Canadian Journal of Zoology
Influence of field technique, density, and sex on home range and overlap of the southern red-backed vole (Myodes gapperi)
Journal: Canadian Journal of Zoology
Manuscript ID cjz-2018-0338.R1
Manuscript Type: Article
Date Submitted by the 14-Jun-2019 Author:
Complete List of Authors: Tisell, Honora; University System of New Hampshire, Natural Resources and the Environment Degrassi, Allyson; Shenandoah University, Environment and Society DepartmentDraft Stephens, Ryan; University System of New Hampshire, Natural Resources and the Environment Rowe, Rebecca; University System of New Hampshire, Natural Resources and the Environment
Is your manuscript invited for consideration in a Special Not applicable (regular submission) Issue?:
core area, mark-recapture, myodes gapperi, radiotelemetry, Keyword: RODENTS;SHREWS < Taxon, southern red-backed vole, space use
https://mc06.manuscriptcentral.com//cjz-pubs Page 1 of 35 Canadian Journal of Zoology
Influence of field technique, density, and sex on home range and overlap of the southern
red-backed vole (Myodes gapperi)
Honora B. Tisell1, Allyson L. Degrassi2, Ryan B. Stephens1, Rebecca J. Rowe1
1 University of New Hampshire, Natural Resources and the Environment, 114 James Hall, 56
College Road, Durham, NH 03824, USA.
2 Shenandoah University, Environment and Society Department, 1460 University Drive,
Winchester, VA 22601 Draft
Correspondent:
Honora B. Tisell; [email protected]; 707-266-3754
University of New Hampshire, Natural Resources and the Environment, 114 James Hall, 56
College Road, Durham, NH 03824, USA.
1
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 2 of 35
Influence of field technique, density, and sex on home range and overlap of the southern red- backed vole (Myodes gapperi)
Honora B. Tisell, Allyson L. Degrassi, Ryan B. Stephens, Rebecca J. Rowe
Abstract
Home range is shaped by an individual’s interactions with the environment and conspecifics, and both size and placement may vary in response to population fluctuations. The method used to collect locational data may also affect home range estimates. We examined the effect of density, sex, and field method on home range of southern red-backed voles (Myodes gapperi Vigors, 1830) inhabiting eastern hemlock forests (Tsuga canadensis L. Carrière).
Twelve mark-recapture grids were used to census M. gapperi from 2014 to 2017. In 2017, individuals were radio collared. Home rangeDraft size, core area size, and shared space were calculated using kernel density estimators from both mark-recapture and radiotelemetry data.
Density effects on home range and core area were analyzed and differences between sex and field method were compared. We found 1) density did not affect home range size, 2) male home range was larger than female home range, 3) females shared space more frequently and to a greater extent with males than other females, and 4) home range estimates were not significantly different between mark-recapture and radiotelemetry. Male home range was, however, larger under radiotelemetry, and may reflect a truncation effect when mark-recapture grid size is smaller than male home range.
Keywords: core area, mark-recapture, Myodes gapperi, Northern Forests, radiotelemetry, rodent,
Southern red-backed vole, space use
2
https://mc06.manuscriptcentral.com//cjz-pubs Page 3 of 35 Canadian Journal of Zoology
Introduction
An individual’s home range represents the space they use and reflects a number of
important ecological processes and behaviors. The placement and size of the home range is an
indicator of a species’ resource requirements (Burt 1943; Powell and Mitchell 2012; Spencer
2012) and is shaped by environmental factors, both abiotic conditions, such as temperature and
water availability (Getz 1968), and biotic conditions, such as the distribution of food and habitat
resources (Gore 1988; Börger et al. 2008). Intraspecific interactions, such as competition and
sociality, can also lead to differences in home range size and placement between sexes or ages
(Jorgensen 1968; Bondrup-Nielson and Karlsson 1985; Bondrup-Nielsen and Ims 1986).
Changes in population density can affect the strength of intraspecific interactions, particularly for
small mammals which often have markedDraft seasonal or inter-annual population fluctuations
(Boonstra and Krebs 2012). For example, at high density, intraspecific competition is greater for
resources, such as food and space, which often leads to smaller home ranges (Adams 2001;
Efford et al. 2016). Because home ranges reflect an animal’s resource needs, movement, and
social behavior (Jorgensen 1968) the study of home range is fundamental to understanding
population dynamics and responses to changing environments (Spencer 2012). Furthermore,
population density and home range (size and overlap) interact to influence ecological processes
that are mediated through contact rates, including disease transmission and genetic structure
(Kitchen et al. 2005; Sanchez and Hudgens 2015).
Identifying an individual’s home range can be challenging because the area is not
uniformly used. Areas in which an individual spends disproportionately more time are
considered core areas (Samuel et al. 1985; Millspaugh et al. 2012a). These core areas often
reflect nesting sites, resting areas, or areas of high resource abundance (Samuel et al. 1985;
3
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 4 of 35
Millspaugh et al. 2006; Spencer 2012). Overall home range size is limited by the distance an individual can travel and readily return to their core area (Börger et al. 2008).
Modelling home range and distinguishing core areas requires multiple spatial or locational observations from the same individual. The ability to differentiate between areas of high and low use may vary depending on the tracking method employed (Samuel et al. 1985).
Mark-recapture and radiotelemetry are two common field methods used to gather locational data.
Choice of method can depend on the research question and terrain of the study site (Millspaugh et al. 2012b; Fuller and Fuller 2012). An individual’s space use is strongly tied to their body size and the scaling relationships among body size, space use, and energetic requirements (Jetz 2004;
Millspaugh et al. 2012a) often dictates the method used to determine home range. In general, because smaller bodied mammals moveDraft shorter distances than larger bodied mammals (Jetz
2004), both mark-recapture and radiotelemetry may be appropriate for measuring their home range. For small-bodied rodents, only a few studies have compared the two field methods for measuring home ranges (e.g. Jones and Sherman 1983, Microtus pennsylvanicus (Ord, 1815);
Desy et al. 1989, Microtus ochrogaster (Wagner, 1842); Ribble et al. 2002, Peromyscus boylii
(Baird, 1855) and P. truei (Shufeldt, 1885)). However, these studies used minimum convex polygon estimators for home range size and this technique cannot give intensity of use estimates
(Harris et al. 1990) and therefore cannot calculate a core area. More recent approaches including utilization distribution (UD) kernel density estimators can calculate the intensity of use of an individual at any given point within their home range (Van Winkle 1975; Powell and Mitchell
2012) and thus can be used to identify core areas (Samuel et al. 1985).
