Abstract The Natura 2000 network forms the cornerstone of the biodiversity conservation strategy of the European Union and is the largest coordinated network of protected areas (PAs) in the world. Here, we demonstrated that the network fails to adequately cover the marine environment and meet the conservation target of 10% set by the Convention on Biological Diversity. The relative percentage of marine surface cover varies significantly among member states. Interestingly, the relative cover of protected seascape was significantly lower for member states with larger exclusive economic zones. Our analyses demonstrated that the vast majority (93%) of the Natura 2000 sites that cover marine waters include both a terrestrial and a marine component. As a result, the majority of the protected surfaces is adjacent to the coastline, and decreases offshore; only 20% of Natura marine PAs is at depths >200 m. The lack of systematic planning processes is further reflected by the great variability in the distances among protected sites and the limited number of shared Natura sites among member states. Moreover, <40% of the marine sites have management plans, indicating the absence of active, or limited management in most sites. This work highlights the gaps in coverage and spatial design of the European conservation network in the marine environment, and raises questions on the unevenly treatment of marine vs. terrestrial areas. Introduction Protected areas (PAs) form the cornerstone of biodiversity conservation. The new Strategic Plan for Biodiversity 2011–2020 of the Convention on Biological Diversity promotes the expansion of the global coverage of PAs, through Aichi Target 11, to reach at least 17% of land and 10% of marine waters by 2020. In human dominated regions, the establishment of large PAs is challenging due to multiple human uses of land and sea. In such crowded environments, networks of smaller multi-purpose PAs are a feasible alternative for biodiversity conservation and the achievement of Aichi Targets (Gaines et al., 2010; Grorud-Colvert et al., 2014; PISCO and UNS, 2016). To protect its biodiversity, the European Union (EU) has created the Natura 2000 network of PAs (Evans, 2012). The Natura 2000 network is not a system of strict nature reserves from which all human activities are excluded (Tsiafouli et al., 2013). Although it includes strictly protected nature reserves, in most of its extent extractive activities are allowed but regulated (Iojă et al., 2010; Evans, 2012). It is one of the world’s most extensive networks of conservation areas that currently counts more than 27 200 sites (EEA, 2012; EC, 2016). Its major advantage over other multi-national PA networks is that it was designed based on two EU legislative instruments, the Habitats Directive (EC, 1992) and the Birds Directive (EC, 2009). Thus, under the umbrella of these two directives all EU member states follow the same rules for the inclusion of a site within the network and have the same obligations regarding the conservation of priority species and habitats. Currently, the Natura 2000 network covers >18% of the terrestrial EU territory (EC, 2016). The effectiveness of PA networks largely depends on their spatial properties. The size of individual sites, their locations and spacing are critical parameters which apart from structural elements of a network reflect its potential effectiveness. The size of individual sites determines whether the area of the enclosed habitats could ensure viable populations. The spacing among the sites of the network is critical as it allows or not the dispersal and connectivity of species populations. These structural elements (i.e. size and spacing) should be based on estimates of dispersal and connectivity to ensure population persistence. The actual location of network sites could ensure representativeness of various biodiversity features of concern. In theory, the Natura 2000 network is composed of sites that were defined based on the same principles and, thus, should ensure a high representativeness and cover the minimum area required for species persistence. However, the spatial properties (e.g. connectivity patterns) were not set as a direct criterion for the design of the network (Opermanis et al., 2012). In practice, Natura 2000 sites were largely selected on an individual basis, and issues such as dispersal and movements of recruits and adult stages of populations were not considered (Johnson et al., 2008; Mazaris et al., 2013). This gap in the consideration of the spatial structure and configuration of the network downgrades the ‘ecological coherence’, which is one of the main objectives of the Natura 2000 network (Articles 3 and 10 of Habitats Directive). Continuity and connectivity among the Natura 2000 sites is necessary both within countries as well as across countries (Johnson et al., 2008; Lagabrielle et al., 2014). Conservation outcomes can be maximized when countries that share natural environments plan spatially explicit actions in collaboration (Opermanis et al., 2012; Mazaris et al., 2013; Mazor et al., 2013; Opermanis et al., 2013). Whereas the efficacy and spatial properties of the Natura 2000 network has been investigated on land (Opermanis et al., 2012; Maiorano et al., 2015), this has not been the case for the marine part of the network. Consequently, major gaps exist in the research on the marine part of the Natura 2000 network (Orlikowska et al., 2016), including the relative coverage of marine habitats and species (Giakoumi et al., 2011, 2012; Trochet and Schmeller, 2013). The European marine environment (Mediterranean Sea, Baltic Sea, Black Sea, North Sea, northeast Atlantic Ocean) is subjected to multiple anthropogenic pressures (Benn et al., 2010; Coll et al., 2012; Korpinen et al., 2012; Micheli et al., 2013; Giakoumi et al., 2015) that modify native community compositions and impact ecosystem functionality and services (Claudet and Fraschetti, 2010; Katsanevakis et al., 2015). Therefore, it is important to assess whether one of the most extensive conservation networks of the world, i.e. the Natura 2000 network, provides an adequate coverage of the marine environment and safeguards the sustainability of European marine ecosystems and their biodiversity (Maiorano et al., 2007; Sundblad et al., 2011; Qiu and Jones, 2013). An