Evaluating the Functional Connectivity of Natura 2000 Forest Patch for Mammals in Romania

Evaluating the Functional Connectivity of Natura 2000 Forest Patch for Mammals in Romania

Available online at www.sciencedirect.com ScienceDirect Procedia Environmental Sciences 32 (2016) 28 – 37 International Conference – Environment at ...

2MB Sizes 0 Downloads 35 Views

Available online at www.sciencedirect.com

ScienceDirect Procedia Environmental Sciences 32 (2016) 28 – 37

International Conference – Environment at a Crossroads: SMART approaches for a sustainable future

Evaluating the Functional Connectivity of Natura 2000 Forest Patch for Mammals in Romania Mihaita- Iulian Niculae*, Mihai Razvan Nita, Gabriel Ovidiu Vanau, Maria Patroescu Centre for Environmental Research and Impact Studies, University of Bucharest, 1 st N. Balcescu Blvd., Bucharest, 010041, Romania

Abstract Maintaining and improving landscape connectivity represents a central objective in biodiversity conservation. The present study aims to evaluate the functional connectivity of the Romanian Natura 2000 forest sites for mammals, and to determine the importance of individual forest patches in maintaining landscape connectivity. We established three groups of mammals according to the dispersion distance and the average home-range size: small mammals (1km; 1ha), intermediate (10km; 100ha) and large mammals (100km; 1000ha). For measuring the connectivity we used a graph theory approach and the software CONEFOR 2.6. The importance of each patch as an indicator for the connectivity of forest surfaces was determined using a binary index, the Harary index (dH). Forest surfaces included in the Natura 2000 network present a high connectivity for terrestrial mammals with a large dispersion value and home range when compared with other categories of mammals. Furthermore, results evidence that the connectivity objective of the Natura 2000 network is not totally fulfilled, especially for protected forest surfaces, requiring the focus of future activities on increasing the connectivity of the network. © Published by Elsevier B.V This © 2016 2016The TheAuthors. Authors. Published by Elsevier B.V.is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ECOSMART 2015. Peer-review under responsibility of the organizing committee of ECOSMART 2015 Keywords: functional connectivity; Harary index; CONEFOR; Natura 2000; forest patch; Romania

1. Introduction The loss of connectivity in forest habitats is one of the main problems for biodiversity conservation and conservation planning [1, 2, 3, 4], mainly due to habitat loss and fragmentation [5]. Maintaining and increasing landscape connectivity is important [6, 7], ensuring the dispersion of individuals and gene flow, essential in order to

* Corresponding author. Tel.: +4-021-310-3872; fax: +4-021-310-3872. E-mail address: [email protected]

1878-0296 © 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ECOSMART 2015 doi:10.1016/j.proenv.2016.03.009

Mihaita-Iulian Niculae et al. / Procedia Environmental Sciences 32 (2016) 28 – 37

