Road effects on bat activity depend on surrounding habitat type
Graphical abstract
Introduction
Roads are widespread features of many countries, and they form dense worldwide networks, such that there are few road-free areas remaining. In Europe, at least a quarter of the continent's surface is located within 500 m of the nearest transport infrastructure (Torres et al., 2016). The road network strongly contributes to landscape fragmentation or degradation by dissecting continuous areas that have high conservation values, and this fragmentation is considered to be a major driver for negative impacts on natural worldwide populations (Fahrig and Rytwinski, 2009; Forman et al., 2003; Grilo et al., 2012; Jaeger et al., 2005). The most perceptible of such impacts is wildlife-vehicle collisions. However, the combination of road avoidance and road barrier effects may contribute to reduced gene flow between roadside populations. Roads also decrease the suitability of adjacent habitats, by altering the physical (e.g., noise and light) and chemical environments. These changes will likely increase the risk of local extinctions (Ascensão et al., 2016; Jackson and Fahrig, 2011; Reed et al., 2007; Westemeier, 1998). Additionally, these negative effects are usually cumulative in the long-term (Balkenhol and Waits, 2009) and can extend their influence to distances of several hundred metres away from roads and into the habitat matrix (e.g., road-effect zone) (Forman et al., 2003). The spatial extension of the impacts of roads on surrounding areas has been studied for many species and can range from <100 m to over 5000 m, depending on the road type, the crossed habitat and the traits of the species (Benítez-López et al., 2010; Bennett et al., 2013; Rotholz and Mandelik, 2013). Currently, most of the studies of road effects on biodiversity focus on high-traffic transport infrastructures, highways and motorways. However, approximately 50% of worldwide roads are low- and medium-traffic roads. Thus, it is surprising that for these types of roads, which comprise the majority of the world's roads, information about their potential impacts on wildlife is still lacking (Clevenger and Waltho, 2000; Dodd et al., 2004; Yanes et al., 1995).
However, roads may also attract wildlife, as they provide a favourable habitat along roadside verges (e.g., shelter and/or movement corridors for a large number of taxa - Davies and Pullin, 2007; Penone et al., 2012) and provide foraging opportunities (e.g., carrion for scavengers – Santos et al., 2016). For some of these species, roads and their verges can become ecological traps because the attracted individuals have a higher risk of becoming road kill (Bernes et al., 2017). Several studies support that the role of roadside verges, either as detractors or attractants, may depend on the surrounding landscape. O'Farrell and Milton (2006) demonstrated that, when crossing high-quality habitats, road verges do not provide habitats for several threatened shrew species. Similarly, Galantinho et al. (2017) found no effects on the density and survival of a forest-dwelling small mammal species (Apodemus sylvaticus) when comparing roadside verges and roadless areas in a Mediterranean forest area, which is a high-quality habitat for this species. Conversely, in intensive agricultural landscapes, roadside verges are often the only remnants of semi-natural habitats and have an important role as corridors or refuges of biodiversity (Penone et al., 2012; de Redon et al., 2015).
A primary concern for biodiversity conservation is to understand the extent to which roads affect the persistence and population of each species (Forman and Alexander, 1998). Rytwinski and Fahrig (2013) showed that species with slow life histories and large home ranges can be particularly affected by roads. For example, bats have very low reproductive rates, can have long lives and need large areas for foraging. Thus, transport infrastructures are likely to induce significant negative impacts on this group of mammals.
Several studies have shown that bats suffer frequent road kills (Gaisler et al., 2009; Lesiński, 2007; Lesiński et al., 2010; Medinas et al., 2013; Russell et al., 2009; Secco et al., 2017), have reduced foraging activity near roads with street lighting (Hale et al., 2015; Stone et al., 2009; Stone et al., 2012) or intense traffic noise (Luo et al., 2015; Schaub et al., 2008; Siemers and Schaub, 2011) and have restricted access to some habitats when these areas are dissected by roads (i.e., barrier effect) (Kerth and Melber, 2009). Other studies have shown that gleaning species are less active and that species richness decreases, when approaching the major roads (Zurcher et al., 2010; Berthinussen and Altringham, 2012; Kitzes and Merenlender, 2014). In a study on railways, Vandevelde et al. (2014) showed that the commuting activity of aerial hawking bats is higher over verges than in the surrounding intensive agricultural habitat. However, because of the very low train traffic at night and the narrow railway corridors, the negative effects of these infrastructures on bats are potentially weaker than those of roads. Thus, bats may show contrasting behaviours, in regard to transport infrastructures, depending on their type, features and surrounding habitats. Their response is also likely to be dependent on the functional groups of the bats (hereafter called guilds).
