Many butterfly and moth species persist in networks of interlinked populations, known as metapopulations1.  Habitat fragmentation is a threat to butterfly and moth metapopulations across the UK. The problems occur as large areas of habitat are split into smaller, fragmented patches as a result of land-use change, such as intensive agriculture or urban development.  Habitat fragmentation leaves species increasingly at risk from environmental change as smaller fragments of habitat mean butterfly populations are small and vulnerable, and the increased isolation of patches reduce the chance of a re-colonisation if the population in that habitat fragment goes extinct. 

Reducing the vulnerability of butterfly metapopulations is a key strategy for conserving species, and often involves management at a landscape-scale to make sites bigger, better, and more joined up2,3. However, in fragmented landscapes, there are often competing demands for space and limited scope to create new habitats, enlarge existing habitat patches or improve connectivity. Furthermore, some of our most threatened species are habitat specialists with very specific habitat needs, that are only found in particular habitats. Creating or restoring habitats for these species in highly modified landscapes can be complex and costly. 

Research by the University of Exeter and Butterfly Conservation, published in the journal Oikos4, uses the Lulworth Skipper to demonstrate how management of habitat at a site level (management within individual fragments of habitat to improve habitat suitability) can help overcome challenges in delivering landscape-scale conservation in highly fragmented landscapes. The Lulworth Skipper is an example of a habitat specialist species, and it reaches the northern most part of its range in the UK, where it is only found along the south coast of Dorset. In Dorset, this butterfly only inhabits chalk and limestone grasslands that have plenty of tall Tor-grass, which is the food plant of their caterpillars5.We wanted to understand how key habitat characteristics for the Lulworth Skipper (vegetation height and food plant cover) influenced population size on each site, then predict how habitat changes affect the distribution of the butterfly across the entire landscape because of changes in population size.

Every site in the UK with a known population of Lulworth Skipper was visited in 19785, 19976, and 20107. During visits, data was collected on whether the site was occupied by the butterfly, the population size, vegetation height, and food plant cover. We then repeated the survey in 2017/18 but included potential habitats up to 5km outside the existing distribution to detect expansion. Over the four decades of surveys, there was no evidence of significant westward or northward expansion (which might have been expected if patches that were once too cool become more suitable due to climate change). There was also no evidence of the distribution contracting and becoming smaller.

We then investigated which habitat conditions resulted in the largest and smallest population sizes of Lulworth Skipper to understand what represented ‘optimum’, ‘high’ and ‘marginal’ habitat. Population densities (number of butterflies per hectare) were largest in habitats with medium-long vegetation heights (20-30cm) and over 90% cover of the larval food plant. The optimum habitat with the highest population densities was at a vegetation height of 26.2cm and 100% cover of the larval food plant. We found that when vegetation height was under 8cm, over 46cm, or food plant cover was less than 30%, population densities were low – this represented marginal habitat.  

Projections of the distribution of Lulworth Skipper 40 years into the future were made to assess landscape-scale impacts of changing the vegetation height or food plant cover on individual sites (see figure). Projections showed that if the habitat stayed the same as it was in 2017/18  a high proportion of sites remained occupied (panel 1). However, though currently unoccupied sites outside the core distribution in the west were occasionally colonised, these sites often quickly went extinct once colonised (due to marginal habitat, small site size or isolation). If habitat in sites was changed to ‘marginal’ quality (panel 2), the Lulworth Skipper was projected to become extinct from all but a small number of ‘core’ sites and long-term survival of the butterfly in the UK was diminished. If we set the vegetation height and food plant cover in every site at ‘optimum’ (panel 3) we observed population expansion outside the current range to the new patches in the west.  As population density was at its highest under these scenarios, the chance of extinction is reduced and colonisations increased (without making physical changes to connectivity between patches).

Figure showing Lulworth Skipper habitat patches (dots) along the south coast of Dorset from Burton Bradstock in the west to Swanage.

Figure showing Lulworth Skipper habitat patches (dots) along the south coast of Dorset from Burton Bradstock in the west to Swanage in the east. The dots are coloured by how likely the site was occupied in 40 years’ time (the darker the colour the more likely) based on different habitat scenarios; 1) Observed (2017/18), 2) Marginal and 3) Optimum. 