stop the populational decline of certain species [8]. This can be done by setting up new corridors that will help organism dispersion or new habitats in critical areas [9]. To find ways to improve the landscape connectivity and thus the conservation status, specific methods to assess the existant landscape connectivity have to be selected. These methods evaluate both the structural and the functional connectivity through patch analysis [10]. The functional connectivity shows the degree in which the landscape facilitates or obstructs the movement of organisms between the landscape patches [11]. One of the methods used in landscape ecology is the graph theory [3, 12-15], allowing landscape connectivity assessment and representation. The graph theoretical methods (network analisys) requires indicators to identify key components of the landscape their importance for the connectivity [9]. Base elements for building landscape graphs are: the number of nodes (N) or habitat patches, and links (L) or connecting elements [3, 13, 16, 17]. A link between two patches allows an organism to move between the two connetcted patches [3]. Links can be well defined physical ecological corridors or favorable land cover [9, 17]. The importance of the nodes or links for a habitat graph, in order to identify conservation priorities, is evaluated by the use of topological indicators [9]. Recent methods sugest including indices to characterize the of graph nodes (habitat patches), such as Hararry index [14]. Globally, forest habitat loss and fragmentation is one of the main causes that influence the populations of forest species, as they have specific demands for size and quality of their habitat, also low mobility in some cases [18-20]. Many of the forest mammals do not cross easily large open spaces [21]. Nature 2000 is a European Union network of protected areas aimed at improving the conservation status of species and habitats of community importance [22]. In the Nature 2000 network of protected ares, forest habitats are among the most unchanged ones, providing living conditions for different species of mammals [6], many included in Habitats Directive [23] list of species of community importance. More than half of the Natura 2000 sites are covered with forest. It is extremely important for the Natura 2000 network to maintain or improve the conservation status of the forest included in sites of community importance [24]. The Joint Research Center of the European Comission has looked into the protected areas, including forest ones, fragmentation and connectivity [6]. These activities are mentioned in the European Biodiversity Strategy to 2020 and are necessary to reach targets 1, 2 and 3, aimed and reducing and improving species and habitats status in accordance with the EU legislation, improving ecosystems and ecosystem services by creating green infrastructures and restauration of damaged ecosystems, ensuring the sustainability of forest areas through better forest management planning [25]. Management measures for the forest included in Natura 2000 sites have to address the biodiversity conservation problems, taking into consideration the structure and surface of the forest, regeneration capacity, environmental vulnerability [26]. Furthermore, management measures have to provide answers for connectivity issues inside and between the protected areas [6]. In Romania, the development of the Natura 2000 network of protected areas started in 2007 and determined significant changes in the biodiversity conservation policies [27]. Although the Romanian legislation regarding the Nature 2000 network of protected areas is quite comprehensive, implementing it at national level proved a difficult task [24]. One of the most important ideas behind the Nature 2000 project is to ensure the connectivity of the most important areas for biodiversity conservation at national and European level. Our study aims at establishing a scientific method useful to territorial planners and conservation specialists as an instrument of evaluating the connectivity of the protected areas network for groups of species. The objectives of the present study are: a) to assess the functional connectivity of the Romanian Natura 2000 forest sites for terrestrial mammals, b) to determine the importance of individual forest patches in maintaining the landscape connectivity. Finaly, we identify the priority areas for activities of conservation using the general principles from the spatial planning for conservation. 2. Methods 2.1 Study area The study area is the Romanian territory with its five European biogeographical regions: Alpine, Continental, Pannonian, Steppic and the Black Sea [27]. Of these regions, the largest in Romania is the Continental region,

29

30

Mihaita-Iulian Niculae et al. / Procedia Environmental Sciences 32 (2016) 28 – 37

(~56.5% of the total surface of the country, 238,391 km 2). The Alpine region is ~21%, the Steppe and the Black Sea ~16.5% of the total surface of the country, while the Panonian region is the smallest with ~6% [28]. Forest cover 70,760 km2, which is ~29.5% of the Romanian territory [29]. Most of the forest is to be found in the mountain and hilly areas of the Alpine and Continental regions. Agricultural fields cover 55.5% of the country’s surface, while artificial surfaces, semi natural areas, wetlands and water bodies cover 15% of the total, as show the CORINE Land Cover, level I data [29]. 2.2 Materials and methods Forest areas included in the Sites of Community Importance (SCI) of the Natura 2000 network have been highlighted in order to evaluate the functional connectivity. Of the 383 SCIs in Romania, 313 sites protect forest habitats included in Anexe I of the Habitat Directive [23] and are the subject of this study. The surface of the forest included in SCIs is around 32% of the total forest cover in Romania. Approximately 56 % of this surface is occupied by the forest patches we analysed (Fig. 1).

Fig. 1 Distribution the of Sites of Community Interest and the forest sites in Romania

We extracted information about forest distribution from the CORINE Land Cover data set, year 2000 (29). Forest surface is 9.5% of the total surface of the country. The spatial distribution of the SCIs was extracted from the European Union official Nature 2000 network database in vector format (available at http://www.eea.europa.eu/data-and-maps/data/natura-5#tab-gis-data) (30). These areas cover 17.4% of the total surface of the country. The study evaluated the connectivity of the forest patches for land mammals at diffent territorial scales, considering the size of their home range and the species’ dispersion distance. Using these two inputs, mammals were classified in three categories: small mammals (Scenario 1, home range at least 1 ha, dispersion distance maximum 1 km), intermediate mammals (Scenario 2, home range at least 100 ha, dispersion distance 10 km) and large mammals (Scenario 3, home range at least 1000 ha and dispersion distance 100 km). The classification relates to small, intermediate and large body mass (31). For every scenario, the network includes forest surfaces that fulfill the minimum home range criterium as an indicator of dispersion distance (32): 1 ha for small mammals, 100 ha for

