In¯uence of prerelease experience on reintroduced black-footed ferrets (Mustela nigripes)

advertisement
Biological Conservation 89 (1999) 121±129
In¯uence of prerelease experience on reintroduced black-footed
ferrets (Mustela nigripes)
Dean E. Biggins a,*, Astrid Vargas b, Jerry L. Godbey a, Stanley H. Anderson c
a
U.S. Geological Survey, Midcontinent Ecological Science Center, 4512 McMurry Avenue, Fort Collins, Colorado 80525, USA
U.S. Fish and Wildlife Service, National Black-footed Ferret Conservation Center, 410 Grand Avenue, Suite 315, Laramie, Wyoming 82070, USA
c
U.S. Geological Survey, Wyoming Cooperative Fish and Wildlife Research Unit, University of Wyoming, Laramie, Wyoming 82071, USA
b
Received 24 June 1998; received in revised form 24 November 1998; accepted 4 December 1998
Abstract
Captive breeding is increasingly being used to create supplies of endangered animals for release into natural habitats, but rearing
strategies vary and debates arise over which methods are most ecient. We assessed postrelease behaviors and survival of three
groups of black-footed ferrets, each with di€erent prerelease experience. Eighteen ferret kits 60 days of age were moved with their
dams from cages to 80-m2 outdoor pens with prairie dog burrows. These animals were compared to animals reared in standard
cages (n ˆ 72), some of which were given experience killing prairie dogs (n ˆ 32). Ferrets were released onto white-tailed prairie dog
(Cynomys leucurus) colonies in Wyoming, USA, in fall, 1992. Radio-tagged cage-reared ferrets made longer nightly moves and
dispersed further from release sites than their pen-reared counterparts. The band return rate was 4-fold higher for pen-reared animals than for cage-reared animals during surveys conducted about 1 month after release. We recommend routine use of quasinatural outdoor pens for prerelease conditioning of black-footed ferrets. Published by Elsevier Science Ltd.
Keywords: Behavior; Black-footed ferret; Mustela nigripes; Reintroduction; Survival
1. Introduction
Populations of black-footed ferrets declined throughout this century to near extinction by the late 1970's
(Biggins and Schroeder, 1988). A small population was
discovered near Meeteetse, Wyoming, USA in 1981
(Fig. 1), but that population was decimated by canine
distemper and plague (Yersinia pestis) in 1985 (Forrest
et al., 1988). The last animals were removed by 1987 to
establish a captive population (Miller et al., 1996), with
the ultimate goal of reintroduction.
Captive-raised animals of many species have had poor
success after reintroduction into their natural habitat
(see reviews in Wemmer and Derrickson, 1987; Grith
et al., 1989; Beck et al., 1994; Wilson and Stanley-Price,
1994). To maximize survival, reintroduction candidates
must be able to procure food and shelter, develop antipredator skills, interact properly with conspeci®cs, and
orient (navigate, migrate, and/or disperse) in a structurally complex environment (Kleiman, 1989; Box, 1991).
Prerelease ``training'' given to several mammal species
* Corresponding author. Tel.: +1-970-226-9467; fax: +1-970-2269230; e-mail: dean-biggins@usgs.gov.
0006-3207/99/$ - see front matter Published by Elsevier Science Ltd.
PII: S0006-3207(98)00158-X
has altered behaviors in ways assumed to be bene®cial
to survival (Kleiman et al., 1986; Stanley-Price, 1989;
Phillips, 1990; Soderquist and Serena, 1994). Golden
lion tamarins pre-conditioned in a naturalistic wooded
area had higher survivorship after release in the Brazilian forests (Beck, 1991). Masked bobwhites (Colinus
virginianus ridgwayi) cross-fostered to wild neutered
Texas bobwhites (C. v. virginianus) and birds trained for
physical ®tness and anti-predatory behaviors survived
better than untrained birds (Ellis et al., 1978).
Laboratory experiments have detailed additional
e€ects of environmental enrichment. Behaviors of
domestic gerbils (Meriones unguiculatus), for example,
were altered toward the wild phenotype after one generation of rearing in tunnels (Clark and Galef, 1980).
Compared to rats (Rattus sp.) raised in simple cages,
rats raised in more complex environments were found to
have more synaptic connections and heavier brains
(Rosenzweig, 1979), and were better at negotiating
mazes (Greenough and Juraska, 1979) and escaping
predator models (Renner, 1988). Behavioral di€erences
between breeds of dogs (Canis familiaris) were no longer
present when the animals were reared in a quasi-natural
setting (Scott and Fuller, 1965).
122
D.E. Biggins et al. / Biological Conservation 89 (1999) 121±129
To gain knowledge and experience in preparation for
black-footed ferret releases, studies were ®rst conducted
on the closely related Siberian polecat (Mustela eversmanii) to investigate the development of food searching
and predator avoidance behaviors (Miller et al., 1990a,
b). When introduced to a novel arena, polecats familiar
with underground burrow systems spent less time
aboveground than did their cage-reared counterparts.
Studies of captive polecats were followed by trial releases of neutered polecats reared in cages and pens similar
to those described below, and of neutered wild-caught
polecats translocated from China (Biggins et al., 1990,
1991). Behaviors and survival di€ered among some
groups. Predation from coyotes (Canis latrans), badgers
(Taxidea taxus), and avian predators caused rapid loss
of both captive-reared and wild-caught polecats.
After several years of successful captive breeding, 49
ferrets were released in Shirley Basin, Wyoming (Fig. 1),
in the fall of 1991. No comparative evaluations of rearing or release techniques, however, were made using
those animals. All ferrets were raised in cages, and were
released with postrelease support in the form of cages
that held animals during an adjustment period of about
10 days; after release, a ferret could return to its cage for
food and refuge.
