airplane engine noise), and demonstrates the capacity to synthetically and efficiently describe the complexity of bird soundscapes. Finally, this index offers new opportunities for the monitoring of songbird communities faced with the challenge of human-induced disturbances and other proxies like climate and land use changes.

Introduction

Animals acoustic cues are influenced by the physical and energetic context (see f. i. Richards and Wiley, 1980) in which organisms live and, consequently, probably reflect these circumstances. The possibility to measure the sounds produced by animals represents the obligatory route to approach the ecological dynamics of this informative structure (Derryberry, 2009, Farina et al., accepted for publication).

New tools for monitoring natural systems have been recently provided by technological advances such as the opportunity to automatically record sounds from animal communities (f. i. birds: Celis-Murillo et al., 2009, Haselmayer and Quinn, 2000, Rempel et al., 2005, Scott et al., 2005; mammals: Gedamke and Robinson, 2010; amphibia: Meek, 2010). Audio-recordings have many benefits, for instance the irrelevant disturbances caused by the operators during field surveys and the opportunity to remotely listen the recorded sounds and post-process the collected acoustic information. The approach can also minimize observer errors by using a single interpreter, thus providing a potentially permanent record of surveys and solving the logistical problems that are often encountered in field studies, including the limited availability of expert ornithologists (Hobson et al., 2002).

Audio recordings are the basis of acoustic ecology, a recently developed ecological field of research that focuses on the relationship between the sounds of the environment (the soundscape) and the listener. The soundscape, defined as any acoustic environment, whether natural, urban, or rural, can be composed of three fundamental elements: the biophony (non-human biological sounds such as the vocalizations of birds, amphibians and other animals), the geophony (physical features of the environment such as the wind blowing through a forest or the burbling of water in a stream), and the anthrophony (human-induced noise from whatever source) (Krause et al., 2003, Pijanowsky et al., accepted for publication).

According to Schafer (1977), who was its initial promoter, the soundscape approach suggests that we should try to hear the natural acoustic environment as a musical symphony and, further, that we own responsibility for its composition and preservation from its bigger enemy: the pollution produced by noise. In recent years in particular, the increase in human mechanical noise (the anthrophony) has tended to mask the fine textures of the natural acoustic environment, forcing vocal creatures to displace or activate unusual adaptations (Dooling and Popper, 2007, Nemeth and Brumm, 2009, Rheindt, 2003, Slabbekoorn and Peet, 2003).

Nowadays, the study of the animal soundscape represents a field of growing interest because of the implications it has for the assessment of human–landscape interactions. Unfortunately, it continues to be a difficult subject to investigate, due to the wide variety of information available in each acoustic environment, and the difficulty that there is in identifying indices with which to quickly interpret the wide range of audio-registration data. Indeed, besides all the new technologies and software developed in recent years, the analysis of the natural sounds often remains a time-consuming process and this reduces the possibility to extract useful ecological data from this widely rich informative context. Several authors have successfully produced techniques based on bio-acoustic cues of a single species (e.g. Bardeli et al., 2010, Frommolt et al., 2008, Klinck et al., 2008, Wolf, 2009), while indices have been rarely calibrated for the monitoring of entire animals communities (e.g. Sueur et al., 2008a, Sueur et al., 2008b).

In particular, an attempt to quickly measure the singing behaviour of a bird community was made by Farina and Morri (2008) who elaborated an index to rapidly quantify the typical complexity of the biotic songs of a soundscape, despite the presence of constant human-generated-noise. The aim of the present work is to further test this methodology by using the index, which we describe in this paper with the acronym ACI (Acoustic Complexity Index).

Section snippets

Acoustic Complexity Index (ACI)

The Acoustic Complexity Index, elaborated by Farina and Morri (2008), was created in order to produce a direct and quick quantification of the birds vocalizations by processing the intensities registered in audio-files.

