Baselines for Establishing meta-population connectivity of Eurasian lynx populations in the Alps, Dinarics and Balkan Handbook on suitability and connectivity of the space for Eurasian lynx in the area March 2024 Contents 1 Introduction 4 2 The Eurasian lynx 5 Dinaric lynx population 6 Balkan lynx population 8 3 General threats and Conservation status of the Eurasian lynx 10 4 Habitat suitability 12 Habitat fragmentation and connectivity 13 5 Isolation and Inbreeding 16 6 Lynx Movement 17 Dispersal 19 7 Methodological approaches to study habitat suitability, connectivity, and 21 viability of lynx populations 8 Modelling habitat suitability and connectivity for lynx in the Alps, Dinarics 25 and Balkan region Habitat suitability modelling and distribution of core and other suitable habitat patches 25 Extra-territorial movements 29 Connectivity 36 Connectivity of potential stepping stone patches 39 9 Conclusions 44 10 References 48 3 Introduction 1 The Eurasian lynx 2 One of the most radical changes to the European landscape in recent The Eurasian lynx ( Lynx lynx) is a middle-sized, spotted felid and one of the centuries has been the creation of vast urban and agricultural areas and four species belonging to the Lynx genus. It is considered to have one of the subsequent expansion of infrastructure networks. After more than the largest east-west distribution ranges in felids. It occurs along forested 5000 years of intensive human activity, only 2% of the original primaeval mountain ranges in South-eastern and Central Europe and from Northern and forest remains (Iuell et al. 2003). In Europe’s human-dominated and highly Eastern Europe through the Boreal forest belt of Russia, down into Central fragmented landscape, the dispersal of large mammals such as the Eurasian Asia and the Tibetan Plateau (Kaczensky et al. 2012, Nowell and Jackson 1996, lynx is hindered by natural (e.g. big rivers, deep valleys) and anthropogenic Sunquist and Sunquist 2002). The global population trend of the Eurasian barriers such as extensive urban and agricultural areas and the subsequent Lynx is estimated as stable with no severe fragmentation in the boreal range expansion of transportation infrastructure networks (Breitenmoser et al. (Breitenmoser et al. 2015). Subspecies in the southwest of its range (Europe 2000, Potočnik et al. 2023). In addition, dispersing sub-adult lynx show a strong and Asia Minor) are generally small and widely separated. The European lynx tendency to establish home ranges in areas adjacent to their conspecifics population (excluding Russia and Belarus) has been estimated at 9,000-10,000 (Zimmermann et al. 2005). Thus, the combination of anthropogenic and (Breitenmoser et al. 2015). Its native distribution stretches from Scandinavia ecological factors makes it unlikely that lynx will spontaneously colonise and Fennoscandia in the north, the Carpathian Mountains in the east and the new patches in the Alps, Dinaric Mountains and the Balkans. A priority goal southwest Balkan Peninsula. The Balkan lynx population is thought to be for lynx conservation is therefore to connect the existing lynx populations stable with only 20-39 individuals remaining (Melovski et al.2015). Densities in the Alps with the Jura and Dinaric Mountains (Molinari-Jobin et al. 2003), are typically 0.69-2.39 resident adults per 100 km², although higher densities possibly also with the populations in the Vosges, the Bohemian, Bavarian of up to 5/100 km² have been reported from Turkey, Eastern Europe and Austrian Forest and the Balkan populations, and in the long term possibly parts of Russia and lower densities of 0.24/100 km² from some reintroduced even with the remnant populations in the Carpathians (European Commission populations and from Scandinavia (Jedrzejewski et al. 1996, Sunde et al. 2000, 2013). Natural dispersal alone would probably not be sufficient to establish Schmidt et al. 2011, Pesenti & Zimmermann 2013, Avgan et al. 2014, Gimenez this interconnectivity, making translocations and reintroductions necessary et al.2019, Dula et al. 2021, Mengüllüoğlu et al. 2018, Palmero et al. 2021). (e.g. Zimmermann and Breitenmoser 2007, Molinari-Jobin et al. 2010). Once widespread throughout Europe, the Eurasian lynx disappeared from Central and Southern Europe and many other parts of the continent during the 18th and 19th centuries, as a consequence of direct persecution, habitat loss through forest destruction, expansion of cultivated land, and the excessive reduction of wild ungulates (Breitenmoser 1998, Schadt et al. 2002, Zimmermann 2003, Potočnik et al. 2009). Except for the Carpathian Mountains, it also survived in a small area in the Balkans with a stronghold in North Macedonia, Albania and Kosovo. Since the end of the nineteenth century, forests have regenerated in many mountainous regions of Europe (Breitenmoser 1998, Zimmermann 2004), and the wild ungulate populations have recovered quickly (Apollonio et al. 2010). The improvement of the ecological conditions as well as protective legislation was favourable for the return of large carnivores as lynx populations reintroduced in Central Europe in the 1970s and 1980s, mostly sourcing animals from Slovakian Carpathians (Von Arx et al., 2004, Mueller et al 2020) still persist in the Jura Mountains, Northwestern Alps, Dinarics, Bohemian-Bavarian-Austrian forest and Vosges (Breitenmoser 1998, Chapron et al. 2014). The population sizes have fluctuated over the years, but distribution has not significantly expanded by natural colonisation. Following the first reintroductions, lynx were translocated to Harz (2000), Northeastern Switzerland in 2001 (Ryser et al 2004), to the Kalkalpen (Austria) in 2011-2013 and to Palatinate forest (2016 - 2019) . Lynx’ current distribution in Central and Southeastern Europe seems to be mainly limited to sites that were used for reintroductions and translocations where they were successful. The total has been estimated at only about 3,000 individuals, with little connectivity between subpopulations localised around mountain ranges (Chapron et al., 2014). 4 5 Dinaric lynx population breeding event in the population would be equivalent to a direct brother-sister mating, which could result in Lynx in Dinaric Mountains had become extinct at the considerable inbreeding depression, likely affecting beginning of the 20th century, was reintroduced to the viability and fecundity of the Dinaric lynx. To prevent south-eastern Slovenia in 1973 (known as the Dinaric population extinction, reinforcement and restoration lynx population) with only six founders, some of them of lost population connectivity have been proposed related, from Slovakian Carpathians (Čop & Frković, (Breitenmoser et al., 2007; Zimmerman & Breitenmoser, 1998; Kos et al., 2004). Although little experience and no 2007; Kramer-Schadt et al., 2011; Sindičić et al., 2013, guidelines were available for carnivore recovery programs Lucena-Perez, 2020, Port et al., 2020). In the Dinaric (Breitenmoser et al. 2001), data on signs of presence like lynx population, three reinforcement projects started sightings, reproduction events, scats, prey kills or attacks in 2013 (ULyCA, Urgent Lynx Conservation Action) in on domestic animals were collected opportunistically, 2017 (LIFE Lynx Project, LIFE16 NAT/SI/000634) and but only mortalities were recorded systematically since in 2023 (ULyCA2 Project) to improve the genetic status the reintroduction, both in Slovenia and Croatia (Čop and create a stepping-stone population in the SE Alps, and Frković 1998). The monitoring data made it possible within dispersal distance to the Dinaric population, to to follow the forefront of the expansion of the growing boost connectivity to neighbouring areas. population in subsequent years. Eight years after the reintroduction, young dispersing lynx or adult territorial The population reinforcement, which included the lynx were recorded in all directions (but mainly along the translocation of 12 individuals to the Dinaric part of Dinaric Mountains), at distances from 36 to 100 km from Slovenia and Croatia to enhance the genetic diversity the release site (Čop and Frkovič 1988). The maximum of the population, was completed during 2019-2023 distance of recorded area of presence of lynx in BiH (Fležar et al., 2024). During the reinforcement process from the release site was around 390 km, while from the in the Dinaric Mountains, the mean lynx population northwest (NE Italy) was around 140 km. The proximate density increased for 44.3% according to yearly cause for the faster expansion towards southeast is not monitoring surveys (from 0.88 ± 0.15 to 1.27 ± 0.15 clear; however, it is obvious that fenced highway Ljubljana independent lynx / 100 km2 with the highest increase – Trieste/Koper represents strong barrier for dispersing in the last survey (Fležar et al. 2024, Krofel et al. in lynx (Skrbinšek 2004, Kuralt et al. 2023, 2024). Given preparation). the apparent reduced necessity or ability of subadults, especially females, to cross the highway and reproduce However, without gene flow, natural or assisted, the in almost a half of century after the reintroduction, it is problem of inbreeding cannot be completely solved unlikely that lynx will be able to spontaneously establish in the long term, as the effective population size would new reproductive areas towards the SE Alps. invariably remain low, with resulting high genetic drift causing inbreeding to keep accumulating. Lynx in the While the reintroduction initially appeared to be south-eastern Alps formally/administratively belong to successful, the small founding population led to the Alpine population (Kaczensky et al. 2013), although inbreeding, which resulted in signs of stagnation in they demographically and genetically represent the the 1990s, which turned into drastic decline and local same population since they have been colonised first extinctions after 2000 (Kaczensky et al., 2012; Sindičić et by lynx from the reintroduction in Slovenia in 1973. Thus al., 2013, Fležar et al. 2021). By the 2010s, signs of lynx in order to improve the connection between the Dinaric presence became increasingly scarce, and extinction and Alpine populations, between 2021 and 2023 10 of the population became a tangible possibility. animals were also translocated to the southeast Alps Genetic studies of the Dinaric lynx after 2010 showed in Slovenia and Italy (Fležar et al., 2023a, Hočevar et that the population had the lowest genetic diversity al. 2024, Krofel et al. in preparation). With established and the highest inbreeding of all studied Eurasian lynx reproductions in this area since 2021 (Fležar et al. 2024), populations, although all reintroduced populations are SE Alpine lynx play an important role as the “stepping inbred (Breitenmoser & Obexer-Ruff, 2003; Sindičić et al., stone” that could with further spread enhance potential 2013, Rueness, 2014, Krojerová-Prokešová, 2019, Mueller connections with the reintroduced populations in et al., 2022), with the exception of the reintroduced the northwestern Alps and Austria (Kalkalpen). If this Harz lynx population in Germany (Mueller et al., 2020) will not lead to the movement between the Dinaric where zoo animals were used as founders. In the Dinaric and Alpine population, then further conservation lynx, the average inbreeding coefficient exceeded 0.26 measures (assisted dispersal, translocations, improved (Sindičić et al., 2013) and even F = 0.316 (Pazhenkova permeability of Ljubljana - Trieste highway - green et al. 2023, 2024). This means that an average random bridges) will be necessary. 6 7 Balkan lynx population Albania, lynx occur on Munella Mt. and its surroundings in central-north Albania (Trajçe et al. 2014) and Shebenik-Jablanica NP on the eastern border with North Macedonia and The Balkan lynx population was extirpated from most of the Balkan countries and Polis-Guri I Zi-Valamara in the south-west of the country. It occupies mixed deciduous confined to a small population in the South-west Balkans; Albania, North Macedonia and evergreen forests in the mountainous areas in the south-western Balkans. Deciduous and Kosovo (Melovski, 2022). It was described as a subspecies balcanicus in 1941 (Bureš, forests consist of predominantly European beech and several oak species (Quercus spp.), 1941), later solidified taking three different facets into account: morphology (Mirić, 1978), mixed forests comprise more than 18% (mainly beech–fir mixed forests), nearly 10% are conservation (Melovski et al. 2015) and genetics – unique haplotypes (Gugolz et al. 2008, shrublands and around 1% are coniferous trees (Macedonian fir, Abies borisii-regis and Cómert et al. 2018, Bazzicalupo et al. 2022). The population of the Balkan lynx has been European spruce, Picea abies) (Ivanov et al. 2018). The altitude at which Balkan lynxes occur intrinsically small for at least the past 150 generations (Bazzicalupo et al. 2022). Already ranges from 500 to 1,800 m, with rare exceptions when they venture into high mountain experiencing few bottlenecks in the last 100 years, its genetic resistance is ever so weak pastures above 1,800 m (up to 2,100 m) to cross territories or hunt chamois. in withstanding the rapid environmental change. The next steps of its recovery will most likely also involve a genetic rescue mission in order to strengthen its genetic variability. Reproduction was detected in Munella and Polis-Guri I Zi-Valamara (Melovski et al. 2015). Given that the Balkan lynx is genetically and taxonomically unique question is which Reports of lynx sightings in northern Albania (Albanian Alps) have not been confirmed by subspecies is a better candidate for such a measure. So, we need to know the phylogeny photos taken by locals. In North Macedonia lynx sightings have been reported in western and the phylogeographic and current genetic makeup of the Balkan lynx and its closest part, mainly in the areas in and between Mavrovo, Galičica and Pelister National Parks, neighbours; the Carpathian and the Caucasian lynx. Recent publications have covered but also in Shar Planina National Park, Jablanica Mountains, Stogovo-Karaorman, Ilinska these facets (Gugolz et al. 2008, Cómert et al. 2018, Bazzicalupo et al. 2022), however Plakenska Mountains and Jakupica Massif. In December 2010, lynx were discovered during there is still lack of sufficient knowledge on which subspecies is better ecological fit for the a camera-trapping survey, revealing individuals in the central-northern part of North given environment. However, based on the prey preference (roe deer being the main prey), Macedonia (Jasen PA) (Melovski et al. 2013). The sightings were confirmed by camera local prey availability (lower lagomorph and higher ungulate availability) and habitat use trapping and telemetry studies in 2020 and 2021. In Kosovo, a camera trap photo confirmed (predominant use of the mixed and broadleaved forests) it has been suggested that the L. the presence of two lynx in the Prokletije Mountains in March 2015, which were detected l. carpathicus is ecologically more similar to the L.l. balcanicus and therefore likely better until 2022. In Montenegro, a baseline survey in 2013 found that two individuals had been suited for the environment of south-western Balkans (Melovski et al. 2022a). Potentially, killed in 2002 on the southern border with Albania and Kosovo (Prokletije Mountains). Montenegro and Greece are also sharing this scattered and fragmented population. In Their current presence is, however unlikely. In Greece, isolated, unconfirmed sightings are reported from the border regions of Greece with North Macedonia and Albania. The suspected presence of lynx in the Nestos river delta in eastern Greece, close to the Turkish border (Panayotopoulou and Godes 2004), has never been confirmed by reliable evidence, so their current presence in Greece is unlikely (Melovski et al. 2015). The Balkan lynx population is estimated at 20-39 adult individuals (Melovski et al. 2015), and the density fluctuates between 0.8 to 2 individuals per 100 km² in the core area (Mavrovo NP in North Macedonia) using deterministic camera-trapping surveys conducted from 2008 until 2022 in seven occasions (Melovski pers comm, after CMS proposal, 2023). The population is considered stable, but no systematic abundance estimates have been done outside this core area. I. Figure I. The historic and the current distribution of Balkan lynx population (Melovski 2022) 8 9 General threats and Conservation 3 Red List criteria. Many populations of wide-spread subspecies could be hampered due to unsustainable development and fragmentation without realising it because of their seemingly status of the Eurasian lynx intact distribution range. The Eurasian Lynx is protected by the EU Habitats Directive: Annex II (designation of special areas of conservation for these species, which must be managed according to the ecological needs of the species) and Annex IV (strict protection – protected from killing, disturbance or destruction of their habitats). The general threats to the lynx in Europe are low acceptance due to The Convention on the Conservation of European Wildlife and Natural Habitats (Bern Convention) conflicts with hunters and livestock breeders, illegal killing, habitat loss and lists the Eurasian Lynx under Appendix III (protected fauna species - special protection through fragmentation mainly due to infrastructure development, poor management ‘appropriate and necessary legislative and administrative measures’, of the listed wild fauna structures and incidental mortality (Kaczensky et al. 