Predicted impacts of transport infrastructure and traffic on bird conservation in Swedish Special Protection Areas

Jan Olof Helldin

doi:10.3897/natureconservation.36.31826

Abstract

The ecological impacts of roads and railways extend into the surrounding landscape, leading to habitat degradation and reduced wildlife densities within an area that is considerably larger than the actual road or railway corridor. For birds, an extensive meta-analysis has identified an average of 20% density reduction within 1 km from the infrastructure. I investigated to what extent this density reduction can be expected to compromise the habitat quality and conservation value of Swedish Natura 2000 areas designated for the protection of birds (Special Protection Areas; SPAs). The majority (63%) of Swedish SPAs are, to some extent, found within this 1 km road/railway effect zone (REZ). The total overlap between SPA and REZ is approximately 126,000 ha or 4.2% of the country’s SPA area. There are, however, large differences amongst bio-geographical regions. In the southern (continental) and coastal regions combined, 25.8% of the total SPA area fall within REZ, representing an estimated 4–7% reduction in bird abundance within SPAs. The probability of overlap with REZ is higher for larger SPAs. However, the proportion of overlap is higher for smaller SPAs and, accordingly, smaller sites can be assumed to experience a greater impact from transport infrastructure and traffic. The impacts on Natura 2000 sites are particularly concerning as this network of protected areas is a cornerstone for maintenance and restoration of biodiversity within the EU. I recommend placing a stronger emphasis in the management of Natura 2000 sites on the threats to wildlife conservation caused by transport infrastructure and traffic. Special attention should be paid to sites with a large overlap with the REZ and sites hosting particularly vulnerable taxa or habitats. Infrastructure owners and managers should make their best efforts to minimise and compensate for the negative impacts of roads and railways and associated traffic in SPAs and other protected areas.

Keywords

Effect zone, Natura 2000, railway, road, Sweden

Introduction

Ecological impact of transport infrastructure

Infrastructure development is recognised as one of the significant drivers of global biodiversity loss and, with increasing traffic and expanding infrastructure networks worldwide, the pressure on biodiversity is expected to increase in coming decades (EEA-FOEN 2011, EEA 2012, OECD 2012). The impacts of transport infrastructure on wildlife are well described (Forman et al. 2003, van der Ree et al. 2015) and include loss of habitat, traffic casualties, creation of physical barriers, disturbance by noise, light and other visual cues, spread of chemicals, dust and alien species, changes in hydrology and microclimate and accidental spills. Most of these impacts extend into the surrounding landscape, leading to degradation and fragmentation of habitats and, for some species, to restricted movements, increased mortality and avoidance of a zone around the infrastructure (Forman et al. 2003, EEA-FOEN 2011, van der Ree et al. 2015, Tulloch et al. 2019).

Due to these impacts, the population densities of many wildlife species are reduced within a distance from larger infrastructures (Rytwinski and Fahrig 2015). For example, the population densities may be reduced up to about 1 km distance for birds (Forman and Deblinger 2000, Forman et al. 2002, Benítez-López et al. 2010) and anurans (Eigenbrod et al. 2009) and up to 5 km for mammals (Benítez-López et al. 2010). Not only large infrastructure but also minor and unpaved roads may have a considerable impact on some species (e.g. van Langevelde and Jaarsma 2009, Benítez-López et al. 2010, Shanley and Pyare 2011). Based on an extensive meta-analysis, Benítez-López et al. (2010) showed that the mean bird and mammal abundance in an effect zone around infrastructures is reduced by 20–30% and with an increasing reduction with proximity to the infrastructure (Fig. 1).

Figure 1.

Mean species abundance (MSA) of birds as a function of distance to infrastructure (logistic regression). Open dots represent the pooled results of a meta-analysis per distance interval ±SE. The solid black line denotes the estimated curve for the decline of MSA in proximity to infrastructure; dashed lines are the 95% upper and lower limits of the confidence bands of the curve. Figure from Benítez-López et al. (2010).

Accordingly, in regions with dense infrastructure networks, large natural areas may be situated within this effect zone and therefore may be impoverished in species sensitive to traffic and transport infrastructure. For example, in the United States, the road effect zone covers 15–22% of the total land area and more than 60% of some particularly exposed biomes, such as coastal regions and river basins (Forman 2000, Riitters and Wickham 2003). In Spain, a country with intermediate road density with European standards, it is expected that reduced bird densities, due to transportation infrastructure, will affect 49% of the country and lead to reduced mammal densities in as much as 96% of the country (Torres et al. 2016). Some habitats of particular importance to biodiversity in Europe, such as wetlands, semi-natural grasslands and temperate broad-leaved forest, may be disproportionately affected by roads because of the landscape structure (Helldin et al. 2013, Karlson and Mörtberg 2015, Torres et al. 2016). Disturbances (noise and visual cues) tend to be stronger and extend further from the road in open habitats as compared to forest (Forman and Deblinger 2000, Reijnen and Foppen 2006). There is growing concern globally about the impact of roads and traffic on wildlife populations in protected areas and other biodiversity hotspots (Forman and Deblinger 2000, Ament et al. 2008, Selva et al. 2011, Laurance and Balmford 2013, Bager et al. 2015, Gadd 2015, Seshadri and Ganesh 2015).

