Connecting soundscape to landscape: Which acoustic index best describes landscape configuration?
Introduction
Soundscape ecology is the study of sounds in the landscape (‘soundscape’) and is based on how sounds from biological, geophysical and anthropogenic sources can be used to understand natural and human systems at multiple temporal and spatial scales (Pijanowski et al., 2011a). Biophony, geophony and anthrophony are terms used to characterize sounds that occur in the landscape (Pijanowski et al., 2011a). Biophony refers to the sounds produced by living organisms, usually sounds that are used by animals as a means of communication. This may include birds, amphibians, insects, mammals, fish, amphipods, and crustaceans in both terrestrial and aquatic systems. Geophony is the collection of sounds caused by physical processes such as wind, water flow, thunder, rainfall, and earth movement. The sound created when humans use mechanical devices is referred to as anthrophony (or technophony). This includes the sounds that come from stationary machines such as fans and air conditioners, and mobile machines used for transportation and construction such as aircraft, cars, trucks, boats, building cranes, bulldozers etc.
There has been considerable interest and research to develop and compute acoustic indices that represent the characteristics of the soundscape. Early research in this field led to the application of landscape metrics (reviewed in Turner, 1989) to the soundscape using acoustic diversity indices (Gage et al., 2001, Napoletano, 2004). These indices were based on the quantification of spectrogram images, calculated by dividing the spectrum into frequency bins and using automated processing of multiple spectrograms (Gage and Napoletano, 2004). A computation approach using the power density spectrum (Welch, 1967) was then developed and used to characterize temporal changes in the soundscape in Sequoia National Park (Krause et al., 2011). The computation of acoustic metrics from multiple recordings was further developed to compute soundscape power (Matlab code can be obtained from the authors). Subsequently, the normalized difference soundscape index (Joo, 2009, Kasten et al., 2012) was created to estimate the relative amount of biophony and anthrophony in the soundscape by computing the ratio of anthrophony to biophony found in field-collected acoustic recordings.
Farina et al. (2005) examined landscape ecology from a cognitive perspective and described new thinking about how organisms perceive landscapes according to signals and signs in the context of energy flows within the landscape. The acoustic complexity index was developed based on the observation that many biotic sounds, such as bird songs, are characterized by an intrinsic variability of intensities, while human-generated noise is often constant in intensity (Pieretti et al., 2011). Pieretti et al. (2011) found that this index correlates with the number of bird vocalizations, while efficiently filtering airplane noise. The acoustic complexity index has been used to describe avian soundscapes (Farina et al., 2011), relate avian soundscapes to vegetation complexity (Farina and Pieretti, 2014) and describe the influence of traffic noise (Pieretti and Farina, 2013). It is calculated as the average absolute fractional change in spectral amplitude, averaged over all frequency bins for the entire recording. Similarly, Boelman et al. (2007) developed a bioacoustic index which was a function of both the spectral amplitude and the number of frequency bands in a sound recording. This index was shown to be strongly correlated with avian abundance in Hawaiin forests experiencing weed invasion.
Acoustic diversity indices have also been developed to facilitate automated surveying of ecosystems for rapid biodiversity appraisal (Sueur et al., 2008b, Sueur et al., 2012). The acoustic entropy index is one such index and is computed as the product of both the temporal (acoustic energy dispersal within a recording) and spectral entropies (acoustic energy dispersal through the spectrum) following application of the Shannon index (Sueur et al., 2008b). Simulations revealed a correlation between the acoustic entropy index and species diversity and in field studies this index was found to be sensitive to disturbance in Tanzanian forests (Sueur et al., 2008b). The acoustic diversity index (Villanueva-Rivera et al., 2011) is a modification of spectral entropy and is also calculated using the Shannon index, while the acoustic evenness index uses the Gini coefficient as a measure of evenness (Villanueva-Rivera et al., 2011).
The theoretical underpinning of the application of acoustic indices is that communities with more audible species have a greater acoustic diversity and that biodiversity will correlate positively with acoustic diversity (Gage et al., 2001, Qi et al., 2008). Despite the existence of a suite of acoustic indices, few comparative studies have been undertaken. Towsey et al. (2014) provided a thorough investigation of multiple indices relative to a comprehensive avifauna census dataset. However, the focus of their study was to develop a computer assisted sampling methodology to obtain a more efficient estimate of species richness than random sampling alone, rather than to evaluate acoustic indices relative to landscape condition or configuration.
While it has been proposed that there is an intrinsic relationship between the soundscape and the landscape (Pijanowski et al., 2011b), there have been few studies that have tested this explicitly (see Bormpoudakis et al., 2013, Tucker et al., 2014). Furthermore, recent studies in urban environments have highlighted the importance of land use planning regarding the evaluation of the soundscape using a landscape perspective (Kuehne et al., 2013, Votsi et al., 2012). However, while a range of studies have recorded and analyzed acoustic signals produced by birds, insects and other audible organisms to assess the effects of disturbance on biodiversity (Blumstein et al., 2011, Depraetere et al., 2012, Laiolo, 2010, Proppe et al., 2013, Sueur et al., 2008b), a lack of standardized methods to evaluate landscape characteristics has probably inhibited research on linking soundscape with landscape configuration. Recently, an ecological condition framework that assesses landscape characteristics has been developed to meet biodiversity offset policy demands (Eyre et al., 2011). Tucker et al. (2014) conducted an evaluation of fragmented spotted gum forests in eastern Queensland, Australia using this framework and found that there was a significant relationship between the soundscape and the size and connectedness of forest patches, but other landscape features such as road fragmentation and land use were not studied. Consequently, our study aims to investigate the patterns of six acoustic indices and relate these patterns to an array of landscape features and ecological condition in nineteen fragmented forest sites in south-eastern Australia.
