ORIGINAL RESEARCH published: 11 October 2019 doi: 10.3389/fevo.2019.00373 Echolocation and Stratum Preference: Key Trait Correlates of Vulnerability of Insectivorous Bats to Tropical Forest Fragmentation Silvia Fraixedas Núñez 1*† , Adrià López-Baucells 2,3,4*† , Ricardo Rocha 3,4,5 , Fábio Z. Farneda 3,4,6 , Paulo E. D. Bobrowiec 4 , Jorge M. Palmeirim 3 and Christoph F. J. Meyer 3,4,7 1 Institute for Environmental Science and Technology, Autonomous University of Barcelona, Bellaterra, Spain, 2 Granollers Museum of Natural Sciences, Granollers, Spain, 3 Department of Biology, Faculty of Sciences, Centre for Ecology, Evolution and Environmental Changes – cE3c, University of Lisbon, Lisbon, Portugal, 4 Biological Dynamics of Forest Fragments Project, National Institute for Amazonian Research and Smithsonian Tropical Research Institute, Manaus, Brazil, 5 Conservation Science Group, Department of Zoology, University of Cambridge, Cambridge, United Kingdom, 6 Department of Ecology, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil, 7 School of Science, Engineering and Environment, Edited by: University of Salford, Salford, United Kingdom Mark A. Elgar, The University of Melbourne, Australia Habitat loss and fragmentation rank high amongst the most pressing threats to Reviewed by: Leonardo Ancillotto, biodiversity. Understanding how variation in functional traits is associated with species University of Naples Federico II, Italy vulnerability in fragmented landscapes is central to the design of effective conservation Orly Razgour, University of Southampton, strategies. Here, we used a whole-ecosystem ecological experiment in the Central United Kingdom Amazon to investigate which functional traits of aerial-hawking insectivorous bats best *Correspondence: predict their sensitivity to forest fragmentation. During 2014, bats were surveyed using Silvia Fraixedas Núñez passive bat recorders in six continuous forest sites, eight forest fragments, eight fragment silvia.fraixedas99@gmail.com Adrià López-Baucells edges, and eight forest clearings. The interaction between functional traits, environmental adria.baucells@gmail.com characteristics, and species distribution was investigated using a combination of RLQ † These authors have contributed and fourth-corner analyses. Our results showed that echolocation call structure, vertical equally to this work stratification, and wing aspect ratio were the strongest predictors of sensitivity to forest fragmentation. Frequency of maximum energy, body mass, and relative wing Specialty section: This article was submitted to loading did not show any correlation with the environmental variables. Bat species Conservation, with constant-frequency calls were associated with high vegetation density, being a section of the journal more susceptible to forest fragmentation than species with frequency-modulated calls. Frontiers in Ecology and Evolution Vertical stratum preference was also correlated with vegetation structure, indicating that Received: 20 February 2019 Accepted: 17 September 2019 understory species were more sensitive to forest loss than canopy species. Finally, Published: 11 October 2019 species with high aspect ratio wings were linked to forest edges and clearings. Our Citation: findings suggest that species functional traits determine the vulnerability of aerial- Núñez SF, López-Baucells A, Rocha R, Farneda FZ, Bobrowiec PED, hawking insectivorous bats toward fragmentation and, similarly, environmental conditions Palmeirim JM and Meyer CFJ (2019) determine if a species is likely to become locally extinct due to fragmentation. Preserving Echolocation and Stratum Preference: structurally complex forests will be crucial to ensure the long-term persistence of the Key Trait Correlates of Vulnerability of Insectivorous Bats to Tropical most sensitive and vulnerable species of this bat ensemble in fragmented landscapes Forest Fragmentation. across the Neotropics. Front. Ecol. Evol. 7:373. doi: 10.3389/fevo.2019.00373 Keywords: passive recorders, bioacoustics, Chiroptera, deforestation, species traits, wing Frontiers in Ecology and Evolution | www.frontiersin.org 1 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats INTRODUCTION In the Neotropics, aerial-hawking insectivorous bats (hereafter AIB) represent 30-50% of local bat assemblages The exponential increase of the human population and growing (Kalko, 1998; MacSwiney et al., 2008; Estrada-Villegas per capita consumption of resources are causing widespread et al., 2010). Although AIB are usually classified into one habitat loss and degradation, which are threatening the survival ensemble, they exhibit high variability in foraging and behavioral of numerous species worldwide (Dirzo et al., 2014; Laurance strategies. However, AIB so far have rarely been considered et al., 2014). In fact, human-caused environmental changes are in fragmentation studies as most of them relied on mist-net among the most significant and immediate threats to biodiversity captures, a sampling method poorly suited for this group (Ewers and Didham, 2006; Dobrovolski et al., 2011; Haddad (Estrada-Villegas et al., 2010; Silva and Bernard, 2017). Although et al., 2015), and their impacts on biota are predicted to be acoustic methods are the most suitable to sample AIB, the more pronounced in species-rich tropical areas (Malhi et al., costs and time commitment involved in acoustic surveys is 2014). Across the tropics, human-induced habitat modification is still considerable. Moreover, the echolocation calls of many creating increasingly fragmented landscapes composed of smaller tropical species and the variation among them have not yet been and isolated primary habitat patches (Haddad et al., 2015; Taubert adequately documented, and reference call libraries for tropical et al., 2018), usually embedded in a matrix that is often a regions are scarce (MacSwiney et al., 2008; López-Baucells mosaic of different habitats that affect both species diversity et al., 2016). As a result of these limitations, AIB continue to be and composition (Ewers and Didham, 2006; Kupfer et al., 2006; underrepresented in inventories and ecological studies (Cunto Kennedy et al., 2011; Loureiro and Gregorin, 2015). Matrix and Bernard, 2012), and there is a lack of data about their quality and permeability are key determinants of biodiversity vulnerability to habitat fragmentation (but see Estrada-Villegas persistence in degraded landscapes (Ewers and Didham, 2006; et al., 2010; Bader et al., 2015a,b). Moreover, other ecological Kennedy et al., 2011; Mendenhall et al., 2014; Rocha et al., 2018). characteristics like hunting strategies remain poorly studied Matrix permeability refers to the capacity of any landscape to (Marques et al., 2013). Despite the aforementioned limitations, promote movement of animal populations and to sustain key the recent improvement of acoustic recording devices and new ecological services and processes (Ray et al., 2002). In general, data analysis approaches have stimulated the inclusion of AIB in low matrix permeability negatively affects local biodiversity as surveys and monitoring programmes (Browning et al., 2017). it reduces connectivity and increases the effects of isolation of Here, we used an ecosystem-wide fragmentation experiment remnant patches (Ricketts, 2001; Estrada-Villegas et al., 2010). in the Brazilian Amazon to investigate which functional Understanding how habitat alterations affect biodiversity traits are linked to fragmentation sensitivity in AIB. We and ecosystem functioning is crucial for the creation of employed the analytical framework introduced by Dray effective conservation strategies (Cadotte et al., 2011). Since et al. (2014) that integrates RLQ and fourth-corner methods most ecological processes are determined by species functional to examine trait-environment links and their association traits (Cardinale et al., 2012), trait-based approaches have with patterns of species distribution. RLQ is a multivariate lately become popular to predict the anthropogenic effects ordination technique that summarizes the joint structure on ecosystem functioning in fragmented tropical landscapes. among three data tables (R—environmental characteristics, Functional traits are measurable characteristics of organisms L—species abundances, Q—species traits), whereas the fourth- or