The effects of housing developments and agriculture on the ecology of white-tailed deer and mule deer
in the Gallatin Valley, Montana
by William Otto Vogel

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Fish
and Wildlife Management
Montana State University
© Copyright by William Otto Vogel (1983)

Abstract:
Investigations of white-tailed deer and mule deer were conducted from the spring of 1981 through the
summer of 1982 on over 1000 km^2 in the Gallatin Valley, Montana. The Gallatin Valley is a
primarily agricultural high mountain valley with many streams flowing through it. White-tailed deer,
quite scarce 10 years ago, currently comprise over half of the deer population. In recent years
considerable housing development has occurred. An inverse relationship existed between housing
density and deer observed. This relationship appeared to have a threshold around 4-8 houses per km^2 .
Deer use was greatest on the moderately arable soils while housing development occurred on the more
arable soils. 'Deer use peaked on land with intermediate densities of houses indicating that development
is occurring on deer habitat while marginal deer habitat is left undeveloped. Deer were less likely to be
active during midday if they were near housing developments. White-tailed deer were more secretive,
more nocturnal, had greater reproductive capabilities, a shorter life expectancy and appeared better
adapted to human activities than mule deer. Reproductive data collected from carcasses yielded 1.62
fetuses per doe for white tails and 1.23 for mule deer. In conjunction with fawn-doe ratios an estimated
fawn mortality of 40% was calculated for the first 90 days in white-tailed deer and 54% for mule deer.
The presence of hybrids was also noted. 



THE EFFECTS OF HOUSING DEVELOPMENTS AND AGRICULTURE 
ON THE ECOLOGY OF WHITE-TAILED DEER AND MULE 

DEER IN THE GALLATIN VALLEY, MONTANA

by
William Otto Vogel

A thesis submitted in partial fulfillment 
of the requirements for the degree

of

Master of Science 
in

Fish and Wildlife Management

MONTANA STATE UNIVERSITY 
Bozeman, Montana

June 1983



MAIN UR,

N  ^ ^  3' j

Cop. 3 ii

APPROVAL

of a thesis submitted by

William Otto Vogel

This thesis has been read by each member of the thesis committee 
and has been found to be satisfactory regarding content, English 
usage, format, citations, bibliographic style, and consistency, and 
is ready for submission to the College of Graduate Studies.

Dhti Chairperson, Graduate Committee

Approved for the Major Department

nrt
Date I Head, Major Department

Approved for the College of Graduate Studies

Date Graduate Dean 7



iii

STATEMENT OF PERMISSION TO USE

In presenting this thesis in partial fulfillment of the require­
ments for a master's degree at Montana State University, I agree that 
the Library shall make it available to borrowers under rules of the 
Library. Brief quotations from this thesis are allowable without 
special permission, provided that accurate acknowledgment of source 
is made.

Permission for extensive quotation from or reproduction of this 
thesis may be granted by my major professor, or in his absence, by 
the Director of Libraries when, in the opinion of either, the proposed 
use of the material is.for scholarly purposes. Any copying or use of 
the material in this thesis for financial gain shall not be allowed 
without my written permission.

5

U ■ .
Date /<r /9̂ 1
Signature_

£7



V

ACKNOWLEDGEMENT

I would like to express my gratitude to the people who have 
contributed to the completion of this study: Dr. Harold D. Picton, 
Montana State University, for his advice, guidance and help in the 
preparation of this thesis; Dr. William Gould and Dr. Robert Eng for 
review of the manuscript; Mr. David Pac for assisting me with and 
teaching me many techniques involved with trapping deer and other 
aspects of. wildlife management that are not found in any text book; 
Mr. James Stradley, Gallatin Flying Service, for his expertise in 
aviation and abilities in locating deer. I would also like to thank: 
Dr. Robert Moore, Dr. Richard Mackie and Dr. Lynn Irby for their 
assistance in many ways; Sean Wickens for assistance in soil mapping; 

Wayne Kasworm for his assistance with trapping operations; and local 
landowners for their cooperation and tolerance; especially Mr. James 
T. Paugh, Mr. Richard Gee, Mr. Donald T. Saisbury, Mr. Lester Schultz 
Mr. E. D. Blackwood, Mr. Jack Nash, Mr. Wilbur Springs, Mr. Louis 
Kamp, Mr. Jon Alberida, and Mr. Keith Noyes. I would also like to 

thank Alice for cooking, cleaning and for support and my brother 
Herbert for his assistance and companionship during the first field 

season. I would especially like to thank my parents without whom 
none of this would have been possible for me.



J
vi

TABLE OF CONTENTS
Page

VITA...............................................   iv
ACKNOWLEDGEMENT............................   v
TABLE OF CONTENTS................................   vi
LIST OF TABLES.........................................  viii
LIST OF FIGURES...................................    xii
ABSTRACT.........................................   xiv
INTRODUCTION...........................................  I
DESCRIPTION OF STUDY AREA..............................  3

Topography 
History...
Climate...
Vegetation......................................... 10
Large Mammals...................   12

METHODS..............................................    14
RESULTS................................................  18

Trapping................................... :.....  18
Movements.............   19

Reese Creek....................................   24
Pine Butte....................................  27
Penwell.......................................  27
Sourdough............  28
Valley Center................................. 29

Population Structure............................... 29
Interrelationships ............................ . .. 35
Food Habits........................................  36
Use of Cover Types.................................  37
Use of Hay Stacks..................................  38
Activity Patterns..............   40
Effects of Housing.................................  47
Effects of Roads..........................    54
Use of Soils.............................    55

CO Lf) OO



DISCUSSION..........    56
Movements and Distribution.........     56
Population Structure....    58
Soils.......................       59
Use of Vegetation................   60
Activity Patterns... ..................   62
Development Effects.................    63
Hunting.........      69

CONCLUSIONS AND RECOMMENDATIONS .......................  70
REFERENCES CITED.......................................  72

vii

TABLE OF CONTENTS— Continued

Page

APPENDIX 77



viii

LIST OF TABLES
Table Page
1. Land use in acres harvested in Gallatin County

from 1950 to 1979................................. 7
2. Population of Gallatin County from 1950 to 1980.... 8
3. The changes in farms and farm size in Gallatin

County............................................  8

4. Climatological data from the Montana State
University weather station at Bozeman, MT for 
November through May, 1971-1982........   9

5. Distribution of snow cover in the Gallatin Valley
from October 1981 to April 1982.................. . 10

6. The location, date, and cause of death of 5 
marked deer in the Gallatin Valley during 1981
and 1982............       18

7s. Home ranges and movements of radio-collared deer 
of both species in the Gallatin Valley in 1981 
and 1982.....     26

8. Sex and age proportions of deer in the Gallatin 
Valley in late August 1981 from a 3 day
aerial survey, in percent of classifiable deer.... 30

9. Sex and age proportions of deer in the Gallatin 
Valley in 1981 and 1982 derived from flights 
during all seasons, in percent of classifiable
deer.........;...................................  31

10. Sex and age proportions of deer in the Gallatin
Valley derived from observations along routes 
during 1981 and 1982, in percent of classifiable 
deer..............................................  31

11. Sex and age proportions of deer carcasses 
collected from area roads in the Gallatin
Valley from 1976 to . 1982.......................... 31



ix

12. The ratio of young per adult female for 
white-tailed deer seen during flights made 
thoughout the year in the Gallatin Valley
during 1981 and 1982............................ . 32

13. Productivity of female deer of different ages 
from counts of fetuses in carcasses collected 
between January I and June 14 in the Gallatin
Valley between 1976 and 1982...................... 33

14. A comparison of the diets of mule deer and 
white-tailed deer in % volume of the rumen in 
and immediately around the.Gallatin Valley from
1976 to 1982.......... ............................ 37

15. The seasonal use of cover types by both species 
of deer in the Gallatin Valley in percent of
deer seen in that season during 1981 and 1982..... 39

16. The percent of deer using hay stacks during
different seasons and snow depths in the Gallatin 
Valley during 1981 and 1982....................... 39

17. Observed deer activity during different temporal 
conditions in the average number of deer seen per 
survey of 21 plots in the Gallatin Valley during
1981 and 1982............  ...... ........... ..... 42

18. The percentage of both species of deer that 
were observed during the day and night in plots 
associated with different housing densities in
the Gallatin Valley during 1981 and 1982......... . 47

19. Deer use by both species along routes relative
to housing in the Gallatin Valley during 1981 and
1982 ............................................. 49

20. Average housing density for 4 development
types in the Gallatin Valley, 1982................. 50

21. Relative deer usage, by both species, of 
development types on the basis of ..availability 
by 3 methods of sampling in the Gallatin
Valley during 1981 and 1982.......... .......... .. 50

LIST OF TABLES— Continued

Table Page



X

22. The relative use of areas by both species of deer 
relative to the development type and the number 
of houses within 800 m in the Gallatin
Valley during 1981 and 1982.......................  51

23. Deer avoidance of housing in different development
types in the Gallatin Valley during 1981 and 1982.. 52

24. Comparison of average distance to cover and 
average number of houses within 800 m for
3 methods of observation of both species of
deer in the Gallatin Valley during 1981 and 1982... 52

25. The use of cover by white-tailed deer and mule 
deer in relation to housing density within 900 
m of the survey plots in the Gallatin
Valley during 1981 and 1982.......................  53

26. Deer use of cover along routes relative to the 
development type and the amount of housing 
within 800 m in the Gallatin Valley during
1981 and 1982..................................... 54

27. Relative distribution of deer within different 
distances of roads as measured by 2 methods
of observation in the Gallatin Valley during
1981 and 1982..................................... 55

28. Description of deer marked in the Gallatin
Valley, Montana during 1981 and 1982.............. 77

29. The monthly food habits in % by volume of 
white-tailed deer as determined from rumen 
analysis of deer carcasses collected
between 1976 and 1982 in the Gallatin Valley...... 79

30. The monthly food habits in % by volume of mule 
deer as determined from rumen analysis of 
deer carcasses collected between 1976 and
1982 in the Gallatin Valley......................  79

31. Relative usage by deer of both species and 
characteristics of cover types in the Gallatin
Valley during 1981 and 1982....................... .80

LIST OF TABLES— Continued

Table Page



xi

LIST OF TABLES— Continued
Table Page
32. The avoidance of housing by deer of both species 

relative to the distance between deer and housing 
in the Gallatin Valley during 1981 and 1982....... 81



xii

LIST OF FIGURES
Figure Page

1. A map of the study area in the Gallatin Valley,
Montana........................................... 4

2. Distribution of white-tailed deer observations 
during the summers of 1981 and 1982 in the
Gallatin Valley ................................. 20

3. Distribution of white-tailed deer observations 
during the winters of 1981 and 1982 in the
Gallatin Valley ..........   21

4. Distribution of mule deer observations during 
the summers of 1981 and 1982 in the Gallatin
Valley ........................................... 22

5. Distribution of mule deer observatins during 
the winters of 1981 and 1982 in the Gallatin
Valley ........................................... 23

6. Distribution of marked deer relocations of both
species in the Gallatin Valley in 1981 and 1982.... 25

7. Monthly frequency of deer carcasses throughout
the year in the Gallatin Valley from 1976 to 1982.. 34

.8. Monthly proportion of yearling males among carcasses 
collected throughout the year in the Gallatin Valley 
from 1976 to 1982................................. 34

