91
MOVEMENTS AND RESOURCE USE BY MOOSE IN
TRADITIONAL AND NONTRADITIONAL HABITATS
IN NORTH DAKOTA
James J. Maskey Jr.1,2 and Rick A. Sweitzer1,3
1Department of Biology, 10 Cornell Street, Stop 9019, University of North Dakota, Grand Forks,
North Dakota 58202, USA; 2Department of Biology, University of Mary, 7500 University Drive, Bismarck,
North Dakota 58504, USA; 3Great Basin Institute,16750 Mt. Rose Hwy, Reno, Nevada 89511, USA
ABSTRACT: In the past several decades, moose (Alces alces) have expanded their range in North
Dakota from primarily forested areas to the prairie/agriculture mosaic of the state. As a result, moose are
now well- established in a large portion of North Dakota, yet little is known about their ecology in the
state. We examined the home ranges, habitat selection, and diets of moose in both traditional (forested)
and nontraditional ranges (prairie/agricultural) and inferred whether range expansion is the result of
agriculture-related landscape changes. From 2004 to 2006, we placed GPS radio-collars on a total of 14
moose in two study areas: Turtle Mountains (forested) and Lonetree (prairie/agricultural). Total and
seasonal home ranges were larger for Lonetree moose, and moose in both study areas selected strongly
for wooded habitat. In both study areas seasonal diets ranged from 65 to 99% woody browse, with forbs
15% of summer diets. In the Lonetree area row crops made up the second highest consumed forage in
fall (12%) and winter (29%) diets. Larger home ranges in the Lonetree area may reflect the low avail-
ability and scattered distribution of wooded habitat. Further, the strong selection for planted woodlands
and the high proportion of woody browse and row crops in the diet of Lonetree moose suggests that
conversion of the native prairie to agriculture has facilitated range expansion by moose in North Dakota.
ALCES VOL. 55: 91–104 (2019)
Key Words: Alces alces, browse, cropland, diet, habitat selection, home ranges, North Dakota, prairie,
woodland
INTRODUCTION
Moose (Alces alces) are native to North
Dakota with their traditional range encom-
passing the aspen (Populus tremuloides) and
bur oak (Quercus macrocarpa) forests of the
Turtle Mountains and Pembina Hills along
the northern edge of the state (Knue 1991).
While moose were extirpated from North
Dakota by the late 1800s, they had begun to
re-establish a population in the state by the
1960s. After re-colonizing their historic
range, by the 1980s moose had expanded
their range to include large expanses of for-
mer tall and mixed grass prairie that had
been greatly modified by conversion to agri-
culture and widespread planting of tree rows
to reduce wind erosion subsequent to the
Dust Bowl years of the 1930s (Knue 1991,
Licht 1997).
The colonization and range expansion
by moose in North Dakota are likely the
result of conversion of the native prairie
landscape to an agricultural mosaic that pro-
vides suitable cover and forage otherwise
absent in unaltered tall or mixed grass prairie
habitats. Although moose are known to per-
sist in other landscapes modified by humans
such as clear-cuts and agricultural areas
within forested landscapes (Leptich and
Gilbert 1989, Rempel et al. 1997, Schneider
and Wasel 2002), the agriculture-dominated
landscape of the northern Great Plains
MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER ALCES VOL. 55, 2019
92
represents a unique habitat for the species
that was not occupied prior to human-induced
habitat change. While numerous prior efforts
have provided insight into moose move-
ments and resource use in traditional habi-
tats (Kearney and Gilbert 1976, Leptich and
Gilbert 1989, Cederlund and Sand 1994,
MacCracken et al. 1997, Labonte et al.
1998), the ecology and behavior of moose in
the prairie ecoregions of North America is
relatively unknown.
The purpose of this project was to inves-
tigate the ecology of moose in both tradi-
tional woodland habitats and the recently
colonized prairie region of North Dakota,
including how this species may be taking
advantage of landscape alterations to extend
its range. The specific objectives were to 1)
examine seasonal and annual movements
and habitat use of moose in the prairie and
woodland regions of North Dakota, 2) inves-
tigate the diet of moose in prairie and wood-
land regions of North Dakota, and 3)
compare movements, habitat use, and diets
of moose in these two regions. To meet these
objectives, we selected study areas that were
representative of traditional and more
recently colonized habitats. First, the for-
ested Turtle Mountains comprise a major
portion of the historic range of moose in
North Dakota and was one of the areas first
recolonized upon their return (Knue 1991,
Seabloom et al. 2011). Second, the Lonetree
Wildlife Management Area (WMA) is an
agricultural mosaic characteristic of the hab-
itats more recently colonized by moose. It is
representative of most of the landscape of
eastern and central North Dakota, which
now comprises much of the primary range of
this species in the state.
STUDY AREA
The Turtle Mountains (48° 57ʹ N, 99°
53ʹ 00” W; Fig. 1) are located along the
Canadian border and are characterized by
hilly wooded terrain and numerous small
lakes and wetlands with interspersed agri-
cultural fields, pastureland, and hay fields,
especially near the southern edge of the
area. The forest of the Turtle Mountains is
comprised primarily of aspen and bur oak
along with green ash (Fraxinus pennsylvan-
ica), paper birch (Betula papyrifera), bal-
sam poplar (Populus balsamea) , and box
elder (Acer negundo), with an understory of
chokecherry (Prunus virginiana), hazel
(Corylus cornuta), and several species of
willow (Salix spp.). Typical herbaceous
species include sarsaparilla (Aralia nudi-
caulis), alfalfa (Medicago sativa), brome
(Bromus spp.), fescue (Festuca spp.),
wheatgrass (Agropyron spp.), sedges (Carex
spp.), baneberry (Actea spp.), false lily
of the valley (Maianthemum canadensis),
wild vetch (Vicia americana), and Virginia
anemone (Anemone virginiana) (Stevens
1966, Bakke 1980, ND Forest Service
2003).