species (morphological, physiological, phenological, or corner method tests for trait–environment relationships behavioral) that determine their performance, fitness, and (Dray et al., 2014). ecological functions in a particular habitat (Violle et al., 2007). We predicted that: (i) species with high frequency Because some species are more susceptible to habitat loss than calls are more sensitive to fragmentation than those with others (Davies et al., 2000; Henle et al., 2004), identifying which low frequency calls; (ii) the main echolocation call types traits make species more sensitive to habitat fragmentation is (modulated/quasi-constant/constant frequency calls) are essential to minimize local extinctions (Barbaro and Van Halder, related to species-specific fragmentation sensitivity; (iii) 2009; Hanspach et al., 2012). fragmentation filters bat assemblages according to species’ The tropics harbor the greatest diversity of bats (Mickleburgh morphological and flight-related traits such as body size et al., 2002; Altringham, 2011), which fulfill key ecological or wing morphology; and (iv) preferences with regard to functions. They act as pollinators and seed dispersers of a large vertical foraging stratum influence the adaptability of bats number of plant species, promoting the restoration of forest in to deforestation. disturbed areas. Moreover, they are both important prey and biological control agents of animal populations (Kunz et al., MATERIALS AND METHODS 2011). The recent increase in published studies on tropical bats has considerably advanced our knowledge about the effects Study Area of habitat loss and fragmentation on this taxon at different The study was conducted at the Biological Dynamics of Forest ecological levels (Fenton, 1997). However, most studies have Fragments Project (BDFFP), located in the Central Amazon, analyzed biodiversity changes from a taxonomic point of view, approximately 80 km north of Manaus (2◦ 30′ S, 60◦ W), Brazil with few focusing on a functional trait-based dimension (Farneda (Figure 1). Commencing in 1979, the BDFFP is the longest- et al., 2015, 2018a; but see Cisneros et al., 2015; Frank et al., 2017; running experimental study of habitat fragmentation (Laurance Wordley et al., 2017). et al., 2018). The study area covers 1,000 km2 of moist tropical Frontiers in Ecology and Evolution | www.frontiersin.org 2 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats FIGURE 1 | Map of the study area in the central Amazon. Image © DigitalGlobe/CNES/Airbus. Experimental areas within the BDFFP landscape are delineated by a white border (Km 41 and Cabo Frio for continuous forest and Dimona, Colosso, and Porto Alegre for forest fragments); continuous forest is represented in light blue and secondary vegetation matrix in dark blue. The image in the upper right corner shows the location of the study area. forest characterized by a mosaic of primary terra firme rainforest Sound Recordings and primary forest fragments (1, 10, and 100 ha) embedded Acoustic surveys were carried out in eight forest fragments in a matrix of late-stage secondary regrowth (∼30 years old). during the first year after the re-isolation that took place in 2013– In the early 1980s, the region underwent an intense process of 2014 (Rocha et al., 2017c), and in six control sites in continuous forest clearing, resulting in a set of fragments of primary forest primary forest (Figure 1). Automatic bat recorders Song Meter 2 isolated from continuous forest by distances of 80–650 m. Since (SM2Bat+) equipped with omnidirectional ultrasonic SMX-US their creation, these experimental fragments were re-isolated 4 microphones (Wildlife Acoustics, Inc. Massachusetts, USA) were to 5 times by clearing a 100 m-wide strip around each fragment. used to acoustically sample bats. Recorders were placed at 1.5 m The most recent re-isolation occurred in 2013–2014, resulting height, in the interiors (N = 8) and edges (N = 8) of all forest in a landscape comprised of both forest-dominated habitats and fragments, as well as in the adjacent clearings (N = 8) (for more small man-made clearings around the fragments (Rocha et al., information regarding the experimental re-isolation of the forest 2017c). Yearly rainfall fluctuates between 1,900 and 3,500 mm, fragments see Rocha et al., 2017c). With the six recorders placed with a dry season from June to October and a wet season from in the interior of continuous forest we had a total of 30 sampling November to June (Ferreira et al., 2017). Forest canopy height sites. Detectors were installed for three consecutive nights per varies from 30 to 37 m, with some emergent trees reaching 55 m recording site (from 18:00 to 6:00 h), and all sites were sampled (Laurance et al., 2011). twice during both the dry and wet season in 2014. Our experimental design was constrained by the overall landscape context and idiosyncrasies of the BDFFP, especially in Acoustic Analysis terms of the number and locations of fragments and continuous Kaleidoscope Pro Software v4.0.4 (Wildlife Acoustics, In. forest sites, which precluded us from following a paired study Massachusetts, USA) was used to manually classify all design, i.e., pairing fragments with nearby continuous forest sites. recordings based on López-Baucells et al. (2016, 2019) and Although the BDFFP landscape is not ideal in every respect from a regional reference call library compiled by A. López-Baucells an experimental design perspective, environmental conditions (unpublished). In the case of 14 species it was possible to across the study landscape can be considered sufficiently similar identify the calls to species level. However, some species had (Laurance et al., 2018). We therefore believe the influence of to be grouped into sonotypes (comprising several species with potentially confounding environmental variables on our results similar calls that could not be reliably distinguished, N = 5; see to be negligible, especially in view of the high vagility of the Table S2). Owing to inherent constraints of acoustic data (i.e., study taxon. the fact that it is not possible to count individuals), we used “bat Frontiers in Ecology and Evolution | www.frontiersin.org 3 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats TABLE 1 | List of functional traits used as predictors of aerial-hawking insectivorous bat vulnerability to habitat fragmentation in the central Amazon. Trait variable Abbreviation Definition References Frequency of FME Frequency with the highest intensity from each echolocation call López-Baucells et al., 2016 maximum energy Call structure Call_QCF, Call_CF Pulse shape, classified into four main categories: quasi-constant frequency López-Baucells et al., 2016 Call_FM (QCF), constant frequency (CF), frequency-modulated (FM), frequency Call FM-QCF modulated - quasi-constant frequency (FM-QCF) Body mass Weight Average body mass of each species (only adults) Nowalk, 1999; Gardner, Foream length FA Distance between the elbow and the wrist (only adults) 2008; Rocha et al., 2017a Vertical stratification VertS Forest stratum where each species tends to forage; categorization based on Mas, 2014 three levels: understory, canopy, and above the canopy Relative wing RWL Wing loading (weight of an individual multiplied by the gravitational acceleration Norberg and Rayner, 1987; loading divided by wing area) divided by body mass raised to one third (Norberg and Davies et al., 2013; Voigt, Fenton, 1988) 2013; Rocha et al., 2017a Aspect ratio AR Square of the wingspan divided by the wing area (Norberg, 1987) Freeman, 1981; Norberg and Rayner, 1987; Davies et al., 2013; Voigt, 2013; Rocha et al., 2017a activity” as a proxy for bat abundance. The sample unit was a species with similar peak frequencies had to be grouped under bat pass (Wickramasinghe et al., 2003), which we considered as the same sonotype, the average FME of all species was calculated any sequence/recording with a minimum of two distinguishable and used in subsequent analyses. echolocation pulses per species during a maximum time of 5 s (Millon et al., 2015). We only considered those pulses whose Echolocation: Call Structure intensity exceeded 10 dB from the background noise. Data This predictor was based on an existing guild classification from all fragments were pooled together independently of their of bats that links habitat type and foraging mode with their size within the habitat category “fragment,” due to the limited echolocation calls (Schnitzler and Kalko, 2001; Schnitzler et al., number of replicates. 