Efficacy of a multiple-capture live trap for small mammals (1).
Belant, Jerrold L. ; Windels, Steve K.
Abstract. We compared the efficacy of Victor" Tin Cat'
and Sherman live traps for capturing small mammals in northern hardwood
and red pine (Pinus resinosa) stands in the north-central Upper
Peninsula of Michigan during 2001. Overall mean capture rates (total
captures/100 adjusted trap nights) by habitat were greater (P [less than
or equal to] 0.030) for Sherman traps than for Tin Cat traps. Capture
rates remained lower for Tin Cat traps in northern hardwood (P = 0.004)
but not red pine (P = 0.936) habitat after adjusting for species
(sciurids) unable to enter them. Greater species diversity values were
obtained using Sherman traps in both habitats. We conclude that in
sampling arrays tested, Victor Tin Cat traps were less effective than
Sherman traps for estimating small mammal abundance and diversity.
INTRODUCTION
Live traps are commonly used to estimate small mammal abundance and
diversity (for example, Von Trebra and others 1998; Carey and Wilson
2001). Differential trapping success among types and sizes of small
mammal live traps has been demonstrated (Slade and others 1993;
O'Farrell and others 1994). Consequently, interpretation of results
for a particular trap type can only be made with some knowledge of the
trap's performance relative to other trap types.
Multiple-capture live traps have been used in several recent
studies of small mammal ecology and distribution, with the commercial
Victor Tin Cat trap (Woodstream Corp., Lititz, PA) used most commonly
(for example, Bowman and others 2001 a,b,c). One proposed benefit of
using these traps is a reduction in the number of traps necessary to
sample an area due to their multiple-capture capability. Tin Cat traps
reset after each capture and can hold several small mammals
simultaneously. The website for the manufacturer of the Tin Cat trap
(www.woodstream.com) states these traps can hold up to 30 mice. Studies
by Bowman and others (2001 a,b,c) used arrays of five Tin Cats in a
cross pattern spaced 35 m apart, which presumably sampled the same area
as 25 Sherman traps set in a 5 x 5 grid with 10 m spacing, or a circle
of about 50 m radius (J. Bowman, Ontario Ministry of Natural Resources,
personal communication).
Another potential benefit of Tin Cat traps is the increased
likelihood of multiple captures due to the presence of a small mammal in
the trap. An animal in the trap may help other individuals overcome a
neophobic response. Residual odors of conspecifics in traps has enhanced
capture efficacy for several small mammals species (Rowe 1970; Drickamer
1984; Tobin and others 1994).
Despite its increasing use in ecological studies, the efficacy of
the Victor Tin Cat has not been thoroughly tested against the more
commonly used Sherman live trap (H. B. Sherman Traps Inc., Tallahassee,
FL). Our objective was to compare small mammal capture rates and species
composition between Victor Tin Cat and Sherman live traps in previously
used trap arrays to determine the effectiveness of Tin Cat traps in
estimating abundance and diversity of small mammals.
MATERIALS AND METHODS
The study was conducted in the north-central Upper Peninsula of
Michigan, on lands administered by Pictured Rocks National Lakeshore and
Hiawatha National Forest. Trapping was conducted in two habitats:
northern hardwoods with a dominant overstory of sugar maple (Acer
saccbarum) and American beech (Fagus grandifolia), and a red pine (Pinus
resinosa) plantation.
We used collapsible Sherman live traps (Model LFG) designed for
single captures (Fig. 1). Sherman traps were 8.0 x 8.8 x 23.4 cm with
one tapered 6.5 to 7.0 x 8.0 cm entrance. Victor Tin Cat live traps
(Model 308) were 26.7 x 15.9 x 4.8 cm with a clear plastic lid and two
2.8 x 3.5-cm entrances. Tin Cat traps were designed for multiple
captures.