2012). species). The Balkan lynx, as a subspecies, is listed under Appendix II (Strictly protected fauna species) in 2017 during the 37th meeting of the Standing Committee of the Convention. At the European level, a regional assessment has been made in the IUCN Red List of Threatened Species (von Arx 2018) and a number of European or regional strategies have been The Eurasian Lynx is included on CITES Appendix II and protected under the Bern Convention produced, e.g. the Action Plan for the Conservation of the Eurasian Lynx (Lynx lynx) in Europe (Appendix III). The Balkan Lynx is protected under Appendix II of the Bern Convention. The (Breitenmoser et al. 2000), the Pan-Alpine Conservation Strategy for the Lynx (Molinari-Jobin EU Habitat Directive protects the Eurasian Lynx in each state of the European Union under et al. 2003), the Conservation Strategy and National Action Plans for the conservation of the Annex II, (except the Estonian, Latvian and Finnish populations) and Annex IV (except the Critically Endangered Balkan Lynx (Council of Europe 2011), the Key Actions for Large Carnivore Estonian population). Populations in Europe (Boitani et al. 2015) or the Lynx in the Alps: Recommendations for an internationally coordinated management (Schnidrig et al. 2016). The conservation measures At 14th meeting of the conference of the parties of the Convention on migratory Species for the Balkan lynx have been implemented as part of the Balkan Lynx Recovery Programme, in February 2024 the Eurasian lynx got accepted to be included in the Appendix 2 of the a partnership project between non-governmental organisations from North Macedonia, Bonn Convention and the Balkan lynx as a subspecies, included in the Appendix 1 of the Albania and Kosovo, which was launched in 2006 under the expert guidance (Breitenmoser et convention. The proposal was supported by the EU, among a few other states. The concerted al. 2008). The programme is ongoing and represents an interdisciplinary approach to species actions listed from this proposal are to be implemented in the next two years. They include conservation. However, none of these plans, which were mainly drawn up by experts, led to the conservation strategies and action plans for balcanicus and carpathicus subspecies and desired improvement in formal transboundary cooperation or population-wide conservation knowledge gathering through baseline survey data for isabellinus and dinikii. The lynx listing and management coordination. under the convention is expected to increase the global awareness of its conservation status and support different conservation programmes, strengthen the monitoring activities in IUCN classifies the Eurasian Lynx as Least Concern on the global level given its wide range the range countries, provide possibilities for identifying green infrastructure to ensure the and stable populations in the north of Europe and its wide distribution in southern Siberian invaluable migration of the species, transboundary cooperation between range countries for woodland stretching through Russia from the Ural Mountains to the Pacific, as well as Central implementation of conservation measures and action plans, act in a prompt manner to recover Asia and the Tibetan plateau (Bao 2010, Bersenev et al. 2011, Kaczensky et al. 2012, Moqanaki et native populations that are at threat, motivate research of populations where data is missing, al. 2010, Matyushkin and Vaisfeld 2003). A recent assessment of the status of Eurasian Lynx in as well as strengthening the institutional capacities of all relevant national and international Europe shows that some isolated subpopulations remain Critically Endangered or Endangered stakeholders in regards to the monitoring and conservation activities (CMS 2023). (Kaczensky et al. 2012). Among the subspecies, L. lynx lynx and L. lynx wrangeli are likely to be considered Least Concern, whereas the status of the other subspecies is either unknown or should be considered within the threat categories. Only the Balkan lynx (Lynx lynx balcanicus) has been assessed at the subspecies level, so far and was listed as Critically Endangered in 2015. The population of the latter is estimated to be less than 50 mature individuals distributed mainly in North Macedonia, Albania and few individuals in Kosovo. There has not been recent evidence coming from Greece or Montenegro. However, no systematic monitoring is conducted in these two countries where dispersing individuals could have already appeared. Based on the population size estimates, the IUCN Red List assessment classifies the Balkan Lynx as Critically Endangered (CR: D) as the number of mature/adult individuals is estimated to be less than 50. The population is estimated to be 27-52 independent (adult and sub-adult) animals, corresponding to about 20–39 mature individuals. (Melovski et al. 2015). Currently, its distribution is restricted to three countries: North Macedonia, probably hosting around 70% of the population and Albania and Kosovo, with the rest of the individuals. The range is divided into two nuclei, indicating population fragmentation. The main threats involve poaching, prey depletion, habitat destruction and inbreeding (Bazzicalupo et al. 2022). Other subspecies of the Eurasian lynx are in a need for thorough conservation evaluation according to the IUCN 10 11 Habitat suitability 4 it can also prey on semi-domestic reindeer (Rangifer tarandus) and white-tailed deer (Odocoileus virgianus). In areas with low roe deer density Eurasian lynx diet can seasonally shift to other types of prey like rodents and birds (Krofel et al. 2011). In a research of lynx diet in Dinaric forests with low density of ungulates, rodents represented a considerable part of the diet (7.7%) especially at peaks of their population dynamics. This proves that Eurasian lynx is able to adapt to various food sources. Given present high population densities of ungulate species across Central Europe including Slovenia and Croatia (e.g. Apollonio at al. 2010) it is assumed that prey availability is mostly not a limiting factor for its habitat suitability in the Alps Top predators are generally not very sensitive to a particular habitat structure, and Northern Dinarics but could be an issue for several parts of the Balkans (e.g. Macedonia, vegetation or ecosystem type (Mladenoff et al. 1995). But among the Serbia, Greece, Albania) (Apollonio at al. 2010). European large carnivores, Eurasian lynx is certainly the one with the most specific demands regarding habitat and prey (Breitenmoser 1998). However, Habitat fragmentation and connectivity lynx can adapt to semi-natural landscape and their permanent disturbances Habitat for any species is divided into “habitat patches”, areas with favourable conditions for (Breitenmoser-Würsten et al. 2001). The Eurasian lynx is present in large the species that are separated by “matrix”, areas where individuals can move through but will continuous lowland forest areas with more than 50% of forest cover. It is linked not permanently reside, and “barriers”, through which individuals are more or less difficult or to forest areas with high amounts of forest fringe (Breitenmoser et al. 2000). The even unable to pass (Andrén 1994, Iuell et al. 2003, Bird Jackson and Fahrig 2013, Potočnik et Eurasian lynx can also tolerate interruptions by open land habitat patches and al. 2019, 2023). This fragmentation can be caused by natural features like rivers, high mountain land use types such as pastures or agriculture. Telemetry studies in 1990s in ridges or seas and divides species range into populations and subpopulations. However, the Swiss Alps (Breitenmoser-Würsten et al. 2001) showed that re-introduced human developments are changing the landscape, decreasing habitat, introducing new barriers and pushing fragmentation to the point where it is currently recognized as one of the lynx originating from highly forested Carpathian Mountains, already adapted main threats for many endangered species and a critical obstacle to species recovery (Andrén to open areas, when compared to the first telemetry studies in the early 1970s 1994, Fahrig 2001, 2003). (Haller and Breitenmoser 1986). Intensive land use is tolerated as long as there is enough connected forest area for retreat (Breitenmoser 1998, Schadt et al. In addition, roads, railways and waterways impose movement barriers on many animals, 2002).That was supported in a continental scale study with data of 434 lynx barriers that can isolate populations and lead to long- term population decline. Habitat individuals (Oeser et al. 2023). They confirmed that lynx use refuge habitats fragmentation, the splitting of natural habitats and ecosystems into smaller and more more intensively with increasing landscape modification across spatial scales, isolated patches, is recognised globally as one of the biggest threats to the conservation of biological diversity (Iuell et al 2003, Bird Jackson and Fahrig 2013, Fahrig 2003, 2007). Habitat selecting forests most strongly in otherwise open landscapes and rugged fragmentation is mainly the result of different forms of landuse change. The construction terrain in mountainous regions. Moreover, higher forest availability enabled and use of transport infrastructure is one of the major agents causing this change as well as lynx to place their home ranges in more human-modified landscapes. Human creating barriers between otherwise continuous habitat. On the other hand, barriers causing pressure and refuge habitat availability also shaped temporal patterns of habitat fragmentation have a long-term effect that is not that easy to detect (Iuell et al. 2003, lynx habitat use, with lynx increasing refuge habitat use and reducing their Bird Jackson and Fahrig 2013). use of human-modified areas during periods of high exposure (daytime) or high vulnerability (postnatal period) to human pressure. Urban areas, agricultural landscapes and infrastructure networks divide natural habitats into small, isolated patches and create barriers between the remaining patches. This can affect It is crucial to assess and mitigate the negative effect of habitat fragmentation on lynx species in two ways: firstly, habitat patches can be so reduced in size that they can no longer populations and facilitate genetic exchange among isolated (sub)-populations or demes in support viable populations of important species, and secondly, the remaining patches can Central and Southeastern Europe. Knowledge on the amount and distribution of suitable habitat be so isolated that individuals have little chance of moving between patches. The inability to available to a particular lynx population and to the obstacles it is exposed to is important for move between patches renders species vulnerable to local and regional extinction. Although improvement of our understanding of lynx population connectivity within each population humans began fragmenting nature many centuries ago, the rapid increase in the density of and across habitat patches at the metapopulation level. Apart from habitat suitability and transportation networks in the 1900s and the impact of improved accessibility have greatly connectivity studies, we provide also information on lynx home range size and movement accelerated these effects. activity, including dispersion, as a critical part of its ability to occupy sufficient interconnected areas to compensate for demographic variations and subsequently support genetic exchange The barrier effect, especially of (fenced) roads and railroads, is probably their greatest negative between (sub)-populations, ensuring viability of the metapopulation. ecological impact. The dispersal ability of individual organisms is one of the key factors for the survival of species. The ability to move across a landscape in search of food, shelter or mating Presence and availability of food/prey sources is an important parameter determining habitat is negatively affected by barriers that cause habitat isolation. The impact on individuals affects suitability for animal species. Lynx diet varies greatly depending on prey availability and population dynamics and often threatens the survival of species. accessibility. Although other species within Lynx genus developed specializations for hunting lagomorphs, Eurasian lynx staple prey in Central Europe are roe deer (Capreolus capreolus) Habitat loss and excessive fragmentation is a well-documented threat to wildlife (e.g. and Alpine chamois (Rupicapra rupicapra) as well as other ungulate species like red deer Andrén 1994, Hagan et al. 1996). As habitat is reduced, wildlife populations decrease in size (Cervus elaphus) and European mouflon (Ovis aries musimon). In other parts of its distribution, and become more isolated. The extinction risks may be reduced by rescue effect due to 12 13 dispersal between local populations (Hanski et al. 1996). Connectivity between suitable habitat patches depends on the number of dispersers available in the population, the distance between the source and the target populations, and the dispersal ability of the species under consideration (Wiens 1997). The Eurasian lynx, a charismatic large carnivore, is recovering in most of the European populations as a result of different management strategies applied on, often, well diverse scenarios of different intensities of human-pressure (Chapron et al. 2014). However, the viability of recovering populations and the well-being of the populations that have best withstood human pressure depend very much on appropriate decision-making in conservation strategies. Consequently, it is important to improve the understanding of the requirements of lynx in the current context of population recovery and likely expansion, including the specific spatial needs for the species. The Central and South-Eastern European lynx populations are relatively isolated, and only limited movement occurs between some populations (Zimmermann and Breitenmoser 2007, Potočnik et al. 2009). In the fragmented mountainous regions of the Alps and Dinarics dispersal is constrained by barriers including high mountain peaks, anthropized valleys, canyons and glaciers, fenced highways, large rivers as well as settlements, agricultural, industrial and other urban areas. The ongoing refugee crisis in Europe has seen many countries rush to construct border security fencing to divert or control the flow of people (Linnel et al. 2016). The process of border fencing can represent an important additional threat to wildlife because it can cause additional fragmentation of habitat, reducing its connectivity and lower effective population size. Further colonisation of Central, South-Eastern and Eastern Alps through natural or “human managed” expansion of lynx individuals from the Dinaric population in Slovenia, Italy and Croatia should be one of the priorities of lynx conservation in Central Europe. Connectivity between habitat patches is a critical issue for long-term survival of any wildlife population, as it directly affects not only its dynamics and chances of long-term survival, but also its possibilities for expansion. This makes improving of habitat connectivity between the Dinaric Mountains and the Alps, which will ensure the adequate number of dispersals and maintain gene flow, critical for establishing a viable lynx (meta)population in the Alps, but very challenging considering the needs and desires of humans. The impact of lynx translocations in the Dinaric and SE-Alps has been evaluated on the viability and connectivity of isolated lynx populations within a stepping-stone system (Sánchez Arribas et al. 2023, in preparation). Models have shown lynx translocations positively impacted the demography and connectivity on a local scale, but not at the regional level. Translocations in Dinaric lynx population improved the connectivity of the lynx sub-population in the SE Alps, increasing its viability. Increased urbanisation of lynx inhabited areas and development of large transport infrastructure such as highways has accentuated this challenge in Slovenia and the neighbouring countries over the recent years. The cheapest and most effective way to preserve connectivity is to prevent development in small, critical areas that connect large habitat patches. An effective way to do this is to provide correct information for environmental impact assessment (EIA) that would include habitat connectivity for the Eurasian lynx in spatial planning, and conserve the most critical locations. This is becoming increasingly important as these locations are typically located on cheaper land between already developed areas, and are often the most desirable locations among investors for expansion of industrial and urban areas. While legislation and procedures concerning spatial planning are well developed, there is still a gap in expert knowledge when it comes to ensuring connectivity between habitat patches for large carnivores. 