2003). Following this, we established the following categories: (A) frequency-modulated signals (FM): usually emitted by Environmental Variables cluttered space foragers; (B) frequency modulated—quasi- Vegetation structure was assessed at each sampling site within constant frequency signals (FM-QCF): typical of edge-foraging three plots of 100 m2 (5 × 20 m) located next to the detector species; (C) quasi-constant frequency signals (QCF): mostly locations. Four environmental variables were quantified: number emitted by open space foragers; and (D) constant frequency of stems, tree height (height in m of all trees ≥10 cm diameter signals (CF): emitted by cluttered space foragers. Species were at breast height), shortest distance to water (distance in km assigned to each category following López-Baucells et al. (2016). between each site and the nearest water body calculated using Google Earth), and habitat type (qualitative variable including Body Mass four categories: continuous forest interior, fragment interior and Body mass plays a key role in the physiology, ecology, and fragment edge, clearing). diversification of species and also influences life-history traits of individuals (Safi et al., 2013). Species with large body size Species Traits tend to have smaller populations and larger home ranges and The relative scarcity of knowledge on AIB, together with the are therefore more prone to extinction (Purvis et al., 2000). difficulty of gathering reliable information about most of the Body mass was measured in captured bats (data collected at species, led us to not consider traits such as diet or roosting the BDFFP between 2012 and 2014; Rocha et al., 2017b,c) and strategy/guild (Borkin and Parsons, 2011; Bohnenstengel, 2012; averaged for each species (excluding juveniles and reproductive Farneda et al., 2015; Coronado et al., 2018), and to focus instead females). Missing data for Molossops spp., Peropteryx kappleri on only six functional traits as potential predictors of AIB and P. macrotis, Promops spp., and Pteronotus personatus were vulnerability to habitat fragmentation (Table 1): obtained from the literature (Eisenberg and Redford, 1989). Echolocation: Frequency of Maximum Energy Forearm Length Frequency of maximum energy (FME), also known as peak Body size has been widely used in trait-based analyses due to frequency, is the most intense frequency in the call (López- its influence on the physiology, ecology, and morphology of Baucells et al., 2016). Amongst other environmental and acoustic organisms (Peters, 1986; Penone et al., 2018). Larger bats tend factors, it determines how quickly the intensity of a call will to emit lower frequency calls and typically forage in more open diminish and therefore, will influence the kind of habitat in which habitats, while smaller species use higher frequency signals that the bat will be able to fly and forage. Data on FME for all species allow them to move in cluttered spaces (Kalko and Schnitzler, were obtained from López-Baucells et al. (2016). Whenever 1993; Thiagavel et al., 2017; Penone et al., 2018). We used forearm Frontiers in Ecology and Evolution | www.frontiersin.org 4 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats length as a measure of body size. Data was obtained from our own TABLE 2 | Summary of the RLQ analysis. data and the literature (Gardner, 2008). Axis 1 Axis 2 Vertical Stratification Eigenvalues (%) Eigenvalues (%) Bat activity and richness differ among tropical forest strata, whereby species are generally divided into canopy, sub-canopy, (A) or understory specialists. However, this classification is mostly RLQ ordination 0.77 87.61 0.09 10.25 based on studies on phyllostomid bats (Carvalho et al., 2013; Correlation: L 0.38 – 0.14 – Marques, 2016). We assigned species to three different strata Covariance 0.93 – 0.22 – based on Mas (2014), who studied vertical stratification at Projected variance: Q 2.48 5.16 the BDFFP using acoustic data: (1) more recordings in the Projected variance: R 2.11 – 3.83 – understory than in the other strata, (2) bats flying significantly (B) more often between 20 and 30 m but still under the canopy Q table (Hill-Smith PCA) 3.2 35.65 2.55 28.36 (sub-canopy), and (3) bats flying above the canopy. R table (Hill-Smith PCA) 2.89 48.29 1.38 23.09 L table (CA) 0.39 43.11 0.14 15.26 Wing Morphology (A) Eigenvalues and total co-inertia (%) for the first two axes of the RLQ ordination, Maneuverability and flight speed in bats are constrained by projected variance of tables Q and R and correlation and covariance for the L matrix. (B) wing morphology (Marinello and Bernard, 2014), which usually Eigenvalues and total co-inertia (%) for the first two axes of the three matrices Q (functional determines access to suitable foraging habitat (Arita and traits table), R (environmental variables table) and L (species abundance table). Fenton, 1997). We calculated aspect ratio and relative wing loading of each species following Norberg and Rayner (1987). model 4 assumes no association between L and Q and the null Measurements were obtained from extended-wing photographs hypothesis here is that species composition is independent of of bats captured at the BDFFP (Rocha et al., 2017b,c; Torrent species traits (Dray et al., 2014). To adjust p-values for multiple et al., 2018) and analyzed with ImageJ (National Institute of comparisons, we used the false discovery method (Benjamini Health, USA). Missing data were obtained from the literature (see and Hochberg, 1995; Dray et al., 2014), determining significance references in Table 1). based on 50,000 permutations. Given the ecological focus of our study and following the recommendations by De Bello et al. Statistical Analysis (2015) regarding when to apply phylogenetic corrections in Several analytical methods have historically been used to functional trait-based analyses, we decided not to incorporate assess the associations between species traits and environmental phylogenetic relatedness. All calculations were carried out using gradients (Peres-Neto et al., 2017). Amongst them, RLQ and R v3.4.1 (R Core Team, 2017), specifically package ade4 (Dray fourth-corner analysis are considered excellent approaches to and Dufour, 2007). assess trait–environment relationships at the species level and have been widely used in ecological and conservation research RESULTS (Kleyer et al., 2012). Dray et al. (2014) recently developed an analytical framework that combines both methods and The RLQ analysis showed that the first axis of the ordination so far is considered the most appropriate approach to study (considering the three tables R, L, and Q) explained 87.6% of the relation between environmental variables and functional the variance and the second axis 10.2%. The first and second traits. Applied to our dataset, it tests the relation of the data axes of the Hill-Smith PCA of the functional traits table (Q contained in three different matrices: table L (species activity table) accounted for 35.6 and 28.4% of the variance respectively. levels per sampling site), table Q (functional traits of species) Regarding the R table (environmental variables), the first axis of and table R (environmental variables per sampling site). A the PCA explained 48.3% of the variance, and the second axis correspondence analysis (CA) is applied to table L, providing 23.1%. Finally, in the L table (species activity) the first axis of the a common organization of the samples and the species (Dray correspondence analysis (CA) explained 43.1% of the variance et al., 2014). Both the table of environmental variables (R) and the second 15.3% (Table 2). and traits (Q) contained a mix of qualitative and quantitative In the PCA of the R table (Figure 2), sites grouped according variables and were therefore analyzed using Hill-Smith principal to the gradient made up by the different habitat categories, with component analysis (PCA; Hill and Smith, 1976). The RLQ all forest interior sites (continuous forest and forest fragments) method maximizes the covariance between the traits and the well-separated from clearings and fragment edges along the environmental variables through the species relative