The southern red-backed vole (Myodes gapperi Vigors, 1830) is wide spread throughout
Canada and the northern United States (Merritt 1981), and is typically associated with montane
4
https://mc06.manuscriptcentral.com//cjz-pubs Page 5 of 35 Canadian Journal of Zoology
boreal forests (Miller and Getz 1972). Like many small mammals, it is short-lived, highly fecund
and is known to have years of high and low population density (Boonstra and Krebs 2006;
Sullivan et al. 2017). Females are thought to be territorial and exclude conspecific females from
their defended home range (Perrin 1979b; Bondrup-Nielson and Karlsson 1985). Myodes gapperi
is associated with eastern hemlock (Tsuga canadensis L. Carrière) in northern forests (Miller and
Getz 1972; Yamasaki et al. 1999). These forests are currently threatened by the introduced
hemlock woolly adelgid (Adelges tsugae Annand, 1928) (Orwig et al. 2002; Evans 2004),
underscoring the need to document this species’ space use and better understand and manage for
response to habitat disturbance (Ellison et al. 2005). In this study, we examined 1) whether M.
gapperi density influenced home range or core area size and the amount of shared area, 2)
whether there were differences in homeDraft range and core area size and overlap between sexes, and
3) whether home range and core area estimates differed by the field methods of mark-recapture
and radiotelemetry. We predict that M. gapperi home range and core area will be density
dependent because resources are limited, that there will be sex differences in home range and
core area size and overlap because males and females have different reproductive strategies, and
that the two field methods will result in different estimates of home range because mark-
recapture is spatially constrained while radiotelemetry can follow an animal’s movement freely.
Methods
Study area
Myodes gapperi were captured at the Bartlett Experimental Forest, White Mountain
National Forest, New Hampshire (44° 3′ 7.2′′ N, 71° 17′ 25.1′′ W). Twelve mark-recapture grids
were established in each of three forest types, softwood (n = 4), mixed (n = 4), and hardwood (n
= 4). Forest type was defined by the percentage of hardwood basal area on grids, with hardwood
5
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 6 of 35
grids averaging 91% ± 6.2 and mixed and softwood grids averaging 52.5% ± 8.1 and 25.5% ±
12.7 hardwood basal area, respectively. Dominant hardwood species included American beech
(Fagus grandifolia Ehrh.) and red maple (Acer rubrum L.) and dominant softwood species include eastern hemlock and red spruce (Picea rubens Sarg.). Grids were standardized as an 8 x
8 station array with 15 m spacing (64 stations, 1.1 ha). One Sherman live trap (H. B. Sherman
Co., Tallahassee, Florida) was placed at each station. If trap stations were disturbed by black bear (Ursus americanus Pallas, 1780) activity, those trap nights were considered lost and deducted from the total effort calculated per grid.
Mark-recapture
Mark-recapture surveys were conductedDraft at each grid over four consecutive days in each summer month (June - August) from 2014 - 2017. This timing coincides with the peak breeding season for M. gapperi. Traps were baited with bird seed and polyester fiber was provided for insulation. Traps were checked twice a day in the morning and evening. Captured individuals were identified to species, and the sex, mass (g), age class (juvenile, sub-adult, and adult), and reproductive condition recorded. Age was determined based on body size and reproductive condition (e.g., testes position and nipple size). Individuals were marked with a unique identifier using an ear tag (model 1005-1; National Band and Tag Company, Newport, Kentucky, USA) and a passively integrated transponder (PIT; model HPT9 [9 mm x 2.1 mm] BioMark, Inc,
Boise, Idaho, USA).
6
https://mc06.manuscriptcentral.com//cjz-pubs Page 7 of 35 Canadian Journal of Zoology
Radiotelemetry
In 2017, VHF radio collars (BD-2XC 0.75 g; Holohil Systems Ltd., Carp, Ontario,
Canada) were placed on both adult male and female M. gapperi (n = 23; mass ≥ 19g) that were
captured on mark-recapture grids. Radio collars were ≤ 4% of each individual’s body mass to
ensure collars did not interfere with daily activity and behavior (Millspaugh et al. 2012b). Collars
were equipped with a Tygon sleeve that fastened around the individual’s neck to avoid chafing.
Individuals were collared from July to August with collars on for an average of 16 days
± 2.24. Individuals were tracked twice daily while collared except in cases where weather
conditions prohibited tracking, or the signal could not be located. Locations were marked with
GPS coordinates (Garmin GPSMAP 64s). The protocol for small mammal trapping and
radiotelemetry was approved by the UniversityDraft of New Hampshire Animal Care and Use
Committee (protocol # 120708, 140304, 160507) and followed guidelines outlined by the
American Society of Mammalogists (Sikes et al. 2016).
Home range and core area
Home ranges were modelled with a utilization distribution (UD) kernel density model
(Anderson 1982; Worton 1989) which determines the probability of finding an individual at any
given point within their home range (Van Winkle 1975). Models were calculated for each adult
male and female that was captured ≥ 5 times either on the mark-recapture grids within a year or
with radiotelemetry. Five locations were required for the modelling package ‘adehabitatHR’ to
calculate a home range using UD kernel density models (Calenge 2006). Five captures per
individual also better ensures estimates are based on resident adults and not dispersing subadults.
7
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 8 of 35
The home range was delimited at the 95% UD isopleth to ensure outliers did not skew area estimates (Anderson 1982).
Core areas within the home range are those that have a proportionally higher intensity of use (Samuel et al. 1985; Vander Wal and Rodgers 2012). Core areas were calculated following
Vander Wal and Rodgers (2012) by plotting the UD area (size of region within UD isopleth) against the volume and calculating the point at which the slope of the fitted line was equal to 1; this point defines the boundary of the core area. Core areas and home ranges were calculated using the ‘adehabitatHR’ package (Calenge 2006) in the R software (R Core Team 2017).
