Northeastern Naturalist 86 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 22001144 NORTHEASTERN NATURALIST V21o(l1. )2:18,6 N–1o0. 71 Offshore Acoustic Monitoring of Bats in the Gulf of Maine Trevor S. Peterson1,*, Steven K. Pelletier1, Sarah A. Boyden1, and Kristen S. Watrous1 Abstract - Although bats have been observed from offshore ships, are known to cross large bodies of water, and have been observed seasonally on remote islands, little information has been collected regarding their distribution, species composition, and behavior at offshore locations. Techniques for monitoring bats over long periods are limited, and the typical logistical constraints for studies of nocturnal, flying species are amplified in open-water habitats. To our knowledge, this acoustic study represents the first attempt to monitor bats on a regional scale in an offshore environment. Long-term acoustic monitoring of 16 sites in the Gulf of Maine confirmed the presence of long-distance migratory and cave-hibernating bat species at coastal sites; large, forested islands; small, isolated rocks lacking woody vegetation; and an offshore weather buoy. Patterns of presence, species composition, and activity were highly variable among sites but consistently indicated that bats were most active and widespread between mid-August and mid-September, and that bats regularly occurred offshore. Sporadic presence of certain species during this period, surrounded by multiple nights with no activity, also suggests that bat presence offshore may be associated with migratory flight or use of remote islands as temporary roosts or stopover sites during seasonal movements. Introduction Historic logs of lighthouse keepers and chance observations from fishing vessels and ships constitute the majority of written records that document bats offshore (defined here as at least 4.8 km or 3 nautical miles from any land mass) in the Atlantic Ocean. Early published observations of North American bats at sea in the Atlantic include 3 reports of flocks of up to 200 individuals (Allen 1923, Carter 1950, Thomas 1921) and 12 accounts of 1–3 bats (Griffin 1940, Mackiewicz and Backus 1956, Nichols 1920, Norton 1930, Peterson 1970) flying past or temporarily roosting on ships 8–800 km from shore (Appendix 1). Lasiurus borealis (Müller) (Eastern Red Bat) and Lasionycteris noctivagans (Le Conte) (Silver-haired Bat) were the only species identified in these accounts, although not all observations could be assigned to species. All but 1 sighting in these accounts occurred between mid-August and the end of September, and multiple authors (e.g., Nichols 1920, Norton 1930) specified that weather was calm in the days preceding the observations, suggesting that bats had not been blown offshore. Anecdotal reports of bats at remote islands confirm that bats are capable of long water-crossings, and apparently do so on a seasonal basis (Appendix 2). Merriam (1887) reported regular spring and fall observations of Silver-haired Bats by the lighthouse keeper at Mount Desert Rock, a barren island in the Gulf of Maine, 1Stantec Consulting Services Inc., 30 Park Drive, Topsham, ME 04086. *Corresponding author - trevor.peterson@stantec.com. Manuscript Editor: Allen Kurta Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 87 42 km from the mainland, noting that bats were absent at other times of the year. Cryan and Brown (2007) analyzed 38 years of records from Southeast Farallon Island, CA, where Lasiurus cinereus (Palisot de Beauvois) (Hoary Bat) was observed consistently between August and October, with observations most frequent during September. Though present in April and May during 1 year, Hoary Bats were generally absent from the island during other times of year. Furthermore, occurrence of bats during spring and fall migration is well documented on Bermuda, 1000 km from the mainland, although midsummer observations are more sporadic, suggesting lack of a consistent resident population (Van Gelder and Wingate 1961). Fossilized bats that are 400,000 years old indicate that this offshore migratory pattern may have been established as early as the middle Pleistocene (Grady and Olsen 2006). Miller (1897) reported the presence of Eastern Red, Hoary, and Silver-haired Bats at Highland Light on Cape Cod, MA, between late August and mid-September during 2 consecutive years, noting that these species were absent at other times. Merriam’s (1887) collection of coastal observations of Hoary and Silver-haired Bats and Cryan’s (2003) study of museum records further suggest autumnal concentrations of migratory species along the Atlantic Coast (Appendix 3). One of the few telemetric studies of movement by bats recently confirmed migration of Silverhaired Bats across and along the shoreline of Lake Erie (McGuire et al. 2012). Thus, with few exceptions, presence of bats offshore has been observed only between August and October, and reports involved the Hoary Bat, Eastern Red Bat, and Silver-haired Bat—the species that account for most mortalities at wind farms in eastern North America (Arnett et al. 2008, Kunz et al. 2007). To document patterns of bat occurrence offshore, we conducted acoustic monitoring at a series of mainland coastal sites, islands of varying size and remoteness, and a buoy in the Gulf of Maine, mostly during summer and autumn of 2009–2011. Primary objectives of monitoring included testing the effectiveness and practicality of offshore acoustic surveys, and characterizing patterns of bat presence and species composition at diverse sites. Because bats must cross extensive open water to reach remote islands, describing patterns of presence at a variety of islands may help us understand when, how often, and possibly why bats occur offshore. As far as we are aware, these surveys constitute the most thorough acoustic survey of bats offshore in North America to date and provide a baseline for comparison to the results of future monitoring and for the assessment of potential risk to bats by development of offshore wind-energy facilities. Study Area and Methods Study area description and site characterization We conducted acoustic monitoring for varying amounts of time at 16 locations in the Gulf of Maine, including 12 islands, 3 coastal sites, and 1 buoy that was maintained by the Northeastern Regional Association of Coastal and Ocean Observing Systems (Table 1). These locations spanned a linear distance of 380 km (Fig. 1). We monitored 4 locations during each of the 3 years, 3 during 2 years, and the remaining 9 sites during a single year. Northeastern Naturalist 88 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 Vol. 21, No. 1 Table 1. Monitoring sites for acoustic surveys of bat during 2009–2011 in the Gulf of Maine. Area = area of island in ha; distance = distance