nena masthead
NENA Home Staff & Editors For Readers For Authors

Road and Habitat Interact to Influence Selection and Avoidance Behavior of Bats in Indiana
Roxanne D. Pourshoushtari, Benjamin P. Pauli, Patrick A. Zollner, and G. Scott Haultone

Northeastern Naturalist, Volume 25, Issue 2 (2018): 236–247

Full-text pdf (Accessible only to subscribers. To subscribe click here.)

 

Access Journal Content

Open access browsing of table of contents and abstract pages. Full text pdfs available for download for subscribers.



Current Issue: Vol. 30 (3)
NENA 30(3)

Check out NENA's latest Monograph:

Monograph 22
NENA monograph 22

All Regular Issues

Monographs

Special Issues

 

submit

 

subscribe

 

JSTOR logoClarivate logoWeb of science logoBioOne logo EbscoHOST logoProQuest logo

Northeastern Naturalist 236 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 22001188 NORTHEASTERN NATURALIST 2V5(o2l). :2253,6 N–2o4. 72 Road and Habitat Interact to Influence Selection and Avoidance Behavior of Bats in Indiana Roxanne D. Pourshoushtari1,2,*, Benjamin P. Pauli1,3, Patrick A. Zollner1, and G. Scott Haulton4 Abstract - Research on the ecology of bats and roads has largely covered impacts of major highways, but varying types of roads and surrounding habitats may differ in their impacts on bat activity. We conducted 28 mobile acoustic surveys in and around Indiana state forests from May to August 2012. We employed Manly’s selection ratios to examine levels of bat activity along different types of roads through various habitats, and the interaction of road and habitat, using an exact chi-squared test. Activity was higher than expected along unpaved local roads and roads that were lined with open-canopy forest and forest edges, whereas activity was lower than expected along state highways and “other” roads (e.g., service roads), as well as roads within open areas. The influence of a road on activity was dependent on surrounding habitat features. For example, activity on unpaved local roads was greater than expected when surrounded by closed-canopy forest, but lower than expected when surrounded by human development. Inventory and monitoring programs might be improved if they consider the interacting roles of road and habitat type in influencing how bats select their environment. Introduction Roads and other anthropogenic developments have major impacts on the density (Russo and Ancillotto 2016), diversity (Fensome and Mathews 2016), and behavior (Trombulak and Frissell 2000) of vertebrates. Roads also cause direct mortality from vehicular collisions and may have indirect effects, such as loss, degradation, disturbance, and fragmentation of habitats (Fensome and Mathews 2016). Nevertheless, roads might also have positive effects on some species of wildlife, such as providing carrion to scavengers with the ability to avoid oncoming traffic or by offering relief from predation to animals with predators that are not able to avoid vehicular collisions (Fahrig and Rytwinski 2009). Bats in the US comprise a suite of species of considerable conservation concern that may be vulnerable to threats posed by roads (Bennett and Zurcher 2013, Bennett et al. 2013, Berthinussen and Altringham 2012). Roads can act as barriers and limit access to suitable habitat (Fensome and Mathews 2016). Roads may be a significant barrier to bats that forage close to surfaces and are not willing to fly in the open, although bats that forage in the open and are more willing to cross motorways 1Department of Forestry and Natural Resources, School of Agriculture, Purdue University, West Lafayette, IN 47907. 2Current address - Department of Biology, Angelo State University, San Angelo, TX 76904. 3Department of Biology, Saint Mary’s University of Minnesota, Winona, MN 55987. 