Wildcat Image courtesy of Peter Trimmings CC BY-SA 2.0 Generic license Originally posted to Flickr.
Within Britain, the wildcat is now only found in Scotland. The species relies on a mosaic of habitat types, with broadleaved or mixed forest being core (Stahl and Leger, 1992), but its range in Scotland also encompasses a high proportion of coniferous woodland, with young plantations, in particular, being used because of lower deer grazing intensity and high prey densities (Kilshaw, 2011). Open areas, such as marginal farmland and grasslands, also provide hunting opportunities (Easterbee et al., 1991; Silva et al., 2013). They are important habitats in parts of the distribution (e.g., the north east of Scotland), but are avoided elsewhere (Kilshaw, 2011). At a fine scale, habitat fragmentation may be beneficial for wildcats: areas with high percentage cover from coniferous forest are avoided, whereas smaller patches of forest next to areas of grassland are used more frequently. Habitat requirements are, however, unlikely to be a limiting factor for wildcats: the main, and increasing, threat is hybridisation with domestic and feral cats (Littlewood et al., 2014; Kilshaw et al., 2016).
IUCN Red List (GB: CR; England: n/a; Scotland: [CR]; Wales: n/a; Global: LC). • National Conservation Status (Article 17 overall assessment 2013. UK: Bad; England: n/a; Scotland: Bad; Wales: n/a)
Four papers reporting over-winter estimates of population density were identified by the literature search (Table 8.1a). One of these contained replicate population density values from other papers. Two papers contained assessments of the factors affecting population density, and one — which provides the best data available on wildcat occurrence — provided information on positive and negative sites, and could therefore be used to estimate percentage occupancy (Kilshaw et al., 2016). As all of them considered multiple habitats, the occupancy value is not habitat-specific.
Kitchener et al. (2005) developed a method to distinguish between pure wildcats, hybrids and domestic cats using pelage characteristics. Whilst these characteristics correlate with genetic differentiation (Kilshaw et al., 2010), it is still difficult to classify individuals with certainty, particularly in populations with extensive introgression between wildcats and domestic cats, as is the case throughout the Scottish range (Beaumont et al., 2001; Macdonald et al., 2004b). There is a high probability that some of the density estimates and presence records included in our analysis are from feral cats and hybrid wildcats, particularly since relaxed inclusion criteria tend to be applied to camera-trap records. Population size and distribution are therefore highly likely to be overestimated.
The wildcat makes use of a mosaic of habitat types, so the population density estimates reported in the literature relate to extensive regions, rather than to specific habitats separately. It is therefore not informative to assess the proportion of the population found in each habitat type. All of the surveys that provided population density estimates focused on areas particularly suitable for wildcats, so these densities are likely to be higher than the average for the whole of the species’ range. Similarly, the study from which the occupancy data were derived (Kilshaw et al. (2016)) was conducted in areas thought likely to contain wildcats. The density and percentage occupancy values applied in the review are therefore likely to be overly optimistic.
Population size was estimated to be 3,500 in Harris et al. (1995). This estimate was produced before the development of the pelage scoring system (Kitchener et al., 2005), however, and so it includes contains hybrid wildcats. This is likely to be the largest source of error in assessing relative trends in wildcat population size. A comparison of methods is, nevertheless, provided for completeness. The population estimate in Harris et al. (1995) was based on two population density estimates (3km-2 and 0.8km-2) assigned to occupied 100km2 squares, depending on the frequency of sightings in each square. Both of these density estimates are higher than the median used for the current population size estimate.
Nationally, there are changes between the two reviews in the estimated availability of key habitats (broadleaved woodland and coniferous woodland), generated by a combination of true change and methodological differences, irrespective of any range change. The adjusting of results to reflect more probable temporal changes in the composition of the British landscape — using differences between the 1990 and 2007 Countryside Surveys (Carey et al., 2008) — generates a small increase in population size which falls within the confidence limits of the original. Methodological issues relating to changing habitat availability are therefore unlikely to influence the assessment of temporal trends.
Comparison of population sizes between the two reviews shows a sharp decline in population size: the estimate from Harris et al. (1995) is well above the higher confidence interval of the current estimate. This comparison does not, however, take any account of feral hybrids.
The population of wildcats in Scotland is widely reported to be under threat of extinction, mostly from hybridisation with feral and domestic cats. This threat has been present for much longer than the last 20 years (Stahl and Leger, 1992) but poses an increasing threat with time (Macdonald et al., 2004b; Kilshaw et al., 2016). A population size of approximately 400 was estimated in the mid-2000s by extrapolating from samples taken from free-living wildcats collected during the 1990s (Macdonald et al., 2004b). A more recent population size of 115-314 wildcats was estimated by Kilshaw (2015), which is in line with the current estimate and represents a decline since 1995 and the mid-2000.
Future trends seem to suggest Population and Range is in decline ,habitat remains stable.