9. The diel activity patterns of radio-monitored 
white-tailed deer and mule deer in the Gallatin 
Valley during 1981 and 1982....................... 41

10. The efficiency of observing both species of 
deer from routes in the Gallatin Valley during
1981 and 1982..................................... 41

11. The efficiency of observing deer of both 
species during the summer and fall from routes 
driven in the Gallatin Valley during 1981
and 1982..........................................  43



xiii

12. The efficiency of observing deer of both species 
deer during the winter and. spring from routes 
driven in the Gallatin Valley during 1981 and
1982...............................................   43

13. The diel activity patterns of radio-monitored 
white-tailed deer during the summer and fall
in the Gallatin Valley during 1981 and 1982....... 44

14. The diel activity patterns of radio-monitored 
white-tailed deer during the winter and spring
in the Gallatin Valley during 1981 and 1982....... 44

15. The diel activity patterns of radio-monitored 
white-tailed deer during semi-lunar periods
in the Gallatin Valley during 1981 and 1982........  45

16. The average number of deer of both species 
seen per plot per survey in relation to the 
number of houses within 900 m (.564 mile 
radius), in the Gallatin Valley during 1981
and 1982, assuming a cover type preference of 1.0.. 48

17. A map of marked deer locations in the Reese Creek 
area and the distribution of surrounding houses
and roads......................................... 82

18. A map of marked deer locations in the Pine Butte 
area and the distribution of surrounding houses
and roads......................................... 83

19. A map of marked deer locations in the Penwell area 
and the distribution of surrounding houses and
roads............................................. 84

20. A map of marked deer locations in the Sourdough 
area and the distribution of surrounding houses
and roads.......................................... 85

21. A map of marked deer locations in the Valley center 
area and the distribution of surrounding houses 
and roads

LIST OF FIGURES— Continued

Figure Page

86



xiv

ABSTRACT

Investigations of white-tailed deer and mule deer were conducted 
from the spring of 1981 through the summer of 1982 on over 1000 km2 
in the Gallatin Valley, Montana. The Gallatin Valley is a primarily 
agricultural high mountain valley with many streams flowing through 
it. White-tailed deer, quite scarce 10 years ago, currently comprise 
over half of the deer population. In recent years considerable 
housing development has occurred. An inverse relationship existed 
between housing density and deer observed. Jhis relationship appeared 
to have a threshold around 4-8 houses per km . Deer use was greatest 
on the moderately arable soils while housing development occurred on 
the more arable soils. 'Deer use peaked on land with intermediate 
densities of houses indicating that development is occurring on deer 
habitat while marginal deer habitat is left undeveloped. Deer were 
less likely to be active during midday if they were near housing 
developments. White-tailed deer were more secretive, more nocturnal, 
had greater reproductive capabilities, a shorter life expectancy and 
appeared better adapted to human activities than mule deer. Repro­
ductive data collected from carcasses yielded 1.62 fetuses per doe 
for white tails and 1.23 for mule deer. In conjunction with fawn-doe 
ratios an estimated fawn mortality of 40% was calculated for the 
first 90 days in white-tailed deer and 54% for mule deer. The 
presence of hybrids was also noted.



I

INTRODUCTION

Between 1963 and 1973 there was a 0.65 million hectare (ha) 
decrease in the amount of agricultural land in Montana. Sixteen 
percent of this was removed in 1972 - 1973 alone. Suburban tract 
acres represented 24% of this decrease (Hayden 1975). In recent 
years the Gallatin Valley has undergone many of these changes. The 
number of farms has decreased as has the total acreage of farmland. 
The rural segment of the population has had a substantial increase. 
Most of this increase has been accommodated on small acre tracts.

Pac and Mackie (1981) mention a "pronounced shift from urban areas to 

the rural countryside surrounding them"£ and further state that this 
was "especially true for cities situated in western Montana mountain 
valleys, such as Bozeman".

Hayden (1975) states "An intensive study of the relationship of 
white-tailed deer (Odocoileus virginiahus) to land development in 

western Montana would provide valuable information to land use plan­
ners, biologists, and others". Mackie and Pac (1980) found that 
there is a lack of quantification of the effects of subdivisions on 
deer and more detailed information is required. According to long 
term residents, white-tailed deer were scarce in the Gallatin Valley 

several years ago but now thrive in what was previously considered to 
be mule deer (Ch hemioniis) habitat.



2

In the spring of 1981 field work was initiated to identify and
relate the basic biological and ecological parameters of an increasing

!
deer population to habitiat, environmental characteristics and specific 
potential limiting factors such as farming practices, rural housing 
developments and associated human activities. Distribution patterns 
of different segments of the deer population were investigated to 
further develop and test hypotheses relating to deer habitat inter­
actions and population regulation. An attempt was made to reveal 
mechanisms of the encroachment by whitetails into the Valley and to 
determine when this will become self-limiting. This study should help 
quantify the effects of housing density and develop guidelines for 
management of mountain valley agricultural habitats. This project 
was funded by the Montana Department of Fish, Wildlife and Parks 
through the Pittman-Robertson project W-120-R-13 and 14 and the 
Montana Agricultural Experiment Station.



3

DESCRIPTION OF STUDY AREA 

Topography

The study area, an intemontane basin in Gallatin County in
2southwest Montana, covered over I', 000 square kilometers (km ) of the 

Gallatin Valley (Figure I). It is part of the Three Forks structural 
basin which was formed in pre^Oligocene times. The depression of the 
Valley continued in the Oligocene and Miocene. While the basin was 
being formed interruption of its drainage resulted in the creation of 
a lake. During the Tertiary an enormous deposition of fine wind borne 
volcanic dust fell on the lake and surrounding land. The Valley 
eventually became laden with 1,200 meters (m) of loess and with sand, 
silti and clay eroded from higher land. Most of these sediments were 
fine grained and the majority of today's soils are derived from them. 
After the formation of an outlet for the impounded waters, extensive 
surface erosion and cutting by streams occurred. In Quarternary 
times a mantle of alluvium was deposited over a considerable portion 
of the Valley (DeYoung and Smith 1936, Hackett 1960).

The soils of the main Valley are well developed, high in organic 
matter and primarily alluvial and eolian in nature. All the soils in 
the Valley appear to have a calcium carbonate layer within the soil 
profile. Some areas (Figure I), such as those along the East Gallatin



4

river north of Central Park and along Reese Creek, are poorly drained 
a gravelly, well-drained area surrounds Belgrade (Figure I). The 
soils of the Valley rim are less well developed.

LEGEND

----  Peved Rood

Stream

Town

M  I km

Figure I. A map of the study area in the Gallatin Valley, Montana



5

Elevations in the study area ranged from 1283 m to 1985 m. The 
West Gallatin River drains the Gallatin and Madison ranges, from the 
northwest corner of Yellowstone National Park to its entrance into 
the Valley near Gallatin Gateway. It then continues along the western 
edge of the Valley until it is joined by the East Gallatin River 
forming a sharp cliff where, at their confluence, they impinge upon 
the Horseshoe Hills. The Gallatin River leaves the Valley through a 
narrow gorge in its northwest corner. Three miles downstream it 
connects with the Madison and Jefferson Rivers to form the Missouri. 
The East Gallatin River is the main tributary of the West Gallatin, 
draining the east and south sides of the Valley.

In July of 1805, Lewis and Clark described, "a distant range of 
lofty mountains running their snowclad tops above the irregular and 
broken mountains which lie adjacent to this beautiful spot"
(Burlingame 1964). These ranges include the Gallatin Range, to the 

south, which averages 2,700 to 3030 m in maximum elevation, and the 
Bridger Range, to the northeast, which is linear and its crest trends 
north at an average of 2550 to 2750 m. The Horseshoe Hills, to the 
north, are a series of northeast trending ridges that rise about 300 
m above the Valley floor. Steeply sloping coalescing, alluvial fans 
from the Gallatin and Bridger ranges form the south and east sides of 
the Valley (Hackett I960).

History

In July of 1805, Meriwether Lewis - after climbing the high 
cliff which rises along the north bank of the Gallatin at Three



6
Forks - described the Gallatin as "passing through a smooth extensive 
green meadow of fine grass" and "meandering in several streams". He 
also observed that a "high, wide, and extensive plain succeeds the 
meadow and extends back several miles from the river" (Burlingame 
1964). On July 13 of 1806, William Clark stated that he "started up 
a butiful navigable stream (The Gallatin), saw a large gangue of elk 
in the plains and deer in the river bottoms". He further noted 
seeing deer several times and made reference to "common deer", which 
were probably white-tailed deer. On July 14 they stopped about 3 
miles east of the present site of Bozeman and made mention of seeing 
elk (Cervus elaphus), deer, antelope (Antilocapra americana), and a 
great deal of buffalo (Bison bison) sign (Burlingame 1976). Passing 
through the Valley in 1863, J. Stewart noted seeing both white-tailed 
deer and mule deer (Koch 1941).

The major settlement in the Valley, Bozeman, was established as 
a town in 1864. The agricultural history of the Valley began soon 
after the opening of the mines near Virginia City, but it was not 
until 1885, when the Northern Pacific Railway reached Bozeman, that 
rapid development of the agricultural lands of the Valley began 
(DeYoung and Smith 1936). Since 1960 the amount of irrigated land in 
the county has increased by 35% while the amount of non-irrigated 
farm land has decreased (Table I). Since 1950, winter wheat has 
shown a steady decrease, spring wheat has varied and barley has made 
a steady increase in acres harvested, .as has irrigated alfalfa hay.

During the interval 1950 - 1980 the number of residents in the 
county has increased 196% (Table 2). Although the total number of



7

residents has steadily increased, the most noticable change is the 
53.4% increase in rural residents from 1970 to 1980. The increase in 
the number of rural residents concurrent with a decrease in the 
number of farms indicates that most of these people are dwelling on 
small acre tracts (Table 3). In March of 1973 there were 6,307 ha of 
subdivisions in Gallatin County and had by the summer of 1974 in­
creased to 8,100 ha (Hayden 1975). From 1950 to 1970 the decrease in 
the numbers of farms and the increase in size resulted in the mainte­
nance of the same total acreage of farmland. Since 1970 it appears 
that the total acreage of farmland has been decreasing.

Table I. Land use in acres harvested in Gallatin 
Cdunty from 1950 to 1979. 1

Crop Type 1950 1960 .1970 1979
Nonirrigated 96,709 99,300 79,500
Irrigated 76,933 66,521 73,400 89,800
Winter wheat 68,500 49,400 47,000 28,800
Spring wheat 25,400 11,600 4,800 14,300
Barley 28,400 30,100 38,000 45,800
All hay 57*900 61,700 78,100 .74,100
Irrigated
Alfalfa 21,900 29,300 34,400 35,000

# Frost-free
Days
T— Efrrrrrr-Tr-T:

69 120 116
1 Montana Agricultural Statistics (1950 - 1981)



8

Table 2. Population of Gallatin County from 1950 to 1980 1

Year
Total
Population

Percent
Change

Rural
Population

Percent
Change

1950 21,902 ----- 10,577
1960 26,045 18.9 12,684 19.9
1970 32,505 24.8 13,835 9.1
1980 42,865 31.9 21,220 53.4
1 U.S. Bureau of the Census (1950 - 1980)

Table 3. The changes in farms and farm size in Gallatin County
1950 1970 1978

Number of farms 1,129 831 759
Percent of land 58.3 59.0 55.0
Mean farm size 2 832 1,144 1,170
Total land 2 938,714 950,857 888,246

T Montana Agricultural Statistics (1950 - 1981) 
2 Acres

Climate

The climate of the Valley is continental in nature, characterized 
by extremes in daily and seasonal temperatures, with long cold winters 
and short cool summers. Killing frosts are unpredictable; The 
average frost free period in Bozeman is 114 days. The average annual 

temperature at Montana State University is 5.2 °C. During the winter, 
warm "Chinook winds" are of variable occurrence, at times causing the . 
direct evaporation of snow (DeYoung and Smith 1936).