A fixed kernel with smoothing bandwidth selected by reference was used to decrease bias for home range estimations and to accommodate our small sample sizes of location data (Worton
1989; Seaman et al. 1999; Kie 2013). ADraft fixed kernel uses a fixed value over the plane of location points (Kie 2013) and results in fine scale detail in the UD home range estimate (Worton 1989;
Vander Wal and Rodgers 2012). The reference bandwidth controls the size of the kernel based on the density of locations points and assumes the kernel density UD has a bivariate normal distribution (Worton 1989; Kie 2013).
Size of home range and core area were compared between males and females using a
Welch two-sample t-test. A linear regression was used to assess whether number of locations impacted home range estimates under both methods and for both sexes. Home ranges (95% isopleth) and core areas were exported as shapefiles and projected into QGIS (QGIS
Development Team 2018) to calculate overlap between individuals for each year. Home range and core area overlap was derived by calculating the percent of area that was shared between one individual and their neighbors and was evaluated separately for neighbors of the same and
8
https://mc06.manuscriptcentral.com//cjz-pubs Page 9 of 35 Canadian Journal of Zoology
different sex. A Welch two-sample t-test was used to determine whether the amount of overlap
was significantly different between mark-recapture and radiotelemetry.
Population density
Mark-recapture data were used to estimate the population density of M. gapperi
separately for each grid in each summer using the POPAN parameterization (Schwarz and
Arnason 1996) of the Jolly-Seber model (Jolly 1965; Seber 1965; Krebs 1999). Calculations
were performed in the R software (R Core Team 2017) package ‘RMARK’ (Laake 2016) which
has an integrated interface to program MARK (White and Burnham 1999). The Jolly-Seber
model assumes the population is “open” between monthly trapping sessions in which individuals
may be born, die, emigrate, or immigrateDraft (Jolly 1965; Seber 1965; Krebs 1999). The time within
the trapping session must be negligible compared to the time between trapping sessions (Krebs
1999). Our three mark-recapture trapping sessions were each 4-days separated by an interval of
three weeks, satisfying such a requirement. The Jolly-Seber model also assumes that each
individual 1) has the same probability of entering a trap (are not “trap happy” or “trap shy”), 2)
has the same probability of surviving between trapping sessions, and 3) is marked and the mark
is not lost (Krebs 1999). POPAN parameterized Jolly-Seber models further assumes births have a
multinomial distribution and that captures are derived from a super population (Schwarz and
Arnason 1996). A 30-meter boundary strip was used to augment the grid and account for the
effective area sampled (Burnham and Anderson 2004).
We tested eight POPAN parameterized Jolly-Seber density models with constant or time-
dependent parameters. Models were ranked using Akaike’s information criterion (AIC). Because
of the small sample sizes, we used the second-order AIC (AICc) which includes a bias-correction
9
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 10 of 35
term (Burnham and Anderson 2004). The highest ranked model for all grids had constant
survival (φ), constant catchability (p) and constant probability of entrance (b). Models with time-
dependent parameters were either ranked low or did not converge.
Analysis of variance (ANOVA) was used to test if density changed with year or forest
type. Linear regression models were used to test whether density was affected by the percent
basal area of eastern hemlock and whether the size and overlap of home range and core area
estimates from mark-recapture grids changed with density. All analyses were conducted in the R
software (R Core Team 2017).
Results
Home range and core area estimates Draft
A total of 417 unique Myodes gapperi were captured over 4 summers from 2014 to 2017.
Forty-two adult females and 35 adult males were captured ≥ 5 times (average 8.08 ± 2.88)
allowing home range during the breeding season to be calculated (Table 1). Of the 77 home
ranges calculated, 12% were on hardwood grids, 28% on mixed grids, and 60% on softwood
grids. Twenty-three M. gapperi were collared in the summer of 2017 on 5 mark-recapture grids
(3 mixed, 2 softwood). Of these, home range was modeled for 21 individuals (10 females, 11
males) that had ≥ 5 locations (average 16.2 ± 2.15).
There was no significant difference in the size of home range (95% isopleth) or core area
calculated by mark-recapture and radiotelemetry for females or males (Fig. 1, Table 2). All core
areas were between the 59-62% isopleth for both methods. Core area was on average 32.80% ±
0.34 of the home range area. A Welch two-sample t-test showed that male home range is significantly larger than female home range, about 5.5 times larger (t = -6.5751, df = 50.234, P <
10
https://mc06.manuscriptcentral.com//cjz-pubs Page 11 of 35 Canadian Journal of Zoology
0.0001), and that male core area is significantly larger than female core area, about 6 times larger
(t = -6.237, df = 49.678, P < 0.0001). Number of locations did not affect the home range estimate
for mark-recapture males or females or radiotelemetry females but did affect the home range
estimate for radiotelemetry males (r2 = 0.5188, F = 9.702, df = 9, P = 0.0124) with more
locations leading to a decrease in the size of the male home range (Fig. 2).
Patterns of shared space use did not differ significantly between the methods of mark-
recapture and radiotelemetry (Table 3) and were pooled to examine proportion of overlap within
and among sexes (Table 4). Eighty-five percent of marked females occurred on grids with
another individual (either male or female) and 95% of males occurred on grids with another
individual. Of these individuals, less than half (45%) of females overlapped core areas with
another female whereas 86% of males overlappedDraft female core areas. Females shared their home
range and core area with males at a significantly greater percentage than they shared area with
other females (Fig. 3, Table 4). Figure 4 shows representative home range placement and overlap
for mark-recapture and radiotelemetry.
Density
Density estimates from mark-recapture data ranged from 1.09 voles/ha to 26.29 voles/ha
across the grids and years (Fig. 5). Density was only estimated for grids on which voles were
captured; of the 12 grids on average 8.75 contained voles each year, with hardwood grids
typically not capturing any voles. A chi-squared test for outliers (Dixon 1950) found the highest
value of density, 26.29, to be an outlier (χ2 = 20.176, P < 0.001). With that outlier removed, the
average density among grids and years was 5.49 ± 0.69 M. gapperi per hectare. Density did not
differ significantly among years (F = 0.664, df = 30, P = 0.581). Density also did not vary among
11
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 12 of 35
softwood and mixed forests when hardwood stands were removed due to low samples size (F =
1.636, df = 28, P = 0.211), nor was density influenced by the percentage basal area of eastern
hemlock (r2 = 0.014, F = 0.404, P = 0.530) among these grids.
