South Dakota State University South Dakota State University
Open PRAIRIE: Open Public Research Access Institutional Open PRAIRIE: Open Public Research Access Institutional
Repository and Information Exchange Repository and Information Exchange
Electronic Theses and Dissertations
1987
Habitat Selection and Sexual Suggestion of Rocky Mountain Habitat Selection and Sexual Suggestion of Rocky Mountain
Bighorn Sheep in Custer State Park, South Dakota Bighorn Sheep in Custer State Park, South Dakota
Larry J. Layne
Follow this and additional works at: https://openprairie.sdstate.edu/etd
Part of the Natural Resources and Conservation Commons
Recommended Citation Recommended Citation Layne, Larry J., "Habitat Selection and Sexual Suggestion of Rocky Mountain Bighorn Sheep in Custer State Park, South Dakota" (1987). Electronic Theses and Dissertations. 163. https://openprairie.sdstate.edu/etd/163
This Thesis - Open Access is brought to you for free and open access by Open PRAIRIE: Open Public Research Access Institutional Repository and Information Exchange. It has been accepted for inclusion in Electronic Theses and Dissertations by an authorized administrator of Open PRAIRIE: Open Public Research Access Institutional Repository and Information Exchange. For more information, please contact [email protected].
HABITAT SELECTION AND SEXUAL SEGREGATION OF ROCKY
MOUNTAIN BIGHORN SHEEP IN CUSTER STATE PARK, SOUTH DAKOTA
BY
LARRY J. LAYNE
A thesis submitted in partial fulfillment of the requirements for the degree
Master of Science Major in Wildlife and Fisheries Sciences
(Wildlife Option) South Dakota State University
1987
HABIT,�T SELECTIO�I A1\JD SEXUAL SEGREGATION OF ROCKY MOUNTAH:
BIGHORN SHEEP IN CUSTEq STATE PARK, SOUTH DAKOTA
This thesis is app:Jved as a cr2dit3ble and indep�ndent
investigation by th� candidate for the degree Master of Science,
and is acceptable for meeting the th8sis requir2ments for this
degree. Acceptance of this th2sis does not i�ply that the
conclusions re3ched by t�e candij�te are necessarily the
conclusions of the majcr depart�ent.
Thomas/R.Ac'cabe Thesis .:.\dvisor
Char 1 �s G. · �s·63f�t Head Dept. of �ildlife and Fisheries Sciences
JJate
t\CK.t.�OWLEDGEMENTS
I would especially like to thank Dr. T. R. McCabe
for 311 of his encouragement and assistance during this
study, and for his help i� analysis and intarpretation cf
the data.
I am grateful to W. Winter, R. Noem, a�d w. Jackson
for their help in the field and contir.ued support for the
study. Also, to �. Bland and G. Brundige for their halp
with vegetation measurements and to all the students who
provided assistance with trapping.
My thanks go to T. R. McCabe, C. G. Scalet, K. H.
Higgins, G. R. Larson, and R. M. Todd for editing the
manuscript. They provided many helpful ccmments and
insights.
I especially thank my grandmother, T. James, for
all the support and advice she has provided, for which
this degree would not have been possible.
Funding was provided by a grant from the Foundation
for North American Wild Sheep, Custer State Park, and
Federal Aid administered by South Dakota Department of
Game, Fish and Parks.
i i i
CONTENTS
ACKNOWLEDGEMENTS. .iii
CONTENTS . . . . . . . . . . . . • . • • . . . . . . . . . . . • . . . . . . . . . . . . . . . . . iv
LIST OF TABLES . ... . .v
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
STUDY AREA .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . • 6
METHODS AND MATERIALS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
RESULTS . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
DISCUSSION. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . ' .34
MANAGEMENT RECOMMENDATIONS. . . . . . . . . . . . . . . . . . . . . . . . . .
LITERATURE CITED ... . . . .. . . .
. 46
48 . . . . . . . . . . . . . . . . . . . . . . . . . .
APPENDIX .. . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
iv
LIST OF TABLES
Table 1. Number of vegetation sampling plots for each habitat type located within each sheep herd range in Custer State Park, SD, from August through September, 1985 ................... 19
Table 2. Discriminant analysis classification summary for each habitat type, for all sheep ranges included, in Custer State Park, SD, from August through September, 1985 ................... 19
Table 3. Proportion and area (ha) of each habitat type found within each bighorn sheep herd home range in Custer State Park, SD, 1985 ............. 25
Table 4. Selection and avoidance of habitat types by Rocky Mountain bighorn sheep herds in Custer State Park, SD, from July through October, 1985 ........................................ 25
Table 5. Selection and avoidance of habitat types by pooled ram and ewe bighorn sheep herds in Custer State Park, SD, from July through October, 1985 ........................................ 27
Table 6. Proportion (no. of observations) of behavior categories for bighorn ram and ewe sheep herds in Custer State Park, SD, from July through October, 1985 ............................... 27
Table 7. Relative frequency (no. observations) of resting activity by Rocky Mountain bighorn sheep herds in each habitat type in Custer State Park, SD, from July through October, 1985 ............. 28
Table 8. Relative frequency (no. observations) of feeding activity by Rocky Mountain bighorn sheep herds for each habitat type in Custer State Park, SD, from July through October, 1985 ... 30
Table 9. Mean distances (no. observations) [group size] between locations of foraging sheep and escape terrain for all bighorn sheep herds for each habitat type in Custer State Park, SD, from July through October, 1985 ....................... 30
v
Table 10. Selection and avoidance of habitat types used for feeding by Rocky Mountain bighorn sheep herds in Custer State Park, SD, from July through October, 1985 ...................... 31
Table 11. Selection and avoidance of habitat types used for resting by Rocky Mountain bighorn sheep herds in Custer State Park, SD, from July through October, 1985 ...................... 33
Table 12. Expected number of occurrences for each habitat type by Rocky Mountain bighorn sheep herds in Custer State Park, SD, from July through October, 1985 .............................. 39
Table Al. Occurrence of habitat occupation by the east end ewe herd in Custer State Park, SD, from July through October, 1985 ................ 52
Table A2. Occurrence of habitat occupation by the west end ewe herd in Custer State Park, SD, from July through October, 1985 •............... 52
Table A3. Occurrence of habitat occupation by the Grace Coolidge ewe herd in Custer State Park, SD, from July through October, 1985 ............ 53
Table A4. Occurrence of habitat occupation by the southeast ram herd in Custer State Park, SD, from July through October, 1985 ............ 53
Table AS. Occurrence of habitat occupation by the southwest ram herd in Custer State Park, SD, from July through October, 1985 ............ 54
Table A6. Occurrence of habitat occupation for pooled ewe herds in Custer State Park, SD, from July through October, 1985 ...................... 54
Table A7. Occurrence of habitat occupation for pooled ram herds in Custer State Park, SD, from July through October, 1985 ...................... 55
vi
LIST OF FIGURES
Figure 1. Map of ram and ewe home ranges from July through August in Custer State Park, South
Figure 2.
Figure 3.
Figure 4.
Figure 5.
Figure Al.
Figure A2.
Figure A3.
Figure A4.
Dakota, 1985 ................................ 7
Canonical representations of the first and second discriminant functions for habitat types in Custer State Park, SD, 1985. Canl uses vegetation variables grass, forb, duff, bare ground, and canopy coverage. Can2 uses vegetation variables forb, rock, canopy coverage, and duff ......................... 20
Mean(+ se) of ground cover, canopy cover, and horizontal density for all bighorn sheep ranges combined for each habitat type in Custer State Park, SD, 1985 ............................. 21
Mean(+ se) of tree basal area/ha for all bighorn sheep ranges combined for each habitat type in Custer State Park, SD, 1985 ........ 22
Mean(+ se) of number of tree stems/ha for all bighorn sheep ranges combined for each habitat type in Custer state Park, SD, 1985 ........ 23
Mean(+ se) of ground cover, canopy cover, and horizontal density for all bighorn sheep ranges for mixed grass/forb habitat type in Custer State Park, SD, 1985 ............... 56
Mean(+ se) of ground cover, canopy cover, and horizontal density for all bighorn sheep ranges for riparian habitat type in Custer State Park, SD, 1985 ...................... 57
Mean(+ se) of ground cover, canopy cover, and horizontal density for all bighorn sheep ranges for Ponderosa pine/no understory habitat type in Custer State Park, SD, 1985 ...................................... 58
Mean(+ se) of ground cover, canopy cover, and horizontal density for all bighorn sheep ranges for Ponderosa pine/grass forb habitat type in Custer State Park, SD, 1985 ....... 59
vii
Figure AS.
Figure A6.
Figure A7.
Figure A8.
Figure A9.
Figure AlO.
Figure All.
Figure A12.
Figure A13.
Figure Al4.
Mean (+ se) of ground cover, canopy cover, and horizontal density for all bighorn sheep ranges for steep rocky/ponderosa pine habitat type in Custer State Park, SD, 198 5 . . .. . . . 60
Mean(+ se) of ground cover, canopy cover, and horizontal density for all bighorn sheep ranges for steep rocky/grass forb habitat type in Custer State Park, SD, 198 5 . .. .. .. . . . . . 61
Mean (+ se) of tree basal area/ha for all bighorn sheep ranges for riparian habitat type in Custer State Park, SD, 198 5 . . . . . . . . . . . . 62
Mean (+ se) of tree basal area/ha for all bighorn sheep ranges for Ponderosa pine/no understory habitat type in Custer State Park, SD, 19 8 5 ...................... , •••..••••..• 6 3
Mean (+ se) of tree basal area/ha for all bighorn sheep ranges for Ponderosa pine/grass forb habitat type in Custer State Park, SD, 1985 ...................................... 64
Mean (+ se) of tree basal area/ha for all bighorn sheep ranges for steep rocky/ponderosa pine habitat type in Custer State Park, SD, 198 5 . . . . .. . . . . . . . . . . . . . . . 6 5
Mean (+ se) of nt.:.mber of tree stems/ha for all bighorn sheep ranges for riparian habitat type in Custer State Park, SD, 198 5 .. . . . . 66
Mean (+ se) of number of tree stems/ha for all bighorn sheep ranges for Ponderosa pine/no understory habitat type in Custer State Park, SD, 198 5 . . . . . . . . . . . . . . . . . . . . . 67
Mean (+ se) of number of tree stems/ha for all bighorn sheep ranges for Ponderosa pine/grass forb habitat type in Custer State Park, SD, 1985 . . .. . . . . . . . . .. . . . .. . . . . .. . . 68
Mean (+ se) of number of tree stems/ha for all bighorn sheep ranges for steep rocky/ponderosa pine habitat type in Custer State Park, SD, 198 5 . . . . . . . . . . . . . . . . . . . . . 69
vi i i
ABS'IRACT: One hypothesis which explains segregation between Rocky Mountain bighorn ram and ewe herds was examined using the Custer State Park bighorn sheep population. Three predictions of this hypothesis, similarity of habitat types, similarity of habitat utilization, and high home range fidelity, were assessed using physical habitat measurements and locations of radio-tagged individuals of both sexes. Habitat types were similar, as classified �ith discriminant analysis, with respect to each type, across all of �he sheep ranges. Members of each sheep herd maintained high fideli�y to their respective home range. Habitat utilization differences were inconsistent among the herds, regardless of sex, except for selection against doghair ponderosa pine habitat type by all herds. Utilization of feeding and resting areas were also selected or avoided inconsisten�ly, except for avoidance of doghair ponderosa pine for both resting and feeding, and also ponderosa pine/no understory for feeding. Mixed grass/forb, ponderosa pine/grass forb, and riparian habitat types were used most frequently for feeding, and steep rocky/ponderoasa pine most frequently for resting. Rams foraged at significantly greater distances {F = 4.99, p = 0. 0009) from escape terrain than ewes although group size of rams was always small (1-5 individuals) and ewe group sizes ranged from small to large (up to 29 members) . These results supported 2 of the 3 hypothesis predictions; similarity of habitat types and fidelity of home range. However, the most important prediction, similarity of habitat utilization between the sexes, was not supported and consequently the hypothesis of mi�imizing habitat competition between the sexes was rejected as not providing a complete explanation for sexual segregation. An alt:rnative hypothesis, where rams and ewes segregate in order to reduce agonistic interactions between them, is presented. Future directions to more completely test the competing �ypotheses which explain sexual segregation in bighorn sheep are suggested.
ix
INTRODUCTION
Spatial segregation between Rocky Mountain bighorn
sheep (Ovis canadensis canadensis) ram and ewe herds has
been documented by Geist (197 1), Geist and Petocz (1977),
Shank (1979), Morgantini and Hudson (1981), and Hogg
(1983), where ram herds consist of mature males
approximately 4 years old and older and ewe herds are
comprised of all other individuals. Geist ( 197 1), Shank
(1979), and Hogg (1983) have shown that segregation
occurs year-round except during the breeding season
(November and December) when the sexes congregate on a
traditional breeding range (cf. Geist 1971, p 209).
