Bioacoustics as a tool for red squirrel conservation
This post is greater than 6 months old - links may be broken or out of date. Proceed with caution!
By Will Cresswell, PhD candidate University of Bristol
Bioacoustic monitoring is a burgeoning field in conservation with great potential coming from technological advances. This technique focuses on listening to the natural soundscape and inferring the presence of certain species based on what can be heard. During my PhD project at Bristol University (supported and funded by Rainforest Connection, the Mammal Society, and Huawei) we aim to develop this framework as a monitoring technique for the conservation of red squirrels.
Red squirrels are widespread in European temperate woodlands, however in the UK they face sharp declines, being classed as endangered in the Mammal Society’s Red List for Britain’s Mammals. The underlying cause of this decline is interspecies competition with the invasive grey squirrel. Grey squirrels suppress Red populations through resource exploitation and disease-mediated competition, due to their role as carriers of squirrelpox virus. This has caused red squirrel ranges to decline and grey ranges to expand in the areas they overlap. Conservation action is urgently required to stop further declines and to prevent the eventual extinction of red squirrels in the UK.
The most important requirement for effective conservation action is information, and for red and grey squirrels this is certainly the case. Data is needed to understand where grey squirrel ranges are expanding to and to inform management to protect Red ‘strongholds’ where the populations are holding on. Traditional, manual survey methodologies are labour intensive, requiring experienced observers. As well as this, when using indirect indicators like dreys and signs of feeding signs you are unable to accurately discern the two species. To avoid these issues camera traps are often used, set up in front of feeding stations (see pic. 1) although these setups also have their own shortcomings. Camera traps are power/data hungry and as a result require frequent visits to change batteries and SD cards making long-term monitoring more difficult (and expensive). More importantly there is a risk associated with using feeding stations in areas where Reds and Greys overlap (the areas that are most in need of monitoring) as these sites can act as a point of squirrel pox transmission. Acoustic monitoring has potential to solve many of these issues as it is completely non-invasive, recorders can be low-cost and able to record autonomously for long periods of time. There is also scope to automate large parts of data analysis.
In our project we are testing exactly this by attempting to use soundscape recordings to survey for the two species of squirrel and particularly to monitor for grey squirrel invasions into Red stronghold sites in the Yorkshire Dales and the Lake District. To collect these recordings we are testing a combination of methods. Firstly ‘AudioMoths’ a small, cheap battery powered recorder developed by British research group Open Acoustic Devices, designed specifically for bioacoustic monitoring (see pic. 2). Secondly, we are using recorders designed and built by our project partners Rainforest Connection. Originally designed to automatically detect illegal logging in tropical rainforests by listening for the sound of chainsaws, these recorders have 3G connectivity allowing for data to be sent remotely and solar panels to keep the recorders running with little-to-no maintenance (see pic. 3). Using the gathered soundscape recordings, I look for instances of the two squirrel species vocalising to demarcate their presence. In the future, we hope to develop AI models to automate extracting these instances of vocalisations.
Red Squirrel Loud Chatter and Grey Squirrel Quaa
In order to identify when the squirrels are vocalising a deep understanding of their vocal repertoire and behaviour is required and this is a major hurdle for the project. Throughout the project we have gathered an extensive collection of recordings of both captive and wild individuals of the two species and have been able to carefully analyse them and develop a library of their different call types. Many assume these are two fairly quiet species, however alone and undisturbed they can both be very vocally active. The grey squirrel is (fittingly for its larger stature) much louder and more shrill in its vocalisations. A sound most would recognise as familiar is their ‘Quaa’ vocalisation, a drawn-out scream-like sound somewhat similar to a Crow’s call. The loudness and identifiable waveform signature of this call makes it very useful in acoustic monitoring. Red squirrels are much softer in their vocalisations. Reds have some loud calls such as those used when engaged in mate-chases in the breeding season. However, a more common call seen when two individuals are interacting is a highly variable chatter call. This call sounds highly modulated almost like a machine beeping and whirring and is an extremely engaging call to hear. The fact that the two species are so distinct in the vocalisations they produce limits any confusion between the two species and helps to make this a viable way to monitor the two species.
There are two more years of fieldwork scheduled for this project and we hope to by the end of this period have outlined a useful methodology for red squirrel conservation, as well as having developed a deeper understanding of red squirrel communication – a relatively unexplored aspect of the scientific literature.
You can contact Will on di18446@bristol.ac.uk
The Mammal Society
Established in 1954 we are a UK based charity devoted to the study and conservation of the mammals of the British Isles. We aim to raise awareness of the issues mammals face and share our scientific research so they can be best protected in the future. Although a small team we are mighty, with six dedicated staff members helping mammals across the nation.
More from The Mammal Society