The distribution of precipitation in the Valley is uneven. At 
the eastern and southern boundaries an increase in elevation as well 
as the nearness of the mountains causes ah increase in annual precipi­



9

tation. The character of the vegetation and color of the surface 
soils indicate a considerable decrease in precipitation to the north 
and west of Bozeman. The average annual precipitation at Bozeman is 
46.6 cm, much of which comes in the form of snow. The 11 year mean 
winter snowfall was 223 cm (Pac et al. 1982). May and June are 
normally the months of heaviest rainfall.

The winter of 1980 - 1981 was a mild winter, which fell outside 
the biometeorological normal range (Table 4). The winter of 1981-1982 
within the biometeorological normal, had a large amount of snowfall 
but a lower than average maximum accumulation on the ground, perhaps 
due to the Chinooks experienced between snowfalls. The relationship 
between snow on the ground and elevation is presented in Table 5.

Table 4. Climatological data from the Montana State University 
weather station at Bozeman, MT for November through 
May, 1971-1982 I

Mean Temp. 
Year °F

2
Ppt. 
2

Snowfall
in.
2

Max. Snow 
Depth 
in. 2

Days or 
Ground 
2

t Severity
Index 
2 3

71/72 33.9 7.7* 81.8 10 119 - 329
72/73 31.3 8.4* 92.6 14 135 +1335
73/74 34.3 10.0 89.9 14 116 +2025
74/75 30.5* 12.1* 118.6* 17* 161* +2286
75/76 35.2 9.9 95.2 12 86* - 810
76/77 37.4* 7.9* 57.8* 9* 88* -2391*
77/78 33.9 10.0 60.5* 12 111 + 790
78/79 29.9* 10.7 87.4 18* 159* +7843*
79/80 33.3 10.9 100.7 20* 144 + 752
80/81 37.6* 12.9* 55.6* 8* 74* -2871*
81/82 32.8 11.5 125.3* 11 129 + 818
Mean 33.6 10.2 87.8 13.2 120 + 859
S 2.51 1.68 23.07 3.84 29.12 2847.61
1 (Pac et al. 1982)
2 * Indicates that values differ from 11 year mean by at least I

standard deviation and thus are outside the biometeorological 
normal.

(Picton and Knight 1971) positive values indicate greater severity3



10

Table 5. Distribution of snow cover in the Gallatin Valley from 
October 1981 to April 1982 I

Weather.
Station

Maximum . 
Snow Depth

Average of the Monthly 
Maximum Snow Depth

Elevation

Belgrade
Airport 23.0 cm 13.0 cm 1358 m
Bozeman 6 W
Experiment
Farm 25.0 cm 16.8 cm 1456 m
Bozeman 
Mt. St. Univ. 28.0 cm 21.1 cm 1481 m
I Montana Climatological Data (1981-1982)

Vegetation

The distribution of plant life in the study area is complex due 
to the variation in terrain and climate and is further complicated by 
an interspersion of agricultural fields. These fields consist mainly 
of hay and small grain and irrigated pastures are common with dairies 
being a major land use. Plant nomenclature is according to Hitchcock 

and Cronquist (1973).
The definitions of 17 cover types are:
1. Cottonwood type. Along most banks of major rivers and 

streams, cottonwood (Populus trichocarpa) commonly formed a 
canopy. Rocky Mountain juniper (Juniperus scopulorum) 
could sometimes be found in this type.

2. Aspen type. Aspen (Populus tremuloides) was generally 
found along some smaller streams and wet areas.

3. Deciduous forest type. Deciduous forest consisted of Alnus 
spp. and Populus spp.; and was found along rivers or on a 
slope immediately below and adjacent to the conifer type.

4. Conifer forest type. The conifer type was found at higher 
elevations around the rim of the valley and consisted 
mainly of Douglas fir (Pseudotsuga menziesii) and lodgepole 
pine (Pinus contorta).

I



11

5. Mixed forest type. A mixture of the conifer and deciduous 
forest types was referred to as mixed forest.

6. Willow type. Willow (Salix spp.) stands usually occupied 
areas along rivers, streams, and low wet parcels of land.

7. Mixed brush type. A mixed brush type often bordered old 
ditches, streams, canal banks, and draw bottoms ■, and con­
tained shrubs such as serviceberry (Amelanchier spp.), 
dogwood (Cornua spp), hawthorn (Craetugus spp.); chokecherry 
(Primus spp:), current (Kibes spp.), rose (Rosa spp.), and 
snowberry (Symphoricarpos spp.). It sometimes had grasses

. such as basin wildrye (Elymus cinereus) and reed canary 
grass (Phalaris arundinacea) associated with it along 
roadside ditches.

8. Skuiikbush - Juniper type. On many rocky slopes of inter­
mediate to high elevations, such as those found in the 
northeast portion of the study area, the skunkbush sumac 
(Rhus trilobata) and Juniper type was common. Big sagebrush 
(Artemesia tridentata) antelope bitterbrush (Purshia 
tridentata) and limber pine (Firms flexilis) were often 
associated with the Rhus/Juniper type.

9. Range - Grassland type. According to DeYoung and Smith 
(1936), the native range/grassland type has felt the intro­
duction of many grasses such as Poa spp., and Phleum spp. 
Before the advent of settlement and agricultural practices 
such as irrigation, a moderate amount of silver sagebrush
(Artemesia cana) and an abundant growth of grass, carpeted 
the slopes, benches, and parts of the lower areas in the 
southern and eastern parts of the Valley. Idaho fescue 
.(Festuca idahoensis), blue-bunch wheatgrass (Agropyron 
spicatum), prairie junegrass (Koelaria cristata) and some 
western wheatgrass (Agropyron smithii) can still be found 
on unplowed hill sides. Big sagebrush originally covered 
most of the more arid uplands and still dominates on rough 
land. On the more arid areas to the west and northwest, 
drought resistant species such as needle and thread (Stipa 
comata), red three-awn (Aristida longiseta) thread-leaved 
sedge (Carex filifolia), blue-grama (Bouteloua gracilis) 
and rabbit brush (Chrysothamnus spp.) can be found. Salt 
grass (Distichlis spp.), alkaligrass (Puccinellia nuttaliaha) 
and cheatgrass brome (Bromus tectorum) grow on some of the 
alkali areas. Weeds such as thistle (Cirsium spp.) are 
also common in this type.

10. Marsh type. The marsh type occupied low lying wet areas 
with plants such as cattails (Typha spp.), rushes (Juncus 
spp.) and sedges (Carex spp.).



12

11. Riparian type. The term riparian type was applied to 
bodies of water and their immediate surroundings such as 
gravel bars.

12. Hay type. Hay was used to describe ’wild hay' consisting 
of orchard grass (Dactylis glomerata) and brome grass
(Bromuis spp.) and that seldomly appeared to have a legume 
accompanying it.

13. Alfalfa type. Alfalfa fields composed of Medicago sativa 
or Ml falcata were occasionally grown with a companion crop 
of small grain during the year of establishment. This type 
was often irrigated or sub-irrigated.

14. Small grain type. The small grain type was predominantly 
wheat (Triticum aestivum), rye (Secale cereale), barley 
(Hordeum vulgare), and oats (Ayena sativa). Wild mustard 
was commonly interspersed in wheat fields.

15. Potato type. Potato fields were common on the western edge 
of the study area.

16. Lawn type. The lawn type was generally composed of sod­
forming grasses such as Kentucky blue-grass (Poa pratensis); 
which was kept short by frequent mowing.

17. Non-vegetative type. Non-vegetative habitat refered to 
buildings and the spaces between them such as sidewalks, 
streets, alleyways and parking lots.

Large Mammals

Nomenclature of mammals is according to Hoffmann and Pattie 

(1968). The Valley contains a substantial number of resident white­
tailed deer and mule deer. Mackie and Pac (1980) have delineated 
several mule deer winter ranges along the west slope of the Bridger 
range. Pronghorn can be observed fairly regularly in the northeast 
region of the study area. Moose (Alces aIces) and elk are often seen 
near the edges, especially the southern edge, of the Valley during 
the winter. In the past moose and elk used areas further into the



13

Valley according to long term residents. Carnivores such as black 
bear (Ursus americanus), mountain lion (Felis concolor), and coyotes 
(Canis latrans) were also occasionally observed during the study.

Domestic livestock consisted mainly of cattle, horses and sheep. 
The Valley also has a substantial population of domestic dogs.



14

METHODS

Deer were classified to sex, species and age class when possible. 
Locations were recorded using the Universal Transverse Mercator (UTM) 
coordinate system and time was recorded in ^military time. Habitat 
characteristics such as cover type, slope, exposure, and distance to 
concealment cover were recorded. Concealment cover was considered as 
any amount of vegetation or geomorphic structure that concealed 50% 
of a deer at 50 m. The activity that the animal was involved in, 
interspecific interactions and any special behavioral characteristics 
were also recorded< Environmental conditions, such as the phase of 
the moon, lunar illumination, temperature, wind velocity, cloud 
cover, ground cover and type of precipitation were noted. Parameters 
such as elevationj soil type, distance to roads, highways, towns, and 
the number of farm houses and non-farm houses in the vicinity along 
with the overall level of development were generated from maps. A 
summary map was constructed from investigations bn the ground during 
the study.

White-tailed deer were trapped in 1981 in 2 helicopter drive net 
operations. In 1982, rocket nets and Clover traps were used to 
capture both species of deer. AVM radio beacons and Telonics motion 
sensing collars were placed oil adult does while individually recog­
nizable collars were placed oh other deer.



15
Instrumented deer were monitored for 24 hour (hr.) periods using 

an omnidirectional antenna. The activity level of the deer, moving 
or still, was estimated by the variability in the strength of the 
signal during 5 minute sampling periods. The observer would move 
from one location to another throughout the night, monitoring dif­
ferent deer, in conjunction with other measurement activities. A 
group of I to 4 radios was monitored at least every 2 hrs. Happe 
(1982) reported that 0, 1-11, and 11+ changes in strength per 50 
pulses indicated bedding, feeding, and travelling, respectively. In 
addition the activity level of the 10 deer wearing the motion sensing 

collars could also be estimated using the frequency of change between 
the 2 modes of pulsing: fast and slow. Estimation of activity level 
by telemetry was verified by both night and day time observations of 
instrumented deer while simultaneously monitoring their signal. 
Locations were obtained by triangulation with a 3 element Yagi, 
hand-held antenna.

Instrumented deer were located during 44 flights made at 10 to 
14 day intervals or as the weather permitted. Intensive surveys, 
each requiring 2 or 3 mornings of flying, were flown in the spring, 
late summer, and early winter. . Summer was considered to be the 
months of July and August, fall, September and October, and spring, 
April through June. The season of winter contained the remaining 5 
months, November through March.

