Horse power: hoofed animals as agents of rewilding

Concerns about land use and climate change have resulted in increasing interest in “rewilding” to restore habitats that have been heavily affected by past agricultural or other human-mediated practices. Such projects can be done passively by removing livestock or other grazers and relying on natural successional changes in the landscape or by actively reintroducing plant and animal species that were historically part of the ecosystem. Another strategy is to introduce large herbivores that retain traits thought to be present in their ancestors before domestication, which are thought to make them more resilient to seasonal and changing environments with minimal or no supplemental feeding. These species are often more specialist grazers compared to traditional livestock, allowing increased plant diversity, which could in turn increase the abundance of pollinators and other beneficial insects. Maintaining animals at much lower densities than commercial herds could also allow some benefits of nutrient supplementation, without overwhelming the normal soil balance or reducing microbial diversity. However, despite the theoretical and in some cases proven benefits of increasing habitat heterogeneity through grazers, few studies have yet explicitly tested the impacts of different types of co-occurring and free ranging herbivores on plant and insect biodiversity or soil health in relation to their use of the landscapes.

This multi-disciplinary project will work in collaboration with the Bamff Estate rewilding project (Bamff Wildland), which introduced beavers to their estate more than 20 years ago for habitat creation purposes; one of the first such demonstration projects in the UK. Bamff Wildland are now investing further in habitat restoration through the introduction of low-density herds of cattle, pigs and horses to sustainably transform land-use practices (Figure 1). They have already piloted the introduction of highland cattle and pigs but this project will be focused on the collective impacts of also introducing Exmoor ponies, which can be maintained as semi-wild herds, with minimal interventions beyond annual health checks. Near the start of 2024, ponies will be relocated from the Cochno estate at the University of Glasgow to Bamff, providing an opportunity to compare their diets before and after the translocation event and to track what parts of their new ecosystem they use over time. Global positioning system (GPS) collars will be used to monitor interactions between the cattle, pigs and horses to test hypotheses about habitat partitioning and synergies. For example, we can test whether different species use different pre-existing habitats, whether they terraform their own habitats into which they become entrenched, or whether they use the same habitats in different ways. Tracking the animals will also enable targeted monitoring of changes in above-and below-ground biodiversity and soil health in areas where the animals are grazing relative to a gradient of lesser used areas. The overall aim would be to obtain insights that could be used to develop stable strategies for rewilding management.

Click on an image to expand

Image Captions

Figure 1. Rewilding projects, such as Bamff Wildlands, aim to transform land use from: (a) traditional livestock grazing to (b) less intensive grazing by heartier and more independent megaherbivores, such as Exmoor ponies. Habitats can be restored by: (c) tree planting (top of the hill in image) after removing livestock from large areas; or (d) excluding herbivores from smaller areas to encourage natural regeneration. Restoration of (e) woodlands or (f) more complex habitats can be further facilitated by reintroduction of native ecosystem engineers such as beavers.


The project is multi-disciplinary because it could ultimately involve integration of ecological survey techniques, movement modelling, advanced statistical inference, and environmental DNA approaches to monitoring diet and/or changes in biodiversity using deep sequencing methods and big data analyses. Other possibilities would include monitoring animal health, welfare, and behaviour when left under semi-wild conditions. The studentship will thus form the trunk of a vertically integrated project also involving associated masters or undergraduate student projects focusing on particular aspects, and working in collaboration with existing PhD students at Bamff. Opportunities would be explored for the student to spend time at other rewilding project locations across the UK and in Europe, in order to learn and share best practices. Deborah Davy, a recent PhD student, would act as advisor about working with the horses and how best to integrate this project with other schemes that are currently under development.

A suggestion of possible activities across the 3 years of data collection is outlined below but there would be extensive potential for the student to customise the project to their own area of interests.

Project Timeline

Year 1

Undertake literature review (October – January). Extract DNA and optimise methods for diet analysis from faecal samples of ponies (collected before their move from Cochno) using deep sequencing based metabarcoding (February -April). Obtain training in bioinformatics approaches for handling the big data resulting from deep sequencing. Compare diets of ponies after translocation with those of the cattle and pigs collected in the first spring and summer (from faecal samples collected before the start of the PhD project) and autumn from faecal samples collected by the PhD student). Optimise methods for eDNA analysis from environmental samples (soil, air, water) using deep sequencing approaches. Conduct pilot analyses on data from initial GPS collars placed on the lead individual from the herds of cattle and ponies and on one of the pigs. Obtain training in modelling approaches and geographic information systems to allow interpretation of animal movement data in relation to habitat characteristics. Obtain training in vegetation and insect surveys. Put GPS collars on additional individuals in the spring.

Year 2

Continuation of data analysis from year 1 (diet analyses and movement modelling data). Produce detailed maps of land use by the ponies, horses and pigs, in order to plan targeted biodiversity surveys in the spring, using a combination of transect surveys, observational studies and eDNA analyses. Assessment of relative soil quality and microbial community structure in: 1) agricultural plots; 2) rewilding areas used by cattle, ponies or pigs; and 3) rewilding areas not as heavily used by the herbivores. Diet analysis based on samples collected at multiple timepoints throughout year 2.

Year 3

Potential for extended visits to other rewilding projects, in the UK or Europe. This could include additional sample collection for diet analysis or other collaborative initiatives. Diet analysis based on samples collected at multiple timepoints throughout year 3; integration of data across three years and preparation of manuscript for publication. Integration of movement data collected across the three years to assess patterns of land use within and between seasons and between species.

Year 3.5

All lab work and analyses should have been completed and the focus should be on writing up the results for the thesis and for publications.

& Skills

The primary supervisor team has expertise in: evolutionary genetics in relation to conservation (Mable); ecological restoration and resilience (Willby); and spatial ecology and movement modelling (Matthiopoulos). Training would be provided in each of these broad areas, as well as identifying specific gaps where the student could benefit from targeted training courses.

Field sampling and biodiversity survey techniques
Statistics and advanced inference using R or other platforms
Movement modelling using Bayesian techniques
GIS mapping
Molecular laboratory techniques

References & further reading

Matthiopoulos, J., J. Fieberg, and G. Aarts. (2020). Species-habitat associations: Spatial data, predictive models, and ecological insights, 2nd Edition. University of Minnesota Libraries Publishing. Retrieved from the University of Minnesota Digital Conservancy, https://hdl.handle.net/11299/217469.

Pettorelli, N., & Bullock, J. M. (2023). Restore or rewild? Implementing complementary approaches to bend the curve on biodiversity loss. Ecological Solutions and Evidence, 4, e12244. https://doi.org/10.1002/2688-8319.12244

Riotte-Lambert, L., and J. Matthiopoulos. (2019). Environmental predictability as a cause and consequence of animal movement. Trends in Ecology and Evolution, 35(2), 163-174. https://doi.org/10.1016/j.tree.2019.09.009

Thomassen, E. E., Sigsgaard, E. E., Jensen, M. R., Olsen, K., Hansen, M. D. D., Svenning, J.-C., & Thomsen, P. F. (2023). Contrasting seasonal patterns in diet and dung-associated invertebrates of feral cattle and horses in a rewilding area. Molecular Ecology, 32, 2071–2091. https://doi.org/10.1111/mec.16847

Apply Now