Mihaita-Iulian Niculae et al. / Procedia Environmental Sciences 32 (2016) 28 – 37

intermediate mammals and 1000 ha for large mammals (31). Surfaces that do not conform have been excluded for the specific analysis of every category. Connectivity of the forest landscape has been evaluated using indices based on the binary connection model: number of components (NC), number of links (NL) and Harary index (H) [16]. A component represents a connected region, a set of nodes in which a path exists between every pair of nodes of a specific network, or an isolated node [7, 16]. Links represent connections between habitat nodes in network, the higher the values, the more connected the landscape and the better connected the forest patches are the higher value of the H [16]. To quantify landscape connectivity and the importance of forest patches for maintaining landscape connectivity for each scenario, we used the graph theory approach [14, 30] and the Conefor 2.6 with graphical user interface soft [31], available at http://www.conefor.org/coneforsensinode.html) with inputs the pattern of forest patches and dispersal distance. The nodes that constitute the network correspond to the SCIS protected forest patches, with the minimum surface each group of mammals was attributed (1 ha, 100 ha, 1000 ha). For the three landscape scenarios, the resulting number of nodes was 2945 (Scenario 1), 821 nodes (Scenario 2) and 218 for the last scenario. Distance between nodes was calculated from the patches limits (edge to edge) [7, 30, 32]. In order to measure the importance of each node (forest patches) as an indicator for forest connectivity, the Harary index (H) proved very useful [7, 14, 16, 33]. The importance value of the Harary index (dH) for each node was calculated and analysed in the territorial context [33]. Importance of each node for maintaining forest landscape connectivity for a specific I indix was calculated as:

where I is the general value when all nodes in the lanndscape are considered and Iremove is the general value of the index when a specific node is removed [16], folowing the loss of that forest habitat. Harary index is calculated as the sum of the inverse values of the topological distance (shortest path) between every two nodes (or all pairs of vertices) of the graph [7, 14, 34]: where n is the total number of nodes and nlij is the number of links in the shortest path between patches i and j

[16]. If two nodes, corresponding to two forest patches, are located in two different components, then their topological distance is infinity (component is a group of connected patches) [14, 16]. Values of dI can be positive or equal to 0. The higher its value is, the more important that node is for the landscape connectivity [9, 16]. For each of the three scenarious the values of the dH were distributed across five categories, showing the spatial distribution of the nodes that are the most important for maintaining the connectivity. Alongside the five categories, in the case of the intermediate and small mammals, nodes with 0 dH were also highlighted. Each case had been represented in graphics showing the relation between the dH range and the number of patches, also between the dH range and total dA, to explain the importance of each node [33]. dA is not a connectivity index and shows the proportion of a specific node to the total surface [16]. 3. Results and discussion 3.1Functional connectivity of the Romanian Natura 2000 forest sites for terrestrial mammals Analysing the values of binary indices calculated for the landscape in the three scenarios, a very large NC is observed for the small mammals NC, 621, indicating a reduced degree of connectivity, compared with the NC for the intermediate mammals, 91. The connectivity of forest patches is favorable to large mamals, where NC is 1, all nodes being interconnected. The number of links is 3603 for the large mammals, 2359 for intermediate mammals, and 3830 for small mammals, where the number of connected forest patches is higher. As for the Harary index, the highest values are recorded for the small mammals (30742) and large mammals (10925).

31

32

Mihaita-Iulian Niculae et al. / Procedia Environmental Sciences 32 (2016) 28 – 37

An important aspect regarding the connectivity is the surface of the forest patches. The spatial distribution and characteristics of the protected forest (included in SCIs) in each scenario (Table 1) show major differences between the number of patches, but also between the average surface of each node and the number of SCIS that form the network of ptotected areas in each case. Table 1 Spatial analysis of protected forest

Group of mammals

Nr. of forest patches

Small mammals Intermediate mammals Large mammals

2945 821 218

Total surface of protected forest (km2) 22492.86 21994.38 20135.68

Mean area of individual node (km2)

