Distribution and Abundance of the Endangered Hawaiian Hoary Bat, Lasiurus Cinereus Semotus, on the Island of Hawai'i1

Distribution and Abundance of the Endangered Hawaiian Hoary Bat, Lasiurus cinereus semotus, on the Island of Hawai'i1

DAVID S. JACOBS2

ABSTRACT: The endangered Hawaiian hoary bat, Lasiurus cinereus semotus (H. Allen), has an island-wide distribution on the island of Hawai'i. However, abundance estimates suggest that its endangered status is appropriate. Although distributed across wide ranges in elevation, temperature, and rainfall, this bat is most often associated with native vegetation. Its diurnal and seasonal activity patterns argue against migration within the island of Hawai'i. These patterns suggest that the Hawaiian bat uses more than one foraging site and that its use offoraging sites is influenced by changes in insect biomass.

THE HAWAIIAN HOARY BAT, Lasiurus cinereus occurred regularly and to estimate the abun semotus (H. Allen), is the only extant species dance of this bat in such areas. Both visual ofbat in the Hawaiian Archipelago. Although observation and echolocation censuses were it has been on the endangered species list since used. 1970 (Tomich 1986), no recovery plan exists. This is partly because very little is known MATERIALS AND METHODS about the natural history or the distribution and abundance ofthis bat. Information on its The survey was undertaken on the island distribution is limited to reports of incidental of Hawai'i over a 3-yr period (1990 to 1992) sightings on the island of Hawai'i (Baldwin for a total of360 hr spread over 101 days. Bat 1950, Bryan 1955, Tomich 1986) and Maui echolocation calls were monitored by driving (Duvall and Gassman-Duvall 1991) or comes at speeds of 25-40 km hr-1 along roads from surveys peripheral to surveys conducted accessible to two-wheel-drive vehicles (Figure for forest birds (Kepler and Scott 1990). The 1). The echolocation calls were monitored only focused survey on the distribution and using a QMC Mini Bat Detector (QMC In abundance ofthe Hawaiian hoary bat was an struments Ltd., 229 Mile End Road, London echolocation census conducted by J. H. Fullard E1 4AA, England) tuned to a frequency of (unpublished) on Kaua'i. The little that is 30 khz, which is the peak frequency of the known about the distribution and abundance Hawaiian bat's echolocation calls (Belwood of the Hawaiian hoary bat is confusing. For and Fullard 1984). Each section of road was example, it is reported as both common traversed at least twice (once in each direc (R. C. L. Perkins 1913, as cited in Baldwin tion) between 1830 and 2030 hr each night of 1950) and rare (Baldwin 1950, Kepler and survey. Sections of road were surveyed on at Scott 1990) at higher elevations on the island least two successive nights during September of Hawai'i. and October of 1990. The direction that sur The purpose of this study was to locate veys were begun on successive nights was areas on the island of Hawai'i where bats reversed to reduce the effects oftime ofnight. Visual observations were also carried out at localities selected on the basis ofthe type of 1 Manuscript accepted 8 April 1993. habitats in which bats were previously seen 2 Department of Zoology, University of Hawaii at Manoa, Honolulu, Hawai'i 96822. Current address: foraging (Tomich 1986; Hawaiian Heritage Department of Zoology, University of Cape Town, Program, 1992, database of rare and endan Rondebosch 7700, Cape Town, South Africa. gered plants, animals, and natural communi- 193 194 PACIFIC SCIENCE, Volume 48, April 1994 ties of the Hawaiian islands, The Nature remaining 14 sites, bats were observed on two Conservancy, Honolulu, Hawai'i.). Surveys or more consecutive nights, during September at these sites