The Lonetree WMA is large, encom-
passing 134 km2 in the central part of the
state (47°30ʹ N, 100°15ʹ W; Fig. 1). It con-
sists of farmland initially purchased by the
U.S. Bureau of Reclamation to be used
as part of the Missouri River Garrison
Diversion Project (Garrison Diversion
Project 2019). Following the cancellation of
that portion of the project, management of
the land was turned over to the North Dakota
Game and Fish Department. Habitats
include native mixed grass prairie, corn
(Zea mays) and sunflower (Helianthus ann-
uus) food plots (range = 6–31 ha), numerous
seasonal and semi-permanent wetlands,
small impoundments along the Sheyenne
River, and planted woodlands in the form of
linear tree rows or larger block plantings
(Smith et al. 2007). The surrounding area is
comprised primarily of pasture and hay land
ALCES VOL. 55, 2019 MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER
93
as well as crop fields consisting mostly of
small grains. Planted tree rows and wood-
lots are present with some natural wood-
lands occurring in woody draws along the
Missouri Escarpment that marks the border
between the Northern Glaciated Plains and
Missouri Coteau ecoregions (USEPA 1996).
Typical grassland plants found in the area
include prairie junegrass (Koeleria macran-
tha), indiangrass (Sorghastrum nutans),
needle and thread grass (Hesperostipa
comata), brome, wheatgrass, and alfalfa.
Common tree species in planted and/or
native woodlands include green ash, box
elder, American elm (Ulnus americana),
Russian olive (Elaeagnus angustafolia),
plum (Prunus spp.), apple (Malus spp.),
chokecherry, fireberry hawthorn (Crataegus
chrysocarpa), serviceberry (Amelanchier
alnifolia), and willow.
METHODS
Study animals
Global positioning system (GPS) radio-
collars (Lotek Wireless Inc. Newmarket,
Ontario, Canada) were placed on 14 adult
moose (5 cows, 1 bull in the Lonetree WMA;
4 cows, 4 bulls in the Turtle Mountains) in
January 2004–2006. Only moose in the
Lonetree WMA study area were captured in
2004, with subsequent expansion to the
Turtle Mountains in 2005 and 2006. Moose
were captured from helicopter with the use
of a net gun. Capture operations were per-
formed by Leading Edge Aviation (Lewiston,
Idaho, USA), and all methods were approved
by the University of North Dakota (IACUC
Project #0506-3). Collars were set to acquire
a location every 4 h, and location data were
stored on board. After ~ 52 weeks, radio-col-
lars were recovered when moose were
Fig. 1. Location of the Lonetree Wildlife Management Area and Turtle Mountains study areas in North
Dakota, USA. Boundaries delineate North Dakota counties.
MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER ALCES VOL. 55, 2019
94
recaptured or after they dropped off via
timed-release mechanisms. Moose were
periodically monitored by aerial and ground-
based VHF telemetry.
Home range estimation
Total and seasonal moose home range
sizes (km2) were estimated with a 95%
fixed-kernel estimator, with seasons defined
as winter (1 January–30 April), summer
(1 May–31 August), and fall (1 September–31
December) for all analyses. We carried out
these calculations using all available loca-
tions for non-dispersing moose, with disper-
sal defined as locations for a moose that
deviated from other grouped relocations for
that animal (Dodge et al. 2004). Seasonal
home range sizes were estimated for each
moose for all seasons for which at least 30
locations were available, and total home
ranges were estimated for moose with at
least 30 locations in every season (Seaman
et al. 1999). Moose were considered to
exhibit seasonal migrations if < 25% of their
seasonal home ranges overlapped (Dodge
et al. 2004). Location data were input into
ArcMap 9.2 (ESRI Inc., Redlands,
California, USA) and home range calcula-
tions were performed using the Home Range
Extension (Rodgers and Carr 1998).
At the time of our study, least squares
cross validation (LSCV) was the most rec-
ommended technique to determine the opti-
mal smoothing parameter for fixed-kernel
home range estimation (Worton 1995,
Seaman and Powell 1996, Seaman et al.
1999). However, we experienced similar
problems with this method as reported by
others (Silverman 1986, Hemson et al.
2005); LSCV was unable to calculate a
smoothing parameter for most sets of loca-
tions, and if it did, the multi-modal nature of
the locations produced home ranges that
were dramatically under-smoothed. To deal
with these problems, we used biased cross
validation (BCV) to calculate the smoothing
parameters for all fixed-kernel home range
estimations (Wand and Jones 1995, Rodgers
and Carr 1998). Although BCV has not been
commonly applied to estimate the smooth-
ing parameters for home range estimates, the
statistical literature has demonstrated its util-
ity in selecting kernel bandwidth, as well as
its potential superiority to LSCV (Sain et al.