Integrity of the Natura 2000 network

The Natura 2000 network of protected areas is a key tool in the maintenance and restoration of biodiversity within the European Union (EU). The network consists of Special Protection Areas (SPAs), designated according to the Birds Directive and Special Areas of Conservation (SACs), designated according to the Habitats Directive (EEC 1992, EEC 2010). Under these directives, the network is supposed to provide protection for vulnerable wildlife and habitats. One important motivation for designating Natura 2000 sites, particularly in coastal and other lower-elevation areas, is the protection against negative impacts of urbanisation and infrastructure development (EEA 2012). There are currently more than 26,400 sites in the Natura 2000 network, accounting for ca. 18% of EU’s land territory.

The network is, however, biased toward highland areas, and lowland areas are under-represented (Oldfield et al. 2004, Maiorano et al. 2007). In addition, the average size of the Natura 2000 sites is quite low and particularly so in lowlands (Maiorano et al. 2007, Gaston et al. 2008, EEA 2012). As smaller sites are more susceptible to pressure from land use and human activities surrounding them, major concerns are expressed about the capacity of existing protected areas to maintain their biodiversity values (Chape et al. 2005, Gaston et al. 2008, Maiorano et al. 2008, Kati et al. 2015).

Though many assessments for the effectiveness of the Natura 2000 network in protecting species have been reported in recent years, a vast majority of these relied on rather static population data, such as species’ geographic distribution, species presence/absence or habitat suitability (e.g. Araújo et al. 2007, Maiorano et al. 2007, 2015, Sánchez-Fernández et al. 2008, Iojă et al 2010, López-López et al. 2011, Gruber et al. 2012, Albuquerque et al. 2013, D’Amen et al. 2013, Lison et al. 2013, Rubio-Salcedo et al. 2013, Trochet and Schmeller 2013) and, accordingly, were designed to assess the ecological representation of the network rather than tracking changes in population densities due to environmental impacts. In view of the many negative population trajectories reported for both avian and non-avian species in the EU (EEA 2015), it appears necessary to analyse the ecological functionality of the Natura 2000 network with regard to pressures both within and outside the designated areas, but few studies have taken this course.

Frequency of transport infrastructure within Natura 2000 sites was investigated by Tsiafouli et al. (2013), showing that as a European average, roads are present in 29% of the sites, with a higher frequency in the countries in the south. A preceding study of Greek Natura 2000 sites, by Votsi et al. (2012), showed that 85% of sites are bisected by roads. Insufficient functional connectivity of the Natura 2000 network has been reported by Gurrutxaga et al. (2011) and Opermanis et al. (2012), suggesting that dispersal barriers exist between sites, for example in the form of large roads (Gurrutxaga et al. 2011). With regard to disturbance, the European Environmental Agency estimated that almost 20% of Natura 2000 areas are presently adversely affected by high levels of environmental noise, largely owing to major transport infrastructure (EEA 2016). In an assessment of the impact on Natura 2000 sites of major traffic arteries, planned or under construction as part of EU’s TEN-T framework, Byron and Arnold (2008) estimated that 379 SPAs (8% of sites) and 953 SCIs (4% of sites) would be adversely affected by these new traffic arteries, with potential effects also on the coherence of the Natura 2000 network.

Aim of the study

Sweden is one of the European countries that is least fragmented by transport infrastructure and built-up areas (EEA-FOEN 2011) and the physical impact of infrastructure is generally not well acknowledged in Swedish nature conservation. It was not until recently that national status reports for biodiversity have described the threats from infrastructure and traffic on species conservation (Bernes 2011, Naturvårdsverket 2015a) and the current national conservation action plan contains few requirements for infrastructure managers (Naturvårdsverket 2015b). Biodiversity is generally insufficiently described in impact assessments for transportation infrastructure plans and projects (Wärnbäck 2013, Karlson et al. 2014). Few, if any, management plans for protected areas address the full array of potential ecological impacts of transport infrastructure on the areas’ conservation status or management (Helldin and Tytor 2017). This ignorance is not unique for Sweden, as it appears to be largely similar in most EU member states (Tsiafouli et al. 2013, EEA 2015; but see, for example, Selva et al. 2011, Votsi et al. 2012).

In order to illustrate and highlight the impacts of infrastructure on protected areas in particular, I estimated the frequency and proportion of Swedish SPAs situated within the predicted effect zone for birds around existing larger transport infrastructure (roads and railways) and hence cannot be expected to reach their full conservation potential due to infrastructure impacts. I included only SPAs in the estimation, i.e. areas designated specifically for the protection of birds, because the effects of roads and railways on birds are well described in literature and apparently can impact the majority of bird species (Reijnen and Foppen 2006, Benítez-López et al. 2010, Rytwinski and Fahrig 2015). I assessed the conservation value of SPAs that is lost due to infrastructure in terms of reduced predicted bird abundance. Due to the large geographic variation over the country in density of the infrastructure network and proportion of area within an SPA, I separated the analyses between biogeographical regions. I tentatively explored the association between the infrastructure impact on an individual SPA and its size and dominating habitats. In this paper, I present the results of these analyses and propose improvements for the management of Swedish SPAs.