to mainland in km; % = percent land area within 4.8 km; location = location of detector, height = height of detector in m. Site Latitude/ longitude Area Distance % Location Height Habitat surrounding detector Appledore Island 42°59'13"N, 70°36'55"W 40 10.2 0.7 Observation tower 18 Tower in complex of buildings, surrounded by shrubby forest Frenchboro 44°5'53"N, 68°22'2"W 594 10.5 7.7 Tree 3 Coniferous forest, along shoreline Gloucester Buoy 42°31'12"N, 70°33'36"W n/a 8.9 0.0 Buoy 2 Buoy, in open water Great Duck Island 44°8'31"N, 68°14'44"W 86 9.8 1.3 Lighthouse 5 Lighthouse tower, in field near shoreline Halfway Rock 43°39'21"N, 70°2'12"W 0.9 8.3 0.7 Lighthouse 23 Wave-swept rock, with no woody vegetation Isle au Haut 44°3'52"N, 68°39'4"W 2753 16.7 31.8 Lighthouse 12 Short lighthouse on rocky shoreline, away from treeline Kent Island 44°34'53"N, 66°45'19"W 99 30.2 2.0 Temporary tower 6 Mowed, grassy field, with stunted spruce trees Matinicus Rock 43°47'1"N, 68°51'18"W 10 32.9 0.7 Lighthouse 14 Wave-swept rock, with no woody vegetation Metinic Island 43°52'46"N, 69°7'37"W 140 8.7 1.6 Temporary tower 6 Grassy field, with scattered shrubs Monhegan Island 43°45'53"N, 69°18'58"W 208 17.1 2.3 Lighthouse 14 Lighthouse, with associated building and lawn surrounded by coniferous trees Mt. Desert Rock 43°57'51"N, 68°8'28"W 0.8 41.6 0.0 Lighthouse 17 Wave-swept rock, with no woody vegetation Owl’s Head 44°5'30"N, 69°2'38"W n/a n/a 19.1 Temporary tower 2 Shrubs and herbaceous vegetation on steep shoreline, with scattered spruce trees Petit Manan Island 44°22'3"N, 67°51'50"W 6 4.0 1.5 Lighthouse 30 Mowed grassy field, with occupied buildings and scattered shrubs Petit Manan Point 44°24'28"N, 67°54'10"W n/a n/a 22.9 Tree 2 Deciduous forest, with low canopy surrounding a mowed field Schoodic Peninsula 44°20'10"N, 68°3'42"W n/a n/a 18.5 Rooftop 7 Mowed field, with buildings and scattered hardwoods, surrounded by coniferous forest Seguin Island 43°42'26"N, 69°45'29"W 26 4.1 4.2 Lighthouse 16 Mowed grassy field, occupied residence, and shrubs Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 89 We recorded site location using GPS, and calculated area, distance to the nearest mainland for each island, as well as percent land area within 4.8 km of each site based on the Medium Resolution Shoreline datalayer of the National Oceanic and Atmospheric Agency (NOAA 2000) using ArcGIS (Environmental Systems Research Institute, Redlands, CA). The islands and the buoy were located ≈4–42 km from the nearest mainland. The percentage of land within a 4.8-km radius ranged from 0% at the buoy to 31.8% at Isle au Haut (T able 1). Islands also varied considerably in size and in type and amount of vegetation. Small islands (<15 ha; Halfway Rock, Matinicus Rock, Mount Desert Rock, and Petit Manan Island) were remote and lacked trees, whereas medium-sized islands (15–200 ha; Metinic Island, Appledore Island, Kent Island, Seguin Island, and Great Duck Island) supported undergrowth and limited patches of forest. In contrast, large islands (>200 ha; Frenchboro, Monhegan Island, Isle au Haut) were forested or partly forested. Coastal sites included 3 forested peninsulas: Owl’s Head, Schoodic Peninsula, and Petit Manan Point. The single buoy (Gloucester Buoy) was southeast of Gloucester, MA, 8.9 km from the nearest land (Fig. 1). Collection and analysis of data We used various acoustic-detection/recording units, including Anabat SD1, Anabat SD2, and Anabat II, with a separate zero-crossings analysis and interface module (Titley Scientific, Queensland, Australia). Heights and supporting structures for the Figure 1. Locations of offshore acoustic surveys of bats during 2009–2011. Northeastern Naturalist 90 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 Vol. 21, No. 1 equipment varied among sites. We attached 9 units on lighthouses or other permanent structures (5–30 m above surface level); 2 on portable guyed towers (6 m above surface level); and 3 units on tripods or tree trunks, 1 on a building, and 1 on an anchored buoy (2–3 m, 7 m, and 2 m above surface level, respectively). We deployed 2 detectors adjacent to each other at most locations, to minimize potential loss of data if detectors malfunctioned. Detectors were powered by sealed 12-V batteries (2.5–22.0 A-h) charged by two 10-W solar panels, with recharging regulated by a low-capacity charge controller. We programmed most detectors to record from 1700 or 1800 hours through 0800 hours (Eastern Daylight Time) every night, although a few operated continuously during all or a part of the monitoring period. We placed acoustic units in boxes made from polyvinylchloride, with an attached 90° elbow at the bottom, as in J.S. Johnson et al. (2011). The microphone of the detector faced into the elbow and received incoming calls generated by flying bats. We tested microphones before and after deployment, using a single detector and a source of ultrasound (Bat Chirp, Nevada Bat Technology, Las Vegas, NV), to ensure that microphones were functioning properly. We set sensitivity of the detectors in the field at 6.25–7.0, which was about 1 unit below the position at which detectors emitted constant noise. After downloading data, we reviewed each dataset to determine the starting and ending date of actual data collection and the possible cause and timing of failure for any systems. We scrutinized periods of many successive nights with no recorded files and considered them valid only if subsequent nights indicated normal operation. If 2 detectors at a site collected valid data for a portion of the period, we reviewed a subset of recordings from each detector and assigned the betterperforming detector (i.e., the unit with higher number of passes, more complete dataset, higher-quality recordings) as the primary detector from which we analyzed data. We defined any night with at least 1 detector deployed at a site as an attempted night, and any night during which at least 1 detector functioned properly as a valid monitoring-night. To assess variability among adjacent detectors, we calculated a Pearson correlation coefficient for the number of bat passes (natural log-transformed) recorded by primary and backup detectors during a subset of nights during which both units functioned properly. We viewed each recorded call file as a sonograph and used Analook 3.7w software (Corben 2009) to determine whether the file included a bat pass, which was defined as 2 or more discrete, frequency-modulated or quasi-constant-frequency pulses characteristic of bats and readily distinguishable from mechanical or environmental noise. We identified passes with 5 or more pulses to species or species group based on visual comparison to reference libraries of known species calls, following methods outlined in J.S. Johnson