<tt>Hello list,</tt><br>
<br>
<tt>My dissertation is now finished, and available online at
www.carnivoreconservation.org (currently on the main page, lower-right
side, and will eventually be available through the "Theses" link) or
www.uvm.edu/~bmitchel.</tt><br>
<br>
<tt>TITLE: Coyote Vocal Communication and Its Application to the Selective
Management of Problem Individuals</tt><br>
<br>
<tt>ABSTRACT: Livestock depredation by coyotes severely affects ranchers, and
the existing evidence implicates breeding coyotes in the majority of
livestock losses. Management approaches that target these problem
individuals will be the most effective way to reduce livestock losses. This
dissertation examines coyote long-range vocal communication and the likely
usefulness of recorded vocalizations for selective coyote control.<br>
The information content of barks and howls is important because
coyotes may recognize vocalizing individuals. This could cause coyotes to
respond differently to playbacks depending on the individuals used. It is
also important to understand how vocal characteristics change over
biologically relevant distances, since these changes provide insights into
the practical communicative significance of long-range vocalizations. I
investigated whether coyote barks and howls were individually distinctive
using 293 barks and 280 howls from 7 coyotes. Barks and howls were
individually specific: discriminant analysis correctly classified the
barks of 5 coyotes 69% of the time, and the howls of 6 coyotes 79% of the
time. Howl characteristics did not degrade with distance, and discriminant
analysis was 75% accurate at assigning howls recorded at multiple distances
to 6 individuals. Bark characteristics were unstable with distance and it
is unlikely that barks could be used for individual recognition. Howls and
barks probably serve separate functions: howls are optimized to convey<br>
information, while barks are suitable for attracting attention and for
facilitating distance estimation. Effective playbacks should incorporate
both types of vocalization so that the complementary information they
contain is available to listeners.<br>
A year-long experiment investigated the selectivity and efficacy
of a variety of acoustic stimuli for calling coyotes. Transients rarely
responded vocally, and territorial coyotes commonly responded to group
coyote vocalizations. During optimal conditions, vocal response rates were
over 55% for territorial males, 42% for alpha females, 11% for beta
females, and below 4% for transients. Territorial coyotes were more likely
to approach playbacks than transients, and coyotes more readily approached
group howls than other playback types. When conditions were optimal,
approach response rates were 47% for alphas, 49% for betas, and 27% for
transients. These results suggest that playbacks can be used for selective
coyote control.</tt><br>
<br>
<pre style="margin:
0em;">---------------------------------------------------------------------------
Brian R. Mitchell
Post-Doctoral Associate
University of Vermont
The Rubenstein School of Environment and Natural Resources
George D. Aiken Center
81 Carrigan Drive
Burlington, VT 05405-0088
(802) 656-2496
[EMAIL PROTECTED]
---------------------------------------------------------------------------</pre><br>
|