Methods

A map of the effect zone for birds around the existing larger Swedish roads and railways (REZ in the following) was produced. I used a standardised effect distance of 1 km from the infrastructure, following the results from a meta-analysis presented by Benítez-López et al. (2010). To my knowledge, this is the most comprehensive analysis of infrastructure impacts on bird densities, including 49 bird datasets and 201 bird species. As most studies in the meta-analysis were conducted in biomes that occur in Sweden, i.e. taiga, temperate broadleaf forest or alpine/tundra (39 of the 49 datasets) and on species that occur in Sweden (105 of the 201 species), I judged the results to be relevant to a Swedish perspective. The results have also previously been applied to assess the impacts of the road network on birds in Sweden (Karlson and Mörtberg 2015) and Europe (Torres et al. 2016). I consider the assumption of a 1 km effect distance to be conservative because i) individual studies in the meta-analysis indicated reduced bird populations at larger distances and ii) impacts at greater distances may not necessarily result in direct population declines, but yet be of ecological significance. Infrastructure data were obtained in December 2015 from Open Street Map (http://openstreetmapdata.com), using only the following road classes: primary road, secondary road, tertiary road, motorway, trunk road and railway (thus excluding minor roads for which ecological effects are less well known).

The REZ map was overlaid with all Swedish SPAs to calculate the area and proportion of each site situated within REZ. The area and habitat distribution of SPAs were obtained in December 2015 from European Environment Agency’s Natura 2000 database (http://www.eea.europa.eu/data-and-maps/data/natura-2000-eunis-database). SPAs were separated depending on biome (based on a combination of the Natura 2000 database and Global Biomes data from the CIESIN; http://sedac.ciesin.columbia.edu/data/set/nagdc-population-landscape-climate-estimates-v3/maps?facets=theme:climate) and proximity to coast (data on coastline obtained in December 2015 from Open Street Map; http://openstreetmapdata.com) on the following terms:

continental region: sites with >50% of the area within EU continental region,

mixed-forest region: sites with >50% of the area within EU boreal region and in CIESIN broadleaf and mixed-forest region,

boreal region: rest of sites within EU boreal region but with no part within EU alpine region or

alpine region: sites with at least some part of the area within EU alpine region

in combination with

coastal: mainland sites with at least some part within 20 km from coast of mainland Sweden (including mainland islands Öland and Gotland) or

inland: the rest of mainland sites.

The alpine region in Sweden is only inland, i.e. no alpine coastal sites exist. As only two continental sites are inland, all continental sites were pooled in one region. In addition, an off-coastal region was formed including all sites with no contact with mainland Sweden, irrespective of terrestrial biome. Hereby, a total of seven biogeographical regions were obtained (Table 1 and Fig. 2).

Figure 2.

Special Protection Areas (SPAs) in Sweden divided by biogeographical region (left map) and larger roads and railways with predicted effect zone (REZ; right map).

Table 1.

Download as

CSV

XLSX

Predicted impacts on SPAs. Number, area and proportion of Swedish SPAs within an effect zone of 1 km from larger transport infrastructure (REZ) and predicted total reduction in bird abundance (with 95% confidence interval) due to the effects. Results are given for the entire country and by biogeographical region. Detailed results for each SPA are provided in appendix available online at http://triekol.se/earlier-work/infrastructure-impacts-on-protected-areas/.

Region	Total no. of SPAs	Total area of SPA (ha)	No. of SPAs affected to		Total area of SPA in REZ (ha)	Proportion of total SPA in REZ (%)	Reduction in bird abundance (% with C.I.)
Region	Total no. of SPAs	Total area of SPA (ha)	>0%	>50%	Total area of SPA in REZ (ha)	Proportion of total SPA in REZ (%)	Reduction in bird abundance (% with C.I.)
Continental	41	53,331	41	18	19,202	36.0	7.2 (4.3–11.9)
Mixed-forest coastal	94	103,064	71	24	20,865	20.2	4.0 (2.4–6.7)
Mixed-forest inland	161	258,804	119	43	34,622	13.4	2.7 (1.6–4.4)
Boreal coastal	22	21,712	15	8	5,968	27.5	5.5 (3.3–9.1)
Boreal inland	130	162,924	63	30	13,122	8.1	1.6 (1.0–2.7)
Alpine (only inland)	26	1,984,005	16	0	28,242	1.4	0.3 (0.2–0.5)
Off-coastal	64	415,308	14	0	3,913	1.0	0.2 (0.1–0.3)
All Sweden	538	2,999,149	339	123	125,946	4.2	0.8 (0.5–1.4)

To assess the bird conservation value of SPAs that is lost due to infrastructure, I used an average of 20% (C.I. 12–33%) reduction of bird abundance within the 1 km effect distance from the infrastructure, as indicated by the results presented by Benítez-López et al. (2010; Fig. 1). GIS analyses were conducted using ArcGIS version 10.2 and QGIS version 2.6. To explore the differences between biogeographical regions in the proportion of SPA area within REZ, an ANOVA with Tukey´s post-hoc test was conducted, with region as explanatory variable. To explore how the degree of impact on an SPA is associated with its size and dominating habitat, two different analyses were performed for each region. For the probability for an SPA to overlap to any degree with REZ (response variable either 0 or 1), a generalised linear model with a logit link function (logistic regression) was used, and for the proportion of overlap with REZ (i.e. a response variable from >0 to 1), a beta-regression model (Ferrari and Cribari-Neto 2004) was used, including only sites with overlap. In both types of models, explanatory variables were 1) SPA size, 2) proportion of forest habitat, 3) proportion of wetlands and 4) proportion of agricultural land and grasslands. SPA size was log-transformed to improve normality and all variables were standardised to make parameter estimates comparable. Model selection was based on AIC and the final models were the ones with a combination of explanatory variables resulting in the lowest AIC. Statistical analyses were conducted using programme R version 3.4.2.