commenced at dusk and lasted and October (Table I). They can therefore be until dark, a period ofabout I hr. The number regarded as sites used regularly by the bats (at of bats at each site was conservatively esti least at some times of the year [see Temporal mated by counting the maximum number of Patterns]). bats that could be seen at anyone time. The largest number ofbats seen in the hour Counts were made every 10 min, and the from dusk to dark at each of the 14 sites is a largest count in the hour of the survey was conservative estimate (Table I). For example, recorded. Because these bats dart in and out at the Red Cinder Road site (at about the of foliage and one's field ofview, they are not 70-mile marker on Highway II), the maxi all visible at the same time. mum number of bats observed was 5, but 12 Three of the sites (Kipahoehoe, Red Cin different individuals were captured at this site der Road, and Ocean Vie~) Were also sur over 14 trap nights. veyed after dark using a bat detector. During In September 1990 I saw a total of 15 bats that time surveys were conducted every half flying, singly or in pairs, upslope from below hour from 1900 to 2400 hr, and every hour Highway II over Kipahoehoe Natural Area from 2400 to 0600 hr, for two nights at the Reserve and disappearing over the horizon first two sites and six nights at Ocean View. high up the slope of Mauna Loa. Similar When using the bat detector each detection sightings have been made over the Manuka was counted as a single bat. A fourth site Natural Area Reserve about 10 miles (16 km) (Wai'ono Ranch) was also surveyed from just to the south ofKipahoehoe (Fujioka and Gon before dawn to about 1 hr after sunrise, on 1988). two consecutive days. Sites where bats were seen on the first night Echolocation Monitoring were surveyed again at least once. With the exception ofOcean View, which was surveyed Almost all ofthe bats detected by echoloca from May 1991 to October 1992; Pohakuloa, tion monitoring were concentrated south of which was surveyed from May to October Kailua (Kona) between the 80- and 96-mile 1992; and Hawai'i Volcanoes National Park, markers along Highway II (Figure I). A total which was surveyed in October 1992, all sites of 18 bats (i.e., 18 detections) were detected were surveyed between September and Octo between the 80- and 96-mile markers over two ber 1990, and from February to August the nights of survey (4.25 ± 1.78 bats per tra following year. The insect fauna at Ocean verse). Not a single observation or echoloca View was sampled by means of a 22-watt tion detection was made in northwestern and ultraviolet insect light trap (BioQuip Prod southwestern Kohala on the leeward side of ucts), from February to October 1992. the Island. However, 40% ofall bat sightings occurred on the drier leeward side of the RESULTS Island south of Kailua, compared with 26% on the wetter windward side. Visual Observation Bats were seen at 16 of the 26 sites visited Context ofObservations (Figure I). Two of the 16 sites were visited ECOLOGICAL. The bats were all in flight and only once. These were located at Pu'uhonua obviously foraging at the 16 sites where they o Honaunau Historical Park on the leeward were seen. They displayed the darting, weav coast (maximum count of two bats in I hr of ing flight, with sudden turns and dives, char visual observation) and at the entrance to acteristic of foraging insectivorous bats that the Kulani Correctional Facility (maximum catch their prey on the wing. Bats detected count of three bats in I hr of visual observa (using the bat detector) along Highway II tion) at the end ofStainback Highway. At the may have been foraging as well. Visual obser- LEGEND Echo Location Survey 0 No Bats • Bats Seen l), Bats Heard