1994, Wand and Jones 1995, Rodgers and
Carr 1998). We compared total home range
sizes between study sites with a two-sample
t-test. For all moose with home range esti-
mates for all seasons, we also compared sea-
sonal home range sizes among seasons and
study sites with repeated-measures ANOVA.
When necessary, home range sizes were nat-
ural log transformed to meet the assumptions
of parametric tests. All statistical compari-
sons were performed in the statistical pack-
age R 2.6 (R Core Development Team 2007).
Habitat selection
We first estimated the extent of the area
available to moose in each study area by
constructing 99% fixed-kernel home ranges
for each moose, and then combined the home
ranges for each study site into a single poly-
gon. We then used land cover data from the
United States Geological Survey’s Gap
Analysis Program (GAP) compiled from
1992 to 1999 (Strong et al. 2005) as well as
National Wetland Inventory 1:24,000 digital
quadrangles (USFWS 2000) to determine
habitat types available to moose. To make
the comparison of habitat use between study
sites possible, land cover data were collapsed
into 4 habitat types using Spatial Analyst
in ArcMap 9.2 (ESRI Inc., Redlands,
California, USA). These were defined as
woodland (all planted and naturally occur-
ring woodlands), wetland (temporary, sea-
sonal, permanent, and semi-permanent
wetlands), grasslands (planted non-native
grasses, hay fields, old fields, and planted
ALCES VOL. 55, 2019 MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER
95
or naturally occurring prairie), and crops
(all planted row crops or grains).
Next, because preliminary analysis indi-
cated that the coarse spatial resolution (30-
m) of the GAP data were insufficient to detect
small areas of habitat, we improved the reso-
lution of terrestrial habitat layers by re-digi-
tizing data based on 1-m resolution aerial
photos of each study site (National
Agricultural Imagery Program, USDA-FSA
2005). We did this by overlaying GAP habi-
tat layers onto the aerial photos in ArcMap,
then manually re-digitizing the GAP layers to
conform to the habitat boundaries indicated
on the photos. We modified the wetland hab-
itat layer by considering all temporary and
seasonal wetlands to be part of the terrestrial
habitat in which they were imbedded, as
these wetlands are typically inundated only
during the spring and do not provide a source
of emergent or submergent aquatic vegeta-
tion (USFWS 2000). We also adjusted wet-
land habitat availability to account for the
presence of several larger lakes in the 2 study
areas. Because the deep-water areas in these
lakes were unlikely available to moose, we
created a 100 m buffer layer that extended
from the shoreline into each lake. This dis-
tance was chosen as a conservative estimate
of the extent of the littoral zone, where water
depth was shallow and emergent and sub-
mergent plants would occur. The area of this
buffer layer was considered the amount of
lake habitat available to moose. We measured
the area of each habitat type in each study
area using the X-Tools extension for ArcMap
9.2 (ESRI Inc., Redlands, California, USA)
and then determined the proportional avail-
ability of each habitat (Table 1).
All locations for individual moose at
each study site were separated into seasons.
We then calculated Manly’s standardized
selection ratios for each moose with ≥30
locations in a season (Manly et al. 2002,
Osko et al. 2004). This method produces
selection ratios that represent the probability
of a moose using a particular habitat if all
habitats were equally available, and the
selection ratios for all habitats sum to 1.
Because there were four habitat types in this
study, a selection ratio of 0.25 indicates
non-selection (not different than by chance),
while a selection ratio >0.25 represent posi-
tive selection for a habitat type.
Diet
In 2005–2006 we collected 5 samples of
fresh moose feces monthly from each study
area by searching several locations distrib-
uted across each study site, with no more
than 2 samples/month collected from a single
location (e.g., from the same clear-cut).
Samples were combined to generate a series
of 2-month composite fecal samples. Samples
were sent to the Wildlife Habitat and Nutrition
Laboratory at Washington State University
(Pullman, Washington, USA) for microhisto-
logical determination of plant fragments and
estimates of diets to the genus and species
level (Van Vuren 1984). Forage plants were
classified into 5 categories: woody browse,
grasses and sedges, forbs, crops, and other
(fruits, nuts, aquatic vegetation). Results of
this diet analysis were grouped by season
based on the same criteria used for home
range and habitat use analyses.
RESULTS
Home range and movement
A high rate of collar failure in 2005 pre-
vented the calculation of all seasonal and
Table 1. Proportional availability of each of the four
major habitat types in the Lonetree and Turtle
Mountains study areas in North Dakota, USA.
Woodland Wetland Grass Crop
Lonetree 0.025 0.053 0.400 0.522
Turtle Mountains 0.451 0.170 0.252 0.127
MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER ALCES VOL. 55, 2019
96
annual home ranges. Estimates of home
range size (95% fixed kernel method) ranged
from 59.2 to 262.6 km2 (n = 5) in the Lonetree
WMA study area which were larger than in
the Turtle Mountains (9.6–47.7 km2, n = 4;
t5.3= 3.7, P = 0.01; mean number of loca-
tions = 2709). Seasonal home range estimates
were also larger in Lonetree than the Turtle
Mountains (F1,25 = 13.3, P = 0.0012, mean
number of locations = 807), ranging from
18.8 to 292.8 km2 and 1.0 to 44.7 km2, respec-
tively. Home range size did not differ among
seasons (F2, 25 = 0.1, P = 0.91). One moose
was excluded from comparisons because its
30 locations occurred in a single month.