Section snippets
Study sites
The study area was situated in South-east Queensland, Australia; a region characterized by a subtropical climate, fast growing population and increasing urban and peri-urban pressures including reduced native forest cover and habitat fragmentation. Nineteen sites were selected in forest patches ranging in size from 3 ha to 44,110 ha (see Supplementary Material 1 for site location details). Ten sites were located in patches of remnant spotted gum (Corymbia citriodora ssp. variegata) open forest
Time-of-day acoustic indices
The pattern of each acoustic index is plotted over hour of day during the soundscape recording period averaged over the ten spotted gum sites (Fig. 1A) and nine scribbly gum sites (Fig. 1B). The overall patterns for the same acoustic index are similar across the two forest types. The ACI and the BIO have similar patterns; they rise rapidly at the dawn chorus, decline until the evening chorus, increase slightly then decline until the dawn chorus is reached in the morning (about 0600 h). H and ADI
Discussion
Automated soundscape recording and analysis technologies have the potential to serve as a powerful conservation planning tool for measuring the influence of disturbance, fragmentation and declining ecological condition on biodiversity. While studies have shown that anthropogenic sound has a negative impact on biodiversity (e.g. Bayne et al., 2008, Proppe et al., 2013, Reed and Merenlender, 2008) and that the integration of soundscape and landscape studies can result in improved landscape
Acknowledgements
The authors gratefully acknowledge Tom Tarrant for expert analysis of bird calls, Teresa Eyre and Annie Kelly from the Queensland Department of Environment and Heritage Protection for assistance with biocondition and access to spatial data, and Peter Young for botanical advice and assistance on locating benchmark sites. We thank Jerome Sueur for advice on using seewave-R and Luis Villanueva-Rivera for incorporating the Normalized Difference Sound Index into the soundecology package. We would
References (57)
- et al.
Do vegetation patch spatial patterns disrupt the spatial organization of plant species?
Ecol. Complex.
(2009) - et al.
Monitoring animal diversity using acoustic indices: implementation in a temperate woodland
Ecol. Indic.
(2012) - et al.
Sonic environment and vegetation structure: a methodological approach for a soundscape analysis of a Mediterranean maqui
Ecol. Inform.
(2014) - et al.
The soundscape methodology for long-term bird monitoring: a Mediterranean Europe case-study
Ecol. Inform.
(2011) - et al.
The remote environmental assessment laboratory's acoustic library: an archive for studying soundscape ecology
Ecol. Inform.
(2012) The emerging significance of bioacoustics in animal species conservation
Biol. Conserv.
(2010)- et al.
Climate change meets habitat fragmentation: linking landscape and biogeographical scale levels in research and conservation
Biol. Conserv.
(2004) - et al.
A new methodology to infer the singing activity of an avian community: the Acoustic Complexity Index (ACI)
Ecol. Indic.
(2011) - et al.
The use of combinations of acoustic indices to analyse audio-recordings of the environment
Ecol. Inform.
(2014) - et al.
Distance-based assessment of open country Quiet Areas in Greece
Landsc. Urban Plan.
(2012)
Linking vegetation type and condition to ecosystem goods and services
Ecol. Complex.
Ecological neighborhoods scaling environmental patterns
Oikos
lme4: Linear Mixed-Effects Models Using Eigen and S4. R Package Version 1
Impacts of chronic anthropogenic noise from energy-sector activity on abundance of songbirds in the boreal forest
Conserv. Biol.
Acoustic monitoring in terrestrial environments using microphone arrays: applications, technological considerations and prospectus
J. Appl. Ecol.
Multi-trophic invasion resistance in Hawaii: bioacoustics, field surveys and airborne remote sensing
Ecol. Appl.
Spatial heterogeneity of ambient sound at the habitat type level: ecological implications and applications
Landsc. Ecol.
Acoustic signals, diversity and behaviour of cicadas (Cicadae, Hemiptera)
Spatial dynamics of the knob-tailed gecko Nephrurus stellatus in a fragmented agricultural landscape
Landsc. Ecol.
BioCondition: A Condition Assessment Framework for Terrestrial Biodiversity in Queensland. Assessment Methodology Manual Version 2.1
Effects of habitat fragmentation on biodiversity
Annu. Rev. Ecol. Evol. Syst.
Cognitive landscape and information: new perspectives to investigate the ecological complexity
BioSystems
Beyond fragmentation: the continuum model for fauna research and conservation in human-modified landscapes
Oikos
Envirosonics Equipment and Operations Manual
Assessment of ecosystem biodiversity by acoustic diversity indices
J. Acoust. Soc. Am.
Linear Mixed-Effects Model, Linear Mixed-Effects Models Using R
Biodiversity sampling using a global acoustic approach: contrasting sites with microendemics in New Caledonia
PLOS ONE
An overview of methods used to assess vegetation condition at the scale of the site
Ecol. Manage. Restor.
Cited by (215)
Short-time acoustic indices for monitoring urban-natural environments using artificial neural networks
2024, Ecological IndicatorsThe effect of latitude on the efficacy of acoustic indices to predict biodiversity: A meta-analysis
2024, Ecological IndicatorsComplex networks for analyzing the urban acoustic environment
2023, Ecological InformaticsComparison of alien Robinia pseudoacacia stands with native forest stands across different taxonomic groups
2023, Forest Ecology and Management