Long-term assessment of the success of a European badger reintroduction

Wildlife reintroductions are largely considered a major tool in conservation wherever natural re-colonisation is impossible or unlikely (Tear et al. 1993, Ostermann et al. 2001). Despite extensive use, there is still no general agreement on how to measure the success of a reintroduction (Seddon 1999, Estrada 2014). Based on the goal of a viable population after reintroduction (Caughley and Gunn 1996, Dodd 2005), a species-specific, sufficient amount of time should have passed prior to a reintroduction being regarded as “biologically successful” (Seddon 1999, Fischer and Lindenmayer 2000, Albrecht et al. 2011). However, although essential for assessing the efficacy of most reintroduction projects of vertebrates (Semlitsch 2002, Muths and Dreitz 2008, IUCN/SSC 2013), long-term monitoring has seldom been accomplished (Sutherland 2000, Estrada 2014).

Being widespread throughout Europe and western Asia (Kranz et al. 2016), the European badger (Meles meles L., 1758) has been the object of only two reintroduction attempts (Roper 2010). Following widespread decline of badger populations throughout the Netherlands, between 1992 and 1994 a total of 21 captive-reared individual badgers were reintroduced into two areas, in the provinces of Overijssel and Friesland, respectively. The high proportion of badgers leaving the area of release and high mortality hampered the project in the first area, while five animals succeeded in settling in Friesland (Mulder 1996). Overall results led the Authorities to concentrate conservation efforts on habitat defragmentation measures (Vink et al. 2008).

With the aim of restoring the natural biodiversity of the protected area, the badger was reintroduced into the Regional Park of Montevecchia and Curone Valley (Lombardy, northern Italy), where it had been extirpated by human persecution since the 1980s (Balestrieri et al. 2006). The Park (23.6 km², 200–550 m a.s.l.) – covered by deciduous mixed forests (Castanea sativa, Quercus petraea, Q. peduncolata; 30% ca. of the total surface), pastures (20%) and, in the southern section, intensely urbanised and cultivated (maize, wheat) areas -, is surrounded by an intensively urbanised belt, which had been judged an insurmountable obstacle to the natural recolonisation of the protected hill by badgers (Balestrieri et al. 2016).

Between 2001 and 2004, four badger family groups (a total of 12 animals) were trapped from setts dug into railway embankments 20–40 km from the Park, and moved to an acclimatisation enclosure built in the middle of the release area. Post-release monitoring showed that, at the end of 2004, badger signs were spread over a ca. 10 km² area, suggesting that the project had succeeded in re-establishing the species (Balestrieri et al. 2006). Nonetheless, badgers are relatively long-living animals, and the outcome of the reintroduction program immediately after the releases may have been drastically different from that assessed following a longer period (Estrada 2014). With the aim of assessing the status of the badger population 10 years after the end of the reintroduction programme, we monitored the protected area by camera-trapping. Monitoring was carried out in the northern and central sections of the Park, which were considered more suitable to badgers, between 15 April and 22 June 2015, in order to record cubs.

For each session, 10 unbaited, digital scouting pocket cameras (SG550 with Passive Infra-Red motion sensor) were deployed for 10 days within a 2 km² large unit, for a total surveyed area of ca. 14 km² (Figure 1). As the range of reintroduced badgers varied between 100 and 200 ha (Balestrieri et al. 2006), grid size was deemed appropriate to monitor badger occurrence efficiently (Kays and Slauson 2008).

Figure 1:

Distribution of camera-traps positive and negative for badgers 10 years after their reintroduction in the Regional Park of Montevecchia and Curone Valley (Lombardy region, N Italy).

The 1×1 km grid was superimposed on the kilometric grid of digitalised, 1:10,000 Regional Technical Maps.

To space cameras as uniformly as possible, for each unit the distance between contiguous cameras was calculated as the inverse of the square root of the ratio between the number of trapping sites (10) and the size of the unit (2 km²) (Rowcliffe et al. 2008). Anyway, as spatial autocorrelation is irrelevant even when cameras are placed much closer to each other (Kays et al. 2011), the resulting distance (0.45 km) was borne in mind as a general rule, subordinate to the deployment of cameras proportionally to each habitat area (woods, cultivated fields, shrubs and pastures), the availability of suitable trapping-sites and accessibility. Cameras were tied to trees 30–50 cm above ground level, possibly along animals’ trails, and set to record 15 s long videoclips, with a 30 s interval between two successive recordings. Consecutive records of the same species at the same site were deemed independent when there was at least 30 min interval between them (Kelly and Holub 2008). Signs of badger presence – setts, latrines and paths – were also recorded and georeferenced.