None of the moose exhibited seasonal
migrations. One dispersed from the Lonetree
WMA in March 2004, with the 5 others
remaining in the general vicinity. Radio-
collars were recovered successfully from 7
animals (the 8th failed) in the Turtle
Mountains; all remained in the Turtle
Mountains throughout the study.
Habitat use
Moose strongly selected for wooded
habitat in all seasons in both study areas;
conversely, no selection for cropland or
grassland habitats was measured in either
study area. Moose in the Turtle Mountains
also selected for wetland habitats during the
summer (Table 2).
Moose diets
Moose consumed mostly woody plants in
both the Lonetree (≥65%) and Turtle Mountains
(≥83%) areas in all seasons of the year (Fig. 2).
Consumption of woody browse was particu-
larly high in the Turtle Mountains; for exam-
ple, moose consumed 99% woody browse
during winter, primarily aspen (36%) and wil-
low (20%). Willow and aspen were also
important components of the diets in the Turtle
Mountains during summer (15 and 12%) and
fall (19 and 23%). Bur oak stems and leaves
were also common forage items in these sea-
sons, representing 20% and 23% of summer
and fall diets, respectively. In the Lonetree
area, Russian olive was the most common
woody browse consumed in all seasons and
was 50% of the fall diet, followed by willow
(10% in summer) and cottonwood (11% in
winter). Row crops (primarily corn) were also
a major component of the diets in the Lonetree
area during fall (12%) and winter (29%; Fig. 2).
In contrast, row crops were absent from sam-
ples collected in the Turtle Mountains, although
alfalfa was an important component in summer
and fall diets (13%), representing 90% of forbs
consumed in these seasons. Grasses (≤3%) and
fruits and nuts (≤1%) were minor components
of the diet in both study areas, while emergent
and submergent aquatic vegetation were ≤1%
of the diet during the open water seasons of
summer and fall.
Table 2. Mean Manly’s standardized selection ratios (SE) for four habitat types based on data from 13
GPS-collared moose in the Lonetree and Turtle Mountains study areas in North Dakota, USA. Bold
numbers indicate positive selection (> 0.25) for a habitat type.
Study site Season n Woodland Wetland Crop Grassland
Lonetree Winter 6 0.95(0.008) 0.013(0.005) 0.013(0.003) 0.024(0.004)
Summer 6 0.89(0.016) 0.048(0.018) 0.024(0.006) 0.034(0.005)
Fall 5 0.84(0.033) 0.067(0.017) 0.051(0.017) 0.038(0.009)
Turtle Mountains Winter 7 0.76(0.01) 0.16(0.021) 0.031(0.017) 0.048(0.012)
Summer 4 0.56(0.06) 0.30(0.039) 0.015(0.007) 0.13(0.068)
Fall 4 0.54(0.1) 0.21(0.052) 0.10(0.040) 0.15(0.085)
ALCES VOL. 55, 2019 MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER
97
DISCUSSION
Sample size and radio-collar
performance
The low density of moose limited the
number of animals that could be studied in
the Lonetree WMA. Observations from
fixed-wing aircraft indicated that only 5
moose were in the vicinity in 2004, and all
were successfully captured and radio-
collared that year; likewise, in subsequent
years we successfully radio-collared nearly
every known moose in the area. Radio-
collar failures limited our ability to make
comparisons between moose in their tradi-
tional range and the prairie habitats. Overall,
9 of 22 radio-collars failed prematurely,
Fig. 2. Seasonal diet composition (%) of moose in the Lonetree (a) and Turtle Mountains (b) study
areas in North Dakota, USA.
MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER ALCES VOL. 55, 2019
98
with 6 of 10 in the Turtle Mountains includ-
ing 4 of 5 deployed in 2005. We do not
believe that these failures generated any
obvious sources of bias within our data.
More importantly, we radio-collared a large
proportion of the moose in the study areas
and our results provide novel information
about habitat and forage use in the North
Dakota landscape.
Migration and dispersal
Our results indicate that moose in North
Dakota are largely non-migratory. Elsewhere,
moose may migrate to avoid deep snows at
high elevation or to seek conifer forests that
provide thermal cover or reduced snow
depth (Pierce and Peek 1984, Ballard et al.
1991, Hundertmark 1998, Thompson and
Stewart 1998, Poole and Stuart-Smith 2006).
Because moose range in North Dakota lacks
these characteristics, elevational differences
do not lead to variability in snowfall or
depth, and conifer forests are absent.
Additionally, habitat selection and diet com-
position of moose in both study areas indi-
cated that moose selected for wooded
habitats and consumed primarily woody
browse in all seasons. Other habitats such as
croplands realized increased seasonal use,
but these habitats were interspersed within
the mosaic of woodland patches used year-
round by moose. As a result, moose did not
need to migrate to gain access to seasonally
preferred forage.
Home range
Detailed comparisons of home range
sizes across studies are difficult because of
differences in the number of locations col-
lected and the variety of estimators used.
However, home range size is expected to be
a function of the energetic requirements of
an animal and the spatial distribution of nec-
essary resources (McNab 1963, Elchuk and
Weibe 2003, Mitchell and Powell 2004).
Thus, where required resources are widely
dispersed, home range size will be larger.
Mean total home range size (160.5 km2,
SE = 38.9) for Lonetree moose was near the
upper range of averages reported for non-mi-
gratory moose (174–290 km2; Grauvogel
1984, Ballard et al. 1991, Stenhouse et al.
1995). Ballard et al. (1991) attributed large
home ranges to the high proportion of
unavailable habitat within home ranges.
