Intra- and interspecific responses to Rafinesque’s big-eared bat (Corynorhinus rafinesquii) social calls. Page: 3 of 8
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to UNT Digital Library by the UNT Libraries Government Documents Department.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
DE- A109-00SR22188
Journal Article
Responses to social calls of C. rafinesquii
We recorded calls at the entrance from 20:38 EDT to 20:52 EDT
and then moved the receiver back approximately 15 m from the
entrance to record bats as they exited the roost until 21:15 EDT;
there were no apparent qualitative differences between calls
recorded in the roosts or as bats exited. On 8 May 2008 we
placed the AR125 receiver in the base of a roost tree at SRS that
contained six Rafinesque's big-eared bats. We oriented the re-
ceiver at approximately 450. Calls were recorded from 20:36
EDT to 21:11 EDT. We used SCAN'R software (Binary Acoust-
ics Technology, Tucson, Arizona) to delete all files that did not
contain bat calls. Remaining call files were examined in Bat
Sound (Pettersson Elektronik AB, Uppsala, Sweden) and typical
calls for each site were copied into the playback file for that
site. We selected 85 calls for the LM file and 135 calls for the
SRS file.
Playback Procedures
Playback experiments were conducted from mid-June
through mid-August 2008. Two nets were set at each site. Rafi-
nesque's big-eared bats rarely forage > 1 km from their roosts
(Hurst and Lacki, 1999; Menzel et al., 2001). We set nets near
known roost sites to ensure that there were a sufficient number
of bats within the vicinity to respond to playback calls but far
enough away to avoid catching bats as soon as they exited the
roost. At LM we set single high 6-m wide nets across trails lead-
ing to the two mine entrances. The nets were approximately
50-100 m from the entrances and approximately 100 m from
each other. At SRS we set two double high 6-m wide nets per-
pendicular to a road; the net sites were approximately 500 m
from six known roost sites (five trees and one building). The
nets were approximately 60 m from each other. The CNP net
site was along a road near (200-500 m) a number of trees that
had been used for roosting in 2007. Two single high 6-m nets
were set perpendicular to the road and were approximately 50 m
from each other. Playback experiments were conducted for four
nights at LM and SRS but for only one night at CNP because we
were not able to record bats there.
Each trial lasted four hours. We opened both nets at dusk
and monitored each continuously. We designated one of the nets
as the Experimental net by a coin toss; the other net served as
the Control. Two BAT AT 100 ultrasonic transmitters (frequency
range 20-120 kHz; 100 dB SPL at 1 m) on tripods were orient-
ed at approximately 450 from horizontal and placed at the cen-
ter of the Experimental net, one on each side of the net. The
transmitters faced in opposite directions and were perpendicular
to the net (00 and 1800). Beginning at dusk, the playback se-
quence was broadcast for 5 min in a continuous loop followed
by 5 minutes of no playback. During the 5 min period of no
playback, we rotated the transmitters so that they were oriented
at 600 and 2400 for the second 5 min playback period and then
oriented at 1200 and 3000 for the third period of playback. This
sequence was repeated so that each orientation was sampled
twice during the hour. At the end of the first hour of netting, we
moved the microphones to the Control net which became the
Experimental net for Hour 2; the Experimental net during Hour
1 became the Control net during Hour 2. At the end of Hours 2
and 3, the microphones were again switched so that each net
served as a Control and an Experimental net two times in a night.
Bats at LM and SRS were exposed to calls recorded at their own
site as well as at the other site; we used the SRS calls at CNP.
For each capture we recorded time of capture, whether the trans-
mitter was on or off, and the species, age (adult or juvenile), andsex of each bat. Bats were banded with numbered alumi-
num lipped bands (Porzana Ltd., East Sussex, UK) to monitor
recaptures.
General bat activity in the vicinity of the nets was recorded
at both nets using AnabatII bat detectors connected to CF-
ZCAIM recorders (Titley Scientific, Stones Corner, Australia).
Anabats were placed approximated 15-20 m from the net and
oriented toward the net. The AT100 transmitters are highly
directional and we were careful not to set them so that they
faced the Anabats. Calls recorded by the Anabat detectors were
filtered in Analook (Version 4.9j) to remove insect and other
noise (Britzke, 2003). Files that passed this filter were used as
an index of bat activity. We then used an identification filter
(Britzke and Murray, 2000) to select search phase calls. We used
a combination of quantitative and qualitative methods to identi-
fy calls to species. First we used a discriminant function model
to identify species (Britzke, 2003) and then qualitatively evalu-
ated each identified pass to ensure or correct the identification.
Qualitative identification was particularly important because
C. rafinesquii and Myotis austroriparius were not well repre-
sented in the call library.
Statistical Analysis
Call parameters for each playback file were extracted using
SCAN'R software. Two harmonics were evident in most of the
social calls recorded at both sites. However, SCAN'R exports
parameters for one or the other harmonic, depending on which
is encountered first. Thus, statistical comparisons were conduct-
ed for each harmonic. We determined the means and standard
errors of 6 call parameters for each site and used two-sample
t-tests to determine whether call parameters differed between
LM and SRS. Parameters were duration, maximum frequency,
minimum frequency, bandwidth (maximum frequency - mini-
mum frequency), dominant frequency (frequency at the strong-
est sound pressure level), and lower slope (from the high end of
the characteristic frequency or knee to the low end of the char-
acteristic frequency). We used PROC STEPDISC (SAS, 2002)
to determine the variables that discriminated between calls from
the two sites. All six variables were included in the analysis and
site (LM or SRS) was the class variable. We used the default cri-
terion for entry and removal (a = 0.15). A binomial test (PROC
FREQ) was used to test whether the proportion of captures in
the Control and Experimental nets differed on nights when we
captured bats. Median tests (PROC NPAR1 WAY) were used to
test whether the number of bat passes and calls differed between
Control and Experimental nets.
RESULTS
Call Structures
Calls recorded at LM and SRS were generally
similar in structure (Fig. 1) but differed from echolo-
cation calls (Fig. 2). Echolocation calls tended to
have a constant frequency component at the begin-
ning of the call (Fig. 2) which the social calls lacked.
The lower harmonic of calls of the LM social calls
was significantly longer, had a higher maximum
frequency, and a wider bandwidth than calls of SRS2010
10-13-P
331
Upcoming Pages
Here’s what’s next.
Search Inside
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Loeb, Susan, C. & Britzke, Eric, R. Intra- and interspecific responses to Rafinesque’s big-eared bat (Corynorhinus rafinesquii) social calls., article, July 1, 2010; New Ellenton, South Carolina. (https://digital.library.unt.edu/ark:/67531/metadc837246/m1/3/: accessed March 28, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.