additional challenge related to the spatial properties of marine sites of the network is to assess whether site selection, often driven by conservation focus, is tightly connected to terrestrial planning and management (Agardy, 1994; Jones, 2001). In the past, many marine sites have been subjected to protection as small extensions of the boundaries of terrestrial-protected sites. For example, the inclusion of a bay within a protected site could offer a simple detectable boundary represented by a direct line between the headlands; Gubbay, 1995). Over the last decades, a great progress towards systematic marine conservation planning and prioritization has been made (Leslie et al., 2003; Qiu and Jones, 2013). Nevertheless, many MPAs represent an extension of terrestrial PAs (i.e. a marine buffer zone surrounding a protected land; Day et al., 2012), cover only a limited area of marine habitat or have various use zones (Wood et al., 2008; Costello and Ballantine, 2015). In the Natura 2000 network, sites which have a marine component covering >5% of their total area are listed as marine sites (EEA, 2017), despite often being designated based solely on terrestrial ecological features (Giakoumi et al., 2011). Therefore, it remains to assess whether sites designated based on such criteria can provide benefits for marine biodiversity conservation. Given that the basic objective of the Natura 2000 network is to ensure the coherence of PAs, we aimed to estimate the extent of the marine part of the network and identify its spatial properties related to the location of sites and their structural connectivity. This study aims to investigate whether: (i) there are any spatial biases in the location of the protected sites, (ii) the Natura 2000 network was biased in terms of habitat representativeness, and (iii) the current structure and configuration of the network ensures cross-boundary coherence. Methods The spatial database on the distribution of the European PAs composing the Natura 2000 network was derived by the European Environment Agency (EEA; http://www.eea.europa.eu/data-and-maps/data/natura-7; accessed 8 October 2016). As the Natura 2000 sites were not clearly characterized as marine, coastal, or terrestrial, we initially selected all sites containing marine areas. To identify these sites, we used the digital terrestrial terrain of Europe, also derived by EEA (http://www.eea.europa.eu/data-and-maps/data/eea-coastline-for-analysis; accessed 8 October 2016). We preferred to use datasets provided by EEA in both cases to ensure the maximum level of compatibility between the different spatial datasets. A bathymetry map with 30 arc-second intervals was derived by GEBCO-General Bathymetric Chart of the Oceans (http://www.gebco.net/about_us/contact_us/; accessed 4 November 2016). Natura 2000 sites were designed with the objective to protect species and habitats listed in two different directives (i.e. the Birds Directive and the Habitats Directive). As a result the surface of many Natura 2000 sites defined for the protection of birds may overlap with sites independently defined to protect habitats. Here, to avoid any overestimation of protected surface, we dissolved the boundaries of the sites to generate contiguous polygons of protected marine surface. Then, the area of the protected marine environment by the Natura 2000 sites was calculated. By splitting this information within the Exclusive Economic Zone (EEZ) and the territorial waters (i.e. extending at most 12 nautical miles—6 nautical miles in the case of Greece) of each member state (derived by: http://marineregions.org; accessed 20 October 2016), we calculated the proportion of national maritime area under protection. As not all member state’s EEZs have been ratified, the median line was used in the case of non-agreed marine borders. Cumulative curves of PAs vs. marine area in relation to bathymetry and distance from the coastline were created to assess how well offshore and deep marine areas are represented in the Natura 2000 network. These curves depict the differences between the distribution of PAs and the distribution of marine area across the range of bathymetric zones and distances from the coast. In a network where PAs would be uniformly distributed across the bathymetric range and across various distances from the coast, the two cumulative curves (i.e. of PAs and marine area) would share a similar pattern; the greater their distance, the higher the deviation from uniformity. Mann-Whiney test was employed to investigate the potential difference in the size of the Natura 2000 sites that are mainly marine (>90% of their total area) over the sites of the network that are predominantly terrestrial sites expanded to the sea (<10% of their total area marine). The analysis was repeated by gradually considering the 20 and 30% of extremes (i.e., >80 vs. <20% marine surface of their total area; >70 vs. <30% marine surface of their total area). We used the Spearman correlation coefficient to test for any significant relationship between the marine surface protected by the network and (i) the total surface of EEZ of each member state, (ii) the number of Natura 2000 sites expanded to the sea, and (iii) the relative cover of protected seascape within each member state. A non-parametric Kruskal-Wallis test was employed to explore for any significant difference among sizes of national marine Natura 2000 sites. The mean size of PAs per country was correlated with their EEZs by using the Spearman correlation coefficient to examine whether the selection of the size was driven by the extent of national marine waters. We further detected all polygons of protected marine surface that are located within the maritime boundaries of two (or more) countries. To provide some information on the spatial continuity between protected surfaces that extend among different countries, we quantified the proportion of marine surface of the Natura 2000 sites across borders, by generating a Cross-border Continuity Index (CCI). In the case a site was shared by two countries we used the quantity: CCI=Areamin12Areatotal, whereas in the case that a site was shared by three countries the metric was calculated as CCI= Areamin13Areatotal, where Areamin is the area of the smallest part of the cross-boundary site and Areatotal is the area of the entire