Results

The overlay of SPAs and REZ showed that 339 of Sweden´s 538 SPAs (63%) have at least some part and 123 (23%) have most of their area within REZ (Table 1). In terms of area, a total of approximately 126,000 ha or 4.2% of the total SPA area in the country lies within REZ.

However, national level figures on the impacted area give a crude picture, as the results pointed to large differences amongst the biogeographical regions (Table 1). The proportion of SPA within REZ differed among regions (F = 13.8, p <0.001), with SPAs in the alpine and off-coastal regions having a significantly smaller proportion of their area in REZ than the other regions, and boreal inland SPAs having a significantly smaller proportion than continental SPAs. Alpine and off-coastal areas have a number of large SPAs and hold most of Sweden´s total SPA area, but have sparse networks of large (terrestrial) transport infrastructure and, accordingly, they reduce the national average. Continental, mixed-forest coastal and boreal coastal regions, however, are comparably more impacted; in these three regions combined, a total of 46,046 ha or 25.8% of the total SPA area fall within REZ. In terms of bird conservation value, present infrastructure impacts are predicted to cause a 4–7% reduction in bird abundance in SPAs in the three most impacted regions (continental, mixed-forest coastal and boreal coastal) and an average of 1% reduction when all SPAs in the country are taken into consideration (Table 1).

The overlap models with the lowest AIC included SPA size or habitat composition in all regions except the continental (Table 2). As the most strongly emergent pattern, larger SPAs appear to have a higher probability of overlap with REZ (in three regions), but a lower proportion of overlap with REZ (in all regions except continental). Another pattern, however less emergent, is that the overlap with REZ is larger in SPAs with more agricultural land and grasslands (in four regions) and smaller in SPAs with more forest habitat (in two regions). The pattern regarding overlap with SPAs with more wetland is inconclusive, and in general, only a minor part of the variation in overlap could be explained by SPA size and dominating habitat.

Table 2.

Download as

CSV

XLSX

Variables associated with road/railway effect zone (REZ) overlap with Swedish SPAs divided by biogeographical region. Values given are mean estimates of coefficients of logistic regressions and beta-regressions with standard errors (SE) and probabilities (P). Values are only given for variables that were included in the final model. The right column gives the variation explained by the final model.

Region	Log(size)		Forested area		Wetland		Agri & grassland
Region	Estimate (SE)	P	Estimate (SE)	P	Estimate (SE)	P	Estimate (SE)	P
Logistic regression (probability of overlap)									Explained deviance
Continental †	–	–	–	–	–	–	–	–	–
Mixed-forest coastal	0.45(0.27)	0.094							0.03
Mixed-forest inland					0.49(0.21)	0.019	0.91(0.38)	0.016	0.07
Boreal coastal			-1.06(0.57)	0.065					0.16
Boreal inland			-2.61(0.75)	<0.001	-2.06(0.70)	<0.001			0.21
Alpine (only inland)	1.31(0.64)	0.039			1.23(0.64)	0.056			0.23
Off-coastal	1.07(0.40)	0.008			0.79(0.32)	0.015			0.18
Beta-regression (proportion of overlap)									Pseudo R²
Continental ‡	–	–	–	–	–	–	–	–	–
Mixed-forest coastal	-0.56(0.15)	<0.001					0.44(0.15)	0.002	0.26
Mixed-forest inland	-0.64(0.12)	<0.001					0.47(0.11)	<0.001	0.40
Boreal coastal	-0.85(0.34)	0.013							0.36
Boreal inland	-0.56(0.16)	<0.001	-0.48(0.16)	0.004			0.43(0.16)	0.009	0.41
Alpine (only inland)	-0.78(0.11)	<0.001			-0.38(0.17)	0.022			0.50
Off-coastal	-1.09(0.24)	<0.001					0.40(0.19)	0.036	0.38

Discussion

The results indicate that a significant proportion of Swedish SPAs, both in terms of area and number of sites, lies within a predicted effect zone for birds around present larger transport infrastructure (REZ) and therefore can be expected not to reach their full potential as a bird habitat. The reduction due to transport infrastructure impacts may not be dramatic when seen in the country as a whole, with only around 4% of the total SPA area affected, corresponding to ca. 1% reduction in predicted bird abundance within SPAs. However, for more urbanised parts of the country, with a denser infrastructure network, the predicted impact and reduction is nearly an order of magnitude larger and may well be one of the main factors determining bird abundance in protected areas. This is the case in the southern (continental) and coastal regions of the country, where the urbanisation and landscape fragmentation is at a similar level to that of most western and central European countries (EEA-FOEN 2011).

At the level of individual SPAs, smaller sites tend to have a higher proportion of overlap with REZ and, accordingly, can be assumed to experience a greater impact than larger sites. This pattern emerges in all biogeographical regions, except the continental where most SPAs are small and indeed impacted to a large degree. In effect, the larger impact on smaller sites amplifies the bias against area protection in the lowlands, i.e. the southern and coastal regions. This is in line with general concerns previously expressed about the small size of many protected areas in Europe and about the impacts from transport infrastructure, traffic and other urban development in the landscape surrounding them (Shafer 1995, Gaston et al. 2008, Maiorano et al. 2008, Kati et al. 2015). However, as indicated by Helldin and Tytor (2017), these concerns are not well expressed in the management plans for Swedish SPAs and appear particularly underestimated in the regions where the impacts are the largest.