Four of the 77 individuals occurred on the grid with the highest (outlier) density and their
home ranges were included in these analyses of density-dependence as removing them did not
affect the overall pattern. Density did not influence female home range size (r2 = 0.06785, F =
2.911, df = 40, P = 0.09571) or male home range size (r2 = 0.05037, F = 1.75, df = 33, P =
0.1949). Density did not affect female core area (r2 = 0.06961, F = 2.993, df = 40, P = 0.09134)
or male core area (r2 = 0.04024, F = 1.384, df = 33, P = 0.2479). The proportions of shared area for core area and for home range were not density dependent for either males or females1. Draft Discussion
This study assessed whether home range or core area size were density dependent or
varied between mark-recapture and radiotelemetry field methods for both female and male
southern red-backed voles. We found no marked fluctuations in the average density of M.
gapperi across the four summers sampled (~ 5-6 voles/ha each summer). However, there was
spatial variation in density among grids, most notably with one mark-recapture grid in one
summer (2016) reaching high-density (26.29 voles/ha) while a grid a short distance away had
low-density (1.64 voles/ha). Although these two grids were in different forest types (softwood
and mixed, respectively), the percent basal area of hemlock, for which M. gapperi has an
affinity, was similar (ca. 40%) on both grids. In contrast to many studies on terrestrial vertebrates
(Adams 2001), we found that home range and core area size were not density dependent.
1 Fig. S1 12
https://mc06.manuscriptcentral.com//cjz-pubs Page 13 of 35 Canadian Journal of Zoology
Another study looking across years in Alberta, Canada also found that home range of M. gapperi
did not change with density in mixed forests of trembling aspen (Populus tremuloides Michx),
white spruce (Picea glauca (Moench) Voss), and black spruce (Picea mariana (Mill.) Britton,
Sterns & Poggenb) (Bondrup-Nielsen 1987). Longer-term studies of M. gapperi have found both
stable densities across years (Mihok 1979; Fuller 1985; Boonstra and Krebs 2012), and boom
and bust 4-year (Elias et al. 2006; Fauteux et al. 2015) and 7-year (Sullivan et al. 2017) cycles.
Density independence from home range dynamics suggests environmental factors (e.g.
seasonality, Beer 1961; Boonstra and Krebs 2012 or habitat quality and type, Bondrup-Nielson
1987) or interspecific interactions (e.g. competition, Dufour et al. 2015 or different foraging
approaches, Mitchell and Powell 2004) may be structuring space use among M. gapperi in this
region. We found that density did not affectDraft home range size during the breeding season, which
could indicate that environment is important in determining the population size of this species.
With eastern hemlock ecosystems under threat of hemlock woolly adelgid (Evans 2004), more
research is needed to understand how this species’ density and population persistence may be
affected.
Amount of home range or core area overlap can show how individuals within a species
interact with each other and indicate the degree of territoriality of a species (Jorgensen 1968).
Female M. gapperi shared more area with males than other females. Females had very little
overlap between core areas (~18%) which suggests that females actively defend or avoid other
female’s core areas. Previous studies using mark-recapture grids have also suggested that female
M. gapperi are territorial (Perrin 1979b; Bondrup-Nielson and Karlsson 1985). Females are
likely defending their nests during the summer breeding season , which are located within the
13
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 14 of 35
core area of their home range and the areas directly around nest sites are important for gathering resources to rear young.
The patterns of shared space we documented during the breeding season among males as well as between males and females are indicative of a promiscuous breeding strategy (Merritt
1981). Males did not exclude other males from home range or core areas, suggesting they are not guarding one particular female. Male home ranges often overlap with multiple females and females were often sharing their home range and core areas with multiple males. These patterns highlight that to males, females are a resource sought out during the breeding season for mating opportunities (Bondrup-Nielson 1986). Many studies use female home range as representative of the species’ space use because male home range also reflects mating behavior when studies are conducted during the breeding season (PerrinDraft 1979a; Fuller 1985). Because male and female M. gapperi are similar in body mass, variation in home range size most likely reflects differences in reproductive strategies with males seeking out multiple mates and thus occupying a larger extent whereas females are tied to a specific nest location and that may subsequently restrict the size of their home range.
Home range and core area estimates were not significantly different between mark- recapture and radiotelemetry methods for either female or male M. gapperi. This is consistent with previous studies of other vole species (Jones and Sherman 1983; Desy et al. 1989), however
Desy et. al (1989) found that mark-recapture derived home ranges for voles were smaller than radiotelemetry derived home ranges, but only for individuals that were caught < 8 times. Desy et. al (1989) did not specify how sex was handled in their analysis and this result could have been a result of sex differences; in our study we saw an effect of number of locations on home range estimates, but only for males. Previous studies on other small rodents found that radiotelemetry
14
https://mc06.manuscriptcentral.com//cjz-pubs Page 15 of 35 Canadian Journal of Zoology
methods estimated larger home ranges than mark-recapture methods (Bergstrom 1988; Ribble et
al. 2002). Studies on small to medium-sized marsupials (Sunquist et al. 1987, common opossum
(Didelphis marsupialis Linnaeus, 1758); Bradshaw and Bradshaw 2002, honey possum (Tarsipes
rostratus Gervais and Verreaux, 1842); Lira and Fernandez 2008, neotropical opossum
(Philander frenatus (Olfers, 1818)) have also found radiotelemetry to yield consistently larger
home range estimates when compared to mark-recapture. These studies, in combination with our
work, suggests differences between mark-recapture and radiotelemetry estimates may be species
or system-specific, or could reflect differences in mark-recapture study design.
Number of locations had an impact on the size of radiotelemetry male home range with
more locations leading to a decrease in home range size. This may be due to males having larger
home ranges and as location points are addedDraft there becomes greater precision in the reference
smoothing bandwidth used to calculate the UD kernel density. Because males typically range
farther than females (Bondrup-Nielsen 1987) they are more likely to range outside of a telemetry
receiver’s ability to pick up the signal. Our data support this with seven of 11 males missing
location data on a given survey session due to inability to locate a signal. Five of these males
returned to within signal strength range in a subsequent day; the remaining 2 may have
experienced a collar failure or a predation event. In contrast, three of 12 females had missing
location data due to inability to locate a radiotelemetry signal however, the signal was never
located again for those three females indicating possible collar failure or predation event.
Because dispersal is sex biased such that males disperse more often and farther (Greenwood
1980) and we restricted collars to adults, dispersal events are unlikely to explain these lost
signals.
15
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 16 of 35
Although home range and core area estimates under mark-recapture and radiotelemetry were not significantly different for males (P = 0.16), male home range and core area estimates were larger when calculated from radiotelemetry (2.1 ha) than from mark-recapture (1.3 ha).
These mark-recapture grids are 1.1 ha in size and our findings suggest grid size may be truncating the male home range such that mark-recapture estimates approximate grid size at 1.3 ha and radiotelemetry estimates exceed mark-recapture estimates by over 50% (Fig. 4). Overall, our findings suggest mark-recapture and radiotelemetry provide similar estimates of home range and core area size for M. gapperi, however we suggest using both mark-recapture and radiotelemetry simultaneously or larger grid sizes (ca. 3 ha) if using mark-recapture alone to calculate accurate estimates of male space use. If space use of females is the only interest, then mark-recapture alone on a 1 ha grid wouldDraft be suitable.
Acknowledgements:
We thank M. Yamasaki and C. Costello of the US Forest Service for support and access to the
Bartlett Experimental Forest. Data were collected with the help of C. Burke, H. Comstock, M.
Cyr, J. Guptill, B. Lewis, R. Migotsky, L. Poland, T. Remick, S. Smith, A. Uccello, and M.
Wilson. Helpful comments from 2 anonymous reviewers improved the quality and clarity of this work. Partial funding was provided by the New Hampshire Agricultural Experiment Station.
This is Scientific Contribution Number 2825. This work was supported by the USDA National
Institute of Food and Agriculture McIntire-Stennis Project (0229197 & 1006881).
16
https://mc06.manuscriptcentral.com//cjz-pubs Page 17 of 35 Canadian Journal of Zoology
References:
Adams, E.S. 2001. Approaches to the study of territory size and shape. Annu. Rev. Ecol. Syst.
32(1): 277–303. doi:10.1146/annurev.ecolsys.32.081501.114034.
Anderson, D.J. 1982. The home range: a new nonparametric estimation technique. Ecology,
63(1): 103–112. doi:10.2307/1937036.
Beer, J.R. 1961. Winter home ranges of the red-backed mouse and white-footed mouse. J.
Mammal. 42(2): 174. doi:10.2307/1376827.
Bergstrom, B.J. 1988. Home ranges of three species of chipmunks (Tamias) as assessed by
radiotelemetry and grid trapping. J. Mammal. 69(1): 190–193. doi:10.2307/1381774.
Bondrup-Nielsen, S. 1987. Demography of Clethrionomys gapperi in different habitats. Can. J.
Zool. 65(2): 277–283. doi:10.1139/z87-043.Draft
Bondrup-Nielsen, S., and Ims, R.A. 1986. Reproduction and spacing behaviour of females in a
peak density population of Clethrionomys glareolus. Ecography (Cop.). 9(2): 109–112.
doi:10.1111/j.1600-0587.1986.tb01198.x.
Bondrup-Nielson, S. 1986. Analysis of spacing behaviour of females from a live-trapping study
of Clethrionomys gapperi. Finnish Zool. Bot. Publ. Board, 23(3): 261–267.
Bondrup-Nielson, S., and Karlsson, F. 1985. Movements and spatial patterns in populations of
Clethrionomys species: a review. Finnish Zool. Bot. Publ. Board, 22(3): 385–392.
Boonstra, R., and Krebs, C.J. 2006. Population limitation of the northern red-backed vole in the
boreal forests of northern Canada. J. Anim. Ecol. 75(6): 1269–1284. doi:10.1111/j.1365-
2656.2006.01149.x.
Boonstra, R., and Krebs, C.J. 2012. Population dynamics of red-backed voles (Myodes) in North
America. Oecologia, 168(3): 601–620. doi:10.1007/s00442-011-2120-z.
17
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 18 of 35
Börger, L., Dalziel, B.D., and Fryxell, J.M. 2008. Are there general mechanisms of animal home
range behaviour? A review and prospects for future research. Ecol. Lett. 11(6): 637–650.
doi:10.1111/j.1461-0248.2008.01182.x.
Bradshaw, S.D., and Bradshaw, F.J. 2002. Short-term movements and habitat use of the
marsupial honey possum (Tarsipes rostratus). J. Zool. (Lond.) 258(3): 343–348.
doi:10.1017/S0952836902001486.
Burnham, K.P., and Anderson, D.R. 2004. Model selection and multimodel inference: a practical
information-theoretic approach. Edited By K.P. Burnham and D.R. Anderson. Springer New
York, New York, NY. doi:10.1007/b97636.
Burt, W.H. 1943. Territoriality and home range concepts as applied to mammals. J. Mammal.
24(3): 346. doi:10.2307/1374834. Draft
Calenge, C. 2006. The package “adehabitat” for the R software: A tool for the analysis of space
and habitat use by animals. Ecol. Modell. 197(3–4): 516–519.
doi:10.1016/j.ecolmodel.2006.03.017.
Desy, E.A., Batzli, G.O., and Jike, L. 1989. Comparison of vole movements assessed by live
trapping and radiotracking. J. Mammal. 70(3): 652–656. doi:10.2307/1381444.
Dixon, W.J. 1950. Analysis of extreme values. Ann. Math. Stat. 21(4): 488–506.
doi:10.1214/aoms/1177729747.
Dufour, C.M.S., Meynard, C., Watson, J., Rioux, C., Benhamou, S., Perez, J., et al. 2015. Space
use variation in co-occurring sister species: response to environmental variation or
competition? PLoS One, 10(2): e0117750. doi:10.1371/journal.pone.0117750.
Efford, M.G., Dawson, D.K., Jhala, Y. V., and Qureshi, Q. 2016. Density-dependent home-range
size revealed by spatially explicit capture-recapture. Ecography (Cop.). 39(7): 676–688.
18
https://mc06.manuscriptcentral.com//cjz-pubs Page 19 of 35 Canadian Journal of Zoology
doi:10.1111/ecog.01511.
Elias, S.P., Witham, J.W., and Hunter, M.L. 2006. A cyclic red-backed vole (Clethrionomys
gapperi) population and seedfall over 22 years in Maine. J. Mammal. 87(3): 440–445.
doi:10.1644/05-MAMM-A-170R1.1.
Ellison, A.M., Bank, M.S., Clinton, B.D., Colburn, E.A., Elliott, K., Ford, C.R., et al. 2005. Loss
of foundation species: consequences for the structure and dynamics of forested ecosystems.
Front. Ecol. Environ. 3(9): 479–486. doi:10.1890/1540-
9295(2005)003[0479:LOFSCF]2.0.CO;2.
Evans, R.A. 2004. Hemlock woolly adelgid and the disintegration of eastern hemlock
ecosystems. Park Sci. 22(2): 53–56.
Fauteux, D., Cheveau, M., Imbeau, L., andDraft Drapeau, P. 2015. Cyclic dynamics of a boreal
southern red-backed vole population in northwestern Quebec. J. Mammal. 96(3): 573–578.
doi:10.1093/jmammal/gyv062.
Fuller, M.R., and Fuller, T.K. 2012. Radio-telemetry equipment and applications for carnivores.
In Carnivore Ecology and Conservation. Edited by L. Boitani and R. A. Powell. Oxford
University Press. pp. 152–168. doi:10.1093/acprof:oso/9780199558520.003.0007.
Fuller, W.A. 1985. Demography of Clethrionomys gapperi, parapatric C. rutilus, and sympatric
Peromyscus maniculatus in northern Canada. Ann. Zool. Fenn. 22(3): 229–241.
Getz, L.L. 1968. Influence of water balance and microclimate on the local distribution of the
redback vole and white-footed mouse. Ecology, 49(2): 276–286. doi:10.2307/1934456.
Gore, J.A. 1988. Habitat structure and the distribution of small mammals in a northern
hardwoods forest. Proceedings of the Management of Amphibians, reptiles, and Small
Mammals in North America, Flagstaff, AZ, July 19-21, 1988. pp 319-327.
19
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 20 of 35
Greenwood, P.J. 1980. Mating systems, philopatry and dispersal in birds and mammals. Anim.
Behav. 28(4): 1140–1162. doi:10.1016/S0003-3472(80)80103-5.
Harris, S., Cresswell, W.J., Forde, P.G., Trewhella, W.J., Woollard, T., and Wray, S. 1990.
Home-range analysis using radio-tracking data: a review of problems and techniques
particularly as applied to the study of mammals. Mammal Rev. 20(2/3): 97–123.
doi:10.1111/j.1365-2907.1990.tb00106.x.
Jetz, W. 2004. The scaling of animal space use. Science, 306(5694): 266–268.
doi:10.1126/science.1102138.
Jolly, G.M. 1965. Explicit estimates from capture-recapture data with both death and
immigration-stochastic model. Biometrika, 52(1/2): 225. doi:10.2307/2333826.
Jones, E.N., and Sherman, L.J. 1983. A Draftcomparison of meadow vole home ranges derived from
grid trapping and radiotelemetry. J. Wildl. Manage. 47(2): 558. doi:10.2307/3808537.
Jorgensen, C.D. 1968. Home range as a measure of probable interactions among populations of
small mammals. J. Mammal. 49(1): 104–112. doi:10.2307/1377732.
Kie, J.G. 2013. A rule-based ad hoc method for selecting a bandwidth in kernel home-range
analyses. Anim. Biotelemetry, 1(1): 13. doi:10.1186/2050-3385-1-13.
Kitchen, A.M., Gese, E.M., Waits, L.P., Karki, S.M., and Schauster, E.R. 2005. Genetic and
spatial structure within a swift fox population. J. Anim. Ecol. 74(6): 1173–1181.
doi:10.1111/j.1365-2656.2005.01017.x.
Krebs, C.J. 1999. Ecological methodology. Addison Wesley Longman Inc., Menlo Park,
California.
Laake, J.L. 2016. RMark: An R Interface for Analysis of Capture-Recapture Data with MARK.
Lira, P.K., and Fernandez, F.A. dos S. 2008. A comparison of trapping- and radiotelemetry-
20
https://mc06.manuscriptcentral.com//cjz-pubs Page 21 of 35 Canadian Journal of Zoology
based estimates of home range of the neotropical opossum Philander frenatus. Mamm.
Biol. - Z. Säugetierkd. 74(1): 1–8. Urban & Fischer. doi:10.1016/J.MAMBIO.2008.05.002.
Merritt, J.F. 1981. Clethrionomys gapperi. Mamm. Species, No. 146: 1–9. doi:10.2307/3503900.
Mihok, S. 1979. Demography, behavior, and protein polymorphism in subarctic Clethrionomys
gapperi. PhD Dissertation. The University of Alberta.
Miller, D.H., and Getz, L.L. 1972. Factors infuencing the local distribution of the redback vole,
Clethrionomys gapperi, in New England. Occas. Pap. Biol. Sci. Ser. 2(17): 115–138.
Millspaugh, J.J., Gitzen, R.A., Belant, J.L., Kays, R.W., Keller, B.J., Kesler, D.C., et al. 2012a.
Analysis of radiotelemetry data. In The Wildlife Techniques Manual: Research. Edited by
N.J. Silvy. Johns Hopkins University Press, Baltimore, MD. pp. 480–501.
Millspaugh, J.J., Kesler, D.C., Kays, R.W.,Draft Gitzen, R.A., Schulz, J.H., Rota, C.M., et al. 2012b.
Wildlife radiotelemetry and remote monitoring. In The Wildlife Techniques Manual:
Research. Edited by N.J. Silvy. Johns Hopkins University Press, Baltimore, MD. pp. 258–
283.
Millspaugh, J.J., Nielson, R.M., McDonald, L.L., Marzluff, J.M., Gitzen, R. A, Rittenhouse,
C.D., Hubbard, M.W., and Sheriff, S.L. 2006. Analysis of resource selection using
utilization distributions. J. Wildl. Manage. 70(2): 384–395. doi:10.2193/0022-
541X(2006)70[384:AORSUU]2.0.CO;2.
Mitchell, M.S., and Powell, R.A. 2004. A mechanistic home range model for optimal use of
spatially distributed resources. Ecol. Modell. 177(1–2): 209–232.
doi:10.1016/j.ecolmodel.2004.01.015.
Orwig, D.A., Foster, D.R., and Mausel, D.L. 2002. Landscape patterns of hemlock decline in
New England due to the introduced hemlock woolly adelgid. J. Biogeogr. 29(10–11): 1475–
21
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 22 of 35
1487. doi:10.1046/j.1365-2699.2002.00765.x.
Perrin, M.R. 1979a. The Roles of Reproduction, Survival, and Territoriality in the Seasonal
Dynamics of Clethrionomys gapperi Populations. Acta Oecologica, 24(35): 475–500.
doi:10.4098/AT.arch.79-45.
Perrin, M.R. 1979b. Seasonal variation in the growth, body composition and diet of
Clethrionomys gapperi in spruce forest. Acta Theriol. (Warsz). 24(23): 299–318.
doi:10.4098/AT.arch.79-30.
Powell, R.A., and Mitchell, M.S. 2012. What is a home range? J. Mammal. 93(4): 948–958.
doi:10.1644/11-mamm-s-177.1.
QGIS Development, T. 2018. QGIS Geographic Information System. Open Source Geospatial
Foundation Project. Available fromDraft http://qgis.org/en/site.
R Core, T. 2017. R: A Language and Environment for Statistical Computing. Vienna, Austria.
Available from https://www.r-project.org.
Ribble, D.O., Wurtz, A.E., McConnell, E.K., Buegge, J.J., and Welch Jr., K.C. 2002. A
comparison of home ranges of two species of Peromyscus using trapping and radiotelemetry
data. J. Mammal. 83(1): 260–266. doi:10.1644/1545-
1542(2002)083<0260:ACOHRO>2.0.CO;2.
Samuel, M.D., Pierce, D.J., and Garton, E.O. 1985. Identifying areas of concentrated use within
the home range. J. Anim. Ecol. 54(3): 711. doi:10.2307/4373.
Sanchez, J.N., and Hudgens, B.R. 2015. Interactions between density, home range behaviors, and
contact rates in the Channel Island fox (Urocyon littoralis). Ecol. Evol. 5(12): 2466–2477.
doi:10.1002/ece3.1533.
Schwarz, C.J., and Arnason, A.N. 1996. A general methodology for the analysis of capture-
22
https://mc06.manuscriptcentral.com//cjz-pubs Page 23 of 35 Canadian Journal of Zoology
recapture experiments in open populations. Biometrics, 52(3): 860. doi:10.2307/2533048.
Seaman, D.E., Millspaugh, J.J., Kernohan, B.J., Brundige, G.C., Raedeke, K.J., and Gitzen, R.A.
1999. Effects of sample size on kernel home range estimates. J. Wildl. Manage. 63(2): 739.
doi:10.2307/3802664.
Seber, G.A.F. 1965. A note on the multiple-recapture census. Biometrika, 52(1–2): 249–260.
doi:10.1093/biomet/52.1-2.249.
Sikes, R.S., and the Animal Care and Use Committee of the American Society of Mammalogists.
2016. 2016 Guidelines of the American Society of Mammalogists for the use of wild
mammals in research and education. J. Mammal. 97(3): 663–688.
doi:10.1093/jmammal/gyw078.
Spencer, W.D. 2012. Home ranges and Draftthe value of spatial information. J. Mammal. 93(4): 929–
947. doi:10.1644/12-MAMM-S-061.1.
Sullivan, T.P., Sullivan, D.S., Boonstra, R., Krebs, C.J., and Vyse, A. 2017. Mechanisms of
population limitation in the southern red-backed vole in conifer forests of western North
America: insights from a long-term study. J. Mammal. 98(5): 1367–1378.
doi:10.1093/jmammal/gyx082.
Sunquist, M.E., Austad, S.N., and Sunquist, F. 1987. Movement patterns and home range in the
common opossum (Didelphis marsupialis). J. Mammal. 68(1): 173–176.
doi:10.2307/1381069.
Vander Wal, E., and Rodgers, A.R. 2012. An individual-based quantitative approach for
delineating core areas of animal space use. Ecol. Modell. 224(1): 48–53.
doi:10.1016/j.ecolmodel.2011.10.006.
White, G.C., and Burnham, K.P. 1999. Program MARK: survival estimation from populations of
23
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 24 of 35
marked animals. Bird Study, 46(sup1): S120–S139. doi:10.1080/00063659909477239.
Van Winkle, W. 1975. Comparison of several probabilistic home-range models. J. Wildl.
Manage. 39(1): 118. doi:10.2307/3800474.
Worton, B.. J. 1989. Kernel methods for estimating the utilization distribution in home-range.
Ecology, 70(1): 164–168. doi:10.2307/1938423.
Yamasaki, M., DeGraaf, R.M., and Lanier, J.W. 1999. Wildlife habitat associations in eastern
hemlock - birds, smaller mammals, and forest carnivores. In Proceedings: Symposium on
sustainable management of hemlock ecosystems in eastern North America. Edited by K.A.
McManus, K.S. Shields, and D.R. Souto. USDA Forest Service, Northeastern Forest
Experiment Station. Newtown Square, PA. Gen. Tech. Rep. NE-267. pp. 135-143. Draft
24
https://mc06.manuscriptcentral.com//cjz-pubs Page 25 of 35 Canadian Journal of Zoology
Table 1. Summary of the trap nights (one trap, set for 24-hour period) and captures for the
southern red-backed vole (M. gapperi) from 2014 - 2017. Total captures include number of times
each individual was captured. Home ranges were calculated from individuals with 5 or more
captures (77 home ranges).
Captures Home ranges Year Trap nights Individuals Total Female Male 2014 9192 78 216 7 9 2015 8726 123 324 13 6 2016 9006 104 306 10 8 2017 8738 112 346 12 12 Draft
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 26 of 35
Table 2. Comparison of mark-recapture (MR) and radiotelemetry (RT) methods on the size of home range and core area of the southern red-backed vole (M. gapperi). Area is recorded as mean size in hectares ± SE. A Welch two-sample t-test was conducted to compare estimates between methods.
Sex Distribution Method Mean size (ha) P-value MR (n = 42) 0.402 ± 0.048 Home range 0.7547 RT (n = 10) 0.370 ± 0.043 Female MR (n = 42) 0.135 ± 0.016 Core area 0.6055 RT (n = 10) 0.117 ± 0.014 MR (n = 35) 1.329 ± 0.142 Home range 0.1617 RT (n = 11) 2.107 ± 0.501 Male MR (Draftn = 35) 0.448 ± 0.050 Core area 0.2092 RT (n = 11) 0.701 ± 0.184
https://mc06.manuscriptcentral.com//cjz-pubs Page 27 of 35 Canadian Journal of Zoology
Table 3. Comparison of estimates of shared space use between mark-recapture (MR) and
radiotelemetry (RT) methods for the southern red-backed vole (M. gapperi). Proportion of
overlap was measured for both home range and core area and a two-sample t-test was conducted
separately within and between sexes. Overlap was considered the total proportion of the area
shared. Comparison was conducted on proportions but, for clarity, results are shown as percent ±
SE. Number of overlapping individuals is the average number of individuals that shared space
with each focal individual ± SE.
Focal Shared Mean overlap # overlapping Distribution Method P-value individual individual (%) individuals MR (n = 16) 20.41 ± 4.42 1.37 ± 0.15 Core area 0.3169 RT (n = 3) 9.57 ± 2.71 1.33 ± 0.33 Female Draft MR (n = 27) 39.45 ± 5.65 1.70 ± 0.14 Home range 0.3608 RT (n = 8) 29.23 ± 6.79 2.75 ± 0.45 Female MR (n = 28) 65.27 ± 6.76 1.86 ± 0.13 Core area 0.6060 RT (n = 8) 57.41 ± 15.65 2.75 ± 0.45 Male MR (n = 31) 91.95 ± 2.64 2.32 ± 0.16 Home range 0.1929 RT (n = 9) 77.84 ± 9.69 3.33 ± 0.55 MR (n = 26) 31.41 ± 4.37 1.96 ± 0.17 Core area 0.2974 RT (n = 10) 22.88 ± 6.30 2.10 ± 0.43 Female MR (n = 28) 49.23 ± 5.12 2.57 ± 0.23 Home range 0.1162 RT (n = 10) 64.40 ± 6.45 3.00 ± 0.60 Male MR (n = 30) 57.82 ± 5.04 1.53 ± 0.14 Core area 0.4493 RT (n = 9) 66.19 ± 10.95 3.56 ± 0.44 Male MR (n = 31) 74.06 ± 4.45 1.81 ± 0.15 Home range 0.1703 RT (n = 9) 86.45 ± 5.82 4.00 ± 0.50
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 28 of 35
Table 4. Summary of shared space within and among female and male southern red-backed voles (M. gapperi). Averages reflect areas generated under both mark-recapture and radiotelemetry as no differences were detected between methods (Table 3). Overlap was considered the total proportion of the area shared between members of the overlap sex.
Comparisons were conducted on proportions but, for clarity, are shown as percent ± SE.
Differences between male and female home range and core area were calculated using two- sample t-tests.
Focal Shared Mean overlap # overlapping Distribution P-value individual individual (%) individuals
Female (n = 19) 18.69 ± 3.84 1.37 ± 0.14 Core area < 0.0001 Male (n = 36) 63.52 ± 6.21 2.06 ± 0.15 Female Draft Female (n = 35) 37.14 ± 4.64 1.94 ± 0.16 Home range < 0.0001 Male (n = 40) 88.78 ± 3.06 2.55 ± 0.18 Female (n = 36) 29.04 ± 3.62 2.00 ± 0.17 Core area < 0.0001 Male (n = 39) 59.76 ± 4.59 2.00 ± 0.20 Male Female (n = 38) 53.22 ± 4.24 2.68 ± 0.23 Home range < 0.0001 Male (n = 40) 76.84 ± 3.75 2.30 ± 0.22
https://mc06.manuscriptcentral.com//cjz-pubs Page 29 of 35 Canadian Journal of Zoology
Figure 1. Estimates of home range and core area size utilizing mark-recapture and
radiotelemetry methods for both adult female and male southern red-backed vole (M. gapperi).
Boxplots denote 25th, 50th and 75th percentiles: whiskers represent the lowest and highest datum
within the 1.5 interquartile range. Dots represent outliers. Welch two-sample t-tests were used to
compare methods among sexes of which there were no significant differences.
Figure 2. Linear regressions were used to assess whether home range area was sensitive to the
number of locations used to construct home ranges for southern red-backed vole (M. gapperi)
females (top) and males (bottom). Each data point represents the size of an individual’s home range and number of locations used to calculateDraft that home range area. The shaded area represents the 95% confidence interval. Mark-recapture (left) data are compiled over 2014 - 2017 and
radiotelemetry data (right) are from 2017.
Figure 3. Mean proportion of shared area between individual southern red-backed voles (M.
gapperi). Female-male refers to the proportion of female home range that is shared with male
individuals and male-female refers to the proportion of male home range that is shared with
females. Boxplots denote 25th, 50th and 75th percentiles: whiskers represent the lowest and
highest datum within the 1.5 interquartile range. Dots represent values that fall outside the 1.5
interquartile range.
Figure 4. Representative southern red-backed vole (M. gapperi) home ranges calculated using
locations gathered through mark-recapture (left) and radiotelemetry (right). These data are not
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 30 of 35
paired: for clarity, grids with greater sample sizes have been depicted here. Filled polygons represent female individuals and lines represent male individuals. The white box is the mark- recapture grid boundary.
Figure 5. Density estimates for the southern red-backed vole (M. gapperi) each year. Density was calculated from grids with vole captures > 0 (average 8.75 grids per year). Boxplots denote
25th, 50th and 75th percentiles: whiskers represent the lowest and highest datum within the 1.5 interquartile range. The one dot falls outside the 1.5 interquartile range and was identified as an outlier based on a chi-squared test for outliers. Draft
https://mc06.manuscriptcentral.com//cjz-pubs Page 31 of 35 Canadian Journal of Zoology
Draft
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 32 of 35
Draft
https://mc06.manuscriptcentral.com//cjz-pubs Page 33 of 35 Canadian Journal of Zoology
Draft
https://mc06.manuscriptcentral.com//cjz-pubs Canadian Journal of Zoology Page 34 of 35
Draft
https://mc06.manuscriptcentral.com//cjz-pubs Page 35 of 35 Canadian Journal of Zoology
Draft
https://mc06.manuscriptcentral.com//cjz-pubs