site across the countries. The CCI maximizes to 1 whenever the area of a site is equally divided among the countries. Conversely, the index takes the value of zero (0) whenever a protected surface of a given country is adjacent to the border but does not continue into the neighbouring country. We provided a preliminary assessment of structural links between the sites to further evaluate the spatial distribution of marine areas protected under the scheme of the Natura 2000 network. After merging sites that overlap, we defined the centroid of each marine surface and calculated the nearest distance to the next site within buffer zones of various distances. In an attempt to make a rough estimate of the protection status in each Natura 2000 site, we processed data on the existence of a management plan, provided by the EEA. The lack of information, at an EU level, on the regulations applying in each Natura 2000 site did not allow us to explore the sites’ protection level (i.e. full protection vs. partial protection). Considering the presence of a management plan as an indicator positively correlated to the enforcement of protection, we compared (t-test) the size and % coverage of the marine component of Natura 2000 sites between sites that had and sites that did not have a management plan. Acknowledging that the existence of a management plan by itself does not guarantee the application of conservation measures (e.g. management plans might mainly target the terrestrial component of the site, or the management plans may not be enforced), we examined other datasets provided by the EEA that could provide an indication of whether the marine component of Natura 2000 sites were protected in practice. We considered that information on the relative pressures assigned to each site as reported by experts conducting the National reports on the Natura 2000 network, could be used towards that end. We assumed that if no pressures to marine ecosystems were reported at a given site, this might be indicative of the absence of actual management in the marine environment, as the identification and assessment of pressures would have been a first step for taking management actions. For each marine site of the network, we extracted information on the assigned pressures from human activities at sea, as reported by the EEA. We calculated the total number of assigned pressures/activities and then selected only those assigned under the categories: (i) Marine and Freshwater Aquaculture, (ii) Fishing and harvesting aquatic resources, (iii) Illegal taking/removal of marine fauna. Sites which are properly managed should have had a management plan and at least recognized some of these marine-related activities as pressures to the marine component of the site. Results The Natura 2000 network covers about 11.7% of the territorial waters of EU member states, but <6% of their EEZs. Among the 27 295 Natura 2000 sites, ∼15% (n = 3957) cover either both terrestrial and marine or only marine surfaces (Figure 1). About 40% of these sites overlap with at least one other Natura 2000 site. Figure 1. View largeDownload slide Sites of the European conservation network Natura 2000 covering marine surfaces. Gradual shading illustrates the bathymetric range in the marine realm (0–5000 m). Figure 1. View largeDownload slide Sites of the European conservation network Natura 2000 covering marine surfaces. Gradual shading illustrates the bathymetric range in the marine realm (0–5000 m). The vast majority (∼90%) of the Natura 2000 sites that include marine area are a combination of terrestrial and marine sites. Approximately 31% of these cross-environment sites (n = 1122), have a marine area that covers >75% of the total site area, while for ∼34% (n = 1186) <10% of their surface is marine (Figure 2). We detected a strong bias towards very small marine surfaces linked to terrestrial areas protected within the Natura 2000 network. Figure 2. View largeDownload slide Sites of the Natura 2000 European network of PAs that are expanded to the marine environment. The proportional cover of marine vs. the total surface area of each site is presented. Figure 2. View largeDownload slide Sites of the Natura 2000 European network of PAs that are expanded to the marine environment. The proportional cover of marine vs. the total surface area of each site is presented. The marine part of sites with >90% of their surface being marine, i.e. mainly marine sites, was significantly larger (mean area 343 km2, ranging from 26 to 416 km2 95% CI) than that of the sites whose marine part was <10% of the total area, i.e. predominantly terrestrial sites expanded to the sea (mean area =57 km2, ranging from 42 to 70 km2 95% CI km2) (p < 0.01). Similarly, the total size of the marine part of sites with 80 and 70% of their surface being marine was significantly larger than the marine part of those covering <20 and 30% of marine areas (298 vs. 51 and 269 vs. 52 km2, respectively). The curve of the cumulative area of Natura 2000 sites with the distance from the coastline increased rapidly up to about 40 km from the coastline, but beyond that distance it increased at a much lower rate (Figure 3a). Conversely, the cumulative curve of marine area with the distance from the coastline was much closer to the diagonal (corresponding to a uniform distribution). This spatial bias was even more intense in relation to bathymetry. The great majority of Natura 2000 sites were in shallow waters (<200 m). Despite that >57% of the EEZ marine area spans depths >200 m, only 20% of Natura PA is located there (Figure 3b). Figure 3. View largeDownload slide Cumulative curves of the total and Natura 2000 marine surface with respect to distance from the coastline (a), and depth (b). Figure 3. View largeDownload slide Cumulative curves of the total and Natura 2000 marine surface with respect to distance from the coastline (a), and depth (b). At a national level, we found that most EU coastal countries (15/23) protected <10% of their EEZ (Figure 4). The number of Natura 2000 sites that mainly covered marine surface was rather limited in comparison to sites covering both marine and terrestrial surface (Figure 4). In Greece, <12% of Natura 2000 sites with a marine component (n = 228) protected mainly marine surface (>75% of their total area), whereas in Croatia the majority