et al. (2011). In addition to the migratory species mentioned previously, we also identified cave-hibernating bats including Eptesicus fuscus (Palisot de Beauvois) (Big Brown Bat) and Perimyotis subflavus (Cuvier) (Tricolored Bat). In some cases, we were able to assign calls only to species groups rather than individual species; these groups were Big Brown/Silverhaired Bat, Eastern Red/Tricolored Bat, and Myotis (mouse-eared bats). Passes Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 91 attributed to Myotis could have been produced by M. lucifugus (Le Conte) (Little Brown Bat), M. septentrionalis (Trouessart) (Northern Long-eared Bat), and/or M. leibii (Audubon and Bachman) (Eastern Small-footed Bat). We categorized files containing only 2–4 pulses as high frequency unknown (HFUN; minimum frequency ≥33 kHz) or low frequency unknown (LFUN; minimum frequency <33 kHz), with the threshold of 33 kHz selected so that fragments of calls by Eastern Red Bats would be included in the HFUN category. Eight people trained in this standardized classification method identified calls during the 3-year study. One of 3 primary identifiers inspected all identified passes a second time to ensure proper classification to a species, species group, or unknown category . We considered the presence of 1 or more bat passes of a given category indicative of presence of that species or group for that night. We summarized patterns of acoustic activity for each site and analyzed them among sites to characterize the spatial and seasonal distribution of bats within the study area. We used wind speed and temperature to analyze local patterns of activity at those sites for which these physical data were available. Results We obtained valid acoustic data during 82% of 1258 attempted monitoringnights in 2009, 87% of 1073 attempted monitoring-nights in 2010, and 99% of 820 attempted monitoring-nights in 2011. Survey efforts focused on the period between late-July and late-October during each year, with 56% of 2781 total monitoring-nights with valid data occurring 15 July–15 October. Some detectors also operated outside this period, with monitoring occurring during 28 July–31 December 2009, 1 January–31 December 2010, and 1 January–31 December 2011 (12, 10, and 6 sites, respectively). Detectors collected valid data from 13 to 260 (mean = 99 ± 9) nights by site per year (Table 2). Two detectors operated simultaneously at sites for 1388 total nights. Of 45 unique detectors, 27 (60%) worked properly when demobilized, 7 (16%) memory cards filled to capacity before detectors were demobilized, 5 (11%) detectors stopped recording prematurely for unknown reasons, 4 (9%) battery/charging systems failed, and 1 (2%) entire detector and 1 (2%) microphone were destroyed by saltwater. Although most microphones showed some corrosion, post-survey tests of sensitivity with the Bat Chirp suggested that sensitivity of working detectors was similar to levels before deployment. Data obtained during 809 monitoring-nights allowed us to compare 12 unique pairs of primary and backup detectors. Both detectors recorded 1 or more passes during 330 (41%) nights; neither detector recorded bats during 396 (49%) nights, and only 1 detector recorded bats during 83 (10%) nights. The ratio of passes recorded by the primary detector compared to the backup detector was 1.86 ± 0.13, which demonstrates the benefit of using more than one detector. Nevertheless, during the 330 monitoring-nights when both units recorded activity, the number of bat passes recorded by primary and backup detectors was highly correlated (r = 0.94; P < 0.01). Northeastern Naturalist 92 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 Vol. 21, No. 1 Bats were recorded at every site. A total of 75,058 bat passes was recorded during 949 of 2781 (40%) successful monitoring-nights in 2009–2011. More than 99% of bat passes were recorded between 15 July and 15 October, although only 56% of sampling occurred during this period. Bats were present during at least 30% of nights between 15 July and 15 October at all but 1 site (Table 2), although activity during 15 July–15 October was highly variable among sites, ranging from 1 pass/ night at Monhegan Island in 2009 to 389 passes/night at Seguin Island in 2011 (Table 2). Table 2. Location, dates, and results of acoustic surveys, in the Gulf of Maine, 2009–2011. Week with most passes = midpoint of consecutive 7-night period with most bat passes, # of nights = number of monitoring nights; Total passes = total number of passes; Max passes = maximum number of passes per night; Rate = rate of passes per night between 15 July and 15 October; % = percent of nights with activity between 15 July and 15 October. Week with # of Total Max Site Year Dates of survey most passes nights passes passes Rate % Appledore Island 2010 10 Aug–31 Dec 19 Aug 144 360 31 5.2 70 Frenchboro 2009 28 Aug–3 Nov 3 Sep 68 302 98 6.0 53 Gloucester Buoy 2011 1 Jun–15 Oct 29 Aug 137 84 9 0.9 31 Great Duck Island 2009 17 Aug–20 Oct 29 Aug 65 954 153 15.9 50 2010 27 Aug–15 Dec 26 Sep 111 269 63 5.3 50 2011 1 Jan–31 May 1 May 151 2 2 n/a n/a Halfway Rock 2009 13 Aug–31 Dec 23 Aug 141 287 60 4.4 44 2010 1 Jan–24 Feb 55 0 0 0 0 Isle au Haut 2009 26 Aug–11 Nov 29 Aug 78 237 97 4.6 57 Kent Island 2011 3 July–18 Aug 14 Aug 47 97 18 2.7 57 Matinicus Rock 2009 2 Sep–14 Sep 6 Sep 13 102 43 7.8 54 2010 5 Aug–31 Oct 21 Aug 88 178 24 2.5 32 Metinic Island 2009 29 Jul–27 Oct 13 Aug 91 209 29 2.6 46 2010 16 Jul–16 Oct 20 July 93 213 21 2.3 55 Monhegan Island 2009 16 Sep–27 Dec 6 Oct 103 27 5 0.8 43 2010 19 Jul–4 Sep 30 Aug 48 1157 197 24.1 96 2011 12 Aug–31 Dec 15 Sep 142 1452 235 22.1 82 Mt. Desert Rock 2009 17 Aug–31 Dec 29 Aug 137 597 168 9.7 30 2010 26 Aug–31 Dec 8 Oct 128 277 92 5.4 41 2011 1 Jan–17 Sep 1 Aug 260 366 133 5.2 42 Owl’s Head 2009 11 Aug–14 Nov 17 Aug 93 3208 397 50.5 90 Petit Manan Island 2009 28 Jul–29 Oct 17 Aug 94 117 36 1.5 31 2010 30 Jul–11 Nov 20 Aug 105 94 27 1.2 17 Petit Manan Point 2010 7 Sep–4 Nov 22 Sep 59 4659 978 119.4 87 Schoodic Peninsula 2009 18 Aug–31 Oct 29 Aug 75 314 49 5.3 64 Seguin Island 2009 25 Aug–11 Nov 29 Aug 79 19,833 2494 380.9 81 2010 15 Jul–24 Oct 14 Sep 102 20,216 1989 217.3 90 2011 27 Aug–8 Nov 1 Sep 74 19,447 1775 388.8 82 Overall 2009 28 Jul–31 Dec - 1037 26,187 - 39.5 53 2010 1 Jan–31 Dec - 933 27,423 - 46.4 58 2011 1 Jan–31 Dec - 811 21,448 - 69.5 55 Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 93 Seasonal trends were similar among sites, with average monthly frequency of detection and number of passes/night highest in August (Fig. 2). The average number of hours per night with activity was highest in August during 18 of 27 (67%) distinct survey periods. The midpoints of consecutive 7-night periods with the highest combined number of passes