In this workshop, we will teach participants how to collect long-duration acoustic data using the AudioMoth recorder, and how to use a novel soundscape approach to capture species richness patterns from big acoustic data without the need for species identification.
Collecting long-duration acoustic data has become increasingly common to monitor biodiversity trends. Passive Acoustic Monitoring (PAM) offers several advantages over traditional biodiversity monitoring methods, such as a reduced cost and human effort, and the potential to increase the spatial and temporal scale of sampling. Yet, obtaining species-level information from big acoustic data for broad spatio-temporal scales or wide taxonomic breadth presents numerous difficulties, such as the time-consuming and knowledge-demanding nature of aural annotation, and the paucity of automated species identifiers and reference databases for most taxa and regions. To overcome this hurdle, the field of soundscape ecology attempts to infer ecological information from the soundscape – i.e. the collection of biological, geophysical and human-produced sounds emanating from a landscape – without the need for species identification. The discipline uses the variation of acoustic traits in the landscape to understand ecological processes and nature-human dynamics across spatial and temporal scales. The Acoustic Niche Hypothesis states that acoustic space is a core ecological resource for which soniferous species compete, leading to the partitioning of the acoustic niche in the time-frequency domain to avoid overlap in sound production. Thus, a more speciose community should lead to increased competition and partitioning of acoustic niche space, which can be measured by quantifying the diversity of sounds in the acoustic trait space. In this workshop, we will provide a theoretical and practical overview of soundscape diversity research. First, we will provide a hands-on introduction to collecting acoustic data using the low-cost full-spectrum AudioMoth recorder (https://www.openacousticdevices.info/audiomoth). Second, we will use an existing acoustic dataset to introduce a novel workflow that aims to quantify and decompose the soundscape diversity into its various components. Building on the analytical framework of Hill numbers, we will introduce three metrics that capture different aspects of soundscape diversity: (i) soundscape richness; (ii) soundscape diversity; (iii) soundscape evenness. To demonstrate the potential ecological applications of these metrics, we will assess how they behave along an ecological gradient in the richness of sound-producing organisms. At the end of the workshop, participants will know: (i) how long-duration acoustic data can be collected using the AudioMoth recorder; (ii) the challenges of deriving taxonomic information from big acoustic data: (iii) how soundscape ecology can overcome some of these challenges; (iv) how to process big acoustic data to derive the soundscape richness, evenness and diversity using Hill numbers; (v) how these metrics can be used to capture patterns of species richness in an acoustically complex tropical rain forest environment.
A soundscape approach to analyzing big acoustic data
Sunday July 10th-Morning (4h, 8:00am-12:00m)