Pine martin Image courtesy of Sarah McGuire geograph.org.uk CC BY-SA 2.0 license
The pine marten in western Europe is not dependent on closed-canopy woodland, unlike eastern European populations (Pereboom et al., 2008; Mergey et al., 2011), and it occurs in areas with as little as 4% forest cover (Balharry, 1993). In Scotland, the pine marten is adapted to a landscape with low levels of forest cover; the highest recorded population densities occur in areas with intermediate levels of forest fragmentation (Caryl et al., 2012; Kubasiewicz, 2014). It is also recorded in areas with very low levels of forest cover in the north west Highlands (Balharry, 1993), and in non-wooded habitats such as upland montane areas, semi-natural grassland, and heathland in the Cairngorms (Croose et al., 2013; Moll et al., 2016). High pine marten densities are also recorded in the Irish midlands (3.13km-2), where woodland is particularly sparse and fragmented (Sheehy, 2013). In such regions, home ranges are larger to incorporate the resources required for resting and foraging (Balharry, 1993). The species is also adaptable, and may be able to supplement the resources provided by woodlands, such as denning sites and foraging opportunities, with features found in other habitat types (Caryl et al., 2012).
The dietary composition of the pine marten in Scotland varies seasonally according to the availability of different food sources, including small mammals, carrion, berries and insects (Caryl, 2008). There is a strong preference for the field vole as a primary prey item — in contrast to the preference for bank voles displayed by eastern European populations (Caryl, 2008). This preference is reflected in the incorporation of scrub and tussocky grassland into the home range (Pereboom et al., 2008; Caryl et al., 2012). Milder winters and higher availability of rodents has been linked to higher densities of pine martens in mainland Europe (Zalewski and Jedrzejewski, 2006). These factors may affect population density more than the availability of woodland habitat.
Pine martens were once prevalent throughout mainland Britain. However, by the late 19th century, only a few populations in the north west of Scotland survived (Langley and Yalden, 1977; Ritchie, 2015). Some recovery of suitable habitat, followed by legal protection (Wildlife and Countryside Act (1981); protection for the species was enacted in 1988), has led to a partial recolonisation of the Scottish range over the last few decades (Croose et al., 2013; Croose et al., 2014).
Status - Native
Conservation status. -• IUCN Red List (GB: LC; England: [CR]; Scotland: [LC]; Wales: [CR]; Global: LC). • National Conservation Status (Article 17 overall assessment 2013. UK: Favourable; England: Bad; Scotland: Favourable; Wales: Bad).
Estimates of pine marten density and home range size are all taken from sites in Scotland dominated by coniferous forest, including varying degrees of plantation and semi-natural habitat. Sites also contain some broadleaved woodland. All density estimates were applied to both coniferous and broadleaved woodlands. The population estimate therefore only represents individuals associated with woodland, however expert opinion suggests that in Scotland this will be most of the population (Johnny Birks, pers. comm.).
Several papers found during the literature search contained estimates of pine marten home range size as opposed to density (Balharry, 1993; Bright and Smithson, 1997; Halliwell, 1997; Caryl et al., 2012). Home range size has previously been used as a proxy for density, with ranges being assumed to be contiguous and without overlap within each sex. This approach was also used for the purpose of the current review.
The population size estimate for Wales is based on the number of animals translocated to Wales from Scotland during the 2015 and 2016 Pine Marten Recovery Project (Vincent Wildlife Trust, pers. comm.). Extensive research by the Vincent Wildlife Trust suggests that records in England do not indicate an established population. Therefore, no estimate has been made for England.