14 15 Isolation and Inbreeding 5 Lynx Movement 6 In conservation biology, inbreeding poses a significant threat to endangered Movement is one of the most studied yet least understood concepts in species, particularly those in small, isolated groups, often resulting from the ecology and evolutionary biology. It has been considered as a glue cementing fragmentation or reduction of populations (Frankham et al., 2002). In large subpopulations and allowing connections between usually isolated populations populations, rare deleterious alleles do not pose a major risk due to their rarity, (Waser et al. 2001, Wiens 2001). Movements have consequences for individuals but inbreeding increases the likelihood that these alleles will be expressed, as well as for populations and communities, and their effects on inclusive fitness reducing individual fitness and reproductive success (Charlesworth & are ultimately the selecting forces for dispersal, migration, exploration, and Charlesworth, 1987). However, in inbred individuals, where both the maternal other types of movement that affect the distribution, abundance, and dispersion and paternal lineages meet in a recent ancestor, the opposite is true. In such of individuals (Clobert and Wolff 2001). individuals, there is a high probability that the phenotypic expression of such alleles will reduce individual fitness and reduce survival and reproductive Understanding animal movement is fundamental to interpret spatial-temporal patterns of habitat success (Allendorf & Luikart 2009). This phenomenon, known as inbreeding selection, foraging behaviour, and the interactions between predator and prey (Bell 1990). Animal movements are influenced by intrinsic physiological factors (e.g., hunger and reproduction) and depression, can contribute to the extinction of small populations. Traditionally, the sensory capabilities of organisms. Spatial structure also influences movement as long as there conservation efforts have focused on demographic factors such as population is a perceived difference in quality of the varying cover types as individuals search for resources size and structure (Jamieson & Lacy, 2012). However, it is now clear that such as food, mates, or den sites or use different cover types to avoid intraspecific and interspecific genetic considerations are critical to the long-term success of conservation. agonistic encounters (Zollner and Lima 1997). Genetic rescue has emerged as a strategy to combat inbreeding depression by introducing Eurasian lynx movements are related to the needs of foraging, mating and rearing of young. genes from closely related populations to improve genetic diversity and population fitness Eurasian Lynx have two main types of movement during their lives: dispersal, which occurs when (Tallmon et al., 2004; Bell et al., 2019). It aims to mitigate the risks of inbreeding depression by they are sub-adult to establish their own territories, and movement within their partly huge home strengthening genetic health through gene flow. The correlation between genetic parameters ranges throughout their lives. The latter may show a seasonal pattern depending on the topography such as heterozygosity and demographic outcomes has been widely documented, highlighting (mountains) and seasonal prey availability. Despite its relatively small size, this species uses large the central role of genetics in the health and persistence of populations (Agudo et al. 2012, home ranges, therefore their moving paths are longer, too (Schmidt et al. 1997). The movement Terrell et al. 2016, Velando et al. 2015). Genetic rescue has been shown to be particularly paths of an individual arise from sequential decisions regarding their needs and perceptions of the effective for small, isolated populations struggling with inbreeding. In summary, inbreeding is surrounding habitat, and it is these decisions that ultimately give rise to the functional connectivity a serious threat to endangered species, exacerbated by factors like population fragmentation. of the landscape (Tracey et al. 2013). Eurasian lynx is a highly territorial species and if individuals are Genetic rescue offers a promising solution by introducing genetic diversity from related to maintain their rights to a territory, they need to move fast and widely enough to advertise their populations, thus improving the fitness and long-term viability of endangered populations. presence over as much area and in as short intervals as possible. Movements within established This integrated approach underscores the importance of genetics in conservation biology territories of resident lynx are often cyclical/seasonal, with core areas of their home range being and highlights the need to consider genetic factors alongside demographic ones for effective used more than the rest. Core areas usually have features and resources that are of high value to conservation strategies. the lynx: abundant prey, preserved forests, potential den sites, low anthropogenic disturbance, etc. Home ranges are traversed throughout the animals’ lives to mark, hunt and raise their young. Mothers with young usually stay near the natal den from the end of May to the first half of July and then roam the surrounding areas in search of prey. GPS tracking in Dinaric mountains revealed movements of the lynx were primarily affected by daytime period, time since the last kill/den translocation, lynx demographic category, and their interactions (Krofel et al. 2013). The lynx tended to stay closer to the prey immediately after the kill, but were found increasingly further away, especially during the day, as the time progressed. This effect was especially pronounced in the females with immobile kittens, but was practically nonexistent in the subadult male. There was a notable difference in movement pattern of female lynx during the consumption process in the period of denning (Krofel et al. 2013). While their kittens were immobile, the females were frequently found further away from the kill compared to when they were alone or had mobile kittens, as they kept regularly returning to the den site. This was particularly the case during daytime, when the females spent a lot of time at the den site. 16 17 A home range is the area in which an animal lives and moves on a periodic basis. It is among Dispersal the most basic of ecological parameters that is regularly described for a given species. An understanding of the requirements for use of space is fundamental for species management Dispersal is any movement of individual organisms in which they leave their home area, and conservation (Schwartz 1999). Furthermore, home range size is one of the most important sometimes establishing a new home area. It is a life-history trait that influences genetic diversity, parameters in producing population estimates. It is important to know how much space demographic viability of metapopulations (e.g. by increasing fitness) and range shifts (Tesson individuals need when estimating potential carrying capacities to plan conservation or & Edelaar 2013) and is a crucial parameter in population dynamics, especially for threatened recovery programs (Schmidt et al. 1997). Home range size is not easy to determine. One big subpopulations within a metapopulation (Levins 1970, Hanski 1999). Dispersal alone can problem is that home ranges vary greatly between interspecific and intraspecific samples. contribute to the recovery of a population if the reasons for decline are demographic or genetic While some interspecific variations in home range can be explained in body mass and feeding in nature. Ultimate mechanisms most likely to affect dispersal are environmental variation styles (Guarino, 2002) in many cases patterns of space-use within species vary by factors and demographic structure. Direct mechanisms include genetics, competition, individual of 10 to 1000 (Gompper and Gittleman 1991). Another problem is the variety of concepts, fitness and (breeding) habitat selection. These act through the fitness traits of survival and methodologies and estimators used to determine home ranges within and between species. reproduction. If dispersal enhances these functions, it will be selected for independently of The simplest estimator of a home range from a set of location data is the minimum convex whatever proximate factors may serve to trigger it (Shields 1982). Another evolutionary issue is polygon (MCP) (Mohr 1947) that has been widely used in Eurasian lynx studies, although it the fitness that often follows successful colonisation of empty habitat or the discovery of new has many drawbacks including often overestimating the size of home ranges (Burgman and habitat beyond the species’ current range. Possibility of inbreeding or outbreeding depression Fox 2003). The other estimators, especially in more recent studies, that have been frequently are also potential concerns (Shields 1982). A final evolutionary issue concerns the maintenance employed for constructing utilisation distribution home ranges in lynx are the so-called (fixed of an appropriate level of genetic variability in a population (e.g. Cooper and Kaplan 1982). or adaptive) kernel density estimators (Worton 1989, Burgman and Fox 2003). This is often seen as a population-level process involving the long-term probability of demic survival and reproduction. A distinction can be made between reproductive dispersal, i.e. the subsequent movement between sites or groups, and natal dispersal, i.e. the movement of an individual from its place of birth or previous breeding site to the site where it potentially reproduces (Zimmermann 2004). Ultimate factors are the selective forces that determine the evolution of behaviour. The most important factors that drive an individual into a dispersal are: genetic predisposition to disperse, local population density, habitat change, age of the individual, reproductive status and disturbance perturbation (Zimmermann 2004). The decision to stop dispersal may involve various elements of habitat or patch selection, such as attraction of conspecifics, habitat quality or physiological factors (Wiens 2001). The chances for successful dispersal depend on the connectivity of the landscape and is consequently decreased in intensively used landscapes i.e. matrix by barriers mostly imposed by humans, such as traffic infrastructure and the loss of suitable habitat (Schadt et.al. 2004). Dispersal allows a species to recolonize former habitats after severe range depression. Natal dispersal rate and dispersal distances are generally male biased in mammals and female biased in birds (Greenwood 1980, Dobson 1982, Clarke et al. 1997). However, the significance of dispersal for the spread of a population is less obvious in felids. Natal dispersal patterns are generally male biased for large solitary felids (Smith 1993, Beier 1995, Maehr et al. 2002), whereas the patterns are less clear among the four species of the Lynx genus, with no clear patterns within species and findings ranging from male biased dispersal (Mowat and Slough 1998, Mowat et al. 2000, Schmidt 1998, Janecˇka et al. 2007) to male and female lynx dispersing equally far and with equal frequency (O’Donoghue et al. 1997, O’Donoghue et al. 1998, Ferreras et al. 2004, Zimmermann et al. 2005, Campbell and Strobeck 2006). The long-range dispersal of Eurasian lynx in their second year of life is sex-dependent. Lynx kittens stay with their mother on average for 10 months, after which they disperse. Dispersal age usually varies from 8 to 24 months (Breitenmoser et al. 1993; Schmidt 1998; Zimmermann et al. 2005; Samelius et al. 2012). While females are phylopatric and only occasionally travel long distances, such long-distance movements (often up to several hundred kilometres) are more common in males (Samelius et al. 2012, Herrero et al. 2020). This dispersal pattern in lynx (and other cat species) prevents inbreeding and is also important for the exchange of genetic information and thus for ensuring the genetic health of populations. Dispersal is also associated with the expansion of a species’ range (Thompson and Jenks 2010), which is particularly important for the recolonization of areas where the species has been eradicated. In this respect, it is necessary to maintain or achieve connected populations and suitable habitats to ensure the prevention of inbreeding and to ensure a high level of genetic diversity and thus long-term survival. Outside the populations of the large and continuous boreal forests in Asia, lynx 18 19 populations exist in the form of meta-populations with partly unknown connectivity of sub-Methodological approaches to study 7 populations and exchange of individuals. A study comparing dispersing lynx from populations in the Nordics, Baltics, and Dinaric habitat suitability, connectivity, and Mountains as well as Central Europe found that the mean dispersal distance was 39 kilometres, and 68% of dispersing lynx settled within 50 kilometres (Molinari-Jobin et al. 2010) while study of viability of lynx populations large dataset of GPS tracked reintroduced and wild dispersing lynx showed median dispersing distances of 84 and 83 kilometres (Meyer et al. In preparation). Lynx tend to establish home ranges adjacent to those of other lynx (Zimmermann et al. 2005), which affects their likelihood of establishing new colonies. Thus, while a lynx population may expand in spatial size, solitary lynx are unlikely to disperse and establish entirely new, separate populations (Zimmermann et al. 2007). Defining important areas for conservation based on recognized species’ habitat preferences is crucial for ensuring populations’ viability and Between 1988 and 2001, a comprehensive study of the spatio-temporal behaviour of subadult lynx in two reintroduced populations was conducted in Switzerland (Zimmermann 2004, persistence in a given geographical area. This is equally true for existing Zimmermann et al. 2005, Zimmermann et al. 2007). The study was based on telemetry and populations and their present ranges as is for their future ranges. To other data from 39 juvenile lynx; 22 in the northwestern Swiss Alps and 17 lynx in the Jura assess given species’ habitat preferences and define areas of importance, Mountains. The lynx became independent at the age of 9.3 - 10.6 months (there was no constructing habitat suitability models provides a crucial first step. Habitat significant difference between males and females). Mothers usually left their kittens at the suitability models (also referred to as habitat distribution models, resource edge of their territory, making excursions to the other side of their territory or even out of their selection functions – Guisan et al. 2017) are a widely used analytical tool home range. In most cases, the mother appeared to have abandoned the young. The reason that quantifies the relationship between the distribution of studied species for the separation could be the female’s feeling of not being able to catch enough prey for her kittens (Molinari and Molinari-Jobin 2001). Various aspects of spatio-temporal behaviour (populations) in a given geographical area and various environmental suggest that disintegration of litters of free-ranging lynx is not caused by aggression of the variables that might contribute to their choice of habitat. Apart from female parent, as claimed by Stroganov (1962) and later by Jonsson (1984). After separation, their role in spatial planning for prioritisation of core habitat patches for the subadult animals usually stayed a few days near the place where the separation took conservation of present or future species’ distribution, they also provide a place and then moved on (Zimmermann 2004). Dispersing lynx were recovered mean=41,2 basis for connectivity analyses and assessment of possible connections to km (n=14) (in Jura Mountains) and mean = 24,3 km (n=13) (in North Western Swiss Alps) away different populations – e.g. defining the most suitable area for establishing from their point of origin. In Central Europe, Eurasian Lynx dispersal distances are substantially a stepping stone population and assessing connectivity between that and shorter than those in Scandinavia, although individual variation is considerable. In Central Europe, males dispersed 4.5–129 km, compared to 32–428 km in Scandinavia (Breitenmoser surrounding core population areas in a metapopulation scheme. et al. 1993; Schmidt 1998; Zimmermann et al. 2005; Samelius et al. 2012). Females in Central Habitat suitability models are especially important in species that appear in low densities Europe dispersed 2–81 km compared to 3–215 km in Scandinavia (Samelius et al. 2012). across large extents and are difficult to spot due to their cryptic nature which makes acquiring their actual distribution in space practically impossible – as holds true also for For some subadults the researchers were able to document a transient home range but most large carnivores (Zimmermann and Breitenmoser 2002). Thanks to the rapid development subadults established a definitive home range directly after their dispersal. Subadults from the of telemetry technology and modelling techniques, it is possible to get good estimates north-west Swiss Alps and the Jura Mountains appeared to have the same dispersal potential of their potential distribution and habitat preferences. Next to generalised linear (logistic as there were no observed differences between the two areas in the total and maximum regression) models (e.g. Zimmermann and Breitenmoser 2002; Zimmermann and distances dispersed. However, a larger proportion of individuals in the north-west Swiss Alps, Breitenmoser 2007; Schadt et al. 2002a; Kramer-Schadt et al. 2004; Signer 2010; Skrbinšek all males, moved through unfavourable habitat but all stopped at fenced highways and turned 2004; Cristescu et al. 2019; Potočnik et al. 2020; Hemmingmoore et al. 2020), ecological back, except one male, which left the area. The apparent reduced ability of subadults to cross niche factor analyses (e.g. Zimmermann 2004; Basille et al. 2009; Huck et al. 2010) and other barriers led to circular dispersal (Zimmermann et al. 2007). Within the study, they did not statistical methods, machine learning algorithms have proved to provide an excellent tool detect any positive density dependent effects in lynx dispersal and hence could not confirm for habitat suitability modelling. In the field of large carnivore spatial research, MaxEnt the hypothesis that high population density encourages the expansion of the population. (Phillips et al. 2006; Phillips and Dudík 2008; Becker 2013 and Oeser et al. 2023 for the Similar study of various aspects of lynx natal dispersal was carried out in Scandinavia by Eurasian lynx), Random Forest (Breiman 2001; Ripari et al. 2022 and Oeser et al. 2023 for comparing dispersal patterns of 120 radio-marked lynx in two study areas in Sweden (Sarek the Eurasian lynx) and Boosted Regression Tree (Friedman 2001) have been increasingly and Bergslagen areas) and two study areas in Norway (Hedmark and Akershus areas, Samelius used and are also deemed as the three most powerful models currently available (Elith et et al. 2012). They found, contrary to the Swiss study, that male lynx dispersed farther than al. 2006; Oeser et al. 2023; Valavi et al. 2021). female lynx with mean dispersal distances of 148 and 47 km for male and female lynx that One particular advantage of machine learning algorithm-based habitat suitability models, were followed to the age of 18 months or older. In fact, female lynx often established home like MaxEnt, is their use of presence-only data and ability to work with small datasets ranges that overlapped or partly overlapped that of their mothers. Similarly, the dispersal rate (Phillips et al. 2006), which is crucial for study of species like large carnivores. MaxEnt, used was greater among male lynx than among female lynx, with 100% of the males dispersing for construction of habitat suitability model for the purposes of these guidelines, works on compared with 65% of the females dispersing. 