of the sites (69%) with a marine component (n = 319) actually focused on the seascape rather than the landscape. Overall, no significant association was observed between the relative cover of these sites (% of marine vs. total surface) and surface of EEZ or total marine area protected by the Natura 2000 network (in both cases p > 0.05). The overall marine surface of Natura 2000 sites within a country's jurisdiction increased as the total surface of EEZ increased (rs = 0.526, p < 0.01). Likewise, the number of protected sites increased with respect to the size of the country’s EEZ (rs = 0.372, p < 0.01). Still, the relative cover of protected seascape was significantly lower for the member states with larger EEZs (rs = −0.31, p < 0.01). For example, for countries with long coastlines and expanded EEZs, traditionally considered as maritime states such as Greece, Portugal, Italy, Malta, and Cyprus, the protected marine areas were <5% of their (provisional) EEZs. In contrast, for countries with rather narrow coastline such as Belgium and Germany, the coverage of marine surfaces under protection was >35%. Figure 4. View largeDownload slide Proportion of the Natura 2000 sites that are expanded to the sea, and their relative cover of marine surface (≤25, 25–50, 50–75, ≥75%). Their coverage in the Exclusive Economic Zones is highlighted by red lines. Country codes as follows: BE, Belgium; BG, Bulgaria; CY, Cyprus; DE, Germany; DK, Denmark; EE, Estonia; ES, Spain; FI, Finland; FR, France; GR, Greece; HR, Croatia; IE, Ireland; IT, Italy; LT, Lithuania; LV, Latvia; MT, Malta; NL, Netherlands; PL, Poland; PT, Portugal; RO, Romania; SE, Sweden; SI, Slovenia; UK, United Kingdom. Figure 4. View largeDownload slide Proportion of the Natura 2000 sites that are expanded to the sea, and their relative cover of marine surface (≤25, 25–50, 50–75, ≥75%). Their coverage in the Exclusive Economic Zones is highlighted by red lines. Country codes as follows: BE, Belgium; BG, Bulgaria; CY, Cyprus; DE, Germany; DK, Denmark; EE, Estonia; ES, Spain; FI, Finland; FR, France; GR, Greece; HR, Croatia; IE, Ireland; IT, Italy; LT, Lithuania; LV, Latvia; MT, Malta; NL, Netherlands; PL, Poland; PT, Portugal; RO, Romania; SE, Sweden; SI, Slovenia; UK, United Kingdom. We found a statistically significant difference in the mean size of the marine Natura 2000 sites among member states (Kruskal-Wallis X2 = 110.1, p < 0.01) (Supplementary Figure S1). The mean size of the marine Natura 2000 surfaces was not related to the size of EEZs (p > 0.05), and demonstrated no geographical pattern for the countries with lower, intermediate or large mean sizes. Only 17 PAs covering marine habitats crossed political borders (Supplementary Table S1). For these sites, the values of the index which demonstrates the evenness of surface protection across borders ranged from 0.03 to 0.99. The majority of these cross-boundary sites are near the coastline and only a small surface is shared among countries. Most commonly, Natura 2000 sites fall within the territorial waters of a single country. Examples of good connectedness were found in northern EU countries (e.g. Germany and Denmark, Germany and Poland, Latvia and Estonia). Still, in terms of both spatial properties and extent, the Dogger Bank submerged sandbank represents a PA shared by the United Kingdom, Germany, and the Netherlands under three different national Natura 2000 sites which intersect, resulting in one of the largest MPAs in Europe. In addition, we found that in 28 cases a protected surface of a given country was adjacent to the border but did not continue to the neighbouring country (CCI = 0; Supplementary Table S2). A great variability was observed in the distribution of distances between marine Natura 2000 sites (Figure 5). The mean distance between the nearest sites was rather short (mean = 13.7 ± 18 km), and significantly increased as the farther neighbours were considered. It was not however only the mean distance that increased, but the deviation of the estimated distances around the mean was greatly extended. This demonstrates that even if for some of the PAs neighbouring sites exist in close distances, there are also sites in the network with neighbouring sites located far away. This result illustrates that the spatial structure of the network did not follow any systematic pattern. Figure 5. View largeDownload slide Boxplot on the distribution of distances between pairs of marine areas, covered by Natura 2000 Network. The first box represents the distances of all sites to nearest neighbour, the second plot the distance to the second closer neighbour and so on. Figure 5. View largeDownload slide Boxplot on the distribution of distances between pairs of marine areas, covered by Natura 2000 Network. The first box represents the distances of all sites to nearest neighbour, the second plot the distance to the second closer neighbour and so on. Our analyses revealed that <40% (n = 1579) of the 3957 marine Natura 2000 sites had a management plan. The marine sites with a management plan were significantly smaller and covered significantly smaller marine component than those without a management plan (in both cases p < 0.01). In total, only 515 sites with a marine component covering >75% of the overall area of the site had a management plan. The vast majority (>65%) of the sites located in the Baltic Sea (in Sweden, Estonia, and Denmark) had management plans, followed by sites located in Spain and Italy (∼50% of the national sites), and then by sites in Belgium, Finland, Latvia, and Germany (30–40% of the national sites). For all remaining member states <15% of the marine-protected sites of the Natura 2000 network had formally adopted a management plan. Only in three countries (i.e. Germany, Belgium and Estonia) PAs with management plan covered more than one-fourth of the EEZs, while in the remaining cases this percentage dropped to <10%. Sites with management plan covered mainly shallow waters, with 50% of the total protected surface found at depths <20 m. Similarly, <5% of the European seas deeper than 200 m were actually protected by sites owning a management plan. For 2547 marine sites, no information on any type of pressure or activity related to fisheries and aquatic pressures was reported, with almost half of them (n = 1204) being reported to