In 1992, our objective was to evaluate the e€ect of
three rearing strategies on behavior and survival of
released black-footed ferrets. Some ferrets were raised in
standard cages with either of two levels of experience
killing prey; other ferrets were reared in outdoor pens.
In the fall, 1992, these ferrets were released in Shirley
Basin (Fig. 1) and samples of animals from each
experimental group were monitored via radiotelemetry.
2. Methods
2.1. Captive-breeding and rearing
Black-footed ferrets used in these experimental releases were young-of-the-year raised at the Sybille Wildlife
Research facility operated by the Wyoming Game and
Fish Department, the Henry Doorly Zoo at Omaha, the
Louisville Zoo and the National Zoo's Conservation
and Research Center (Table 1). Captive-breeding and
rearing procedures have been standardized (Williams et
al., 1991), but modi®cations were added for animals
that were to be involved in reintroductions. These animals were divided into the following 3 treatment groups
(Table 1).
(1) ``Standard operating procedures'' (SOP) ferret kits
were reared in cages with a surface area of 1.5 m2 supported on 1 m legs (Vargas, 1994; Vargas and Anderson, 1996). The SOP ferrets initially were given live
hamsters (Mesocricetus auratus) three times weekly
starting at age 8 weeks and were fed prairie dog meat
about 3 times weekly from 60 days of age, until their
release. The SOP group came from all 4 rearing facilities. Four SOP ferrets were hand-raised due to death of
their mother, and one was cross-fostered to a Siberian
polecat.
Table 1
Group sizes and rearing facilities for young black-footed ferrets
released during 1992 in Shirley Basin, Wyoming
Facility of origin
Sybille Omaha
<< ,, << ,,
Louisville
<<
,,
National Total
<< ,,
SOPa
(Radio-tagged)
12
(1)
5
(3)
PDb
(Radio-tagged)
18 14
(11) (6)
32
(17)
PENc
(Radio-tagged)
14 4
(7) (1)
18
(8)
TOTAL
(Radio-tagged)
66
(26)
Rearing:
a
Fig. 1. Historic range of the black-footed ferret, location of the ®nal
known population (Meeteetse, Wyoming, USA) extirpated in 1987 to
protect the remnant ferrets from a disease outbreak and to initiate
captive-breeding, and location of the ®rst reintroduction site (Shirley
Basin, Wyoming). (Range map was adapted from Miller et al., 1996).
4
5
5
(2) (1)
10
(3)
7
(5)
12
(8)
2
0
2
(0)
40
(12)
90
(37)
SOP=ferrets raised in standard indoor cages, and having no
experience with live prairie dogs.
b
PD=ferrets raised cages and given experience with predation on
prairie dogs.
c
PEN=ferrets born in cages, allowed access to outdoor pens with
prairie dog burrows when kits were about 60 days old, and fed live
prairie dogs.
D.E. Biggins et al. / Biological Conservation 89 (1999) 121±129
(2) The second treatment group (PD) was raised at
Sybille in SOP-type cages, fed live hamsters twice weekly
starting at age 8 weeks, and exposed to live white-tailed
prairie dogs (Cynomys leucurus) in weekly trials at 13±
16 weeks. A prairie dog was given to a family group of
ferrets, consisting of kits and their mother, in each trial.
Individual experience varied from witnessing kills and
interacting with the freshly-killed prairie dog carcass to
performing several kills (Vargas, 1994).
(3) The third group (PEN) was raised in outdoor pens
at the Sybille facility. Pens were about 80 m2 in surface
area, with soil to a depth of 2.4 m. The enclosures were
constructed of heavy-gauge wire mesh (``cyclone'' fencing) on 4 sides and the bottom, with roo®ng of solid
materials to shed precipitation. White-tailed prairie
dogs had been introduced previously and had established burrow systems. Pregnant female ferrets whelped
in cages (1.1 m2) within these pens, and were allowed
access to the pens when kits were about 2 months old.
Prairie dogs were added to the pens as needed. At least
one kit from each outdoor litter was individually
exposed to a live prairie dog at 16.5 weeks of age as part of
a study evaluating eciency of killing (Vargas, 1994).
2.2. Marking and radio-collaring
Animals produced at Sybille were anaesthetized, given
a thorough physical examination, vaccinated, marked
with transponder implants (Fagerstone and Johns,
1987) and numerical tattoos in the ear, and radio-tagged
(Fagerstone et al., 1985) several days before they were
shipped to Shirley Basin. Ferrets from other facilities
were handled and radio-tagged after their arrival at the
Shirley Basin ®eld site.
Thirty-seven ferrets were out®tted (Table 1) with collars consisting of a 6-gram transmitter package mounted with Te¯on heat-shrink tubing to a 1 cm wide wool
neckband. The Te¯on coating resists mud buildup and
the wool is degradable, with stretch and wear allowing
the collar to be shed by the animal in several days to
several weeks.
2.3. Release
Ferrets were delivered to the 20,596 ha prairie dog
complex in Shirley Basin (Fig. 1) between 9 September
and 12 October at 16.5±18 weeks, held in release cages
for 10 days after their arrival, and released in the ®rst 2
h after sunset. Supplemental food (prairie dog meat)
was provided in the release cages as long as evidence of
ferret use existed (ferret sightings, scat, disappearing
food) and for several days thereafter.
Two types of release cages, an aboveground version
and a surface/belowground combination, were used.