Nr. of SCI for each scenario

7.64 26.79 92.37

313 225 135

Of the total total number of forest patches included in the analysis, ~7% have and individual surface of over 1000 ha and are distributed over 135 SCI (Scenario 3). The total protected area is 20135 km 2. The forest patches of over 100 ha are ~28% (821 forest patches) of the total number of patches, distributed over 225 SCI (Scenario 2). The protected surface is 21994 km2. The largest number of forest patches, 2945, have an individual surface of over 1 ha and are distributed across 313 SCI (Scenario1). The protected surface is 11.5% larger than the surface attributed to the large mammals’ network and 2% larger than that for intermediate mammals. 3.2 Prioritisation of individual forest patches in maintaining the landscape connectivity The importance of the selected index (dH) for each node was analysed, offering information about the role of each forest patch in maintaining forest landscape connectivity. The maps show the spatial distribution of the index for the three scenarios (Small mammals, intermediate mammals and large mammals). For the small mammals group (Scenario 1; d= 1km), the index is between 0.003 and 26.143. The spatial distribution of the Harary index importance value (dH) indicate that the highest values (dH>10) are specific for the forest patches situated in the central part of the country (9 forest patches), where the Alpine biogeographical region is (Fig 2). Three of the forest patches have values between 21.36 and 26.14. Values of 0 for dH are recorded for 314 nodes, that are considered to be disconnected, as the distance towards other nodes is larger than than the 1 km selected threshold. For the intermediate mammals group (Scenario 2; d= 10 km), dH values are between 0.012 si 14.044. Six forest patches have dH values above 10, and are also located in the Alpine biogeographical region (Fig. 3). 33 nodes have 0 for the dH, meaning that are disconnected and the distance towards other nodes is over 10 km. For the large mammals group (Scenario 3; d=100), index values are 0.517 to 2.438. In this scenario also the nodes with the higher values are situated in the Alpine biogeographical region. No 0 values were recorded in this case, as all the nodes are connected (Fig. 4). Graphics were constructed using dH range and number of forest patches (Fig. 5), dH range and dA (Fig. 6). They show the importance of each node for maintaining the connectivity of the forest landscape and also the characteristics of the nodes in relation to the percentage of each individual surface of the total surface. For the network related to the small mammals group, values of dH<4 and 0.04İdH<0.5, the number of the patches is 1409 patches for the first interval, 1158 patches for the second. Total dA is 17.8 for the first interval, 18.7 for the second. For 1İdH<10 and 10İdH<26.14, there are fewer patches, 27 for the first interval, 9 for the second. Total dA values are higher, 26.9 and 27.8 (Fig 5a; Fig 6a).

Mihaita-Iulian Niculae et al. / Procedia Environmental Sciences 32 (2016) 28 – 37

Fig. 2 Harary index value (dH) for each node for small mammals group (1ha, 1km)

Fig. 3 Harary index value (dH) for each node for intermediate mammals group (100ha, 10km)

33

34

Mihaita-Iulian Niculae et al. / Procedia Environmental Sciences 32 (2016) 28 – 37

Fig. 4 Harary index value (dH) for each node for large mammals group (1000ha, 100km)

dH has mostly values between dH< 0.2 si 0.2İdH< 0.5 for the intermediate mammals group, 314 patches for the fisrt interval, 344 patches for the second), total dA being rather the same (15.2 and 15.9). Only six patches are found between 10İdH<14.05, but the total dA is higher, 33.23 (Fig 5b; Fig 6b). As for the large mammals group, in the intervals 0.5
b

c

Fig. 5 dH Ranges vs. Number of patches in the corresponding dH Ranges for a) small; b) intermediate; c) large mammals

a

b

c

Fig. 6 dH Ranges vs. Total dA in the corresponding dH Ranges for a) small; b) intermediate; c) large mammals

Mihaita-Iulian Niculae et al. / Procedia Environmental Sciences 32 (2016) 28 – 37