have a management plan. Discussion Our results demonstrate that the European conservation network, Natura 2000, covers <6% of the European marine environment, corresponding mainly to coastal and shallow waters. For some countries (e.g. Greece, Cyprus, Italy, Malta), the low coverage of PAs can be partly justified by the continuing disputes over marine boundaries that has impeded the declaration of their EEZs (Katsanevakis et al., 2015), and thus Natura sites are restricted in territorial waters. Moreover, our findings reveal the lack of standardized properties of structural connectivity in the network as well as the existence of very few cross-boundary conservation areas across the EU marine realm. These important issues need special attention in order to increase the effectiveness of the Natura 2000 network in protecting marine biodiversity and achieving international conservation targets. The majority of the Natura 2000 sites covering partially (or fully) marine environments are either extensions of terrestrial sites into the sea or cross-environment sites whose coverage is highly biased towards land. This bias towards terrestrial protection indicates that the site selection of marine Natura 2000 sites was seldom driven by pure marine conservation needs. Rather, terrestrial sites, established for the protection of terrestrial species and ecosystems, were extended to adjacent marine environments often without examining whether marine species and ecosystems would be effectively protected within these sites (Giakoumi et al., 2012). For instance, in Greece, the basic criterion for extending terrestrial Natura 2000 sites into the sea was the presence of the endemic seagrass Posidonia oceanica, most of the time without accounting for the presence, density, or biomass of other marine species and habitats in the area. We acknowledge that securing continuity in protection of both terrestrial and marine environments can facilitate the prioritization of conservation actions across both land and sea (Klein et al., 2012) offering protection for several marine species, e.g. sea turtles (Mazaris et al., 2014; Almpanidou et al., 2016). However, under the current structure of the network most threatened marine species are not represented and adequately protected (Giakoumi et al., 2011, 2012; Trochet and Schmeller, 2013). Considering these findings, a re-assessment of the Natura 2000 marine sites would significantly improve the representation of the marine biodiversity within the framework. Given the various gaps in the network design identified herein, we suggest that such an initiative should not be limited to an extension of current sites or selection of few new sites based on expert opinions, but systematic planning approaches should be applied (e.g. Giakoumi et al. 2011). The lack of representativeness of marine species and habitats is even more pronounced for deep-sea marine features (Olsen et al., 2013). Deep-sea offshore areas host a unique biodiversity which is associated with key ecological and biogeochemical processes. Scientific evidence demonstrates that the conservation of deep-sea biodiversity is a priority for the sustainable functioning of the worlds' oceans (Danovaro et al., 2008). Studies in the European seas have demonstrated that many priority areas for marine biodiversity conservation are offshore and urgently need protection (e.g. Micheli et al., 2013). Although surveillance and enforcement of protection offshore is challenging, cutting edge technology and control systems can provide solutions (Katsanevakis et al., 2015; Hays et al., 2016) for the protection of new Natura 2000 deep-sea and offshore sites. Beyond the representation imbalance between coastal and offshore areas within the present Natura 2000 network, we revealed the great variability in the properties, extent, and spatial coverage of the marine environment by Natura 2000 sites across member states. Notably, the coverage of the network in the Mediterranean Sea is extremely low. One could assume that this pattern could be driven by historical and political reasons (e.g. time since entering in the EU, or initiation time of designating Natura 2000 sites). However, historical and political reasons do not seem to influence the terrestrial coverage among countries (EC, 2016). Therefore, such factors are unlikely to justify the uneven coverage patterns of the marine Nature 2000 network across the EU marine realm. Potentially, differences between national networks reflect inconsistencies in the implementation of EU directives and strategic plans (e.g. the Marine Strategy Framework Directive 2008/56/EC) among member states (Douvere and Ehler, 2009) but also increased territorial conflicts in some regions, such as the Mediterranean Sea (Katsanevakis et al., 2015). Resources should be invested and collaboration among EU countries should be promoted for the expansion of the network in large EEZs, especially in areas with high importance for marine biodiversity (Micheli et al., 2013). By definition, PA networks should ensure the connectivity among protected sites. The importance of connectivity, which is one of the fundamental principles of systematic conservation planning (Moilanen et al., 2009), becomes even more important in the marine environment, where processes are widely spread and some species have particularly large home ranges (Palumbi, 2003). Moreover, in a well-established network the optimal size and shape of individual sites must be considered along with the spacing among them, throughout the seascape, to ensure connectivity for the populations of various species (Moffitt et al., 2011; Green et al., 2014). However, issues related to spatial configuration and connectivity have generally been ignored during the design phase of the Natura 2000 network (Olsen et al., 2013). Our results reinforce this evidence by demonstrating the lack of important spatial properties for the marine part of the Natura 2000 network. Looking at the close vicinity of each PA, we could conclude that the marine sites of the Natura 2000 network are well connected. However, this first look does not reflect the real structure of the network (Supplementary Figure S2) in terms of its population connectivity potential. We found that connectivity patterns are rather biased because when we move to the next closer site