Aboveground cages were 0:9 1:2 m in ¯oor area, were
on 0.9 m legs, and had a plastic tunnel extending to
123
ground level for release of ferrets. The second version
had a similarly-sized cage of wire mesh at ground level,
connected to a belowground nest box that was protected by a sheet metal outer box to deter badgers.
Thirty-six animals were released from belowground
units, including all 8 radio-tagged PEN animals. All
non-Sybille animals (n ˆ 24) were released from aboveground cages.
Radio-tagged ferrets were released at a cluster of 3
locations centered about 5 km northeast of the 4 locations where non-instrumented ferrets were released.
Con¯icting operational objectives prevented the preferred experimental design of intermixing radio-tagged
and non-tagged ferrets (for details, see Miller et al.,
1996; Biggins et al., 1997). Most release locations within
the clusters were separated by about 2 km. Estimated
densities of prairie dogs were 7.3 haÿ1 on the ``good''
habitat (that portion of the colony with density of active
burrows 40 haÿ1) where the radio-tagged ferrets were
released, but only 31% of that colony was rated as
``good'' (Hnilicka and Luce, 1993). The average prairie
dog density on the 3 colonies where non-radioed ferrets
were released (weighted by the number of ferrets
released on each) was 10.2 haÿ1, and 77% of the area
was rated as ``good''.
2.4. Radio-tracking
Radio-tracking, conducted 3 October±26 November,
mostly involved triangulation from ®xed stations, producing positional ®xes by intersecting bearings from 2
stations. The stations were camp trailers ®tted with 11element dual-beam Yagi antennas, situated on hilltops.
Accuracy of each station was tested by taking a sample
of at least 30 paired bearings, one bearing with the telemetry system and another with a transit, to a beacon
transmitter moved in an arc around the station. Di€erences between actual and radio bearings were plotted
against azimuths in an attempt to identify patterns of
error. If bias patterns were obvious, correction formulae
were developed to adjust subsequent bearings. A second
test sample was used to evaluate the e€ect of bias correction on the standard deviation of di€erences between
true and telemetric bearings. If variation was reduced,
the bias correction algorithm was added to the computer program used to process bearings. The standard
deviation of di€erences between pairs of bias-adjusted
telemetric and transit bearings was used to produce a
con®dence interval to predict the distribution of future
bearings. Accuracy of each ®x was indicated by area of
the error quadrangle, formed by intersecting 2 stationspeci®c con®dence interval arcs, and lengths of its diagonals (White and Garrot, 1990).
Multiple beacons (2±4) were used to reference the
stations at least once every 8 h. For stations with identi®able bias patterns, bearings were adjusted for bias
124
D.E. Biggins et al. / Biological Conservation 89 (1999) 121±129
before referencing calculations were made. A laptop
computer with TRITEL (a BASIC program by DEB)
was used in one of the tracking stations to perform
referencing calculations and to process bearings as they
were taken. The program assisted technicians in
selecting appropriate tracking stations by presenting
estimates of accuracy and a simple map of each ®x.
Monitoring was conducted in ®xed stations from sunset
to sunrise.
Mobile stations, trucks with collapsible or permanently erect null-peak yagi antennas, were used to track
rapidly-moving or distant animals. Hand-held equipment often was used to ®nd burrows occupied by
ferrets, shed collars, and predator-killed or injured ferrets. A ®xed-wing aircraft was used on 2 occasions to
locate dispersing animals.
Extremes of dispersal (the straight-line distance from
release site to most distant ®x during the ®rst 3 days
postrelease) and movements (the greatest cumulative
movement over any 3 day period) were estimated for
each animal. Long cumulative moves and dispersals had
mostly negative implications (wasted energy, increased
risk of predation, failure to explore a novel area methodically). Although extreme behaviors may be rare, we
assumed they would disproportionately in¯uence retention of ferrets on good quality habitat and long-term
survival. We agree, therefore, with Gaines and Denny
(1993) that ``...many problems of biological interest
concern the extremes...'' but we chose to use conventional statistics of central tendency and variation to
assess these measures.
To ®nd the longest cumulative movement in a 3 day
period, we summed distances between consecutive ®xes
for 72 h, examining all possible start times. Moves were
calculated without regard to inter®x time interval
(x ˆ 19 min). An animal was assumed to remain at a
point until it was found at a new point. We did not
assess movements for animals monitored <3 days.
The primary objective was to compare groups of animals with di€erent prerelease experience. The SOP
group of radio-tagged ferrets contained 5 hand-fed or
cross-fostered individuals, and represented 3 rearing
facilities. We emphasized the comparison between PEN
and PD groups because of greater uniformity of treatment within these groups, and because both groups
were from Sybille (Table 1).
Preliminary comparisons between sexes and types of
release cages were done to evaluate the potential for
confounding the interpretation regarding rearing
groups. To avoid grossly unbalanced designs, subsets of
the overall dataset were used in these initial comparisons. Comparisons of males and females were made
within the pooled data for PD and SOP groups to avoid
confounding by the PEN group, in which 7 of 8 radiotagged individuals were males. The 2 types of release
cages were compared within the radio-tagged PD group
(10 belowground, 7 aboveground) because all PEN animals were released with the belowground strategy and
all radio-tagged SOP animals were released with the
aboveground strategy.
Ferrets who left the prairie dog colonies where they
were released were classi®ed by departure direction,
either northeasterly (310 ±360 and 0 ±90 ) or southwesterly (91 ±309 ). Habitat conditions were disparate
in these 2 categories. Ferrets departing within the NE
arc could encounter 175 ha of prairie dog colonies
within 20 km, while ferrets departing in the SW arc had
16 752 ha of prairie dog colonies available after departing the colony of release (Hnilicka and Luce, 1993).