20 21 estimating the probability Once the habitat patches are defined, it is important to assess the connectivity between of distribution based on them. Connected paths ensure enough gene flow among parts of populations or between the probability distribution populations, avoiding splitting them into separate segments that become more isolated and of maximum entropy prone to (local) extinction due to loss of genetic variability (Frankham et al. 2010), which is constrained by the given data especially important also for the Dinaric population of Eurasian lynx considering it’s high levels (environmental variables on of inbreeding due to small (and related) reintroduced population and prolonged isolation observed occurrence points (Sindičić et al. 2013; Skrbinšek et al. 2019). Connectivity analyses can be done on the level versus generated random of populations in order to evaluate the permeability or fragmentation of the landscape in background points) (Phillips population’s range or on the level of metapopulations to evaluate the possible gene flow et al. 2006). Habitat suitability among the remote populated patches. The latter is important especially in determining models give a basis for suitable patches for reintroductions or relocations for establishing stepping stone populations, estimating possible core and checking whether they are within range (distance and cost-wise) for dispersing individuals suitable habitat patches in and thus ensuring the long term viability of (meta)populations. Crucial corridors or potential existing populations’ ranges bottlenecks can be defined between habitat patches that need to be protected or established (and therefore estimate the in order to ensure connectivity between those patches, which provides important information potential population size, for spatial planning in management and conservation of species in study. environmental capacity) and in potential stepping Apart from least cost path (LCP) analyses, connectivity models based on circuit theory stone populations’ areas (McRae et al. 2008) have increasingly been used in the past years. Incorporating random (and therefore estimating the walk theory (Newman 2005), they can provide a more accurate description of possible possible size of stepping stone successful dispersal movements through a previously unknown landscape, in contrast to populations in the area, which LCP analyses, which assume knowledge and overview of the landscape (McRae et al. 2008). could mean crucial information The algorithms used in circuit theory based connectivity models use resistance surfaces for viability analysis, as well as that define the costs of individual’s movement across different parts of landscape and focal determining suitable sites for nodes (habitat patches or occurrence points) in case of Circuitscape (Shah and McRae reintroductions). A broader 2008; Anantharaman et al. 2020) or a moving window with a defined radius (based on known scale of habitat suitability dispersal distances) to iterate the Circuitscape algorithm across the landscape in the case models is increasingly of Omniscape (McRae et al. 2016; Bezanson et al. 2017). The output is a cumulative current necessary in order to assess flow map which considers all possible paths, where we interpret current density as the potential future distribution probability of the individual moving across a given location through a random walk across the sites that could represent landscape (McRae et al. 2008). Corridors are then defined as high current density at pinch-important connections points where conservation actions are crucial in maintaining or establishing connectivity for (in terms of stepping a viable (meta)population. A good example is using connectivity models for planning animal stone populations’ areas) crossings across barriers such as highways, often impeding connectivity between habitat between existing Eurasian patches, where intersections of corridors and linear barriers present sites where mitigation lynx populations. However, actions are crucial (as discussed in Kuralt et al. 2023). extrapolation of existing habitat suitability models over An important notion to take into account when designing management and conservation large areas can be difficult due actions is the species’ dispersal characteristics. Dispersal, defined as the movement from to environmental differences the site of origin to the site of reproduction or new settlement (Howard 1960), is an important across geographical extent. life-history trait that concerns not only the dispersing individual but also the population and Using largest possible the species as a whole (Tesson and Edelaar, 2013), especially with regard to sufficient gene datasets and testing different flow between (isolated) populations to prevent inbreeding and the resulting risk of local modelling techniques, extinction (Woodroffe 2003). Dispersal allows individuals to colonise new areas and connect preferably including machine populations into a metapopulation to ensure their long-term survival. Understanding dispersal learning algorithms for large-movements and their prerequisites is crucial for effective conservation management, such scale habitat suitability models as protecting and enhancing landscape connectivity for dispersing individuals, especially in (like it was done in Oeser et al. fragmented and human-dominated landscapes (Woodroffe 2003). 2023) might be the solution for a sound ecologically informed While direct data on dispersers through telemetry studies of dispersing individuals is basis for spatial planning immensely important, it is often difficult to catch and tag dispersing animals within a population. and management at the Thanks to large databases of telemetry data, it is possible to identify dispersal movement metapopulation level. through various methods, one of the widely used being the net square displacement (NSD) method,which uses the straight-line distance between the starting and each subsequent 22 23 location for the movement of each individual – the shape and slope of the curve can explain Modelling habitat suitability and 8 the movement type of the observed individual, with the dispersal fitting a logistic regression model (Bunnefeld et al. 2011) and thus showing a positive NSD slope over time (Meyer et al. unpublished). Additionally, data from translocated animals (which are often equipped with connectivity for lynx in the Alps, a GPS collar) that showed exploratory behaviour upon release (also named post-release dispersal) (Topličanec et al. 2022) or long-distance exploratory movements of remnant or Dinarics and Balkan region translocated animals can be used as a proxy for dispersal movement and consequently immensely useful in creating a large enough dataset of extra-territorial movements for further analyses (as was done in the study made by Meyer et al., unpublished). As space-use can often differ between resident and dispersing individuals (dispersers being known as using also sub-optimal habitat when traversing the landscape, e.g. Hemmingmoore Habitat suitability modelling and distribution of core and other et al. 2020), knowledge of dispersal characteristics is also important in constructing habitat suitable habitat patches suitability models or designing resistance surfaces for future connectivity analyses. Connectivity analyses that consider dispersal abilities of the studied species can provide For the purposes of these analyses, a habitat suitability model was created using MaxEnt an even better ecologically informed basis for spatial planning and translocation actions to machine learning algorithm for a large study area ranging from Jura and the NW Alpine establish a well connected network of stepping stone and core populations with sufficient lynx populations on one side and Balkan and Southern Carpathian lynx populations on natural gene flow provided through dispersal for a viable metapopulation. the other. Telemetry data from 42 individual lynx (from Dinaric, SE Alpine and Kalkalpen populations) were used as occurrence points on the background of five environmental Another important tool informing management and conservation strategies is population variables – altitude, human footprint index, tree cover density, surface roughness and viability analysis that can simulate and predict the viability of the populations through time aspect. The model was later used to define highly suitable habitats, termed ‘core’ and other which provides crucial information on long term viability of established or potential (meta) ‘suitable habitat” patches, based on the suitability values at occurrence points (0.68) and populations. This involves determining the populations’ demographic changes in the future, the arbitrarily set threshold (0.5), respectively, including only patches larger than 10 km2 their survival rate under different scenarios, or identifying variables that are important to avoid using small fragments of core or other suitable habitat. The constructed habitat for their population growth, which may prove crucial in determining future management suitability model (Figure 1) shows that less populated, forested areas at medium altitudes steps or guidelines (genetic or spatial) for vulnerable populations, including reintroduced are preferred, as high altitudes pose a natural barrier limiting their movement while low or reinforced ones (e.g. Pazhenkova and Skrbinšek 2021; Sanchez et al. unpublished; altitudes are usually densely populated. It shows large patches of suitable habitat in the Pazhenkova and Skrbinšek 2024. unpublished; Potočnik et al. 2009; Heurich et al. 2018; massifs of Balkan peninsula, stretching from Dinaric mountains in the northwest to Pindus Kramer-Schadt et al. 2005). Individual-based genetic-demographic models (as recently mountains in the south, to the edge of Carpathians and Balkan mountains in the east done for the Dinaric and SE Alpine population by Pazhenkova et al. unpublished and Pazhenkova and Skrbinšek, 2024) can be used to evaluate the success of past translocations (reintroductions and reinforcements) and inform future genetic management strategies (e.g. additional reinforcements) in terms of predicted long-term viability of populations based on reduction of inbreeding and enhanced genetic variability (Pazhenkova and Skrbinšek 2021, 2024; Pazhenkova et al. unpublished). In the case of the Dinaric population of Eurasian lynx, Pazhenkova and Skrbinšek (2021, 2024) have shown that even though recent reinforcements have increased the probability of population survival, the success is short lived and would need additional reinforcements of 5-10 animals every 10-20 years to maintain a viable population if it continues to live in isolation from surrounding populations. A spatially-explicit individual-based model of population viability recently done on the Alpine populations (Sanchez Arribas et al. unpublished), for example, showed an improvement in the viability and connectivity in the established SE Alpine population and predicted the most reliable patches for future establishment of stepping stone populations together with the minimum number of released individuals in order to connect Alpine populations into a viable metapopulation. Other important variables that can be simulated and predicted through population viability analyses are factors leading to lynx mortality (such as poaching or traffic mortality) which should also be addressed in future management plans (Sanchez Arribas et al. unpublished, Pazhenkova et al. unpublished). Figure 1. Habitat suitability model for Eurasian lynx in Alpine (purple outline) and Balkan (red outline) region. 24 25 and Rhodope mountains in the southeast. Dinaric and other Balkan regions showed more suitable habitat for lynx compared to Oeser et al. (2023) model, however since our model was created using local data it might indicate better fit for that region. In the Alpine region, it shows smaller patches of suitable habitat on the northern and southern edges of the Alps, showing belts of suitable habitat below high altitude mountain ridges and above the valleys. The difference possibly resides in lynx habitat in the Alpine region being restricted by heavily populated valleys on one side and high altitude mountain ridges on the other side, whereas Dinaric mountains are less densely populated even at lower altitudes. Populated Core Other No. Range of No. No. No. males No.females No. males and potential habitat suitable patches patch sizes patches females (core / (core / (core / stepping (km2) habitat (core / [km2] (core >= 200 (core / suitable) suitable) suitable) stone areas (km2) suitable / suitable) km2 suitable) by area by patches by patches (core / by area suitable) 10.25 – 25 0 5 3 0 0 Jura and NW 131.4999 746.5 9348.25 Alps 103 10 – 7 68 44 43 24 3250.25 8 10.25 – 323 1 3 2 1 1 Kalkalpen 520.25 2135.5 1* 2135.5 1* 13 9 13 9 11 – Figure 2. Suitable and core habitat patches for Eurasian lynx in the Alpine region. Populated 16 3 12 7 9 5 SE Alps 2490.75 6932.5 865.9996 patches are shown in green, potential patches in blue, both with darker tones for core and 1* 1426 1* 35 21 35 21 brighter for suitable patches. 10.25 – 51 10 45 36 31 24 Dinaric 8142.5 22056 1327.75 20 10.5 - 21265 1 123 99 119 95 10.5 – 26 2 12 3 6 1 452.75 Balkan 1491.93 8657.75 7 11.25 – 2 72 23 71 22 7037.499 10.25 – 121 8 40 24 22 11 Western – 12691.5 7292.75 35493.99 Eastern Alps 78 10.5 – 12 198 128 176 110 5313.499 10.25 – 17 2 6 4 9 4 506.75 Northern Alps 1070.75 8106.5 34 10.5 – 3 48 32 36 23 3312.57 10.5 – 63 4 19 11 7 3 415.75 Southern Alps 3479.25 14263.99 33 10.75 – 8 78 49 68 40 5313.499 10.25 – 41 2 15 9 6 4 North-eastern 1078.75 2742.75 13123.5 Alps 11 12.5 – 1 72 47 72 47 12691.5 Dinaric - 10.25 – 157 17 100 53 51 19 Balkan – S 1245.75 14882.25 63528.02 Carpathian 10 – (part) 56 12 429 206 333 157 43269.5 Table 1. Core and other suitable habitat areas, number of patches and their sizes and potential Figure 3. Suitable and core habitat patches for Eurasian lynx in Balkan region. Populated number of resident female and male lynx in populated and potential stepping stone areas. patches are shown in green, potential patches in blue, both with darker tones for core and Data for the South Carpathian area is not shown due to using only a part of the whole area brighter for suitable patches. based on study extent. 