The results have important implications for conservation management of the Lulworth Skipper. The Lulworth Skipper requires habitat management that will maintain medium-tall vegetation. Our research defines the vegetation heights and food plant cover needed to support high population densities and defines marginal habitats in which the butterfly is less likely to survive. 

Model projections show how managing habitat to improve the suitability of individual sites is a key conservation tool in landscape-scale conservation, leading to an increased population size and having a similar effect to making sites ‘bigger’ without physically enlarging the patch.  When population sizes are bigger there is more chance of dispersal to other habitat patches. Following dispersal, if the site a species arrives in has high-quality habitat, it has more chance of breeding and establishing a successful population which helps facilitate the distribution expansion of species without the need to physically connect sites (e.g. through wildlife corridors). 

The research shows how managing habitat at the site level can help to overcome some of the challenges of implementing landscape-scale conservation in landscapes with lots of competing pressures (agriculture, development, species with conflicting habitat needs). Though managing habitats within existing sites can be less costly than creating new habitat patches, landscape-scale projects3, funding, and policy (such as agri-environment schemes) are important in order to support management at the site level and encouraging a co-ordinated approach across multiple sites (including those that are small and isolated) of different ownership.

Rachel Jones
Senior Ecologist at Butterfly Conservation and PhD Researcher at the University of Exeter

We thank F. Bell, I. Clark, S. Woodley, K. Pradel, B. Goodyear and G. Pearman for their fieldwork and analysis, J. A. Thomas for supplying the 1978 data. Particular thanks also to the National Trust, the Lulworth Estate and all private landowners that supported the study and permitted access for surveys. The work was funded through a UK Natural Environment Research Council studentship (grant reference: NE/N00857X/1

1)    Hanski, I. 1998. Metapolulation dynamics. – Nature 396: 41– 49. 
Metapopulation dynamics | Nature
2)    Lawton, J. H., Brotherton, P. N. M., Brown, V. K., Elphick, C., Fitter, A. H., Forshaw, J., Haddow, R. W., Hilborne, S., Leafe, R. N., Mace, G. M., Southgate, M. P., Sutherland, W. J., Tew, T. E., Varley, J. and Wynne, G. R. 2010. Making space for nature: a review of England's wildlife sites and ecological networks. – Report to Defra, p. 107.
3)    Ellis, S., Wainwright, D., Berney, F., Bulman, C. and Bourn, N. 2011. Landscape-scale conservation in practice: lessons from northern England, UK. – J. Insect Conserv. 15: 69– 81.
Landscape-scale conservation in practice: lessons from northern England, UK | SpringerLink
4)    Jones, R., Bourn, NAD., Maclean, IMD. and Wilson, RJ. 2023. Landscape-scale dynamics of a threatened species respond to local-scale conservation management. Oikos. 
Landscape‐scale dynamics of a threatened species respond to local‐scale conservation management
5)    Thomas, J. A. 1983a. The ecology and status of Thymelicus acteon (Lepidoptera: Hesperiidae) in Britain. – Ecol. Entomol. 8: 427– 435.
The ecology and status of Thymelicus acteon (Lepidoptera: Hesperiidae) in Britain - THOMAS - 1983 - Ecological Entomology - Wiley Online Library
6)    Thomas, J. A., Bourn, N. A. D., Clarke, R. T., Stewart, K. E., Simcox, D. J., Pearman, G. S., Curtis, R. and Goodger, B. 2001. The quality and isolation of habitat patches both determine where butterflies persist in fragmented landscapes. – Proc. R. Soc. B 268: 1791– 1796.
The quality and isolation of habitat patches both determine where butterflies persist in fragmented landscapes | Proceedings of the Royal Society of London. Series B: Biological Sciences (
7)    Jones, R. E., Pradel, K., Woodley, S. and Bourn, N. A. D. 2013. The changing status of the Lulworth skipper butterfly Thymelicus acteon in Dorset between 1997 and 2010. Butterfly Conservation report no. S13–11. Butterfly Conservation, Wareham.