Two papers containing population density estimates were identified by the literature search. These reported estimates made between September and November (Kubasiewicz, 2014; Croose et al., 2015). A further four papers reported pine marten home range sizes, based on studies of at least one year (the specific timings of individual capture and tracking were not specified) (Balharry, 1993; Bright and Smithson, 1997; Halliwell, 1997; Caryl et al., 2012). Two papers contained information relevant to occupancy by reporting the percentage of surveyed hectads found to contain pine marten scats in east and central Scotland (25%; Croose et al., 2013) and southern Scotland (4%; (Croose et al., 2014). However, the surveys were conducted with relatively low sampling frequency, and an unusually high proportion of DNA extracted from scats could not be identified to species (48%; Croose et al., 2013). The surveys were also conducted at the edge of the species’ range. It is therefore concluded that these reported occupancy rates are unlikely to be representative.
Pine marten kits typically emerge from the natal den in late June and disperse from their mother’s territory between September and mid-November (Harris and Yalden, 2008). The calculated population sizes therefore represent means for the year, with some bias towards the post-breeding population.
Our analysis is restricted to woodland habitats, and this will have tended to underestimate the population size. However, this error is considered unlikely to be serious because most of pine martens in Scotland are thought to incorporate woodland into their home range (Johnny Birks, pers. comm.). Potentially more serious is the lack of occupancy data, and the consequent assumption that pine martens are present in all woodlands within the geographical range: the population size is therefore likely to be overestimated.
The density estimates found in the literature (n=11) were applied to all woodlands within the species’ range. Although most woodland within the species’ distribution is coniferous (75%). deriving separate density estimates for coniferous and broadleaved woodland would be unlikely to improve the estimate materially, because pine martens have large home ranges and use a matrix of different habitats.
The highest densities of pine martens in Scotland were recorded in areas with 20%-35% forest cover (see Kubasiewicz, 2014), but our calculations do not take the importance of local habitat composition into account. Population sizes in areas of high forest cover will therefore tend to be overestimated, and the converse will be true in areas of intermediate cover. Given that average forest cover in Scotland is 17%, these errors are expected to lead to an underestimate of population size. Further surveys to clarify these relationships are recommended, as conclusions are currently based on relatively low sample sizes. Experts consulted for this report suggested that the population size is most likely to be closer to the upper confidence limit of 8,900 individuals (Laura Kubasiewicz, pers. obs.)
The population size given by Harris et al. (1995) was 3,650, comprised of <100 in England, 3,500 in Scotland and <50 in Wales. The estimate for Scotland was based on home range size as a proxy for density: the total area of woodland within the species’ distribution was divided by the area of woodland (1.26km2) found within an average pine marten territory of 4-10km2. Outside the core range in the Highlands, percentage occupancy of 50% was assumed, although no empirical data were available (Balharry, 1993). The current analysis derived similar population estimates to Harris et al. (1995), but it is unclear whether the population is stable because a different methodology was used: densities measured in woodlands with varying degrees of fragmentation were multiplied by the area of woodland within the range (Scotland only). Although these density estimates are likely to be too high because non-woodland habitats in the home ranges were excluded, the calculations were only applied to woodland. This error is unlikely to be serious provided that woodland forms a core part of the home range of most pine martens, and that detectability is good. However, if much of the population lives independently of woodland, or pine martens have low detectability in woodland, then the estimates will be too low. Further study is needed to distinguish between these two possibilities.
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 (see Sections 2.3 and 32.3 for further details). 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 population size that still falls within the confidence limits of the original. Comparisons between the two reviews are therefore unlikely to be affected materially by these methodological issues.
The geographical range estimate for Britain is similar to that reported by the last Article 17 Report (Joint Nature Conservation Committee, 2013a), but is considerably larger than that reported by Arnold (1993). The area of occupancy for England does not represent established populations.
Pine martens have continued to increase their range in Scotland in the last 20 years (figure 8.3a; Croose et al., 2013; Croose et al., 2014), and the median density estimate used for the current analysis is larger than both of the estimates provided by Harris et al. (1995). The population is, therefore, highly likely to have increase.
Future trends. Population and Range increasing. Habitat stable.
There is a page on this site about the Pine Martin, habitat,lifestyle and breeding . Click on the relevant content banner above. Scroll down to view.