By Dr. Marina (Nina) D. A. Scarpelli, Prof. Paul Roe, Dr. David Tucker, and Assoc. Prof. Susan Fuller (Queensland University of Technology, Faculty of Science, Brisbane)

Acoustic sensors have been used for decades to monitor audible fauna. In recent years, advances in passive acoustic sensors which can be left unattended in the field, and decreased costs in hardware and storage solutions, has made it even more appealing to collect environmental sounds. The deployment of multiple sensors, which may be used for targeted research questions or broad-scale initiatives such as the Australian Acoustic Observatory (A2O), allows data to be collected simultaneously in a standardised manner, with the potential to answer a wide range of ecological questions.

Soundscapes, or the sounds of a landscape, can provide valuable ecological information, particularly for processes that drive changes in biodiversity and community dynamics over time. Temporal changes can be captured because acoustic sensors may be left to record continuously for long periods of time. Combining acoustic data with other remotely sensed information can be used to determine associations between changes in fauna and other environmental factors. Satellite imagery and weather stations can provide information about climatic factors and seasonal vegetation change, and allow questions about phenology, or the study of recurrent biological phenomena, to be answered.

Many soundscape studies still focus on identifying species, a process that is extremely time consuming and labour intensive. An alternative approach to analysing environmental sounds is based on techniques that provide more generalised information on recorded fauna, such as taxonomic groups and categorical sound sources (e.g., birds, insects, wind, rain, car). This high-level classification can be valuable for comparing recordings that have different sets of species but share broad groups across seasons and locations. We have recently demonstrated how soundscapes can be used to investigate phenological questions and explore the effects of climate change on animal communication.

We investigated one year of recordings from a single acoustic sensor belonging to the A2O network in subtropical woodland in South-east Queensland (see image 1 below). The dominant taxonomic groups were insects and birds, with both showing clear 24-hour and seasonal acoustic patterns. Insects such as crickets dominated the night soundscape, and birds were the most active group during the day. We also found distinct seasonal differences in the soundscape, with cicadas very active during the day in summer when it was hot and humid, high wind levels over winter, and an increase in bird activity for March, August, and October (see image 2 below). Several environmental factors were shown to be related to variation in calling activity. For example, higher temperatures were associated with a decrease in bird calls, while moon illumination was associated with higher calling activity for nocturnal insects.

Subtropical woodland where the data were recorded (Samford Ecological Facility – QUT) – image from: Samford Ecological Research Facility (64) – Australian Acoustic Observatory | A20

1 day per month of soundscape recordings examples. Each image corresponds to one False-colour Spectrogram with 24-hours of recordings. Time is represented by the x-axis while frequency (in kHz) is represented in the y-axis. Some features are marked by coloured boxes, being: red – cicadas; green – birds; pink – crickets; blue – wind.

We also investigated the possible effects of increasing temperatures and climate change on insect and bird calling activity. We compiled 10 years of temperature data for the study location and compared the temperatures from the year when the recordings were collected with the historic average. Unsurprisingly, we found that higher-than-average temperatures were associated with a decrease in calling activity for both nocturnal insects and birds across different seasons. There was no change in cicada calling activity with changes in temperature.

In this study we demonstrated the utility of using soundscapes to understand ecological processes linked to climatic and seasonal patterns. The practicality of using passive acoustic sensors and their ability to capture large volumes of soundscape data provide a valuable tool for monitoring biodiversity and tracking changes in ecological communities over time. However, it is also important to consider other environmental factors when deploying sensors, and not underestimate the importance of local influences and patterns (e.g., fine-scale flowering events, thunderstorms, human activity). Other remote sensing technologies such as satellite imagery and weather stations can assist with understanding these effects on calling activity.

By analysing the seasonal patterns of environmental sound, we can gain insights into what drives biodiversity and community dynamics over time. These insights are important for understanding and predicting how climatic change and extreme weather events can influence biodiversity and underpin our ability to effectively implement possible mitigation and recovery actions in the future.

The complete paper can be found at: https://doi.org/10.1016/j.scitotenv.2023.163080