26 27 Based on the habitat suitability model, we extracted core and other suitable habitat patches and categorised them into populated (based on lynx populations’ distribution data of Kaczensky et al. 2021) and potential (possible stepping stone populations) habitat patches (Figure 2 and 3) that could present bridges between existing lynx populations. Similar to Schadt et al. (2002b), we also dissolved adjacent patches within 1 km distance and considered them connected as one. As already seen in the habitat suitability model (Figure 1), the patches in the Alpine region (Figure 2) are smaller and more fragmented than patches in the Balkan region (Figure 3). We were able to assess the surface of core and suitable habitat in the populated and potentially populated areas and estimate the number of territorial individuals (male and female separately) that could reside in these patches or areas (similar to Kuralt et al. 2023), using data of home range sizes of Eurasian lynx from literature or accessible telemetry data for the population in question (Breitenmoser et al. 1993; Breitenmoser-Würsten et al. 2001; Potočnik et al. 2020; Kuralt et al. 2023; Melovski et al. 2020) and using the averages of home range sizes of neighbouring populations for potentially populated areas. We clustered patches into 5 populated areas (Jura and NW Alps, Kalkalpen, SE Alps, Dinaric, Balkan) and 4 potential areas (N Alps, S Alps, NE Alps, Dinaric-Balkan-S Carpathian), as shown in Table 1 and in Figure 4 and 5. Apart from determining core and suitable habitat areas, number of patches, theirs size ranges and number of possible individuals these patches or areas can hold (shown in Table 1), we also show core and suitable habitat areas in each country for respective population areas (Figure 4 for Alpine and Figure 5 for Balkan region). Figure 5. Core and other suitable habitat areas in countries for respective population areas in the Balkan region. The areas represented are only part of the areas in question for this study, other populated or potential population areas (such as suitable patches in Slavonia in Croatia or in eastern Slovenia) are excluded. Suitable and core areas in Bulgaria and Greece were only partially included in this study so the resulting areas should be considered with caution. Extra-territorial and non-territorial movements To analyse the extraterritorial movements of individuals from telemetry studies within the LIFE Lynx project, we used the lsmnsd package in R (Bastille-Rousseau et al. 2016) to cluster the movement data according to net square displacement (NSD) values. This approach was later combined with visual inspection of tracks and all extraterritorial/non-territorial movement paths were extracted and further analysed (as explained in Mlinarič et al. in preparation). Extraterritorial or non-territorial movements were categorised as follows: natal dispersal (when young lynx leave their mother’s home range on their way to independence), post-release dispersal (exploratory movements of translocated lynx from release until home range/territory establishment, in some cases also between temporary home ranges) and excursions ( a round-trip exploratory movement of territorial lynx outside of its territory), the latter being of particular interest during the mating season (February to early April) and thus called mating excursions. Out of 30 individuals, 4 were considered territorial and showed no extraterritorial or non-territorial movements. Eight were dispersing from their mothers’ home ranges, 4 in the Alps (see also Figure 6) and 4 in the Dinaric region (see also Figure 7) – the total length Figure 4. Core and other suitable habitat areas in countries for respective population areas in of their dispersal paths (sometimes divided into several sections by temporary home the Alpine region. The areas represented are only part of the areas in question for this study, ranges) ranged from 96 km (Neža, Dinaric) to 860 km (Flori, Alpine); more information on other populated or potential population areas (such as Vosges mountains, BBA population or dispersal paths can be found in Table 4 below. In the case of translocated lynx, 10 out of potential suitable areas in the Apennines) are excluded. 15 individuals showed post-release dispersal movements (in some cases also divided into 28 29 several segments through temporary home ranges), ranging from 74 km (Blisk, Dinaric The total length of excursion paths ranged from 20 km (Maks, Dinaric – Slovenia, excursion - Slovenia) to 481 km (Kras, Dinaric – Croatia), shown in Figure 8 for individuals released from a temporary home range) to 304 km (also Maks, excursion from a temporary home or resettled in Slovenia and in Figure 9 for individuals released and resettled in Croatia. range). We also measured the total distance of extra-territorial movement paths as the The results also roughly correspond to post-release exploratory movement analysis distance between the centroid of home range and the furthest point of the path. Paths done on the same (but not all) individuals by Topličanec et al. (2022) and Hočevar et al. (post-release dispersal and exploratory ) that had a total distance greater than 20 km were (2024). Excursions were detected in 12 individuals, both remnant and translocated alike, categorised as long-distance and are also listed in Tables 2 and 3 below. Long-distance with many individuals making multiple excursions throughout the tracking period, – with excursions were recorded for individuals residing in Slovenia (or had a transboundary translocated lynx Katalin (Dinaric – Slovenia) leading the race with 13 recorded excursions. territory, as was the case for lynx Bojan) and are shown in Figure 10. Name of Extra-territorial Total length [km] Total distance Time frame Name of Extra- Total Total Distance Time frame Comments individual type [km] individual territorial length distance start – type [km] [km] centroid Mihec (R - M) [km] e Excursion 126.02 28.07 8.3.-20.3.2021 Doru (T - M) Klif (R - M) Post-release pd 146.85 44.39 41.5 5.5.-17.6.2019 dispersal e1 Excursion 134.06 35.26 17.3.-7.4.2022 Emil (T - M) Bojan (R - M) *not longitudinal Post-release pd 352.8 54.17 25.93 15.5.-1.10.2021 but more e1 Excursion 33.72 23.98 17.1.-20.1.2020 dispersal polygonal movement Maks (T - M) Kras (T - M) Excursion from *not longitudinal Post-release e1 temporary home 303.67 76.28 26.11.2020- pd 481.14 89.44 61.77 24.3.-17.7.2023 but more 15.2.2021 dispersal polygonal range movement Goru (T - M) Lubomir (T - M) *first polygonal, e1 Excursion 223.68 57.46 1.3.-6.4.2020 Post-release pd 221.36 32 6.44 16.6.-29.8.2022 then trip around dispersal until HR e2 Excursion 219.07 43.68 19.2.-18.3.2021 Sneška (T - F) e3 Excursion 126.61 25.49 8.3.-18.3.2022 Post-release pd1 26.69 19.99 19.99 26.4.-6.5.2023 Blisk (T - M) dispersal Alojzije (T - M) e1 Excursion 100.13 26.57 21.3.-6.4.2023 Post-release pd 125.18 41.46 10.79 14.3.-24.4.2020 Katalin (T - M) dispersal Goru (T - M) e1 Excursion 71.85 36.35 11.3.-11.3.2021 Post-release pd 113.25 40.6 15.97 14.5.-1.6.2019 e2 Excursion 79.27 32.82 15.3.-22.3.2022 dispersion e5 Excursion 59.12 36.18 17.3.-19.3.2023 Katalin (T - M) Post-release e6 Excursion 47.75 26.14 4.4.-6.4.2023 pd 207.41 58.88 25.43 31.3.-20.4.2020 dispersal Table 2. Long-distance exploratory movements – excursions of lynx, collared in the course of Table 3. Long-distance exploratory movements – post-release dispersal of lynx, collared LIFE Lynx project. Total length represents the length of the extra-territorial movement path, in the course of LIFE Lynx project. Total length represents the length of the extra-territorial total distance the distance between home range centroid and the farthest point of the extra-movement path, total distance the distance between established home range centroid and territorial movement path. Exploratory movement is shown in remnant (R) and translocated (T) the farthest point of the extra-territorial movement path, fifth column of the table also shows individuals. In this case, only males (M) showed long-distance exploratory movements and all distance from start to the centroid of established home range. Post-release dispersal was except Maks could be said to go on mating excursions. present in translocated (T) individuals, both male (M) and female (F). 30 31 Name of Extra- Total Total Distance from Time frame Comments 11.948 (mother 16.241 individual territorial length distance center of natal HR Natal HR centroid to HR 25.10.- d2 255.55 (mother type [km] [km] (mother / release site to dispersal centroid) / 9.992 26.12.2022 HR) HR / THR) centroid of HR / (start to HR centroid) [km] THR / end point [km] Niko (D - M) Andrej (A - M) 23.043 (mother HR 21.925 *Not finished, Natal centroid to THR 10.12.- d1 42.67 (mother 31.911 19.648 (mother HR 17.4.2023- but has dispersal centroid) / 18.895 24.12.2020 Natal HR) d 857.9 (mother centroid to latest 24.3.2024 some (start to HR centroid) dispersal HR) point) (ongoing) polygonal 32.412 (centroids) Dispersal movements d2 47.15 / 18.876 (start to 3.5.-19.5.2021 from THR Flori (A - M) centroid) *Not finished, 25.33 (centroids) Dispersal 52.011 33.881 (mother HR 23.4.2023- but has d3 26.08 / 24.104 (start to 10.6.-18.6.2021 Natal from THR d 860.26 (mother centroid to latest 22.3.2024 some centroid) dispersal HR) point) (ongoing) polygonal 54.503 (centroids) Dispersal movements d4 103 / 49.177 (start to 11.8.-3.9.2021 from THR Meri (A - M) centroid) 12.816 (mother HR 27.026 (centroids) Natal Dispersal centroid to THR d5 55.28 / 18.447 (start to 8.10.-19.10.2021 d1 dispersal to 12.7 22.2.-24.2.2023 from THR centroid) / 13.373 centroid) THR (start to centroid) Table 4. Dispersal movements of eight remnant individuals – from SE Alpine (A) and Dinaric Excursion e1 28.23 13.64 18.2.-21.2.2024 (D) population, either female (F) or male (M). Total length represents the length of the extra-from THR territorial movement path, total distance the distance between (natal or temporary) home Dispersal 16.875 12.112 (THR centroid 10.3.-21.3.2024 *Polygonal d2 106.87 range centroid and the farthest point of the extra-territorial movement path, where it is from THR (THR) to end point) (ongoing) movement possible, and the distances between centroid of consecutive home ranges or between start/ Rozi (A - F) end point and the home range in question. The dispersal paths are segmented with several 8.596 (mother HR individuals as they showed home range movements (temporary home range – THR) between Natal 11.78 centroid to THR segments of dispersal movements and before establishing their home range (HR). In segments d1 dispersal to 11.11 (mother 27.1.-1.2.2023 centroid) / 6.53 that do not present a “straight” line until the next (temporary) home range, total distances from THR HR) (start to centroid) centroids either of mother home range or temporary home range are calculated. Dispersal 4.672 (centroids) / d2 23.89 8.3.-14.3.2023 from THR 6.78 (start to centroid) 13.564 (centroids) Dispersal d3 28.83 / 13.52 (start to 24.4.-15.5.2023 from THR centroid) 15.10.2023- Dispersal 16.179 7.413 (THR centroid *Polygonal d4 296.08 22.3.2024 from THR (THR) to end point) movement Figure 6. Dispersal (ongoing) paths of 4 individual lynx Mala (D - F) (Andrej, Flori, Meri and 9.228 0.849 (mother HR *Polygonal Natal Rozi) in the Alpine region. d 134.14 (mother centroid to end 28.4.-1.7.2020 but not dispersal In the case of Meri HR) point) distinctive HR and Rozi, the dispersal Neža (D - F) path is segmented by Natal temporary home ranges 11.929 dispersal (shown as polygons). d1 40.64 (mother 5.3.-16.3.2022 (attempt) / Mother home ranges HR) exploratory (Julija for Andrej, Flori Natal 11.794 5.458 (mother HR and Meri, and Aida for d2 dispersal 95.97 (mother centroid to end 6.4.-11.5.2022 *Killed Rozi) are also shown as (attempt) HR) point) polygons on the map, Valentina (D- F) together with paths of Natal dispersing individuals 11.971 dispersal still in natal home ranges d1 37.81 (mother 4.3.-15.3.2022 (attempt) / (dashed lines). HR) exploratory continue on the next page 32 33 Figure 7. Dispersal paths of Figure 8. Post-release 4 individual lynx (Mala, Niko, dispersal paths of 6 Valentina and Neža) in the individuals released or later Dinaric region. In the case residing in Slovenia. Their of Niko, the dispersal path established home ranges is segmented by temporary or temporary home ranges home ranges (shown as are shown as polygons. polygons). Last detected Sneška’s and Boris’ post- home range (THR5) is split release dispersal paths are in two parts (Slovenian and segmented by temporary Croatian), with several paths home ranges, both also (dark orange) crossing made a short excursion from between them. Valentina the temporary home range. and Neža attempted dispersal twice, the second resulting in (temporary) home range in Valentina’s case and in mortality by poaching in Neža’s case. Mother home range (Teja) is also shown as a polygon Figure 9. Post-release (dark brown) on the map.. dispersal paths of 4 individuals released and later residing in Croatia. Their established home ranges are shown as polygons. Emil, Kras and Lubomir showed some polygonal movement that was not last long or was not concentrated enough to be classified as home range movement. Figure 10. Long-distance excursions of 7 individuals. Their established (or temporary, in the case of lynx Maks) home ranges are shown as polygons. 11 paths are shown in part A, lynx Maks’ excursion is shown separately in part B. 34 35 Connectivity For grainscape analysis, we used long-distance exploratory (excursions and post-We conducted a connectivity analysis using two approaches – Omniscape algorithm release dispersal) tracks and entire dispersal tracks from lynx tracked during LIFE Lynx (McRae et al. 2016; Bezanson et al. 2017) with a moving window radius set to 75 km (based project, described above, together with data from 2 individuals from UlyCA2 project and on Kuralt et al. 2023 and Potočnik et al. 2020 information on longest dispersal distance 3 individuals from Kalkalpen population, in order to calculate cumulative costs of extra-of Eurasian lynx) and Grainscape package in R (Galpern et al. 2023). Resistance surface territorial paths. Grainscape output provided least cost paths connecting suitable habitat was created based on the habitat preferences from habitat suitability model and using patches with various cumulative costs, we extracted the paths with cumulative costs below the approach used in Kuralt et al. (2023). Motorway network was also included as a linear 908 threshold – the median of cumulative costs extra-territorial paths – and the paths with barrier, using the highest possible resistance value (100), with passages (bridges, tunnels, cumulative costs below 5147 threshold – the value at 95th percentile of cumulative costs over- and under-passes and wildlife crossings) on the barrier reducing its resistance value. of extra-territorial paths. The results show 596 (below median threshold) and 796 (below Resulting connectivity maps are shown in Figure 11 (Alpine region) and Figure 12 (Balkan 95th percentile threshold) least cost paths in the Alpine region (Figure 13) and 392 (below region) for the output of Omniscape analysis, and in Figure 13 (Alpine region) and Figure 14 median threshold) and 520 (below 95th percentile threshold) in the Balkan region (Figure (Balkan region) combining both results from Omniscape and Grainscape analysis. 14). The thresholded least cost paths show whether habitat patches could be connected through dispersal (either short-scaled dispersal movement, taken into account through the Omniscape (Figure 11, Figure 12) shows moderately high current density (landscape median threshold, or by exceptional long-distance dispersal, taken into account through permeability) across larger suitable patches, meaning good connectivity with several 95th percentile threshold), and if so, where the most important corridors for connections possible pathways for individuals to choose when travelling across them, and low current are, providing crucial information on areas in need of protection. While the balkan region density on unsuitable areas, meaning low connectivity or lower chances an individual would seems fairly well connected, especially due to large patches of suitable habitat, the alpine pass through them. However, the important information Omniscape current density maps region shows a lack of connection between the western and eastern Alpine populations. provide, is the location of pinch-points or corridors where current density is high – these The reason could lie in the barrier-like areas of low suitability in populated valleys or are seen in the case of smaller or narrower suitable patches, surrounded by otherwise across high-altitude mountain ridges, as was already explained above. unsuitable landscape, or between adjacent suitable patches, thus showing a possible corridor or path connecting neighbouring patches. If areas with moderately high current density call for large-scale protection of suitable habitat, corridors with high current density mean a need for management measures ensuring connectivity among and inside those suitable areas. The cumulative current value of 1.5 was taken as a threshold showing corridors or high current density areas (shown in Figure 13 and Figure 14). Figure 11. Omniscape connectivity results for the Alpine region. Cumulative current density, Figure 12. Omniscape connectivity results for the Balkan region. Cumulative current density, shown on the map, can be interpreted as the probability that a random walking lynx individual shown on the map, can be interpreted as the probability that a random walking lynx individual would pass through a specific cell on the map. High density thus means high chances of would pass through a specific cell on the map. High density thus means high chances of passing through, identifying corridors and narrow paths in need of protection. passing through, identifying corridors and narrow paths in need of protection. 