Habitat quality of prairie dog colonies at these release
sites (Hnilicka and Luce, 1993) was assessed using
transects to estimate density of active burrows (Biggins
et al., 1993).
Preliminary examination of the behavioral datasets
revealed heteroscedasticity and skewed distributions.
Signi®cance of di€erences between groups was assessed
with Kruskal±Wallis and Mann±Whitney tests; values
presented are the chi-square approximations (H) and
associated probabilities (p).
2.5. Spotlighting and capture
In the 2-week period 4±18 November, an e€ort was
made to ®nd, capture and identify as many animals as
possible. Spotlighting was conducted by a crew of 32
observers, who were coordinated to provide as complete
as possible coverage on the prairie dog colonies where
reintroduction occurred, and on selected colonies away
from the reintroduction site. Trapping and handling
(Thorne et al., 1985) was conducted by a few of the
most experienced persons. Tattoo identi®cation was
facilitated by shining a small ¯ashlight though the tattooed ear while the animal was in the trap. Ferrets were
transferred to a cloth sack for transponder identi®cation
with an electronic reader. Animals were released from
the bag at the locations of capture.
Data from spotlighting/recapture consisted of simple
counts (within treatment groups) of animals captured,
with statistical tests on representation of groups. If an
animal was captured more than once during the sampling period, the last date of capture was used to calculate time spent free-ranging. A previous assessment
(Biggins et al., 1998) incorporated these recapture data
into a multi-site summary, but pooled the SOP and PD
categories as variations of cage rearing.
Band recovery rate is the probability of capture times
the number of animals remaining after losses due to
mortality and emigration. Because these components of
recovery rate were not separately estimable in this
study, we could not examine the relationship between
mortality rate and rearing condition using non-telemetric data. Recovery rates, however, gave estimates of
D.E. Biggins et al. / Biological Conservation 89 (1999) 121±129
125
minimum survival. For evaluating signi®cance of
recapture data, the ``exact'' Chi-square procedure
(Berry and Mielke, 1985) was used. Mean numbers of
days that each group had been free-ranging when captured was compared with one-way analysis of variance.
For evaluation of null hypotheses, we used ˆ 0:05.
3. Results
3.1. Radio-telemetry
After bias correction, radio-tracking stations produced bearing readings varying ‹0.29 to ‹0.91 with
90% con®dence, assuming the best case scenario of nonmodulating signal, line-of-sight, and matched vertical
polarity of transmitting and receiving antennas (test
conditions). Reception range for ferret transmitters
varied with elevation of the station and position of the
animal; ranges to 10 km were common from 2 stations
located about 500 m above the surrounding terrain, and
1 animal's signal was received from 27 km. Radiotracking of 37 tagged ferrets produced 1224 locational
®xes. Most data were obtained by triangulation, but 7
animals were found during aerial searches. Ferrets were
telemetrically monitored an average of 6.3 days each.
The mean length of the longest diagonal of estimated
error quadrangles (Saltz and Alkon, 1985) for all ®xes
was 316 m. Lengths of longest diagonals did not di€er
signi®cantly between rearing groups (H ˆ 1:777,
d:f: ˆ 1, p ˆ 0:411), evidence that variation due to
tracking error was distributed equally among groups.
Eleven radio-tagged ferrets were known to have died
during the monitoring period, including 7 killed by
coyotes, 1 killed by a badger, 1 from a diurnal bird of
prey, 1 hit by an automobile, and 1 fatally injured by an
unknown assailant. The latter animal had head injuries
and punctures into the thoracic and abdominal cavities,
similar to injuries observed in cases of predation.
E€ects of type of release cage and sex were not signi®cant for dispersal (cage type, H ˆ 0:000, d:f: ˆ 1,
p ˆ 0:000; sex, H ˆ 0:277, d:f: ˆ 1, p ˆ 0:599) or
cumulative movements (cage type, H ˆ 0:193, d:f: ˆ 1,
p ˆ 0:661; sex, H ˆ 1:636, d:f: ˆ 1, p ˆ 0:201). Thus,
data were merged across levels of these categories for all
further analyses.
Although we did not use parametric statistics for tests
regarding movements or dispersal, standard errors were
larger for the SOP group than for other groups for
those attributes (by factors of 43±175%). The relatively
heterogeneous rearing conditions present in the SOP
group may have contributed to its large variation, reemphasizing the preferred use of the PD group for comparing cage-reared ferrets with pen-experienced ferrets.
Dispersing animals were found in a rectangular area
of about 1200 km2 (Fig. 2). Known overall maximum
Fig. 2. Release area and last known locations of radio-tagged blackfooted ferrets released at Shirley Basin, Wyoming in 1992.
dispersal was >7 km for 19 animals. Eight animals, all
without pen experience, dispersed >15 km, and 4 of
those dispersed >20 km over the entire study period.
Some ferrets began dispersing during the night of
release (up to 10 km), and 6 had dispersed >10 km by
the third night postrelease (up to 24.7 km). Maximum
dispersal during the ®rst 3 days was greater for PD animals than for PEN animals (H ˆ 4:900, d:f: ˆ 1,
p ˆ 0:027), but the di€erence between dispersal of SOP
and PEN animals was not signi®cant (H ˆ 2:149,
d:f: ˆ 1, p ˆ 0:143) (Fig. 3). Dispersal distances were
associated with direction of dispersal for ferrets departing their colonies of release (H ˆ 4:24, d:f: ˆ 1,
p ˆ 0:040); the 10 animals in the NE group dispersed
further than the 17 SW animals (Fig. 4). The NE group
and the SW group, however, did not di€er in maximum
cumulative 3 day movements (H ˆ 0:06, d:f: ˆ 1, p ˆ 0:800).