Similarly, 92% of the landscape in the
Lonetree WMA consisted of grassland and
cropland habitats that moose mostly avoided,
while the wooded habitats that moose
selected for comprised only 2.5% of the
landscape. In contrast, in the Turtle
Mountains with a high proportion of wood-
land habitat (45.1%), moose had smaller
total home ranges (x̄ = 27.7 km2, SE = 10.0)
similar to those (2–43 km2) for other pre-
dominantly wooded areas in eastern North
America (Phillips et al. 1973, Addison et al.
1980, Leptich and Gilbert 1989, Garner and
Porter 1990, Dodge et al. 2004).
We did not observe significant differ-
ences in home range size among seasons,
but seasonal home-range size varied con-
siderably among moose. Although energy
constraints associated with moving through
deep snow or predator avoidance may result
in smaller home ranges during winter
(Phillips et al. 1973, Thompson and
Vukelich 1981, Dussault et al. 2005), snow
depths considered limiting to moose are
rare (> 70 cm; Hundertmark 1998) and
large predators are absent in North Dakota.
Thus, seasonal home ranges were more
likely determined by the distribution of
seasonally-important forage resources
(Doerr 1983, Lynch and Morgantini 1984,
Leptich and Gilbert 1989). As such, the dif-
ferences in size of seasonal home ranges
among moose likely reflected the spatial
pattern of available seasonal resources
where moose resided.
ALCES VOL. 55, 2019 MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER
99
Habitat selection and diet
While the characterization of habitat
types was general in order to facilitate com-
parisons between the study areas, the habitat
selection analyses nevertheless provided
important insight into how moose utilized
available habitats in North Dakota. Numerous
researchers have demonstrated the impor-
tance of a variety of types of woody habitats
in providing forage and/or cover for moose
(e.g., Peek et al. 1976, Peek 1998, Courtois
et al. 2002). Therefore, it was not surprising
that moose in North Dakota exhibited a
strong selection for woody habitats in all
seasons in both study areas. In adjacent
Minnesota and many areas of Canada, moose
inhabited early successional forest created by
periodic fire or insect outbreaks and that are
now maintained largely by forest harvesting
(Peterson 1955, Phillips et al. 1973, Peek
et al. 1976). The forests that cover nearly half
of the Turtle Mountains study area represent
this “typical” moose habitat, and the tree
plantings on and around the Lonetree WMA,
though more scattered across the landscape,
also appear to provide important forest habi-
tat for moose.
Woody browse dominated the diets of
moose in both study areas, similar to prior
research demonstrating the importance of
woodlands in providing forage for moose
(Belovsky 1981, Renecker and Schwartz
1998). Diets in the Turtle Mountains con-
sisted in large part of browse species such as
aspen, willow, birch, juneberry, and cherry
that are typically considered common forage
items of moose. In addition to many of
these traditional browse species, the Turtle
Mountains also had an abundance of bur oak
which was a major component of summer
and fall diets. In the Lonetree area, the most
common woody plants (except willow) are
not typically found in traditional moose
range, and this was reflected in the local diet.
For example, Russian olive is a common
shrub in tree plantings and was the most
abundant browse item (25% of overall diets),
and green ash and box elder, also commonly
planted, were ~11% of the winter and sum-
mer diets.
Woodlands was the only habitat moose
selected for in all seasons, but seasonal
changes in diet and the use of croplands
and wetlands suggest that moose also
selectively used seasonally available for-
age in other habitat types. Thus, selection
ratios may not entirely reflect the impor-
tance of habitats other than woodlands. For
example, while moose avoided grassland
habitats in both study areas, alfalfa was
13% of the summer and fall diets in the
Turtle Mountains, indicating that this forb
was an important supplemental seasonal
forage. Likewise, moose exhibited nega-
tive selection for croplands in all seasons,
even though its use was greater in fall than
in other seasons and corn was an important
part of the fall and winter diets of Lonetree
moose. This apparent lack of selection may
reflect the different composition of crop-
lands in the two study areas. Crops in the
Turtle Mountains consisted almost entirely
of small grains (wheat, barley) that were
not expected to serve as moose forage, and
in Sheridan and Wells Counties where
Lonetree WMA is located, ~26% of the
total land area was planted in wheat and
barley in 2005 with only 2% corn and 3%
sunflowers (USDA 2005). In contrast, if
habitat selection analyses were confined to
the boundary of the Lonetree WMA where
the only croplands were corn and sun-
flower food plots, then moose would show
an overall positive selection for cropland
habitats (Manly’s standardized selection
ratio = 0.29).
Moose avoided wetlands in most sea-
sons, possibly due to avoidance of open areas
during warm daytime temperatures (Olson
et al. 2016). However, relative to winter, we
MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER ALCES VOL. 55, 2019
100
observed an increase in wetland use during
summer and fall which was primarily driven
by Turtle Mountain moose; use of wetlands
was low year-round for Lonetree moose
(Table 2). The increased use of wetlands was
not reflected in the diets, as aquatic vegeta-
tion was a minor component (≤1%) of sum-
mer and fall diets in both study areas.
However, these plants likely play an import-
ant role because it is believed moose con-
sume aquatic plants for their critical minerals
(De Vos 1958, Belovsky and Jordan 1981)
and high digestibility (MacCracken et al.