occurred between 15 August and 15 September in 20 of 27 (74%) distinct survey periods (Table 2). No bats were detected during any of the 518 monitoring-nights at 10 sites before 30 April or after 26 November. The number of bat passes was highly variable among nights at most sites, with isolated peaks in activity often surrounded by periods of multiple nights with no detections. The highest number of passes recorded at a site during 1 night was 2494 passes recorded at Seguin Island on 31 August 2009. The 5 nights with the most recorded passes per site accounted for 17% (Seguin Island) to 77% (Frenchboro and Isle au Haut) of total recorded passes per site (mean = 50 ± 4%). In contrast, only 1 pass was recorded during 6% (Seguin Island) to 45% (Gloucester Buoy) of nights with bat activity (mean = 26 ± 3%) at other sites. We identified a total of 12,400 (17%) bat passes to species and 33,459 (45%) to species group. We documented Hoary Bats at every site, and Silver-haired Bats and Eastern Red Bats each occurred at all but 1 site (Table 3). Silver-haired Bats and Eastern Red Bats were the most commonly detected species at the buoy and each of the Figure 2. Monthly frequency of bat presence and number of bat passes per monitoring-night during acoustic surveys in the Gulf of Maine, averaged across sites. Numbers above each bar figure represent the number of sites surveyed per month, with the number of monitoringnights in parentheses. Error bars represent ± 1 Standard Error . Northeastern Naturalist 94 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 Vol. 21, No. 1 Table 3. Number of identified passes and percent composition by bat species or species group of identified passes during acoustic surveys in the Gulf of Maine, 2009–2011. Percent of monitoring nights between April and November with presence of each species or group is in parentheses. Percent Number Big Brown/ Eastern Red/ Site of passes Big Brown Silver-haired Silver-haired Hoary Eastern Red Tricolored Tricolored Myotis Appledore Island 138 1 (1) 12 (10) 12 (9) 12 (9) 38 (21) 0 26 (15) 0 Frenchboro 199 0 36 (4) 32 (6) 1 (2) 2 (3) 0 0 31 (29) Gloucester Buoy 24 0 4 (1) 8 (2) 33 (6) 50 (5) 0 4 (1) 0 Great Duck Island 715 <1 (1) 1 (3) 8 (5) 5 (3) 4 (5) <1 (1) 4 (6) 77 (21) Halfway Rock 133 0 24 (6) 22 (5) 8 (6) 22 (13) 0 2 (2) 22 (6) Isle au Haut 159 0 26 (9) 44 (10) 1 (1) 6 (5) 0 5 (6) 18 (23) Kent Island 54 0 0 0 4 (4) 0 0 0 96 (30) Matinicus Rock 141 0 21 (10) 29 (11) 3 (1) 33 (16) 0 11 (4) 4 (2) Metinic Island 216 0 7 (5) 17 (9) 13 (9) 6 (7) 0 4 (4) 52 (25) Monhegan Island 1321 <1 (<1) 2 (7) 6 (6) 2 (7) 16 (26) 0 31 (32) 43 (26) Mt. Desert Rock 754 <1 (<1) 33 (8) 16 (3) 1 (2) 23 (7) 0 <1 (1) 26 (2) Owl’s Head 1756 5 (11) <1 (4) 26 (22) 1 (9) 4 (20) 0 3 (18) 60 (62) Petit Manan Island 113 4 (1) 21 (2) 8 (3) 11 (4) 13 (5) 0 7 (2) 36 (5) Petit Manan Point 3511 0 1 (5) <1 (7) <1 (5) 1 (9) 0 4 (29) 94 (54) Schoodic Peninsula 149 3 (5) 9 (5) 40 (8) 13 (7) 7 (8) 0 3 (3) 26 (29) Seguin Island 36476 1 (8) <1 (13) 3 (31) 1 (37) 27 (54) <1 (2) 49 (58) 18 (34) Overall 45859 1 (2) 2 (3) 5 (5) 2 (6) 23 (9) <1 (<1) 41 (8) 27 (8) Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 95 treeless sites (Halfway Rock, Matinicus Rock, Mount Desert Rock, and Petit Manan Island). Myotis was also documented at all but 2 sites and was the most commonly identified species group at the 3 coastal sites and 4 of the forested islands (Great Duck Island, Kent Island, Metinic Island, and Monhegan Island). We recorded passes of Big Brown Bats at 8 locations, and Tricolored Bats occurred at only 2 sites. Overall, we categorized 29,199 (39%) as HFUN or LFUN, although the average percent of passes categorized as HFUN or LFUN per site was 46 ± 3%. Although timing of activity varied among nights, the highest number of passes per hour occurred during the 2nd hour past sunset most frequently for all sites combined (Fig. 3) and between the 1st and 4th hour past sunset at all individual sites. We also detected bats up to 88 min before sunset and 118 min past sunrise, although additional daytime activity could have occurred outside the typical sampling schedule of 1700–0800 h. Of 155 total bat passes recorded before sunset, 141 (91%) occurred ≤60 min before sunset. Most pre-sunset recordings (150 passes) occurred at Mount Desert Rock during 19 nights in May (3 nights, 3 passes), August (6, 113), September (8, 30), and October (2, 4). Species or groups recorded before sunset included the Eastern Red Bat (n = 47), Silver-haired Bat (n = 20), and Myotis (n = 9). Fifty-two bat passes occurred after sunrise, 75% of which occurred ≤60 min past sunrise; post-sunrise recordings occurred at Mount Desert Rock (n = 18), Halfway Rock (n = 12), Matinicus Rock (n = 7), and 4 additional sites. The Eastern Red Bat was the most commonly identified species after sunrise (19 of 34 passes identified to species or species group). Several isolated peaks in activity occurred at Mount Desert Rock, the most isolated and farthest offshore of the sites. For example, we recorded 143 passes Figure 3. Number of nights, averaged across sites (n = 16), when peak activity (highest number of passes per hour) occurred during a particular hour past sunset. Error bars represent ± 1 Standard Error. Northeastern Naturalist 96 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 Vol. 21, No. 1 identified as Myotis and 25 passes categorized as HFUN at this site during the first 3 hours past sunset on the night of 15 September. No bats were documented during the remainder of the night or during the 5 nights preceding or the 13 nights following this date. Average wind speeds during 5 days prior to this date, as measured by a buoy 10 km away, were light (≤4.5 m/s) and variable in direction, although wind speed increased to 9.4 m/s and became northeast during the night of 15 September. In 2010, 83% of 277 bat passes at Mount Desert Rock occurred on 5 nights between mid-September and mid-October. No bats were detected between 10 June and 26 July 2011 at Mount Desert Rock, whereas 62% of 366 bat passes recorded during 2011 at this site occurred between 2 and 4 August. In other cases, individuals from long-distance migratory species (Eastern Red Bat, Hoary Bat, and Silver-haired Bat) were detected at a site during only 1 or a small number of nights. Hoary Bats, for example, were recorded at Matinicus Rock on only 1 night during the 13-night survey period in 2009 and at no point during 88 nights in 2010. At Halfway Rock, 75% of 32 call sequences by Silver-haired Bats that were detected during the 110-night survey in 2009 occurred on the night of 5 September. At Frenchboro, 94% of passes by Silver-haired Bats (n = 71) that were recorded during the 68-night survey in 2009 occurred either between 1932 and 1946 hours on the night of 28 