[FIGURE 1 OMITTED]
The experimental design in each habitat was a randomized
split-block with 6 replicates. Each block consisted of 25 Sherman traps
and 5 Tin Cat traps. We placed Sherman traps in a 5 x 5 grid with 17.5-m
spacing for an cffective trapping area of about 7,650 [m.sup.2]. We
placed a Tin Cat trap in the center of each split and remaining traps 35
m from center in the cardinal directions; this array was designed to
sample a 50-m radius around the center trap (effective trapping area of
7,850 [m.sup.2]; Bowman and others 2001 a,b,c). Blocks and splits within
blocks were separated by > 25 m; blocks were >25 m from habitat
edge. All traps were placed in "most likely runway" positions,
baited with rolled oats and peanut butter, and rebaited as necessary.
Prebaiting was not conducted. Cotton was placed in traps to provide
bedding. Traps were set initially during morning and checked 3 times
each day for 4 days, resulting in 100 and 20 unadjusted trap
nights/block for Sherman and Tin Cat traps, respectively. Individuals
captured were identified to species and released at their respective
capture sites. We followed animal care and use guidelines outlined by
the American Society of Mammalogists (1987).
To standardize trapping effort, Sherman traps that were sprung were
adjusted using 0.5 time intervals (that is, trap nights; Belant 1992;
Beauvais and Buskirk 1999). Tin Cat traps allowed multiple captures and
were not adjusted for unless disturbed or missing. We calculated mean
number of captures/ 100 adjusted trap nights for each trap type in both
habitats. We also calculated mean capture rates excluding sciurids (red
squirrel [Tamiasciurus hudsonicus], eastern chipmunk [T. striatus], and
least chipmunk [Eutamius sciurus]), as these species were too large to
enter Tin Cat traps.
Because of unequal variances, we used Wilcoxon rank sum tests (Zar
1984) to compare mean rank scores of capture rates between trap types in
each habitat using PROC NPAR1WAY (SAS Institute Inc. 1990). Means are
reported with [+ or -] 1.0 standard error; statistical significance was
established at P <0.050.
We used Species Richness, Shannon-Weiner, and Simpson's
indices to compare species diversity of small mammals captured with each
trap type in each habitat (Colinvaux 1986; Kirkland 1990). Species
Richness (S) is a measure of the number of species documented by
capture. The Shannon-Weiner index (H') measures the probability of
selecting the identity of an individual taken from the sample at random
using the equation:
H' = -[summation] [p.sub.i], ln [p.sub.i]
Where [p.sub.i] is the proportion of the total number of
individuals in the ith species for i = 1 to n.; H' increases with
species diversity. Simpson's index (l) is the probability that any
two individuals selected at random will be the same species and uses the
equation:
[lambda] = [summation] [p.sup.2.sub.i]
for i = 1 to n. Simpson's index is an inverse measure of
diversity in that species diversity increases as [lambda] decreases.
RESULTS
We captured 253 individuals in 1,440 unadjusted trap nights, 141 in
northern hardwood and 112 in red pine. In the northern hardwood habitat,
139 were captured in Sherman traps and 2 were captured in Tin Cat traps.
We captured 112 individuals in Sherman traps and 9 in Tin Cat traps in
the red pine stand. Deer mice (Peromyscus maniculatus) and eastern
chipmunks were the species most commonly captured in both habitats (Fig.
2).
[FIGURE 2 OMITTED]
Overall mean capture rates were greater for Sherman traps than for
Tin Cat traps in northern hardwood (Z = 2.87, P = 0.004) and red pine
habitats (Z = 2.16, P = 0.030; Table 1). After adjusting for sciurid
captures, capture rates remained higher for Sherman traps in northern
hardwoods (Z = 2.85 P = 0.004) but not in the red pine habitat (Z =
0.80, P = 0.936).
We captured 7 species overall, 3 in northern hardwood and 7 in red
pine. In the northern hardwood habitat, 3 species were captured in
Sherman traps and 1 species was captured using Tin Cat traps. We
captured 7 species in Sherman traps and 2 species in Tin Cat traps in
the red pine stand. Greater species diversity values were obtained using
all 3 indices from Sherman traps in both habitats (Table 2).