Six routes, consisting of 405 km of road, were driven at 40 to 
48 km/hr. at purposely varying times, dates, and in alternate direc­
tions through the course of the study. After a sighting had been



16

made, the vehicle was brought to a stop and more detailed information 
was obtained with 10 x 50 binoculars and a 15 to 60 power spotting 
scope. Routes were driven on 117 occasions.

Twenty-one plots, consisting of an open field bordered by cover, 
were monitored by day and by night to detect any diel differences in 
habitat usage. These plots were observed with a spotlight equipped 
with a red filter and binoculars during the hours of darkness. Each 
plot was checked 25 times. A passive night vision device (Javelin 
221) was available if needed.

A grid of 272 control points based on the DTM coordinate system 
covering the entire study area was used for comparison of deer obser­
vations made during flights.

Another series of 626 control points was placed along vehicle 
routes for comparison of deer observations made while driving routes. 
All routes were divided into 1.6 km intervals. A random digit, 
generated from a random number table, determined the distance in 
divisions of 0.16 km (one tenth of a mile) along the road where the 2 
control points were placed. The distance to which a standing deer 

could be seen to each side of the road determined the boundary of the 
corridor. The distance into the corridor that the control point was 
placed was also determined by the random digit. The outside of all 
turns of 90° degrees or more were assigned a control point to offset 
the 0% chance of obtaining a control point through the above method. 
The SPSS and BMDP statistical packages were used for much of the 
analysis.



17

Deer carcasses were collected by the author during the study and 
by personnel of the Montana Department of Fish, Wildlife, and Parks 
Research Bureau in Bozeman during the 5 years prior to the study. 
Necropsies were performed on all deer to the extent allowed by damage 
or theft of parts.

The lower jaw was removed from adult deer to determine their age 
(Severinghaus 1949, Robinette et al. 1957). Rumen samples were 
collected for analysis of food habits (Saunders 1955). Condition was 
estimated from 3 parameters: the amount of fat on the brisket, the 
amount of intestinal fat, and the color and texture of the marrow in 
the femur. Females were examined for reproductive activity and the 
sex, weight, and crown-rump length were determined for all fetuses.



18

RESULTS

Trapping

In late March and early April of 1981, 16 white-tailed deer were 
trapped in 2 separate helicopter drive net operations. During the 
winter 1981 - 1982, 21 white-tailed ahd 9 mule deer Were trapped 
using Clover traps and rocket nets. The best results in using Clover 
traps and rocket nets were obtained during periods of extreme cold and 
deep snow. Sixteen adult does were equipped with radio beacons, 6 
with A.V.M. radio beacons and 10 with Telonics motion sensing trans­
mitters. . Four of the 10 were mule deer. Twenty-nine other deer were 
fitted with individually recognizable collars.

No mortality was experienced during the trapping operations.
All deer that were marked were relocated at least once. For a des­
cription of all marked animals see appendix Table 28: Two radio
equipped does and 3 collared males were killed during the course of 
the study (Table 6). All remaining radio-equipped deer were known to 
be alive at the end of the study.

Table 6. The location, date and cause of death of 5 marked deer 
in the Gallatin Valley during 1981 and 1982.

Deer
Number

Location Date Cause Sex

0015 Fine Butte 11/ /81 Wounded M
0032 . Sourdough 10/12/82 Vandalized F
0043 Valley Center 10/27/82 Harvested M
0037 Sourdough 11/22/82 Harvested M
0046 Bozeman 12/14/82 Vehicle Collision F



19

Movements

About two-thirds of the deer observed in flights and routes 
combined were whitetails. Mule deer were abundant in many locations 
such as around the edges of the Valley and in the center of the 
Valley southeast of Belgrade (Figures 2, 3, 4 and 5). Mule deer from 
the surrounding mountains frequently moved down into the Valley, 
especially along ridges and slopes, early in the spring. White-tailed 
deer movements, with the onset of winter, varied with their location. 
In some areas they did not change their distribution on a large scale 
while in others they made some major adjustments. In some areas, 

like Pine Butte and Sourdough, there were altitudinal movements which 
decreased the surrounding disturbances, increased conifer cover and, 
especially by Pine Butte, provided forage on small grain fields kept 

snow free by wind. In another area, near Reese Creek, unmarked white- 
tails seemed to move down in elevation as the snow accumulated.



20

LlOIND

Deer Observation

I km

Figure 2. Distribution of white-tailed deer observations during
the summers of 1981 and 1982 in the Gallatin Valley.



21

IE C iN O

D » *r  O b servation

I km

Figure 3. Distribution of white-tailed deer observations during
the winters of 1981 and 1982 in the Gallatin Valley.



Figure

KOEND

D *«r Observetion

I km

4. Distribution of mule deer observations during the summers
of 1981 and 1982 in the Gallatin Valley.



23

LlOlND

Deer Observelion

I km

Figure 5. Distribution of mule deer observations during the winters
of 1981 and 1982 in the Gallatin Valley.



24
Reese Creek

Three radio and 4 visual collars were put on whitetail does in 
this area (Figure 6). Movements were restricted mostly to Reese 
Creek and Bear, Hays, and North Cottonwood Creeks to the north.
There were few disturbances to the deer in these areas. During the 
fawning season the females appeared to concentrate in North Cotton^ 
wood and upper Bear Creek.

During the early winter of 1981-1982 many deer spent the days in 
Hays Creek which is free of roads and houses, and their nights in 
Bear Creek near a barley field that had been damaged by hail and 
never harvested properly* Once the snow became deep, deer discon­
tinued their use of this field.

In early June of 1981 a radioed deer and her marked yearling 
travelled 11 km up into Dry ;Creek. Another radio-collared deer was 

located in February at the junction of the East and West Gallatin 
Rivers, over 16 km froin where it was usually found. It returned to
its biocenter by the next night. Polygon home range areas for this

2herd were the largest of all the herds, averaging 36.3 km (Table 7 
and Appendix Figure 17).



25

' i : *ee«e Creek

LEGEND

TOWN

PAVED ROAD

MARKED DEER 
RELOCATION

Figure 6. Distribution of marked deer relocations of both species 
in the Gallatin Valley in 1981 and 1982.



26

Table 7. Home ranges and movements of radio-collared deer of both 
species in the Gallatin Valley in 1981 and 1982.

Deer Age1 Polygon Maximum Mean Shape
Number Home Distance Distance Index of

Range Between Between Polygon2
Area Fixes Fixes
km2 km km

REESE CREEK - White-tailed deer
01 2.5* 20.50 4.81 1.05 4.2
02 4.5* 41.53 15.85 1.37 4.9
03 3.5* 46.95 11.08 1.30 7.4
PINE BUTTE - White-tailed deer
08 2.5* 14.89 7.10 1.41 1.3
09 ■ 2.5* 9.22 5.22 0.31 3.7
10 3.5* 6.44 3.85 0.36 2.5
PENWELL - White-tailed deer
26 1.5 3.1 2.4 0.64 1.3
30 2.5 2.3 2.02 0.77 0.8
35 1.5 1.94 1.77 0.77 0.8
SOURDOUGH - White-tailed deer
28 3.5 1.17 1.90 0.61 0.4
32 1.5 0.31 0.73 0.28 0.3
46 1.5 8.76 7.59 1.55 0.7
VALLEY CENTER - Mule deer
17 2.5 3.51 2.92 0.35 2.1
21 2.5 10.11 4.24 0.29 2.0
22 8.5 1.52 2.00 0.39 0.8
45 3.5 0.82 1.5 0.29 0.5

* Ages determined in Spring of 1981.
All other ages determined during the winter of 1982.
The shape index was derived by dividing the polygon home range 
by the standard diameter; smaller values are correlated with 
a more linear shape.



27
Pine Butte

Nine whitetails were marked in this area (Figure 6). The 3
-

radio-equipped deer showed distinct winter and summer ranges. All . 
three wintered on the benches above and to the west of the West 
Gallatin River where they fed in small grain fields and relocations 
were well scattered. Two radio-equipped deer summered on a small 
area of the river; they fed mainly in an alfalfa field by night and 
stayed on I of 2 islands during the day. Relocations of these animals 
were very concentrated around their activity centers. The other 
radio-equipped Whitehall spent the summer on a small area along 
Middle creek.

A heck-banded female fawn spent the winter of 1981-1982 south of
Pine Butte and west of Gallatin Gateway as a yearling and spent the
following summer where Middle Creek is crossed by Cottonwood Road.
Several other young marked deer were observed as far as the base of
Mount Ellis on the east side of Sourdough Creek, 22 km from Pine
Butte. One wounded male traveled 23 km to the northeast, returned
and died within a month. Polygon home ranges were also large for

2this area averaging 10.2 km (Table 7 and Appendix Figure 18).

Penwell

There were 11 marked whitetails in this area (Figure 6). Two 
of the radios were on year round residents whereas the third could 
not be relocated from May 1982 until January of 1983, when it was 
relocated slightly over I km from the place, the deer was originally 
trapped. This led me to believe that it had spent the summer and



28

fall outside the valley. A marked female fawn moved to North Cotton­
wood Creek in the spring of 1982 while what was assumed to be its
mother remained in the Penwell area. Polygon home ranges averaged 

22.5 km for this area (Table 7 and Appendix Figure 19).

Sourdough

There were 10 marked animals in this area (Figure 6). There
were large differences in the size of the home ranges of the 3 radioed

odeer (0.31, 1.17, 8.76 km ) and the habitats included within those
home ranges. One spent all winter and summer up on the slope on the
west side of Sourdough Creek in a mixed forest until it was shot.
One moved only 1.6 km north into the Valley and spent all summer

along Sourdough Creek. The third moved to the outskirts of Bozeman in
early summer and remained there with I exception until it was killed
on Interstate 90. During the hunting season it moved back to within
2 km of the slope only to return the 10 km to Bozeman. A young
marked male was also found on the same area near Bozeman.

These deer appeared to concentrate on a small area during a
normal winter (1981-1982) and spread south along Sourdough Creek
during the summer. During the following winter these deer were
scarce if not absent from the area where they spent the previous
winter. Movements were mainly restricted to the Sourdough Creek
drainage and a draw east of Bozeman. Polygon home ranges averaged 

23.4 km (Table 7 and Appendix Figure 20).



29
Valley Center

A population of mule deer exists along Middle Creek in the 
center of the Valley. This area is occupied almost exclusively by 
mule deer during the winter and they become concentrated on small 
areas in rather large groups. Four radios and 5 other collars were 
placed on deer in this area (Figure 6). Three radioed deer remained 
extremely close to the trap site. One of the youngest radio-equipped 
deer, a 2 year old, moved 4 km south-east in early June, where it 
spent the summer and fall. Its winter 1982 - 1983 fixes were inter­
mediate between those of the summer 1982 and the winter of 1981 - 
1982.

Males trapped in this area had larger movements than did the
females. No males were fitted with transmitters and thus relocations
were limited. Polygon home range areas for radio-collared does in

2this area averaged 4.0 km (Table 7 and Appendix Figure 21).

Population Structure

A greater percentage of males and fawns were observed in an 
intensive 3 day aerial survey in late August 1981 than during 
flights throughout the year (Tables 8 and 9).

Relatively fewer males were seen along routes than from the air 
(Table 10). Of the deer seen during flights in the summer and fall 
of 1981 and 1982, 16.6% of the mule deer and 17.3% of the white-tailed
deer were males.