The longest cumulative movement in a 3 day period
was 49.6 km, made by a radio-tagged male of the PD
group who ®rst moved 10 km to the east, then reversed
course, travelling during the next two nights to a point
26.5 km northwest of the release site. Another male of
the PD group initially moved about 17 km west of the
reintroduction site, and was found in pine forest in the
Shirley Mountains. He was relocated back to the release
site, and was lost to trackers when he had reached a
point 23.5 km east of the release site. Six ferrets moved
>30 km in 3 days, and 19 ferrets moved more than 7
km in a single night.
Seventy-seven percent of the cage-reared (PD and SOP)
animals moved >7 km within a 12 h period and 50%
sustained that rate for 72 h (>21 km). PEN animals
exhibited less extreme 3 day maximum cumulative moves
(H ˆ 6:80, d:f: ˆ 1, p ˆ 0:009) than did animals of the
126
D.E. Biggins et al. / Biological Conservation 89 (1999) 121±129
Fig. 3. Maximum postrelease dispersal (distance separating release
site and most distance telemetric ®x during the ®rst 3 days postrelease)
of radio-tagged black-footed ferrets with 3 rearing histories (medians,
ranges, 25th±75th percentiles). The SOP ferrets were raised in standard
indoor cages, and had no experience with live prairie dogs. The PD
ferrets were cage-reared, and had experience with predation on prairie
dogs. The PEN ferrets were born in cages, were allowed access to
outdoor pens with prairie dog burrows when kits were about 60 days
old, and were fed live prairie dogs. The PEN and PD groups di€ered
signi®cantly (H ˆ 4:900, d:f: ˆ 1, p ˆ 0:027), but the PEN and SOP
groups did not (H ˆ 2:149, d:f: ˆ 1, p ˆ 0:143).
Fig. 5. Maximum cumulative 72 h movement (over any 3 day period)
made by radio-tagged black-footed ferrets with 3 rearing histories
(medians, ranges, 25th±75th percentiles). The SOP ferrets were raised
in standard indoor cages, and had no experience with live prairie dogs.
The PD ferrets were cage-reared, and had experience with predation
on prairie dogs. The PEN ferrets were born in cages, were allowed
access to outdoor pens with prairie dog burrows when kits were about
60 days old, and were fed live prairie dogs. The PEN and PD groups
di€ered signi®cantly (H ˆ 6:80, d:f: ˆ 1, p ˆ 0:009), but the PEN and
SOP groups did not (H ˆ 0:893, d:f: ˆ 1, p ˆ 0:345).
been confounded. Using a more conservative comparison of Sybille PD and PEN animals only, the return
rate for PEN animals was higher than the return rate for
PD animals (X2 ˆ 8:32, d:f: ˆ 1, p ˆ 0:007). Nonradioed ferrets, released on better habitat, also can be
treated as a separate subset of ferrets, and again the
PEN animals had the highest return rate (X2 ˆ 8:17,
d:f: ˆ 2, p ˆ 0:015).
Fig. 4. Maximum postrelease dispersal (distance between release site
and most distant telemetric ®x) of radio-tagged ferrets moving
(southwesterly) further into the prairie dog complex and (northeasterly) away from the prairie dog complex (medians, ranges, 25th±
25th percentiles). The e€ect of dispersal direction was signi®cant
(H ˆ 4:24, d:f: ˆ 1, p ˆ 0:040).
PD group, but cumulative movements of the PEN and
SOP groups did not di€er signi®cantly (H ˆ 0:893,
d:f: ˆ 1, p ˆ 0:345) (Fig. 5).
3.2. Spotlighting and capture
The November spotlighting/capture e€ort detected 14
animals that were released in 1992, 4 of which had been
radio-tagged (all had lost their collars). Neither transponders nor ear tattoos were totally reliable for identi®cation of animals; ear tattoos were not legible for some
animals, and transponder loss occurred in captivity and
postrelease. At least one of the two identi®ers, however,
was usable for every animal recaptured.
Seven of the 14 animals captured were from the PEN
group, the highest band return rate (39% of animals in
that group) of the three groups (X2 ˆ 9:85, d:f: ˆ 2,
p ˆ 0:006). Mean elapsed times from release to recapture were similar (about 1 month) for the three groups
(F ˆ 1:55; d:f: ˆ 2; 11; p ˆ 0:25). Because all PEN animals came from Sybille and SOP animals came from 4
facilities, e€ects of rearing and origin facility could have
4. Discussion
Movements of males exceeded movements of females
for free-ranging ferrets near Meeteetse, Wyoming (Biggins et al., 1985), contrasting with the similar movements and dispersal of released males and females in
this study. Perhaps individuals of both sexes were overwhelmed by release into a strange area, and other sources of variation masked intersexual di€erences (Biggins
et al., 1998). Over time, patterns of movement unique to
each sex would be expected to develop; the duration of
this study was too short to gather such information.
The longest total movement in one night by a radiotagged native black-footed ferret at Meeteetse, Wyoming
slightly exceeded 7 km (Biggins and Fagerstone, 1984).
By comparison, 19 released ferrets moved (cumulatively) >7 km in one night and 8 ferrets dispersed >15
km, perhaps re¯ecting release of naive animals into a
novel environment. We may have underestimated dispersal for some animals, because dispersing animals
sometimes were lost to radiotrackers, and some probably continued to disperse after losing their collars.
The geographical distribution of ferrets captured in
November is consistent with habitat selection by ferrets.