The hypothesis on which the ACI formula is based lays on the observation that many biotic sounds, such as bird songs, are characterized by an intrinsic variability of intensities, while some types of human generated noise (such as car passing or airplane transit) present very

Acoustic environment at the study area

The soundscape of all of the 20 stations was dominated by bird song (biophony) and the anthropogenic noise from the airplane flight paths which crossed the study area (anthrophony). On average, 14 airplane transits for every 2 h of recording were registered. Fig. 3 sets out a spectrogram of an audio scene polluted by this kind of anthropogenic noise. Other typical audio-scenes are provided as Supplementary Content, together with their spectrogram representation (Wavesurfer v.1.8.5 Sjölander and

Discussion and conclusions

Human activities are often the cause of dramatic changes to populations of wild animals, and we therefore need basic information about the extent of this impact if we are to make decisions about the most efficient ways of ensuring their conservation. Although methods and techniques are still a work in progress, the analysis of biotic sounds and their interrelations could produce new interpretations of community coalescence mechanisms and represent a new tool with which to monitor the complexity

Acknowledgements

We are extremely grateful to Prof. Catia Grimani for her assistance with the mathematics’ terminology of the index and her encouragement. We also thank the Editor F. Muller and two anonymous referees for helpful comments on the manuscript which were much appreciated.

References (31)

  • R. Bardeli et al.

    Detecting bird sounds in a complex acoustic environment and application to bioacoustic monitoring

    Pattern Recognition Letters

    (2010)
  • J. Gedamke et al.

    Acoustic survey for marine mammal occurrence and distribution off East Antarctica (30-80(E) in January–February 2006

    Deep-Sea Research Part II: Topical Studies in Oceanography

    (2010)
  • A. Celis-Murillo et al.

    Using soundscape recordings to estimate bird species abundance, richness, and composition

    Journal of Field Ornithology

    (2009)
  • Derryberry, E.P., 2009. Ecology Shapes Birdsong Evolution: Variation in Morphology and Habitat Explains Variation in...
  • R.J. Dooling et al.

    The Effects of Highway Noise on Birds

    (2007)
  • Farina, A., Lattanzi, E., Malavasi, R., Piccioli, L., Pieretti, N., accepted for publication. Bird soundscape and the...
  • Farina, A., Morri, D., 2008. Source-sink e eco-field: ipotesi ed evidenze sperimentali. Atti del X congresso nazionale...
  • K.H. Frommolt et al.

    Advantages and Disadvantages of Acoustic Monitoring of Birds. Realistic Scenarios for Automated Bioacoustic Monitoring in a Densely Populated Region

    Computational Bioacoustics for Assessing Biodiversity. Proceedings of the International Expert Meeting on IT-based Detection of Bioacoustical Patterns, vol. 234

    (2008)
  • J. Haselmayer et al.

    A comparison of point counts and sound recording as bird survey methods in Amazonian southeast Peru

    Condor

    (2000)
  • K.A. Hobson et al.

    Acoustic surveys of birds using electronic recordings: new potential from an omni-directional microphone system

    Wildlife Society Bulletin

    (2002)
  • H. Klinck et al.

    Detection of leopard seal (Hydrurga leptonyx) vocalizations using the Envelope-Spectrogram Technique (TEST) in combination with a Hidden Markov model

    Canadian Acoustics – Acoustique Canadienne

    (2008)
  • Krause, B., 1999. Loss of Natural Soundscapes Within the Americas. Wildlife Sanctuary Inc., 10pp. (Online) Retrieved 12...
  • B. Krause et al.

    Testing Biophony as an Indicator of Habitat Fitness and Dynamics. Sequoia National Park (SEKI) Natural Soundscape Vital Signs Pilot Program Report

    (2003)
  • B.F. Manly

    Randomization, Bootstrap, and Monte Carlo Methods in Biology

    (1997)
  • P.D. Meek

    An automated data logger for simultaneously recording amphibian vocalizations with weather variables

    Pacific Conservation Biology

    (2010)
  • Cited by (411)

    View all citing articles on Scopus
    View full text