1997). The latter may represent a potential
limitation of this study because it may cause
underrepresentation of aquatic plants in fecal
samples. Alternatively, the relative increase
in wetland use in summer and fall may have
been independent of forage requirements and
triggered by thermoregulatory behavior or to
avoid insects (De Vos 1958, Belovsky and
Jordan 1981).
The combined results of home range,
habitat use, and diet analyses provide
insight into the factors influencing space
use by moose in both traditional and prairie
habitats in North Dakota. While wooded
habitats appear to be critical for moose
throughout their range in North Dakota,
other seasonally available resources such as
corn and alfalfa may provide supplemental
food sources. Further, the strong selection
for planted woodlands and use of crops as a
food source in the Lonetree WMA support
the hypothesis that range expansion by
moose is the direct result of landscape
modifications occurring since European
settlement.
Management implications
Moose are a prized big game species in
North Dakota, with >15,000 hunters apply-
ing annually for a once-in-a-lifetime license
(North Dakota Game and Fish Department
2019). This study provides ecological
information about the state’s moose popula-
tion that will help managers make informed
decisions to maintain and enhance this
unique wildlife resource. While moose have
expanded their range to include areas of
North Dakota that were historically prairie,
the woodland habitats that they depend on
constitute a very small proportion of the
overall landscape in these areas, thereby
requiring moose to have large home ranges
to acquire sufficient resources. As a result,
managers should recognize that prairie habi-
tats are likely capable of supporting fewer
moose than forested areas, and that the con-
tinued persistence of prairie populations of
moose will be dependent on the maintenance
of forest habitat. Additionally, the planted
woodlands and food plots of the Lonetree
WMA may make this area particularly
attractive to prairie moose. The continued
management of this and other WMAs to pro-
vide food and cover for wildlife should help
support the state’s moose population in
non-traditional range where availability of
preferred habitats is limited.
ACKNOWLEDGMENTS
We thank North Dakota EPSCoR, the
North Dakota Game and Fish Department,
the U.S.D.I. Bureau of Reclamation, the
U.S. Fish and Wildlife Service, the North
Dakota Chapter of The Wildlife Society,
the University of North Dakota Biology
Department, and the Wheeler Scholarship
for funding this project. We would also like
to thank W. Jensen, R. Johnson, R. Kreil,
and S. Peterson for permitting us to cap-
ture moose and conduct work on North
Dakota Game and Fish lands. Additionally,
we are grateful to R. Newman, B. Rundquist,
and J. Vaughan for their helpful input on
this manuscript. Finally, gratitude is
expressed to J. Smith, E. Pulis, T. Manuwal,
J. Faught, and J. Rubbert for their assis-
tance in the field.
ALCES VOL. 55, 2019 MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER
101
REFERENCES
Addison, R. B., J. C. WilliAmson, B. P
sAundeRs, and d. FRAseR. 1980. Radio-
tracking of moose in the boreal forest of
northwestern Ontario. Canadian Field-
Naturalist 94: 269–276.
BAkke, e. l. 1980. Movements and Habitat
Use of Ruffed Grouse in the Turtle
Mountains, North Dakota. M. S. Thesis.
University of North Dakota, Grand
Forks, North Dakota, USA.
BAllARd, W. B., J. s. WhitmAn, and d. J.
Reed. 1991. Population dynamics of
moose in south-central Alaska. Wildlife
Monographs 114: 1–49.
Belovsky, G. e. 1981. Food plant selection by
a generalist herbivore: the moose. Ecology
62: 1020–1030. doi:10.2307/1937001
_____, and P. A. JoRdAn. 1981. Sodium
dynamics and adaptations of a moose
population. Journal of Mammalogy 62:
613–627. doi:10.2307/1380408
CedeRlund, G. h., and h. sand. 1994.
Home-range size in relation to age and
sex in moose. Journal of Mammalogy
75: 1005–1012. doi:10.2307/1376618
CouRtois, R., C. dussAult, F. Potvin, and
G. dAiGle. 2002. Habitat selection by
moose (Alces alces) in clear-cut land-
scapes. Alces 38: 117–192.
de vos, A. 1958. Summer observations on
moose behavior in Ontario. Journal of
Mammalogy 39: 128–139. doi:10.2307/
1376618
dodGe, W. B., JR, s. R. WinteRstein, d. e.
BeyeR, JR., and h. CAmPA iii. 2004.
Survival, reproduction, and movements
of moose in the western Upper Peninsula
of Michigan. Alces 40: 71–86.
doeRR, J. G. 1983. Home-range size, move-
ments, and habitat use in two moose, Alces
alces, populations in southeastern Alaska.
Canadian Field-Naturalist 97: 79–88.
dussAult, C., J. P. ouellet, R. CouRtois, J.
huot, l. BReton, and h. JoliCieR. 2005.
Linking moose habitat selection to limit-
ing factors. Ecography 28: 619–628.
doi:10.1111/j.2005.0906-7590.04263.x
elChuk, C. l., and k. l. WieBe. 2003. Home
range size of northern flickers (Colaptes
auratus) in relation to habitat and paren-
tal attributes. Canadian Journal of Zoology
81: 954–961. doi:10.1139/z03-077
GARneR, d. l., and W. F. PoRteR. 1990.
Movements and seasonal home ranges
of bull moose in a pioneering Adirondack
population. Alces 26: 80–85.
GARRison diveRsion PRoJeCt. 2019. History
and federal legislation. (accessed September 2019).