Habitat quality in the immediate vicinity of the release
site may have a€ected the likelihood that an animal
would remain there. During the recapture period, 1 of
D.E. Biggins et al. / Biological Conservation 89 (1999) 121±129
the 14 ferrets captured was found on the colony where
radio-tagged ferrets were released. The individual found
on that presumed poorer habitat was a radio-tagged
ferret that had been recaptured and translocated back
to its initial release site shortly before the recapture
e€ort began; it had initially dispersed about 17 km to
the west and, without our intervention, probably would
not have been present at the reintroduction site during
the survey. The other 3 (formerly) radio-tagged individuals located in the November survey had moved to
colonies with higher estimates of burrow density and
prairie dog density. Nevertheless, ®nding fewer ferrets
on the telemetry area could also be explained by the
poorer conditions for spotlighting there; access roads
were fewer and of poorer quality, and topographic relief
was more pronounced, circumstances expected to
decrease detectability of ferrets (Marinari, 1992). Our
data suggesting in¯uence of density of active prairie
dog burrows and prairie dogs should be regarded as
tentative.
Recapture data must be interpreted cautiously
because surveys were conducted only in the core portion
of the reintroduction area. Interpretation of our band
recovery rate was complicated by interanimal variation
in times of release and recapture. Concern about the
e€ect of time is reduced, however, by having releases of
all groups distributed over much of the release period,
and by the similarity between groups in mean times that
recaptured animals had been free-ranging.
The radio-tagged subsample helped explain the 4-fold
di€erence in recovery rates of PEN and PD animals.
PEN ferrets, by making shorter aboveground movements and dispersals, may decrease their probability of
encountering aboveground predators (Miller et al.,
1992). Dispersing ferrets frequently found themselves
away from prairie dog colonies, where food and cover
were scarce, and the ferrets expended energy traveling;
thus, dispersing ferrets may su€er higher mortality from
predation and starvation.
What factor(s) in the outdoor pen environment could
lead to improved adaptation to the wild? There are
many possibilities (Biggins et al., 1998), but few experiments have been conducted to isolate their separate
in¯uences on ferrets. Ferrets raised outdoors were subjected to oscillations in climatic conditions, and were
pre-exposed to soil microorganisms and parasites such
as ¯eas and ticks. Perhaps PEN animals were more
cautious because of less frequent stimulation by humans
or were imprinted on burrows as refugia from having
been raised in them. Pen-raised black-footed ferrets may
possess improved motor coordination or may be in better physical condition from having more exercise during
their rearing. If they were in better physical condition,
they should have been capable of longer moves than
cage-reared ferrets, making their signi®cantly shorter
moves more striking.
127
Ferrets gained hunting experience in pens, and during
pre-release trials, cage-raised black-footed ferrets
exposed to live prairie dogs (PD group) were signi®cant
more pro®cient at killing this prey than SOP animals
exposed to live hamsters only (Vargas, 1994). These
superior predatory abilities, however, did not translate
into a post-release survival advantage for the PD group.
Similarly, cage-raised golden lion tamarins trained in
food processing and locomotory skills did not di€er signi®cantly from untrained counterparts (Beck et al., 1988).
5. Recommendations
As many ferret kits as possible should be exposed to
quasi-natural environments (in outdoor pens) before
release. Monetary costs of this strategy should be evaluated relative to the number of surviving, free-ranging
ferrets in short and long time scales. Nevertheless,
monetary outlays and counts of surviving ferrets are
proximate costs and bene®ts. Ultimate costs of not
maximizing survivorship could include irretrievable loss
of genetic diversity due to slower expansion of the ferret
population, and erosion of survivorship skills due to
increasing the generational time spent in captivity by a
large proportion of the ferret population.
Finally, it seems advisable to avoid releasing ferrets in
areas of low prairie dog density, at the edges of colonies,
or on small colonies at the edges of larger aggregations
of colonies, the cautionary note on reliability of habitat
data notwithstanding.
Acknowledgements
This project involved a great deal of cooperation from
diverse groups, and multiple pages would be needed to
identify all who contributed. Many individuals of the
Wyoming Game and Fish Department contributed to
the success of the captive breeding program that led to
these reintroductions, and for managing the complex
logistics of the reintroductions themselves. Zoos that
have volunteered to participate in the captive breeding
program are to be commended; these presently include
Omaha, Louisville, the National Zoo's Conservation
and Research Center, Colorado Springs, Phoenix, and
Toronto. We greatly appreciate the assistance of the
spotlight-capture crew who came from several states
and agencies including the Wyoming Game and Fish
Department, the Park Service and Forest Service (South
Dakota), the Bureau of Land Management (Wyoming,
Colorado and Montana), the Fish and Wildlife Service
(Wyoming, Colorado, Montana, and South Dakota),
the University of Wyoming and Yale University. We are
indebted to the landowners of Shirley Basin who
128
D.E. Biggins et al. / Biological Conservation 89 (1999) 121±129
allowed us access to their ranches. Sincere thanks are
due to the Legacy Program of the Department of
Defense, Wildlife Preservation Trust International,
Chevron USA, and the National Fish and Wildlife
Foundation for their contributions that augmented
funding from the US Fish and Wildlife Service. We
appreciate the critical reviews of John Oldemeyer,
Richard Reading, and Roger Powell.
References
Beck, B.B., 1991. Managing zoo environments for reintroduction. In:
American Association of Zoological Parks and Aquariums Annual
Conference Proceedings. September 15±19, San Diego, California,
pp. 436±440.