Assuming that badger home ranges are delimited by hard boundaries, such as hill ridges, major roads, urbanised areas or other inhospitable terrain (Powell and Mitchell 1998), their size could be broadly assessed on the basis of both the distribution of setts and signs and results of camera-trapping. In hilly areas of northern Italy, home range borders as assessed by Dirichlet tessellations have been demonstrated to match well with both geographical barriers/features and the location of latrines (Balestrieri et al. 2016).

During a total of 577 trap-days, we collected 671 independent videos of vertebrates, of which 94 were of the European badger, 128 of the red fox (Vulpes vulpes L., 1758) and 35 of the stone marten (Martes foina Erxleben, 1777). Badgers were recorded by 20 camera-traps, spread over 12 out of the 15 grid-squares totally or partially surveyed (Figure 1).

Overall trapping success – one badger videoclip per 6.1 camera-days -, was similar to that obtained, by the same methods, for an isolated hilly area of the central plain of the River Po (Lombardy region; one videoclip per 6.4 camera-days, Balestrieri et al. 2016), where badger density was assessed to be rather high with respect to mean values available for most European countries (0.93–1.4 adult individuals/km²; Balestrieri et al. 2016). Moreover, trapping success was higher than that recorded in riparian woods of the Po plain (one videoclip per 8.6 camera-days; authors’ unpubl. data).

Eighteen videoclips recorded cubs, providing evidence of the occurrence of at least two litters, one ranging in the northern tip of the Park and one in the north-eastern section of the study area. Litter size was two and three cubs, respectively. Average values for litter size range between 2.4 (southern England; Neal 1977) and 2.5 (Switzerland; Do Linh San et al. 2003).

Ten latrines and four badger setts were recorded (Figure 2), of which one coincided with the sett dug by the first badger group released in the Park (Balestrieri et al. 2006).

Figure 2:

Distribution of badger signs of presence (camera-trap records, setts and latrines) and estimated badger home ranges in the Regional Park of Montevecchia and Curone Valley (Lombardy region, N Italy).

Based on the distribution of badger videoclips and signs of presence, the occurrence of five badger social groups could be assessed, the home ranges of which may be broadly drawn as shown in Figure 2.

Considering that, in northern Italy, group size averages 2.75 badgers/group (Balestrieri et al. 2016), the current badger population in the northern section of the protected area would include ca. 14 adult individuals, corresponding to a density of 0.9–1.15 individual/km².

According to Seddon (1999), any reintroduction includes a sequence of three major aims: the survival of released individuals, breeding by the reintroduced stock and its offspring, and persistence of the re-established population. Camera-trapping allowed to demonstrate that, 10 years after the end of the reintroduction project, badgers still occur in the reintroduction area and have successfully colonised all suitable habitats. Breeding probably occurs regularly, as suggested by both the increase in badger numbers and range since 2004, and records of litters.

Several factors can affect reintroduction success – the number per unit area and year, origin (wild-caught vs. captive-reared) and genetic diversity of founding individuals, habitat suitability, breeding behaviour (early breeders with large litters vs. late breeders with small litters) and diet (herbivores vs. carnivores and omnivores; Griffith et al. 1989), making it hard to compare different projects and reinforcing the need for long-term assessment monitoring. The success of badger reintroduction in our study area is likely to have depended on the translocation of wild-caught, pre-established social groups, which may have reduced post-release erratic movements and enhanced both acclimatisation and reproductive success of the first generation (see also Balestrieri et al. 2006, Roper 2010). Dispersal from the release site was also limited by the gentle-release technique, as reported for reintroduced martens (Davis 1983). Although, in general, the release of a few individuals per year might be too low to expect the establishment of a viable population (Williams et al. 2007), small releases have been demonstrated to be effective also for other mustelids (e.g. Lutra lutra L., 1758; Prigioni et al. 2009).

The current distribution of badgers is an a posteriori confirmation of the habitat suitability estimate obtained by the feasibility study (Balestrieri et al. 2006).

Based on available knowledge on badger range-size and sett density in northern Italy (Remonti et al. 2006a,2006b, Balestrieri et al. 2016), the reintroduced population may have saturated available habitats. If, on one hand, habitat saturation supports the hypothesis that badgers have overcome the delicate phases of “establishment” and “growth”, which are associated to high risk of reintroduction failure (Robert et al. 2015), on the other hand, anthropogenic barriers that surround the protected hill may currently hinder the dispersal of offspring.

Following the Dutch example (Vink et al. 2008), the construction of wildlife corridors, guiding fences and underpasses should be implemented to allow the safe crossing of major roads and railways. Defragmentation measures would reduce traffic casualties and allow badgers to colonise suitable surrounding areas, such as the wooded hills which raise north of the Park.