The functions of predicted bird abundance and distance to infrastructure described by Benítez-López et al. (2010) have previously been applied twice to assess the ecological impact of an infrastructure network at a larger geographical scale. Karlson and Mörtberg (2015) presented an assessment of the impacts of roads on habitats of high diversity value in Sweden (irrespective of their protection status), concluding that natural grasslands and southern broadleaved forest are likely to be particularly impacted; on a national level, 13–19% and 16–24%, respectively, of the total areas of these habitats are found within the predicted road effect zone for birds. Torres et al. (2016) estimated a 19.0% (CI: 9.6–25.6%) reduction in national bird numbers due to transport infrastructure for Spain, when considering all land, not only protected areas. They too concluded that some habitats (most notably farmland and maritime wetlands) might be disproportionately affected by transport infrastructure. In relation to these previous assessments, the present study is unique in that it points out the impacts specifically on protected areas, i.e. areas where nature conservation should be a top priority.

The assessment was aimed at providing a general picture and was therefore simplified in several respects. A fixed-width REZ is less realistic, as the actual effect depends on the local context, such as the habitat distribution, topography, species and ecological processes involved (Forman and Deblinger 2000, Riitters and Wickham 2003, Biglin and Dupigny-Giroux 2006), or the road characteristics (e.g. Reijnen and Foppen 2006, Rytwinski and Fahrig 2015). Additionally, the greater reduction in bird densities near the infrastructure within REZ (Fig. 1) provides an opportunity to assess the decline in bird abundance within the effect zone in individual sites in greater detail than conducted here (e.g. Torres et al. 2016, Tulloch et al. 2019).

Furthermore, the analysis of impacts on different habitats within SPAs was rather coarse, since the Natura 2000 database does not provide habitat maps. Therefore, I could not explore to what degree EU priority habitats (habitat types of community interest; EEC 1992) are distributed disproportionately within REZ.

Because the study was focused on transport infrastructures in terrestrial environments (roads and railways), the low predicted impact on off-coastal protected areas is not surprising. A corresponding analysis of the frequency of shipping and proximity to marine fairways would be of relevance to assess traffic impacts in marine reserves. Such an analysis would, however, need a different approach than the one here described, as the ecological effects of shipping are not yet comprehensively described and effect distances are less well established (Pirotta et al. 2019).

Implications for the management of protected areas

The present study underlines the concern about the impact of transport infrastructure on wildlife in protected areas in general (Forman and Deblinger 2000, Ament et al. 2008, Selva et al. 2011, Tsiafouli et al. 2013) and in the smaller areas in particular (Shafer 1995, Maiorano et al. 2008). Following article 4 of the Birds Directive, EU Member States must take appropriate steps to avoid habitat deterioration and significant species disturbance within SPAs (EEC 2010). Accordingly, a stronger emphasis on keeping natural areas free from the impacts of heavy traffic and new roads and railways has been proposed for conservation and transport planning (Selva et al. 2011, 2015, Laurance and Balmford 2013, IENE 2015).

Management plans for Natura 2000 sites should better acknowledge and consider the threats to wildlife conservation caused by both present transport infrastructure and new development projects in and near sites (Cortina and Boggia 2014, Helldin and Tytor 2017). Assessments of the effectiveness of individual Natura 2000 sites in maintaining biodiversity should include monitoring of population density and demography of species of special conservation concern (Gaston et al. 2008).

In order to correctly address the impacts from transport infrastructure, managers of Natura 2000 sites should conduct more detailed assessments of REZ in the respective site, based on habitat maps, species occurrences, and local road or railway characteristics (traffic, corridor width, etc.), to serve as a basis for setting priorities in conservation planning and action. Such site specific assessments could take into account the greater reduction in population density in close proximity to the transport infrastructure, and impacts on some wildlife species extending further than the 1 km used in the current study (Reijnen and Foppen 2006, Benítez-López et al. 2010). For example, Tulloch et al. (2019) describe a method to produce site- and taxon-specific REZs, incorporating local biodiversity distribution and expert opinion on the impacts on different taxa. Special attention to road effects should be paid in protected areas with a large overlap with REZ, in areas hosting particularly vulnerable taxa, and in areas with pronounced impacts on EU priority habitats.

Conservation authorities should ensure that infrastructure owners and managers make their best efforts to minimise the negative impact of nearby roads and railways and related traffic. Technical mitigation of impacts of transport infrastructure on birds could include preventing bird-vehicle collisions (e.g. with flight diverters), planning the timing of infrastructure maintenance and construction work to avoid particularly sensitive periods, providing crossing structures, and reducing noise and visual impacts through walls, berms or adapted paving. Such road mitigation measures for birds have been implemented on a few sites in Sweden, but are still lacking over the vast majority of roads and railways (Trafikverket 2017), and moreover, are not always considered in new construction projects. Traffic calming, speed reduction, or road closure (permanent or temporary) would also provide reductions in road mortality, disturbance, and barrier effects. Finally, compensatory measures such as habitat improvement or additional area protection could reinforce remnant populations and restore vital ecological processes.