Miloii N i

FIGURE I. Distribution of Hawaiian hoary bat sightings and echolocation detections on the island of Hawai'i. Circles indicate 26 sites surveyed for bats: open circles indicate sites at which no bats were seen; closed circles indicate sites where bats were seen on at least two consecutive nights. Dashed lines indicate the roads along which echolocation surveys were conducted; triangles indicate areas where bats were heard during the echolocation surveys. 196 PACIFIC SCIENCE, Volume 48, April 1994

TABLE I 1990 appeared to be in transit. These bats NUMBERS OF BATS AT EACH FORAGING SITE ON were flying very high (> 150 m) and did not THE ISLAND OF HAWAI'I display the darting and weaving flight charac teristic of a foraging bat. % OF VISITS BATS WERE NO. OF NO. OF TEMPORAL PATTERNS. Diurnal activity pat LOCATION OBSERVED VISITS BATS terns. The commencement and duration of Kipiihoehoe Natural 100 5 10 foraging activity varied slightly from site to Area Reserve site. For example, during September and Oc (Highway 11,90- to tober (1990 and 1991) bats started foraging at 92-mile markers) Red Cinder Road at 1800 hr or immediately Manukii Natural Area 100 5 4 thereafter. At the Ocean View and Kipiiho Reserve (Highway II, 80- to 82-mile ehoe sites, on the other hand, they first ap markers) peared between 1830 and 1845 hr. At the latter Red Cinder Road (off 77 13 5 two sites the bats remained active until about Highway II, near 0200 hr. There was, however, a peak in activ 70-mile marker) Highway II (70-mile 60 5 7-10 ity just before sunset (four bats visible in the marker) hour from dusk to dark), which dropped Kaimii Bay (Puna) 100 2 3-4 slightly as the night progressed (two bats Kalapana (Puna, over 100 2 5 detected per hour). At Red Cinder Road, new lava flow) activity came to an almost complete stop at Kal6pii State Park 100 2 2 (Hiimiikua) about 1900 hr. In five nights ofsurveying, only Honoka'a (St. George's 100 5 2-3 one bat was detected briefly after 1900 hr-at Cemetery, Lehua 2100 hr on two separate occasions. Pohakuloa Street) (Bradshaw Military Airfield) was different in Waimanu Bay (North 100 2 5 Kohala) that there did not appear to be a drop in Ke6kea Bay (North 100 2 II activity. The number of bats remained con Kohala) stant at least until 0230 hr when observation Ocean View 100 6 4 ceased. (intersection of Early morning foraging was observed. Be Hukilau and Tree Fern avenues, South tween three and five bats were observed on Kona) two consecutive mornings foragingjust before P6hakuloa (Bradshaw 100 7 7-10 d~wn and for at least 45 min after sunrise at Military Airfield) Wai'ono Ranch (Kailua-Kona). At Kipiiho Wai'ono Ranch 100 4 3-5 (Kailua-Kona) ehoe, the only other site at which bats were Hawai'i Volcanoes 100 2 2-3 observed foraging at dawn, a single bat was National Park seen on one of the two mornings ofsampling. Early morning foraging in bats has also been NOTE: Data are results ofsurveys conducted between August and October (see text). reported on Maui (Duvall and Gassman Duvall 1991). Seasonal activity patterns. There appeared to be a drop in activity at most foraging sites vations offoraging bats before nightfall along between February and August (Table 2). At Highway II, especially those parts ofthe road Kipiihoehoe, Manukii, and Red Cinder Road, at or near Kipiihoehoe and Manukii Natural bat sightings dropped to three bats over three Area Reserves, suggest that the bats detected nights of sampling. Bats did not spend much after nightfall along this section of road may time at these sites. Most of them made one to have been foraging as well. three foraging circuits before disappearing. At The 15 bats observed flying over Kipii Ocean View, sightings dropped to one bat hoehoe Natural Area Reserve in September over four nights of sampling in February and Hawaiian Hoary Bat on Island of Hawai'i-JAcoBs 197

TABLE 2 NUMBERS" OF BATS AT VARIOUS FORAGING SITES FOR DIFFERENT MONTHS

LOCATION FEBRUARY MAY JUNE AUGUST SEPTEMBER OCTOBER

Ki"pahoehoe Natural Area Reserve 0.5 ± 0.5 0 I 10 ± 0.0 (2) (2) (I) (2) Manuka Natural Area Reserve 0.5 ± 0.5 I ±O.O 0 4 ± 0.0 (2) (2) (I) (2) Red Cinder Road 0.5 ± 0.5 0 0 5 ± 0.0 3.6 ± 2.2 (2) (2) (I) (2) (II) Highway II (70-mile marker) 0 0 0 5.7 ± 2.6 (2) (2) (I) (2) Kaimii Bay 0 0 0 3.5 ± 0.5 (2) (2) (I) (2) Kal6pa State Park 0 0 0 2 ± 0.0 (2) (2) (I) (2) Ke6kea Bay 0 0 0 II ± 0.0 (2) (2) (I) (2) Ocean View 0.25 ± 0.5 3.4 ± 0.5 3.4 ± 0.5 5.6 ± 2.5 4.5 ± 1.7 3.5 ± 1.7 (4) (5) (2) (4) (4) (2) P6hakuloa (Multipurpose Range 0.75 ± 0.5 0.2 ± 0.5 0.7 ± 1.0 0.5 ± 0.6 2.7 ± 2.1 Complex)' (4) (4) (4) (4) (4)