These studies also have shown that the sexes remain
separated during the winter, when segregation should be
least likely. During winter, resources, particularly
forage, are most limiting and when the sexes would be
expected to remain congregated after the breeding season,
such as at available foraging sites. However, both Geist
and Petocz (1977) and Morgantini and Hudson (1981)
reported that the sexes remained segregated on a
continuous winter range, even during severe winters.
Four hypotheses have been developed to explain
sexual segregation of bighorn sheep. Shank (1979)
...
2
believed sexual segregation was a result of differential
habitat requirements due to sexual dimorphism.
Segregation has been suggested as an anti-predator
strategy for males which are physically weakened after
the breeding season (Geist and Bromely 1978). Morgantini
and Hudson (198 1) suggested that sexual segregation
reduces frequency of agonistic interactions among rams
during the post-breeding season and maximizes fitness of
rams by conserving energy when reproduction is not
possible. Geist and Petocz (1977) explained segregation
as a mechanism for rams to minimize habitat competition
with pregnant ewes and their prospective lambs, thus
increasing survival of lambs, and thereby maximizing
reproductive fitness of rams.
Morgantini and Hudson {198 1) concluded that
reduction of agonistic behavior among rams after the
breeding season provided the best explanation for sexual
segregation during winter, and each of the other
hypotheses could be considered additional benefits
accrued as a result of this behavior. However, this
hypothesis is inadequate to explain why segregation is
maintained during the rest of the year when energy intake
is maximum and behaviors among rams would be least
detrimental. The hypothesis of increased predation risk
proposed by Geist and Bromely (1978) is also inadequate
to explain maintenance of segregation between the sexes
during the sWTU11er and fall when body condition of the
sexes should be in best condition. The hypothesis by
Shank (1979) adequately explained why segregation may be
maintained throughout the year, but because differential
habitat requirements were not demonstrated, he concluded
that the sexes did not partition their range in a manner
that best satisfied sex-specific requirements . Only the
hypothesis of Geist and Petocz {1977) of minimizing
habitat competition between the rams and ewes best
explains why segregation should be maintained throughout
the year, except during the breeding season .
3
Geist and Petocz (1977) emphasized reduced
competition for forage resources, however habitat
provides other resources, some of which may be competed
for (Anderson and Shugart 1974). Habitat is defined
primarily as a place where an organism lives and
secondarily as how that place is characterized
(Wittenberg 1981, Ricklefs 1973, Brown and Gipson 1983).
Bighorn sheep habitat may be divided into a physical
component and a forage component, with the forage
component described by the composition of forage species
(eg. grasses and forbs), and the relative importance each
contributes to the diet, and the physical component,
which consists of both biotic and abiotic descriptors and
their spatial distributions; in essence, the visual
appearance of the habitat. It is unknown whether
intersexual competition for the structual component or
the forage component is more important in bighorn sheep.
However, Shannon et al. (1975), Morgantini and Hudson
(1981), Geist (1971), and Geist and Petocz (1977) showed
that habitat structure was important in explaining the
distribution of ram and ewe herds, and Geist {1974)
considered the 3-dimensional structure of habitat to be
an important factor in ungulate social evolution.
The objective of this study was to examine the
predictions of the hypothesis that segregation reduces
habitat competition between ram and ewe herds in bighorn
sheep during the summer and fall. The predictions were,
if rams and ewes segregate in order to reduce habitat
competition between them, then they should: 1) occupy
home ranges which share similar physical characteristics
for each habitat type, 2) utilize those types with
similar frequencies, and 3) maintain fidelity of home
range occupation both spatially and temporally. In this
study, habitat characterization focused on the physical
component.
4
The bighorn sheep population found in Custer State
Park (CSP) is particularly well suited to examining this
hypothesis. The present population is derived from 22
5
individuals (15 ewes, 7 rams) transplanted from Wyoming
in 1964 (W. Jackson unpubl. rep. 1981) . Between 1964 and
1985, the herd increased to a level of at least 90
animals and separated into 6 distinct herds. Because the
currently occupied areas were vacant of bighorn sheep at
the time of the transplant, an assumption was made that
these separate herds have established their home ranges
to best satisfy each herd's habitat choice. Given that
the sheep had free choice to establish a home range
anywhere within the park boundaries, their observed home
ranges should reflect what they considered as the optimal
areas to define a home range. Additionally, the ranges
of all herds are assumed to be qualitatively equivalent,
where at least the minimum amount of nutrition is
available to all members of this bighorn sheep
population.
6
STUDY AREA
Research was conducted in CSP, which is located in
the southeast portion of the Black Hills, South Dakota.
Bighorn sheep ewes primarily inhabit French Creek Canyon,
located in the central portion of the Park, and rams
occupy ranges peripheral to those of the ewes (Fig. 1) .
French Creek Canyon is approximately 19 km long and
ranges from 70 to 140 m in depth. It is characterized by
steep, rugged walls with adjacent rolling meadows and
ponderosa pine (Pinus ponderosa) forests. The Park
headquarters region, located in the center of the Park,
contains a few meadows and steep, forested hills with
rock outcrops. The southwestern portion of the Park is
characterized by hilly, forested terrain with some
extensive rock outcrops adjacent to a few relatively
large meadows. The southeastern portion of the Park
south of and adjacent to French Creek Canyon is
characterized by less steeply rolling, but forested,
terrain with few rock outcrops and few meadow areas.
Physiographically, the headquarters region and southwest
area appear similar, while the southeast area is the
least rugged of any of the areas used by CSP bighorn
sheep. Elevations within CSP bighorn sheep ranges vary
from 1160 to 1707 m, rising steadily from the southeast
to the north and west.
•
.--..,...;. c::::
f i��-e Loca�ic� of Fre�cn Cre��. Ca�ycn ano �am ano ewe grc�:: ���! rancei iG Cus:er Stace Far�. SoJtn oa�oca. 1985. E-..e gr:::o.:c!: _{ZZj Grace Ccc l i o·;:i=. � E.:1sc Enc a:-.c Gra:::e Ccci ,·:::;;:::!. �we�-:: [r,c. F.a,., g:-o·J::'3: illScuc:-,eia�:.� Sc..w:�, .. ec:. �cc::le: 0--1 m;iec.
7
8
l\'!ETHODS .?I.ND MATERIALS
The 3 ewe herds were designated east end (EE),
west end (WE) , and Grace Coolidge (GC) herds (Brundige
198 5). Divisions for the 3 ram herds were northeast
(NE), southeast (SE) , and southwest (SW) . The
sumrr.er/fall ranges of each herd are indicated in Figure
1. Only the GC range extensively overlapped that of the
EE herd on French Creek Canyon and members of both ewe
herds intermingled in this overlap area (Brundige 1985) .
Otherwise, there was almost no range overlap among the
herds and individuals remained associated with their
respective herd and range. Accordingly, sampling schemes
for vegetation and movements were divided with respect to
each sheep herd.
Division of Habitats
Habitats for the sheep herds had previously been
divided into 7 types based on visual appearance (Brundige
198 5) . These divisions were based on presence or absence
of trees, presence or absence of understory, degree of
slope, and degree of rockiness. Mixed grass/forb habitat
type was designated for meadows and consisted of mostly
grasses with some forbs, usually no rocks present, and no
trees present. Riparian habitats were those
characterized by vegetation found near streams and could
include grasses, forbs, rocks, and a mixture of several
9
deciduous tree species and/or ponderosa pine. A
permanent water source was necessary to maintain riparian
areas. Ponderosa pine/grass forb was similar to mixed
grass/forb except that ponderosa pine, or, rarely, bur
oak (Quercus macrocarpa), was present. This habitat was
typical of a coniferous parkland. The ponderosa pine/no
understory category consisted of a ponderosa pine forest
with only an occasional forb or grass present, few rocks,
and was usually located on slopes of less than 40%.
D9ghair/ponderosa pine habitats were similar to ponderosa
pine/no understory, except they were much more dense and
where found contained trees of a uniform height. The
characteristics of this habitat could best be described
as a stand of ponderosa pine whose stems average less
than 15 cm in diameter at breast height (DBH) , grow 1. 5 m
or less apart, and average approximately 6 m in height.
Rocks could be present but no understory was present.
Steep rocky/grass forb areas were typical of open slopes
usually steeper than 35%, with rocks, but no trees
present. Steep rocky/ponderosa pine was similar to steep
rocky/grass forb except ponderosa pine was present. A
description of plant species composition for each habitat
type is given by Brundige (1985) for the EE and GC ranges
for 1984.
10
Characterization of Habitat Types
Physical characteristics for each of the above
habitat types were sampled in all of the sheep ranges.
For each range, 3 or 5 sampling plot centerpoints were
chosen randomly for each habitat type. Within each sheep
range, centerpoints were chosen by assigning a number to
every sheep location in each type obtained during July,
198 5. Numbered locations were randomly chosen to
determine to define a sampling plot centerpoint. In
cases where fewer than 3 sheep locations were available
for a particular habitat type at time of sampling,
centerpoints were chosen within types without locations
near types which had sheep locations. The number of
centerpoints (3 or 5) per habitat was subjectively
determined by the relative abundance of that habitat for
a given range. For types which had relatively small
total areas, 3 points were selected; otherwise 5 were
used. Three sampling plot centerpoints were used for all
of the habitat types in the northern portion of the GC
range located in the Park Headquarters region of CSP
(Fig. 1). Plot centerpoints were positioned within the
respective habitat type to avoid the inclusion of areas
that appeared to be transition zones. Thus, each
centerpoint was assumed to be located in a homogeneous
sample of the habitat type selected.
11
At each sampling centerpoint, a 10 x 10 m plot was
delineated with the edges lying in the 4 cardinal compass
directions. Within this plot, tree density and basal
area, ground cover, overhead canopy, horizontal
obstruction, slope and aspect were measured. Elevations
were estimated to the nearest 10 m from u. S. Geological
Survey (USGS) 7.5 min. quadrat maps.
Mean tree density was estimated by counting the
number of stems for each species in each habitat type and
converting this number to stems/ha. Similarly, mean tree
basal area was estimated by measuring DBH to the nearest
0. 5 cm, converting DBH to basal area, and expressing as
cm 2/ha.
overhead canopy was estimated using a convex
spherical densiometer (Forestry Suppliers, Inc. Jackson,
MS) placed in the center of the plot and using 4
readings. The densiometer was kept level by attaching it
to a camera tripod.
Mean horizontal obstruction of vision was
estimated by using a density checkerboard, 1.5 x 1. 5 m
and marked in alternating black-and-white, 15 cm squares
giving a total of 100 squares. The number of squares
covered were counted and the 4 readings were averaged to
give an estimate of horizontal obstruction for the plot.
A square was considered covered if at least 1/2 of it was
obstructed from vision. The board was read from a
kneeling position to approximate the height of eyesight
for an average bighorn sheep (Risenhoover and Bailey
1985) . The board was placed at a distance of 15 m and
read from the center of the plot in each of the 4
cardinal directions.
12
Ground cover was estimated using line transects
and a 20 x 50 cm guadrat (Daubenmire 1959) . Within each
sampling plot, 4 - 5 m transects were randomly placed in
one of 8 compass directions (N, NE, E, SE, S, SW, W, NW) .
No transect was allowed to overlap any other transect.
Five guadrats were placed parallel and next to each
transect line at 1 m intervals. Placement of the guadrat
with respect to the side of the transect line was
determined by a coin toss . A total of 20 guadrats for
each sampling plot was used.
Ground cover was divided into 7 categories:
grass, forb, shrub, log, duff, rock, and bare ground.
The grass category was comprised of grasses and grass
like forms including Carex spp. Forbs were defined as
herbaceous broadleaved species. Shrubs included plants
with more than 1 woody stem and ponderosa pine seedlings
less than 20 cm tall. Dead and downed trees or branches
greater than 3 cm in diameter were classified as logs.
Duff included dead plant parts and any dead woody
material less than 3 cm in diameter. This category
consisted mostly of shed ponderosa pine leaves. Rocks
were considered as such if they were 3 cm or greater in
their longest axis. The bare ground category included
mostly bare soil and rocks less than 3 cm. Each of
these categories were measured by percent cover within
the quadrat.