Acknowledgements

I am grateful to Vadym Sokol for assisting in GIS data retrieval and GIS analyses, and to Victor Johansson for assisting in the statistical analyses. I am also grateful to Lars Nilsson, Mårten Karlson, Jochen Jaeger and an anonymous reviewer for providing helpful comments on earlier drafts of this paper. I thank Ana Benítez-López and her co-workers for kind permission to reproduce their published figure. The study was financed by the Swedish Transport Administration as part of the research programme TRIEKOL (http://triekol.se/home-eng/).

References

Albuquerque FS, Assunção-Albuquerque MJT, Cayuela L, Zamora R, Benito BM (2013) European Bird distribution is “well” represented by Special Protected Areas: Mission accomplished? Biological Conservation 159: 45–50. https://doi.org/10.1016/j.biocon.2012.10.012

Ament R, Clevenger AP, Yu O, Hardy A (2008) An Assessment of Road Impacts on Wildlife Populations in U.S. National Parks. Environmental Management 42(3): 480–496. https://doi.org/10.1007/s00267-008-9112-8

Araújo MB, Lobo JM, Moreno JC (2007) The effectiveness of Iberian protected areas in conserving terrestrial biodiversity. Conservation Biology 21(6): 1423–1432. https://doi.org/10.1111/j.1523-1739.2007.00827.x

Bager A, Borghi CE, Secco H (2015) The influence of economics, politics and environment on road ecology in South America. In: van der Ree R, Smith DJ, Grilo C (Eds) Handbook of Road Ecology. John Wiley & Sons Ltd, UK, 407–413. https://doi.org/10.1002/9781118568170.ch50

Benítez-López A, Alkemade R, Verweij PA (2010) The impacts of roads and other infrastructure on mammal and bird populations: A meta-analysis. Biological Conservation 143(6): 1307–1316. https://doi.org/10.1016/j.biocon.2010.02.009

Bernes C (2011) Biodiversity in Sweden. Monitor 22, Swedish Environmental Protection Agency, Stockholm, 280 pp.

Biglin K, Dupigny-Giroux L-A (2006) Mapping the road-effect zone to assess impacts of proposed road segments. Journal of Conservation Planning 2: 1–16.

Byron H, Arnold L (2008) TEN-T and Natura 2000: the way forward. An assessment of the potential impact of the TEN-T Priority Projects on Natura 2000. Report from BirdLife International. http://www.birdlife.org/eu/pdfs/TEN_T_report2008_final.pdf

Chape S, Harrison J, Spalding M, Lysenko I (2005) Measuring the extent and effectiveness of protected areas as an indicator for meeting global biodiversity targets. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 360(1454): 443–455. https://doi.org/10.1098/rstb.2004.1592

Cortina C, Boggia A (2014) Development of policies for Natura 2000 sites: A multi-criteria approach to support decision makers. Journal of Environmental Management 141: 138–145. https://doi.org/10.1016/j.jenvman.2014.02.039

D’Amen M, Bombi P, Campanaro A, Zapponi L, Bologna MA, Mason F (2013) Protected areas and insect conservation: Questioning the effectiveness of Natura 2000 network for saproxylic beetles in Italy. Animal Conservation 16(4): 370–378. https://doi.org/10.1111/acv.12016

EEA (2012) Protected areas in Europe - an overview. EEA report No 5/2012. European Environment Agency, Copenhagen.

EEA (2015) State of nature in the EU; Results from reporting under the nature directives 2007-2012. EEA report No 2/2015. European Environment Agency, Copenhagen.

EEA (2016) Quiet areas in Europe; The environment unaffected by noise pollution. EEA report No 14/2016. European Environment Agency, Copenhagen.

EEA-FOEN (2011) Landscape fragmentation in Europe; Joint EEA-FOEN report. EEA report No 2/2011. European Environment Agency and Swiss Federal Office for the Environment. Luxembourg, Publications Office of the European Union.

EEC (1992) Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora. Official Journal of the European Communities No L 206/7-50.

EEC (2010) Directive 2009/147/EC of the European Parliament and of the Council of 30 November 2009 on the conservation of wild birds. Official Journal of the European Union No L 20/7-25.

Eigenbrod F, Hecnar SJ, Fahrig L (2009) Quantifying the Road-Effect Zone: Threshold Effects of a Motorway on Anuran Populations in Ontario, Canada. Ecology and Society 14(1): 24. https://doi.org/10.5751/ES-02691-140124

Ferrari S, Cribari-Neto F (2004) Beta Regression for Modelling Rates and Proportions. Journal of Applied Statistics 31(7): 799–815. https://doi.org/10.1080/0266476042000214501

Forman RTT (2000) Estimate of the area affected ecologically by the road system in the United States. Conservation Biology 14(1): 31–35. https://doi.org/10.1046/j.1523-1739.2000.99299.x

Forman RTT, Deblinger RD (2000) The ecological road-effect zone of a Massachusetts (U.S.A.) suburban highway. Conservation Biology 14(1): 36–46. https://doi.org/10.1046/j.1523-1739.2000.99088.x

Forman RTT, Reineking B, Hersperger AM (2002) Road traffic and nearby grassland bird patterns in a suburbanizing landscape. Environmental Management 29(6): 782–800. https://doi.org/10.1007/s00267-001-0065-4

Forman RTT, Sperling D, Bissonette JA, Clevenger AP, Cutshall CD, Dale VH, Fahrig L, France R, Goldman CR, Haenue K, Jones JA, Swanson FJ, Turrentine T, Winter TC (2003) Road ecology - Science and solutions. Island Press, Washington, 504 pp.