4Division of Forestry, Indiana Department of Natural Resources, Indianapolis, IN 46204. *Corresponding author - rox11492@gmail.com. Manuscript Editor: Allen Kurta Northeastern Naturalist Vol. 25, No. 2 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 237 with traffic may put themselves at a greater risk of mortality (Kerth and Melber 2009). Bats are more likely to turn away from roads when vehicles are present (Zurcher et al. 2010), so a greater number of encounters with automobiles increases avoidance behavior (Bennett et al. 2013); but even in the absence of vehicles, some bats avoid roads, due to the characteristic gap in tree cover created by a roadway (Bennett and Zurcher 2013). Other factors, such as excessive anthropogenic noise, may also contribute to a reduction in bat-foraging activity and success (Schaub et al. 2008). Artificial lighting (e.g., street lamps) can influence bat behaviors such as commuting and foraging, and artificial lighting can hinder foraging by some species (Stone et al. 2015). Small roads with little traffic, though, may actually be beneficial to bats, providing useful commuting routes (Abbott et al. 2012) or foraging opportunities (Zimmerman and Glanz 2000). Bats use various habitats, although wooded areas, edges, and riparian zones generally are important habitat for bats in the eastern US (Sparks et al. 2005, Whitaker et al. 2007). Forests provide bats with foraging (Grindal and Brigham 1999) and roosting habitat (Hayes 2003), as well as cover to escape from predators (Zimmerman and Glanz 2000). Activity levels of insectivorous bats tend to be positively correlated with overall foliage density (Bullen and Dunlop 2012), whereas open areas such as agricultural fields are often avoided, due to lack of vegetative cover (Henderson and Broders 2008). Urbanization and the associated loss of vegetation reduce insect density, which can lead to decreased populations of the insect taxa that are food sources for foraging bats, in turn reducing bat density in such areas (Russo and Ancillotto 2015). Variations in road type and habitat independently affect bats, but these factors can interact with one another. For instance, the impacts of roadways on bats may be exacerbated by the resultant modification of natural habitat (Russo and Ancillotto 2015). Habitat destruction and fragmentation increase with road density, and these combined factors can lead to transformation of the physical environment adjacent to the road (Trombulak and Frissell 2000), isolation of populations, and loss of preferred habitat (Berthinussen and Altringham 2012). Although bats are greatly influenced by both roadways and composition of the surrounding habitat, whether a particular type of road has a uniform influence on bat activity across habitats is not as well understood. The goal of this project was to use acoustic-transect surveys to measure the variation in activity of bats along different types of roads that were surrounded by different habitats. Our specific objectives were to: quantify the influence of road type on overall activity, identify the influence of habitats surrounding roads on overall activity, and determine if bats respond to combinations of different types of road and habitat in an additive or interactive manner. We predicted that bats would avoid large roads, while selecting for small roads. We hypothesized that bats would be more active along roads in forested areas and near forest edges, while avoiding open and developed areas. We also hypothesized that bats would respond to the combinations in an interactive manner, such that activity levels over a single road type would vary through different habitats and vice versa. Northeastern Naturalist 238 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 Vol. 25, No. 2 Field-site Description We conducted acoustic surveys of bats in and near 13 state forests of Indiana. Our 14 sampling transects were located along roads within and up to 8 km outside of forest boundaries (center of all areas 39°51'36.00''N 85°54'57.60''W; Fig. 1). Our northern study areas were within the Central Till Plain natural region, whereas southern areas were within the Southwestern Lowlands, Shawnee Hills, Highland Rim, and Bluegrass natural regions (Homoya et al. 1985). The state forests and surrounding forested areas are composed of conifers, mixed forest, Quercus (oak)– Carya (hickory), mixed oak, Acer spp. (maples), and other hardwood stands; the most common types are oak–hickory and mixed oak (Shao et al. 2014). Open areas consisted of various agricultural and grassy fields or bodies of water, while developed areas were rural neighborhoods (R.D. Pourshoushtari, pers. observ.). There were 9 bat species whose known range overlapped our study area and thus potentially could be detected during our acoustic surveys (Whitaker et al. 2007). These included Eptesicus fuscus (Palisot de Beauvois) (Big Brown Bat), Myotis septentrionalis (Trouessart) (Northern Long-eared Bat), Myotis lucifugus (Le Conte) (Little Brown Bat), Myotis sodalis Miller and Allen (Indiana Bat), Figure 1. The 14 routes through state forests and surrounding areas in Indiana along which we conducted acoustic surveys of bats during summer 2012. Routes were 40–48 km in length and consisted of public roads on state forest property or within 8 km of forest boundaries. Northeastern Naturalist Vol. 25, No. 2 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 239 Lasiurus borealis (Müller) (Eastern Red Bat), Lasiurus cinereus (Palisot de Beauvois) (Hoary Bat), Perimyotis subflavus (F. Cuvier) (Tricolored Bat), and Nycticeius humeralis (Rafinesque) (Evening Bat). Recent captures of Myotis leibii (Audubon and Bachman) (Eastern Small-footed Bat) in Indiana indicate its potential presence as well (Gikas et al. 2009). Methods We conducted mobile surveys on public roads from 31 May to 7 August 2012, following standardized methods of Britzke and Herzog (2009). We sampled along each of the transects 1–3 times, with most sampled twice, resulting in 28 mobile surveys (imbalances in sampling effort were due to equipment malfunction; Fig. 1). Each transect was 40–48 km long. We drove the routes at speeds of 24–32 km/h, starting 20 min after sunset, on nights with temperatures suitable for bat activity (≥12.8 °C; Wolbert et al. 2014), no precipitation or fog, and forecasted wind speeds of less than 24 km/h (USFWS 2012). We used an Anabat SD2 (Titley Electronics, Ballina, NSW, Australia) to record echolocation calls, with a roof-mounted microphone angled 5–15° from vertical, and stored the recorded data on an iPaq PDA device (Hewlett-Packard, Palo Alto, CA). We logged the location of each ultrasonic recording with a GPS (CompactFlash SiRF STAR III, GlobalSat, New Taipei City, Taiwan). Using AnalookW (Titley Electronics, Ballina, NSW, Australia), we manually filtered recorded files and removed files containing only noise. We did not attempt species-level identification; doing so would have resulted in a large reduction in our sample size, due to unidentifiable bat calls and the limited number of echolocation pulses in certain files. We recorded the starting and ending points of segments of roads and habitats with a hand-held GPS unit (GPSmap 60C, Garmin, Olathe, KS). We used geographic infromation system (GIS) data from the Indiana Department of Transportation (INDOT) to classify segments of roadway into 1 of 5 types (Table 1): local roads (LCL/P or LCL/U), which were defined as local, neighborhood, and rural roads, combined with our visual determination of whether a road was paved or unpaved; US highways (USHW), which included primary roads with limited access and unseparated US highways; state highways (STHW), which were defined as secondary and connecting roads and unseparated state highways; and other (OTH) roads, which included driveways, service roads, trails requiring 4-wheel-drive vehicles to pass, or undefined special road features (US Census Bureau 2017). We considered LCL/U, LCL/P, and OTH small roads and USHW and STHW large roads. We did not quantify the amount of traffic, but there was a clear distinction in traffic volumes that we encountered on large (high-volume) versus small (low-volume) roads. We classified habitats adjacent to roadways into 7 categories (Table 1), based on the most prominent characteristics that we observed (e.g., forests, open lands, and development) within 15 m of the road because this was the greatest distance that we could safely assume would fall within the sampling range of the detector (Titley Scientific 2017). Open-canopy forest (OCF) was wooded on both sides, with a break in the tree canopy over the road, whereas closed-canopy forest (CCF) Northeastern Naturalist 240 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 Vol. 25, No. 2 was similar, but with no break in canopy over the road. We defined open (OPN) as regions without tree cover, such as agricultural fields or bodies of water, on both sides of the road. We considered developed areas (DEV) to be sites of human development on both sides of the road, which, in our study area, consisted of neighborhoods with a range of 5–10 houses or areas with several farm buildings. The last 3 types of habitat were forest edges facing open or developed areas on the opposite side of the road (F/O and F/D) or open on 1 side with developed on the other (O/D). We marked a change