Beck, B., Castro, I., Kleiman, D., Dietz, J., Rettberg-Beck, B., 1988.
Preparing captive-born primates for reintroduction. Int. J. Primatology 8, 426.
Beck, B.B., Rappaport, L.G., Stanley Price, M.R., Wilson, A.C., 1994.
Reintroduction of captive-born animals. In: Olney, P., Mace, G.,
Feistner, A. (Eds.), Creative Conservation: Interactive Management
of Wild and Captive Animals. Chapman and Hall, London, pp.
265±286.
Berry, K.J., Mielke Jr, P.W., 1985. Subroutines for computing exact
Chi-square and Fisher's exact probability tests. Ed. and Psych.
Meas. 45, 153±159.
Biggins, D.E., Fagerstone, K.A., 1984. Movements and activity of
black-footed ferrets (Mustela nigripes). American Society of Mammalogists, 64th Annual Meeting, Arcata, California, Abstract.
Biggins, D.E., Godbey, J.L., Hanebury, L.R., Luce, B., Marinari,
P.E., Matchett, M.R., Vargas, A., 1998. The e€ect of rearing methods on survival of reintroduced black-footed ferrets. J. Wildl. Manage. 62, 643±653.
Biggins, D.E., Hanebury, L.R., Miller, B.J., 1991. Trial release of
Siberian polecats (Mustela eversmanni)Ðprogress report, 1 May
1991. Unpublished, Midcontinent Ecol. Sci. Center, Fort Collins,
Colorado.
Biggins, D.E., Hanebury, L.R., Miller, B.J., Powell, R.A., 1990. Trial
release of Siberian polecats (Mustela eversmanni)Ðprogress report,
28 February 1990. Unpublished, Midcontinent Ecol. Sci. Center,
Fort Collins, Colorado.
Biggins, D.E., Miller, B.J., Hanebury, L.R., Oakleaf, R., Farmer,
A.H., Crete, R., Dood, A., 1993. A technique for evaluating blackfooted ferret habitat. In: Oldemeyer, J.L., Biggins, D.E., Miller, B.J.
(Eds.), Proceedings of the Symposium on the Management of
Prairie Dog Complexes for Reintroduction of the Black-footed
Ferret. U.S. Fish and Wildl. Ser. Biol. Rep. 93-1, Washington, DC,
pp. 73±88.
Biggins, D.E., Miller, B.J., Clark, T.W., Reading, R.P., 1997. Management of an endangered species: the black-footed ferret. In:
Me€e, G.K., Carroll, C.R. (Eds.), Principles of Conservation Biology, 2nd ed. Sinauer Associates, Sunderland, Massachusetts, pp.
420±426.
Biggins, D.E., Schroeder, M.H., 1988. Historical and present status of
the black-footed ferret. In: Uresk, D.W., Schenbeck, G.L., Cefkin,
R. (Eds.), Eighth Great Plains Wildlife Damage Control Workshop
Proceedings. USDA For. Ser. Gen. Tech. Rep. RM-154. Fort Collins, Colorado, pp. 93±97.
Biggins, D.E., Schroeder, M.H., Forrest, S.C., Richardson, L., 1985.
Movements and habitat relationships of radio-tagged black-footed
ferrets. In: Anderson, S.H., Inkley, D.B. (Eds.), Black-footed Ferret
Workshop Proceedings. Wyoming Game and Fish Dep, Cheyenne,
Wyoming, pp. 11.1±11.17.
Box, H.O., 1991. Training for life after release: simian primates as
examples. In: Gipps, J.H.W. (Ed.), Beyond Captive Breeding: Reintroducing Endangered Mammals to the Wild. Clarendon Press,
Oxford, U.K, pp. 111±124.
Clark, M.M., Galef, B.G., 1980. E€ect of rearing environment on
adrenal weights, sexual development, and behavior of gerbils: an
examination of Richter's domestication hypothesis. J. Comp. Physiol. Psych. 94, 857±863.
Ellis, D.H., Dobrott, S.J., Goodwin Jr, J.G., 1978. Reintroduction
techniques for masked bob-whites. In: Temple, S.A. (Ed.), Endangered Birds. U. Wisconsin Press, Madison, pp. 345±354.
Fagerstone, K.A., Biggins, D.E., Campbell III, T.M., 1985. Marking
and radio-tagging of black-footed ferrets (Mustela nigripes). In:
Anderson, S.H., Inkley, D.B. (Eds.), Black-footed Ferret Workshop
Proceedings. Wyoming Game and Fish Dep, Cheyenne, pp. 10.1±
11.10.
Fagerstone, K.A., Johns, B.E., 1987. Transponders as permanent
identi®cation markers for domestic ferrets, black-footed ferrets, and
other wildlife. J. Wildl. Manage. 51, 294±297.
Forrest, S.C., Biggins, D.E., Richardson, L., Clark, T.W., Campbell,
T.M.III, Fagerstone, K.A., Thorne, E.T., 1988. Population attributes for the black-footed ferret (Mustela nigripes) at Meeteetse,
Wyoming, 1981±1985. J. Mammal. 69, 261±273.
Gaines, S.D., Denny, M.W., 1993. The largest, smallest, highest, lowest, longest, and shortest: extremes in ecology. Ecology 74, 1677±
1692.
Greenough, W.T., Juraska, J.M., 1979. Experience induced changes in
brain ®ne structure: Their behavioral implications. In: Hahn, M.E.,
Jensen, C., Dudek, B.C. (Eds.), Development and Evolution of
Brain Size: Behavioral Implications. Academic Press, New York,
pp. 263±294.