Gadd ME (2015) Expected effects of a road across the Serengeti. In: van der Ree R, Smith DJ, Grilo C (Eds) Handbook of Road Ecology. John Wiley & Sons Ltd, UK, 455–464. https://doi.org/10.1002/9781118568170.ch56

Gaston KJ, Jackson SF, Nagy A, Cantú-Salazar L, Johnson M (2008) Protected Areas in Europe; Principle and Practice. Annals of the New York Academy of Sciences 1134(1): 97–119. https://doi.org/10.1196/annals.1439.006

Gruber B, Evans D, Henle K, Bauch B, Schmeller D, Dziock F, Henry PY, Lengyel S, Margules C, Dormann C (2012) “Mind the gap!” – How well does Natura 2000 cover species of European interest? Nature Conservation 3: 45–62. https://doi.org/10.3897/natureconservation.3.3732

Gurrutxaga M, Rubio L, Saurac S (2011) Key connectors in protected forest area networks and the impact of highways: A transnational case study from the Cantabrian Range to the Western Alps (SW Europe). Landscape and Urban Planning 101(4): 310–320. https://doi.org/10.1016/j.landurbplan.2011.02.036

Helldin J-O, Tytor S (2017) Hur uppmärksammas ekologiska effekter av vägar och järnvägar inom naturvården? – innehållsanalys av bevarandeplaner för Natura 2000-områden. Report from Calluna AB. http://www.calluna.se/userfiles/files/5_Innehållsanalys_SPA_bevarandeplaner_2017(1).pdf [In Swedish]

Helldin J-O, Collinder P, Bengtsson D, Karlberg Å, Askling J (2013) Assessment of traffic noise impact in important bird sites in Sweden - a practical method for the regional scale. Oecologia Australis 17(1): 70–84. https://doi.org/10.4257/oeco.2013.1701.05

IENE (2015) Protect remaining roadless areas: The IENE 2014 declaration. Nature Conservation 11: 1–4. https://doi.org/10.3897/natureconservation.11.5630

Iojă CI, Pâtroescu M, Rozylowicz L, Popescu VD, Verghelet M¸, Zotta MI, Felciuc M (2010) The efficacy of Romania’s protected areas network in conserving biodiversity. Biological Conservation 143(11): 2468–2476. https://doi.org/10.1016/j.biocon.2010.06.013

Karlson M, Mörtberg U (2015) A spatial ecological assessment of fragmentation and disturbance effects of the Swedish road network. Landscape and Urban Planning 134: 53–65. https://doi.org/10.1016/j.landurbplan.2014.10.009

Karlson M, Mörtberg U, Balfors B (2014) Road ecology in environmental impact assessment. Environmental Impact Assessment Review 48: 10–19. https://doi.org/10.1016/j.eiar.2014.04.002

Kati V, Hovardas T, Dieterich M, Ibisch PL, Mihok B, Selva N (2015) The challenge of implementing the European network of protected areas Natura 2000. Conservation Biology 29(1): 260–270. https://doi.org/10.1111/cobi.12366

Laurance WF, Balmford A (2013) A global map for road building. Nature 495(7441): 308–309. https://doi.org/10.1038/495308a

Lison F, Palazon JA, Calvo JF (2013) Effectiveness of the Natura 2000 Network for the conservation of cave-dwelling bats in a Mediterranean region. Animal Conservation 16(5): 528–537. https://doi.org/10.1111/acv.12025

López-López P, Maiorano L, Falcucci A, Barba E, Boitani L (2011) Hotspots of species richness, threat and endemism for terrestrial vertebrates in SW Europe. Acta Oecologica 37(5): 399–412. https://doi.org/10.1016/j.actao.2011.05.004

Maiorano L, Falcucci A, Garton EO, Boitani L (2007) Contribution of the Natura 2000 Network to Biodiversity Conservation in Italy. Conservation Biology 21(6): 1433–1444. https://doi.org/10.1111/j.1523-1739.2007.00831.x

Maiorano L, Falcucci A, Boitani L (2008) Size-dependent resistance of protected areas to land-use change. Proceedings. Biological Sciences 275(1640): 1297–1304. https://doi.org/10.1098/rspb.2007.1756

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(4): 986–995. https://doi.org/10.1111/cobi.12535

Naturvårdsverket (2015a) Målövergripande analys av miljömålen; Underlag till fördjupad utvärdering av miljömålen 2015. Swedish Environmental Protection Agency, Report No 6692. [In Swedish]

Naturvårdsverket (2015b) Styr med sikte på miljömålen - Naturvårdsverkets fördjupade utvärdering av miljömålen 2015. Swedish Environmental Protection Agency, Report No 6666. [In Swedish]

OECD (2012) OECD Environmental Outlook to 2050; The Consequences of Inaction. OECD Publishing, Paris. https://doi.org/10.1787/9789264122246-en

Oldfield TEE, Smith RJ, Harrop SR, Leader-Williams N (2004) A gap analysis of terrestrial protected areas in England and its implications for conservation policy. Biological Conservation 120(3): 307–313. https://doi.org/10.1016/j.biocon.2004.03.003

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. https://doi.org/10.1016/j.biocon.2012.04.031

Pirotta V, Grech A, Jonsen ID, Laurance WF, Harcourt RG (2019) Consequences of global shipping traffic for marine giants. Frontiers in Ecology and the Environment 17(1): 39–47. https://doi.org/10.1002/fee.1987