4. Discussion Following the analysis, the study shows that the protected forest patches are well connected for large mammals (all forest patches are connected, dH>0), while for small and intermediate mammals connectivity is lower, as it is proven by the number of unconnected forest patches (314 nodes for small mammals and 33 nodes for intermediate mammals; dH<0). This is shown also by NC values, higher for small and intermediate mammals (lower connectivity). For the small mammals group’s network, the large number of forest patches, alongside the mean area of an individual node, shows a high degree of fragmentation of the forest patches, compared to the large mammals group. The major difference regarding the number of patches and the mean area of an individual node between the two groups is the total surface of the protected forest (for a large mammal is 11.5% smaller). Loss and fragmentation of forest has an important impact on mammal’s distribution across the landscape. It is also a factor increasing the risk of local extinction in some patches [18]. For these reasons, the conservation efforts should concentrate on reducing the landscape fragmentation. As for the importance of the forest patches in maintaining connectivity, specific values are relevant for each scenario. The higher the dH values, the more important their role in the linkage of this patches with the other patches in proximity is [33]. The three forest patches for the small mammal’s group analysis with the highest dH values (between 21.36 and 26.14) represent the nodes most important for landscape connectivity. These are located in a series of SCIs: Zarandul de Est, Zarandul de Est, Drocea, Defileul Crisului Alb, Muntii Metaliferi, Frumoasa, Gradistea Muncelului-Ciclovina, Defilul Jiului, Nordul Gorjului de Est, Nordul Gorjului de Vest, Domogled Valea Cernei, For the intermediate mammals group, the forest with the highest dH index is found in Defileul Jiului SCI, Nordul Gorjului de Vest SCI, Domogled-Valea Cernei SCI, Platoul Mehedinti SCI. For the large mammals group, the highest dH (2.438671), is found for the forest in Fagaras Mountains SCI, Piatra Craiului SCI, Leaota-Bucegi SCI, Raul Targului-Argesel-Rausor SCI, where the landscape connectivity is better. All these SCIs are located in the Alpine biogeographical region. In the European Union, the efficacy of the Natura 2000 network is evaluated at biogeograhical regions level [27]. The dH index values spatial analysis for the three scenarios show the highest values are recorded for the forest patches (nodes) situated in the Alpine biogeographical region, in the Carpathians mountain range (Meridional Carpathians and Occidental Carpathians). The Alpine biogeographical region is considered a critical region for forest landscape connectivity in SCIs of Romania. For this region, a number of key SCIs for maintaining connectivity were identified. These must be targeted with priority when conservation measures are considered. Comparing dH ranges with the number of and total dA in the corresponding dH ranges, the fragmentation of forest surfaces for small and intermediate mammals was highlighted. In the first scenario, where dH<0.4 and 0.4İ dH<0.5, a high number of patches (1409 for the first interval and 1158 patches for the second) is linked to a dA of less than 20 in each case, which signifies the forest patches are very fragmented. This means these nodes are less important for maintaining landscape connectivity. Where dHı10, there are only 9 patches, and the dA is highest (~27), signaling patches important for maintaining connectivity. The same is the case for the interrmediate mammals. Forest patches included in large mammals forest patches network were a subset of the intermediate mammal’s network, and these were a subset of small mammals forest patches network [30]. The importance of each patch can be correlated with its size, surface playing an important rol for connectivity [33], but in most cases which is more important depends on the matrix representation and dispersal distance [5, 35]. Most of the time, the analysis of the connectivity done by using distance underestimates the role of patches with smaller surfaces [5]. Smaller patches are often critical elements in the network [35] functioning as stepping stone between the large patches, situated farther away [36]. Even if the number of small surface forest patches inside Romanian SCIs is rather large, their rol is essential for maintaining landscape connectivity. Forest patches less than 1000 ha in protected areas cand be used even by large mammals as stepping stones for getting to larger habitats. This is also true for the less than 100 ha forest patches in the case of the intermediate mammals. To improve more the connectivity, when larger patches are impossible to maintain, at least smaller patches can be implemented and protected, to function as stepping stones, increasing the existing landscape connectivity and

35

36

Mihaita-Iulian Niculae et al. / Procedia Environmental Sciences 32 (2016) 28 – 37