abundance first axis. In the PCA of the Q ordination table (Figure 3), (Dray et al., 2014). While RLQ consists of a multivariate analysis no clear species groupings could be distinguished. However, of the associations between the three tables, the fourth-corner all species were mostly spatially organized based on three of method tests the significance of bivariate associations. We used their functional traits, call structure (CF or FM calls), forearm two permutation models in the RLQ-fourth corner analysis. length (FA) and frequency of maximum energy (FME). The Model 2 tests the null hypothesis that there is no relationship eigenvalues of the first axis of the RLQ analysis from species between the distribution of species and environmental conditions traits and environmental variables (Figure 4) showed that the (i.e., no association between R and L matrices). In contrast, variables that explained most variability in species occurrence Frontiers in Ecology and Evolution | www.frontiersin.org 5 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats FIGURE 2 | Biplots from the Hill-Smith PCA of the environmental gradients (R table). (A) Environmental variables. (B) Sites. C, clearing; FF, forest fragment; CF, continuous forest; E, edge. among sites were call structure (CF) and vertical stratification, and two categories of the habitat type variable, forest fragment and continuous forest. FIGURE 3 | Biplots from the Hill-Smith PCA of the functional traits (Q table). Fourth corner analysis revealed a highly significant (A) Species. Red, Mormoopidae; Pink, Furipteridae; Green, Vespertilionidae; relationship between species distribution and environmental Orange, Emballonuridae; Blue, Molossidae (see Table S1 for species variables (model 2, p < 0.0001), and a non-significant association abbreviations). (B) Traits (see Table 1 for trait abbreviations). between species composition and functional traits (model 4, p = 0.17). Based on 50,000 iterations, the total inertia of the RLQ analysis indicates a non-significant general relation (p = 0.17) functional traits are also likely to influence the vulnerability between environmental variables and functional traits in the of species to the ongoing process of habitat destruction and combined models 2 and 4. fragmentation. However, trait-mediated environmental filters are Finally, the combination of RLQ and fourth-corner still seldom considered in studies based on human-modified analysis yielded three significant (p < 0.05) trait-environment tropical landscapes. This is especially the case for poorly relationships (Figure S1): CF call structure showed a positive known animal groups, such as AIB. We used autonomous correlation with the first axis (AxR1) for the environmental bat detectors and a modern statistical approach to identify gradient; vertical stratification revealed a negative correlation how the traits of Neotropical aerial-hawking insectivorous bats with the AxR1 axis for the environmental gradient; and the correlate with environmental features and species distributions assemblages recorded in the forest fragments were significantly in a fragmented landscape. Our results revealed that some traits associated with the AxQ1 axis that represented the pool of bat were significantly linked to environmental variables, potentially functional traits. affecting species vulnerability to habitat fragmentation. We found a strong association between echolocation call type with DISCUSSION features of AIB foraging habitat that are likely to be affected by forest fragmentation. The diversity of species functional traits in a given assemblage The responses of phyllostomid bats to habitat modification is a key predictor of an ecosystem’s functioning, resistance, have been extensively studied across the Neotropics (Meyer and resilience (Petchey and Gaston, 2006), and therefore, et al., 2016; Frank et al., 2017) and in particular at the BDFFP, Frontiers in Ecology and Evolution | www.frontiersin.org 6 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats FIGURE 4 | Eigenvalues of species traits and environmental variables along the first axis of the RLQ analysis. See Table 1 for trait abbreviations. where recent long-term studies have provided important insights Froidevaux et al. (2016) studied the relation between bat guilds into the dynamics of bat responses to landscape fragmentation and forest structure through the combination of LiDAR and associated with matrix regeneration (Farneda et al., 2018a,b; acoustic sampling. The authors found that guild-specific activity Rocha et al., 2018). Farneda et al. (2015) found that, in the case was strongly influenced by 3D forest structure, highlighting the of Amazonian phyllostomid bats, body mass and trophic level importance of echolocation in habitat selection. Similarly, Jung are good predictors of fragmentation sensitivity. However, this et al. (2012) found that bat assemblages in Germany were affected issue is poorly studied in AIB, and the present study is one of by canopy structure, height, surface roughness and edge fraction. the few investigating the link between these species’ traits and Bats belonging to the same guild might not use calls with vulnerability to fragmentation. similar frequencies, but unquestionably tend to share similar echolocation call shapes, which are highly related to habitat type and foraging mode (Schnitzler and Kalko, 2001; Denzinger Echolocation Call Type and Schnitzler, 2013). Following a widely used classification of Although bat echolocation calls are specifically adapted to bats into guilds (Schnitzler and Kalko, 2001), we used the term different habitat types (Schnitzler et al., 2003), only a few studies “clutter foragers” for those species in our study that emit CF have analyzed the correlation between echolocation traits and calls (e.g., Pteronotus alitonus and Pteronotus cf. rubiginosus). habitat characteristics (but see Wordley et al., 2017). These They emit pulses of long duration that allow them to obtain authors linked echolocation and morphological traits to habitat accurate information about the environment, and detect small, use in the Paleotropics and found that functional traits filtered mobile prey in dense vegetation (Barclay, 1999; Barclay et al., bat assemblages in tea plantations. 1999; López-Baucells et al., 2017; Pavan and Marroig, 2017). A recent study tested the relation between echolocation traits These species might be more susceptible to the landscape-wide and forest structure as quantified through LiDAR technology increase in open areas and the reduction of vegetation density (Blakey et al., 2017). In accordance with our results, the authors associated with forest fragmentation. Our results show that CF did not find a significant correlation between forest structure species are more abundant in forest interiors than in clearings variables and frequency of maximum energy. Similarly, Frank or fragment edges, regardless of whether the area was continuous et al. (2017) found that this echolocation parameter was not forest or a fragment. This suggests that vegetation cover is a major predictive of a bat’s habitat usage. Although Wordley et al. (2017) determinant of their habitat use and implies a capacity for these found bats with highest FME to be mostly affiliated with forests, species to persist even in relatively small forest patches. Wordley they suggest that this variable might be less useful than others et al. (2017) obtained similar results in India, with bats using to predict bat vulnerability as bat call libraries in the tropics constant frequency calls being much more common in protected are not comprehensive. The consistency between our results and forests than in altered habitats. A better knowledge of the extent findings from similar studies (Blakey et al., 2016; Frank et al., to which these species can adapt to habitat fragmentation and 2017), indicates that this trait overall does not seem to be a key modification, or which kinds of remnant habitats offer suitable predictor of bat vulnerability to habitat modification. environmental conditions for their long-term persistence is Frontiers in Ecology and Evolution | www.frontiersin.org 7 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats essential to improve their conservation. Larger distances between and this may also be true for AIB. In fact, insectivorous