August (n = 30) or between 1914 and 1934 hours on the night of 5 September (n = 37). Several sequences identified as either Silver-haired Bat or Big Brown Bat were recorded at Frenchboro in the early morning hours on the night of 4 September 2009 but not during the 3 nights prior or the remainder of the month after 5 September. Of the 41 passes made by Silver-haired Bats at Isle au Haut in 2009, 51% occurred on the night of 26 August, and 29% occurred on the night of 5 September. Multiple weather buoys in the study area documented winds shifting to the northeast and increasing to ≈10 m/s during the night of 5 September 2009, although winds were light and variable for several days p receding this date. Long-distance migratory species occurred infrequently at most sites between 15 July and 15 October, with Silver-haired Bats and Hoary Bats detected during an average of 7 ± 1% and 8 ± 2% of monitoring-nights by site, respectively, during this period. Eastern Red Bats were detected during 15 ± 4% of monitoring-nights by site. However, each of these species occurred at 50–100% of surveyed sites during certain nights each year between mid-August and mid-September (Fig. 4). No storm events occurred immediately prior to these nights with widespread presence of long-distance migrating bats. Discussion Our acoustic surveys confirmed presence of bats across a diverse set of offshore and coastal sites in the Gulf of Maine and demonstrated that passive acoustic surveys can be an effective tool to monitor bats offshore. Bats were detected at every location, although the average number of bat passes varied substantially among sites. Passive acoustic surveys do not allow differentiation between multiple recordings of an individual animal versus single recordings of multiple bats, and passage rates can be influenced by factors such as volume of air space sampled, microphone orientation, Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 97 detector sensitivity, and weather variables (Hayes 2000). These factors cannot necessarily be standardized among sites over a long-term monitoring period and likely contributed to some of the inter-site variability that we documented. However, they are less likely to affect trends across seasonal and temporal scales at individual sites or determinations of presence or absence of bats at sites during certain intervals. Deploying 2 detectors at most sites enabled selection of the most complete dataset, increased the proportion of nights with valid data, and allowed differentiation between long periods with no bat presence versus malfunction of equipment. Figure 4. Percent of surveyed sites with activity of long-distance migratory bats during acoustic bat surveys in the Gulf of Maine 2009 (top), 2010 (middle), and 2011 (bottom). Note dates when multiple species were present at 50% or more sites. Dates are limited to 15 August–15 October. LACI = Hoary Bat, LANO = Silver-haired Bat, LABO = Eastern Red Bat. Northeastern Naturalist 98 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 Vol. 21, No. 1 Despite potential variability in sensitivity of detectors or detectability of bats, acoustic monitoring documented distinct activity patterns among sites. Consistency of bat presence among and within nights, and high number of bat passes at coastal sites and certain large islands (e.g., Appledore Island, Monhegan Island, and Seguin Island) suggested the presence of resident bats for much of the 15 July–15 October period. The unusually high number of bat passes recorded at Seguin Island (Table 2) was likely due to the characteristics of the lighthouse where the detector was deployed. Seguin Island was the only site with a constantly illuminated lighthouse, and resident caretakers noted insects flying around the light, and regularly reported bats foraging around the light. Thus, the high volume of recorded acoustic activity could represent repeated passes by a small number of resident bats or attraction to the lighthouse by bats during migration or foraging. Bats were present irregularly at other sites (e.g., Mount Desert Rock, Halfway Rock, Matinicus Rock, and Petit Manan Island), suggesting use by transient or migrating bats, possibly as a temporary roosting, foraging, or stopover site. Similar to Merriam’s (1887) observation that bats were present at Mount Desert Rock only during spring and fall migration, we documented bats at this site only during spring and fall, with a period of 45 consecutive nights in mid-summer with no detections. We consider the lack of bat passes across multiple nights a reliable indicator of absence of bats at treeless sites, because the detectors were deployed in the only available roosting habitat, and therefore bats should have been detected if present. Absence of natural roosting habitat and reliable fresh water at several of the treeless sites would also preclude long-term residence. Although some uncertainty exists regarding use of echolocation by bats during migration, bats appear to recognize water bodies innately using echolocation (Greif and Siemers 2010) and use echolocation for spatial orientation as well as foraging (Schnitzler et al. 2003). The only study to make direct observations of bats flying offshore demonstrated that all observed bats used echolocation, although calls were slightly lower in frequency than typical calls for the species (Ahlén et al. 2009). The presence of bats during 30% or more of the nights between 15 July and 15 October at all sites except Petit Manan Island in 2010, including 31% of nights at a remote buoy, indicates that bats occur offshore regularly during this 3-month period (Table 2). Number of recorded bat passes, frequency of presence, and hours per night with activity were all consistently highest in August, compared to other months. Although our survey effort generally targeted late summer and fall, less than 1% of bat passes occurred during the 1223 monitoring-nights (44% of total survey effort) before 15 July or after 15 October. Even at sites with apparent resident bats, increases in activity and appearance of migratory species during August and September suggested an influx of migrating bats, although juvenile bats begin to fly in mid-summer and likely contribute to increases in bat passes observed along the coast and offshore islands. Stopover of bats at offshore or coastal sites during migration has been documented in North America at Southeast Farallon Island off California (Cryan and Brown 2007); at Long Point, ON, Canada (McGuire et al. 2012); and Assateague Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 99 Island, MD (J.B. Johnson et al. 2011). Ahlén et al. (2009) documented Scandinavian bats roosting temporarily on a variety of offshore structures, including wind turbines, and cited several European