We investigated the effect of distance to low-medium traffic roads, as well as the importance of road-surrounding habitats, on bat activity patterns in a Mediterranean landscape. It was assumed that bats reduce their activity even in proximity to low-medium traffic roads. We also hypothesized that the characteristics of the surrounding habitats would modulate the road-effect zone, with this zone this being larger in open agricultural areas, when compared with woodlands. Moreover, the road-effect zone should also be guild specific. Short- (SRE) and mid-range (MRE) echolocators, which are more adapted to fly in woodland and vegetated edges (Frey-Ehrenbold et al., 2013), would avoid roads that cross woodlands, but could use vegetated verges of roads that are imbibed in open areas. Additionally, the long-range echolocators (LRE) bats, which generally forage in wide-open spaces, as well as above the tree canopy level, should be less affected. To test these hypotheses, we identified the road-effect zone for the overall bat community, SRE, MRE and LRE bats, as well as for Pipistrellus kuhlii (the most common bat species). We also evaluated whether the road-surrounding habitats (woodlands or open agricultural field) had an effect on the road-effect zone.
Section snippets
Study area and sampling design
This study was conducted in southern Portugal (38°32′24″ to 38°47′33″N; −08°13′33″ to −07°55′45″W) within a landscape dominated by cork (Quercus suber) and holm oak (Quercus rotundifolia) savannah-like woodlands (the Portuguese “montado”), which alternated with arable fields, olive groves and vineyards (Fig. 1). The climate is Mediterranean, with the mean temperature ranging from 5.8 °C to 12.8 °C in the winter season (January), and from 16.3 °C to 30.2 °C in the summer season (July), as well
Results
A total of 29,239 bat passes were recorded from 360 detector-nights (20 transects × 6 locations × 3 nights), which corresponded to over 1080 survey hours. Of these, 24,455 (83.6%) passes could be identified to the species level, while 4158 (14.2%) were assigned to a single-genus complex, 262 (0.1%) were classified as a multi-genus complex and 364 (0.1%) could not be identified (Table A.1).
Bat passes corresponded to eight genera, namely Pipistrellus (N = 23,853; 82.6%), Eptesicus (N = 984;
Habitat effects
We found that there is a strong association between bat activity in the vicinity of roads and habitat type. For the bat guilds and species that was individually analysed, bat activity was much higher in the woodlands. This is consistent with the well-documented high importance of this habitat for European bat species, both as foraging (e.g., Davidson-Watts et al., 2006; Russ and Montgomery, 2002; Vaughan et al., 1997) and roosting areas (e.g., Boughey et al., 2011; Dietz et al., 2009; Russo and
Conclusion
Our study provides the first insights into how low-medium traffic roads affect bat activity. Overall, the results emphasize that road-effects on bats are non-linear and depend on the functional characteristics of the species and the road-habitat contexts. We highlighted that trade-offs that are posed by roads between feeding/commuting opportunities and road kill risk must be taken in account during road planning and in implementation of mitigation measures. This is a priority, and it represents
Acknowledgements
We are indebted to all who kindly provided assistance during field work, and to SS for helpful suggestions on an earlier draft of the manuscript, SB for aiding with call identification. DM was financed by Fundação para a Ciência e Tecnologia (FCT) with doctoral grant SFRH/BD/104861/2014. AMB and HR were supported by FCT and FEDER/COMPETE 2020 through an “Investigador FCT” contract (IF/00266/2013 and IF/04313/2017, respectively) and AMB also supported by an exploratory project (CP1168/CT0001).
References (86)
- et al.
When flyways meet highways – the relative permeability of different motorway crossing sites to functionally diverse bat species
Landsc. Urban Plan.
(2012) - et al.
The impacts of roads and other infrastructure on mammal and bird populations: a meta-analysis
Biol. Conserv.
(2010) - et al.
Effects of landscape-scale broadleaved woodland configuration and extent on roost location for six bat species across the UK
Biol. Conserv.
(2011) - et al.
Habitat use by bats in a Mediterranean ecosystem in Israel — conservation implications
Biol. Conserv.
(1998) - et al.
Differential habitat selection by Pipistrellus pipistrellus and Pipistrellus pygmaeus identifies distinct conservation needs for cryptic species of echolocating bats
Biol. Conserv.
(2006) - et al.
Relative effects of road mortality and decreased connectivity on population genetic diversity
Biol. Conserv.
(2011) - et al.
Predicting when animal populations are at risk from roads: an interactive model of road avoidance behavior
Ecol. Model.
(2005) - et al.
Species-specific barrier effects of a motorway on the habitat use of two threatened forest-living bat species
Biol. Conserv.
(2009) Birds on roadsides verges and the effect of mowing on frequency and distribution
Biol. Conserv.
(1981)- et al.
Ecological niche modelling of three pipistrelle bat species in semiarid Mediterranean landscapes
Acta Oecol.
(2013)