bats remnant patches in landscapes that are heavily fragmented would have distinct diets that include a broad diversity of insect likely constitute a substantial barrier for these species, reducing groups or adjust their activity patterns to new environmental their survival capacity (Ethier and Fahrig, 2011). conditions (Kalka and Kalko, 2006; Lawer and Darkoh, 2016; Kemp et al., 2019; Rocha et al., 2019) and there is evidence that Vertical Stratification some of these are substantially impacted by forest fragmentation We found a strong link between the preferred stratum in which (Didham et al., 1996; Golden and Crist, 1999; Rösch et al., 2013; each bat species tends to forage and the vegetation structure of Benítez-Malvido et al., 2016). Diet type may thus influence the the forest. Primary tropical forests are characterized by marked vulnerability of AIB to fragmentation, but we could not include vertical complexity that creates a pattern of differential use of this trait in our study due to the scarcity of knowledge about space by organisms (Bernard, 2001; Marques et al., 2015). Bats the diet of the species involved. Another potentially critical trait move through space three-dimensionally, and as a consequence, for bat species that should be included in further studies is forest structure and vegetation clutter may affect not only roosting preference. Some bat species require mature forests for horizontal patterns of bat activity but also vertical patterns (Hayes roosting and therefore will be highly vulnerable to habitat loss, and Gruver, 2000; Jung et al., 2012; Mas, 2014). In fact, most of fragmentation and degradation (Voss et al., 2016). In this context, the foraging activity of bats occurs in the canopy, the upper forest it is important to improve our understanding of the natural layer (Kalko and Handley, 2001; Vetter et al., 2011; Marques history of AIB, to allow studies that include a more complete set et al., 2015; Marques, 2016). Deforestation may not have a large of predictors that potentially influence the vulnerability of species influence on bats foraging in open spaces because they mostly fly to fragmentation. above the forest canopy (Marques et al., 2015). However, for bats that typically forage in cluttered habitats, a reduction of tree cover might force these species to abandon the area due to the lack of CONCLUSIONS suitable foraging habitat. We found that a combination of species functional traits and Caveats and Future Research Directions environmental characteristics were relevant predictors of the The fact that we pooled the data from the different fragment sizes sensitivity of AIB toward fragmentation, a finding that may apply (due to the limited number of replicates per size class dictated across many human-modified Neotropical landscapes. In this by the BDFFP experimental design) might have slightly biased context, the type of echolocation calls used by the species proved the results due to differential responses of some AIB toward to be particularly important. Species with QCF echolocation fragments of different sizes and associated edge effects (López- calls, adapted to open habitats or forest edges are likely to be Baucells, 2018). However, because of the complexity of the less affected by forest fragmentation. In contrast, species with statistical approach employed here, we considered it preferable CF calls, typically understory foragers, with low wing loading not to add another layer of complexity by taking fragment size and aspect ratio, are potentially more vulnerable. The latter are into account, which would have made our results not only thus potentially most threatened by deforestation, although the statistically less robust but also more difficult to interpret. fact that they persist even in fragments as small as 1 ha is Patterns of habitat use in AIB are generally influenced by two encouraging for those working to prevent local extinctions of main functional traits, wing morphology and echolocation call these species. Promoting natural forest regeneration and human- structure (Almeida et al., 2014; Marques et al., 2015). Bats with assisted restoration seem to be promising management strategies high wing loading and aspect ratio (long and pointed wings) tend in the Neotropics for maintaining high taxonomic and functional to emit quasi-constant frequency calls, suitable to detect prey at bat diversity at the landscape-scale (Farneda et al., 2018a,b, long distances in open habitats. In contrast, bats with low wing 2019; Rocha et al., 2018). Our results indicate that preserving loading and aspect ratio (short and rounded wings) emit either the structural complexity of tropical forests is important to modulated or constant frequency calls, characterized by a shorter facilitate the persistence of the most fragmentation-sensitive range of detection, more suitable for cluttered habitats (Aldridge species and to avoid general functional trait loss and associated and Rautenbach, 1987; Emrich et al., 2014). This intricate relation loss in ecosystem services provided by bats in human-modified among wing morphology, echolocation call design, habitat type tropical landscapes. and feeding ecology hampers the clear separation of the influence of each functional trait on fragmentation vulnerability. Studies on tropical bats that have tried to unravel the relation between DATA AVAILABILITY STATEMENT functional traits and environmental gradients so far mostly The datasets generated for this study are available on request to targeted phyllostomid species, therefore neglecting the acoustic the corresponding author. component (Meyer et al., 2008; Farneda et al., 2015; Garcia- Morales et al., 2016). More research is still needed to reveal to which extent and under which circumstances a particular trait is AUTHOR CONTRIBUTIONS more important than another. Farneda et al. (2015) showed that diet type is an important All authors listed have made a substantial, direct and intellectual predictor of vulnerability to fragmentation in phyllostomids, contribution to the work, and approved it for publication. Frontiers in Ecology and Evolution | www.frontiersin.org 8 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats FUNDING contribution to figure design and to Sara Fraixedas and Álvaro Fernández-Llamazares for help with proof reading. We further Funding was provided by the Portuguese Foundation for Science acknowledge comments by two reviewers which helped improve and Technology to CM (PTDC/BIA-BIC/111184/2009), RR the manuscript. Research was conducted under permit 26877- (SFRH/BD/80488/2011), and AL-B (PD/BD/52597/2014). FF and 2 issued by the Instituto Chico Mendes de Conservação da PB was supported by a CAPES fellowship (8888.370067/2019- Biodiversidade (ICMBio). 00). This is publication 769 in the Technical Series of the BDFFP. ACKNOWLEDGMENTS SUPPLEMENTARY MATERIAL We would like to thank the volunteers and field assistants The Supplementary Material for this article can be found that participated in data collection as well as the BDFFP for online at: https://www.frontiersin.org/articles/10.3389/fevo. logistical support. We are also thankful to Isabel Núñez for her 2019.00373/full#supplementary-material REFERENCES Borkin, K. M., and Parsons, S. (2011). Sex-specific roost selection by bats in clearfell harvested plantation forest: improved knowledge advises management. Acta Aldridge, H., and Rautenbach, I. L. (1987). Morphology, echolocation and resource Chiropterol. 