studies documenting migratory departures from coastal locations in Sweden. Bats’ ability to conserve energy by entering torpor may enable them to use stopover sites that lack extensive foraging habitat (McGuire et al. 2012). Based on our acoustic data, we were unable to differentiate between bats flying past offshore sites from bats using these locations as roosting or foraging habitat during stopovers. Long-distance migratory species were present at most sites during some nights in late-August and early September. This widespread occurrence was particularly noteworthy on 26–28 August and 5 September 2009. No unusual weather events occurred before these nights, although winds became northeasterly and increased during both periods, possibly due to the passage of weather fronts. Such patterns could indicate widespread movement of long-distance migratory species through the region driven by weather or other factors. For example, Sjollema (2011) documented higher bat activity along the mid-Atlantic coast during calm nights, and J.B. Johnson et al. (2011) demonstrated a positive effect of higher temperatures and lower wind speeds on bat activity off the coast of Maryland. Similarly, Cryan and Brown (2007) found that low wind speeds, low lunar illumination, and high cloud cover predicted arrivals and departures of bats at Southeast Fa rallon Island. Anecdotal reports of bats offshore in North America mention only long-distance migratory species and indicate migration as the suspected reason for offshore presence (Appendices 2, 3). In addition to these species, and in contrast to other acoustic surveys conducted along the mid-Atlantic coast of the United States (J.B. Johnson et al. 2011, Sjollema 2011), we documented widespread presence of Myotis among sites, although frequency of presence for this genus was highly variable. Big Brown Bats and Tricolored Bats were also detected, though infrequently. The regular presence of several bat species at remote islands and an offshore buoy between 15 July and 15 October suggests that the presence of bats offshore is not merely accidental, and that bats may be attracted to such sites for roosting or foraging, or may simply pass by these sites during offshore flight. We did not attempt to differentiate echolocation types such as feeding buzzes versus searchphase calls, although islands may also provide reliable foraging habitat for bats. Numerous types of insects occur offshore (Cheng and Birch 1978), and large-scale insect migrations have been documented in coastal areas and offshore (e.g., Russel et al. 1998, Srygley and Dudley 2008, Wikelski et al. 2006), providing a potentially reliable food supply for bats above open water. Ahlén et al. (2009) also observed offshore aerial and surface foraging and suggested bats were foraging on abundant surface-dwelling crustaceans as well as aerial insects. Although our surveys occurred primarily at night, we documented isolated bouts of bat activity before sunset and after sunrise, with some passes occurring >1 hour before sunset or after sunrise, most often involving Eastern Red Bats and at remote, treeless sites. Such events are presumably the most likely to have been observed historically, because bats would have been difficult to observe in Northeastern Naturalist 100 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 Vol. 21, No. 1 darkness. These observations also highlight the potential need to consider daytime flight of bats offshore. Overall, our results were consistent with anecdotal offshore observations of bats in terms of species composition and seasonal timing (Appendices 1–3). Although we could not determine numbers of individual bats detected, historical accounts document the presence of individual bats and large groups of bats offshore. We documented a greater diversity of bat species and more consistent presence of bats offshore than historical accounts alone would suggest. In agreement with other recent acoustic surveys conducted along the mid-Atlantic coast (J.B. Johnson et al. 2011, Sjollema 2011), extensive acoustic monitoring in the Gulf of Maine demonstrated that bats routinely fly offshore and are present regularly at remote islands in autumn, possibly using these sites for migratory stopover, temporary roosts, navigational landmarks, or foraging habitat. Although our monitoring focused on late summer and fall, apparent seasonal trends offshore align with regional distribution and movement of long-distance migratory bats on land. Frequent presence of bats offshore, combined with observations of bats roosting on and around wind turbines at offshore wind farms in Europe (Ahlén et al. 2009) suggest that bats could be at risk of mortality from offshore wind turbines along the Atlantic Coast of North America. Ongoing acoustic monitoring at the sites included in this study and additional sites along the mid-Atlantic coast and Great Lakes will provide an opportunity to conduct robust quantitative analyses and modeling of species-specific patterns of bat presence and activity as they relate to weather variables, distance from shore, site characteristics, and season. Improved understanding of factors that drive offshore presence, abundance, and behavior of bats, and the variability of these factors among years, will help predict and avoid potential impacts to bats from offshore wind-energy developments. Acknowledgments Many individuals and organizations provided logistical support for this study, including access and transportation to sites. Specifically, we thank the US Fish and Wildlife Service (Maine Coastal Islands National Wildlife Refuge, Rockland, ME), US National Park Service, US Coast Guard, American Lighthouse Commission, College of the Atlantic, Bowdoin College Scientific Station, Cornell University, Friends of Seguin Island, Isle au Haut Lighthouse Commission, Maine Coast Heritage Trust, Monhegan Lighthouse Museum Association, Northeastern Regional Association of Coastal and Ocean Observing Systems, Schoodic Education and Research Center Institute, Shoals Marine Laboratory, Stone Laboratory, University of Maine, and University of New Hampshire. Joel Perkins of Stantec was instrumental in designing, deploying, and maintaining offshore acoustic detectors. Literature Cited Ahlén, I., H.J. Baagøe, and L.B. Bach. 2009. Behavior of Scandinavian bats during migration and foraging at sea. Journal of Mammalogy 90:1318–1323. Allen, G.M. 1923. The Red Bat in Bermuda. Journal of Mammalogy 4:61. Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 101 Arnett, E.B., W.K. Brown, W.P. Erickson, J.K. Fiedler, B.L. Hamilton, T.H. Henry, A. Jain, G.D. Johnson, J. Kerns, R.R. Koford, C.P. Nicholson, T.J. O’Connell, M.D. Piorkowski, and R.D. Takersley, Jr. 2008. Patterns of bat fatalities at wind-energy facilities in North America. Journal of Wildlife Management 72:61–78. Broders, H.G., G.M. Quinn, and G.J. Forbes. 