GRAuvoGel, C. A. 1984. Seward Peninsula
moose population identity study. Federal
Aid in Wildlife Restoration Final Report.
Alaska Department of Fish and Game,
Juneau, Alsaka, USA.
hemson, G., P. Johnson, A. south, R.
kenWARds, R. RiPley, and d. mACdonAld.
2005. Are kernels the mustard? Data from
global positioning system collars suggests
problems for kernel home-range analysis
with least-squares cross validation.
Journal of Animal Ecology 74: 455–463.
doi:10.1111/j.1365-2656.2005. 00944.x
hundeRtmARk, k. J. 1998. Home range, dis-
persal, and migration. Pages 303–335 in
A. W. Franzmann and C. C. Schwartz,
editors. Ecology and Management of
North American Moose. Smithsonian
Institution Press, Washington, DC, USA.
keARney, s. R., and F. F. GilBeRt. 1976.
Habitat use by moose and white-tailed
deer on sympatric range. Journal of
Wildlife Management 40: 645–657.
doi:10.2307/3800559
knue, J. 1991. Big game in North Dakota: a
short history. North Dakota Game and
Fish Department, Bismarck, North
Dakota, USA.
lABonte, J., J. P. Quellet, and R. CouRtois.
1998. Moose dispersal and its role in the
maintenance of harvested populations.
Journal of Wildlife Management 62:
225–235. doi:10.2307/3802282
lePtiCh, d. J., and J. R. GilBeRt. 1989.
Summer home range and habitat use by
http://www.garrisondiv.org/about/HistoryFederalLegislation/
http://www.garrisondiv.org/about/HistoryFederalLegislation/
http://www.garrisondiv.org/about/HistoryFederalLegislation/
MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER ALCES VOL. 55, 2019
102
moose in northern Maine. Journal of
Wildlife Management 53: 880–885.
doi:10.2307/3809581
liCht, d. s. 1997. Ecology and Economics of
the Great Plains. University of Nebraska
Press, Lincoln, Nebraska, USA.
lynCh, G. m., and l. e. moRGAntini. 1984.
Sex and age differential in seasonal
home-range size of moose in northcentral
Alberta. Alces 20: 61–78.
mACCRACken, J. G., v. vAn BAllenBeRGhe, and
J. m. Peek. 1997. Habitat relationships of
moose on the Copper River Delta in coastal
south-central Alaska. Journal of Wildlife
Management 61: (suppl. 136) 1–52.
mAnly, B. F. J., l. l. mCdonAld, d. l.
thomAs, t. l. mCdonAld, and W. P.
eRiCkson. 2002. Resource Selection
by Animals: Statistical Design and
Analysis. 2nd edition. Kluwer Academic
Publishers, Dordrecht, the Netherlands.
mCnAB, B. k. 1963. Bioenergetics and the
determination of home range. American
Naturalist 97: 130–140. doi:10.1086/
282264
mitChell, m. s., and R. A. PoWell. 2004. A
mechanistic home range model for opti-
mal use of spatially distributed resources.
Ecological Modelling 177: 209–232.
doi:10.1016/j.ecolmodel.2004.01.015
noRth dAkotA FoRest seRviCe. 2003.
Legendary forests: State forests guide.
North Dakota Forest Service, Bottineau,
ND, USA.
noRth dAkotA GAme And Fish dePARtment.
2019. Moose hunting. (accessed September
2019).
olson, B. t., s. k. Windels, R. A. moen,
and n. P. mCCAnn. 2016. Moose modify
bed sites in response to high tempera-
tures. Alces 52: 153–160.
osko, t. J., m. n. hiltz, R. J. hudson, and
s. m. WAsel. 2004. Moose habitat
preferences in response to changing avail-
ability. Journal of Wildlife Management
68: 576–584. doi:10.2193/0022-541X
(2004)068[0576:MHPIRT] 2.0.CO;2
Peek, J. m. 1998. Management of moose hab-
itat. Pages 377–401 in A. W. Franzmann
and C. C. Schwartz, editors. Ecology
and Management of North American
Moose. Smithsonian Institution Press,
Washington, DC, USA.
_____, d. l. uRiCh, and R. J. mACkie. 1976.
Moose habitat selection and relation-
ships to forest management in northeast-
ern Minnesota. Wildlife Monographs
48: 1–65.
PeteRson, R. l. 1955. North American
Moose. University of Toronto Press,
Toronto, Ontario, Canada.
PhilliPs, R. l., W. e. BeRG, and d. B. siniFF.
1973. Moose movement patterns and
range use in northeastern Minnesota.
Journal of Wildlife Management 37:
266–278. doi:10.2307/3800117
PieRCe, J. d., and J. m. Peek. 1984.
Moose habitat use and selection patterns
in north-central Idaho. Journal of
Wildlife Management 48: 1335–1343.
doi:10.2307/ 3801794
Poole, k. G., and k. stuARt-smith. 2006.
Winter habitat selection by female
moose in western interior montane for-
ests. Canadian Journal of Zoology 84:
1823–1832. doi:10.1139/z06-184
R CoRe develoPment teAm. 2007. R: a
language and environment for statisti-
cal computing. R Foundation for
Statistical Computing, Vienna, Austria.
(accessed
September 2019).
RemPel, R. s., P. C. elkie, A. R. RodGeRs,
and m. J. GluCk. 1997. Timber manage-
ment and natural disturbance effects on
moose habitat: landscape evaluation.