in either type of road or habitat if the change extended for at least 15 m along the route, so as to limit potentially recording bats in a different category of road or habitat than where the detector was located. Our system yielded a total of 35 possible combinations of road and habitat. We measured the lengths of the roads and habitats for all routes using ArcMap 10.1 (ESRI, Redlands, CA). We overlayed the GPS locations included with the call files along the routes, which allowed us to determine the number of files within each type of road, habitat, or road–habitat combination. To determine whether bats were active in environments in proportion to their availability on the landscape, we calculated the expected number of call files for each category, based upon its total length along each route and the number of times that that the route was surveyed. We determined individual effects of either type of road or habitat via calculation of Manly’s selection ratios (wi) and Bonferroni critical values (Manly et al. 2002). We calculated selection ratios using the adehabitatHS package in R (Calenge 2006). Selection ratios significantly >1 indicated that bats were selecting for that feature, while values significantly less than 1 indicated avoidance. We analyzed the interaction Table 1. Different types of roads and habitats along 14 routes associated with state forests in Indiana during summer 2012. We classified roads based on a combination of GIS data from the Indiana Department of Transportation (INDOT) and our observations of whether or not a road was paved. We defined habitats based on the prominent landscape features within 15 m of the roadway edge. wi = Manly’s selection indices, SE = standard error, and P = probability for each type. Significant results after Bonferroni correction are indicated by an asterisk (*). Definition wi SE P Road LCL/P Local, neighborhood, rural road, or city street/paved 0.97 0.02 0.03 LCL/U Local, neighborhood, rural road, or city street/unpaved 1.13 0.03 less than 0.001* USHW Primary road/U.S. highway (paved) 1.09 0.14 0.51 STHW Secondary road/state highway (paved) 0.87 0.04 0.004* OTH Underpass, driveway or service road, no INDOT 0.53 0.15 0.001* classification/paved or unpaved Habitat OCF Forest on either side of road, gap in canopy above road 1.18 0.04 less than 0.001* CCF Forest on either side of road, no gap in canopy above road 1.03 0.04 0.49 OPN Open field, shrubland, or open water on both sides of road 0.81 0.03 less than 0.001* DEV Human development on both sides of road 0.85 0.06 0.02 F/O Forest on 1 side and open on the other 1.11 0.04 0.004* F/D Forest on 1 side and developed on the other 0.97 0.08 0.69 O/D Open on 1 side and developed on the other 0.87 0.07 0.05 Northeastern Naturalist Vol. 25, No. 2 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 241 between road and habitat using a Pearson’s chi-squared test for independence, resampled 100,000 times using a Monte Carlo approach (Agresti et al. 1979). We also calculated standardized Pearson residuals (Byers et al. 1984) to determine the significance of individual combinations of variables. All statistical tests were conducted in RStudio v.1.0.136 (R Development Core Team 2014). Results Our 28 mobile surveys covered 1200 km in total driven distance. The most common types of roads were LCL/P (55.7%) and LCL/U (29.4%), with smaller contributions from STHW (12.6%), USHW (1.7%), and OTH (0.7%; Table 2). The only unpaved surfaces in any survey were on local roads and OTH. There were greater proportions of OPN (24.7%), OCF (22.5%), F/O (20.6%), and CCF (15.8%) than there were of DEV (6.3%), O/D (5.6%), and F/D (4.5%; Table 2). Two road– habitat combinations were never present—USHW through CCF and OTH through F/D. During our 28 mobile surveys, we recorded 3220 files containing sounds of echolocating bats. We recorded the calls on all road and habitat types and in nearly every combination of the two. Our analyses of selection indices for road type indicated that bats were selecting or avoiding different roads relative to availability (χL 2 = 30.2, df = 4, P < 0.0001). Furthermore, Manly’s selection ratios indicated greater activity than expected along LCL/U and less than expected along STHW and OTH (Table 1, Fig. 2A). Manly’s indices for habitats also revealed selection or avoidance based upon landscape features (χL 2 = 68.6, df = 6, P < 