no standardized pattern is obtained. This finding suggests that at broader scales the highly connected areas appear as isolated clusters. The efficiency of MPAs should be assessed by investigating the spacing among protected sites at the scale of marine neighbourhoods (i.e. scale of population spread) (Palumbi, 2004). A linear configuration, such as the one composed of coastal sites may not be adequate for ensuring a high level of connectivity. In addition, even though it is not feasible to define the ideal size applicable to all species and sites (McLeod et al., 2009), the dominance of very small marine surfaces in the Natura 2000 network highlights the need for defining minimum size requirements. Currently, the protected surface may cover the needs of some species at the local scale, but it is probably insufficient for other species with larger area requirements. Conversely, the distance of some sites might be adequate to facilitate movement and connectivity for some large, highly mobile animals but such distances could pose a critical barrier for short-distance dispersers. Thus, besides the proximity between sites, other features of the network should be taken into consideration, i.e. spatial structure, the relationship between the size of sites and the spacing among them, interconnectedness, and multi-step connections among sites. For example, MPAs in the Mediterranean exhibit low connectivity while a high number of clusters was identified with few number of paths connecting distant sites (Andrello et al., 2013). Therefore, the assessment of the spatial properties of the network should be addressed at various spatial scales and favour the selection of new Natura 2000 sites that will increase the overall connectivity of the network. Moreover, to achieve efficient conservation outcomes, the continuity and linkages among Natura 2000 sites should be promoted by enhancing the collaborative design of sites in neighbouring countries. The lack of international cooperation represents one of the main obstacles that jeopardize conservation targets and efficiency (Rands et al., 2010; Ramos et al., 2015). This should not be the case for the Natura 2000 network, which is a multi-national initiative based on common directives applied in all EU member states. The Habitats and Birds directives and other legislative tools standardize the designation of Natura 2000 sites, monitoring processes, and financial mechanisms (e.g. LIFE projects; http://ec.europa.eu/environment/life/) leading to the harmonization of conservation efforts across countries. Although many terrestrial areas in Europe are continuously protected across national borders (Opermanis et al., 2012), very few sites extend from the territorial waters of one country to the other. Globally, many challenges have been associated to the development and efficient operation of transnational MPAs (Mackelworth, 2016). At the same time, many conservation benefits associated to transboundary collaboration have been identified (Mackelworth, 2012; Mazor et al., 2013). The common procedures, rules and financial instruments that exist within the EU should encourage the designation of transnational marine Natura 2000 sites whenever this is a requirement for increasing the efficiency of marine biodiversity conservation. A thorough understanding of the implications and potential benefits of a PA network, such as the Natura 2000, for marine biodiversity conservation, requires delving into the differences in regulations among individual sites. Critical questions that need to be addressed include the number and properties (e.g. spatial) of sites that are fully, strongly or weakly protected, and the type of activities regulated in those sites (e.g. fishing, boating, recreational diving) (Lubchenco and Grorud-Colvert, 2015). However, performing such an analysis at broad scales is often limited by data availability. Here, we roughly assessed the level of conservation effort, by processing data on the existence of management plans and the officially reported data on marine-related pressures and activities. This assessment rather revealed serious issues. The fact that pressures on the marine component of the sites are often overlooked could be attributed to the conservation focus of the Natura 2000 network, which seems to be mainly terrestrial. The apparent lack of management for the majority of marine sites included in the network raises serious concerns regarding the effectiveness of the marine part of the Natura 2000 European network in conserving marine biodiversity. In conclusion, the establishment of PA networks represents a critical step towards conserving ecological mechanisms and ecosystem functions, yet, focusing solely on coverage targets (e.g. protecting 10% of marine waters by 2020) sets up a poor conservation guideline. MPA networks must be representative (including as many biodiversity features as possible), adequate (including enough proportions of species populations to ensure their persistence), connected (ensuring the connections among species populations), and cost-efficient (accounting for the costs of restricting human activities) in order to provide substantial conservation benefits and foster our capacity to cope with future changes. Furthermore, to achieve effective conservation, policy and decision makers should not only aim at the enlargement of the Natura 2000 network in the marine environment but should also ensure the effective management of the currently designated Natura 2000 sites. Supplementary data Supplementary material is available at the ICESJMS online version of the article. Funding The work of A.D.M. was partially supported by the Action “Staff Mobility for Training”, co-funded by the Erasmus+ Programme of the European Union. The work of S.G. was supported by an ANR fellowship (ANR-16-ACHN-0016-01). References Agardy M. T. 1994. Advances in marine conservation: the role of marine protected areas. Trends in Ecology and Evolution , 9: 267– 270. Google Scholar CrossRef Search ADS PubMed Almpanidou V., Schofield G., Kallimanis A. S., Türkozan O., Hays G. C., Mazaris A. D. 2016. Using climatic suitability thresholds to identify past, present and future population viability. Ecological Indicators , 71: 551– 556. Google Scholar CrossRef Search ADS Andrello M., Mouillot D., Beuvier J., Albouy C., Thuiller