36 37 Figure 13. Connectivity of Alpine region. Results from Omniscape above the 1.5 threshold (black) and from Grainscape LCPs with thresholds at median (red lines) and 95th percentile (purple lines) of extra-territorial paths’ cumulative costs are presented, together with populated and potential suitable and core habitat patches. The map shows less connections with shorter (median threshold) paths between West and East Alpine region, but the connection is established with longer (95th percentile) paths. Connectivity of potential stepping stone patches The identification of potential areas for the establishment of stepping stone populations is an important step in the management strategy aimed at a metapopulation scheme. We chose the size of 10 home ranges of male lynx as a threshold for potential stepping stone population patch and obtained 5 patches in the Alpine region and 2 patches in the Balkan region with a sufficiently large and sufficiently connected area of suitable habitat. Their size and the area of core habitat within these areas are shown in Table 5. Habitat patches [population region] Suitable area [km2] Core area [km2] Alpine region Patch 1 [N Alpine] 3146.7 759.7 Patch 2 [N Alpine] 3312.7 277.7 Patch 3 [NE Alpine] 12678.5 2742.7 Patch 4 [S Alpine] 5313.5 1395.5 Patch 5 [S Alpine] 3742 1250.7 Balkan region [Dinaric – Balkan – S Carpathian] Patch 1 43269.5 9947.7 Figure 14. Connectivity of Balkan region. Results from Omniscape above the 1.5 threshold Patch 2 11482.8 3413.5 (black) and from Grainscape LCPs with thresholds at median (red lines) and 95th percentile (purple lines) of extra-territorial paths’ cumulative costs are presented, together with Table 5. Potential stepping stone population patches with size larger than 10 lynx male home populated and potential suitable and core habitat patches. Population patches are already range sizes. Their suitable areas and core areas are provided, as well as in which population seemingly well connected through large and abundant suitable patches. region they classify. 38 39 Figure 15. Schematic representation of connectivity between potential stepping stone Figure 16. Schematic representation of connectivity between potential stepping stone population patches and populated patches in the Alpine region. Populated patches (dark red) population patches and populated patches in the Balkan region. Populated patches (dark are connected with potential patches (other colours) via LCP links below median threshold red) are connected with potential patches (blue-purple) via LCP links below median threshold (A) or 95th percentile threshold (B). The green nodes represent centroids of smaller parts (A) or 95th percentile threshold (B). The green nodes represent centroids of smaller parts included in the populated or potential patches, their sizes represent the area of the parts. included in the populated or potential patches, their sizes represent the area of the parts. Green lines represent schematic LCPs between these nodes that are below the respective Green lines represent schematic LCPs between these nodes that are below the respective threshold, white lines represent those above. Patches in question are labelled with numbers threshold, white lines represent those above. Patches in question are labelled with numbers that correspond to those in Table 5 above. that correspond to those in in Table 5 above. 40 41 We ran Grainscape with the above-mentioned thresholds for these patches (which Grainscape treats as several smaller patches) and plotted the resulting connectivity in Figure 15 for the Alpine region and Figure 16 for the Balkan region. As seen previously (in Figures 13 and 14), West and East Alps are poorly connected through short dispersal paths, especially in the more fragmented northern side, while longer paths provide enough connections between potential and populated areas. In the Balkan region, on the other side, the areas already seem to be well connected due to large and abundant suitable patches, as mentioned above, even though the connections between Dinaric and Carpathian population is narrowed down in the eastern part to the potential suitable areas in south-eastern Serbia and western Bulgaria, at the western edges of Balkan Mountains. An important point to consider when assessing connectivity are possible linear barriers that could impede landscape permeability and stop dispersers from reaching the adjacent populations or habitat patches. Motorways present such linear barriers that combine the effect of roads with increased mortality with perceived risk while also often being fenced and thus prevent individuals from crossing. Examination of the paths constructed with Grainscape analysis between all suitable patches and their intersections with highways provided an interesting perspective on the importance of available and suitable highway crossings, as shown in Table 6. There are a large number of crossings in the form of overpasses or underpasses that represent a narrow crossing (usually around 2 m) that already has its own traffic volume (primary, secondary or even tertiary roads) that could prove useful for lynx individuals (as in the case of lynx Maks and its regular crossings of the Ljubljana-Trieste highway (Seidl 2023)), however, such narrow crossings are usually not used and movement across highways is still limited. Potential corridors from connectivity analyses also provide important information on critical points where the construction of Figure 17. Movement of dispersing lynx Niko at the last recorded (temporary) home range. wildlife crossings is needed to improve landscape permeability. Polygonal home range movement is obstructed by the border fence on the western part of the Croatian part of the home range where the border fence is present, leading to splitting the home range into Croatian and Slovenian parts. As this effect could be also due to areas Region and paths Number of paths Number of Number of Number of with steeper slopes (cliffs) in the valley of Kolpa river, slope is also added to the map. Unlike intersections with crossings crossings > 10 the home range movements, more directional dispersal paths can cross the fenced border, as motorway within those m / wildlife intersections crossings seen with the middle (yellow) dispersal path – showing also the different perception of barriers between resident and dispersing behaviour. Alpine Median threshold 596 94 106 74 / 1 95th percentile threshold 796 144 178 125 / 1 Balkan Median threshold 392 34 34 28 / 1 95th percentile threshold 520 72 66 50 / 1 Table 6. Number of LCP paths from Grainscape analysis per region and per threshold setting with intersections with motorways and motorway crossings. As LCP paths have a resolution of 500x500 m cells, multiple smaller crossings can fall within one intersection between LCP path and motorway. Another type of linear barriers that sadly do not provide a logical solution such as wildlife crossings are border fences, in this case especially important in the Balkan region due to externalisation of borders of European Union states and the moving of Schengen border more to the south – thus also urging states to implement measures like border fence construction. Like fenced highways, these too can mean a barrier that importantly impedes landscape permeability for large mammals, including Eurasian lynx (as seen in an example of lynx Niko in Figure 17) and measures limiting their construction, especially in suitable and core areas, are crucial. 42 43 Conclusions 9 The situation in the south-western Balkans, within the current native range of the Balkan lynx, is still relatively well connected in terms of fragmentation. However, the non-EU countries are in a phase of rapid development to meet their increasing economic and energy needs. This potentially means fragmentation due to transportation (e.g. highways) and the construction of infrastructure for hydropower (artificial lakes on rivers). Due to the very mountainous terrain in the western parts of North Macedonia and eastern Albania, the main distribution patches of the Balkan lynx are bypassed by large infrastructure projects, but the future dispersal potential could be affected if such projects are implemented The central and south-eastern European lynx populations are relatively isolated, and only without crossing opportunities. limited movement occurs between some populations (Zimmermann and Breitenmoser 2007, Oeser et al. 2023). In the fragmented mountainous regions of the Alps, Dinarics and Inbreeding depression poses a serious threat to small populations as it leads to the fixation the rest of Balkan peninsula dispersal is constrained by barriers including high mountain of deleterious mutations and a decrease in the survival probability. While the creation of peaks, deep valleys, canyons and glaciers, fenced highways, large rivers as well as connectivity and subsequent natural gene flow between populations is an ideal long-settlements, agricultural, industrial and other urban areas. The ongoing refugee crisis in term solution, its practical implementation under real-life conditions is often challenging. Europe has seen many countries rush to construct border security fencing to divert or The significant reduction in the inbreeding coefficient and increase in genetic diversity control the flow of people (Linnel et al. 2016). The process of border fencing can represent following translocations suggest that population reinforcement, as observed in the Dinaric an important additional threat to wildlife because it can cause additional fragmentation lynx (Pazhenkova et al. 2024, Pazhenkova in prep.), can effectively mitigate the negative of habitat, thus reducing its connectivity and lower effective population size. All small consequences of inbreeding. Stochastic modelling underlines the importance of genetic and isolated populations of lynx are already suffering, or may suffer in the future, from management, as simulations without translocations predicted a decline in population the loss of genetic variation. Most reintroduced populations show low genetic diversity size and an increased risk of extinction within the next three decades. The population (Breitenmoser-Würsten and Obexer-Ruff 2003, Kaczensky et al. 2012, Schmidt et al. 2011, reinforcement efforts implemented as part of the LIFE Lynx project have significantly Sindicic et al. 2013, Mueller et al. 2022, Pazhenko and Skrbinšek 2024, Pazhenko et al. in delayed the detrimental effects of inbreeding and genetic erosion, making a crucial preparation), which is due to inbreeding and genetic drift. But even isolated autochthonous contribution to the population’s survival. Reinforcement of populations by translocating populations – all of which experienced severe bottlenecks in the 19th and/or 20th century individuals from larger populations is proving to be a viable strategy for reducing inbreeding, – can suffer from genetic deterioration if they remain isolated. increasing genetic diversity and potentially saving populations from extinction. However, the effectiveness of population reinforcement hinges on a thorough understanding of the The existing highway network in Central and Western Europe poses a serious connectivity genetic status of the target population and the long-term consequences of translocation, problem for the already fragmented and small reintroduced lynx populations. In particular, which can be achieved through close genetic monitoring. The selection of optimal great efforts are being made to connect the Dinaric population in Bosnia, Croatia and translocation is discussed in more detail in Pazhenkova et al. 2024. Slovenia with the Alpine population in Italy. The highway connecting Ljubljana and Trieste is a permanent barrier with few crossing possibilities (Kuralt et al. 2023, Kuralt et al. in Lynx populations in the Alps and neighbouring areas have been demographically stable, preparation, Sanchez et al. in preparation). Connecting the remaining Alpine populations but the lack of connectivity between these populations raises the question of whether (in Switzerland, France and Austria) remains a challenge and will probably mainly depend they will survive in the long term without active management. Lynx translocations have on translocations and reintroductions, as happened in the Kalkalpen National Park (Upper been shown to be beneficial in small populations, as rescue effects by natural immigration Austria) in 2011 and 2013 (Fuxjäger 2014). The most important area for the Alpine lynx have had minimal impact due to low local connectivity (Sánchez Arribas et al. 2023, in population is in the north-western Alps (western Switzerland), followed by north-eastern preparation). However, adding individuals in their model simulations did not result in Switzerland and the south-eastern Alps (Italy and Slovenia). These populations are the result sufficient connectivity between populations in the SE Alps and Dinaric populations to of reintroductions in the early 1970s with very few founder animals, and both populations meet the 50/500 rule (Franklin 1980), suggesting that habitat fragmentation and human-have reached a high inbreeding coefficient. Two other smaller nuclei are located in the associated risks hinder dispersal. Indeed, observations of lynx dispersing into new areas Chartreuse (France) and in the Kalkalpen region (Schnidrig et al. 2016). There is still no are rare (Drouet-Hoguet et al. 2021, Zimmermann et al. 2005, 2007). This finding underlines reproducing lynx core in the German Alps and the closest lynx subpopulations are located the need to create landscape corridors, assisted dispersal and the further connection of in north-eastern Switzerland (distance 70 km) and Slovenia (distance 180 km), apart from the populations through stepping-stone reintroductions (Molinari et al. 2021) to achieve the population in the Šumava ecosystem (Bohemian Forest), which is, however, separated metapopulation structures (McManus et al. 2015, Sharma et al. 2013). The creation and from the Alps by open agricultural areas (Schnidrig et al. 2016). Although the Alpine lynx maintenance of a lynx metapopulation requires the cooperation of all affected countries population is still far from being (genetically) viable, this is the only mountain range in in the area, which must be organised under the auspices of international treaties. Western and Central Europe that could harbour an isolated viable population given its suitable habitat. The Alps are therefore a future stronghold for the species and also crucial Balkan lynx has been intrinsically small for at least the past 150 generations (Bazzicalupo et for connectivity with neighbouring populations, e.g. the Dinaric, Bohemian-Bavarian-al. 2022). Already experiencing few bottlenecks in the last 100 years, its genetic resistance Austrian, Black Forest and Jura populations (von Arx et al. 2021; Molinari-Jobin et al. 2021). is ever so weak in withstanding the rapid environmental change. The next steps of its The overarching goal is to establish a large Central European metapopulation (Bonn Lynx recovery will most likely involve a genetic rescue mission in order to strengthen its genetic Expert Group 2021). However, the strong anthropogenic fragmentation of otherwise good variability. Given that the Balkan lynx is genetically and taxonomically unique it has been habitat patches may require a partially managed metapopulation (e.g. assisted dispersal, questioned which subspecies is better candidadate for such a measure, however using genetic rescue, stepping-stones), which requires a range-wide strategy and reasonable Carpathian males, mimicking recent gene flow, was suggested (Melovski et al. 2022). cooperation between all range states concerned. 44 45 Recent genetic studies (Gugolz et al. 2008, Cómert et al. 2018, Bazzicalupo et al. 2022) indicate a closer relationship with the Carpathian subspecies compared to the Caucasian subspecies, but analyses of ecological traits of the Balkan and neighbouring populations established quite clearly that the Carpathian subspecies has much better ecological fit for the phylogeographic and current genetic makeup of the Balkan subspecies (Melovski et al. 2022). IUCN/ SSC Guidelines (2013) provide a clear direction on taking extreme caution when mixing different genetic lineages due to potential outbreeding depression. Historically, the lynx populations in the Alps, on the Balkan Peninsula and in the Carpathian region were interconnected (Kratochvil and Vala 1968). As a short-term and urgent measures, reinforcements in existing populations of Carpathian lynx and especially Balkan lynx should have been a paramount priority. Apart from further reinforcements, reintroductions and assisted migration/dispersal, the management goals should be directed also towards a natural connection of Dinaric - SE Alpine population with other lynx populations in Europe (Pazhenkova and Skrbinšek 2024). The population “stepping stone” established in the Julian Alps within the LIFE Lynx project served this purpose. The Julian Alps are within the average dispersal distance from the current lynx population in the Dinaric Mountains of Slovenia, but improving connectivity between these areas would help maintain adequate natural gene flow between the stepping-stone nucleus and the core population, for which permeability of the Ljubljana-Trieste highway is of particular importance (Kuralt et al. 2023, Pazhenkova and Skrbinšek 2024, Kuralt et al. in preparation, Mlinarič et al. in preparation). In the long term, further stepping stone nuclei should be created in the identified habitat patches in the Alps and the Balkans in order to connect the Dinaric-SE Alpine population with other, currently isolated lynx populations in the Alps (Molinari-Jobin et al. 2003) and with a Balkan lynx in the south. Our connectivity analysis of the habitat patches revealed that the Dinaric population and (remnant) Carpathian population from eastern Serbia and western Bulgaria could colonize the same large habitat patch extending from eastern Bosnia and Herzegovina to southern Serbia and serve as a stepping stone to the Balkan lynx population. This would create a functional metapopulation across the southern Balkans, the Dinaric Mountains and the Alpine arc and ensure gene flow, reducing the need for further translocations from the Carpathians. 46 47 References 10 56.) Hanski, I. 1998. Metapopulatin dynamics. Nature, 396, 41–49. Rhode Island, USA, pp. 75-107. 57.) Hemmingmoore H, Aronsson M, Åkesson M, Persson J, Andrén H. 2020. 