Location Sampling
13
Individually radio-marked sheep were located from
July through October 198 5. Five ewes from the GC herd, 4
from the EE herd, and 8 from the WE herd were radio
collared. One ram each from the SE, SW, and NE ram herds
had a solar transmitter attached to its ear. Daylength
was divided into 2 periods: early (0500 hrs to 1300
hrs) , and late (1300 to 2100 hrs) . For ewes, 4 locations
per daylength period per herd were determined using
randomly selected, radio-marked sheep. A radio-tagged
ram was located up to 4 times in each daylength period
that it was sampled. The interval between successive
locations for all herds was a minimum of 1 hour.
Habitat type and map location, as designated by
Universal Transverse Mercator (UTM) longitude and
latitude on USGS 7. 5 min. quadrat maps, were recorded for
each location of a radio-marked sheep. Also recorded
were percent slope, aspect, elevation, group size, and
14
age and sex composition of any associated sheep.
Activity categories, including resting, feeding, moving,
nursing, and standing, were observed in order to
calculate activity budgets and determine whether there
was any differential habitat use among the sheep herds
for each activity considered.
seasonal Home Range Estimation
Seasonal home ranges for each herd were estimated
from locations of radio-marked sheep in each herd. The
minimum polygon method was used to statistically estimate
a home range area and boundaries. Locations that caused
the estimated area to include portions of land where
sheep were never seen, or which lacked any evidence of
their presence, were not used. such locations were
assumed to be excursions of individuals from the
respective sheep herd (Geist 1971) and therefore they
were not included. After home range boundaries were
delineated onto USGS 7.5 min. quadrat maps using the
minimum polygon method, topographic features were
subjectively included to define the home range boundaries
and inclusive habitat types used in utilization
estimation. Using this method, the home range usually
included an entire hillslope to the top of a ridge rather
than only partial hillslopes or valleys. The
subjectively delineated home range encompassed the entire
15
area estimated by the minimum polygon method and is
probably a more realistic representation than the minimum
polygon method itself (MacDonald et al. 1979).
Estimating Proportion of Use
The proportion of each habitat type found in each
sheep herd home range was measured from USGS 7. 5 min.
quadrat maps. Habitat types were outlined from aerial
photographs onto clear acetate sheets, and then traced
onto the quadrat maps using a Map-0-Graph (Art-0-Graph,
Minneapolis, MN) to correct for scale and photograph
distortion. These areas were measured in cm2 using an
area meter (Model LI-3000 Area Meter, Lambda Instrwnents
Corporation) and converted to hectares. Precision of the
meter was +0.02 cm2. Proportion of each habitat area was
calculated by dividing the total area of the home range
into the area covered by a habitat type. The mean of the
slopes for each type was used as a correction factor for
proportion of actual area covered by each habitat, since
slope means for each type across all ranges were found to
be not statistically different (F = 1. 54, p = 0.0756, df
= 22).
Selection or avoidance was estimated using the
proportion of area for each habitat type within the home
range, and the proportion of locations, according to the
methods of Neu et al. (1974) and Byers et al. (1984).
16
Only those locations used to estimate the home range area
were used in estimating proportion of use.
Geist (1971) noted that resting and feeding areas
may be particularly important in determining suitability
of ranges for sheep. Thus, frequencies for each of these
activities were compared among the 7 different habitats
of each sheep herd to determine any differences in
utilization of these habitats. Risenhoover and Bailey
(1985) have also discussed the importance of feeding
areas with respect to group size and distance from escape
cover, where escape cover is defined as precipitous,
rocky terrain (Geist 1971). Accordingly, distance from
escape cover at first sighting for a location were
visually estimated and compared among sheep herds for
each type used for feeding or resting.
Statistical Analyses
Habitat types were analyzed using ground cover and
tree plot measurements and compared using discriminant
analysis. Means for each variable from each sampling
plot were used as observations for this analysis. The
only variable excluded from these analyses was shrub
cover, since coverage was less than 5% in all habitats
sampled. Mean area of each habitat type and mean
distances to escape cover were compared using analysis of
variance (ANOVA, Sokal and Rohlf 1981). Frequency
17
comparisons of numbers of sheep groups observed among
habitat types were made using chi-squared test of
independence (Sokal and Rohlf 1981). Discriminant
analysis and 'ANOVA were performed using appropriate
programs contained in Statistical Analysis Systems
(Goodnight 1986) software. Significance levels for 'ANOVA
and chi-squared tests were 0.05, and 0.10 for
discriminant analysis.
For the methods of Neu et al. (1974) and Byers et
al. (1984) for estimating habitat utilization, expected
proportion of usage {Pio> was calculated by dividing
total range size into each respective habitat area.
Expected frequencies of utilization were then found by
multiplying Pio times the sum of the number of
observations. Significance level to determine
utilization greater than, less than, or in proportion to
habitat type availability was 0.05 .
18
RESULTS
Only 5 locations were obtained for the NE ram herd
during the study period. Therefore, data for this herd
were excluded from habitat and utilization analyses.
Division of Habitat Types
A total of 120 vegetation plots were measured,
which included all habitat types found in each sheep
range (Table 1). Each type was classified as distinct
from any other type within each sheep herd range (Table
2). Also, a type found in one range was not classified
differently from that same type which occurred in any of
the other ranges. These results indicated that observed
differences among the habitat types both between and
within the sheep ranges were also differentiated
quantitatively, according to the variables measured, for
not less than 93% of the observations for each type.
Amount of overlap and relative degree of heterogeneity
for each habitat type using canonical representations of
the first and second discriminant functions are shown in
Figure 2. Habitat types for all ranges combined are
characterized in Figures 3 through 5. Habitat type
descriptions for each sheep herd range are shown in
Appendix I.
Table l. NUJ!'.ber of vegetacion sampl�ng plots for each habitat type located within each sheep herd range in �uster State Park, SD, from AJgust through September, i985.
Sheep Herd east end west end Gra�e southeast southwest
ewe e..,c Ccc lidge ram ram H;;.bi:at '!'ypc herd !".erd e,.,·e he:-d herd herd Total
rr:ixe'.! :,rass/ forb 5 5 5 5 23
pcnderosa pine/ no underscory 5 5 5 5 23
ponder::isa pine/ grass forb 5 5 5 5 23
riparian 5 5 0 16
steep rocky.' grass forb 6 2 0 14
steep rocky/ pcnderosa pine 5 5 5 21 -------------------------------------------·--------------------------'i"ctal 31 27 :s 21 26 120
Table 2. Discri�inant analysis classification summary for each habitat type for all sheep ranges included in Custer State Park, SD, from August through September, 1985. Habitat types are: MF - mixed grass/!orb, PB -ponderosa pine/no understory, PG - pcnde:osa pine/grass forb, RI -riparian, SG - steep rocky/grass fcrb, a�j SP - steep rocky/ponderosa pine.
Number of observations and (percents) classified into types
Pf: : .. F.: SG Sr Tctal
MF 22 0 0 0 l 0 23 ( 95. 65 l 0 0 0 ( 4. 3 5 l 0
PE 0 22 0 0 0 23 0 (95.65) I 4. 3 5 l 0 0 0
?G 0 :2 '.) 0 0 23 0 I 4. 3 :': i I 9:. 6:: i 0 c c
R1 l 0 c 15 0 0 16 ( E.: 5} 0 0 ( 93. 75 l 0 0
SG 0 0 0 0 14 0 H 0 0 0 0 (100.00) 0
SP 0 l 0 0 0 20 2 l 0 I 4. 76) 0 0 0 ( 95. 2q
19
?
7
J
2
0
-1
• l • I � I • I • I ..
I +
I • I • 1
• • .. 4
I • I • I • I •
.
4
" .. , ...
5
s
.. .. '\ 4
4 4
,\
� s
..
.J
J J ] J �J
J.:: � 3
" J :; ·.; 1
J
5 5
s 5
s s
s 5
'5
s
.J
J J
.1
.l J )
J
J
s
s
]
Ii 6
6
6 6
6
2 2
66
"' 6
2
2
2
2 2 :?
2 2
z
2
2 2
z
1 l 1
I I
I I I I I
I I l
l
Co
______ ...,__ _______ ·----------·------------- .. ---------------·---------·-----------...... ------- t, - • -;: o 2 • 6
CANI
Figure 2. Cannonical representations of the first and second discriminant functions for habitat types in Custer State Park, South Dakota, 1985. 1 mixed/grass forb, 2 = riparian, 3.= ponderosa pine/grass forb, 4 = ponderosa pine/no understory, 5 = steep rocky/ponderosa pine. 6 = steep rocky/grass forb. Canl uses vegetation variables grass, forb, duff, bare ground, and canopy coverage. Can2 uses vegetation variables forb, rock, canopy coverage, and duff.
N
0
w (,!)
ll! w >
0 .... w 0 ll! w n.
100
90
BO
70
60 1k �
. ��l 50
40
� @� :!O I il,
� �� '0 1 8� 1 1,1 ,,r
"! fJ 1 0 i� 't
Jt:L M::8 o . .r. i.,, _ 1r 1--· MF" ---� PB --j 1-- PG --I 1-- AI SG - ··-J 1-- SP
r i gur:e 3 . M , ,,1n (+ s!' ) o f 1round covr.1· , c<1nopy cove r , a nd hor: i zon ta l dt! n s i t: y f.01· 'I l l b f qti<l l n sher,p r:<'l ngei:: comb i nr,,J f o r r, ,lch habi tat type i n t:ustcr :;\',H<: P,1 1: k , S D , 1 '1 14 ', . MF" � m i xed /gr,1 ;s s f o i l> , PB = pondr rosa r i ne / n o unde r s t ur y , {''-� " pon<1<' r" r. ,, pine /rir&"'!I f or b , P T · 1 i p;n \ 'l n , SG : ,-• .:ep TO•: !: ·; l q r a s s forb, SI-> � "t�er rnr.l<y / pond,.TOl!III p i ne .
Li fl"re n c,rnopy l') Du f t ,::] !"o r b f.1 a , ,.,.. ., f'.'l Hor i r.ontal il"I Loq I'll! Ror.k ;"'! Sh rub l'i r:ro•rnd [j Co·,e.· rJ f::1 t·l � Dens i ty 1.J U l1
TYPE
< w n: <!'.
�.00000
,100000
....1 200000 < V1 < (l'l
,', \,
f- --- · PO •• • 1 PG
f i gure 4 . Mc,.r, ( .!_ !'!c l of t t·c:,: '"""'b l n .. ,, ,,...,. "·•1".h h"bl n, t '" Y! '" ponder.», ,, p i r,..: / no und-,rstory . �_; p :-:: s t ,....,n rnf"'\.'.y/ s,,:,nd'1rosa f' •
� .. - R J
b" "· ' l -'l r<: n/ha for a l l b i ghorn sheep 1·«wws , - . . . � , ,. ,. $ f" ,, t ,... Pd t· � . !;D, l '>A :1 . PB -
) ·•}f. ... 1•:t'O�d 1 · 1 1 1 1..: / (JCtl.S!.. for t , # , ... , � 1 l f�. , 1 i rt J t �
f:sJ Plnu,. p0ndero"a � Ou., rcus ,,,, , . ·roc .. r p .. � fl<' t u l i< p.:i p 1 {c rou ..
f� Pop•J l u s t t· <'nm lo! d.-,s � M l sc,, l l · !l""'• 1s sp.-,c ics .
TYPF
N N
. 1 I
'"( :c 'Vl
2000
1 �;00
� 1 000 I-I.fl
0 z
500
0
1-- PB - · · I
F'lgur.,. 5 . "'"·"' ( • S P. ) ot numb•? 1 ra nges comb i nr·l r;;r �ach h" b i t , , • ponderos.-,, p i n· · . . , . ) •inderstory , !'< : SP = "teP.p ro, , 1,: , / pondero>1A pl n, · .
j · •• · ll I -· l I · SP --- .. ..
· • f t t �, : str.ms /h" (or a l l h i ih.:i,·n ,,1,, ,,.. p • ,T"' 1 n Cu: : t,<' t SI .d i ,; l',H k. . �0 . I 4 ti ', , PB
t 1<H1d , : ro:t.:�, p i n� /q ctt : . :; t -.H h , i < 'l 1 i 1 · , 1 r 1.. t 1 r, .
� P l nus pond•· ro,;11 � Quercui, m.1c roc1<. rp11 12'.2) f\('tula P"Pi tt:rous
� Popu lus trcmu loldcs· � M l scc l l ,rneous .species .
TvPF
N w
24
Although each habitat type was found to be similar
in structure among all ranges , the proportions of area
for each of these habitats were not similar. There were
no consistent patterns among ewe ranges and ram ranges
except for riparian (Table 3). The amount of riparian
habitat was lower in ram ranges than ewe ranges (F =
3 8. 57, p < 0.0001, df = 12) . This difference may have
been because the EE, WE, and part of the GC ranges
included French Creek Canyon, the only extensive area of
riparian habitat in any of the sheep ranges.
Habitat Utilization
Proportions of use and corresponding estimates of
selection for or against each of the 7 habitat types, by
the methods of Byers et al. (1984), for each sheep herd
are contained in Appendix I. Both the WE and GC ewe
herds selected for mixed grass/forb , while the EE herd
used this habitat in proportion to its availability
(Table 4). All 3 ewe herds selected against doghair
ponderosa pine. The EE and WE herds also selected
against ponderosa pine/no understory. The WE herd was
the only herd that selected for riparian.
For the SW ram herd riparian, steep rocky/grass
forb, and doghair ponderosa pine types were selected
against while the other habitats were utilized in
proportion to their availability. The SE ram herd
:" b:.e J. Pr ·:>por-: ion and area { �.a ) c f e!c!'-. habi tat t ype found 1,,,; i thin each t ;\c : � 5tce? herd no�e : a�ge ! : =� J � ! y : �ro�g� October , 1 9 S 5, !� :�ster S a�e Park, Sou':.� �akc ta.
� : xeC g : l S S /
: : r ::
;:,cn:ierosa ;,ir:e I no ur.de � s cory
�c=-i:!.e!"csa ;,ine, grass fcrb
r � paria.n
st.;;;e;: =ockv1 gr�ss :: or:O ·
s: eep rccky/ i:-=�.de: r osa pir.e
cog,.a : r pc:-:dercsa pine
Tc�al ho�e ra�;e s:.z.e ( ha )
e ast end e...,e !"le:·d
o . : r I " :; . l : I
o . : a 9 ( , 39. 7 9 l
:l. 2 � j < : i. 7. :i S l
0.0 .; 9 ( 1 8 . 7 3 )
0 . 0 2 2 I 1 0 . : C l
0 . � � :. 1 1 ;. s . s e I
0.0 2 6 ( 1 2. 6 8 l
4 9 3 . 95
�est end Grace southeast <!·.·e Coolidge r am � e � d eve herd her�
� - � � s 0 . 0 2 €, 0.0 5 6 1 : 7 . � 6 ) ( 1 S. 3 4 l I :: S. 6 3 l
0 . j 5 9 0. 2 5 4 0 . �96 ( .; j6. · :; ) ( 16 1 . 98 ) ( 3 1 6.5 2 )
C . L J j 0.1 66 0.2 06 ( :. 1 3 . 4 9 ) ( 1 19 . 42) ( : 3 1 . 0 7 )
C.0 3 � 0.0 2 5 0.009 I :: ': . 32 l ( 17. 9 0 l ( :- . 9 7 l
o . o : o 0.009 0.0 : , i c l.S6 J ( 6. 2 0 l ( 3 �. 5 2 J
0 . 3 :9 0 . j 6 9 0.101 ( 3 52.� 7 ) ( 264.0 9 1 ( 6 � . ) .; )
) . l � 2 0.1 : 3 �·. 0: :, ( 1 56., 0 J ( 109. 7 :, l I 4 7. 7 3 J
l l 0 4.05 7 1 7.62 6 3 5.7 8
sout•·.,,;est ram �erj
0.0 5 5 ' :; l . 3 � )
0 . 1 7 9 ( 1 9 6.5 7 )
0 . ! C � ( 3 3 8.97 )
o . 0)6 i 2.: :i i
O . C c l ( 2 3 . 1 5 )
J. 3 7 3 '. 4 1 2 . 6 a ;
-:, . -:--: c ( 6 5 . 9 2 )
l l 0 7.2S
Tal: lc 4. Se l;;;ct. ic:i i:::r a::d aga i:lst l �.a!:itat types by :lock;· �ou,.t a : n bighorn s�eep �erds in Cus ter S t a te Park, SD , f rom July through Octo�e r, 195:,.
Sheet:, Herd east e�d west e:id Grace sou t heast southwest
�we e",e Cc,olidge ra1r, ram herd �erd e�e herd herd herd - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
;,i: xed g r ass/ : orb
;,cnde:�sa ;,ir.e / no u:iders tory
ponderosa pine/ grass !crt:
stee;: roe��·,: .:;r:1s.s f e r�
s : cei:; rocky/ pci.'de rcsa pi r,e
dc�::ai r p-:-r.de rosa ;,ine
0 +
0 0
0
0
... 0 0
0 0
0 0
0
0 0
0
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -O i d ca:es :i o select�cn
_ ::i ca:es se :e::t �c:i - � - ? :·. a � i : a t :ype 1 j cates se:ec:icr. a � = : �s: a �aritac type
..
25
likewise selected agains t doghair ponderosa pine and
riparian but, in addition, selected against ponderosa
pine/no unders tory, while selecting for ponderosa
pine/grass forb habitat.
26
Pooling the 3 ewe herds and both ram herds
resulted in a difference of habitat utilization than
those found among individual sheep herds (Table 5). For
the pooled ewe herds, both the mixed grass/forb and
riparian habitats were selected for while doghair
ponderosa pine and ponderosa pine/no unders tory were
selected agains t. Doghair ponderosa pine and riparian
were selected agains t by the pooled ram herds, while
ponderosa pine/grass forb was selected for.
Activity Budgets
Feeding was the mos t frequent activity encountered
for all herds except the SW ram herd (X2 = l S . 88, p <
0. 01 , df = 3, Table 6) . Feeding was the second mos t
frequent activity for the SW herd and was probably an
artifact of small sample size. Standing was not a
frequent activity for any herd. The remainder of the
activities observed were divided between res ting and
moving and were similar among the herds (X 2 = 3.38, p =
0.4981, df = 4).
Res ting areas for all herds (Table 7) were found
most frequently in steep rocky/ponderosa pine habitat,
Table 5 . Selecticn a�d avoidance o: l':abitat t'.{Pes by pooled ram and ewe bi;horn sheep herds i n Custer State Park , SO, from July tr.rough October , 1 98 5 .
Habi tat !'ype pool ed
e·..ie herds pool ed
ram herd -- - - - - - - - - -- -- - - - - - - - - - - -- - - - - - - - - - - - - - ---- - - - - - �- - - - -
pcr.derosa pine; r.c unde rs tory
ponde rosa pine/ grass f:;;rb
r ipar ian
s : eep r::ck.1!
g: ""ss f orl::
s t eep rock:rt pcr.dercsa ;:, :.ne
ponderosa pine
0
0
0
0 0
- · · - - - - -- - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - -- - - - - - - - -- - - - - - - - - - - - - - - -
O ird�cates no se lection : nd i cates selection fer a habi ta� :,�e indicates se .ection aga inst a hazi:at type
Table 6 . Proport ion t ro . cf observatio�s ) o f behavior categories for bighorn raT. and ewe sheep he rds in Custer State Par k , SD, f rom July through October , 1 985 .
Activi ty
Re st:.ng
Moving
Standing
Feed ing
':o � a l �tiJ';.'::er
Obse rvat ior:s
east enc e'-'e herd
� . 2 4 '7
( 1 8 )
0 . l � 1 ( : 1 1
;J . 06 8 I 5 l
0 . 5 . H ( ) 9 l
i l
Sheep Herd west end Grace southea st
e1<,e Coo lidge n,1n herd e�e herd herd
0 . loO 0 . 1 � 1 0 . 206 ( 1 6 ) ( B ) ( 7 }
0 . 1 60 o . 208 0 . 36 2 l i c l I 1 1 l ( 8 )
0 , :;57 0 . 1 5 1 0 . 0 2 9
( 6 ) ( 9 l ( 1 1
O * S i 3 {) . � 9 : 0 . 5 2 S ( :, : ) ( 2 6 l ( :. s )
89 5 ) 3 �
a - - represents no cbserved occur re�ces .
southwe s t ra..i:"l herd
0 . 500
( 7 )
o . 2 1 4
( ) )
_ _ a
0 , ? 1!6 H I
1 4
2 7
';al:le 7 . Re l a t i ve f ::eq-.:e r.cy ( no . obse:-va t i ons ) of res ting act i vity I::: Rc;:ky �ount a i n bighorn s r.ee;: he : ds fer each r.abi tat type i n Cus ter S t a � e Pa r k , S D , from July th::ough OCtobe i; , 1 9 8 5 .
H'lbi : a : Ty;:,e
r.,ixed g: a s s / forb
ponderosa pine / r.o -...nde ?:- s-::cry
ponderosa pine / grass fo:-b
r _par ia�
s teep ::ocky/ ; , ass forb
s teep rocky/ ;: � r.t;?rosa pine
r:!cg� 5 ,;. �
;:..o::. :: : :- : 3-:! p : :-.�
east e :-.d e;,;e � e : d
0 . 00 ( 0 )
0 . l l ( : )
0 . 2 9 ( 5 )
o . oc ( l
0 . 1 1 \ - )
0 . 50 ( 9 )
,· . c ,: ( ( i
�es : e:-.d G:ace e�e Cooli d;e her� c:..tc r. t: : :i
O . Oo .l . 1 3 ( 1 ) ( l )
•J . 06 0 . 25 ( l ) 1 .: )
O . Oo 0 . 25 ( 1 ) ( 2 )
0 . 06 o . co
( l ) ( 0 1
() . 00 0 . 00 ( c ) ( 0 )
0 . 7 :, 0 . J 8 ( l : ) \ ) )
0 . O? c . so ( ( ) { 0 1
s::�t�e a s t ra..T. l:'!?rd
o . oo ( 0 )
'.) . 1 4 l � )
0 . 57 ( 4 )
C· . ".10 ( O J
o . oo t O l
0 . 2 9 ( 2 '
c '· ( )
sout�.west ram herd
0 . 00 ( 0 )
0 . 2 9 ( 2 )
0 . 1 .; ( l l
o . co ( 0 )
o . oo ( 0 )
0 . 57 ( 4 1
0 . 1) 0
( 0 ) -- - · - - - �------ - - ----- - - - - - - - - - - - - -- - - - - - - - - - - - - -- - - _ _ _ _ ... _ _ _ _ _ _ _ _ _ _
28
29
except for the SE ram herd , which used ponderosa
pine/grass forb ( x2 = 13 . 0 4 , p < 0. 05, df = 6) .
Utilization analyses revealed that only the WE ewe herd
selected for steep rocky/ponderosa pine as resting sites
( Table 8). Only doghair ponderosa pine was consistently
selected against by all herds.
Feeding areas were less consistent among herds
( Table 9) , and included riparian , ponderosa pine/grass
forb, and mixed grass/forb habitats as the most
frequently used feeding habitats . The EE ewe herd spent
most of its feeding time in ponderosa pine/grass forb but
also utilized steep rocky/ponderosa pine and mixed
grass/forb to a lesser extent . The WE ewe herd fed
mostly in riparian and mixed grass/forb and, to a much
lesser extent, in ponderosa pine/grass forb. The GC ewe
herd fed most frequently in mixed grass/forb and also
ponderosa pine/grass forb and riparian habitats. Only
ponderosa pine/grass forb was used extensively for
feeding by the SE ram herd while the SW ram herd used
mixed grass/forb more frequently and, to a lesser extent,
both ponderosa pine/grass forb and riparian, although
again, this may only be an artifact due to small sample
size . Selection of feeding areas, determined using the
methods of Byers et al. ( 1984 ) , generally followed trends
found in frequency comparisons ( Table 10 ) . All herds
.....
Tab le 8 . Select ion and avoidance l o f habitat types used for resti ng s i tes by Rocky Mountain bi ghorn sheep herds in Custer State Park , SD, f rom July thrcugh october , 1 98 5 .
mixet grass , forb
ponderosa p ine/ no •-r.dersto:-y
ponce res a pine/ grass forb
r i p;; d a n
steep rocky/ grass !'c:-b
stee? rocky/ p::cr.c::;;,::::,sa pir.e
doghai r por.::erosa pine
Sheep He rd east end ._,est end Grace southea s t south...,es t
e1,1e e·.;e Cool idge r am r am herd herd e...,e herd herd herd
0 0
0 0
0 0 0
0
0
0 0
0
0
0
0
0
0
0
0
--------------------------- ------------------------------ -----------l 0 indica:es no selection
+ 1ndi�ate s se lection for a habitat t1-pe i n� ica:es select ion against a hab:ta: tj-pe
Table 9 . Relative f requency I nc. observations ) o f f eeding activity by Rocky Mountai n bighorn sheep herds for e ach habit a t type i n Custer State Park , SD , f rom July through Octobe r , 1 9 8 5 .
