Exploring Avian Soundscapes in Ancient and Restored Woodlands

Exploring Avian Soundscapes
I had the pleasure of staying on Kilchoan Estate for three weeks in March 2024 to collect data for my undergraduate dissertation in Animal & Conservation Biology at Edinburgh Napier University. My study focused on the fascinating world of ecoacoustics, exploring bird community differences between ancient and restored woodlands (Fig. 1), and the possibility of monitoring birds using audio recorders, comparing their detection performance to traditional survey methods.

Figure 1. The ancient woodland (left) and the restored woodland (right).

The focus of my project

Monitoring biodiversity trends is essential amidst ongoing declines, as well as to assess habitat restoration progress. New technologies are providing alternative methods for ecological monitoring, including the use of audio recorders. These recorders offer a promising solution due to their cost-effectiveness and non-invasiveness. All sounds, ranging from birdsong to the whistling of winds, collectively form an environment’s soundscape – its auditory landscape. Researchers are exploring the potential of ecoacoustics, a field examining the soundscape of environments, in ecological monitoring. Birds are particularly suitable for ecoacoustic surveys due to their vocal nature, and they also serve as valuable bioindicators, providing insights into ecosystem health.

My research objectives were to:

  1. Evaluate variations in bird species richness, diversity and community composition between the ancient and restored woodland.
  2. Assess differences in bird species detection between human observers and acoustic recorders.
  3. Investigate how well acoustic indices predict bird diversity patterns.

Data collection and processing

Each day, my coursemate Arianna, who kindly assisted me with my fieldwork, and I ventured out at sunrise, the peak time for bird activity. We conducted point counts along designated transects in the ancient and restored woodland, recording all birds heard and seen within a specified radius and time frame before moving to the next point. Additionally, I deployed AudioMoth recording devices (Fig. 2) to capture the woodland soundscapes throughout the morning. Despite their compact size, AudioMoths have impressive recording capabilities, and can capture audio for days, even weeks, on a single charge.

An AudioMoth recording device
Figure 2. An AudioMoth recording device attached to a tree branch.
When returning from the field, I analysed the recordings in a software called Audacity. Listening to the recordings and viewing spectrograms (Fig. 3) helped me identify birds by their songs and calls, allowing me to create a species list for each recording. Additionally, I generated acoustic index values from the sound files. Acoustic indices summarise the complex characteristics of acoustic data into a single numerical value, so the value from a recording with many vocalising bird species will be very different from a quieter recording. These representative values may facilitate comparisons of soundscapes across different habitats or over time, providing valuable insights into changes within the bird community and the overall soundscape.
An annotated spectrogram
Figure 3. An annotated spectrogram showing the songs of a coal tit and a blue tit. The left scale shows the frequency in hertz, and the top scale shows the time in minutes and seconds.

Initial findings and planned analysis

The Eurasian chaffinch was the most common species in both woodlands, followed by the Eurasian blue tit in the ancient woodland and the European robin in the restored woodland. While a similar number of bird species occurred in both woodlands, the specific species differed. During my surveys, treecreepers, long-tailed tits and great spotted woodpeckers were only found in the ancient woodland, whereas bullfinches and dunnocks were exclusive to the restored woodland. I will now conduct statistical tests to determine if there are differences in the number of bird species, their diversity and the bird community composition between the ancient and restored woodland. Even though the species richness was similar, I expect the bird diversity and community composition might vary between the two habitats.

Some of the bird species

Figure 4. Some of the bird species found during my surveys. Despite the chaffinch being the most commonly recorded species, I didn’t manage to capture a single photo of one. From left to right: Great tit, dunnock, Eurasian treecreeper.

Moreover, I plan to compare the number of species detected by the acoustic recorder with those detected by human observers to assess the efficacy of the recorder. If the recorders prove equally adept, they could be a valuable tool due to their ability to operate for extended periods and potentially capture elusive species. Additionally, I will validate whether acoustic indices are representative of the actual bird species richness and bird diversity at the two woodland sites, which would further enhance the utility of the recorders.

Exploring the surroundings

When I wasn’t conducting fieldwork or analysing audio files, I took the opportunity to explore the estate, as there was so much to see! The trees were covered in beautiful mosses and lichens, lots of plants were awakening after the winter, and buzzards, golden eagles and white-tailed eagles soared overhead. I was also fortunate to spot a couple of porpoises, and observed an otter hunting for its lunch in the loch (Fig. 5).

wildlife encounters
Figure 5. Two amazing wildlife encounters!
My three weeks on the estate were truly enjoyable, and I want to extend my thanks to all the staff, with a special thanks to Luke and Laura for their guidance during my stay. I really look forward to following the inspiring work that is being done at the estate, and I hope to return one day.

Animal & Conservation Biology student at Edinburgh Napier University