86.) Linnell, J.D., Trouwborst, A., Boitani, L., Kaczensky, P., Huber, D., Reljic, S., Evaluating habitat suitability and connectivity for a recolonizing large Hayward, M.W., 2016. Border security fencing and wildlife: the end of the carnivore. Biological Conservation, 242: 108352. transboundary paradigm in Eurasia? PLoS Biology, 14 (6), e1002483. 58.) Herfindal, I., Linnell, J. D. C., Odden, J., Nilsen, E. B., & Andersen, R. 2005. 87.) Linnell, J.D.C., Andersen, R., Kvam, T., Andren, H., Liberg, O., ‐Odden, J., Prey density, environmental productivity and home‐range size in the Moa, P.F., 2001. Home range size and choice of management strategy for Eurasian lynx (Lynx lynx). Journal of Zoology, 265(1): 63–71. lynx in Scandinavia. Environmental management 27(6): 869-879. 59.) Herrero, A., Heikkinen, J. & Holmala, K. 2020. Movement patterns and 88.) Macdonald, D.W., Johnson, D.D.P., 2001. Dispersal in theory and practice: habitat selection during dispersal in Eurasian Lynx. Mamm Res 65: 523– consequences for conservation biology. In: Clobert, J., Danchin, E., 533. Dhondt, A.A., Nichols, J.D. (eds.): Dispersal. Oxford University Press, New 60.) Heurich M., Schultze-Naumburg J., Piacenza N., Magg N., Červený J., York: 358-372. Engleder T., Herdtfelder M., Sladova M., Kramer-Schadt S. 2018. Illegal 89.) Maehr, D.S., Land, E.D., Shindle, D.B., Bass, O.L., Hoctor, T.S., 2002. Florida hunting as a major driver of the source-sink dynamics of a reintroduced panther dispersal and conservation. Biological Conservation, 106 (2): 187-lynx population in Central Europe. Biological Conservation, 224: 355–365. 197. 1.) Allendorf, F. W., Luikart, G. 2009. Conservation and the genetics of 28.) Chapron, G., Kaczensky, P., Linnell, J.D.C., von Arx, M., Huber, D., Andrén, 61.) Hočevar et al. 2024. Monitoring of survival, movement, reproduction and 90.) Magg, N., Müller, J., Heibl, C., Hackländer, K., Wölfl, S., Wölfl, M., Heurich, populations. John Wiley & Sons. H. et al. 2014. Recovery of large carnivores in Europe’s modern human-predation of translocated lynx and other lynx equipped with GPS collars. M., 2016. Habitat availability is not limiting the distribution of the 2.) Anantharaman R., Hall K., Shah V. B., Edelman A., 2020. Circuitscape in dominated landscapes. Science, 346, 1517–1519. Technical report. Ljubljana, March 2024, 59 pp. Bohemian-Bavarian lynx Lynx lynx population. Oryx, 50 (4): 742-752. Julia: High performance connectivity modelling to support conservation 29.) Charlesworth D, and Charlesworth B. 1987. Inbreeding depression and its 62.) Howard W. E. 1960. Innate and Environmental Dispersal of Individual 91.) Matyushkin, Y.N. and Vaisfeld, M.A. 2003. The lynx – regional features of decisions. JuliaCon Proceedings, 1, 58. evolutionary consequences. Annual Review of Ecology, Evolution, and Vertebrates. The American Mind Naturalist, 63, 1: 152-161. ecology, use and protection. Nauka, Moscow, Russia. 3.) Andrén H. 1994. Effects of habitat fragmentation on birds and mammals Systematics Vol. 18:237-268. 63.) Huck M., Jędrzejewski W., Borowik T., Miłosz-Cielma M., Schmidt K., 92.) McManus, J. S., Dalton, D. L., Kotzé, A., Smuts, B., Dickman, A., Marshal, J. in landscapes with different proportions of suitable habitat: A review. 30.) Chubaty A., Galpern P., Doctolero S. 2023. grainscape: Landscape Jędrzejewska B., Nowak S., Mysłajek R. W. 2010. Habitat suitability, P., & Keith, M. 2015. Gene flow and population structure of a solitary top Oikos 71: 355–366. Connectivity, Habitat, and Protected Area Networks. R package version corridors and dispersal barriers for large carnivores in Poland. Acta carnivore in a human‐dominated landscape. Ecology and Evolution, 5(2): 4.) Avgan, B., Zimmermann, F., Güntert, M., Arikan, F., and Breitenmoser, U. 0.4.4, https://CRAN.R-project.org/package=grainscape Theriologica, 55: 177-192. 335–344. 2014. The first density estimation of an isolated Eurasian Lynx population 31) Clobert, J., Wolff, J.O., 2001. Introduction. Dispersal (eds. Clobert, J., 64.) IUCN/SSN, I. 2013. Guidelines for reintroductions and other conservation 93.) McRae B. H., Dickson B. G., Keitt T. H., Shah V. B. 2008. Using circuit theory in south-west Asia. Wildlife Biology 20: 217-221. Danchin, E., Dhondt, A. A., Nichols, J. D.), Oxford University Press, New translocations. Animal Welfare, 22(4), 489–490. https://doi.org/10.1017/ to model connectivity in ecology, evolution, and conservation. Ecology, 5.) Bao, W. 2010. Eurasian Lynx in China - present status and conservation York, pp. 17 - 21. S0962728600005637. 89, 10: 2712-2724. challenges. Cat News Special Issue 5: 22-25. 32.) CMS 2023. Proposal for the inclusion of the eurasian lynx (lynx lynx) 65.) Iuell, B., Bekker, G.J., Cuperus, R., Dufek, J., Fry, G., Hicks, C., Hlavác, V., 94.) McRae B. H., Popper K., Jones A., Schnidel M., Buttrick S., Hall K., Unnasch 6.) Basille M., Herfindal I., Santin‐Janin H., Linnell J. D., Odden J., Andersen R., in appendix II and balkan lynx (Lynx l. balcanicus) in appendix I of the Keller, V.B., Rosell, C., Sangwine, T., Tørsløv, N., Wandall, Maire, B. (Eds.), R. S., Platt J. 2016. Conserving Nature’s Stage: Mapping Omnidirectional Høgda K. A., Gaillard J. M. 2009. What shapes Eurasian lynx distribution convention, UNEP/CMS/COP14/Doc.31.4.1/Rev.1. 2003. Wildlife and Traffic: A European Handbook for Identifying Conflicts Connectivity for Resilient Terrestrial Landscapes in the Pacific Northwest. in human dominated landscapes: selecting prey or avoiding people?. 33.) Cömert, N., Carlı, O., Dinçtürk, H. B. 2018. The missing lynx of Eurasia at and Designing Solutions. Habitat Fragmentation due to Transportation Portland, Oregon, The Nature Conservancy: 47 pp. Ecography, 32, 4: 683-691. its Southern edge: a connection to the critically endangered Balkan lynx. Infrastructure. COST 341 pp. 95.) Melovski D., Ivanov G., Stojanov A., Avukatov V., Gonev A., Pavlov A., 7.) Bastille-Rousseau G., Potts J. R., Yackulic C. B., Frair J. L., Ellington E. H., Mitochondrial DNA Part A, 29(8): 1269–1275. 66.) Jamieson I.G. and Lacy R.C. 2012. Managing genetic issues in Breitenmoser U., von Arx M., Filla M., Krofel M., Signer J., Balkenho, N. Blake S. 2016. Flexible characterization of animal movement pattern 34.) Council of Europe. 2011. Conservation Strategy and National Action Plans reintroduction biology. in: Ewen J.G. Reintroduction Biology: Integrating 2020. First insight into the spatial and foraging ecology of the critically using net squared displacement and a latent state model. Movement for the conservation of the Critically Endangered Balkan Lynx. T-PVS/Inf Science and Management. Wiley-Blackwell: 441-475. endangered Balkan lynx (Lynx lynx balcanicus, Buresh 1941). Hystrix, 31, ecology, 4: 1-12. (2011) 33. Council of Europe, Strasbourg, France, 30 pp. 67.) Jedrzejewski, W., Jedrzejewska, B., Okarma, H., Schmidt, K., Bunevich, A.N. 1: 26-34. 8.) Bazzicalupo E, Lucena-Perez M, Kleinman-Ruiz D et al. 2022. History, 35.) Cristescu B., Domokos C., Teichman K. J., Nielsen S. E. 2019. Large and Milowski, L. 1996. Population dynamics (1896-1994), demography, 96.) Melovski, D. 2022. Distribution, ecology and conservation priorities of demography and genetic status of Balkan and Caucasian Lynx lynx carnivore habitat suitability modelling for Romania and associated and home ranges of the lynx in Bialowieza Primeval Forest (Poland and the critically endangered Balkan Lynx (Lynx lynx balcanicus Bureš, 1941). (Linnaeus, 1758) populations revealed by genome-wide variation. Divers predictions for protected areas. PeerJ, 7: e6549. Belarus). Ecography 19: 122-138. Doctorate thesis. University of Goettingen. 122 pp. Distrib 28(1):65–82. 36.) Čop, J., Frkovic, A., 1998. The re-introduction of the lynx in Slovenia and 68.) Jędrzejewski, W., Nowak, S., Schmidt, K., Jędrzejewska, B., 2002. The wolf 97.) Melovski, D., Ivanov, Gj., Stojanov, A., Trajce, A., Hoxha, B., von Arx, M., 9.) Becker T. 2013. Modelling Eurasian lynx distribution and estimation its present status in Slovenia and Croatia. Hystrix, 10 (1), 65-76. and the lynx in Poland—results of a census conducted in 2001. Kosmos, Breitenmoser-Wursten, Ch., Hristovski, S., Shumka, S. and Breitenmoser, of patch and population size in the Alps. Master thesis. University of 37.) Drouet-Hoguet, N., Chenesseau, D., Kunz, F., & Zimmermann, F. 2021. 51 (4): 491-499. U. 2013. Distribution and conservation status of the Balkan Lynx (Lynx London, London, United Kingdom. Situation of the lynx in the Jura Mountains. CATnews Species Issue, 14: 69.) Kaczensky P., Linnell J. D. C., Huber D., von Arx M., Andren H., Breitenmoser lynx balcanicus Bureš, 1941). IV Congress of Ecologists of the Republic of 10.) Bell, W.J. 1990. Searching behavior patterns in insects. Annual Review of 29–34. U., Boitani L. 2021. Distribution of large carnivores in Europe 2012 - 2016: Macedonia with international participation Special Issue. Ohrid, Republic Entomology 35, 447-67. 38.) Duľa M. et al. 2021. Multi-seasonal systematic camera-trapping reveals Distribution maps for Brown bear, Eurasian lynx, Grey wolf, and Wolverine of Macedonia. 11.) Bezanson J., Edelman A., Karpinski S. and Shah V. B. 2017. Julia: A fresh fluctuating densities and high turnover rates of Carpathian lynx on the [Dataset]. Dryad. 98.) Melovski, D., Krofel, M., Avukatov, V. et al. 2022. Diverging ecological traits approach to numerical computing. SIAM review, 59, 1: 65-98. western edge of its native range. Scientific Reports 11: 9236. 70.) Kaczensky, P., Chapron, G., von Arx, M., Huber, D., Andrén, H. and Linnell between the Balkan Lynx and neighbouring populations as a basis for 12.) Bird Jackson H, Fahrig L. 2013. Habitat Loss and Fragmentation. In: 39.) Elith J., H. Graham C., P. Anderson R., Dudík M., Ferrier S., Guisan A., J. (eds). 2012. Status, management and distribution of large carnivores - planning its genetic rescue. Mamm Biol 102: 1697– 1708. Encyclopedia of Biodiversity (Second Edition), Editor(s): SA Levin, Hijmans R. J., Huettmann F., Leathwick J. R., Lehmann A., Li J., Lohmann bear, lynx, wolf & wolverine - in Europe. 99.) Mengüllüoğlu D, Ambarlı H, Berger A, Hofer H. 2018. Foraging ecology of Academic Press: 50-58. L. G., Loiselle B. A., Manion G., Moritz C., Nakamura M., Nakazawa Y., 71.) Kaczensky, P., Chapron, G., von Arx, M., Huber, H., Andrén, H., Linnell, J., Eurasian lynx populations in southwest Asia: Conservation implications 13.) Boitani L., Alvarez F., Anders O., Andren H., Avanzinelli E., Balys V., ... & Overton J. McC. M., Townsend Peterson A., Phillips S. J., Richardson K., 2003. Status, management and distribution of large carnivores - bear, for a diet specialist. Ecology & Evolution 8: 9451–9463. Zlatanova D. 2015. Key actions for Large Carnivore populations in Europe. Scachetti-Pereira R., Schapire R. E., Soberón J., Williams S., Wisz M. S., lynx, wolf and wolverine in Europe. Report. European Comission, pp. 72. 100.) Mladenoff, D.J., Sickley, T.A., Haight, R.G., Wydeven, A.P., 1995. A regional Institute of Applied Ecology (Rome, Italy). Report to DG Environment, Zimmermann, N. E. 2006. Novel methods improve prediction of species’ 72.) Kitchener, A. C., Breitenmoser-Würsten, Ch., Eizirik, E., Gentry, A., landscape analysis and prediction of favorable gray wolf habitat in the European Commission, Bruxelles. 120 pp. distributions from occurrence data. Ecography, 29, 2: 129-151. Werdelin, L., Wilting, A., et al. 2017. A revised taxonomy of the Felidae. northern Great Lakes region. Conservation Biology, 9 (2): 279-294. 14.) Bonn Lynx Expert Group. 2021. Recommendations for the conservation of 40.) Fahrig L 2001. How much habitat is enough? Biological Conservation 100: The final report of the Cat Classification Task Force of the IUCN/SSC Cat 101.) Mohr, C.O., 1947. Table of equivalent populations of North American small the Eurasian Lynx lynx lynx in Western and Central Europe. Conclusions 65–74. Specialist Group. Cat News Special Issue 11, 80 pp. mammals. The American Midland Naturalist, 37 (1): 223-249. from the workshop of the “Bonn Lynx Expert Group” in Bonn, Germany, 41.) Fahrig L. 2003. Effects of habitat fragmentation on biodiversity. Annual 73.) Kramer-Schadt S., Revilla E., Wiegand T. 2005. Lynx reintroductions 102) Molinari-Jobin A., Molinari P., Breitenmoser-Würsten Ch., Wölfl M., Stanisa 16–19 June 2019. Cat News Special Issue 14, 78–86. Review of Ecology, Evolution, and Systematics 34: 487–515. in fragmented landscapes of Germany: Projects with a future or C., Fasel M. ... & Breitenmoser U. 2003. The Pan-Alpine Conservation 15.) Boyce, M. S. 1992. Population Viability Analysis. Annual Review of Ecology, 42.) Fahrig L. 2007. Non-optimal animal movement in human-altered misunderstood wildlife conservation?. Biological Conservation, 125, 2: Strategy for the Lynx. Nature and Environment, No. 130. Convention on Evolution, and Systematics, 23, 481–506. landscapes. Functional Ecology 21: 1003–1015. 169-182 the Conservation of European Wildlife and Natural. 16.) Breiman L. 2001. Random forests. Machine Learning, 45, 1: 5–32, 43.) Fležar U., Hočevar L., Sindičić M., Gomerčić T., Konec M., Slijepčević V., 74.) Kramer-Schadt S., Revilla E., Wiegand T., Breitenmoser U. 2004. 103.) Molinari-Jobin, A., Marboutin, E., Wölfl, S., Wölfl, M., Molinari, P., Fasel, 17.) Breitenmoser U, Slough BG, Breitenmoser-Würsten C. 1993. Predators of Bartol M., Bojte B., Črtalič J., Jan M., Kljun F., Molinari-Jobin A., Pičulin A., Fragmented landscapes, road mortality and patch connectivity: M., Kos, I., Blažič, M., Breitenmoser, C., Fuxjäger, C., Huber, T., Koren, cyclic prey: is the Canada lynx victim or profiteer of the snowshoe hare Gotar T., Javornik J., Portas Perez R., Potočnik H., Rot A., Skrbinšek T., … modelling influences on the dispersal of Eurasian lynx. Journal of Applied I., & Breitenmoser, U. 2010. Recovery of the Alpine lynx Lynx lynx cycle? Oikos 66:551–554. Krofel M. 2024. Surveillance of the reinforcement process of the Dinaric Ecology, 41, 4: 711-723 metapopulation. Oryx, 44(2): 267–275. 18.) Breitenmoser U., Breitenmoser-Würsten Ch., Okarma H., Kaphegyi - SE Alpine lynx population in the lynx-monitoring year 2022-2023: final 75.) Kramer-Schadt, S., Revilla, E., & Wiegand, T. 2005. Lynx reintroductions 104.) Molinari P., Breitenmoser U., Černe R., Fuxjäger C., Weingarth K., Ryser A. T., Kaphgyi-Wallmann U. & Müller U. M. 2000. Action Plan for the report. Technical report. Ljubljana, March 2024, 89 p. in fragmented landscapes of Germany: Projects with a future or & Molinari-Jobin A. 2021. The contribution of stepping-stone releases for conservation of the Eurasian Lynx (Lynx lynx) in Europe. Nature and 44.) Fležar U., Hočevar L., Sindičić M., Gomerčić T., Konec M., Slijepčević V., misunderstood wildlife conservation? Biological Conservation, 125(2): enhancing lynx distribution. Cat News Special Issue 14: 46–49. Environment, No.112. Convention on the Conservation of European Bartol M., Bojte B., Črtalič J., Jan M., Kljun F., Molinari-Jobin A., Pičulin A., 169–182. https://doi.org/10.1016/j.biocon.2005.02.015 105.) Mowat, G.K, Poole, E., O’Donoghue, M., 2000. Ecology of lynx in Wildlife and Natural Habitats (Bern Convention). Strasbourg, France, 70 Gotar T., Javornik J., Portas Perez R., Potočnik H., Rot A., Skrbinšek T., … 76.) Kramer-Schadt, S., Revilla, E., Wiegand, T., & Breitenmoser, U. 2004. northern Canada and Alaska. In: Ruggiero, L.F. et al. (eds.): Ecology and pp. Krofel M. 2023. Surveillance of the reinforcement process of the Dinaric— Fragmented landscapes, road mortality and patch connectivity: conservation of lynx in the United States. University Press of Colorado, 19.) Breitenmoser U., Kavczensky P., Dötterer M., Breitenmoser‐Würsten C., SE Alpine lynx population in the lynx-monitoring year 2021-2022 (p. 73) Modelling influences on the dispersal of Eurasian lynx. Journal of Applied Boulder, USA: 265-306. Capt S., Bernhart F., Liberek M. 1993. Spatial organization and recruitment [Technical report]. Ecology, 41(4): 711–723. https://doi.org/10.1111/j.0021-8901.2004.00933.x 106.) Mowat, G., Slough, B.G., 1998. Some observations on the natural history of lynx (Lynx lynx) in a re‐introduced population in the Swiss Jura 45.) Frankham, R. et al. 2002. Introduction to Conservation Genetics. 77.) Kramer-Schadt, S., Revilla, E., Wiegand, T., & Grimm, V. 2007. Patterns for and behaviour of the Canada lynx, Lynx canadensis. Canadian Field-Mountains. Journal of Zoology, 231, 3: 449-464. Cambridge University Press, Cambridge. parameters in simulation models. Ecological Modelling, 204(3–4): 553– Naturalist, 112 (1): 32-36. 20.) Breitenmoser, U. and Breitenmoser-Würsten, Ch. 2008. Der Luchs: Ein 46.) Frankham R., Ballou J. D., Briscoe D. A. 2010. Introduction to Conservation 556. 107.) Mowat, G.K, Poole, E., O’Donoghue, M., 2000. Ecology of lynx in Grossraubtier in der Kulturlandschaft. Salm Verlag, Wohlen/Bern. Genetics, 2nd ed. Cambridge, Cambridge University Press. 78.) Kramer-Schadt, S., Revilla, E., Wiegand, T., Breitenmoser, U., 2004. northern Canada and Alaska. In: Ruggiero, L.F. et al. (eds.): Ecology and 21.) Breitenmoser, U., Breitenmoser-Würsten, C., & Capt, S. 1998. Re-47.) Frankham, R. 2010. Challenges and opportunities of genetic approaches Fragmented landscapes, road mortality and patch connectivity: conservation of lynx in the United States. University Press of Colorado, introduction and present status of the lynx (Lynx Lynx) in Switzerland. to biological conservation. Biological Conservation, 143(9): 1919–1927. modelling influences on the dispersal of Eurasian lynx. Journal of Applied Boulder, USA, 265-306. Hystrix, the Italian Journal of Mammalogy, 10(1), 17–30. 