Reijnen R, Foppen R (2006) Impact of road traffic on breeding bird populations. In: Davenport J, Davenport JL (Eds) The Ecology of Transportation: Managing Mobility for the Environment. Springer, Dordrecht, the Netherlands, 255–274. https://doi.org/10.1007/1-4020-4504-2_12

Riitters KH, Wickham JD (2003) How far to the nearest road? Frontiers in Ecology and the Environment 1(3): 125–129. https://doi.org/10.1890/1540-9295(2003)001[0125:HFTTNR]2.0.CO;2

Rubio-Salcedo M, Martínez I, Carreño F, Escudero A (2013) Poor Effectiveness of the Natura 2000 Network Protecting Mediterranean Lichen Species. Journal for Nature Conservation 21(1): 1–9. https://doi.org/10.1016/j.jnc.2012.06.001

Rytwinski T, Fahrig L (2015) The Impacts of Roads and Traffic on Terrestrial Animal Populations. In van der Ree R, Smith DJ, Grilo C (Eds) Handbook of Road Ecology. John Wiley & Sons Ltd, UK, 237–246. https://doi.org/10.1002/9781118568170.ch28

Sánchez-Fernández D, Bilton DT, Abellán P, Ribera I, Velasco J, Millán A (2008) Are the endemic water beetles of the Iberian Peninsula and the Balearic Islands effectively protected? Biological Conservation 141(6): 1612–1627. https://doi.org/10.1016/j.biocon.2008.04.005

Selva N, Kreft S, Kati V, Schluck M, Jonsson B-G, Mihok B, Okarma H, Ibisch PL (2011) Roadless and Low-Traffic Areas as Conservation Targets in Europe. Environmental Management 48(5): 865–877. https://doi.org/10.1007/s00267-011-9751-z

Selva N, Switalski A, Kreft S, Ibisch PL (2015) Why keep areas road-free? The importance of roadless areas. In van der Ree R, Smith DJ, Grilo C (Eds) Handbook of Road Ecology. John Wiley & Sons Ltd, UK, 16–26. https://doi.org/10.1002/9781118568170.ch3

Seshadri KS, Ganesh T (2015) Road ecology in south India: issues and mitigation opportunities. In van der Ree R, Smith DJ, Grilo C (Eds) Handbook of Road Ecology. John Wiley & Sons Ltd, UK, 425429. https://doi.org/10.1002/9781118568170.ch52

Shafer CL (1995) Values and Shortcomings of Small Reserves: Dealing with the smallest habitat fragments when some of them are all that is left. Bioscience 45(2): 80–88. https://doi.org/10.2307/1312609

Shanley CS, Pyare S (2011) Evaluating the road-effect zone on wildlife distribution in a rural landscape. Ecosphere 2(2): 16. https://doi.org/10.1890/ES10-00093.1

Torres A, Jaeger JAG, Alonso JC (2016) Assessing large-scale wildlife responses to human infrastructure development. Proceedings of the National Academy of Sciences of the United States of America 113(30): 8472–8477. https://doi.org/10.1073/pnas.1522488113

Trafikverket (2017) Trafikverkets Miljörapport 2017. Swedish National Transport Administration, Report No 2018: 144. [In Swedish] https://trafikverket.ineko.se/se/trafikverkets-miljörapport-2017

Trochet A, Schmeller DS (2013) Effectiveness of the Natura 2000 network to cover threatened species. Nature Conservation 4: 35–53. https://doi.org/10.3897/natureconservation.4.3626

Tsiafouli MA, Apostolopoulou E, Mazaris AD, Kallimanis AS, Drakou EG, Pantis JD (2013) Human Activities in Natura 2000 Sites: A Highly Diversified Conservation Network. Environmental Management 51(5): 1025–1033. https://doi.org/10.1007/s00267-013-0036-6

Tulloch AIT, Gordon A, Runge CA, Rhodes JR (2019) Integrating spatially realistic infrastructure impacts into conservation planning to inform strategic environmental assessment. Conservation Letters 2019: e12648. https://doi.org/10.1111/conl.12648

van der Ree R, Smith DJ, Grilo C (2015) Handbook of Road Ecology. John Wiley & Sons Ltd, UK, 522 pp. https://doi.org/10.1002/9781118568170.ch1

van Langevelde F, Jaarsma CF (2009) Modeling the effect of traffic calming on local animal population persistence. Ecology and Society 14(2): 39. https://doi.org/10.5751/ES-03061-140239

Votsi N-E, Mazaris AD, Kallimanis AS, Zomeni MS, Vogiatzakis IN, Sgardelis SP, Pantis JD (2012) Road effects on habitat richness of the Greek Natura 2000 network. Nature Conservation 1: 53–71. https://doi.org/10.3897/natureconservation.1.2086

Wärnbäck A (2013) Beaktande av biologisk mångfald och ekosystemtjänster i miljökonsekvensbeskrivningar och miljöbedömningar. Annex 5 to SOU 2013:68, Synliggöra värdet av ekosystemtjänster – Åtgärder för välfärd genom biologisk mångfald och ekosystemtjänster. http://www.regeringen.se/49bba7/contentassets/ba53cd9f18b74f348eb0ff31e8280d60/synliggora-vardet-av-ekosystemtjanster-sou-201368 [In Swedish]