Grith, B., Scott, J.M., Carpenter, J.W., Reed, C., 1989. Translocation as a species conservation tool: Status and strategy. Science 345,
447±480.
Hnilicka, P., Luce, B., 1993. Habitat evaluation of Primary Management Zone 1 in the Shirley Basin/Medicine Bow black-footed ferret
management area. In: Oakleaf, B., Luce, B., Thorne, E.T., Williams,
B. (Eds.), Black-footed Ferret Reintroduction in Shirley Basin,
WyomingÐ1992 Annual Completion Report. Unpublished,
Wyoming Game and Fish Dept., Cheyenne, pp. 31±38.
Kleiman, D.G., 1989. Reintroduction of captive mammals for conservation. Bioscience 39, 152±161.
Kleiman, D.G., Beck, B.B., Dietz, J.M., Dietz, L.A., Ballou, J.D.,
Coimbra-Filho, A.F., 1986. Conservation program for the golden
lion tamarin: Captive research and management, ecological studies,
educational strategies, and reintroduction. In: Benirschke, K. (Ed.),
Primates: The Road to Self-sustaining Populations. Springer-Verlag, New York, pp. 959±979.
Marinari, P.E., 1992. Detectability of black-footed ferrets (Mustela
nigripes) using spotlighting. MS Thesis, U. Wyoming, Laramie.
Miller, B., Biggins, D., Hanebury, L., Conway, C., Wemmer, C., 1992.
Rehabilitation of a species: The black-footed ferret (Mustela
nigripes). In: Ninth Annual Proceedings of the National Wildlife
Rehabilitation Association. Chicago, Illinois, pp. 183±192.
Miller, B., Biggins, D., Wemmer, C., Powell, R., Hanebury, L., Horn,
D., Vargas, A., 1990a. Development of survival skills in captiveraised Siberian polecats (Mustela eversmanni): I. Locating prey. J.
Ethol. 8, 89±94.
Miller, B., Biggins, D., Wemmer, C., Powell, R., Calvo, L., Hanebury,
L., Wharton, T., 1990b. Development of survival skills in captiveraised Siberian polecats (Mustela eversmanni): II. Predator avoidance. J. Ethol. 8, 95±104.
Miller, B., Reading, R., Forrest, S., 1996. Prairie night. Smithsonian
Inst. Press, Washington, D.C.
Phillips, M.K., 1990. Measures of the value and success of a reintroduction project: Red wolf reintroduction project in Alligator River
National Wildlife Refuge. Endangered Species Update 8, 24±26.
D.E. Biggins et al. / Biological Conservation 89 (1999) 121±129
Renner, M.J., 1988. Learning during exploration: the role of behavioral topography during exploration in determining subsequent
adaptive behavior. Int. J. of Comp. Psych. 2, 43±56.
Rosenzweig, M.R., 1979. Responsiveness of brain size to individual
experience: Behavioral and evolutionary implications. In: Hahn,
M.E., Jensen, C., Dudek, B.C. (Eds.), Development and Evolution
of Brain Size: Behavioral Implications. Academic Press, New York,
pp. 263±294.
Saltz, D., Alkon, P.U., 1985. A simple computer-aided method for
estimating radio-location error. J. Wildl. Manage. 49, 664±668.
Scott, J.P., Fuller, J.L., 1965. Genetics and Social Behavior of the
Dog. U. Chicago Press, Chicago, Illinois.
Soderquist, T.R., Serena, M., 1994. An experimental reintroduction
program for bush-tailed phascogales (Phascogale tapoatafa): the interface between captivity and the wild. In: Olney, P., Mace, G., Feistner,
A. (Eds.), Creative Conservation: Interactive Management of Wild
and Captive Animals. Chapman and Hall, London, pp. 431±438.
Stanley-Price, M.R., 1989. Animal reintroductions: The Arabian oryx
in Oman. Cambridge U. Press, Cambridge, U.K.
Thorne, E.T., Schroeder, M.H., Forrest, S.C., Campbell III, T.M.,
Richardson, L., Biggins, D.E., Hanebury, L.R., Belitsky, D.,
Williams, E.S., 1985. Capture, immobilization and care of black-
129
footed ferrets for research. In: Anderson, S.H., Inkley, D.B. (Eds.),
Black-footed Ferret Workshop Proceedings. Wyoming Game and
Fish Dep, Cheyenne, pp. 9.1±9.8.
Vargas, A., 1994. Ontogeny of the black-footed ferret (Mustela
nigripes) and e€ects of captive-upbringing on predatory behavior
and post-release survival. Ph.D. Diss., U. Wyoming, Laramie.
Vargas, A., Anderson, S.H., 1996. Growth and development of captive-raised black-footed ferrets (Mustela nigripes). Am. Midl. Nat.
135, 43±52.
Wemmer, C., Derrickson, S., 1987. Reintroduction: the zoologists
dream. In: Annual Proceedings, American Association of Zoological Parks and Aquariums, Wheeling, W. Virginia, pp. 48±65.
White, G.C., Garrot, R.A., 1990. Analysis of Wildlife Radio-tracking
Data. Academic Press, San Diego, California.
Williams, E.S., Thorne, E.T., Kwiatkowski, D.R., Anderson, S.L.,
Lutz, K., 1991. Reproductive biology and management of
captive black-footed ferrets (Mustela nigripes). Zoo Biol. 10, 383±
398.
Wilson, A.C., Stanley-Price, M.R., 1994. Reintroduction as a reason
for captive breeding. In: Olney, P., Mace, G., Feistner, A. (Eds.),
Creative Conservation: Interactive Management of Wild and Captive Animals. Chapman and Hall, London, pp. 243±264.
Download