W., Manel S. 2013. Low connectivity between Mediterranean marine protected areas: a biophysical modeling approach for the dusky grouper Epinephelus marginatus. Plos One , 8: e68564. Google Scholar CrossRef Search ADS PubMed Benn A. R., Weaver P. P., Billet D. S., Van Den Hove S., Murdock A. P., Doneghan G. B., Le Bas T. 2010. Human activities on the deep seafloor in the North East Atlantic: an assessment of spatial extent. Plos One , 5: e12730. Google Scholar CrossRef Search ADS PubMed Claudet J., Fraschetti S. 2010. Human-driven impacts on marine habitats: a regional meta-analysis in the Mediterranean Sea. Biological Conservation , 143: 2195– 2206. Google Scholar CrossRef Search ADS Coll M., Piroddi C., Albouy C., Ben Rais Lasram F., Cheung W. W., Christensen V., Karpouzi V. S. et al. 2012. The Mediterranean Sea under siege: spatial overlap between marine biodiversity, cumulative threats and marine reserves. Global Ecology and Biogeography , 21: 465– 480. Google Scholar CrossRef Search ADS Costello M. J., Ballantine B. 2015. Biodiversity conservation should focus on no-take Marine Reserves: 94% of Marine Protected Areas allow fishing. Trends in Ecology and Evolution , 30: 507– 509. Google Scholar CrossRef Search ADS PubMed Danovaro R., Gambi C., Dell'Anno A., Corinaldesi C., Fraschetti S., Vanreusel A., Vincx M. et al. 2008. Exponential decline of deep-sea ecosystem functioning linked to benthic biodiversity loss. Current Biology , 18: 1– 8. Google Scholar CrossRef Search ADS PubMed Day J., Dudley N., Hockings M., Holmes G., Laffoley D. D. A., Stolton S., Wells S. M. 2012. Guidelines for Applying the IUCN Protected Area Management Categories to Marine Protected Areas. IUCN, Gland, Switzerland. Douvere F., Ehler C. N. 2009. New perspectives on sea use management: initial findings from European experience with marine spatial planning. Journal of Environmental Management , 90: 77– 88. Google Scholar CrossRef Search ADS PubMed EC. 1992. Council Directive on the conservation of natural habitats and of wild fauna and flora. European Commission, Directive 92/43/EEC, OJ L 206. EC. 2009. Directive of the European Parliament and the Council on the conservation of wild birds. European Commission, Directive 2009/147/EC, OJ L 20. EC. 2016. European Commission. Nature and Biodiversity Newsletter. Available from http://ec.europa.eu/environment/nature/info/pubs/natura2000nl_en.htm (accessed November 2016). EEA. 2012. European environmental Agency. Protected areas in Europe: an overview. EEA Report N° 5/2012. EU, Luxembourg. EEA. 2017. Natura 2000 Barometer. http://ec.europa.eu/environment/nature/natura2000/barometer/index_en.htm (accessed November 2016) Evans D. 2012. Building the European Union’s Natura 2000 network. Nature Conservation , 1: 11– 26. Google Scholar CrossRef Search ADS Giakoumi S., Grantham H. S., Kokkoris G. D., Possingham H. P. 2011. Designing a network of marine reserves in the Mediterranean Sea with limited socio-economic data. Biological Conservation , 144: 753– 763. Google Scholar CrossRef Search ADS Giakoumi S., Halpern B. S., Michel L. N., Gobert S., Sini M., Boudouresque C. F., Gambi M. C. et al. 2015. Towards a framework for assessment and management of cumulative human impacts on marine food webs. Conservation Biology , 29: 1228– 1234. Google Scholar CrossRef Search ADS PubMed Giakoumi S., Katsanevakis S., Vassilopoulou V., Panayotidis P., Kavadas S., Issaris Y., Kokkali A. et al. 2012. Could European marine conservation policy benefit from systematic conservation planning? Aquatic Conservation: Marine and Freshwater Ecosystems , 22: 762– 775. Google Scholar CrossRef Search ADS Gaines S. D., White C., Carr M. H., Palumbi S. R. 2010. Designing marine reserve networks for both conservation and fisheries management. Proceedings of the National Academy of Sciences , 107: 18286– 18293. Google Scholar CrossRef Search ADS Green A. L., Fernandes L., Almany G., Abesamis R., McLeod E., Aliño P. M., White A. T. et al. 2014. Designing marine reserves for fisheries management, biodiversity conservation, and climate change adaptation. Coastal Management , 42: 143– 159. Google Scholar CrossRef Search ADS Grorud-Colvert K., Claudet J., Tissot B. N., Caselle J. E., Carr M. H., Day J. C., Friedlander A. M. et al. 2014. Marine Protected Area Networks: Assessing Whether the Whole Is Greater than the Sum of Its Parts. PLoS ONE , 9: e102298. Google Scholar CrossRef Search ADS PubMed Gubbay S. 1995. Marine Protected Areas—Past, Present and Future , pp. 1– 14. Springer, Netherlands. Google Scholar CrossRef Search ADS Hays G. C., Ferreira L. C., Sequeira A. M., Meekan M. G., Duarte C. M., Bailey H., Bailleul F. et al. 2016. Key questions in marine megafauna movement ecology. Trends in Ecology and Evolution , 31: 463– 475. Google Scholar CrossRef Search ADS PubMed Iojă C. I., Pătroescu M., Rozylowicz L., Popescu V. D., Vergheleţ M., Zotta M. I., Felciuc M. 2010. The efficacy of Romania’s protected areas network in conserving biodiversity. Biological Conservation , 143: 2468– 2476. Google Scholar CrossRef Search ADS Johnson M. P., Crowe T., McAllen R., Allcock A. 2008. Characterizing the marine Natura 2000 network for the Atlantic region. Aquatic Conservation: Marine and Freshwater Ecosystems , 18: 86– 97. Google Scholar CrossRef Search ADS Jones P. J. 2001. Marine protected area strategies: issues, divergences and the search for middle ground. Reviews in Fish Biology and Fisheries , 11: 197– 216. Google Scholar CrossRef Search ADS Katsanevakis S., Levin N., Coll M., Giakoumi S., Shkedi D., Mackelworth P., Levy R. et al. 2015. Marine conservation challenges in an era of economic crisis and geopolitical instability: the case of the Mediterranean Sea. Marine Policy , 51: 31– 39. Google Scholar CrossRef Search ADS Klein C. J., Jupiter S. D., Selig E. R., Watts M. E., Halpern B. S., Kamal M., Roelfsema C. et al. 2012. Forest conservation delivers highly variable coral reef conservation outcomes. Ecological Applications , 22: 1246– 1256. Google Scholar CrossRef Search ADS PubMed Korpinen S., Meski L., Andersen J. H., Laamanen M. 2012. Human pressures and their potential impact on the Baltic Sea ecosystem. Ecological Indicators , 15: 105– 114. Google Scholar CrossRef Search ADS Leslie H., Ruckelshaus M., Ball I. R., Andelman S., Possingham H. P. 2003. Using siting algorithms