0.0001). Post-hoc tests indicated significant avoidance of OPN and selection for OCF and F/O (Table 1, Fig. 2B). Table 2. Lengths (km) and proportions (%) of different types of roads and habitats, along 14 routes associated with state forests in Indiana, during summer 2012. Total habitat types are represented by the cells in the bottom row and road-type totals are represented by the cells in the last column of the table. See Table 1 for definitions of roads and habitats. Habitat Total length Road OCF CCF OPN DEV F/O F/D O/D of road LCL/P 135.5 83.9 180.5 44.1 143.6 31.5 50.0 669.1 (11.3%) (6.9%) (15.0%) (3.7%) (12.0%) (2.6%) (4.2%) (55.7%) LCL/U 85.9 105.9 80.2 2.5 70.5 3.2 3.9 352.1 (7.2%) (8.8%) (6.7%) (0.2%) (5.9%) (0.3%) (0.3%) (29.4%) USHW 8.0 - 1.9 0.2 3.3 4.6 1.5 19.5 (0.7%) - (0.2%) (less than 0.1%) (0.3%) (0.4%) (0.1%) (1.7%) STHW 40.4 0.4 29.8 28.7 24.9 14.2 11.7 150.1 (3.4%) (less than 0.1%) (2.5%) (2.4%) (2.1%) (1.2%) (1.0%) (12.6%) OTH 0.6 1.5 3.6 0.1 3.0 - 0.4 9.2 (less than 0.1%) (0.1%) (0.3%) (less than 0.1%) (0.3%) - (less than 0.1%) (0.7%) Total length 270.4 191.7 296.0 75.6 245.3 53.5 67.5 1200 of habitat (22.5%) (15.8%) (24.7%) (6.3%) (20.6%) (4.5%) (5.6%) (100%) Northeastern Naturalist 242 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 Vol. 25, No. 2 The exact chi-squared test for independence provided evidence that there are interactive, rather than purely additive, effects of road type and habitat combinations on bat activity (χ2 24 = 488.04, P < 0.0001). Bonferroni-corrected standardized Pearson residuals indicated greater than expected activity on LCL/U within CCF, Figure 2. Observed and expected proportion of echolocation calls for types of (A) road and (B) habitat, along 14 routes associated with state forests in Indiana, during summer 2012. See Table 1 for definitions of roads and habitats. Asterisks (*) denote significant selection or avoidance for a particular road or habitat. Northeastern Naturalist Vol. 25, No. 2 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 243 LCL/P along O/D, STHW within OCF or DEV, and USHW along F/D (Table 3). Combinations of road and habitat that exhibited less activity than expected were LCL/U within DEV or along O/D, or F/D edges. Paved local roads or STHW within CCF also had less activity than expected (Table 3). Discussion We detected reduced bat activity along state highways, which could be due to a combination of factors. First, avoidance by bats of motorways with intense traffic could be a response to the physical properties of large roads, including larger gaps in tree canopy and greater crossing distances (Fensome and Mathews 2016). Second, motor vehicles produce loud sounds that might reduce the bats’ forging efficiency, leading to an avoidance of areas with more noise (Schaub et al. 2008, Siemers and Schaub 2011). Third, some bats shun areas flooded by sources of anthropogenic light, such as vehicles and street lamps (Stone et al. 2015). Finally, high traffic volume increases the likelihood of collisions between vehicles and bats (Fensome and Mathews 2016), which could lead to a reduced density of bats and lower levels of activity. Investigating the mechanisms underlying our observations of bat activity along state highways might be a fruitful area for future research. Although we considered both US highways and state highways large roads, we detected no avoidance of US highways; nevertheless, US highways represented only 1.7% of all roads in our study, which could have affected our ability to detect a statistical effect. We also demonstrated significant selection for unpaved local roads, which consisted of small gravel or dirt roads that likely lacked many of the potentially negative characteristics of larger roads, such as high traffic, large gaps in the canopy, and excessive noise or light. Furthermore, these roads may have acted as habitat edges and corridors, which many species of bats favor (Fensome and Mathews 2016). We also detected avoidance of “other” roads, which are like unpaved local roads in that they are small and tend to have low traffic volume. However, this category was under-represented in our study area, comprising less than 1% of our roads; consequently, rarity may have contributed to this unexpected result. Table 3. Results of the exact chi-squared tests that indicate an interactive effect between types of habitat and road along 