• Mean ± SD, sample size in parentheses. ·See texl for year. 'Data from Hawaiian Heritage Program database (see text). to four bats over three nights of sampling in or detected at any of the other sites between March. From May to August the numbers of February and August during the survey bats increased at this site and returned to the reported here (Table 2). Bat numbers re September/October (1991) levels (Table 2), mained constant at Pohakuloa (Bradshaw with the exception of a decrease in activity Military Airfield) from May to October. during July and October (Figure 2). The drop in activity at Ocean View seems to be corre lated with a drop in insect biomass (Figure 2). Environmental Correlates In the Multipurpose Range Complex of the HABITAT CHARACTERISTICS. The 14 forag Pohakuloa Military Training Area, along the ing sites can be divided into four habitat Bobcat trail (1800 m elevation), between one categories: native vegetation, exotic vegeta and three bats were seen and/or detected over tion, mixed (native and exotic) vegetation, a sampling period of four nights for each and open ocean (Table 3). Of the 81 bats month between February and August 1992 observed at these sites (Table I), 44% were (Table 2). This number increased to II bats associated with native vegetation, 16% with over four sampling nights in August (Hawai exotic vegetation, 9% with mixed vegetation, ian Heritage Program database). Incidental and 25% foraged over open ocean. Bats were sightings of bats at Hakalau Forest National therefore associated more with native vegeta Wildlife Refuge (1890 m elevation) by Jaan tion than with exotic vegetation, contrary to Lepson (pers. comm.) followed the same pat what Kepler and Scott (1990) found. tern. Although bats were seen between Janu ary and August, most sightings occurred from TEMPERATURE. The foraging sites of the September to December. No bats were seen Hawaiian bat extend over an elevational 1.2 6 EJ Insects

~ Bats 1.0 5 c: Q) Q) (/)

0.8 4

TABLE 3 ELEVATION AND VEGETATION ASSOCIATED WITH EACH BAT FORAGING SITE

LOCATION ELEVATION (m) VEGETATION TYPED

Kipiihoehoe Natural Area Reserve 386 '0 'hia lowland mesic forest N Manukii Natural Area Reserve 548 '0 'hia lowland mesic forest N Red Cinder Road 593 '0 'hia lowland mesic forest/eucalyptus M Highway II (70-mile marker) 593 Eucalyptus trees/exotic shrubs E Kaimii Bay Sea level Open ocean Kalapana 60 Open lava Kal6pii State Park 549 '0 'hia/eucalyptus forest M Honoka'a 354 Eucalyptus/macadamia orchard E Waimanu Bay Sea level Open ocean Ke6kea Bay Sea level Open ocean Ocean View 665 'O'hia lowland mesic forest N P6hakuloa 1,830 Mamani-naio (Sophora chrysophylla-Myoporum N sandwicense) open woodland Wai'ono Ranch 1,070 '0 'hia forest interspersed with grass fields N Hawai'i Volcanoes National Park 1,220 Koa (Acacia koa) forest N

aN, nalive; E, exotic; M, mixed (native and exotic). Hawaiian Hoary Bat on Island of Hawai·i-.JACOBS 199 range of 0-1830 m (Table 3). The climatic wicense) forest, wind speed ranged from 3 1 conditions experienced by these bats therefore to 6 km hr- , occasionally gusting to 11 km 1 range from tropical to temperate. At eleva hr- , on 60% of the nights surveyed. tions below 1500 m, temperatures above 35°C are infrequent and the range between the coldest and warmest months averages 6.5°C. The average temperature at Hilo (sea level) in DISCUSSION August is 24.1°C and in January is 21.6°C. Visual observations and echolocation The average temperature at Mauna Loa Ob monitoring suggest that the Hawaiian hoary servatory (3398 m) in January is 5°C and in bat is fairly widespread around the island of August is 8°C (Blumenstock and Price 1972). Hawai'i. Although bats appear to be more However, at times the upper slopes and sum abundant on the drier leeward side of the mits of Mauna Loa and Mauna Kea are Island, they may not be as rare on the wind covered in snow during winter months. Cold ward side of the Island as previously thought air formed immediately above this snow cover (Kepler and Scott 1990). The largest foraging descends into areas like Pohakuloa and aggregation (II bats) was seen in September Hakalau, causing temperatures as low as 1990 over Keokea Bay on the windward coast. - 15°C (Blumenstock and Price 1972). The Large groups of bats have been reported on Pohakuloa and Hakalau sites are at eleva the windward side of the island: 12 bats over tions of about 1830 and 1890 m, respec Hilo Bay in September 1963 and 22 bats at tively. The result is that these sites often Honoka'a landing in October 1964 (Tomich experience frost during winter. 1986). A more extensive survey of that coast line needs to be undertaken. I have received RAINFALL. Bats were observed most fre reports from surfers of bats foraging over quently in areas with moderate rainfall. Eight water all along that coast. of the 14 foraging sites had a mean annual There seems to be a decrease in bat activity rainfall ranging from 1016 mm to 1905 mm. from February to August (Table 2). The drop Four had a mean annual rainfall ranging from in activity occurs over a wide elevational and 1905 to 2540 mm. Pohakuloa had a mean geographical range. In addition, bats are pres annual rainfall of 508 mm (rainfall figures ent as early as May at Ocean View and from Armstrong [1973]). Pohakuloa (Bradshaw Military Airfield), two Bats at Red Cinder Road and Kipahoehoe sites with very different elevations and geo sites generally were not active when it rained. graphical locations. These patterns argue In fact on nights when it rained just before or against an elevational or geographic migra just after the time the bats were usually seen tion within the island of Hawai'i. at those sites, the foraging aggregations did The drop in bat activity with the drop in not appear and at most one or two bats were insect biomass at Ocean View suggests that seen. However, at Kalopa State Park on the these bats may use a number of foraging sites windward side of the Island, two bats were over a course of a night or a few days. The observed foraging in a light rain for at least particular sites used depend on local insect 30 min. biomass. Radio telemetry data (unpublished WIND. In general there was little to no data) indicate that bats do use more than one wind at the inland foraging sites. The tall foraging site per night and that a bat may stay vegetation characteristic of most of these away from a previously visited site for up to sites provided a natural windbreak. However, 2 weeks. If this is the case, the apparent drop these bats are fairly strong flyers. I have in activity from February to August at some observed them flying without much difficulty sites may be associated with times when bats in fairly strong winds. At Pohakuloa, where the were foraging elsewhere because oflow insect vegetation was predominantly low mamani biomass at those sites on nights when they naio (Sophora chrysophylla-Myoporum sand- were surveyed. 200 PACIFIC SCIENCE, Volume 48, April 1994