in the design of marine reserve networks. Ecological Applications , 13: S185– S198. Google Scholar CrossRef Search ADS Lagabrielle E., Crochelet E., Andrello M., Schill S. R., Arnaud‐Haond S., Alloncle N., Ponge B. 2014. Connecting MPAs–eight challenges for science and management. Aquatic Conservation: Marine and Freshwater Ecosystems , 24: 94– 110. Google Scholar CrossRef Search ADS Lubchenco J., Grorud-Colvert K. 2015. Making waves: The science and politics of ocean protection. Science 350: 382– 383. Google Scholar CrossRef Search ADS PubMed Mackelworth P. 2012. Peace parks and transboundary initiatives: implications for marine conservation and spatial planning. Conservation Letters , 5: 90– 98. Google Scholar CrossRef Search ADS Mackelworth P. 2016. Marine Transboundary Conservation and Protected Areas , Routledge, London and New York. Maiorano L., Amori G., Montemaggiori A., Rondinini C., Santini L., Saura S., Boitani L. 2015. On how much biodiversity is covered in Europe by national protected areas and by the Natura 2000 network: insights from terrestrial vertebrates. Conservation Biology , 29: 986– 995. Google Scholar CrossRef Search ADS PubMed Maiorano L., Falcucci A., Garton E. O., Boitani L. 2007. Contribution of the Natura 2000 network to biodiversity conservation in Italy. Conservation Biology , 21: 1433– 1444. Google Scholar CrossRef Search ADS PubMed Mazaris A. D., Almpanidou V., Wallace B. P., Pantis J. D., Schofield G. 2014. A global gap analysis of sea turtle protection coverage. Biological Conservation , 173: 17– 23. Google Scholar CrossRef Search ADS Mazaris A. D., Papanikolaou A. D., Barbet-Massin M., Kallimanis A. S., Jiguet F., Schmeller D. S., Pantis J. D. 2013. Evaluating the connectivity of a protected areas’ network under the prism of global change: the efficiency of the European Natura 2000 network for four birds of prey. Plos One , 8: e59640. Google Scholar CrossRef Search ADS PubMed Mazor T., Possingham H. P., Kark S. 2013. Collaboration among countries in marine conservation can achieve substantial efficiencies. Diversity and Distributions , 19: 1380– 1393. Google Scholar CrossRef Search ADS McLeod E., Salm R., Green A., Almany J. 2009. Designing marine protected area networks to address the impacts of climate change. Frontiers in Ecology and the Environment , 7: 362– 370. Google Scholar CrossRef Search ADS Moffitt E. A., White J. W., Botsford L. W. 2011. The utility and limitations of size and spacing guidelines for designing marine protected area (MPA) networks. Biological Conservation , 144: 306– 318. Google Scholar CrossRef Search ADS Moilanen A., Wilson K. A., Possingham H. P. (Eds.). 2009. Spatial Conservation Prioritization: Quantitative Methods and Computational Tools , Vol. 6. Oxford University Press, Oxford. Micheli F., Levin N., Giakoumi S., Katsanevakis S., Abdulla A., Coll M., Fraschetti S. et al. 2013. Setting priorities for regional conservation planning in the Mediterranean Sea. PLoS One, 8: e59038. Olsen E. M., Johnson D., Weaver P., Goñi R., Ribeiro M. C., Rabaut M., Macpherson E. et al. 2013. Achieving ecologically coherent MPA networks in Europe: science needs and priorities. In Marine Board Position Paper 18 . Ed. by Larkin K. E., McDonough N.. European Marine Board, Ostend, Belgium. Opermanis O., MacSharry B., Aunins A., Sipkova Z. 2012. Connectedness and connectivity of the Natura 2000 network of protected areas across country borders in the European Union. Biological Conservation , 153: 227– 238. Google Scholar CrossRef Search ADS Opermanis O., MacSharry B., Evans D., Sipkova Z. 2013. Is the connectivity of the Natura 2000 network better across internal or external administrative borders? Biological Conservation , 166: 170– 174. Google Scholar CrossRef Search ADS Orlikowska E. H., Roberge J.-M., Blicharska M., Mikusiński G. 2016. Gaps in ecological research on the world's largest internationally coordinated network of protected areas: a review of Natura 2000. Biological Conservation , 200: 216– 227. Google Scholar CrossRef Search ADS Palumbi S. R. 2003. Population genetics, demographic connectivity, and the design of marine reserves. Ecological Applications , 13: S146– S158. Palumbi S. R. 2004. Marine reserves and ocean neighborhoods: the spatial scale of marine populations and their management. Annual Review of Environmental Resources , 29: 31– 68. Google Scholar CrossRef Search ADS PISCO & UNS. 2016. The Science of Marine Protected Areas, 3rd edn. Mediterranean. 22 pp. www.piscoweb.org (last accessed November 2016). Qiu W., Jones P. J. 2013. The emerging policy landscape for marine spatial planning in Europe. Marine Policy , 39: 182– 190. Google Scholar CrossRef Search ADS Ramos J., Soma K., Bergh Ø., Schulze T., Gimpel A., Stelzenmüller V., Mäkinen T. et al. 2015. Multiple interests across European coastal waters: the importance of a common language. ICES Journal of Marine Science: Journal du Conseil , 72: 720– 731. Google Scholar CrossRef Search ADS Rands M. R., Adams W. M., Bennun L., Butchart S. H., Clements A., Coomes D., Entwistle A. et al. 2010. Biodiversity conservation: challenges beyond 2010. Science , 329: 1298– 1303. Google Scholar CrossRef Search ADS PubMed Sundblad G., Bergström U., Sandström A. 2011. Ecological coherence of marine protected area networks: a spatial assessment using species distribution models. Journal of Applied Ecology , 48: 112– 120. Google Scholar CrossRef Search ADS Trochet A., Schmeller D. 2013. Effectiveness of the Natura 2000 network to cover threatened species. Nature Conservation , 4: 35– 53. Google Scholar CrossRef Search ADS Tsiafouli M. A., Apostolopoulou E., Mazaris A. D., Kallimanis A. S., Drakou E. G., Pantis J. D. 2013. Human activities in Natura 2000 sites: a highly diversified conservation network. Environmental Management , 51: 1025– 1033. Google Scholar CrossRef Search ADS PubMed Wood L. J., Fish L., Laughren J., Pauly D. 2008. Assessing progress towards global marine protection targets: shortfalls in information and action. Oryx , 42: 340– 351. Google Scholar CrossRef Search ADS © International Council for the Exploration of the Sea 2017. All rights reserved. For Permissions, please email: firstname.lastname@example.org
ICES Journal of Marine Science – Oxford University Press
Published: Jan 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera