The conservation of large carnivores is a formidable challenge for biodiversity conservation. Using a data set on the past and current status of brown bears (Ursus arctos), Eurasian lynx (Lynx lynx), gray wolves (Canis lupus), and wolverines (Gulo gulo) in European countries, we show that roughly one-third of mainland Europe hosts at least one large carnivore species, with stable or increasing abundance in most cases in 21st-century records. The reasons for this overall conservation success include protective legislation, supportive public opinion, and a variety of practices making coexistence between large carnivores and people possible. The European situation reveals that large carnivores and people can share the same landscape.

The central-European population of wolves Canis lupus has its western border of contiguous range in eastern Poland. Protected since 1998, Polish wolves began to expand towards the west. Based on large-scale data on wolf abundance (2000)(2001)(2002)(2003)(2004)(2005)(2006) and geographic information system (GIS) tools, we built a habitat suitability model (HSM) for the species in Poland. The best model, selected by the Akaike information criterion, of the resource selection function (RSF) by wolves included percentage cover of forests, meadows and marshes (positively correlated to wolf abundance) and road density (negatively correlated), and explained 53% of the total variation in the wolf abundance index in a 10 Â 10 km cell grid. That RSF was then used to evaluate the whole country in terms of suitability for wolves. Potential wolf range appeared to consist of 24 patches of good habitat (predicted relative probability of wolf occurrence Z30%), including six patches already occupied and 17 new ones. In total, habitat suitable for wolves covers at least 20-24% of Poland, compared with 16% of the country's area now occupied by the species. HSM was validated with historical data on wolf occurrence in 1950-2006. The areas selected by our model and those inhabited by wolves in at least one decade overlapped in 81-86%. Furthermore, the probability of wolf occurrence predicted by the model correlated positively with the number of decades the wolves were actually recorded in the area. Based on the empirical relationship between patch size and wolf numbers, we estimated that Poland could support a population of 1450-1540 wolves, two to three times larger than the current estimate. HSM needs to be supplemented with a GIS-based connectivity analysis among patches, which in turn should be validated by molecular genetics studies on dispersing wolves. Because of broadly similar geographic conditions, the model may be applicable to eastern Germany, Belarus Republic and the Baltic States (Lithuania,

Based on data collected during the National Wolf Census in 2000–01, we analysed the main habitat factors influencing the distribution and abundance of the wolf, Canis lupus, in northern Poland. The study region forms the western border of the continuous Eastern European range of wolves, although attempts at westward dispersal have been observed. Using Geographic Information System techniques, we measured nine habitat variables and three parameters related to wolf occurrence in 134 circular sample plots (radius 7 km, area 154 km2 each). We compared 72 plots where wolves were recorded and 62 plots with no signs of wolf presence. Wolf plots were characterized by significantly higher forest cover, less fragmentation of forests, lower density of villages, towns, motorways, and railways than wolf‐free plots. We found a positive correlation between the sum of wolf observations in plots and forest cover. The number of domestic animals killed by wolves was higher in areas with higher indices of wolf abundance and lower forest area. In multiple regression analysis, four independent variables explained 59% of the variation in wolf distribution and abundance in northern Poland: straight‐line distance to continuous range of wolves in Eastern Europe; forest cover; forest fragmentation; and length of major motorways. We conclude that protection of wolves in Poland (since 1998) may not be an adequate conservation measure, especially because of the increasing density of highways and express motorways. Existing forest corridors should be protected and new ones should be restored to ensure long‐term conservation of wolves and allow range expansion into Western Europe.

No abstract

Phylogeographic studies of highly mobile large carnivores suggest that intra-specific genetic differentiation of modern species might be the consequence of the most recent Pleistocene glaciation. However, the relative influence of biogeographical processes and subsequent humaninduced population fragmentation requires a better understanding. Poland represents the western edge of relatively continuous distributions of many wide-ranging species, e.g. lynx (Lynx lynx), wolves (Canis lupus), moose (Alces alces) and, therefore, a key area for understanding historic and contemporary patterns of gene flow in central Europe. We examined wolf genetic structure in Poland and in a recently recolonized area in eastern Germany using microsatellite profiles (n = 457) and mitochondrial DNA sequencing (mtDNA, n = 333) from faecal samples. We found significant genetic structure and high levels of differentiation between wolves in the Carpathian Mountains and the Polish lowlands. Our findings are consistent with previously reported mtDNA subdivision between northern lowlands and southern mountains, and add new and concordant findings based on autosomal marker variation. Wolves in western Poland and eastern Germany showed limited differentiation from northeastern Poland. Although the presence of private alleles suggests immigration also from areas not sampled in this study, most individuals seem to be immigrants from northeastern Poland or their descendants. We observed moderate genetic differentiation between certain northeastern lowland regions separated by less than 50 km. Moreover, mtDNA results indicated a southeastern subpopulation near the border with Ukraine. The observed structure might reflect landscape fragmentation and/orecological differences resulting in natal habitat-biased dispersal.