14 routes associated with state forests in Indiana, during summer 2012. Numbers marked with an asterisk (*) indicate statistical significance based on Bonferroni-corrected standardized Pearson residuals. Significantly greater-than-expected activity relative to habitat type and road is indicated by a positive value and negative values indicate significantly less-than-expected activity. See Table 1 for definitions of roads and habitats. Habitat type Road type OCF CCF OPN DEV F/O F/D O/D LCL/P -2.57 -3.79* 1.06 1.32 1.38 3.02 3.63* LCL/U 0.42 9.86* 0.10 -6.63* -1.23 -5.30* -4.70* USHW 3.06 -3.06 -2.18 -1.75 -1.36 6.72* 1.33 STHW 4.00* -7.60* -1.80 9.45* -0.50 -0.05 -0.52 OTH -1.32 0.59 0.88 -0.84 0.62 -0.75 0.46 Northeastern Naturalist 244 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 Vol. 25, No. 2 We found that foraging/commuting bats along roads selected for open-canopy forest and forest edges, while avoiding open areas. Certain species (e.g., Indiana Bats) use wooded cover instead of open fields for commuting, even when it substantially increases flight distance (Murray and Kurta 2004). Bats rely, at least in part, on linear elements for commuting, and many woodland bats forage along habitat edges (Altringham and Kerth 2016). Not only do forest edges and corridor-like habitats provide areas that are more efficient for foraging than cluttered habitats (due to low structural complexity and high densities of insects), they do so without sacrificing the cover of trees, which provide escape from aerial predators (Zimmerman and Glanz 2000). Therefore, our observations concur with current knowledge on habitat selection of bats, specifically that higher levels of activity occurred in open-canopy forest and forest edges, and less activity took place in open areas. Although road type and habitat affected activity independently, our analysis of the interaction between the 2 variables demonstrated non-additive effects. We discovered 2 particularly noteworthy findings based upon post-hoc a nalyses. The first was the difference in use of closed-canopy forest across different types of roads (Table 3). We expected that bats would select for closed-canopy forest regardless of road type because bats consistently use cover while commuting and less frequently cross roads lacking adjacent vegetation (Fensome and Mathews 2016). However, bats did not independently select for closed-canopy forests. Furthermore, activity in closed-canopy forest was only greater than expected in conjunction with unpaved local roads, whereas activity was less than expected when closed-canopy forest occurred alongside paved local roads and state highways. The pattern across closed-canopy forest may be indicative of the impact of paving a road and, likely, the associated increase in traffic volume and speed. Paved roads tend to have greater traffic density, which can lead to more collisions between vehicles and bats (Trombulak and Frissell 2000). Our analysis also indicated less activity than expected along paved roads through closed-canopy forests, suggesting that maintaining vegetation over roads may not be sufficient to retain the suitability of an area. Thus, when management objectives include increasing the quality of habitat for foraging and roosting bats, it may help to maintain some less-developed areas. Management could include not paving roads that are near roosts or are crossed by commuting bats, or installing traffic-calming measures to limit the volume and speed of traffic and associated habitat degradation. The second interesting interaction was the relationship of activity near unpaved local roads across habitats. In general, activity was higher than expected for unpaved local roads, which was unsurprising because these roads exhibited characteristics of linear edge features used by commuting bats (Altringham and Kerth 2016). However, when unpaved local roads occurred in closed-canopy forest, activity was even greater than expected, and conversely, when such roads traversed habitat that included human development (developed, forest/developed edge, and open/developed edge), activity was lower than expected. Urbanization may have a number of negative impacts on bats, such as lower production of juveniles, due to the inherent decrease in quality of habitat (Russo and Ancillotto Northeastern Naturalist Vol. 25, No. 2 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 