30

During the study the author picked up 65 deer carcasses from 
area roads. In the 5 years prior to the study 144 carcasses were 
picked up by personnel of the Montana Department of Fish, Wildlife, 
and Parks Research Bureau in Bozeman. Some of these carcasses were 
collected immediately outside the study area. Ninety percent of the 
deer collected within the study area were road kills. Mule deer were 
proportionately more abundant in the carcasses collected from area 
roads than from any method of sampling. This could be.a function of 
either a greater proportion of deer being mule deer in the past, mule 
deer in surrounding areas being included, or increased vulnerability 
of mule deer. Fawns were also abundant among these carcasses (Table 
11). This is probably due to a lack of experience rather than a 
decrease in production over the years.

Table 8. Sex and age proportions of deer in the Gallatin 
Valley in late August 1981 from a 3 day 
aerial survey, in percent of classifiable deer.

Species Males
%

Females
%

Fawns
%

Species Fraction 
%

Mule deer 
(n=69)

23 49 28 19
White-tailed deer 
(n=293)

16 42 41 81



31

Table 9. Sex and age proportions of deer in the Gallatin Valley 
in 1981 and 1982 derived from flights during all 
seasons, in percent of classifiable deer.

Species Males
%

Females
%

Fawns
I

Species Fraction 
%

Mule deer 
(n=603)

15 72 13 36
White-tailed deer 
(n=1059)

12 50 38 64

Table 10. Sex and age proportions of deer in the Gallatin Valley 
derived from observations along routes during 1981 
and 1982, in percent of classifiable deer.

Species Males Females Fawns Species Fraction
% I % %

Mule deer 
(n=92)

7 56 37 18
White-tailed deer 4
(n=427)

50 46 82

Table 11. Sex and age proportions of deer 
carcasses collected from area roads in the 
Gallatin Valley from 1976 to 1982.

Species Males Females Fawns Species Fraction
% 7» % %

Mule deer 15 50 35 45
(n=66)
White-tailed deer 23 32 44 55
(n=81)



32

Table 12. The ratio of young per adult female for white-tailed 
deer seen during flights made throughout the year in 
the Gallatin Valley during 1981 arid 1982.

Season Year Fawns/Adult Doe
Summer 1981 0.93
Fall 1981 0.83
Winter 1981-1982

\
0.80

Since fawn:doe ratios found in this study were not lower than 
previously, recruitment does not appear to be depressing the popula­
tion growth. In August of 1977 a flight survey yielded fawn:doe 
ratios of 0.82 and 0.54 for white-tails and mule deer respectively 
(Picton unpubI. data). The white-tailed deer fawn:doe ratios decreased 
from the late summer through the remainder of the year (Table 12). 
Early in the summer fawns are small and secretive and therefore 
difficult to see from the air. Observed fawn:doe ratios early in the 

summer are low and rise to a peak in late summer. Early mortality is 

thus extremely difficult to measure.
The productivity of the population was modeled using data from 

necropsied deer and fawn:doe ratios from flights (Table 13). The 
estimated fawn production in June was 1.62 fawns per whitetail doe 
and 1.23 fawns per mule deer doe. This leads one to believe that 
considerable mortality could be occurring between June 14 and the end 
of August. The expected number of fawns in June derived from the 
model was 1.62 per whitetail doe and the observed number in a 3 day 
August aerial survey was 0.98 fawns per doe which indicated a 40% 
mortality of fawns through the first 90 days of life. The expected 
number of mule deer fawns, 1.23, per doe and the observed in the



33

August flight survey, 0.57, indicated a 54% fawn mortality during 
this time period for mule deer. Deer in the Valley appear to be on a 
steady increase in spite of this mortality. Fawns were much more 
abundant in the sample of carcasses than in the year round population 
itself. The fawn:doe ratios for carcasses were 138.5 and 69.7 for 
whitetails and mule deer respectively. Males were also abundant in 
the sample of carcasses. Forty-one percent of 27 mule deer fetuses 
and 50% of the 12 white-tailed deer fetuses collected from carcasses 
were males; Males comprised 23% of mule deer adults and 42% of white 
tail adults and 57 and 67 % of the fawns for mule deer and white- 
tails. In 1981 and 1982, 46% of the yearling males collected during 
those 2 years were collected during 4 weeks in late May and early 
June (Figures 7 and 8). A similar trend was found by Puglisi et al. 
(1974).

Table 13. Productivity of female deer of different ages from counts 
of fetuses in carcasses collected between January I and 
June 14 in the Gallatin Valley between 1976 and 1982.
White-•tailed deer Mule deer

Age 1 Percent Z Fetuses 3 Percent 3 Fetuses 6
I 29 * 0.2 (n= 6) 20 * 0.2 (n= 4)
2 38 (n=l4) 2.2 * 27 (n=ll) 0.8 (n= 6)
3 . 22 (n= 8) 2.2 * 15 (n= 6) 1.3 (n= 4)
4+' ' 11 Xn= 4) 2.2 * 39 (n=16) 2.0 (n= 10)
1 Age for the following June.
2 ■* Percent of females that were fawns was estimated from flight

data and not from road kill structure.
3 * Estimates were combined due to a combined sample size in 3

age classes of 4 females. Value reported for age I mule deer 
is the value actually observed.
n = the number of females examined for reproductive activity.



PE
RC
EN
T 
OF
 C
AR
CA
SS
ES

34

5
O
30

10.

■
■

■ ■
■

■ ■
■

■

#

■

JAN FlB MAR APR MAY JUN JUi AUO UP OCT NOV DK

Figure 7. Monthly frequency of deer carcasses throughout the year 
in the Gallatin Valley from 1976 to 1982.

JAN FIB MAR APR MAY JUN JUL AUO SiP OCT NOV M C

Figure 8. Monthly proportion of yearling males among carcasses 
collected throughout the year in Gallatin Valley from 
1976 to 1982.



35

Interrelationships

Mule deer and white-tailed deer were not often seen together and 
areas of use by the 2 species sometimes appeared to be separated by 
seemingly non-existent barriers. In I case a small creek separated 
the 2 species of deer and in another case it was a road.

Hybrid carcasses (Wishart 1980) were picked up from roads and 
analyzed. Two trial funs of protein electrophoresis were made. 
Ambiguous results were obtained which was not suprising in view of 
the length of time that some of these animals were dead before col­
lection and the close relationship of the 2 species (Baccus et.al. 
1983). In one of these runs the whitetail samples appeared unique 
in comparison with the hybrid and the mule deer samples. On 14 of 15 
occasions, potential hybrids seen in the field were in the company of 
whitetail does. This is in direct contrast with Wishart (1980) 
who stated that hybrids are usually borne by mule deer since white- 

tail does are too elusive for mule deer males. Wishart (1980) states 
that hybrids are fertile and their offspring are viable.

Dogs were observed chasing deer on 2 occasions; once from the 
air in the morning and once at night. No direct mortality was docu­
mented during the study although Nyberg (1980) documented direct loss 

of deer due to dogs and Gavitt et. al. (1974) postulated that some 
dogs could become proficient at taking fawns. No coyotes were ob­
served chasing deer during the study although they were often seen 
close together. Bear seen in the vicinity of deer were indifferent



36

to their presence. A mountain lion, observed from the air, was 
stalking a radio-collared whitetail female. The deer was aware of 
the cat's presence and merely avoided passing through the cover.

Food Habits

Diet was estimated from rumen samples collected from necropsied 
deer. Mule deer collected in mountainous environments surrounding 
the Valley were evaluated separately. There appeared to be as much 
variation between mule deer diets inside the Valley and those sur­
rounding the Valley as there was between mule deer and whitetail 
diets. Mule deer use of browse was greater than that by whitetails. 
Mule deer appeared to feed on forbs including alfalfa slightly more 
than whitetails while whitetails fed more on grass, including small 
grain, than mule deer (Table 14). Deer use of both alfalfa and small 
grain was highest from September through March, the period when 
intake was also high. The composition of diet by month is given in 
Appendix Tables 29 and 30. Use of alfalfa was difficult to determine 
from rumen samples since it was so rapidly digested. Use of the 

leaves of both alfalfa and small grain were probably grossly underesti 
mated in these figures but if they were underestimated for all samples 
equally the seasonal patterns of use Should be accurate. The Pearson 
correlation coefficient for the volume of rumen as an indicator of 
intake and percent fruit, seed and grain was +.3712 with a P = .000.



37

Table 14. A comparison of the diets of mule deer and white-tailed 
deer in % volume of the rumen in and immediately around
the Gallatin Valley from 1976 to 1982.

Species and Browse Forb Grass Fruit Alfalfa Small
Location % % % % % Grain %
Mule deer 
(outside) 
(n=18)

58.2 11.9 28.3 0.1 2.2 2.6

Mule deer
(inside)
(n=44)

46.7 25.2 25.6 1.6 5.9 2.2

Whitehall
(inside)
(n=66)

38.2 16.2 38.5 5.3 \ 2.9 9.5

Use of Cover Types

Use of the various cover types was related to the amount of 
associated cover as well as the number of houses in and surrounding 
these cover types (Appendix Table 31). Cover types with more associ­
ated cover apparently received greater use by deer. Some of the 
cover types were not equally available to both species of deer. 
Conifers were generally found along the edges of the Valley and were 

thus more available to mule deer. Aspen was used considerably by 

whitetail males especially in the Reese Creek area.
The size of fields also appeared to effect use. The centers of 

large fields would be less likely to be used as was the case in small 
grain fields. There were seasonal changes in the use of cover types 

(Table 15).



38

Although no mule deer were observed feeding in alfalfa fields 
along routes, mule deer were observed feeding in alfalfa extensively 
at other times and on alfalfa hay stacks as well (Appendix Table 31). 
This was a function of the location of the routes used in analysis of 
cover type use and the distribution of mule deer.

Use of Hay Stacks

Deer were only observed feeding on hay stacks during the winter, I
and spring (Table 16). Deer use of hay stacks increased with increas­
ing snow depth. Mule deer appeared to use stacks more and to feed 
more aggressively upon them than did whitetails. Use of stacks by 
white-tailed deer was greater at night than during the day.

The hay baling and stacking systems and the location of the 
stacks both were factors in the amount of depredation. Loose stacks 
without fencing were particularly vulnerable. Rectangular bales 

usually weigh around 20 or 25 kilograms (kg) while round bales average 
around 500 - 600 kg although smaller round bales and larger rec­
tangular hay bales can be found. Rectangular bales were usually 

stacked in large stacks and experienced less decay than if left 
scattered. Round bales, when left scattered in the field, were 
heavily used by deer. Deer appeared to use these round bales when 
bedded, probably for thermal cover. The surface area exposed to 
feeding deer was greater thus reducing agonistic behavior and allowing 
a greater number of deer to feed at the same time. Interspersion of 
types was greater under this system if hay bales are considered a



39

Table 15. The seasonal use of cover types by both species of deer in 
the Gallatin Valley in percent of deer seen in that season 
during 1981 and 1982.