Journal of Wildlife Management 61:
517–524. doi:10.2307/3802610
ReneCkeR, l. A., and C. C. sChWARtz. 1998.
Food habits and feeding behavior.
Pages 403–439 in A. W. Franzmann and
C. C. Schwartz, editors. Ecology
and Management of North American
Moose. Smithsonian Institution Press,
Washington, DC, USA.
https://gf.nd.gov/hunting/moose
https://gf.nd.gov/hunting/moose
http://www.R-project.org/
ALCES VOL. 55, 2019 MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER
103
RodGeRs, A. R., and A. P. CARR. 1998.
Home Range Extension (HRE) for
ArcViewGIS. Ontario Ministry of
Natural Resources Centre for Northern
Forest Ecosystem Research, Thunder
Bay, Ontario, Canada.
sAin, s. R., k. A. BAGGeRty, and d. W.
sCott. 1994. Cross-validation of multi-
variate densities. Journal of the American
Statistical Association 89: 807–817. doi:
10.1080/01621459.1994.10476814
sChneideR, R. R., and s. m. WAsel. 2000.
The effect of human settlement on the
density of moose in northern Alberta.
Journal of Wildlife Management 64:
513–520. doi:10.2307/3803249
seABloom, R. W., J. W. hoGAnson, W. F.
Jensen. 2011. The mammals of North
Dakota. Institute for Regional Studies,
North Dakota State University, Fargo,
North Dakota, USA.
seAmAn, d. e., and R. A. PoWell. 1996. An
evaluation of the accuracy of kernel den-
sity estimators for home range analysis.
Ecology 77: 2075–2085. doi:10.2307/
2265701
_____, J. J. millsPAuGh, B. J. keRnohAn, G.
C. BRundiGe, k. J. RAedeke, and R. A.
Gitzen. 1999. Effects of sample size on
kernel home range estimates. Journal of
Wildlife Management 63: 739–747.
doi:10.2307/3802664
silveRmAn, B. W. 1986. Density Estimation
for Statistics and Data Analysis. Chapman
and Hall, London, United Kingdom.
smith, J. R., R. A. sWeitzeR, and W. J. Jensen.
2007. Diets, movements, and conse-
quence of providing wildlife food plots
for white-tailed deer in central North
Dakota. Journal of Wildlife Management
71: 2719–2726. doi:10.2193/2006-379
stenhouse, G. B., P. B. lAtouR, l. klutny,
n. mACleAn, and G. GloveR. 1995.
Productivity, survival and movement
of female moose in a low-density
population, Northwest Territories,
Canada. Arctic 48: 57–62. doi:10.14430/
arctic1224
stevens, o. A. 1966. Plants of Bottineau
County, North Dakota. North Dakota
School of Forestry, Bottineau, North
Dakota, USA.
stRonG, lAuRenCe l., h. thomAs
skleBAR, and kevin e. keRmes. 2005.
North Dakota GAP analysis project.
Northern Prairie Wildlife Research
Center, Jamestown, North Dakota.
Online. (Version 12JUN2006;
accessed September 2019).
thomPson, i. d., and R. W. steWARt. 1998.
Management of moose habitat.
Pages 377–401 in A. W. Franzmann and
C. C. Schwartz, editors. Ecology and
Management of North American
Moose. Smithsonian Institution Press,
Washington, DC, USA.
_____, i. d., and m. F. vukeliCh. 1981. Use
of logged habitats in winter by moose
calves in northeastern Ontario. Canadian
Journal of Zoology 59: 2103–2114.
doi:10.1139/z81-287
united stAtes dePARtment oF AGRiCultuRe.
2005. National Agricultural Statistics
Service website. U. S. Department of
Agriculture, National Agricultural
Statistics Service, North Dakota Field
Office, Fargo, ND, USA. (accessed September
2019).
_____-Farm Service Agency. 2005.
Statewide NAIP photography 2005.
USDA-FSA Aerial Photography Field
Office, Salt Lake City, Utah, USA.
(accessed
September 2019).
united stAtes enviRonmentAl PRoteCtion
AGenCy. 1996. Level III ecoregions of
the continental United States. National
Health and Environmental Effects
Research Laboratory Map M-1, vari-
ous scales. United States Environmental
Protection Agency, Corvallis, Orgeon,
USA.
united stAtes Fish And WildliFe seRviCe.
2000. National Wetlands Inventory
http://www.npwrc.usgs.gov/projects/ndgap/
http://www.npwrc.usgs.gov/projects/ndgap/
http://www.nass.usda.gov/nd/
http://www.nass.usda.gov/nd/
http://www.apfo.usda.gov
MOVEMENTS AND RESOURCE USE BY MOOSE – MASKEY AND SWEITZER ALCES VOL. 55, 2019
104
website. U.S. Department of the Interior,
Fish and Wildlife Service, Washington,
DC, USA.
(accessed September 2019).
vAn vuRen, d. h. 1984. Summer diets of
bison and cattle in southern Utah. Journal
of Range Management 37: 260–261.
doi:10.2307/3899151
Wand, m. P., and m. C. Jones. 1995. Kernel
Smoothing. Chapman and Hall. London,
United Kingdom.
WoRton, B. J. 1995. Using Monte Carlo sim-
ulation to evaluate kernel-based home
range estimators. Journal of Wildlife
Management 59: 794–800. doi:10.2307/
3801959
http://www.fws.gov/nwi/