One multiple capture (2 deer mice) was recorded for Tin Cat traps;
only individual captures were recorded for Sherman traps.
DISCUSSION
Overall capture rates and species diversity were lower with Tin Cat
than with Sherman traps. Only 4% of total individuals and one-half of
non-sciurid species were captured using Tin Cat traps. Although capture
rates were similar between trap types in red pine habitat, few
individuals were captured in Tin Cat traps. Thus, small differences in
the number of individuals or species captured using Tin Cat traps could
have a large effect on estimates of abundance and species diversity.
Further, the small opening size of Tin Cat traps does not allow for
capture of sciurids which are frequently important for assessing small
mammal community abundance and diversity.
Olfaction is important in the social biology of small mammals
(Stoddart 1974) and residual odors from small mammals in traps have been
reported to enhance trapping efficacy (Drickamer 1984; Tobin and others
1994). The potential benefit of the presence of small mammals in Tin Cat
traps as attractants to enhance capture rates through multiple captures
was not observed in this study. Residual odor of small mammals captured
was present in both trap types. Therefore, visual and olfactory cues
were not advantages for using Tin Cat traps in our study.
Small mammal capture rates and trap efficacy vary markedly across
studies. Location, season, and trap type, among other factors, influence
capture rates (Wiener and Smith 1972; Williams and Braun 1983; Mengak
and Guynn 1987). Indeed, multiple studies comparing identical trap types
have reached different conclusions on capture effectiveness (Sealander
and James 1958; Wiener and Smith 1972). Tin Cat and Sherman live traps
have reportedly had similar capture rates in a previous study (J.
Bowman, Ontario Ministry of Natural Resources, personal communication).
In trap arrays and habitats we compared, however, we do not recommend
Tin Cat over Sherman live traps. While three days of prebaiting prior to
trapping was used previously (Bowman and others 2001 a,b,c), we do not
believe that prebaiting in our study would have changed the relative
efficacy of the trap types. Although equipment costs and trapping effort
were greater with Sherman traps, greater numbers of individuals and
species captured using Sherman traps will likely result in improved
estimates of small mammal abundance and diversity.
Acknowledgments. We thank J. Bowman for critical review of this
manuscript. A. Kavalunas, N. Z. Lapinski, R. Posey, and K. A. Stanley
assisted with fieldwork. K. Doran (Hiawatha National Forest) provided
access to study sites. Funding was provided by the National Park Service
Pictured Rocks Science Center, Michigan Technological University, and
the Michigan Department of Natural Resources Natural Heritage Fund.
(1) Manuscript received 15 May 2005 and in revised form 29 October
2006 (#05-22).
(2) Present address: National Park Service, Voyageurs National
Park, 3131 Highway 53 South, International Falls, MN 56649
LITERATURE CITED
American Society of Mammalogists. 1987. Guidelines for the Capture,
Handling, and Care of Mammals. Lawrence (KS): Amer Soc of Mammalogists.
47 p.
Beauvais GP, Buskirk SW. 1999. Modifying estimates of sampling
effort to account for sprung traps. Wildl Soc Bull 27:39-43.
Belant JL. 1992. Efficacy of three types of live traps for
capturing weasels, Mustela spp. Can Field-Natural 106:394-7.
Bowman J, Forbes G, Dilworth T. 2001a. Landscape context and
small-mammal abundance in a managed forest. For Ecol Manage 140:249-55.
Bowman J, Forbes G, Dilworth T. 2001b. The spatial component of
variation in small-mammal abundance measured at three scales. Can J Zoo1
79:137-44.
Bowman J, Forbes G, Dilworth T. 2001c. Spatial and temporal
patterns of an irrupting population of deer mice. J Mammal 82:567-72.
Carey AB, Wilson SM. 2001. Induced spatial heterogeneity in forest
canopies: responses of small mammals. J Wild/Manage 65:1014-27.
Colinvaux PA. 1986. Ecology. New York: John Wiley. 725 p.
Drickamer LC. 1984. Capture of two species of Peromyscus at live
traps baited with male and female odors. J Mammal 65:699-702.
Kirkland GL Jr. 1990. Patterns of initial small mammal community
change after clearcutting of temperate North American forests. Oikos
59:313-20.
Mengak MT, Guynn DC Jr. 1987. Pitfalls and snap traps for sampling
small mammals and herpetofauna. Am Mid/Natural 118:284-8.
O'Farrell MJ, Clark WA, Emerson FH, Jaurez SM, Kay FR,
O'Farrell TM, Goodlett TY. 1994. Use of a mesh live trap for small
mammals: are results from Sherman live traps deceptive? J Mammal
75:692-9.
Rowe FP. 1970. The response of wild house mice (Mus musculus) to
live-traps marked by their own and by a foreign mouse odour. J Zool
(London) 162:517-20.
SAS Institute Inc. 1990. SAS/STAT User's Guide. Release 6.03.
Cary (NC): SAS Institute Inc. 1686 p.
Sealander JA, James D. 1958. Relative efficiency of different small
mammal traps. J Mammal 39:215-23.
Slade NA, Eifler MA, Gruenhagen NM, Davelos AL. 1993. Differential
effectiveness of standard and long Sherman livetraps in capturing small
mammals. J Mammal 74:156-61.
Stoddart DM. 1974. The role of odor in the social biology of small
mammals. In: Birch MC, editor. Pheromones. Amsterdam (Netherlands):
North Holland Publ. p 297-315.
Tobin ME, Sugihara RT, Engeman RM. 1994. Effects of initial rat
captures on subsequent capture success of traps. Proceedings of the
Vcrtebrate Pest Conf. 16:101-5.
Von Trebra C, Lavender DP, Sullivan TP. 1998. Relations of small
mammal population to even-aged shelterwood systems in sub-boreal spruce
forest. J Wildl Manage 62:630-42.
Wiener JG, Smith MH. 1972. Relative efficiencies of four small
mammal traps. J Mammal 53:868-73.
Williams DF, Braun SE. 1983. Comparison of pitfall and conventional
traps for sampling small mammal populations. J Wildl Mgmt 47:841-5.
Zar JH. 1984. Biostatistical analysis. 2nd edition. Englewood
Cliffs (NJ): Prentice Hall. 718 p.
Jerrold L. Belant AND Steve K. Windels (2), National Park Service,
Pictured Rocks Science Center, N8391 Sand Point Road, Munising, MI
49862; School of Forest Resources and Environmental Science, Michigan
Technological University, 1400 Townsend Drive, Houghton, M149931
TABLE 1
Small mammal capture rates (individuals/l00 adjusted trap nights)
using Sherman and Victor Tin Cat live traps, north-central
Upper Peninsula of Michigan, July 2001.
Sherman
Habitat With sciurids
x SE
Northern
hardwood 8.1 2.7
Red pine 6.0 0.9
Sherman
Habitat Without sciurids
x SE
Northern
hardwood 4.7 0.4
Red pine 2.8 0.8
Sherman
Habitat Tin Cats
x SE
Northern
hardwood 0.9 0.6
Red pine 2.5 0.9
TABLE 2
Small mammal species diversity using Species Richness (S),
Shannon-Weiner (H'), and Simpson's ([lambda]) indices calculated
from captures using Sherman and Victor Tin Cat live traps,
north-central Upper Peninsula of Michigan, July 2001.
Sherman
With sciurids
Northern S H' lambda
hardwood 3 0.79 0.40
Red pine 7 1.54 0.27
Sherman
Without sciurids
Northern S H' lambda
hardwood 2 0.54 0.65
Red pine 4 0.84 0.55
Tin Car
Northern S H' lambda
hardwood 1 0 1.00
Red pine 2 0.61 0.56