2003. Species status and the spatial and temporal patterns of activity of bats in southwest Nova Scotia, Canada. Northeastern Naturalist 10:383–398. Broders, H.G., G.J. Forbes, S. Woodley, and I.D. Thompson. 2006. Range extent and stand selection for roosting and foraging in forest-dwelling Northern Long-eared Bats and Little Brown Bats in the Greater Fundy Ecosystem, New Brunswick. Journal of Wildlife Management 70:1174–1184. Buresch, K. 1999. Seasonal pattern of abundance and habitat use by bats on Martha’s Vineyard, Massachusetts. M.Sc. Thesis. University of New Hampshire, Durham, NH. Carter, T.D. 1950. On the migration of the Red Bat (Lasiurus borealis borealis). Journal of Mammalogy 31:349–350. Cheng, L., and M.C. Birch. 1978. Terrestrial insects at sea. Journal of the Marine Biological Association of the United Kingdom 57:995–997. Corben, C. 2009. Analook 3.7W Software. Available online at http://users.lmi.net/corben/ anabat.htm#Anabat%20Contents. Accessed 2 January 2010. Cryan, P.M. 2003. Seasonal distribution of migratory tree bats (Lasiurus and Lasionycteris) in North America. Journal of Mammalogy 84:579–593. Cryan, P.M., and A.C. Brown. 2007. Migration of bats past a remote island offers clues toward the problem of bat fatalities at wind turbines. Biological Conservation 139:1–11. Czenze, Z., S.N.P Wong, and C.K.R. Willis. 2011. Observations of Eastern Red Bats (Lasiurus borealis) 160 km from the coast of Nova Scotia. Bat Research News 52:28 –30. Grady, F.V., and S. Olsen. 2006. Fossil bats from quaternary deposits on Bermuda (Chiroptera: Vespertilionidae). Journal of Mammalogy 87:148–152. Greif, S., and B.M. Siemers. 2010. Innate recognition of water bodies in echolocating bats. Nature Communications 1:107. doi:10.1038/ncomms1110. Griffin, D.R. 1940. Migrations of New England bats. Bulletin of the Museum of Comparative Zoology 86:217–246. Hayes, J.P. 2000. Assumptions and practical considerations in the design and interpretation of echolocation-monitoring studies. Acta Chiropterologica 2:225–236. Johnson, J.B., and J.E. Gates. 2008. Bats of Assateague Island National Seashore, Maryland. American Midland Naturalist 160:160–170. Johnson, J.B., J.E. Gates, and N.P. Zegre. 2011. Monitoring seasonal bat activity on a coastal barrier island in Maryland, USA. Environmental Monitoring and Assessment 173:685–699. Johnson, J.S., K.S. Watrous, G.J. Giumarro, T.S. Peterson, S.A. Boyden, and M.J. Lacki. 2011. Seasonal and geographic trends in acoustic detection of tree-roosting bats. Acta Chiropterologica 13:157–168. Kunz, T.H., E.B. Arnett, W.P. Erickson, A.R. Hoar, G.D. Johnson, R.P. Larkin, M.D. Strickland, R.W. Thresher, and M. Tuttle. 2007. Ecological impacts of wind energy development on bats: Questions, research needs, and hypotheses. Frontiers in Ecology and the Environment 5:315–324. Mackiewicz, J., and R.H. Backus. 1956. Oceanic records of Lasionycteris noctivagans and Lasiurus borealis. Journal of Mammalogy 37:442–443. McConnell, S. 2007. Observation of Eastern Red Bats (Lasiurus borealis) mating in coastal Virginia. Banisteria 30:36–37. Northeastern Naturalist 102 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 Vol. 21, No. 1 McGuire, L.P., C.G. Guglielmo, S.A. Mackenzie, and P.D. Taylor. 2012. Migratory stopover in the long-distance migrant Silver-haired Bat, Lasionycteris noctivagans. Journal of Animal Ecology 81:377–385. Merriam, C.H. 1887. Do any Canadian bats migrate? Evidence in the affirmative. Transactions of the Royal society of Canada 4:85–87. Miller, G.S., Jr. 1897. Migration of bats on Cape Cod, Massachusetts. Science 118:541–543. Nichols, J.T. 1920. Red Bat and Spotted Porpoise off the Carolinas. Journal of Mammalogy 1:87. Norton, A.H. 1930. A Red Bat at sea. Journal of Mammalogy 1 1:225–226. National Oceanic and Atmospheric Administration (NOAA). 2000. NOAA Medium Resolution Shoreline Documentation. Available online at http://egisws02.nos.noaa. gov/ArcGIS/rest/services/NOAA_Shoreline/NOAA_Medium_Resolution_Shoreline/ MapServer. Accessed 26 September 2013. Peterson, R.L. 1970. Another Red Bat, Lasiurus borealis, taken aboard ship off the coast of Nova Scotia. Canadian Field-Naturalist 84:401. Russel, R.W., M.L. May, K.L. Soltesz, and J.W. Fitzpatrick. 1998. Massive swarm migrations of dragonflies (Odonata) in Eastern North America. American Midland Naturalist 140:235–342. Schnitzler, H-U., C. F. Moss, and A. Denzinger. 2003. From spatial orientation to food acquisition in echolocating bats. Trends in Ecology and Environment 18:386–394. Sjollema, A.L. 2011. Bat activity in the vicinity of proposed wind-power facilities along the mid-Atlantic coast. M.Sc. Thesis. Frostburg State University, Frostburg, MD. Srygley, R.B., and R. Dudley. 2008. Optimal strategies for insects migrating in the flight boundary layer: Mechanisms and consequences. Integrative and Comparative Biology 48:119–133. Tenaza, R.R. 1966. Migration of Hoary Bats on South Farallon Island, California. Journal of Mammalogy 47:533–535. Thomas, O. 1921. Bats on migration. Journal of Mammalogy 2:167. Van Gelder, R.G., and D.B. Wingate. 1961. The taxonomy and status of bats in Bermuda. American Museum Novitates 2029:1–9. Wikelski, M., D. Moskowitz, J.S. Adelman, J. Cochran, D.S. Wilcove, and M.L. May. 2006. Simple rules guide dragonfly migration. Biology Letters 2:325–32 9. Zimmerman, G.S. 1998. Inventory and habitat use of bats along the central coast of Maine. M.Sc. Thesis. University of Maine, Orono, ME. Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 103 Appendix 1. Evidence of offshore occurrence of resident northeastern bat species. EFLN = Big Brown Bat or Silver-haired Bat, LABO = Eastern Red Bat, LACI = Hoary Bat, LANO = Silver-haired Bat, MYSP = bats of the Myotis genus. Species Date Location Notes Citation LABO 25 August 1953 39°38'N, 70°19'W Individual captured in rigging of ship Mackiewicz and Backus (about 145 km SSE 1956 of Montauk Point, NY) LABO October 1969 42°30'N, 66°10'W Adult female aboard ship Peterson 1970 (144 km S of Yarmouth, NS, Canada) LABO 3 September 1919 “Approaching the Capes Observed on ship “about an hour after Nichols 1920 of the Carolinas from the sunrise”; bat “could then not have North, no land yet in view” been driven off-shore by heavy weather” due to calm weather LABO 1 September 1920 “Ship inbound from South Individual found on a ship, with “no Haagner 1921 as cited in Africa, when she was 3 days possibility that the bat had accompanied Griffin 1940 out of Philadelphia” the ship away from the coast.” LABO 29 September 1949 40°10'N, 71°00'W (about About 200 seen flying about ship Carter 1950 105 km offshore) LABO 7 October 1952 42°42'N, 62°58'W (241 km Brown 1953 as cited in