Prehibernation habitat use and foraging activity by endangered Ozark big-eared bats (Plecotus townsenddi ingens).

INTRODUCTION

The Ozark big-eared bat (Plecotus townsendii ingens) has been listed as endangered since 1979 (Bagley, 1984). The highest estimate of the population in Oklahoma since federal listing was 1700 in 1990 (Clark, 1991). The historical range of this subspecies included southwestern Missouri, northwestern and N-central Arkansas and eastern Oklahoma (Harvey, 1992). Currently, only one hibernaculum and one maternity colony are known in N-central Arkansas (Harvey, 1992), and three hibernacula and four maternity colonies (represented by five caves) are known in eastern Oklahoma (Clark, 1991). The Missouri colonies appear to have been extirpated (U.S. Fish and Wildlife Service, 1995).

Little is known of the specific habitat requirements of Plecotus townsendii ingens. Recent telemetry studies of foraging adult females from an Oklahoma maternity colony characterized some aspects of foraging ecology during the maternity season; females used edge habitat more than expected and traveled further from caves to forage as the season progressed (Clark et al., 1993). Clearly, knowledge of habitat requirements throughout the rest of the year is needed to protect and recover this critically endangered subspecies (U.S. Fish and Wildlife Service, 1995).

Our study was designed to follow Clark et al. (1993) and evaluate habitat use and foraging activity by Plecotus townsendii ingens during the period between maternity colony breakup and hibernation. We quantified emergence times, number of foraging sites per bat, distance traveled to foraging sites and size of foraging areas. We expected no change among study periods or between this study and that of Clark et al. (1993) in emergence times, number of foraging sites per bat and size of foraging areas. We hypothesized that distance traveled to foraging sites would increase after the maternity season when females no longer needed to return to maternity caves to nurse young. We also re-evaluated habitat use and expected a continued preference for edge habitat as reported by Clark et al. (1993).

STUDY AREA AND METHODS

Our study was conducted in Adair County, Oklahoma, in the Boston Mountain Plateau subdivision of the Ozark Plateau physiographic region (Madole et al., 1991). Climate was humid subtropical, characterized by hot humid summers and relatively mild winters punctuated by sporadic periods of severe cold. Mean annual temperature was 13-17 C, and mean annual precipitation was 100 cm (Madole et al., 1991). Vegetation in this area was dominated by eastern deciduous oak (Quercus spp.)-hickory (Carya spp.) forests.

Radio telemetry. - Bats were trapped at a maternity cave (used by Clark, 1991) and a transient cave, separated by ca. 1 km. Focus was intended to be at the maternity cave, but low bat numbers made trapping at the transient cave necessary. Bats were captured by draping a mist-net over cave entrances. For each bat captured, age, sex, weight, forearm length and reproductive condition were determined. Skin Bond[R] cement was used to attach 0.65-0.75 g radiotransmitters (Model BD-2B, Holohil Systems Ltd., Ontario, Canada), with frequencies of 150.8-151.8 MHz, between scapulae of bats.

Miniature transmitters have allowed researchers to use radiotelemetry to assess aspects of foraging behavior for insectivorous bats. However, the recommendation that transmitters should be less than 5% of the animal's body mass (Cochran, 1980) is often exceeded (Tidemann et al., 1985; Hickey and Fenton, 1990; Brigham, 1991; Clark, 1991; Adam et al., 1994); in our study, transmitters ranged from [Mathematical Expression Omitted] of body mass, given bat masses of 10.25-15.25 g (Wethington, 1994; Appendix I). Transmitter weight could decrease maneuverability due to increased wing loading (weight/wing area) and potentially could decrease foraging efficiency (Aldridge and Brigham, 1988). However, by observing the behavior of bats with and without transmitters, researchers have shown that careful radio-tagging does not affect emergence times (Wai-Ping and Fenton, 1989), foraging times (Hickey and Fenton, 1990) or foraging success (Hickey, 1992). Although these studies all involved transmitters less than 5% of body mass and our study exceeded this limit slightly, we assumed, as did Clark et al. (1993), that the behavior of our radio-tagged bats did not differ from untagged bats.

Bats were tracked during 11-22 October 1991, 23 August-3 September 1992, and 10-23 September 1992 with TRX-1000s receivers and 3-element Yagi antennae (Wildlife Materials, Inc., Carbondale, Ill.). Directional fixes were determined by the loudest signal method (Springer, 1979) at 2-min intervals. Constraints inherent to tracking bats in rugged topography made it necessary to gather data opportunistically for all bats as available.

Efforts were made to reduce, or at least account for, telemetry error by locating a known transmitter to build radio-tracking skills and become familiar with …