245 2015), so it makes sense that activity would be less than anticipated in the presence of development. Manly’s selection indices did not indicate avoidance of developed habitats exclusively, and avoidance was evident only when our analysis coupled habitat with road type. Although not expected, the lack of avoidance might have been due to the size of developed areas at our study site, where these areas were present as small development-centers within a rural landscape, rather than extensively urbanized centers. Alternatively, lesser avoidance (more use) of developed areas than expected could have been caused by species-specific variation in activity, which was not accounted for in our study. The influence of urban environments on bat populations is likely species- and context-specific (Russo and Ancillotto 2015); thus, increased activity of some urban-tolerant species could have counterbalanced the decreased activity of less-tolerant species. These results highlight the complex nature of habitat use by bats and indicate that independently considering the impacts of either roads or habitats on activity has the potential to draw incomplete conclusions. Future roadside monitoring programs should be designed in ways that account for the interacting roles that habitat and roads have in the selection and avoidance behavior of bats. For instance, if activity in different locations is to be compared, sampling areas with similar combinations of road and habitat will likely provide the most reliable results. Our results also reinforce the need for habitat managers to be cognizant of how road expansion and increases in human development may decrease the suitability of nearby habitat for bats. Measures such as limiting increased development (e.g., bigger neighborhoods or larger paved roads) or perhaps restricting traffic in the evening and early morning when bats would be most active could contribute to greater density and activity of bats and promote a healthier bat community. Acknowledgments We thank the Indiana Department of Natural Resources, Division of Forestry, and the McIntire–Stennis Cooperative Forestry Research Program, for funding this project. We are grateful to J.M. Tonos and R.A. Vanausdall, for assistance in data collection; and R.A. Vanausdall, A.J. Cohen, L.E. D’Acunto, V.J. Bennett, R.J. Spaul, S.H. Smith, and B. Dudek, for feedback on earlier drafts. Literature Cited Abbott, I.M., F. Butler, and S. Harrison. 2012. When flyways meet highways: The relative permeability of different motorway-crossing sites to functionally diverse bat species. Landscape and Urban Planning 106:293–302. Agresti, A., D. Wackerly, and J.M. Boyett. 1979. Exact conditional tests for cross-classifications: Approximation of attained significance levels. Psychometrika 44: 75–83. Altringham, J., and G. Kerth. 2016. Bats and roads. Pp. 35–62, In C.C. Voigt and T. Kingston (Eds.). Bats in the Anthropocene: Conservation of Bats in a Changing World. Springer International Publishing, Cham, Switzerland. 606 pp. Bennett, V.J., and A.A. Zurcher. 2013. When corridors collide: Road-related disturbance in commuting bats. Journal of Wildlife Management 77:93–101. Northeastern Naturalist 246 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 Vol. 25, No. 2 Bennett, V.J., D.W. Sparks, and P.A. Zollner. 2013. Modeling the indirect effects of road networks on the foraging activities of bats. Landscape Ecology 28:979–991. Berthinussen, A., and J. Altringham. 2012. The effect of a major road on bat activity and diversity. Journal of Applied Ecology 49:82–89. Britzke, E.R., and C. Herzog. 2009. Using acoustic surveys to monitor population trends in bats. US Army Engineer Research and Development Center, Vicksburg, MS. 4 pp. Bullen, R.D., and J.N. Dunlop. 2012. Assessment of habitat usage by bats in the rangelands of Western Australia: Comparison of echolocation-call count and stable-isotope analysis methods. Rangeland Journal 34:277–284. Byers, C.R., R.K. Steinhorst, and P.R. Krausman. 1984. Clarification of a technique for analysis of utilization-availability data. Journal of Wildlife Management 48:1050–1053. Calenge, C. 2006. The package “adehabitat” for the R software: A tool for the analysis of space and habitat use by animals. Ecological Modelling 197:516–519. Fahrig, L., and T. Rytwinski. 2009. Effects of roads on animal abundance: An empirical review and synthesis. Ecology and Society 14:21. Fensome, A.G., and F. Mathews. 2016. Roads and bats: A meta-analysis and review of the evidence on vehicle collisions and barrier effects. Mammal Review 46:311–323. Gikas, N.S., J.G. Boyles, A.A. Zurcher, B.L. Walters, and J.O. Whitaker Jr. 2009. The first records of the Eastern Small-footed Bat (Myotis leibii) in Indiana. Proceedings of the Indiana Academy of Science 118:203–204. Grindal, S.D., and R.M. Brigham. 1999. Impacts of forest harvesting on habitat use by foraging insectivorous bats at different spatial scales. Ecoscience 6:25–34. Hayes, J.P. 2003. Habitat ecology and conservation of bats in western coniferous forests. Pp. 81–119, In C.J. Zabel and R.G. Anthony (Eds.). Mammal Community Dynamics in Coniferous Forests of Western North America: Management and Conservation. Cambridge University Press, Cambridge, UK. 732 pp. Henderson, L.E., and H.G. Broders. 2008. Movements and resource selection of the Northern Long-eared Myotis (Myotis septentrionalis) in a forest–agriculture landscape. Journal of Mammalogy 89:952–963. Homoya, M.A., D.B. Abrell, J.R. Aldrich, and T.W. Post. 1985. The natural regions of Indiana. Indiana Academy of Science 94:245–268. Kerth, G., and M. Melber. 2009. Species-specific barrier effects of a motorway on the habitat use of 2 threatened forest-living bat species. Biological Conservation 142:270–279. Manly, B.F., L. McDonald, D. Thomas, T.L. McDonald, and W.P. Erickson. 2002. Resource Selection by Animals: Statistical Design and Analysis for Field Studies, 2nd Edition. Springer Publishing, Dordrecht, Netherlands. 222 pp. Murray, S.W., and A. Kurta. 2004. Nocturnal activity of the endangered Indiana Bat (Myotis sodalis). Journal of Zoology 262:197–206. R Development Core Team. 2014. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available online at http:// www.R-project.org. Accessed 10 June 2015. Russo, D., and L. Ancillotto. 2015. Sensitivity of bats to urbanization: A review. Mammalian Biology 80:205–212. Schaub, A., J. Ostwald, and B.M. Siemers. 2008. Foraging bats avoid noise. Journal of Experimental Biology 211:3174–3180. Shao, G., B.P. Pauli, G.S. Haulton, P.A. Zollner, and G. Shao. 2014. Mapping hardwood forests through a two-stage unsupervised classification by integrating Landsat Thematic Mapper and forest-inventory data. Journal of Applied Remote Sensing 8:083546. Northeastern Naturalist Vol. 25, No. 2 R.D. Pourshoushtari, B.P. Pauli, P.A. Zollner, and G.S. Haulton 2018 247 Siemers, B.M., and A. Schaub. 2011. Hunting at the highway: Traffic noise reduces foraging efficiency in acoustic predators. Proceedings of the Royal Society B: Biological Sciences 278:1646–1652. Sparks, D.W., C.M. Ritzi, J.E. Duchamp, and J.O. Whitaker Jr. 2005. Foraging habitat of the Indiana Bat (Myotis sodalis) at an urban–rural interface. Journal of Mammalogy 86:713–718. Stone, E.L., S. Harris, and G. Jones. 2015. Impacts of artificial lighting on bats: A review of challenges and solutions. Mammalian Biology 80:213–219. Titley Scientific. 2017. AnaBat SD2 Compact Flash Bat Detector User Manual (version 1.8). Titley Scientific, Lawnton, Australia. 108 pp. Trombulak, S.C., and C.A. Frissell. 2000. Review of ecological effects of roads on terrestrial and aquatic communities. Conservation Biology 14:18–30. US Census Bureau. 2017. TIGER/Line shapefiles technical documentation. US Department of Commerce. Available online at https://www2.census.gov/ geo/pdfs/maps-data/data/tiger/tgrshp2017/TGRSHP2017_TechDoc.pdf. Accessed 1 December 2017. US Fish and Wildlife Service (USFWS). 2012. Mobile bat acoustical survey protocol. US Department of the Interior. Available online at https://ecos.fws.gov/ServCat/ DownloadFile/35813?Reference=35782. Accessed 19 December 2017. Whitaker, J.O., Jr., V. Brack Jr., D.W. Sparks, J.B. Cope, and S. Johnson. 2007. Bats of Indiana. Indiana State University Center for North American Bat Research and Conservation, Terre Haute, IN. 59 pp. Wolbert, S.J., A.S. Zellner, and H.P. Whidden. 2014. Bat activity, insect biomass, and temperature along an elevational gradient. Northeastern Naturalist 21:72–85. Zimmerman, G.S., and W.E. Glanz. 2000. Habitat use by bats in eastern Maine. Journal of Wildlife Management 64:1032–1040. Zurcher, A.A., D.W. Sparks, and V.J. Bennett. 2010. Why the bat did not cross the road? Acta Chiropterologica 12:337–340.