Get access

Wildcat occurrence in Scotland: food really matters

Authors

  • André P. Silva,

    Corresponding author
    • Universidade de Lisboa, Centro de Biologia Ambiental, Faculdade de Ciências de Lisboa, Lisboa, Portugal
    Search for more papers by this author
  • Kerry Kilshaw,

    1. Wildlife Conservation Research Unit, Department of Zoology, University of Oxford, Recanti-Kaplan Centre, Tubney, Oxfordshire, UK
    Search for more papers by this author
  • Paul J. Johnson,

    1. Wildlife Conservation Research Unit, Department of Zoology, University of Oxford, Recanti-Kaplan Centre, Tubney, Oxfordshire, UK
    Search for more papers by this author
  • David W. Macdonald,

    1. Wildlife Conservation Research Unit, Department of Zoology, University of Oxford, Recanti-Kaplan Centre, Tubney, Oxfordshire, UK
    Search for more papers by this author
  • Luís M. Rosalino

    1. Universidade de Lisboa, Centro de Biologia Ambiental, Faculdade de Ciências de Lisboa, Lisboa, Portugal
    2. Laboratório de Ecologia Isotópica/CENA, Universidade de São Paulo, Piracicaba, SP, Brasil
    Search for more papers by this author

Correspondence: André P. Silva, Universidade de Lisboa, Centro de Biologia Ambiental, Faculdade de Ciências de Lisboa, Ed. C2, Campo Grande, 1749-016 Lisboa, Portugal.

E-mail: pintodasilva.a@gmail.com

Abstract

Aim

European wildcat Felis silvestris silvestris (Schreber, 1775) populations are suffering considerable threats, making conservation action a priority. In Scotland, the establishment of Special Areas of Wildcat Conservation (SAWC) have been recommended; however, few studies have addressed wildcat ecological requirements in this region. Our goal was to identify the environmental determinants limiting wildcat occurrence at a broad scale in Scotland.

Location

Scotland, Europe.

Methods

We examined data from the recent Scottish wildcat survey (2006–2008) and the National Biodiversity Network (NBN) Gateway. Presence and pseudo-absence data from 71 sampling units (10 km × 10 km) were used to generate models explaining probable wildcat occurrence. Models were built based on three pre-established hypotheses using generalized linear models (GLM-Logit).

Results

European rabbit presence, high rodent diversity and the prevalence of large grassland areas were positively associated with probable wildcat occurrence. Heather moorland, sampling units with few grassland patches or secondary watercourses and higher elevation ranges were associated with probable wildcat absence. We found no evidence that forested areas or human disturbance were influential.

Main conclusions

Our results suggest that wildcats may benefit from heterogeneity within the landscape matrix, reinforcing the idea that the wildcat is not primarily a forest species (as has traditionally been inferred from studies of the species elsewhere). We conclude that less mountainous areas, with a diverse landscape including woodland and grassland supporting rabbits and a diversity of small rodents are a priority for wildcat conservation efforts.

Ancillary