SSE of Liverpool, NS, Canada) Peterson 1970 LABO 17 August 1929 About 209 km “south by “As no offshore gales have been Norton 1930 west from” Cape Sable, NS reported from the regions bordering the Gulf of Maine during the month of August, this bat would seem to have been an early migrant.” LABO 27 November–2 160 km from the coast of NS, “For 2–3consecutive evenings, up to Czenze et al. 2011 December 2010 8 km from Cape Sable Island 3 bats were observed at dusk.” Northeastern Naturalist 104 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 Vol. 21, No. 1 Species Date Location Notes Citation LABO, LANO 3 September 1920 Approximately 32 km off Flock of “about a 100 which caught Thomas 1921 coast of NC up with and settled on Mr. Cheeseman’s ship” LABO, LACI, March–June and Ship-based acoustic surveys Ships traveled up to 166 km from shore; Sjollema 2011 EFLN, MYSP August–October on transects perpendicular to bat activity recorded an average of 8.7 2009, spring and off coast of NJ and MD and km and maximum of 21.9 km from shore fall 2010 at anchor points from MA to NC LANO 25 August 1938 39°09'N, 70°22'W (209 km 3 bats captured on the Atlantis Griffin 1940 SE of Nantucket Island, MA) LANO 19 August 1953 39°36'N, 71°03'W (about 153 Individual “circled the ship several Mackiewicz and Backus km SSE of Montauk, NY) times before coming to rest.” 1956 Unknown 18 August 1929 8 km NNW of Provincetown, Individual “boarded a fishing vessel.” MacCoy 1930 as cited in MA Griffin 1940 Unknown 7 September 1937 45°07'N, 42°36'W (805 km Bat “flew within 15 or 20 feet [of the Griffin 1940 from Cape Race, NL, Canada) ship] but could not be captured.” Unknown First week in 16 km off “the Delaware Ship encountered a “large migration of Allen 1923 September 1902 [River],” DE bats” Unknown 6 September 1907 8 km off Sandy Hook, NJ “Probably a silver-haired bat” Murphy and Nichols 1913 as cited in Griffin 1940 Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 105 Appendix 2. Evidence of occurrence of resident northeastern bat species on remote islands (>4.8 km from the mainland). LABO = Eastern Red Bat, LACI = Hoary Bat, LANO = Silver-haired Bat, PESU = Tricolored Bat. Species Date Location Notes Citation LABO October Bermuda Allen 1923, Van Gelder and Wingate 1961 LACI Autumn, November Bermuda “Observed occasionally at dusk Jones 1884 as cited in during the autumn months … but as Griffin 1940, Merriam it is never seen except at that 1887, Allen 1923, Van particular season it is clear that it is Gelder and Wingate 1961 not a resident” LACI First week of April 1961, Southeast Farallon This species regularly appears on the Cryan 2003, Cryan and Brown early fall 1965, August to Island, CA (42 km island (Cryan 2003) 2007, Farentinos and Hawkins November 1968–2005 W of San Francisco) unpubl. data cited in Tenaza 1966 LANO October– November Bermuda “Only one specimen of this bat is Allen 1923, Jones 1884 as known to have occurred in the cited in Van Gelder and Bermudas. It was taken alive near Wingate 1961, Merriam 1887, Hamilton” Van Gelder and Wingate 1961 LANO Spring and fall Mt. Desert Rock, “A few small dark-colored bats visit Merriam 1887 ME (42 km off the place during the migrations, every coast) spring and fall.” MYSE, MYLU, April–October survey Martha's Vineyard, MYSE and MYLU captured during Buresch 1999 LABO, LACI, MA (6 km from mist netting; remaining species PESU, EPFU, mainland) identified from acoustic surveys LANO PESU 5 November 2004 Bermuda Not clear whether it arrived on Wingate 2005a, 2005b, as island with human assistance cited in Grady and Olsen 2006 Northeastern Naturalist 106 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 Vol. 21, No. 1 Appendix 3. Evidence of occurrence of resident Northeastern bats at coastal sites and nearshore islands (less than 4.8 km from the mainland). EFLN = Big Brown Bat or Silver-haired Bat, EPFU = Big Brown Bat, LABO = Eastern Red Bat, LACI = Hoary Bat, MYLE = Eastern Small-footed Bat, MYLU = Little Brown Bat, MYSE = Northern Myotis, MYSP = bats of the Myotis genus, PESU = Tricolored Bat, LBPS = Eastern Red Bat or Tricolored Bat. Species Season or date Location Notes Citation EFLN, LABO, Late May, mid-June, Kejimkujik National Park, Acoustic and mist netting survey; Broders et al. 2003 LACI, MYLU, mid-July, and Brier Island, and Bon BBSH, LACI, MYLU, MYSE, and MYSE, PESU, early September Portage Island, NS RBTB were observed in early LBPS September; other observations during the summer months EPFU, LABO, July–August 2005, Assateague Island National Mist netting and acoustic monitoring Johnson and Gates LACI, LANO, June 2006, Seashore, MD (ranges from 2008 PESU October 2006 <1 km to 8 km off coast of MD and VA); 38°10'N, 75°10'W EPFU, LABO, July 2005–December Assateague Island National Acoustic monitoring J.B. Johnson et al. LACI, LANO, 2006 Seashore, MD (ranges from 2011 PESU <1 km to 8 km off coast of MD and VA); 38°10'N, 75°10'W LABO 21 August–12 September Highland Light, Cape Cod About 31 individuals observed over Miller 1897 1890, and 25 August–13 9 nights in 1890; about 50 individuals September 1891 observed over 14 nights in 1891 LABO 24 November 2006 Eastern Shore of Virginia Mating observed McConnell 2007 National Wildlife Refuge, at southern tip of Delmarva Peninsula, VA LABO, EFLN, Spring 2009–fall 2010 Lakewood, Ocean Co., NJ; Cape Acoustic monitoring Sjollema 2011 LACI, PESU, May Co., NJ; Lewes, Sussex Co., MYSP DE; Worchester Co., MD Northeastern Naturalist Vol. 21, No. 1 T.S. Peterson, S.K. Pelletier, S.A. Boyden, and K.S. Watrous 2014 107 Species Season or date Location Notes Citation LACI September–December Weehawken and Maplewood, NJ; Merriam 1887 1841– 1887 Long Island, Staten Island, and Riverdale, NY; Baltimore, MD LACI 21 August–12 September Highland Light, Cape Cod, MA About 13 individuals observed in Miller 1897 1890 and 25 August–13 1890; about 3 observed in 1891 September 1891 LACI Sometime during May to “Schoodic Site,” which could Zimmerman 1998 September survey be Schoodic Peninsula, Corea Heath, or Big Moose Island, ME LANO October 1885 Staten Island, NY Merriam 1887 LANO 9 observation nights Highland Light, Cape Cod, MA About 13 individuals observed in Miller 1897 between 21 August and 1890; about 28 observed in 1891 12 September 1890, and 14 observation nights between 25–13 September 1891 LANO 28 October 1889 Highland Light, Cape Cod, MA Killed at lighthouse Miller 1897 MYLE Sometime during May– “Schoodic Site,” which could be Zimmerman 1998 September survey Schoodic Peninsula, Corea Heath, or Big Moose Island, ME MYLU, MYSE Summer–early fall Bay of Fundy National Park, 135 and 142 captures, respectively Broders et al. 2006 NB, Canada MYLU, MYSE, Sometime during May– Mount Desert Island, ME (3 km Zimmerman 1998 EPFU September survey off coast)