13, 373–383. doi: 10.3161/150811011X624848 partitioning in insectivorous bats. J. Anim. Ecol. 56, 763–778. doi: 10.2307/4947 Browning, E., Gibb, R., Glover-Kapfer, P., and Jones, K. (2017). Passive Acoustic Almeida, M. H., Ditchfield, A. D., and Tokumaru, R. S. (2014). Habitat Monitoring in Ecology and Conservation. Woking: WWF. characteristics and insectivorous bat activity. Chiroptera Neotropical 20, Cadotte, M. W., Carscadden, K., and Mirotchnick, N. (2011). Beyond species: 1264–1270. functional diversity and the maintenance of ecological processes and services. Altringham, J. (2011). Bats. From Evolution to Conservation, 2nd Edn. New York, J. Appl. Ecol. 48, 1079–1087. doi: 10.1111/j.1365-2664.2011.02048.x NY: Oxford University Press. Cardinale, B. J., Duffy, J. E., Gonzalez, A., Hooper, D. U., Perrings, C., Venail, Arita, H. T., and Fenton, M. B. (1997). Flight and echolocation in P., et al. (2012). Corrigendum: biodiversity loss and its impact on humanity. the ecology and evolution of bats. Trends Ecol. Evol. 12, 53–58. Nature 489, 326–326. doi: 10.1038/nature11373 doi: 10.1016/S0169-5347(96)10058-6 Carvalho, F., Fabián, M. E., and Menegheti, J. O. (2013). Vertical structure of an Bader, E., Acácio, M., and Monadjem, A. (2015a). The importance of water bodies assemblage of bats (Mammalia: Chiroptera) in a fragment of Atlantic Forest in for insectivorous bats in a Malagasy dry deciduous forest: a case example from Southern Brazil. Zoologia 30, 491–498. doi: 10.1590/S1984-46702013000500004 Kirindy (Cnferef). Malagasy Nat. 9, 88–96. Cisneros, L. M., Fagan, M. E., and Willig, M. R. (2015). Effects of human- Bader, E., Jung, K., Kalko, E. K. V., Page, R. A., Rodriguez, R., and Sattler, modified landscapes on taxonomic, functional and phylogenetic dimensions of T. (2015b). Mobility explains the response of aerial insectivorous bats to bat biodiversity. Div. Distrib. 21, 523–533. doi: 10.1111/ddi.12277 anthropogenic habitat change in the Neotropics. Biol. Conserv. 186, 97–106. Coronado, A., Flaquer, C., Puig-Montserrat, X., Barthe, E., Mas, M., Arrizabalaga, doi: 10.1016/j.biocon.2015.02.028 A., et al. (2018). The role of secondary trees in Mediterranean mature forests for Barbaro, L., and Van Halder, I. (2009). Linking bird, carabid beetle and butterfly the conservation of the forest-dwelling bat Myotis alcathoe. Are current logging life-history traits to habitat fragmentation in mosaic landscapes. Ecography 32, guidelines appropriate? Hystrix 28, 240–246. doi: 10.4404/hystrix-00004-2017 321–333. doi: 10.1111/j.1600-0587.2008.05546.x Cunto, G. C., and Bernard, E. (2012). Neotropical bats as indicators of Barclay, R. M. R. (1999). Bats are not birds–a cautionary note on using environmental disturbance: what is the emerging message? Acta Chiropterol. echolocation calls to identify bats: a comment. J. Mammal. 80, 290–296. 14, 143–151. doi: 10.3161/150811012X654358 doi: 10.2307/1383229 Davies, K. F., Margules, C. R., and Lawrence, J. F. (2000). Which traits of species Barclay, R. M. R., Fullard, J. H., and Jacobs, D. S. (1999). Variation in the predict population declines in experimental forest fragments? Ecology 81, echolocation calls of the hoary bat (Lasiurus cinereus): influence of body 1450–1461. doi: 10.1890/0012-9658(2000)081[1450:WTOSPP]2.0.CO;2 size, habitat structure, and geographic location. Canad. J. Zool. 77, 530–534. Davies, K. T., Bates, P. J., Maryanto, I., Cotton, J. A., and Rossiter, S. J. doi: 10.1139/z99-008 (2013). The evolution of bat vestibular systems in the face of potential Benítez-Malvido, J., Dáttilo, W., Martínez-Falcón, A. P., Durán-Barrón, C., antagonistic selection pressures for flight and echolocation. PLoS ONE. Valenzuela, J., López, S., et al. (2016). The multiple impacts of tropical forest 8:e61998. doi: 10.1371/journal.pone.0061998 fragmentation on arthropod biodiversity and on their patterns of interactions De Bello, F., Berg, M. P., Dias, A. T. C., Diniz-Filho, J. A. F., Gotzenberger, with host plants. PLoS ONE 11:e0146461. doi: 10.1371/journal.pone.0146461 L., Hortal, J., et al. (2015). On the need for phylogenetic ‘corrections’ Benjamini, Y., and Hochberg, Y. (1995). Controlling the false discovery rate: a in functional trait-based approaches. Folia Geobot. 50, 349–357. practical and powerful approach to multiple testing. J. R. Stat. Soc. B Methodol. doi: 10.1007/s12224-015-9228-6 57, 289–300. doi: 10.1111/j.2517-6161.1995.tb02031.x Denzinger, A., and Schnitzler, U. (2013). Bat guilds, a concept to classify the highly Bernard, E. (2001). Vertical stratification of bat communities in primary diverse foraging and echolocation behaviors of microchiropteran bats. Front. forests of Central Amazon, Brazil. J. Trop. Ecol. 17, 115–126. Physiol. 4:164. doi: 10.3389/fphys.2013.00164 doi: 10.1017/S0266467401001079 Didham, R. K., Ghazoul, J., Stork, N. E., and Davis, A. J. (1996). Insects in Blakey, R. V., Law, B. S., Kingsford, R. T., and Stoklosa, J. (2017). Terrestrial fragmented forests: a functional approach. Trends Ecol. Evol. 11, 255–260. laser scanning reveals below-canopy bat trait relationships with forest structure. doi: 10.1016/0169-5347(96)20047-3 Remote Sens. Environ. 198, 40–51. doi: 10.1016/j.rse.2017.05.038 Dirzo, R., Young, H. S., Galetti, M., Ceballos, G., Isaac, N. J. B., and Blakey, R. V., Law, B. S., Kingsford, R. T., Stoklosa, J., Tap, P., and Williamson, Collen, B. (2014). Defaunation in the anthropocene. Science 345, 401–406. K. (2016). Bat communities respond positively to large-scale thinning of forest doi: 10.1126/science.1251817 regrowth. J. Appl. Ecol. 53, 1694–1703. doi: 10.1111/1365-2664.12691 Dobrovolski, R., Diniz-Filho, J. A. F., Loyola, R. D., and De Marco, P. (2011). Bohnenstengel, T. (2012). Roost selection by the forest-dwelling bat Myotis Agricultural expansion and the fate of global conservation priorities. Biodiv. bechsteinii (Mammalia: Chiroptera): implications for its conservation Conserv. 20, 2445–2459. doi: 10.1007/s10531-011-9997-z in managed woodlands. Bull. Soc. Neuchl. Sci. Nat. 132, 47–62. Dray, S., Choler, P., Dolédec, S., Peres-Neto, P. R., Thuiller, W., Pavoine, doi: 10.5169/seals-309720 S., et al. (2014). Combining the fourth-corner and the RLQ methods for Frontiers in Ecology and Evolution | www.frontiersin.org 9 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats assessing trait responses to environmental variation. Ecology 95, 14–21. Hayes, J. P., and Gruver, J. C. (2000). Vertical stratification of bat activity doi: 10.1890/13-0196.1 in an old-growth forest in western Washington. Northw. Sci. 74, 102–108. Dray, S., and Dufour, A. B. (2007). The ade4 package: implementing the duality doi: 10.1046/j.1365-294x.2000.00874-7.x diagram for ecologists. J. Stat. Softw. 22, 1–20. doi: 10.18637/jss.v022.i04 Henle, K., Davies, K. F., Kleyer, M., Margules, C., and Settele, J. (2004). Predictors Eisenberg, J. F., and Redford, K. H. (1989). Mammals of the Neotropics. Chicago, of species sensitivity to fragmentation. Biodivers. Conserv. 13, 207–251. IL: Ecuador. doi: 10.1023/B:BIOC.0000004319.91643.9e Emrich, M. A., Clare, E. L., Symondson, W. O., Koenig, S. E., and Fenton, M. B. Hill, M. O., and Smith, A. J. E. (1976). Principal component analysis of (2014). Resource partitioning by insectivorous bats in Jamaica. Mol. Ecol. 23, taxonomic data with multi-state discrete characters. Taxon 25, 249–255. 3648–3656. doi: 10.1111/mec.12504 doi: 10.2307/1219449 Estrada-Villegas, S., Meyer, C. F., and Kalko, E. K. (2010). Effects of tropical forest Jung, K., Kaiser, S., Böhm, S., Nieschulze, J., and Kalko, E. K. V. (2012). Moving in fragmentation on aerial insectivorous bats in a land-bridge island system. Biol. three dimensions: effects of structural complexity on occurrence and activity Conserv. 143, 597–608. doi: 10.1016/j.biocon.2009.11.009 of insectivorous bats in managed forest stands. J. Appl. Ecol. 49, 523–531. Ethier, K., and Fahrig, L. (2011). Positive effects of forest fragmentation, doi: 10.1111/j.1365-2664.2012.02116.x independent of forest amount, on bat abundance in eastern Ontario, Canada. Kalka, M., and Kalko, E. K. V. (2006). Gleaning bats as underestimated predators of Landsc. Ecol. 26, 865–876. doi: 10.1007/s10980-011-9614-2 herbivorous insects: diet of Micronycteris microtis (Phyllostomidae) in Panama. Ewers, R. M., and Didham, R. K. (2006). Confounding factors in the detection J. Trop. Ecol. 22, 1–10. doi: 10.1017/S0266467405002920 of species responses to habitat fragmentation. Biol. Rev. 81, 117–142. Kalko, E. (1998). Organisation and diversity of tropical bat communities through doi: 10.1017/S1464793105006949 space and time. Zoology 101, 281–297. Farneda, F. Z., Grelle, C. E. V., Rocha, R., Ferreira, D. F., López-Baucells, A., and Kalko, E. K., and Schnitzler, V. (1993). Plasticity in echolocation signals of Meyer, C. F. J. (2019). Predicting biodiversity loss in island and countryside European pipistrelle bats in search flight: implications for habitat use and prey ecosystems through the lens of taxonomic and functional biogeography. detection. Behav. Ecol. Sociobiol. 33, 415–428. doi: 10.1007/BF00170257 Ecography. 