The apparent increase in bat actlVlty in LITERATURE CITED September/October (in terms. of the number of sites at which bats were seen) may be the ARMSTRONG, R. W., ED. 1973. Atlas of result of young of the year fledging. Capture Hawaii. Department of Geography, Uni data (unpublished data) suggest that lactation versity of Hawaii. The University Press of occurs in June and July and young are fledged Hawaii, Honolulu. sometime in August. BALDWIN, P. H. 1950. Occurrence and behav In conclusion, the numbers of bats re ior ofthe Hawaiian bat. J. Mammal. 31(4): ported here for the different sites at different 455-456. times of the year suggest that the Hawaiian BELWOOD, J. J., and J. H. FULLARD. 1984. hoary bat is rare and that its endangered Echolocation and foraging behavior ofthe status is appropriate. The fact that the habi Hawaiian hoary bat, Lasiurus cinereus tats used by this bat are more often than not semotus. Can. J. Zool. 62:2113-2120. associated with native vegetation suggests BLUMENSTOCK, D., and S. PRICE. 1972. Cli that the major threat to its existence in mates of the states: Hawaii. Pages 155 Hawai'i may be the destruction of native 204 in E. A. Kay, ed. Natural history of vegetation. Protection of native vegetation is, the Hawaiian Islands. University ofHawaii therefore, essential to the survival of Lasiurus Press, Honolulu. cinereus semotus in Hawai'i. This may become BRYAN, E. H., JR. 1955. The Hawaiian bat. even more evident when the roosting ecology Elepaio 15: 63-64. of this bat is studied. DUVALL, F., and R. GASSMAN-DuVALL. 1991. No bats on Maui? Look again. Elepaio 51: 1-2. FUJIOKA, K. K., and S. M. GON III. 1988. Observations of the Hawaiian bat, La ACKNOWLEDGMENTS siurus cinereus semotus, in the districts of Ka'u and south Kona, island of Hawaii. J. I am grateful to P. Q. Tomich and Chuck Mammal. 69: 369-371. and Danny Stone for their help and encour KEPLER, C. 8., and J. M. SCOTT. 1990. Notes agement, and to Sheila Conant, Lenny Freed, on distribution and behavior ofthe endan and Michael Hadfield for their valuable com gered Hawaiian hoary bat (Lasiurus cine ments during the preparation of this paper. reus semotus), 1964-1983. Elepaio 50(7): This survey was funded by the Forestry and 1-6. Wildlife Division, Department of Land and TOMICH, P. Q. 1986. Mammals in Hawai'i Natural Resources, Hawai'i. 2nded. Bishop Mus. Spec. Publ. 76: 1-375: