Mapping Habitat Suitability for the Hazel Dormouse to Assist Reintroduction Site Selection
Student Spotlight - Melanie Baker
Master's Research at the University of Liverpool
Email: melanie.baker@liv.ac.uk Twitter: @Me1anieBaker Instagram: @melaniebaker
Lab Group Website: https://www.liverpool.ac.uk/mammalian-behaviour-and-evolution/
Read the Paper: https://doi.org/10.1111/csp2.544
Introduction
Both my personal and research interests have always focussed on reintroduction biology, and I have so far had the pleasure to be involved in two very different, but equally important, projects involving the reintroduction of native UK rodents. I am currently a first year PhD student at the University of Liverpool studying the ecological impacts of beaver (Castor fiber) reintroductions in Cheshire, however, my research journey began as a Masters of Research student studying a much smaller British rodent, the hazel dormouse (Muscardinus avellanarius). Here, I looked into ways to model habitat suitability and how this can be used to assist in the planning of future reintroduction projects. This research is part of my current research group, the University of Liverpool’s Mammalian Behaviour and Evolution, and was linked to the Cheshire Wildlife Trust, and the People’s Trust for Endangered Species (PTES). It was part of a joint study with Emma Cartledge (click here to read her blog) which was finally published last year in Conservation Science and Practice (Cartledge and Baker et al., (2021).
Research Background
The hazel dormouse is the UK’s only native dormouse species. Here, despite various levels of legal protection, populations are still struggling with reports of up to 5.8% annual declines (Goodwin et al., 2017). They are classified as Vulnerable on the Red List for Britain’s Mammals, but perhaps they may soon be classed as Endangered (Scopes et al., 2022). Since the 1990s, a huge amount of effort by the PTES has gone into monitoring 100s of dormouse sites and they have recently reintroduced their 1,000th dormouse. Reintroduction sites have experienced varying levels of success, where the causes of failures are not always clear (White, 2019). One potential is the reintroduction site’s habitat suitability. As habitat specialists, dormice require specific conditions to sustain healthy populations. They have historically been associated with mature, ancient woodland (Bright and Morris, 1992) and evidence suggests populations require at least 10 hectares of habitat (White, 2019). Finding sites that provides all the necessities for dormice is not an easy task. The hopes of my research were to try and alleviate some of the difficulties of reintroduction managers by highlighting woodlands that may contain suitable dormouse habitat.
Habitat Suitability Modelling
Using a method of modelling called Ecological Niche Factor Analysis (ENFA), I created a map of England that could highlight areas of most suitable and least suitable regions of England for dormice. The model requires presence data which was obtained from the National Dormouse Monitoring Project (NDMP) and represents 410 monitoring sites that had recently found adult dormice. Alongside this, information about the environment is provided from open source databases which represented 51 various habitat descriptors such as land cover like grasslands and woodlands, topography like altitude and terrain, anthropogenic factors like roads and buildings, and water bodies like lakes and rivers. When the model runs it takes information from these habitat layers at the points where dormice have been found and compares it to the rest of the country. Eventually, this produces a map of habitat suitability. We can see this in the map below where the darker regions are least suitable, and the lighter regions are areas of most suitable habitat. Clearly the most suitable habitat can be found within the South of England, and this is where we suggest future reintroduction should be focussed.
However, this project was linked to the Cheshire Wildlife Trust, so we wanted to demonstrate the use of this model for the county of Cheshire. To do this we extracted habitat suitability scores from the above map for every broadleaved woodland over 10 hectares in Cheshire and categorised them into most suitable, marginal and least suitable (see the map below).
When it comes to reintroductions site recommendations several approaches can be taken. Either focussing on the largest sites, such as the site with the green star on the map. Or choosing areas where multiple sites may be able to be joined together to form metapopulations, such as the cluster of smaller woodlands in the centre of the map. As this is a model, these points are suggested as starting points for reintroduction site selection. It is important that these sites be assessed in the field by reintroduction managers, for more detailed habitat qualities that remotely sensed data cannot provide, such as diversity of plant species and the ability to manage the woodland.
Assessment of Current Reintroduction Sites
A quick mention here for the other section of the aforementioned paper which was conducted by Emma Cartledge. Emma focussed on the current reintroduction sites to try and explain their population trends. Using the top 10 habitat maps described previously and information collected from the NDMP (time since reintroduction, number of animals reintroduced, number of reintroduction events, survey season, site coordinates and size of reintroduction site), Emma ran generalized linear mixed models to identify factors that most strongly influence reintroduced dormouse populations. Emma found that time since reintroduction decreases the likelihood of finding an adult dormouse in a nest box and you are more likely to find dormice when surveying in autumn over spring or summer. Results highlighted further the importance of broadscale habitat factors. In contrast to the habitat modelling, the chance of finding adult dormice at reintroduction sites decreased with the proportion of surrounding broadleaved woodland. The chance of finding an adults dormouse decreased with increased slope and increased proportion of arable land.
Final Notes
Although the publication of this paper ended my research into hazel dormice, as I had completed my MRes, it is always good to reflect on what future research could be beneficial. The R package used in this modelling technique was actually designed for climactic data. I believe it would be useful to take advantage of this to see how dormice may respond to a changing climate across the UK.
Finally, the hope of every conservation biologist is for their work to make a positive impact, so please do get in touch with Emma or myself if you think this work may be beneficial to you! We are always happy to help.
Acknowledgements
My project changed drastically due to the covid lockdown, and I would like to thank my supervisor, Paula Stockley and my co-author Emma Cartledge for all their support in producing this piece of work. A huge thanks also to the Cheshire Wildlife Trust, volunteers involved with the NDMP and the PTES for providing the data and opportunities for this research. On Emma’s behalf thankyou to her PhD funders from the Natural Environment Research Council as part of the ACCE Doctoral Training Program and the Cheshire Wildlife Trust.
References
Bright, P.W. and Morris, P.A., (1992). Ranging and nesting behaviour of the dormouse Muscardinus avellanarius, in coppice‐with‐standards woodland. Journal of Zoology, 226(4), pp.589-600.
Cartledge, E.L., Baker, M., White, I., Powell, A., Gregory, B., Varley, M., Hurst, J.L. and Stockley, P., (2021). Applying remotely sensed habitat descriptors to assist reintroduction programs: A case study in the hazel dormouse. Conservation Science and Practice, 3(12), p.e544.
Goodwin, C.E., Hodgson, D.J., Al‐Fulaij, N., Bailey, S., Langton, S. and Mcdonald, R.A., (2017). Voluntary recording scheme reveals ongoing decline in the United Kingdom hazel dormouse Muscardinus avellanarius population. Mammal review, 47(3), pp.183-197.
Scopes, E., Goodwin, C., Al-Fulaij, N., White, I., Langton, S., Walsh, K., Broome, A. and McDonald, R., 2022, April. When is a dormouse ‘Endangered’? Continued population decline of Hazel Dormice (Muscardinus avellanarius) in Great Britain.(Poster). In ARPHA Conference Abstracts (Vol. 5, p. e84738). Pensoft Publishers.
White, I., (2019). Hazel Dormouse Reintroduction Sites Report 2019. People’s Trust for Endangered Species. Available at: https://ptes.org/wp-content/uploads/2019/07/20190717_Dormouse-Reintroduction-Report-2019.docx.pdf. (Accessed on: 27 November 2019).