Habitat Type
Sheep Herd east e nd west end Grace southeas t southwes t
ewe ewe Coolidge r am ram herd herd ewe herd herd herd ---- -------------·------------------------·---------------------------
ro : xe:! grass/ 0 . 26 o . n o . j ;> O . l i 0 . 50 f orb ( 1 0 ) 1 1 7 ) ( 9 ) 1 3 ) I 2 l
ponderosa p i ne / 0 . 03 0 . 00 0 . 0 4 o . oo 0 . 00 r.o under story ( l ) 1 0 ) ( l l 1 0 1 ( 0 )
po:1derosa pine/ o . 4 1 0 . 1 6 o . ) 1 0 . 61 0 . 2 5 g rass forb ( 1 6 ) I 8 ) 1 8 I 1 1 1 l 1 1 )
; :.pa:- : .! n 0 . 0 5 ;; . H 0 . 2 3 o . oo o . �s ( ,: ) ( .: 2 1 ( 6 1 ( 0 ) I l )
s �eep rock,· / 0 . 0 5 :l . C ? 0 . 00 0 . 1 1 0 . 00 grass forb ( 2 ) ( : l ( 0 l 1 2 l ( 0 )
s teep rocky/ 0 . 2 1 0 . 06 0 . 08 a . 1 1 0 . 00 ponderosa p i ne ! a l l 3 I ( 2 ) ( 2 ) ( 0 )
doghair 0 . 00 0 . 00 o . oo o . oo 0 . 00 pcndercsa pir.e ( 0 ) ( C l l 0 ) ( 0 1 1 0 ) --------------------------------- --- ---- ------------------------------
3 0
Table 1 0 . Se l ect ion a nd avoidance l of habitat types used for feeding by Rocky Mountain bi ghorn sheep herds in Custer State Par k , SO , f rom July through OCtober , 1 9 8 5 .
Sheep Herd ea s t end wes t end Grace s outhea st southwes t
ewe ewe Coolidge r am ram Hat : : � t Ty-pe herd herd ewe he rd herd herd
m:. xed grass / forb
ponder osa pine/ no understory
ponderosa pine / grass forb
r iparian
steep rock1/ grass forb
steep rocky/ ponderosa pine
dog�ai r p<::r:de�osa pine
0
0
0
0
0
l O �d cates no selection
+ +
0 0
0
0
nd ca�es se lect ion for a habitat tYJ:e nd cates se lection agai n s t a hab i t at type
0
+
0
0
0
0
0
3 1
32
selected against ponderosa pine/no understory and doghair
ponderosa pine types as feeding habitats .
Distances from resting sites to escape terrain
were mostly consistent among the herds , since the most
frequent habitat used for resting also consisted of one
type of escape terrain, steep rocky/ponderosa pine. Mean
distances from these resting sites to escape terrain for
al l herds except the SE ram herd were less than 10 m.
For the SE ram herd, which utilized ponderosa pine/grass
forb most frequently for resting sites , the mean distance
to escape terrain was 83 m.
Mean distances from feeding areas to escape
terrain were more variable than those from resting sites,
depending on which habitats were utilized most frequently
for feeding (Table 11 ) . Mean escape distances among the
ewe herds were not significantly different (F = 5. 02, p =
0. 0632 ) among the habitats most frequently used for
feeding. This was also true for the ram herds (F = 1. 58 ,
p = 0. 2431 ) , but the number of observations for the SW
ram herd (2 ) were too small to make reliable comparisons.
Distances to escape cover from most frequently utilized
feeding habitats were significantly greater for rams than
ewes (F = 4.99, p = 0. 0009 ) .
Table 1 1 . �ear. dis tance ( no. observations) [ group si ze ) between locations cf �ora;ir.g sheep and escape terrain for all bighorn sheep herds for each � . .;.:: i ::. a : : ;-pe : n Cu s ter s : a :.e Park , SD , from July through Octol::er , 1 985.
�a7��
. � �'.3!:-: :. a :. �·,-'FC
,... . ·, .- - ; : � s s . ! c ::0
pc:: d e rcsa pi r.e/ r.o ur.dc r s c o r z
pcr.derosa p � r.e/ g:-ass fcr b
: i ;l r i a�
s:eep !'OCi<:i/ ;rass :orb
s � e�p reek·;: � : :-.�e r :: s i ... . - o
J'" .. · · -
Sheep Herd ease er.d wes : e nd Grace southeast southwes t
ewe e ... ·e Cool idge rarn rar:, her� herd e�e herd herd �erj
-; j . -5
( 1 0 l [ l 4 )
l S . O ( 1 )
[ 11 J
3 7 . 2 3
( : 6 ) r -
. J '
: : . :, ( : j
[ 6 ]
5. 0 ( 2 )
[ 5 j
6 . 3 ( : } [ s l
- • ? I � • •
( 1 7 I [ H J _ _ b
"6. 9 (B J
i 1 � )
: .; . 6a ( : � )
[ 7 J
70 . 0 1 1 l
! 1 o I
o.o ( J l � 9 �
4 j . .;a ci E . 0 � ?. : a. ( 9 ) ( 3 ) ( 2 )
[ 1 3 J ! 3 J [ l J
� 0 . 0 ( l ) [ 2 J
2 , . 5 80. 6a ':1 9. 0 I 5 ) ( 1 1 l ( 1 )
'. 1 1 J [ 5 J [ :, )
20.B l ; . 'j ( 6 ) I : i [ 7 J [ 1 J
20. 0 ( 2 )
[ 3 l
0.0 0 . 0 ( : ) ( 2 l i l c J l } j - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - -- -- - - - - -- - - - - - - - - - - - - - - - - -
a i r.dica tes �::st f r eq-�ent ly utili �ed f oraging habit 3t. -- i r.dica:es r.o observed occur rences.
3 3
3 4
DISCUSSION
Two predictions of the hypothesis that sexual
segregation of bighorn sheep occur to reduce intersexual
habitat competition (maintenance of home range fidelity
between the sexes and similarity of physical
characteristics for each habitat type among sheep ranges)
were supported by the data in this study. Although the
radio- tagged ram in the SW herd was observed on the WE
range, no other members of the SW herd were found on the
WE range. This ram was relatively young (4-years-old)
and might not have completely established his home range,
resulting in a lower ram herd affinity and increased
wandering (Geist 197 1). Otherwise, herd members
maintained high fidelity to their respective range, and
where ranges did overlap, temporal separation was
maintained. Maintenance of separation during the swnmer
and fall supported the argument that intersexual habitat
competition occurred and was reduced by ram and ewes
occupying separate ranges. If reduced habitat
competition were not important , then greater spatial
overlap of ranges would be expected , such as that found
in the Ya Ha Tinda herd by Morgantini and Hudson (198 1)
during the winter, or that the sexes would have
differential habitat requirements.
Similarity of physical characteristics for each
..
3 5
habitat type among sheep ranges demonstrated that
establishment of each sheep range was not restricted to
any particular types for any herd. This lended support
to the assumption that the ranges were not qualitatively
different, at least in terms of 3-dimensional structure.
It also supported the asslli�ption that ranges presently
occupied by CSP bighorn herds were not established
according to differential habitat preferences or
requirements between sexes.
Although forage quality was found to differ
between ram and ewe ranges in a study by Shank (1979) , he
suggested that these differences were too small to
explain segregation based on these differences and found
no basis for suggesting that different habitat
requirements existed between sexes. However, both
Morgantini and Hudson (198 1) and Geist and Petocz (1977)
reported differences in habitat occupation between rams
and ewes during the winter for 2 different populations.
In the herd studied by Geist and Petocz (1977) , rams were
found on areas with more grassy slopes, while ewes
occupied steeper, more rocky terrain. Rams occupied more
rocky and steeper terrain than ewes in the Ya Ha Tinda
herd (Morgantini and Hudson 1981) . If both sexes
required similar habitats, then habitat occupation should
have been similar between the sexes, whether they
-
36
segregated or not , and should have been consistent over
the geographical range that bighorn sheep inhabit
( Morgantini and Hudson 1981). Since habitats occupied by
each sex were not consistent , these two studies support
the conclusion that rams and ewes do not have different
habitat requirements . Therefore , rams and ewes may be
equally likely to establish seasonal ranges over similar
habitat types and potentially compete with each other for
resources within those ranges.
In order to demonstrate that competition for
habitat resources occurred between the sexes , the
prediction was made that habitat types were utilized with
similar frequencies , between the sexes. However ,
utilization analyses for habitat types did not support
this prediction for comparisons among the herds (Table
4). Although most types were utilized in proportion to
their availability for each herd , when there was
selection for or against a habitat type , types selected
were inconsistent among the herds, except for doghair
ponderosa pine.
Utilization analysis between sexes for pooled
herds suggested differences of habitat utilization
between rams and ewes ( Table 5). However , since habitat
utilization was different for each herd, regardless of
sex , pooling was not considered justified to make
3 7
comparisons between sexes .
Simple comparison of habitat type utilization for
all sheep locations may not be adequate in determining
whether intersexual competition occurs or not, since all
behavior categories were lumped for the utilization
analyses. Competition may be critical for only a few
resources, and since Geist (1971) noted that feeding
areas and resting sites were important factors in
determining sheep distribution, habitat utilization
analyses for these two behavior categories would also be
important in evaluating whether competition occurred
between the sexes for these two resources .
Comparison of habitat utilization among the herds
for feeding areas did not support the prediction that any
habitat type selected for feeding areas would be similar
among the herds (Table 8) . Also, the analysis did not
demonstrate that ewes consistently selected for any
particular habitat type for feeding that were different
from rams ; thus, there was no indication that habitat
requirements for feeding areas might be different between
the sexes . Utilization of habitats for resting sites
suggested that some types were not suitable for this
behavior (Table 11) , but lack of selection for any of the
other habitat types did not support either similarity of
selection among the herds or that there were intersexual
...
3 8
differences of habitat selection for resting sites.
One problem with using utilization analysis for
establishing whether habitat competition occurred between
sexes was that only the results of selection for or
against a habitat type were useful to make comparisons.
The test provided no information regarding differential
frequency of use between habitat types, only whether a
type was utilized more than, less than, or in proportion
to its availability. Thus, no information could be
obtained by comparing, among herds, types which were
utilized according to proportion of their availability.
Another problem with using results of habitat
utilization from this study for determining whether
intersexual competition occurred or not is the small
number of observations used to estimate habitat
utilization. Byers et al. (1984 ) suggested that expected
frequency of usage should be 5 or greater, in all
categories, to insure adequate sample size for their
utilization estimator. Expected frequencies of several
habitat types for every sheep herd were less than 5,
possibly leading to biased results, where there were not
enough observations to establish whether selection
actually existed or not (Table 12 ) .
•
Table 1 2 , Expected number o f occurrences o f use for each habita t type by Rocky Mountain bighorn s heep herds in Custer State Park , SD , f rom July through Octcber , 1 98 5 .
Habi tat Type
Sheep Herd east end '"'est end Grace southeast so'.l.th,.,est
ewe ewe Coolidge ram ram herd herd ewe herd herd herd ----- - ------------------ -------------------------------------- --------
m: xed grass/ ! c r b c O li'
ponderosa pine/ no under story 2 1 6 1 3 1 7 2
ponderosa pine/ grass forb 1 7 9 8 7 3
riparian 3 3 1 0 0
steep rocky/ ·;;rass forb 2 2 0 2 0
stee;, rock;-/ ponderosa pine 1 8 2 7 1 9 3
:; :,;;:-. :. : :-por.cerosa pine 2 1 2 8
3 9
40
Comparison among herds of types used most
frequently for resting and feeding provided some basis
for supporting similarity of habitat use between the
sexes. Although utiliz ation analysis showed that only
the WE ewe herd selected for steep rocky/ponderosa pine ,
this type was most frequently used as resting sites for
all but the SE ram herd. This herd used steep
rocky/ponderosa pine second to ponderosa pine/grass forb.
This suggests that steep rocky/ponderosa pine was the
most important habitat type for resting sites among CSP
sheep herds .
While there was less consistency in most
frequently used habitat types for feeding , riparian ,
mixed grass/forb , and ponderosa pine/grass forb were the
most important types for the sheep herds (Table 9 ) .
There was no evidence for intersexual differences of
types used for feeding as each ewe herd and each ram herd
used a different type most frequently. This suggested no
differences in habitat requirements between rams and ewes
in those types most used for feeding . However , since
only these 4 types were most frequently used by all of
the sheep groups , then evidence of intersexual
competition for feeding habitats can be suggested . This
provided support for the hypothesis of segregation to
reduce competition between the sexes and conformed to the
assWTiption that habitat types were used with similar
frequencies by the sexes.
41
Inconsistencies of habitat types used for feeding
by ram and ewe herds may have been influenced by size of
groups within herds . Risenhoover and Bailey (1985) found
group size and distance to escape terrain important in
predicting foraging efficiency and distribution of
bighorn sheep. Small groups (1-5 individuals) were less
efficient than mediWTI (6-10 individuals ) and large (>10
individuals) groups . Small groups were also rarely found
foraging very far from escape terrain, while large groups
were found to forage at greater than expected distances
from escape terrain . Thus, group size may be important
in determining how efficiently bighorn sheep are able to
exploit available foraging areas . If group size is too
small , then foraging areas may be restricted to areas
relatively close to escape terrain. Since a maximWTI of 7
sheep comprised the SE ram herd and the SW ram herd
consisted of 4 individuals , while ewe herd sizes were
much larger (Table 9) , utilization of feeding areas may
have been restricted for the ram herds. However , rams
were found to forage much farther from escape terrain
than ewes , even though average group size of rams was
considered small . Thus , rams were considered not
excluded from foraging areas located at relatively great
4 2
distances from escape terrain because of small group
size . The failure of predicting group si ze and distance
to escape terrain in CSP rams and may have caused
inconsistencies in utili zation of those types used for
feeding .
Results for habitat types used for feeding among
the herds were especially important since forage
competition is probably the most important form of
habitat competition between the sexes , and since
reproductive fitness of females is considered to be most
dependent on foraging efficiency ( Clutton-Brock et al.
1982) . The results from this study supported the
predictions that the sexes occupy home ranges that share
similar physical characteristics for each habitat type
and that spatial and temporal fidelity of home range
occupation is maintained. The data also suggested that
both sexes utilized habitat types with similar
frequencies , but this conclusion is tentative at best ,
given the paucity of data for the ram herds . The data
also led to the conclusion that rams and ewes do not
exhibit differential habitat requirements , a result
previously reached by Shank (1979) . On the other hand ,
Clutton-Brock et al . (1983 } concluded that red deer
(Cervus elaphus } sexes did segregate according to
differences in habitat preferences suggesting that
....
failure to support the hypothesis proposed by Geist and
Petocz (1977) using data from CSP bighorn sheep may be
unique only to this study.
4 3
Although the hypothesis of segregation to minimize
habitat competition between rams and ewes (Geist and
Petocz 1977) can explain maintenance of sexual
segregation year-round, it does not preclude separation
in order to reduce agonistic interactions among rams and
ewes when reproduction is not possible. Geist (1971)
described the forms of each age and sex in a social
context. Basically, older rams ( 8 years and older) are
the mature forms of bighorn sheep. Mature in this case
means physical, psychological, and social maturation.
Older rams regard all smaller rams, adult ewes,
yearlings, and juveniles merely as undeveloped rams and
treat them as such (Geist 1971). Estrous ewes and
subordinate rams react to aggressive advances by larger
males with a set of specific behavior patterns which
allows the aggressor to express its dominance .
Nonestrous females and lambs simply withdraw and leave
the intentions of the aggressive male incomplete. These
encounters may result in the aggressor (male) chasing the
recipient (female) for some distance thus increasing
energy expenditures for ewes and lambs . Geist ( 1971) has
also documented that rams spar for dominance year-round
44
and interprets this to mean that rams may compete for
dominance, not females . Thus, males that have become
dominant in ewe herds can be detrimentally stressful to
pregnant or lactating ewes by remaining in ewe herds and
continually asserting their dominance. Younger rams (3-5
years old) leave female herds only after becoming
dominant to all members of their herd, then join ram
herds for the duration of their lives (Geist 1971). Any
further social interactions with nonestrous ewes is
meaningless for these dispersing rams. Further dominance
can only be attained by interactions with larger males.
It is suggested that rams may segregate from ewes
in order to reduce agonistic behaviors between the sex
groups. By explaining sexual segregation in bighorn
sheep in this manner, segregation can be maintained
during the nonbreeding time of the year without invoking
altruistic behaviors on the part of rams, where rams
occupy poor quality ranges in order that ewe herds may
occupy higher quality ranges. It would also explain
inconsistencies of habitat utilization for feeding and
resting sites found between the herds in this study by
the fact that these inconsistencies would become
unimportant. The other two assumptions, home range
fidelity and similarity of habitat types within the
ranges of both sexes, would follow from maintaining
segregation in order to reduce agon istic interactions
be tween rams and ewes .
4 5
Critical tests to d iscriminate among the
hypotheses which explain segregation are s till needed as
th is study is inadequa te to do this , or even fully
support the hypothesis of reduced hab i ta t compe tition .
Repea ted stud ies as ou tlined here are needed to de term ine
any patterns of cons is tent d is tribu tion and hab ita t
utilization across d ifferent b ighorn sheep herds.
Concurren t data to establish forage quality and /or
quantity differences , be tween the sexes , are needed for
d ifferent herds . A lso, s tud ies to dec ide the importance
of ram -ewe soc ial interac tions dur ing the breed ing and
nonbreed ing seasons would be necessary to de termine
whe ther energy expend itures from these encoun ters would
be great enough to warran t them as an explana tion for
sexual segregation. Regardless of wh ich hypo thes is is
mos t fully supported , the accepted explana tion should be
adequate to comp le te ly explain sexual segregation found
in Rocky Moun tain b ighorn sheep and, probably , all of the
ungulates.
4 6
MANAGEMENT RECOMMENDAT IONS
The behavior and habitat utilization of t he CSP
bighorn sheep population were found to be similar to
other Rocky Mountain bighorn s heep herds . Since ram a nd
ewe herds maintained spatial segregation t hroughout t he
nonbreeding portion of the year, it is concluded t hat a
need for areas large enough to support separate ram and
ewe ranges is necessary for mainte nance o f a bighorn
population . Herd sizes would depend o n t he area o f
available habitat for each herd. Ewe herds apparently
require rugged terrain (ie . escape cover) closer to
foraging areas tha n rams did. Thus, when considering
habitats for bighorn sheep, areas which might appear as
suitable foraging sites may not be utilized a nd could not
be i ncluded as bighorn sheep habitat . Habitat vegetation
types may not need to be similar for ram a nd ewe ranges,
but this cannot be considered conclusive from this study .
CSP rams were found to form 3 distinct herds where
t he herds were not observed to exchange members during
t he summer a nd fall . Since hunting withi n CSP is
allowed, and rams are taken irrespective o f herd
membership , numbers o f each ram herd , rather than just
the ram population, should be monitored care fully to
i nsure that one herd is not accidentally eliminated . I n
t he event one herd was, one range o f bighorn sheep
habi ta t would be effec tively dele ted from the CSP herd
un til individuals unfamiliar with that range learned to
use i t.
47
LITERATURE CITED
Anderson, S .H. and H.H. Shugart, Jr . 1974. Habitat
selection of breeding birds in an east Tennessee
deciduous forest . Ecology . 55 : 828-837.
Brown, J.H. and A .C. Gipson. 1983. Biogeography. c .v.
Mosby Co., St . Louis, MO . 643 pp.
Brundige, G.c . 1985. Lungworm infections, reproduction,
and summer habitat use of bighorn sheep in Custer
State Park, South Dakota. M . S. Thesis, South Dakota
State Univ ., Brookings, SD 54 pp.
Byers, C .R., R.K . Steinhorst, P . R. Krausman. 1984 .
Clarification of a technique for analysis of
utilization-availibility data. J. Wildl. Manage.
48 : 1050-1053 .
48
Clutton-Brock, T .H., F .E. Guinness, and S.D. Alban. 1982 .
Red deer -- behavior and ecology of two sexes . Univ.
Chicago Press, Chicago, IL. 396 pp .
Daubenmire, R. 1959. A canopy-coverage method of
vegetational analysis . Northwest Sci . 33 : 43-64 .
Geist, V . 1 97 1 . Mountain sheep -- a study in behavior and
evolution. Univ. Chicago Press , Chicago, IL . 383 pp.
----- 1974. On the relationship of social evolution and
ecology in ungulates . Amer . Zool. 14 : 205-220.
----- and R.G. Petocz. 1977 . Bighorn sheep in winter : do
rams maximize reproductive fitness by spatial and
........
habitat segregation from ewes? Can . J . Zool .
55 : 1802-1810 .
----- and P .T . Bromley. 1978. Why deer shed antlers . z .
f . Saugetierkunde . 4: 223-231.
Goodnight, J .H. 1982 . ANOVA, DISCRIM Procedures in A .A .
Ray, ed . SAS user ' s guide : statistics . SAS Inst.
Inc ., Cary, N. C . 58 4 pp.
Hogg, J. T. 1983 . A study of social organization, social
behavior, and population dynamics in Rocky Mountain
bighorn sheep on the National Bison Range, Moiese,
Montana . Job Completion Report W-120-R-13, BG-7. 0 .
University of Montana, Missoula, MT . 62 pp.
49
MacDonald, D. W. , F. G . Ball, and N .G. Hough. 1979 . The
evaluation of home range size and configuration using
radiotracking data . in C. L . Arnlarer and D. W .
MacDonald, eds. A handbook on biotelemetry and
radiotracking. Pergammon Press, Ltd . NY.
Morgantini, L. E. and R. J. Hudson. 1981. Sex differential
in use of the physical environment by bighorn sheep.
Can . Field-Nat. 95 : 69-7 4.
Murie, A. 194 4. The wolves of Mt. McKinley, Fauna of the
National Parks of the U. S. , Fauna series No. 5
Washington, D. C ., U. S. Dep . Int., National Park
Service.
Neu, c . w . , C. R. Byers, and J. M. Peek. 197 4. A technique
for analysis of utilization-availibility data . J .
Wildl . Manage . 38 : 541-545 .
so
Nudds , T.D . 1977 . Quantifying the vegetative structure of
wildlife cover . Wildl . Soc . Bul l . 5 : 1 13-117 .
Ricklefs , R . E . 1979 . Ecology. Chiron Press , Inc . New
York , NY 966 pp .
Risenhoover , K . L . and J . A . Bailey . 1 985 . Foraging ecology
of mountain sheep : implications for habitat
management . J . Wildl . Manage . 49 : 797-804 .
Shank , c . c . 1 979 . Sexual dimorphism in the ecological
niche of wintering Rocky Mountain bighorn sheep .
Ph . D . thesis , Univ . of Calgary , Calgary, Alberta .
193 pp .
Shannon, N . H . , R . J . Hudson , B . C . Brink , and w . o . Kitts.
1975 . Determinants of spatial distribution of Rocky
Mountain bighorn sheep . J . Wildl . Manage . 39 : 387-40 1 .
Sokal R . R . and F . J . Rohlf. 1981 . Biometry , 2nd ed . W . H .
Freeman and Company, San Francisco , CA . 9 16 pp .
Stel fox , J . C. 1976 . Range ecology of Rocky Mountain
bighorn sheep . Can . Wildl. Serv . Rep . Ser . 39 .
Wittenberger , J . F . 1981. Animal social behavior .
Wadsworth, Inc . , Belmont , CA . 782 pp .
51
APPENDIX I
-i 1:: : e A ! . Occur re�=e c� ha bi ta t occupa: �an by the e a st e nd ewe herd in C;;s :er St.Hi! Par k , SD , f rom .:uly throu;h .Jctobe r , 1 9 8 S .
Habitat Type
:, : xe= gra ss; !or!::
ponderosa p i ne / no u:-.derstc:-y
ponderosa pine/ gras s forb
r i parian
s teep rocky/ grass forb
steep rocky/ pondercsa pine
dogr.ai r ponderosa p i ne
:'ot a ls
Total Area No . ( ha l Groups
Observed
6 6 . 1 2 1 0
1 ) 9 . 7 9 5
1 1 7 . 55 24
1 8 . 7 3
1 0 . 50 4
1 1 8 . 58 27
1 2 . 68 0
4 8 3 . 9 5 7 2
Actual Proportion o f Usage
C Pi I
0 . 1 3 9
0 . 06 9
0 . 3 3 3
0 . ;, 2 8
C . C 5 6
0 . 3 -; �
0 . 000
Expected Proportion o f Usage
( Pf o l
0 . 1 ) 7
o . 2 a 9a
,:i . 24 3
0 . 0 3 9
c . 0 2 2
Bonferroni I nterva l s
for Pi
0 . 0 2 9.$..?15_0 . 2 4 9
0 . 0 00.$..?25.0 . 1 50
0 . 1 8 4.$..?35_0 . 48 3
O . OOOS,.P4i0 . 0 8C
o . 0005.i::3io . 1 2 e
o . 2 4 5 0 . 2 2 2.$..?6!0 . :: 2 8
0 . 02 6a O . OOOiP7iO . OO O
- - ------- ------- ------ - - - - - - - - - - - - - - - ----- - - - - -- - ·----- ----- - - - -- - - - - - -a ind i ca tes select ion against a habitat type a t the 0 . 0 5 sign i f i ca nce
leve l .