Cover type Spring
N=2146

%

Summer
N=715
I

Fall
N=868
%

Winter
N=4383
%

Cottonwood 3.2 6.7 4.1 8.7
Willow 9.3 16.6 9.7 7.0
Wild Hay 31.2 11.3 43.8 20.5
Alfalfa 13.3 13.6 6.8 26.2
Small Grain 15.8 6.7 7.1 16.2
Native Range 11.0 15.4 3.3 9.1
Juniper slopes 2.9 1.3 0.2 1.3
Conifers 0.6 0.3 0.0 0.6
Marsh 0.7 6.3 2.5 0.8
Mixed Brush 1.7 4.6 2.8 1.7
Aspen 1.9 2.4 4.3 2.0
Mixed Forest 0.6 0.0 1.2 0.7
Riparian 0.5 1.3 1.3 0.2
Deciduous Forest 0.5 2.7 1.6 0.1
Total 100 100 100 100

Table 16. The percent of deer using hay stacks during different 
seasons and snow depths in the Gallatin Valley during 
1981 and 1982.

Snow cover Winter Spring Summer-Fall

WT 1 MD z WT MD WT MD

No Snow 0% 5.1% 0% 0% 0% 0%
(n= 353) (n= 79) (n=863) (n=282) (n=896) (n=381)

Scattered 0% 0% 0% 15.2% 0% 0%
(n= 389) (n=l65) (n=204) (n= 33) (n=175) (n= 0)

2.5-15 cm 4.5% 41.2% 26.3% 1.0% 0% 0%
(n=1591) (n=274) (n=350) (n=103) (n= 88) (n= 0)

15 -30 cm 8.5%
(n= 691)

37.0%
(n=200) — — — —

1
2 WT White-tailed deer 

MD Mule deer



40

separate cover type. Each round bale appeared to have a small area 
down wind from it that tended to have less snow and also offered some 
additional foraging options. The only types of hay that deer were 
known to feed upon were second and third cuttings of alfalfa.

Activity Patterns

The probability of the radio-monitored deer being active under 
different conditions was analyzed. White-tailed deer were active 
more often (60.3%, z test P=.0166) than mule deer (50.9%). Activity 
monitoring showed that white-tailed deer were more likely to be 
active from midnight to noon (57.5%) than were mule deer (41.2%),
(P=.0023), and from 2 hours after sunset to midnight (76.7% and 
37.0%), (P < .0001) (Figure 9). However between noon and I hour 
after sunset mule deer were more likely to be active (85.0%) than 
whitetails (51.5%), (P=.0001). The number of deer seen per one half 
hour gf route driven showed a similar trend for the afternoon and 

evening hours (Figure 10).
Active deer made up 86% of the nocturnal sightings in plots 

during the summer and only 55% during the winter (z test, P=.0002) 
(Table 17). Instrumented deer were active more often overall during 
the winter and spring (59.2%) than during the summer and fall (48.5%) 
(P=.0197). This was especially true for the time period from noon 
till I hour after sunset when deer were active 73.8% and 42.4% of the 

time respectively (P < .0001). This may have been due to the warmer 
temperatures at this time of the day when compared with sunrise to



41

*  WHITE-TAILED DEER
• MULE DEER

SUNRISE

Figure 9. The diel activity patterns of radio-monitored white-tailed 
deer and mule deer in the Gallatin Valley during 1981 and 
1982.

A  W H IT I-TAILED DEiR

•  MULE DEER

SUNSETNO O N

Figure 10. The efficiency of observing both species of deer
from routes in the Gallatin Valley during 1981 and 1982.



42

noon when there was no significant difference (P=.9522). The number 
of deer seen per half hour of route driven was higher for nights 
during the summer than for winter nights (Figures 11 and 12). White­
tailed deer were more crepuscular and less diurnal during the summer 
and fall than during the winter and spring according to results 
obtained from activity monitoring (Figures 13 and 14).

Table 17. Observed deer activity during different temporal
conditions in the average number of deer seen per survey 
of 21 plots in the Gallatin Valley during 1981 and 1982.

Semi-lunar period centered on the:
Time Season Full Moon . New Moon

Bedded Active Bedded Active
Day Winter 4 9 3 33
Day Summer 0 7 0 16
Day Total 4 16 3 49
Night Winter 24 56 74 65
Night Summer I 26 4 5
Night Total 25 82 78 70

The phase of the moon seemed to have a decided affect on the 
activity patterns of white-tailed deer (Figure 15) as it did on both 
species of deer seen in survey plots, both at night and during the 

day. Diurnal activity and nocturnal bedding were more frequently 
observed when the moon was less than half illuminated. During the 
night, 2 hours after sunset to 2 hours before sunrise, both species 
of deer were more active during the semi-lunar period centered on the 
full moon (73.2%) than at other times (50.9%) (P=.0003). During 
spotlight surveys, more deer were seen overall during the semi-lunar 
period centered on the new moon. Progulske and Duerre (1964) found



43

a SUMMIK

•  FALL

NOON

Figure 11. The efficiency of observing deer of both species
during the summer and fall from routes driven in the 
Gallatin Valley during 1981 and 1982.

•  WINTER

A SPRING

SUNSETNOON

Figure 12. The efficiency of observing deer of both species
during the winter and spring from routes driven in 
the Gallatin Valley during 1981 and 1982.



44

• Summer 

A Fall

SUNRISE NOON SUNSET

Figure 13. The diel activity patterns of radio-monitored white-tailed 
deer during the summer and fall in the Gallatin Valley 
during 1981 and 1982.

• Winter

A Spring

SUNRISE NOON SUNSET

Figure 14. The diel activity patterns of radio-monitored white-tailed 
deer during the winter and spring in the Gallatin Valley 
during 1981 and 1982.



45

deer to be less wary on dark nights and could be spotlighted more 
easily. He believes that better results could be obtained by measur­
ing illumination than by recording the phase of the moon and amount 
of cloud cover.

A f u u m o o n

•  NEW MOON

' \ V

SUNRISE SUNSET

Figure 15. The diel activity patterns of radio-monitored white-tailed 
deer during semi-lunar periods in the Gallatin Valley 
during 1981 and 1982.

The amount of activity during the day seemed to decrease with 

the amount of surrounding disturbances. From 3 hours after sunrise 
to 3 hours before sunset deer within 800 m of 10 or less houses were 
active a significantly greater portion of the time (50.0% P=.0003) 
than those within 800 m of 11 or more houses (18.6%). Deer were less 
likely to be active when there were greater than 11 houses within 800 

m during all periods, 50.2% (P=.0000) as compared to 68.4% when there 
were less than 11 within 800m. This does not imply that deer near



46

more houses were less active or that they expended less energy, 
merely that they spent more time bedded and less time being active as 

was also seen by Happe (1982). A radioed doe that spent the summer 

in a draw on the outskirts of Bozeman was only active at night and 
was noted to be active on nights with little or no.illumination.

Fourteen percent of the observations made in plots which had 3 
or fewer houses within 900 m were made during daylight hours whereas 
only 2% of the observations made in plots with 6 - 2 0  houses within 
900 m were made during the day. The radius of 900 m (.564 mile) was 
used since it encompasses an area of I square mile, the standard land 

measurement. The 1424 deer observed in plots were found in developed 
areas more during the night than they were during the day (Table 18).

Table 18. The percentage of both species of deer that were observed
during the day and night in plots associated with different 
housing densities in the Gallatin Valley during 1981 and 
1982.

Housing Density
within 900 m. Day Night

0 - 1 91% 88%
3 - 6 9% 7%
11 - 20 0% 4%
20+ 0% 0%



47

Effects of Housing

Analysis of 21 plots of similar size for deer revealed that deer 

numbers and housing density within a 900 m radius were inversely re­
lated (Figure 16). Deer were never seem in plots with more than 20

' ■ \houses within 900 m during the winter. These plots were only observedito contain deer that were moving through the plot without feeding, or 
bedding. The drastic effects on deer of increasing the housing 
density from I to 2 houses per section was indicated. The effect of 
each additional house is less than the the effect of the house pre­
vious to it. This indicates that 640 houses each on I acre and 

restricted to I section of land while 9 sections are left empty would 
have considerably less effect on deer than 640 houses each on 10 
acres that would occupy the same 10 sections. It appears that deer 
view an area with little distinction of the intermediate situations 
of development. Deer use along routes was also distributed with re­
spect to housing (Table 19). Mule deer observed by all methods had 
fewer houses within 800 m (0.7) than white-tails (4.0, t test, 
a=.000).

The Valley was divided into 4 types on the basis of housing 
density (Table 20). Deer in general appeared to use the low develop­
ment type less than the medium and the high development types based 
on data collected from routes and flights, but not telemetry (Table 
21). This indicates that areas not being developed might be consider­
ed marginal deer habitat. Since telemetered deer remained in the 
general vicinity of the trapping site, there could have been selection



N
U

M
B

E
R

 O
F 

DE
ER

 
PE

R 
PL

O
T

48

•  (n = 7 )

DEER NUMBERS :  -2 0 .7 4  +

(.8911.011 H O U SE S -.0 0 5 4 1 )  +

2 2 .0 9 6  COVER TYPE PREFERENCE.

TOTAL HOUSES W ITHIN  A 9 0 0  m RADIUS

Figure 16. The average number of deer of both species seen per 
plot per survey in relation to the number of houses 
within 900 m (.564 mile radius), in the Gallatin 
Valley during 1981 and 1982, assuming a cover type 
preference of 1.0.



49

Table 19. Deer use by both species along routes relative to housing 
in the Gallatin Valley during 1981 and 1982.

Houses 
within 
800 m

Deer
Observations

Control
Points

Relative 
Deer Usage1

0 619 141 4.39
1 - 2 152 138 1.10
3 - 10 310 200 1.55
10 + 29 132 0.22

Relative deer usage was calculated by dividing the percentage 
of all deer found in the housing category by the percentage 
of all control points in that category. Values greater than 
1.0 indicate preference.

on the part of the trapper to be in less developed areas. Trap sites 
had an average of only 1.2 residential houses within 800 m and 6.8 
within 1600 m. To determine how deer perceive housing density in 
different development types, relative use was calculated from the 
percent use divided by the percent of control points in that develop­
ment type which fell within the given housing density (Table 22).
Deer appeared to be less affected by housing when it was abundant and 
inescapable. This was shown by comparing the number of houses found 
within 800 meters for controls and deer observations made by 3 dif­
ferent methods (Table 23). The relationship between avoidance of 
housing and the distance from deer to that housing appeared linear 
indicating no critical distance (Appendix Table 32). Deer avoided 
farm houses less than other houses. Deer avoidance appeared to be 
less when using data from aerial telemetery than flights or routes 
and greatest for routes (Table 24). This indicated that as deer use



50

Table 20. Average housing density for 4 development types 
in the Gallatin Valley, 1982.

Development Houses within Houses within
Types 1.6 kilometers 3.2 kilometers

Route Grid Route Grid
control control control control

Low .3 1.0 .8 6.0
Medium 9.0 8.0 34.0 37.0
High 32.0 33.0 85.0 84.0
Very High ,103.0 117.0 158.0 174.0

Table 21. Relative deer usage, by both species, of development 
types on the basis of availability by 3 methods of 
sampling in the Gallatin Valley during 1981 and 1982.

Development Routes 1 Flights 1 Aerial Telemetry 1
Type n=1136 n=l662 n= 157

Low 0.50 0.66 1.40
Medium 1.77 1.10 1.35
High 0.73 1.49 0.68
Very High 0.10 0.20 0.18

Relative use was derived from the percentage of deer seen by 
that method that were in each type divided by percent of 
control points for that control method that were in that type. 
Values greater than 1.0 indicate preference.



51

Table 22. The relative use of areas by both species of deer 
relative to the development type and the number 
of houses within 800 m in the Gallatin Valley 
during 1981 and 1982.

DEVELOPMENT TYPE
LOW MEDIUM HIGH VERY HIGH

Houses
within % rel. % rel. % rel.
use use use use use use use
800m 1 I l

0 96.9 1.06 75.4 1.91 3.8 0.64 0.0 - -
1-2 3.1 0.37 16.2 0.46 9.3 0.42 0.0 - -
3-10 0.0 -- 8.1 0.36 84.9 1.92 4.5 0.32
10+ 0.0 — 0.3 0.10 2.0 0.08 95.5 1.13
Total 100.0 100.0 100.0 100.0

Relative deer usage was calculated by taking the percentage of 
all deer that were found in a category and dividing that percent­
age by the percent of control points in that category. Values 
greater than 1.0 indicate preference.

% rel.
use usel

more developed areas they make greater use of available cover and 
become increasingly secretive.

Deer were observed even closer to cover if housing within 900m 
was dense (Table 25). Deer were observed closer to cover during the 
day, an average of 96m, as opposed to 201 m at night. During the 
night deer also appeared to remain closer to cover when bedded (181m) 
than if they were active (214m). The comparison of distance to cover 
for different diurnal activities could not be made since diurnal 
bedding was only observed in the low and medium development types and 
even there only during the winter. Plots with less associated cover 
received frequent use only where housing developments were scarce. 
Deer use of cover relative to the immediately surrounding density of



52

Table 23. Deer avoidance of housing in different development types 
in the Gallatin Valley during 1981 end 1982.

Development
Types

Routes 1 Flights 1 Aerial 1 
Telemetry

Low 0.86 0.44 1.61
Medium 0.34 0.34 0.32
High 0.63 0.88 0.94
Very High 0.51 0.47 0.59

Avoidance was calculated by the average number of houses within 
800 m for deer observations divided by the average number of 
houses within 800m for the respective set of control points. 
Values of less than 1.0 indicate avoidance.

Table 24. Comparison of average distance to cover and average 
number of houses within 800 m for 3 methods 
of observation of both species of deer in the 
Gallatin Valley during 1981 and 1982.

Control Route
Ground

Control
Aerial

Flight Aerial
Telemetry

Houses X X A X X A X A
within (No.) (No.) I (No.) (No.) I (No.) i

400 m 6.8 0.42 0.06 5.7 0.6 0.11 0.92 0.16
800 m 14.7 2.1 0.14 11.5 2.1 0.18 3.9 0.34
1600 m 34.6 10.8 0.31 29.1 8.8 0.30 15.7 0.54

Distance 
to cover 296m 125m 47m 0.23m

A - Avoidance was calculated by the average number of houses 
within a radius for deer observations divided by the average 
number of houses within that radius for the respective set of 
control points. Values of less than 1.0 indicate avoidance.



53

houses differed depending on the level of development in the area 
(Table 26). Caution must be used in assessing distance to cover 
ratings since deer use developed areas more at night and deer tend to 
be further from cover at night.

Table 25. The use of cover by white-tailed deer and mule deer 
in relation to housing density within 900 m of 
the survey plots in the Gallatin Valley during 
1981 and 1982.

Number of 
Houses 
within 
900 m

Average
Distance to Cover 
when Bedded.(m)
(n=486)

Average
Distance to Cover 
during the Day (m)

(n=186)

0 - 1 182 98
3 - 6 173 34
11 - 20 22 0
20+ 30



54

Table 26. Deer use of cover along routes relative to the
development type and the amount.of housing within 
800 m in the Gallatin Valley during 1981 and 1982.

LOW MEDIUM. HIGH VERY HIGH
Houses
within x C X C X C X C
800m (m) 1 (m) l (m) I (m) I

0 67 0.54 124 0.48 19 0.06 -- ' --
1-2 5 0.08 106 0.35 118 0.39 “ “ —  —

3-10 — — — — 116 0.39 315 1.19 30 0.09
10+ — — — — 30 0.08 20 0.07 116 0.30

C was calculated by dividing the average distance to cover for 
deer observations by the average distance to cover for 
controls. Values greater than 1.0 indicate less dependence 
on cover.

Effects of Roads

Interstate 90 was avoided the most while gravel roads were 
avoided the least (Table 27). Figures for the effects of roads along 
routes were not comparable since deer closer to roads were more 
easily observed from the truck. Plots along paved roads were used at 
least as much at night as comparable plots along gravel roads.

Use of Soils

Twenty-nine soil types were identified in the study area accord­
ing to the description given by DeYoung and Smith (1936). These were 
grouped by limitations to cropping (in Leeson 1972) and by moisture 
on a subjective basis. It was discovered from control points that 
the more arable soils had more houses upon them. Deer use was highest



55

Table 27. Relative distribution of deer within different distances 
of roads as measured by 2 methods of observation in the 
Gallatin Valley during 1981 and 1982.

Route Flight
Distance Gravel Paved 1-90 Gravel Paved 1-90
to road (m) I i I l l I

0 - 3 0 0.19 0.0 0.0 0.03 0.0 0.0
31 - 100 0.74 0.55 0.28 0.07 0.16 0.25
101 - 300 0.95 1.72 0.67 0.63 0.42 0.0
301 - 600 1 1.54 0.47 0.79 1.31 0.48 1.10
601 -1200 1.63 1.35 0.39 1.53 1.01 0.11
1201 -2400 3.68 2.28 0.82 1.67 2.13 0.56
2400 +

.

0.0 0.34 1.19 0.43 0.79 1.19

Avoidance of roads was calculated by dividing the number of 
deer observations in a category by the number of applicable 
controls in that category. Values less than 1.0 indicate 
avoidance.

on the intermediately arable soils. Soils with severe limitations 
for agriculture were used 81.5% as much as their abundance while 
those with moderate and slight limitations were used 133.3 and 45.3% 
as much as their abundance. In general, deer appeared to prefer the 
moderately mesic soils.



56

DISCUSSION

Movements and Distribution

In the past, mule deer increases appeared to have originated 
from the surrounding mountains during times of abundance. However, 
mule deer comprise a resident population in the Valley Center area. 
Two or three decades ago, according to long term residents, there 
were mule deer along Middle Creek between the Valley Center area and 
the Gallatin Range. It seems likely that disruption of either travel 
Corridors or the intermediate population might have isolated this 
population. According to Mackie and Pac (1980) the small number of 
deer found east of Bozeman may be remnants of a larger population 
that used these areas prior to extensive human development. It is 
likely that deer from the Bridgets pass through this area as marked 
deer have been found here on the Interstate, and seen southeast of 
Bozeman, and at higher elevations in the Gallatin Range. Some distur 
bance effects may be occurring in the Sypes Canyon area north of 
Bridget Creek. Collared mule deer in this area had erratic and 
extensive movements (Mackie and Pac 1980). There was a shift from a 
resident population of mule deer to white-tailed deer in the upper 
Reese Creek area between 1950 and 1975 (Wilkins pers. comm, and 
Schwarzkopf pers. comm.).



57

Many parts of the Valley have probably always been marginal 
habitat for mule deer. Many factors might have contributed to the 
increase in whitetails. These include greater reproduction by white- 
tails, greater tolerance of development as well as a possible lower 
mortality rate, especially if more hunting was allowed in the past.

Whitetails have recently been observed at increasingly higher 
elevations in the surrounding mountains by the author, Dave Pac, and 
John Cada.

In the Sourdough area mule deer have shifted their use east of 
the creek along the slope while whitetails have moved into the area 
on the west side of the creek. Elk were also observed using the 
slopes on the east side of the creek. There is a substantial housing 
development west of Sourdough Creek.

In two similar areas, Penwell and Valley Center, the distances 
between fixes were greater for white-tailed deer even though mule 
deer occupied a larger home range.

It seemed that human development caused home ranges to become 
more linear (Table 7). This was probably due to cover occurring 
along streams and draws in a linear fashion.

During the early winter of 1982 - 1983 deer that did have 
distinct and separated winter and summer ranges were found in inter­
mediate locations, perhaps due to the mild weather and lack of snow. 
Radio equipped does appeared to congregate near streams during the 
fawning season as well as during the summer.



58

Population Structure

Overall fetal ratios were high indicating that does were re­
ceiving an adequate diet. Whitetail ratios were higher than mule 
deer. The estimated fawn mortality of 40% during the first 90 days 
was comparable to that found in other studies. O1Pezio (1978) found 
a 24% pre- and postnatal fawn mortality in whitetails most of which 
was incurred during the first 3 days in an area relatively free of 
human and predator effects. Cook (1971) found a 72% mortality in 
southern Texas where 93% of this occurred during the first month.

Several causes of mortality were possible in the Valley, which 
include a difficult first parturition due to the young age structure 
of pregnant does, predation, abandonment, adverse weather conditions 
and hay mowing accidents. Death of very young fawns due to vehicle 
collisions was rare. According to 0'Pezio (1978) this is due to 
their sedentary nature.

Males experienced a greater turnover than did females in this 
study. Pac (1979) found that only 30 and 24% of mule deer and white­
tailed deer female and only 10 and 14% of the male carcasses, re­
spectively, were older than 4.5 years. In this study the fawn 
carcasses were 57 and 67% male for mule deer and white tails re­
spectively while the fetal ratio was 4I and 50% respectively. Adult 
males probably comprise between 12 and 20% of the population. This 
greater mortality could be in part due to a substantial number of 
males being born to young does. Pac (1979) reported that 83% of the 
fetuses recovered from fawn carcasses were male while only 40% of



59.

those from adult carcasses were male. Verme and Ozaga (1981) found a 

greater proportion of female fetuses in increasing herds on good 
range and yet pregnant doe fawns had a higher proportion of male 
fetuses. Dapson et al. (1979) said if the age of young does is 
responsible for greater mortality of fawns, more males than females 

would be effected.
Fawns conceive later, would tend to be in less vigorous condition and 
be less experienced than older does. Males also tend to stray more 
and would thus be expected to experience greater mortality.

Movements of marked deer indicate that young males moved further 
than other deer. This was also found by Leopold et al. (1951). Hood 
and Inglis (1974) and Sweeney et al. (1971) stated that when exposed 
to disturbance it was the male that was most likely to extend or move 
its home range. Hawkins et al. (1970) showed that 87% of females 
were sedentary as compared to only 10% of the yearling males.

Soils

In the Gallatin Valley deer densities increased with soil fertil­
ity up to the point where human activities became intense. This is 
similar to the trend reported by Crawford (1950) for raccoon densi­
ties , soil fertility and intensity of agriculture. Not only is 
development on fertile farmland a waste of a valuable resource, it . 
can also be more expensive since these soils are often poor for 
construction of streets, septic tanksj and foundations (Leeson 1972). 
"The use of Gallatin County valley soils is changing from agricultural 
to urban". "Some of.the best soils in Montana are being used.to



60

support roads and houses and to dispose of human wastes (septic tank 

filter fields) rather than to raise crops and livestock" (Stuart 
1974). "The major consumption of land for residential use has occur­
red on prime agricultural land in the Bozeman area" (Anon. 1974).
Farm land often sells for developments at a price much higher than 
can be made profitable by farming practices. Agriculture and wildlife 
should not have to settle for less fertile lands.