1–10. doi: 10.1111/ecog.04507 Kalko, E. K. V., and Handley, C. O. (2001). Neotropical bats in the canopy: Farneda, F. Z., Rocha, R., López-Baucells, A., Groenenberg, M., Silva, I., Palmeirim, diversity, community structure, and implications for conservation. Plant Ecol. J. M., et al. (2015). Trait-related responses to habitat fragmentation in 153, 319–333. doi: 10.1023/A:1017590007861 Amazonian bats. J. Appl. Ecol. 52, 1381–1391. doi: 10.1111/1365-2664.12490 Kemp, J., López-Baucells, A., Rocha, R., Wangensteen, O. S., Andriatafika, Z., Farneda, F. Z., Rocha, R., López-Baucells, A., Sampaio, E. M., Palmeirim, J. M., Nair, A., et al. (2019). Bats as potential suppressors of multiple agricultural Bobrowiec, P. E. D., et al. (2018a). Functional recovery of Amazonian bat pests: a case study from madagascar. Agric. Ecosyst. Environ. 269, 88–96 assemblages following secondary forest succession. Biol. Conserv. 218, 192–199. doi: 10.1016/j.agee.2018.09.027 doi: 10.1016/j.biocon.2017.12.036 Kennedy, C. M., Grant, E. H. C., Neel, M. C., Fagan, W. F., and Marra, P. P. Farneda, F. Z., Rocha, R., López-Baucells, A., Sampaio, E. M., Palmeirim, J. M., (2011). Landscape matrix mediates occupancy dynamics of Neotropical avian Bobrowiec, P. E. D., et al. (2018b). The road to functional recovery: temporal insectivores. Ecol. Appl. 21, 1837–1850. doi: 10.1890/10-1044.1 effects of matrix regeneration on Amazonian bats. Trop. Conserv. Sci. 11, 1–4. Kleyer, M., Dray, S., Bello, F., Lepš, J., Pakeman, R. J., Strauss, B., doi: 10.1177/1940082918777185 et al. (2012). Assessing species and community functional responses to Fenton, M. B. (1997). Science and the conservation of bats. J. Mammal. 78, 1–14. environmental gradients: which multivariate methods? J. Veget. Sci. 23, doi: 10.2307/1382633 805–821. doi: 10.1111/j.1654-1103.2012.01402.x Ferreira, D. F., Rocha, R., López-Baucells, A., Farneda, F. Z., Carreiras, J. M. B., Kunz, T. H., de Torrez, E. B., Bauer, D., Lobova, T., and Fleming, T. H. (2011). Palmeirim, J. M., et al. (2017). Season-modulated responses of Neotropical bats Ecosystem services provided by bats. Ann. N. Y. Acad. Sci. 1223, 1–38. to forest fragmentation. Ecol. Evol. 7, 4059–4071. doi: 10.1002/ece3.3005 doi: 10.1111/j.1749-6632.2011.06004.x Frank, H. K., Frishkoff, L. O., Mendenhall, C. D., Daily, G. C., and Hadly, E. A. Kupfer, J. A., Malanson, G. P., and Franklin, S. B. (2006). Not seeing (2017). Phylogeny, traits, and biodiversity of a Neotropical bat assemblage: the ocean for the islands: the mediating influence of matrix-based close relatives show similar responses to local deforestation. Am. Nat. 190, processes on forest fragmentation effects. Glob. Ecol. Biogeogr. 15, 8–20. 200–212. doi: 10.1086/692534 doi: 10.1111/j.1466-822X.2006.00204.x Freeman, P. W. (1981). A Multivariate Study of the Family Molossidae (Mammalia, Laurance, W. F., Camargo, J. L. C., Fearnside, P. M., Lovejoy, T. E., Williamson, G. Chiroptera): Morphology, Ecology, Evolution. Field Museum of Natural History B., Mesquita, R. C. G., et al. (2018). An Amazonian rainforest and its fragments Lincoln. doi: 10.5962/bhl.title.3128 as a laboratory of global change. Biol. Rev. Cam. Philos. Soc. 93, 223–247. Froidevaux, J. S. P., Zellweger, F., Bollmann, K., Jones, G., and Obrist, M. K. doi: 10.1111/brv.12343 (2016). From field surveys to LiDAR: shining a light on how bats respond Laurance, W. F., Camargo, J. L. C., Luizao, R. C. C., Laurance, S. G., Pimm, S. L., to forest structure. Rem. Sens. Environ. 175, 242–250. doi: 10.1016/j.rse.2015. Bruna, E. M., et al. (2011). The fate of Amazonian forest fragments: a 32-year 12.038 investigation. Biol. Conserv. 144, 56–67. doi: 10.1016/j.biocon.2010.09.021 Garcia-Morales, R., Moreno, C. E., Badano, E. I., Zuria, I., Galindo- Laurance, W. F., Sayer, J., and Cassman, K. G. (2014). Agricultural expansion Gonzalez, J., Rojas-Martinez, A. E., et al. (2016). Deforestation impacts and its impacts on tropical nature. Trends Ecol. Evol. 29, 107–116. on bat functional diversity in tropical landscapes. PLoS ONE 11:e0166765. doi: 10.1016/j.tree.2013.12.001 doi: 10.1371/journal.pone.0166765 Lawer, E. A., and Darkoh, E. L. (2016). Effects of agroecosystems on insect and Gardner, A. L. (2008). Mammals of South America, Volume 1: Marsupials, insectivorous bat activity: a preliminary finding based on light trap and mist Xenarthrans, Shrews, and Bats. Chicago, IL: University of Chicago Press. net captures. Turkish J. Zool. 40, 423–432. doi: 10.3906/zoo-1507-17 doi: 10.7208/chicago/9780226282428.001.0001 López-Baucells, A. (2018). Assessment of the Effects of Forest Fragmentation Golden, D. M., and Crist, T. O. (1999). Experimental effects of habitat on Aerial Insectivorous Bats in the Amazonian Rainforest. University of fragmentation on old-field canopy insects: community, guild and Lisbon (Manaus: Portugal). species responses. Oecologia 118, 371–380. doi: 10.1007/s0044200 López-Baucells, A., Rocha, R., Bobrowiec, P. E. D., Bernard, E., Palmeirim, J., and 50738 Meyer, C. (2016). Field Guide to Amazonian Bats. Manaus: INPA Haddad, N. M., Brudvig, L. A., Clobert, J., Davies, K. F., Gonzalez, A., Holt, R. D., López-Baucells, A., Torrent, L., Rocha, R., Bobrowiec, P. E., Palmeirim, J. M., et al. (2015). Habitat fragmentation and its lasting impact on Earth’s ecosystems. and Meyer, C. F. (2019). Stronger together: combining automated classifiers Sci. Adv. 1:e1500052. doi: 10.1126/sciadv.1500052 with manual post-validation optimizes the workload vs reliability trade-off Hanspach, J., Fischer, J., Ikin, K., Stott, J., and Law, B. S. (2012). Using trait- of species identification in bat acoustic surveys. Ecol. Inform. 49, 45–53. based filtering as a predictive framework for conservation: a case study doi: 10.1016/j.ecoinf.2018.11.004 of bats on farms in southeastern Australia. J. Appl. Ecol. 49, 842–850. López-Baucells, A., Torrent, L., Rocha, R., Pavan, A. C., Bobrowiec, P. E. D., Meyer, doi: 10.1111/j.1365-2664.2012.02159.x C. F., et al. (2017). Geographical variation in the high-duty cycle echolocation of Frontiers in Ecology and Evolution | www.frontiersin.org 10 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats the cryptic common mustached bat Pteronotus cf. rubiginosus (Mormoopidae). Peters, R. H. (1986). The Ecological Implications of Body Size, Vol. 26. Cambridge: Bioacoustics J. 27, 341–357. doi: 10.1080/09524622.2017.1357145 Cambridge University Press. Loureiro, L. O., and Gregorin, R. (2015). Structure of a bat assemblage from Purvis, A., Gittleman, J. L., Cowlishaw, G., and Mace, G. M. (2000). Predicting a fragmented landscape in the state of Minas Gerais, southeastern Brazil. extinction risk in declining species. Proc. R. Soc. Lond. B Biol. Sci. 267, Mastozool. Neotropical. 22, 35–42. 1947–1952. doi: 10.1098/rspb,.2000.1234 MacSwiney, G. C., Clarke, F. M., and Racey, P. A. (2008). What you see is R Core Team (2017). R: A Language and Environment for Statistical Computing. R not what you get: the role of ultrasonic detectors in increasing inventory Foundation for Statistical Computing, Vienna. completeness in Neotropical bat assemblages. J. Appl. Ecol. 45, 1364–1371. Ray, N., Lehmann, A., and Joly, P. (2002). Modeling spatial distribution doi: 10.1111/j.1365-2664.2008.01531.x of amphibian populations: a GIS approach based on habitat matrix Malhi, Y., Gardner, T. A., Goldsmith, G. R., Silman, M. R., and Zelazowski, permeability. Biodivers. Conserv. 11, 2143–2165. doi: 10.1023/A:10213905 P. (2014). Tropical 0. Ann. Rev. Environ. Resour. 39, 125–159. 