Table A 2 . Occurrence o f h abi t a t occupation by the west e nd ewe herd i� Custer Stat e Pa rk , SD , f rom July througr. October , 1 98 5 .
Habita t: Type Tota l Area No . ( h a } Groups
Observed
Actua l Proportion o f Usage
( Pi I
Expected Propor t i on o f Usage
! Pio l
Bonferron i I nterva ls
f or Pi
- - -- - - - ------------- ---- - - - - - - - ---- -- ---- - ---- ----- - - - - - ---- --- - - - - -mi xed grass/ f o r b 27 . 7 6 1 7 0 . 1 % o . 02 sa 0 . 0 B2iP1iO . 3 1 3
ponderosa pi net 0 . 3 59 b no unders t.ory 3 9 6. 7 5 6 0 . 070 0 . 0 00s.F2iO. H4
ponoeros a p i ne / grass forb 1 1 3 . 4 9 1 1 0 . 1 2 8 0 . 1 0 3 0 . 0 3 1.5.p35_0. 2 2 5
r i pa r ia n 35 . 3 2 2 5 0 . 2 9 1 0 . 0 3 :'a O . l 5 35.P �_i0 . 4 Z 2
s tee� reek;·/ gra ss forb : i . 86 2 0 . 02 ) 0 . 020 0 . 0005.p5.5.0 . 06 7
s , eec rock"}/ ponderosa p : ne 3 5� . .; I L S 0 . 2 9 : o . 3 1 9 0 . l 5 9.s_p65_0 . 4 ;; :
dogha ir O . l 4 2b ponderos a pine 1 56 . 4 0 0 0 . 000 0 . 000.5.p7.5.0 . 00C
Tot a ls i 1 0 4 . 0 5 86 - �- - - - ---- - - ---- - - - --- �--- - - - ---- - - - - - - - - - - - � - ----- --- ---------------- -a
b
: ndica �es select ion for a ha bitat type at the 0 . 0 5 s i gn i ficance l eve l . i ndica tes select ion aga inst a hab i ta t tn:e at t he 0 . 0 � s igni f icance leve l .
52
!at " e Al . Occurrence cf �abit a t occ�pat i on by the Grace Cool i dge ewe ��r� . . ;:: _s':er Seate i'ark. , SD, ! ro:n July tr.rough October , 1 98 5 .
Habitat Type
:r.: x�j gra ss / f =: r c
por::!erosa pine/ . . � ur.derstory
p-:.:.:!e rosa pine/ gr.:: S S forb
: : ;:arian
S : i!ep rocky/ 9rass forb
s :e-l?p rocky / F : r . .::erosa p i ne
cc;�.air p:�.:::erosa pine
TC': .l ls
Tota l Area ( ha I
1 e . , �
1 8 1 . 98
1 19 . � 2
1 7 . 90
6 . 20
2 ,: 4 . 09
1 09 . 7:,
5 3 5 . 800
No . Groups
Observed
1 2
1 0
9
6
0
l �
c 5 1
Actual Proporticn
of Usage ( Pi l
0 . 2 3 5
0 . 1 9 6
0 . 1 76
0 . 1 1 8
:i . ooo
::, . 2 7 5
� . 000
Expected Propor t ion
of Usage I P1o l
0 . 02 6 a
0 . 2 5 4
0 . 1 6 6
0 . 0 2 5
0 . 00 9
0 . 36 8
o . 1 s 3b
Bonferroni Interva ls
for Pi
0 . 076.s.P 1£0 . 39 5
0 . 0 4 7S.P2S.O . 3 4 6
0 . 0 3 l_iP3.i0 . 32C
0 . 000s_p4_5.0 . 2 3'?
0 . CCOs_p5.5.0 . ooc,
0 . 106s._p6s_0 . 4 4 3
O . OOOiP7£0 . 00C
: ndicates s election for a habitat type a t the 0 . 05 s ignif icance :eve.:. . :nd�cates se lect ion against a habitat type at t he 0 . 0 5 s igni ficance �e ·le l .
Table A4 . Occurrence of habit a t occupa tion by the southeast ram herd in Custer State Park. , SD , f rom July t hrough October , 1 98 5 .
H abitat: T'fPe
:r :. ;.:�d gra ss , f ::rl::
ponderosa pine/ r.o under story
pcnderos a pi_ne/ ; : ass forb
: ip .. r i an
Heep rock.JI gr a ss forb
s teep rocky/ ?Or.derosa pine
doghair i:onde rosa pi r:e
Totals
':'ot a l Area ( ha J
3 $ . 6 3
3 1 6 . 5 2
1 3 l . 0 7
:'> . 97
) 4 . 5 :
6 4 . H
4 7 . 7 3
4 !: u . 6 7 - - - - - --- � - ----- - - - - - - - - - -
No . Groups
Observed
2 0
0
2
5
0
H
Actual Proportion of Usage
I Pi l
0 . 08S
0 . 1 1 8
0 . :'>83
0 . 00:J
0 . 0 59
O . l fl
0 . 000
Expected P roport ion
of Us age ( Pio )
0 . 0 :1 6
0 . 4 9 8a
0 . 2 0 6b
o . oo g a
0 . 0 5 .;
0 . 1 0 1
o . 01sa
Bonferroni I nt erva l s
for Pi
0 . 000.5.p1.5.0 . 2 1 9
O . O OOs._p2_i0 . 26 6
0 . ) 6l.s_p3.5.0 . Bl 5
O . O OOs_p4i_0 . 000
O . O OOs_p5.5.0 . 1 6 7
0 . 000.5.p5s._O . J l �
o . o oo.5.p7.5.o . o : :
- - - - � - - ---- - --- - - - - - --- - - - - - - - - - ---- - - - --- -indicates select io!'l a;ains: a habitat tYJ;e a t the 0 . 0 5 sign i f icanc� leve l . i nd icates select ic� fer a hab i t a � t ype a t the O . O S s � gn i f �cance leve '. .
5 3
"." c :- :e h � . Occur :-ence o f habi t a t occupa ti c:: cy the so,ith.;e s t r am r.erd l ,. c� ; �e , S:ate Park , SD , f ro� July throJgh OCtobe r , 1 9 8 5 .
Ho:.i tat Type
:'f'li :·:c=. ;:-:tss /
fer::
;;o,.deros a pine/ . . _ 1.nde rs tcry
;:,c�.:lercsa pine / gr ass forb
:- 1 ;-.:! : : a n
s :!:C? rcc.i.;y/ ; !.' .i S S f c rb
st ce? reek:,· I ��:".:.e�osa pine
dcg�:a i r pondercs a pine
::ica l s
Total ll.rea ( ha )
6 1 . 3 4
1 9 6 . 57
3 3 8 . 97
S . 6 5
2 3 . 1 5
4 1 2 . 6 8
6 5 . 9 2
1 1 0 7 . 2 8
No . Groups
Observed
1
0
0
0
1 0
ll.ctual Proporti on
of Usage ( pi )
0 . � 00
0 . 300
0 , 100
0 . 000
0 . 000
o . �oc
0 . 000
Expected Propor ti on o f Usage
( Pi o )
0 . 0 5 5
0 . 1 7 8
0 . 3 0 6
o . o o 8 a
o . o :?l a
0 . 3 7 3
O . OGo a
Bon fer roni Interva l s for Pi
O . OCOiP1i0 . 3 5 5
O . OOOi.P2i0 , 69 0
O . OOOi_P3i0 . 35 5
O . OOOiP4iO . OO O
O . OCOiJ:>5iO . OO O
0 . 07 :' iPoi0 . 92 :i
O . OOOiP7iO , OO O
- - - - -- - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - -
a indicates select ion aga�nst a habi tat t;-;:e at the 0 . 05 s ign i f icance '. e '.'E ? ,
,ajie A6 . occurrence o f ha b i ta t occupat ion for pooled ewe herds in Custer s : � :e Park , SD, f rom Ju ly through OCtobe r , 1 9 8 :i .
Ea::itat Type Tot a l ll.rea No . ( ha ) Groups
Observed
ll.ct·jal Propor t i on o f Jsage
( pi )
Expected Propo r t ion o f Usage
I Pio l
Bonferroni Interval s for Pi
w - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - · - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
r: .:. :·�ec <;r ! S S ,
!c::l:: 1 1 2 . 22
p: �.:ieros a pir.e / r - under s :ory 7 1 8 . 5 2
,-c r:derosa pine / I'.;: ;;. S S f c t"� 3 $0 . 46
:- : � i :· :. a n 7 1 . 95
; l €!'!P re::,:;· I t; :- i ii S fc:- ': 3 6 . 56
s -: ee;: ::eck:r.1 p::�.de ros a pine 7 3 .$ . 1 4
dogha i: pc�.de:-csa pine 2 78 . 8 3
Tc:. a l s 2 30 5 . 6 8
3 9
2 1
� � :, 3
6
6 6
0
209
0 . : 87 o . c � 9 a
0 . 1 00 0 . 3 1 2b
0 . 2 1 1 0 . 1 :i:t
0 . , 5 2 0 . 0 3.i. 8
0 . 1) 2 9 0 . 0 1 7
0 . 3 ! 6 0 . 3 1 9
0 . DC() o . 1 2 1 b
0 . 1 1 �.iP1iO . 2 5 9
0 . 0 4 5i.P2s.0 . 15 6
0 . l 3 5iJ:>3iO . 286
0 . 0 90iP �S.0 . 22 6
O . OOOiP5i0 . 060
0 . .2 29iJ:>6S.0 , 4 02
O . OOCiP7iO . OOO
- - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - -
a r.d i ates se lec:ion for a habi t a t t '{pe a t the O . 05 s igni ficance ,e,ve
I: nd: a :es se lect ion a; a i�s t a habitat type a t the o . o ; s ign i f ica nce �J: �.-e
54
7ab :e A7 . Occurrence of hab �tat cccupa:ion for pooled ra, herds i n Custe: State Par k , SD , f rom July through October , 1 9 8 5 .
� !. >:0:: -;ra s s t ! o ::-;:;
por.de rosa pine/ no under story
pondercsa pine/ g ra ss for b
ripa r ian
steep rocky/ g ra ss forb
steep rocky/ pondercsa pine
dogha ir por.derosa pine
Tctals
Tota l Area ( h a l
96 . 9 7
5 1 3 . 0 9
4 7 0 . 0 4
H . 6 2
57 . 6 7
4 7 7 . 02
1 1 3 , 6 5
17 4 3 . 0 6
No . Groui:s
Observed
7
2 1
0
2
1 0
0
H
r\ctua l Proport ion of Us age
! Pi l
:-- . ·:, � 1
0 . 1 5 9
o . 4 7 7
0 . 000
o . o�s
o . :27
C . GOO
Expected Proport ion of Usa ge
( Pi o l
0 . 0 � 6
0 . 2 9 4
o . 2 1 o a
o . oos b
0 . 0 3 3
0 . 2 7 4
o . o es a
Bon f e r roni Interva ls
for Pi
O . OC 09 1�o . : (l7
0 . 0 1 19>2�0 . 3 0 7
0 . 2 75�P3i0 . 6 80
0 . 02 S9)4iO . OOO
0 . 0 00.1.Ps�O . U O
0 . 0 579>5i0 . 39 7
0 . 00097�0 . 000
- - - -- - - - - - --- - · - - - -- - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
ndica tes se lection for a �abi ta : type a t the O . OS sign i f icance eve l .
b r.dica tes selection a gainst a ha bita� t�-pe at the 0 . 0 5 s ignif icance eve 1 .
5 5
100
90
tiO
70
w 60 C> < 0:: w > 50 0 (.)
.... z w
-IU 0 "' w "-
::io
�o
Duff Forb Grass Hor i zontal Oensi ty
Rock
l'19ure Al . Mean ( + ;;e ) of grou11d cover , canopy cover , and horizout11l ..1.,n,;i ty for a ! l bi�horn .t.1h""P ra11ycs tor mixed 9nu,s/ forb h<ibi t at cy 1,c in Custer State Park, SD , 19HS .
S ei<:.t end ..:we herd , 8 Grae., Coo l idye ewe herd , Bsoucheast ram hc L'd,
[J :rnuthwe:.t r.lm herd , Qwes t ..;nd ewe herd .