communicaton between elements [37]. Implementing Natura 2000 in Romania in a very short time frame, without a proper public consultation, created many negative reactions and adversity [24]. Using the principle of spatial planning for conservation in designing the Natura 2000 network, can generate an optimum network for species conservation, as it was foreseen in the EU Directive. Presently, one of the main objectives of the Natura 2000 network in Romania, the protected areas connectivity, is still not achivied. As the vulnerability of large and small mammals will increase, due to pressure from human activity, connectivity must still be preserved. The landscape graph analysis can be used in various studies to describe quantitatively a landscape as a series of patches interconnected spatialy and functionaly. The Harary index H can be used to cuantify the landscape connectivity statisticaly and ecologicaly [14]. Harary index is showing the conexions between topologically vertices which are close in the graph [14]. Calculating a single index (dH) in order to evaluate how important each node is, constitutes a limitation of the study. Using the Euclidean distance for measuring the accessibility between nodes it is not recommended for terrestrial mammals, as it is not taking into consideration the landscape’s heterogeneity and friction effect [32]. The matrix resistance used in connectivity analysis depends on data availability for species [38] and resistance surfaces based on land cover data. To enhance the relevance of the study, another index must be added, as the Integral index of connectivity (IIC), the probability of connectivity (PC) index and the dPCconnector fraction [7], that can better serve to assess the connectivity [17]. 5. Conclusions The methods used allowed the evaluation of the forest in protected areas and the identification of the most important patches for landscape connectivity relevant for mammals. Results shown that critical areas for connectivity must be addressed with priority with adequate management measures. Also, new surfaces must be identified in order to be included in the network of protected areas and new habitats created, even smaller ones to function as stepping stones. Future studies must also look into the legal context, costs and other aspects. As landscape connectivity is extremely important for mammals. This study shown that there is at least one aspect that the Natura 2000 network failed to address, in the case of forest sites, connectivity for mammals, especially for the low dispersion ones, being low. Acknowledgements We would like to thank the two anonymous reviewers for their comments which will hopefully improve our paper significantly. References 1. Burel, F. and J. Baudry, Habitat quality and connectivity in agricultural landscape: the role of land use systems at various scale in time. Ecological Indicators, 2005. 5(4): p. 305-313. 2. Pascual-Hortal, L. and S. Saura, Integrating landscape connectivity in broad-scale forest planning through a new graph-based habitat availability methodology: application to capercaillie (Tetrao urogallus) in Catalonia (NE Spain). European Journal of Forest Research, 2008. 127: p. 23-31. 3. Pascual-Hortal, L. and S. Saura, Comparison and development of new graph-based landscape connectivity indices: towards the priorization of habitat patches and corridors for conservation. Landscape Ecology, 2006. 21: p. 959-967. 4. Raison, R.J., A.G. Brown, and D.W. Flinn, Criteria and indicators for sustainable forest management. 2001, Wallingford: CAB Publishers. 5. Laita, A., J.S. Kotiaho, and M. Mönkkönen, Graph- theoretic connectivity measures: what do they tell usabout connectivity? Landscape ecology, 2011. 26: p. 951-967. 6. Estreguil, C., G. Caudullo, and J.S. Miguel, Connectivity of Natura 2000 Forest Sites. Executive Report. 2013, European Commission. Joint Research Centre. Institute for Environment and Sustainability: Luxembourg. 7. Saura, S. and L. Pascual- Hortal, A new habitat availability index to integrate connectivity in landscape conservation:comparison with existing indices and application to a case study. Landscape and Urban Planning, 2007. 83(2-3): p. 91-103. 8. Haddad, N.M., et al., Corridor use by diverse taxa. Ecology, 2003. 84(3): p. 609-615. 9. Baranyi, G., et al., Contribution of habitat patches to network connectivity: Redundancy and uniqueness of topological indices. Ecological Indicators, 2011. 11: p. 1301- 1310. 10. Tischendorf, L. and L. Fahrig, How should we measure landscape connectivity? Landscape Ecology 2000. 15: p. 633–641. 11. Taylor, P.D., et al., Connectivity is a vital element of landscape structure. Oikos, 1993. 68(3): p. 571-573.

Mihaita-Iulian Niculae et al. / Procedia Environmental Sciences 32 (2016) 28 – 37