27698 doi: 10.1146/annurev-environ-030713-155141 Ricketts, T. H. (2001). The matrix matters: effective isolation in fragmented Marinello, M. M., and Bernard, E. (2014). Wing morphology of Neotropical bats: landscapes. Am. Nat. 158, 87–99. doi: 10.1086/320863 a quantitative and qualitative analysis with implications for habitat use. Can. J. Rocha, R., Ferreira, D. F., López-Baucells, A., Farneda, F. Z., Carreiras, Zool. 92, 141–147. doi: 10.1139/cjz-2013-0127 J., Palmeirim, J. M., et al. (2017a). Does sex matter? Gender-specific Marques, J. T. (2016). Structuring Amazonian Bat Assemblages: Importance of responses to forest fragmentation in Neotropical bats. Biotropica 49, 881–890. Horizontal and Vertical Dimensions of Habitat. Lisbon: University of Lisbon. doi: 10.1111/btp.12474 Marques, J. T., Ramos Pereira, M. J., Marques, T. A., Santos, C. D., Santana, Rocha, R., López-Baucells, A., Farneda, F., Ferreira, D., Silva, I., Acácio, M., et J., Beja, P., et al. (2013). Optimizing sampling design to deal with al. (2019). Second-growth and small forest clearings have little effect on the mist-net avoidance in Amazonian birds and bats. PLoS ONE 8:e74505. temporal activity patterns of Amazonian phyllostomid bats. Curr. Zool. 1, 1–9. doi: 10.1371/journal.pone.0074505 doi: 10.1093/cz/zoz042/5561099 Marques, J. T., Ramos Pereira, M. J., and Palmeirim, J. M. (2015). Patterns in Rocha, R., López-Baucells, A., Farneda, F. Z., Groenenberg, M., Bobrowiec, the use of rainforest vertical space by Neotropical aerial insectivorous bats: P. E. D., Cabeza, M., et al. (2017b). Consequences of a large-scale all the action is up in the canopy. Ecography 39, 476–486. doi: 10.1111/ecog. fragmentation experiment for Neotropical bats: disentangling the relative 01453 importance of local and landscape-scale effects. Landsc. Ecol. 32, 31–45. Mas, M. (2014). Vertical stratification on insectivorous bats ensembles in central doi: 10.1007/s10980-016-0425-3 Amazon (MSc thesis). Universitat de Barcelona, Barcelona, Spain. Rocha, R., Ovaskainen, O., López-Baucells, A., Farneda, F. Z., Ferreira, D. Mendenhall, C. D., Karp, D. S., Meyer, C. F. J., Hadly, E. A., and Daily, G. C. (2014). F., Bobrowiec, P. E. D., et al. (2017c). Design matters: an evaluation Predicting biodiversity change and averting collapse in agricultural landscapes. of the impact of small man-made forest clearings on tropical bats Nature 509, 213–217. doi: 10.1038/nature13139 using a before-after-control-impact design. Forest Ecol. Manage. 401, 8–16. Meyer, C. F. J., Fruend, J., Lizano, W. P., and Kalko, E. K. V. (2008). doi: 10.1016/j.foreco.2017.06.053 Ecological correlates of vulnerability to fragmentation in Neotropical Rocha, R., Ovaskainen, O., López-Baucells, A., Farneda, F. Z., Sampaio, E. M., bats. J. Appl. Ecol. 45, 381–391. doi: 10.1111/j.1365-2664.2007. Bobrowiec, P. E. D., et al. (2018). Secondary forest regeneration benefits old- 01389.x growth specialist bats in a fragmented tropical landscape. Sci. Rep. 8:3819. Meyer, C. F. J., Struebig, M. J., and Willig, M. R. (2016). “Responses of tropical doi: 10.1038/s41598-018-21999-2 bats to habitat fragmentation, logging, and deforestation,” in Bats in the Rösch, V., Tscharntke, T., Scherber, C., and Batary, P. (2013). Landscape Anthropocene: Conservation of Bats in a Changing World, eds C. C. Voigt and composition, connectivity and fragment size drive effects of grassland T. Kingston (New York, NY: Springer), 63–103. fragmentation on insect communities. J. Appl. Ecol. 50, 387–394. Mickleburgh, S. P., Hutson, A. M., and Racey, P. A. (2002). A review of the global doi: 10.1111/1365-2664.12056 conservation status of bats. Oryx 36, 18–34. doi: 10.1017/S0030605302000054 Safi, K., Meiri, S., and Jones, K. E. (2013). “Evolution of body size in bats,” in Millon, L. J.-,Julien, F., Julliard, R., and Kerbiriou, C. (2015). Bat Animal body size: Linking pattern and process across space, time, and taxonomic activity in intensively farmed landscapes with wind turbines and group, eds F. A. Smith and S. K. Lyons (Chicago, IL: University of Chicago offset measures. Ecol. Eng. 75, 250–257. doi: 10.1016/j.ecoleng.2014. Press), 95–151. 11.050 Schnitzler, H. U., and Kalko, E. K. V. (2001). Echolocation by insect-eating bats: Norberg, U. M. (1987). “Wing form and flight mode in bats,” in Recent Advances in we define four distinct functional groups of bats and find differences in signal the Study of Bats, eds M. B. Fenton, P. A. Racey, and J. M. V. Rayner (Cambridge structure that correlate with the typical echolocation tasks faced by each group. University Press), 43–56. Bioscience 51, 557–569. doi: 10.1641/0006-3568(2001)051[0557:EBIEB]2.0. Norberg, U. M., and Fenton, M. B. (1988). Carnivorous bats? Biol. J. Linnean Soc. CO;2 33, 383–394. doi: 10.1111/j.1095-8312.1988.tb00451.x Schnitzler, H. U., Moss, C. F., and Denzinger, A. (2003). From spatial orientation Norberg, U. M., and Rayner, J. M. V. (1987). Ecological morphology and flight in to food acquisition in echolocating bats. Trends Ecol. Evol. 18, 386–394. bats (Mammalia, Chiroptera) - wing adaptations, flight performance, foraging doi: 10.1016/S0169-5347(03)00185-X strategy and echolocation. Philos. Trans. R. Soc. Lond. B Biol. Sci. 316, 337–419. Silva, C. R., and Bernard, E. (2017). Bioacoustics as an important complementary doi: 10.1098/rstb.1987.0030 tool in bat inventories in the Caatinga drylands of Brazil. Acta Chiropterol. 19, Nowalk, R. M. (1999). Walker’s Mammals of the World. Baltimore, MD: Johns 409–418. doi: 10.3161/15081109ACC2017.19.2.017 Hopkins University Press. Taubert, F., Fischer, R., Groeneveld, J., Lehmann, S., Muller, M. S., Rodig, E., et al. Pavan, A. C., and Marroig, G. (2017). Timing and patterns of diversification in the (2018). Global patterns of tropical forest fragmentation. Nature 554, 519–522. Neotropical bat genus Pteronotus (Mormoopidae). Mol. Phylogenet. Evol. 108, doi: 10.1038/nature25508 61–69. doi: 10.1016/j.ympev.2017.01.017 Thiagavel, J., Santana, S. E., and Ratcliffe, J. M. (2017). Body size predicts Penone, C., Kerbiriou, C., Julien, J. F., Marmet, J., and Le Viol, I. (2018). echolocation call peak frequency better than gape height in vespertilionid bats. Body size information in large-scale acoustic bat databases. PeerJ. 6:e5370. Sci. Rep. 7:828. doi: 10.1038/s41598-017-00959-2 doi: 10.7717/peerj.5370 Torrent, L., López-Baucells, A., Rocha, R., Bobrowiec, P. E. D., and Meyer, C. F. J. Peres-Neto, P. R., Dray, S., and ter Braak, C. J. (2017). Linking trait (2018). The importance of lakes for bat conservation in Amazonian rainforests: variation to the environment: critical issues with community-weighted mean an assessment using autonomous recorders. Remote Sens. Ecol. Conserv. 4, correlation resolved by the fourth-corner approach. Ecography 40, 806–816. 339–351. doi: 10.1002/rse2.83 doi: 10.1111/ecog.02302 Vetter, D., Hansbauer, M. M., Végvári, Z., and Storch, I. (2011). Predictors of forest Petchey, O. L., and Gaston, K. J. (2006). Functional diversity: back to basics and fragmentation sensitivity in Neotropical vertebrates: a quantitative review. looking forward. Ecol. Lett. 9, 741–758. doi: 10.1111/j.1461-0248.2006.00924.x Ecography 34, 1–8. doi: 10.1111/j.1600-0587.2010.06453.x Frontiers in Ecology and Evolution | www.frontiersin.org 11 October 2019 | Volume 7 | Article 373 Núñez et al. Fragmentation Vulnerability in Insectivorous Bats Violle, C., Navas, M. L., Vile, D., Kazakou, E., Fortunel, C., Hummel, I., increases trait filtering in a biodiversity hotspot in India. Biol. Conserv. 210, et al. (2007). Let the concept of trait be functional! Oikos 116, 882–892. 48–55. doi: 10.1016/j.biocon.2017.03.026 doi: 10.1111/j.2007.0030-1299.15559.x Voigt, C. C (2013). Bat flight with bad wings: is flight metabolism affected by Conflict of Interest: The authors declare that the research was conducted in the damaged wings? J. Exp. Biol. 216, 1516–1521. doi: 10.1242/jeb.079509 absence of any commercial or financial relationships that could be construed as a Voss, S. R., Fleck, D. W., Strauss, R. E., Velazco, P. M., and Simmons, N. B. potential conflict of interest. (2016). Roosting Ecology of Amazonian Bats: Evidence for Guild Structure in Hyperdiverse Mammalian Communities. American Museum Novitates. Copyright © 2019 Núñez, López-Baucells, Rocha, Farneda, Bobrowiec, Palmeirim Wickramasinghe, L. P., Harris, S., Jones, G., and Vaughan, N. (2003). and Meyer. This is an open-access article distributed under the terms of the Creative Bat activity and species richness on organic and conventional farms: Commons Attribution License (CC BY). The use, distribution or reproduction in impact of agricultural intensification. J. Appl. Ecol. 40, 984–993. other forums is permitted, provided the original author(s) and the copyright owner(s) doi: 10.1111/j.1365-2664.2003.00856.x are credited and that the original publication in this journal is cited, in accordance Wordley, C. F. R., Sankaran, M., Mudappa, D., and Altringham, J. D. (2017). with accepted academic practice. No use, distribution or reproduction is permitted Bats in the Ghats: agricultural intensification reduces functional diversity and which does not comply with these terms. Frontiers in Ecology and Evolution | www.frontiersin.org 12 October 2019 | Volume 7 | Article 373