:.hrul.l
l1l m
lUO
9 0
1::10
70
w 60
50
w .ao
w
30
.:o
I U
0
I I
� . I
; .
...i..:mi��,L......Llt.ll.,..l::ill"'--l..lLl�J.::,jjloL . ....J..J(.:L,..1..:i--1..l.�....L:l--.L.llui.-_Jl,l __ �--1·.-..-.
l:lare c,mopy Duff Forb Grass Hor izont "l Log Hock :,i,rub <: round Cover Density
tlgure A2 . Me"n ( + ¥v i of ground cover , canopy cover , dnd hor iz on tdl .:i.n.ity for all bighorn ¥h•oP r .. n.,,cs for l ipar ian hdb l t d t cyve in Cu.t..t ..:r �tate Par k , SD , 1 9 U 5 .
S e,u;. t c,nd ewe herd , 8 Gr<1ce
m sou thwc sc rdm herd , c .... ., s t
Cool 1..t._;, ewe herd ,
cud ..:we hcrd .
a ,iOUthedst rdlll he rel ,
U1 -.J
100
90
1:10
70
...... 60
50
-..,.. ..., 40 u (,:; w Q.
30
.!O
1 u
0 ;u.L...-.,..-,.__.J.l:illi�L£:il:JIY,llillU.._.Jllt:11f.1.ll:ilil.L.JJ:il'.:ir,.lli:ll.t....llL� Sar" Canopy Duff l,orb Grass Hori zontal Log kock
Ground cover Densi t y
Figure Al . M,:an I ! a� ) o f g round covci· , canopy cover , and horizontal a,mslty tor all bighorn a.ht:ep Lmqes tor Pondcrosct pine/no undcrstor y habi tat c.yp,: in Custer State Pd , � . sn . I :Ill!> .
S oect.!it t:nd ""'" herd, 8 Grae.,
fil southwt0st. rdlll h<'rd , Q we,.t end ewt0 herd .
Sl,ruL,
U1
(X)
100
90
i:10
70
w 60
50 0 u .... z ...., 40 u a::
30
20
1 0
0 Duff
I
�, """L.JJl.1.W::.u,Ji..u.;L--, . .i.....,_
l:ldt·e C,rnopy Ground Cover
Grdss Hor i zont .. l Dens ity
Forb Nock Log
l:'iqure A. .. . Mc .. u f ! lie ) of qround cover , c .. nopy cov«,: , .. nd hor i zont .. l <1cnlii i t y t or n l . bighorn i.heep 1·an>,1.: s for 1'011<:terosd pi.nc/c,Jrdss t o r b t,.t.bitdt type i , . i:'ust111r s t .. te Pdrk . SD , l 'i !:! S .
ned,;.t end ewe herd , g Grace Coo 1 1..i..,c ewe herd , D ,;ou thed s t '"'" herd ,
� sou thwl!:.t ram herd , Q west cud '-'""<.! herd .
:>hr ul..l
100
90
1:10
70
...... 60
!)O 0 0 ...
z ......
40 0 o::'. 41 Q.
::io
20
10
0 u .. re Canopy Duff Forb Crass Horizon tal Log Rock
(hound Cove r Density figure AS . Mean ( _! se ) of grounCl cove r , canopy cove r , and hori zont .. l ci.:n,;1 ty toe a l J bi<;horn i.htu.:1> 1 .. n\l"" f<ff steep rocky /ponClen:i .... pine i. .. bitd.t tYl •L· i n <'uster State p .. .- � . sn . 1 '111 ', .
8 cas t end ewe he rd, 8 Grace c..,o hdqe ew,: h,:rd , gsoucheds t r..m hero ,
m southwest rdln herd , Qwest end <,w,.: herd .
:.;hrul.>
O'I 0
100
90
BO
70
w 60
Q: w
> 50 0
0
...
z ....
40 u u: w a.
30
.::0
1 0
0
bare c,nopy Dutt For b Grass Hor i zontal Log kock Shrub Grvund Cover Density
r'igui:-c Ali . Mcdn (.! so l of ground cove r , Cdnopy cover , dnd hor ixont " l uons ity t o r .. 1 1 bignorn .,;h""I' ranye" t o r steep r:ocky/yr,,,::.s for b hdb l l ,H tyi• in Custer Stat:� Pdrk , SD . I Y H � .
S ..:.1st: end ewe herd, 8 Grd-:c: Coolidge ewe herd , Esout:Lcd::.l: Cd.o herd ,
gsouchwesc cam h<'rd, Qwc,- 1; end ewe herd .
0\
......
!:>UOUOO 1
400000
<
� 300000
� 0 ....... < .... 0:: < ..J 200000 <( Vl <( a:i
,00000
I-- S£ I-- SW � I-- wE -1
figure A7 . Mean I� se ) of tree bas a l area/ha for a ll bighorn sheep ranges for riparian habitat type in Custer Stat� Par k , SD , 1 9 8 5 . ££ eas t end ew..., herd , GC "' Gr .. .::e Coolidge e"'e he1·d , SE "' sou�hca::,;c 1·am herd , SW " ,.ou thw<.: i.. l 1·Am herd , w t: - west: end. e:"'e herd .
� Pinu,. ponderoiia � Quercu:,; m .. ccocarpd � bet:uld pap.iferou.,,
E\J Populus cremulo.ides � Mbce l laneous species .
AREA
0\ N
4( ::x:
700000
600000
500000
"''::::;, �00000 ::i (.) -
"' ..... � 2100000 ...J � Vl � £D
200000
100000
r-- EE --{ r-- GC --j !--- SE ----{ SW --j
Figure Ad . Medn ( + se ) of tree bu,a l area /ha for a l l b i ghor n shet1p r,mges t o r Ponde rosa p l ne7no unde rstory hab1t<lt type in Custer State P .. , K . SO , 1 9 8 5 . le:!:: = ca,-t end ewe tuird , G..: - Gr ,"�" Coo l idge ewe hte rd , SE = ,:;ou1 i.c.:. s t rum hc: rd , ::; w "' ,;ouc.hwest rl\m 1 1,,r,1 . w i:: " wt>st end cwt: herd .
� Pinus pond.cro•a � Quercus r.i<tcroc .. rpd ['.'.2:1 Bec.ula p<tpi fte.r ou:.
ES:! Popu lu .. t: 1 cmuloides � M 1 scc l l .111t:oui. £yc:cie.s. .
AHl: A
< :x: ' �
6000(J()
!:>00000
400000
� 300000 <( ..... IX <
_.. < Vl < 200000 £D
1 00000
r- t::E --f r- GC --f I-- SE -·-· J I-- SW
l'i,;iunt A'L M..:..in ( + !>c ) of t ree u<.1s.<1l <.1 rc.;./hd L ui' .. 11 bi9!1Qn1 :.l a,.;cp i·awJ..:;; fo1· Pomh:ros,, pJ.t1c/griis>1 tori.> h.it.i tat type.: .in t:u,,;t..:r :;c .. tc Pu.r k , :;u , l !l tl � . t:£ = C<ist end ,:\Jc ht!rd , GC = G.ritcu Cool idgc cwc hcL'd , S J:: = sout l,1.: ... :. t .ram ht:rd , SW " sout hwc,;t t•..in h<:!1'0 , WI:: "' 1,1-,,; c cm.I cwc he.rd .
� l:'1111.11o1 pond.:r..:: .. , � Quarcu:. 111<1c.coc .. rp<1 � tit'!tUl<i v"Pi f ll .Cc.1..,-
b::) Populus t rcmu loidcli
AHl: A
tioovuo
700000
600000
< 500000 :c ........
N-..
� £ 400000 ,c( w a: "" ...I
300000 "" VI ,c( a)
�00000
1 00000
'
t-- EE -"1 f-- GC -"j f-- SE -"1 f-- Sw -"1
F igure AlO . Mc,m I! se ) of t ree basal .. rca/ha for a l l bighorn :;h.:.:cp r .. ng..,,. for steep rocky/ponderosd pine habi tat type iu Custer State Park , SO, 1 98 5 . EE = ua:;t c11,l ewe herd , GC • i;r;;cc C,h1l idgc ewe h"1 d , !;f. = :.:outh<: ... ::. t r.,m he rd , SW = .;outhW!dlit rd.ll'I h,n a , Wt: " west end ewt: t ... : r d .
� Pinu11 ponderO&d � Quercus m .. croc .. rp.. [:'.2l lietula p .. pit erou,;
ESJ Popu lu .. t remuloi<.J.cs E:32 111sc.: ll aneou:; :.p..,.::ie:; .
O"\ U1
-< :I: -....... Ill ::i
401)
300
LJ 200 � Ill
0 z
100
I-- EE --j I-- GC --j I-- SE --j I-- SW --j f-- WE --j
Fi9ure Al l . Mean ( + se ) of number of t re., stems/ha for dll bighorn shccp rdnges for r i pd C ldn-habitat type in Cu� tcr State Pa r k , SD, 1 9 8 5 . ££ = c�st end ew� hc:: 1 ..! , GC -' Grdce Cool idg<· '"""' herd , : ;,.; = !>Ou ( l1cj:. t r....rn hc1·d , !;'.i .. .. .:.uchwe.s t ro1m r,..: n.1 , W£ ;: w�st end ewe herd .
� Pinus pondcros4 � Quercu.s rn ... ci::oc4rp" 22] uecul" p4pi tcrous
C'.J Populus t r cmulo1<les � Miscc l ldnf:ous specie;:. .
.t.AE.t.
.. ooo
3000
< :c ........ Vl
� w 2000 I-
I.II
0 z
1000
j- EE --1 j- GC --1 j- SE --1 j- SW --1 j- wE - j
Figure Al 2 . M..:an I + se ) of nwn!Jer of tree stems/ha tor a l l bighorn sheep r ange,; tor i'onderos;; pine/no unc.Jerstory ha.bitdt type in Cu,;ter St,lt . ..: l:',:.,· k , SIL l �HS . 1::l:: "' cctst end cw..: 1 ... : , <.1 , l.iC Gr:,<':c Cool idge <.:we tu..: rd , SI:: "' liOUtheast ram herd , SW = .:;out llw<.::..t r .:.m h,.;;riJ , Wt:: = we,; t cn<1 ewe h-, L"d .
[::§j l!inus s,ona.:roS4 � Quercus mctcroc.i.rp.. Z2l aet.ulc1 pdpifcrous
E:) Poi,ulus tremulo1des tZ1 Mi;;c1.d ldn<.:ous specioos .
AflfA
1200
< 900
' V)
:f: .... ...... Ill
ci :z 600
300
I-- £E --! I-- GC -I I-- SE --j
Figure Al l . Mo.,.in I ! se ) of m.uuJ;>c i:- of tn,c stems/ha foe a l l bigt,oen st ,.,cp r.sugci. foe l'ondceoi..s pine/gr .. .;:,; forb ha.bi tat type in Cus t.:r St.st" l'a e k , Sli , l9H5 . EE = , . ,, . , t end ewe h,:HI , GC " G , ,.,;., Cuo l idq,· ,::w" herd , St,; = sourh<:<1 :. t e ..,n h,:rd , S W . . southwest r am ho::,rd . W E "' wc :.r ,,11,1 ewe heed .
1.:1:J P1nua pon<J .. roa.a � Qucrcus ltl<lct·oc .. rp.. [::'.2l be tul..s i,.sp 1 t t:i ous
ES) Popu lu� trcmulo1des
AHEA
O"\ 00
2::.00
2000
< 1�00 ::r ......... II)
::.l w .... II)
ci :Z 1000
soo
f-- EE -f F igure Al 4 . Mc.:, n I ! ,;e ) of nwnbcc of cccc SC<.!ms /ha t oe ct l l bi ,i l,al !, ,m..:..:p t dngc:,; t o r :,;teeµ i:ocky/ponderosd pine/gc.;.s:,; forb habJ..c .. c t yf,c iu Cu�t<.: r :., t a t e h•.r k , SLJ , l 9 l:I � . EE = e " s t c 11d ""'' h..: rd , GC = Gr ct<:c Coo l i a,Jc cwi.: h..: 1 il , :c:i:: = .:.uuthcast t .lll\ he rd , SW = s,.,u r hw,, s t i·...i11 hc rd , WE ::. w ... s t cnd ewe 1,..: .-.:.t .
� Plnu,; ponde ro11.a � Oue rcu,; macroc .. rpa � B e t u l ii. p .. pif ., cou,.
£:SJ Popu lus t r emulo1des � Misce l l .. ncou:. spcci<..:s .