48.) Friedman J. H. 2001. Greedy function approximation: a gradient boosting Ecology, 41 (4): 711-723. 108.) Mueller, S. A., Prost, S., Anders, O., Breitenmoser-Würsten, C., Kleven, 22.) Breitenmoser, U., Breitenmoser-Würsten, C., Lanz, T., von Arx, M., machine. Annals of statistics: 1189-1232. 79.) Kratochovil, J., Vala, F. 1968. History of occurence of the Lynx in Bohemia O., Klinga, P., ... & Nowak, C. (2022). Genome-wide diversity loss in Antonevich, A., Bao, W. & Avgan, B. 2015. Lynx lynx (errata version 49.) Fuxjäger C., Laass J. & Molinari-Jobin A. 2012. Eurasian Lynx (Lynx lynx) and Moravia. Acta Sc. Nat. Brno 2(4): 33-48. reintroduced Eurasian lynx populations urges immediate conservation published in 2017). The IUCN Red List of Threatened Species 2015. in the Austrian Alps in period 2005-2009. Acta Biologica Slovenica 55(2): 80.) Krebühl, J., Zimmermann, F., Herdtfelder, M., Idelber-Ger, S., Suchant, R., management. Biological Conservation, 266, 109442. 23.) Breitenmoser-Würsten, C. and Obexer-Ruff, G. 2003. Population genetics 65-69. Drouet-Hoguet, N., Breitenmoser-Würsten, Ch. & Breitenmoser, U. 2021. 109.) Newman M. E. J. 2005. A measure of betweenness centrality based on of two reintroduced lynx (Lynx lynx) populations in Switzerland: a 50.) Gimenez O, Gatti S, Duchamp Ch, Germain E, Laurent A, Zimmermann F, Transboundary cooperation in lynx conservation under the auspice of random walks. Social networks, 27, 1: 39-54. molecular evaluation 25 years after translocation. Progress report. KORA Marboutin E. 2019. Spatial density estimates of Eurasian lynx (Lynx lynx) the Upper Rhine Conference. Cat News Special Issue 14: 55-56. 110.) Nowell, K. and Jackson, P. 1996. Wild Cats. Status Survey and Bericht. Bern. in the French Jura and Vosges Mountains. Ecology & Evolution 9: 11707– 81.) Krofel M., 2012. Predation-related interspecific interactions in Eurasian Conservation Action Plan. IUCN/SSC Cat Specialist Group, Gland, 24.) Breitenmoser-Würsten, C., Vandel, J.-M., Zimmermann, F. and 11715. Lynx (Lynx lynx) in northern Dinaric Mountains. Doctorate thesis. Switzerland and Cambridge, UK. Breitenmoser, U. 2007. Demography of Lynx lynx lynx in the Jura 51.) Gompper, M. E., Gittleman, J.L., 1991. Home range scaling: intraspecific University of Ljubljana. 111.) O’Donoghue, M., Boutin, S., Krebs, C.J., Hofer, E.J., 1997. Numerical Mountains. Wildlife Biology 13: 381-392. and comparative trends. Oecologia, 87: 343-348. 82.) Krofel, M., Huber, D., Kos, I., 2011. Diet of Eurasian lynx Lynx lynx in the responses of coyotes and lynx to the snowshoe hare cycle. Oikos, 80, 25.) Bunnefeld N., Börger L., Van Moorter B., Rolandsen C. M., Dettki H., 52.) Guarino, F., 2002. Spatial ecology of a large carnivorous lizard, Varanus northern Dinaric Mountains (Slovenia and Croatia). Acta Theriologica, 56 150-162. Solberg E. J., Ericsson G. 2011. A model-driven approach to quantify varius (Squamata: Varanidae). Journal of Zoology (London) 258: 449-457. (4), 315-322. 112.) Oeser J., Heurich M., Kramer-Schadt S., Mattisson J., Krofel M., Krojerová- migration patterns: Individual, regional and yearly differences. Journal of 53.) Gugolz, D., Bernasconi, M.V., Breitenmoser-Würsten, C., & Wandeler, P. 83.) Krofel, M., Skrbinšek, T., Kos, I., 2013. Use of GPS location clusters analysis Prokešová J., Zimmermann F., Anders O., Andrén H., Bagrade G., Belotti E., Animal Ecology, 80: 466–476. 2008. Historical DNA reveals the phylogenetic position of the extinct to study predation, feeding, and maternal behavior of the Eurasian lynx. Breitenmoser-Würsten C., Bufka L., Černe R., Drouet-Hoguet N., Dul’a M., 26.) Bunnefeld, N., Linnell, J. D. C., Odden, J., Van Duijn, M. A. J., & Andersen, Alpine lynx. Journal of Zoology, 275, 201-208. Ecological Research, 28 (1), 103-116. Fuxjäger C., Gomerčić T., J‐drzejewski W., Kont R., Koubek P., Kowalczyk R. 2006. Risk taking by Eurasian lynx (Lynx lynx) in a human-dominated 54.) Guisan A., Thuiller W., Zimmermann N. E., 2017. Habitat Suitability and 84.) Kuralt Ž., Črtalič J., Konec M., Molinari-Jobin A., Fuxjäger C.,Potočnik H. R., Kusak J., Kubala J., Kutal M., Linnell J. D. C., Molinari-Jobin A., Männil landscape: Effects of sex and reproductive status. Journal of Zoology, Distribution Models: With Applications in R. Cambridge, Cambridge 2023. Habitat suitability and connectivity models for lynx between and P., Middelhoff T. L., Odden J., Okarma H., Oliveira T., Pagon N., Persson 270(1), 31–39. University Press. within the Southeastern Alps and Dinaric Mountains area. Final report. J., Remm J., Schmidt K., Signer S., Tám B., Vogt K., Kuemmerle T. 2023. 27.) Burgman, M.A., Fox, J.C., 2003. Bias in species range estimates from 55.) Haller, H Breitenmoser, U., 1986. Zur Raumorganisation der in den Ljubljana, University of Ljubljana, Biotechnical faculty, May 2023: 25 pp. Integrating animal tracking datasets at a continental scale for mapping minimum convex polygons: implications for conservation and options for Schweizer Alpen wiederangesiedelten Population des Luchses Lynx 85.) Levins, R., 1970. Extinction. In: Gerestenhaber, M. (ed.): Some Mathematical Eurasian lynx habitat. Diversity and Distributions, 29: 1546–1560. improved planning. Animal Conservation, 6 (1), 19-28. lynx. Zeitschrift fuer Saeugetierkunde, 51 (5): 289-311. Problems in Biology’. American Mathematical Society, Providence, 113.) Palmero, S., Belotti, E., Bufka, L. Gahbauer M, Heibl Ch, Premier J, 48 49 Weingarth-Dachs K, Heurich M. 2021. Demography of a Eurasian lynx Slovenia: 122-147. (Lynx lynx) population within a strictly protected area in Central Europe. 142.) Skrbinšek T., Boljte B., Jelenčič M., Sindičić M., Paule L., Promberger B., Sci Rep 11, 19868. Rigg R., Bartol M., Pop M., Kubala J., Slijepčević V., Molinari P., Molinari-114.) Pesenti E. & Zimmermann F. 2013. Density estimations of the Eurasian Jobin A., Krofel M., Bizjan P., Hočevar L., Oliveira T., Potočnik H., Kos I., lynx (Lynx lynx) in the Swiss Alps. Journal of Mammalogy 94: 73–81. Kljun F., Stronen A. V., Tám B., Črtalič J., Konec M. 2019. Baseline (pre-115.) Panayotopoulou, M. and Godes, C. 2004. Status and conservation of the reinforcement) genetic status of SE Alpine and Dinaric Lynx population. Eurasian Lynx (Lynx lynx) in Europe. In: M. von Arx, Ch. Breitenmoser-Ljubljana, University of Ljubljana: 24 pp. Wuersten, F. Zimmermann, U. Breitenmoser (ed.), KORA Bericht Nr. 19. 143.) Soulé, M. E. 1980. Thresholds for Survival: Maintaining Fitness and 116.) Pazhenkova E., Skrbinšek T. 2021. Guidelines for Ensuring the Long-term Evolutionary Potential. In Conservation Biology: An Evolutionary-Viability of Lynx in the Dinaric Mountains and South Eastern Alps. Action Ecological Perspective: 151–170. D.3 report. Ljubljana, University of Ljubljana, April 2024: 13 pp. 144.) Sunquist, M. and Sunquist, F. 2002. Wild Cats of the World. University of 117.) Pazhenkova, E., Skrbinšek, T. 2021a. Optimal management scenarios for Chicago Press. ensuring viability of lynx in the Dinaric mountains and South eastern Alps. 145.) Tensen, L., Van Vuuren, B. J., Du Plessis, C., & Marneweck, D. G. 2019. Ljubljana. 15 pp. African wild dogs: Genetic viability of translocated populations across 118.) Pazhenkova, E., Skrbinšek, T. 2024. Guidelines for Ensuring the Long-South Africa. Biological Conservation, 234: 131–139. term Viability of Lynx in the Dinaric Mountains and South Eastern Alps. 146.) Tesson S. V. M., Edelaar P. 2013. Dispersal in a changing world: Ljubljana. 12 pp. Opportunities, insights and challenges. Movement Ecology, 1: 1–6. 119.) Phillips S. J., Anderson R. P., Schapire R. E. 2006. Maximum entropy 147.) Topličanec I., Gomerčić T., Černe R., Krofel M., Pop I. M., Kubala J., Tám B., modeling of species geographic distributions. Ecological Modelling, 190: Blašković S., Sindičić M. 2022. Early post-release behaviour of Eurasian 231-259. lynx translocated to the transboundary region of the Dinaric Mountains. 120.) Phillips S. J., Dudík M. 2008. Modeling of species distributions with Journal of Vertebrate Biology, 71, 22064: 1-16. Maxent: new extensions and a comprehensive evaluation. Ecography, 31: 148.) Trajçe, A., Hoxha, B., Trezhnjeva, B. and Mersini, K. 2014. Munella Mountain. 161-175. Summary of findings from the Balkan Lynx Recovery Programme. 121.) Potočnik H., Skrbinšek T., Kos I. 2009. The reintroduced Dinaric lynx Protection and Preservation of Natural Environment in Albania, Tiranë. population dynamics in PVA simulation: the 30 years retrospection and 149.) Valavi R., Guillera-Arroita G., Lahoz-Monfort J. J., Elith J. 2021. Predictive the future viability. Acta biologica Slovenica, 52, 1: 3-18. performance of presence-only species distribution models: A benchmark 122.) Potočnik, H., Al Sayegh-Petkovšek, S., De Angelis, D., Huber, Đ., Jerina, K., study with reproducible code. Ecological Monographs, 92, 1: e01486, Kusak, J., Mavec, M., Pokorny, B., Reljić, S., Rodriguez R., M., Skrbinšek, T., https://doi.org/10.1002/ecm.1486. Vivoda, B., 2019. Handbook for integrating the bear habitat suitability and 150.) von Arx M. 2020. Lynx lynx (amended version of 2018 assessment). The connectivity to spatial planning: prepared within the framework of the IUCN Red List of Threatened Species 2020: e.T12519A177350310. Life Dinalp Bear project. Potočnik, Hubert (ed.). Ljubljana: University of https://dx.doi.org/10.2305/IUCN.UK.2020-3. Ljubljana, 1-62. 151.) von Arx M., Breitenmoser-Würsten Ch., Zimmermann F. & Breitenmoser 123.) Potočnik H., Črtalič J., Kos I., Skrbinšek T. 2020. Characteristics of spatial U. 2004. Status and conservation of the Eurasian Lynx (Lynx lynx) in use and importance of landscape features for recovering populations of Europe in 2001. KORA Bericht Nr. 19, Muri b. Bern, Switzerland, 330 pp. Eurasian lynx (Lynx lynx). Acta Biologica Slovenica, 63, 2: 65-88. 152.) Waser, P.M., Strobeck, C., Paetkau, D., 2001. Estimating interpopulation 124.) Potočnik H, Al Sayegh-Petkovšek S, Črtalič J, Fležar U, Huber Đ, Jelenko dispersal rates. In Carnivore conservation (Gittleman J. L. Funk S. Turinek I, Krofel M, Konec M, Kuralt Ž, Kusak J, Molinari-Jobin A, Molinari M. Macdonald D. W. Wayne K., eds.). Cambridge University Press, P, Pokorny B, Reljić S, Skrbinšek T, Slijepčević V. 2023. Vključevanje Cambridge, United Kingdom: 484-497. povezljivosti in primernosti prostora za risa v prostorsko načrtovanje. 1. 153.) Wiens, J.A., 2001. The landscape context of dispersal. In: Clobert, J., izd. Ljubljana: Biotehniška fakulteta, Oddelek za biologijo, 66 pp. Danchin, E., Dhondt, A.A., Nichols, J.D. (eds.): Dispersal. Oxford University 125.) Ripari L., Premier J., Belotti E., Bluhm H., Breitenmoser-Würsten C., Bufka Press, New York, USA: 96-109. L., Červený J., Douet-Hoguet N., Fuxjäger C., Jędrzejewski W., Kont R., 154.) Woodroffe R. 2003. Dispersal and conservation: a behavioral perspective Koubek P., Kowalczyk R., Krofel M., Krojerová-Prokešová J., Molinarion metapopulation persistence. In: Festa-Bianchet M., Apollonio M. Jobin A., Okarma H., Oliveira T., Remm J., Schmidt K., Zimmermann F., (Eds.): Animal Behavior and Wildlife Conservation. Island Press: 33–48. Kramer-Schadt S., Heurich, M. 2022. Human disturbance is the most 155.) Worton, B.J., 1989. Kernel methods for estimating the utilization limiting factor driving habitat selection of a large carnivore throughout distribution in home-range studies. Ecology, 70 (1): 164-168. Continental Europe. Biological Conservation, 266, 109446. 156.) Zimmermann F. 2004. Conservation of the Eurasian Lynx (Lynx lynx) 126.) Ryser, A., Von Wattenwyl, K., Ryser-Degiorgis, M.P., Willisch, in a fragmented landscape- habitat models, dispersal and potential C., Zimmermann, F., Breitenmoser, U. 2004. Luchsumsiedlung distribution. Doctoral dissertation. Université de Lausanne, Faculté de Nordostschweiz 2001-2003, Schlussbericht Modul Luchs des Projektes biologie et medicine, Switzerland LUNO. Muri bei Bern. KORA Bericht. 22, 1-60. 157.) Zimmermann F., Breitenmoser U. 2002. A distribution model for the 127.) Samelius G, Andrén H, Liberg O, Linnell JDC, Odden J, Ahlqvist P, Eurasian lynx in the Jura Moun-tains, Switzerland. In: Scott J. M., Heglund Segerström P, Sköld K (2012). Spatial and temporal variation in natal P. J., Morrison M. L., Haufler J. B., Raphael M. G., Wall W. A., Samson F. B. dispersal by Eurasian Lynx in Scandinavia. J Zool 286:120–130. (Eds.): Predicting Species Occur-rences: Issues of Accuracy and Scale. 128.) Schadt S., Revilla E., Wiegand T., Knauer F., Kaczensky P., Breitenmoser Island Press, California, Covelo: 653-659. U., Bufka L., Červený J., Koubek P., Huber T., Staniša C., Trepl, L. 2002a. 158.) Zimmermann, .F, Breitenmoser-Würsten, C. &, Breitenmoser, U. 2005. Assessing the suitability of central European landscapes for the Natal dispersal of Eurasian Lynx (Lynx lynx) in Switzerland. J Zool 267: reintroduction of Eurasian lynx. Journal of Applied Ecology, 39, 2: 189-381–395. 203. 159.) Zimmermann, F., Breitenmoser, U., 2007. Potential distribution and 129.) Schadt S., Knauer F., Kaczensky P., Revilla E., Wiegand T., Trepl L. 2002b. population size of the Eurasian lynx Lynx lynx in the Jura Mountains and Rule‐based assessment of suitable habitat and patch connectivity for the possible corridors to adjacent ranges. Wildlife Biology, 13(4), 406-416. Eurasian lynx. Ecological applications, 12, 5: 1469-1483. 160.) Zimmermann, F., Breitenmoser-Würsten, C., Breitenmoser, U. 2007. 130.) Schadt, S.A., 2002. Scenarios assessing the viabililty of a lynx population Importance of dispersal for the expansion of a Eurasian lynx Lynx lynx in Germany. Szenarien für eine lebensfähige Luchspopulation in population in a fragmented landscape. Oryx, 41(3): 358-368. Deutschland (Doctoral dissertation). Technisches Univeritat München, 161.) Zimmermann, F., Breitenmoser-Würsten, C., Breitenmoser, U., 2005. Germany. Natal dispersal of Eurasian lynx (Lynx lynx) in Switzerland. Journal of 131.) Schmidt K. 1998. Maternal behaviour and juvenile dispersal in the Zoology, 267(4): 381-395. Eurasian Lynx. Acta Theriol 43: 391– 408. 162.) Zollner, P.A., Lima, S.L., 1997. Landscape-level perceptual abilities in 132.) Schmidt, K., Jedrzejewski, W. and Okarma, H. 1997. Spatial organization white-footed mice: perceptual range and the detection of forested and social relations in the Eurasian Lynx population in Bialowieza habitat. Oikos: 80: 51-60. Primeval Forest, Poland. Acta Theriologica 42: 289-312. 133.) Schmidt, K., Ratkiewicz, M. and Konopinski, M.K. 2011. The importance of genetic variability and population differentiation in the Eurasian Lynx (Lynx lynx) for conservation, in the context of habitat and climate change. Mammal review 412: 112-124. 134.) Schnidrig R., Nienhuis C., Imhof R., Bürki R. & Breitenmoser U. (Eds) 2016. Lynx in the Alps: Recommendations for an internationally coordinated management. RowAlps Report Objective 3. KORA Bericht Nr. 71. KORA, Muri bei Bern, Switzerland, and BAFU, Ittigen, Switzerland, 70 pp. 135.) Seidl L. 2023. Prehajanje prostoživečih živali prek obstoječih premostitvenih objektov na avtocesti A1 na odseku Vrhnika – Postojna. Master thesis, University of Ljubljana: 61 pp. 136.) Shah V. B., McRae B. H. 2008. Circuitscape: A tool for landscape ecology. Proceedings of the 7th Python in Science Conference, 7: 62–66. 137.) Sharma, S., Dutta, T., Maldonado, J. E., Wood, T. C., Panwar, H. S., & Seidensticker, J. 2013. Forest corridors maintain historical gene flow in a tiger metapopulation in the highlands of central India. Proceedings of the Royal Society B: Biological Sciences, 280(1767). 138.) Shields, W. M., 1982. Philopatry, inbreeding, and the evolution of sex. SUNY press. New York, USA, pp. 245. 139.) Signer, J., 2010. Distribution and Connectivity of Eurasian Lynx (Lynx lynx) in the Alps. ECONNECT, Austrian Environmental Agency, Vienna, Austra, pp 14. 140.) Sindičić M., Polanc P., Gomerčić T., Jelenčič M., Huber Đ., Trontelj P., Skrbinšek T. 2013. Genetic data confirm critical status of the reintroduced Dinaric population of Eurasian lynx. Conservation Genetics 14: 1009–1018 141.) Skrbinšek T. 2004. Model primernega prostora za risa v Sloveniji. In: eng. Lynx in Slovenia, background documents for conservation and management. University of Ljubljana, Biotechnical Faculty, Ljubljana, 50 49 Title, subtitle: Baselines for Establishing meta-population connectivity of Eurasian lynx populations in the Alps, Dinarics and Balkan; Handbook on suitability and connectivity of the space for Eurasian lynx in the area Editors: Hubert Potočnik and Eva Mlinarič Authors: Hubert Potočnik, Eva Mlinarič, Rok Černe, Jaka Črtalič, Urša Fležar, Christian Fuxjäger, Lan Hočevar, Marjeta Konec, Ivan Kos, Miha Krofel, Žan Kuralt, Anja Molinari-Jobin, Paolo Molinari, Elena Pazhenkova, Magda Sindićić, Tomaž Skrbinšek, Ira Topličanec Author of front cover photo: Matej Vranič Graphic design and layout: Agena d.o.o Published by: Biotechnical Faculty of University of Ljubljana, Department of Biology Ljubljana, 2024 Electronic version: http://www.lifelynx.eu ABOUT THE PROJECT Acronym: LIFE Lynx Project title: Preventing the Extinction of the Dinaric-SE Alpine Lynx Population Through Reinforcement and Long-term Conservation Project reference: LIFE16 NAT/SI/000634 FB: facebook.com/ LIFELynx.eu) E-mail: life.lynx.eu@gmail.com With the support of LIFE - the European Union Financial Mechanism. Kataložni zapis o publikaciji (CIP) pripravili v Narodni in univerzitetni knjižnici v Ljubljani COBISS.SI-ID 196410115 ISBN 978-961-7215-05-2 (PDF) Partners Co-financiers