12. Keitt, T.H., D.L. Urban, and B.T. Milne Detecting critical scales in fragmented landscapes. 1997. 1. 13. Jordan, F., et al., Characterizing the importance of habitat patches and corridors in maintaining the landscape connectivity of a Pholidoptera transsylvanica (Orthoptera) metapopulation. Landscape Ecology, 2003. 18: p. 83-92. 14. Ricotta, C., et al., Quantifying the network connectivity of landscape mosaics: a graph- theoretical approach. Community Ecology, 2000. 1(1): p. 89-94. 15. Urban, D. and T. Keitt, Landscape connectivity: a graphtheoretic perspective. Ecology, 2001. 82: p. 1205–1218. 16. Saura, S. and L. Pascual-Hortal, Conefor Sensinode 2.2 User’s Manual: Software for quantifying the importance of habitat patches for maintaining landscape connectivity through graphs and habitat availability indices. 2007, University of Lleida: Spain. 17. Saura, S. and L. Rubio, A common currency for the different ways in which patches and links can contribute to habitat availability and connectivity in the landscape. Ecography, 2010. 33: p. 523-527. 18. McAlpine, C.A., et al., The importance of forest area and configuration relative to local habitat factors for conserving forest mammals: A case study of koalas in Queensland, Australia. Biological Conservation, 2006. 132: p. 153-165. 19. Rochelle, J.A., L.A. Lehmann, and J.B. Wisniewski, Leiden, , eds. Forest Fragmentation: Wildlife and Management Implications. 1999, Brill: Leiden, The Netherlands. 20. Lindenmayer, D.B., et al., The conservation of arboreal marsupials in the montane ash forests of the Central Highlands of Victoria, southeast Australia: I. Factors influencing the occupancy of trees with hollows. Biological Conservation 1990. 54: p. 111–132. 21. Bakker, V.J. and D.H. Van Vuren, Gap-crossing decisions by thered squirrel, a forest-dependent small mammal. Conservation Biology 2004. 18: p. 689-697. 22. Pullin, A.S., et al., Conservation Focus on Europe: Major Conservation Policy Issues That Need to Be Informed by Conservation Sc ience. Conservation Biology, 2009. 23(4): p. 818-824. 23. C, E., Council Directive 92 /43 /EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora . 1992, Official Journal of the European Communities: Official Journal of the European Communities. 24. Merce, O., Natura 2000 and the forests in Romania – management principles, problems and threats. Journal of Horticulture, Forestry and Biotechnology, 2012. 16: p. 139- 144. 25. C, E., COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS.Our life insurance, our natural capital: an EU biodiversity strategy to 2020. 2011, European Commission: Brussels. 26. Velazquez, J., et al., Environmental diagnosis: Integrating biodiversity conservation in management of Natura 2000 forest spaces. Journal for Nature Conservation, 2010. 18(4): p. 309–317. 27. Ioja, I.C., et al., The efficacy of Romania’s protected areas network in conserving biodiversity. Biological Conservation, 2010. 143(11): p. 2468-2476. 28. EEA, Biogeographical Regions. European Environment Agency (EEA). 2015, European Topic Centre on Biological Diversity (ETC/BD). 29. EEA, Corine Land Cover 2000 seamless vector data- Version 17 (12/2013). 2014, European Environment Agency (EEA). The European Topic Centre on Spatial Information and Analysis. 30. Minor, E.S. and T.R. Lookingbill, A multiscale network analysis of protected –area connectivity for mammals in the United States. Conservation Biology, 2010. 24(6): p. 1549- 1558. 31. Saura, S. and J. Torné, Conefor Sensinode 2.2: a software package for quantifying the importance of habitat patches for landscape connectivity. Environmental Modelling & Software, 2009. 24: p. 135-139. 32. Gurrutxaga, M., L. Rubio, and S. Saura, Key conectors in protected forest area networks and impact of highways: A transnational case study from the Cantabrian Range to the Western Alps. Landscape and Urban Planning, 2011. 101: p. 310- 320. 33. Devi, B.S.S., et al., Forest patch connectivity diagnostics and prioritization using graph theory. Ecological modelling, 2013. 251: p. 279287. 34. Li, X.-X. and Y.-Z. Fan, The connectivity and the HararyIndex of a graph. Discrete Applied Mathematics, 2015. 181: p. 167-173. 35. Ziółkowska, E., et al., Effects of different matrix representations and connectivity measures on habitat network assessments. Landscape Ecology, 2014. 29(9): p. 1551-1570. 36. Rubio, L. and S. Saura, Assessing the importance of individual habitat patches as irreplaceable connecting elements: an analysis of simulated and real landscape data. Ecological Complexity, 2012. 1: p. 28–37. 37. Laita, A., M. Monkkonen, and J.S. Kotiaho, Woodland key habitats evaluated as part of a functional reserve network. Biological Conservation, 2010. 143: p. 1212–1227. 38. Zeller, K.A., K. McGarrigal, and A.R. Whiteley, Estimating landscape resistance to movement:a review. Landscape Ecology, 2012. 27: p. 777-797.

37