Use of Vegetation

I found that rumen samples were of greater volume in the winter 
than at other times. The penned deer of Thompson et al. (1973), 
demonstrated a decrease in Dry Matter Intake (DMI) during the winter. 
This is in direct contrast with my findings. Several distinctions 
must be made between the 2 studies. Free-roaming deer forced to 
consume a seasonally changing diet would respond differently than 
penned deer. The second distinction is that the volume of the rumen 
is not the best indicator of DMI and a fuller rumen may be a re­
flection of a longer turnover time. .It is apparent however that 
alfalfa and small grain use were greater' during the winter.and these 
are easily digestible. The last consideration is the effect of 
disturbance. Energy conservation is a strategy often used by deer to 
tolerate winters. They reduce theinanetabolic rate and activity to 
conserve energy. Disturbance could cause a shift to less economical 
activity patterns, elevate metabolism and increase the cost of mainte­
nance. The conservation strategy may not be consistent with the



61

survival of deer existing in developed areas. This could be especial­
ly true where high quality forage can be obtained throughout the 
winter.

Cover type use was based on the amount of associated cover, lack 
of human development, and the supply of seasonally available forage 

(Appendix Table 31). Use of grain fields was observed to be heaviest 
in March and April when and where snow was shallow and winter wheat 
was the first forage to begin "greening-up".

Hay stacks were used only during the winter and spring. They 
were used most when snow was deep and other forage options were 

limited.
Placement of hay stacks near cover facilitated its use by deer.

To prevent this, stacks could be placed closer to houses and further 
from cover when convenient. Fencing and paneling could also be 
employed when feasible to deter deer usage. Baling of hay prevents 
easy access by deer to most of the hay. In areas of high depredation 
it would be advisable to refrain from placing stacks near cover, 
avoid loose stacks ,■ and try to stack rectangular bales rather than 
leaving them scattered in the field. If loose stacks are to be used, 
they should be lined up in rows rather than clumped in groups, to 
prevent deer using them for security while feeding in the center of 
the clump.

For maintenance of root carbohydrate reserves and the health of 
the stands it is often recommended that alfalfa not be cut in the 
month previous to the first killing frost. If this hay is not subse­
quently cut, this "stock-pile system" might prove useful in areas of



62
high deer depredation. The new shoots of alfalfa might lure deer 
away from stacks when snow depth is not excessive. Alfalfa left 

standing can always be grazed by cattle. This system could prevent 
the loss of the cost of harvesting hay that deer might feed on anyway. 

In other words, the loss of standing hay will occur, but the cost of 
harvest will be avoided. In addition to the direct loss of crops by 
consumption, mule deer were observed standing, urinating and de­
fecating on loose stacks of alfalfa hay. Furthermore economic loss 
was also incurred where, due to moist soils, malting barley was 
windrowed and the deposition of fecal pellets caused a lowering of 

the market value.
Hay stack use by mule deer occurred during the day while white­

tailed deer were rarely observed on stacks until night, a pattern 
consistent with deer which fed away from stacks.

Activity Patterns

Whitetails were more nocturnal than mule deer (Figure 9). They 
were more likely to be active during the day in the winter and at 
night during the summer (Figure 13 and 14). This was also demon­
strated by Progulske and Duerre (1964), Montgomery (1963) and 
McCaffery and Creed (1969).

During this study it was determined that deer were more likely 
to be active nocturnalIy if the moon was greater than half illumin­
ated. Skinner (.1929) found that whitetails were active at all hours 
even noon during the winter and near sunrise and sunset during the 
summer or at night during the summer if the moon was shining.



63

Deer appeared to be more nocturnal in the presence of dis­
turbance ; Observed deer numbers and home densities in Utah yielded a 
strong negative regression (R=-.82) while pellet groups analysis did 
not (R=-.18) (Mann pers. comm.)- Dasmann and Taber (1956) indicated 
that black-tailed deer used areas near housing more at night. Mackie 
and Pac (1980) speculated that "disturbances that increase the amount 
or alter the pattern of activity will cause energy reserves to be 
used at a faster rate and decrease the ability of the animal to 
survive and reproduce".

According to Farr and Andrews (1978) subordinate animals showed 

desynchronization of activity rhythm at high density. This effect 
imposed a higher bioenergetic cost upon the subordinate than upon the 
dominant mice. In this study it appeared as if this principal could 
be applied to deer and humans as communicated through housing develop­
ments whereby deer assume the subordinate role and subsequently are 
probably burdened with greater energetic costs.

Development Effects

The curve relating housing density to expected deer numbers in 
Figure 16 demonstrated that when the number of houses was low the 
addition of another house had a pronounced effect on the number of 
deer using that area. However, once housing progressed to a certain 
density an additional house had substantially less effect on deer 
usage. In contrast, Mann (pers. comm.) found a linear relationship 
between deer use and housing density.



64

Deer use of the low development type was high as measured by 
aerial telemetry. This may have been the result of the selection of 
trap sites. The lesser use received by this type as observed from 
flights and routes was probably due to characteristics of the habitat. 
Deer in this category did not appear to avoid houses as much as in 
the next more dense type. This could be a function of fewer houses

I

to avoid or because this type was very open and deer selected areas 
along streams and cover which were also selected for by housing. The 
increased usage of the medium development type might be related to 
the high usage of alfalfa fields since these fields are labor inten­
sive and might be more likely to be found in this type than in the 
low development type. Deet in the medium development type avoided 
housing the most and was probably a result of having more suitable
natural areas away from houses with the proper habitat characteristics
-

to choose from. Deer in the high development type seemed to avoid 
housing less perhaps because deer became accustomed to housing and 
because housing was distributed evenly throughout this type. Thus 
deer could not avoid it unless they left this development type.

Within the extreme development type, deer only needed to avoid a few 
areas with very great densities of housing, such as downtown Bozeman, 
to dramatically increase their avoidance rating. The indication from 
the distribution of deer between development types is that much of 
the development has occurred on deer habitat and that the areas that 
are not developed are marginal deer habitat at best. The closeness 
of. cover to deer in the least disturbed development type could be a



65

function of the distribution of forage in the absence of intensive 
agriculture being concentrated along streams and cover..

The decrease in the nocturnal activity from summer to winter 
displays the need for increased activity during daylight hours in the 
winter. This need for activity during the day coupled with the 
greater need for security during the day indicate a need for winter 
concentration areas to remain undeveloped.

Residential houses were avoided more than farm houses. Geist 
(1971) reported that disturbance tends to have the greatest potential 
to be detrimental if it is unpredictable and frequent. Activities 

around farm houses are fairly predictable and deer became accustomed 
to some of this activity. Farm houses were also more evenly distri­
buted which could also account for this lesser avoidance. Another 
reason might be due to some of the positive effects associated with 
farm houses, such as a higher quality supply of forage. One of the 
most important effects of disturbance is it causes the animal to move 
to a less favorable area. (Geist 1971). According to Geist (1971), 
disturbance is more easily tolerated in species that produce large 
numbers of young, that have a wide dispersal, have a short life 
expectancy, and follow nocturnal habits. Whitetails better fit the 
above description in every way than do mule deer. Geist (1980) 
claimed that whitetails with their excitable nature would not be 
able to tolerate human activities as well as mule deer. However Geist 
worked in a town within a national park under some unique circum­
stances. Under the situation found in the Gallatin Valley it was



66

apparent that whitetails thrived in areas undergoing development 

more easily than mule deer.
Disturbance decreased the amount, changed the time, and influ­

enced the manner in which the area was used by deer. Howe (1973) 
delineated several impacts of residential mountain developments: 
Simplified complex ecosystems; usurped space for food, cover, and 
shelter; increased human activity in adjacent wildlife habitats; 
increased traffic and mortality; increased anti-hunting attitudes; 
increased poaching; and increased harassment and predation by.dogs.

The number of deer present in developed areas during this study 
were dependent on surrounding concentrations of deer in nondeveloped 
habitats. Populations where cover is contiguous are more stable than 
where fragmented. Travel corridors between cover patches should 
increase the benefit of the patches (Picton and Mackie 1980). Cover 
patches should be maintained above some minimum effective size.
Areas of cover are more frequently used if they are large and sepa­
rated from disturbance by distance or a barrier (Hayden 1975). 
Increased topographic relief, tends to protect deer from impacts 
(Picton and Mackie 1980). Patches should not become so removed from 
each other.that the deer are incapable of inter-isolate dispersal in 
order to minimize the probability of simultaneous extinction and to 
maintain biological diversity (Wilcox 1980). Zoological insularity 
of mammal populations of larger land units increased as human economic 
development occurred (Picton and Mackie 1980). Any additional in­
crease in mortality factors, such as fragmentation of cover patches,



67

is less likely to be compensated for by an increase in reproduction 

since this is already approaching the maximum.
Allen and McCullough (1976) noticed that 3 times as many acci­

dents occurred on 2-laned paved roads as on divided highways, while 
fewer occurred on dirt roads. He further stated that the number of 
deer killed is a function of the number of deer crossing and the 
traffic volume. Deer-vehicle collisions are a waste of a resource as 
well as a hazard to human life and property. The public should be 
made aware of this hazard and the importance of driving slower at 
night especially when crossing creeks where icy spots often develop 
during the winter, where fog is common, and where deer are most 
likely to cross. Deer crossings should be well delineated, well 
marked, and fencing should be employed. There could be some improve­
ment along these lines in the Valley. Free and Severinghaus (out of 
Allen and McCullough 1976) showed that fences when used in conjunc­
tion with warning signs and antlerless deer seasons reduced accidents 
44 to 84%. Allen and McCullough (1976) said that control of the deer 
herd through harvest is one of the most effective means of reducing 
accidents. Harvest would not only reduce vehicle collisions but 
would reduce depredation on crops through lowering of deer numbers, 
maintaining natural forage without depletion, and possibly altering 
distribution of deer.

Interstate 90, paved roads, and gravel roads were avoided from 
most to least respectively in this study. It was suspected that 
apparent effects of roads included the effects of associated housing 
and a different array of cover types. Stress along one environmental



68
gradient leads to decreased tolerance along another gradient (Odum 

1971).
Deer observations along routes indicated a greater avoidance of 

houses than aerial observations did. Deer using areas near dis­
turbances concentrated in cover where detectability was low. This 
lead toward apparent, avoidance as the effectiveness of the method of 
observation decreases. Routes were less effective at observing deer 
in cover than flights while aerial telemetry was the most effective. 
This was also seen by comparing the average distances to cover for 
the 3 methods as given in Table 24. Youmans (1979) also noticed a 
greater use by deer of the forested types as indicated by telemetry 
with the least indication of use by ground observations with intermed 

iate use levels indicated by aerial observations. Spotlighting was 
also effective since reflection from the tapetum indicated the loca­
tion of deer in cover. Males were seen less often from roads than 
they were from the air. According to Hood and Inglis (1974) and 
Sweeney et al. (1971), male deer are more susceptible to disturbance. 
Behrend and Lubeck (1968). noticed that deer held their tails up in 
flight more often along roadsides than away, indicating greater 
sensitivity to disturbances along roads. Greater sensitivity to 
housing development by deer along roads was seen in this study as 
depicted in Appendix Table 32. Rost and Bailey (1979) said road 
avoidance was greatest along more heavily traveled roads, greater for 
deer than elk, and for greater deer in shrub habitats than deer in 
pine and juniper habitats.



69

Deer may avoid roads to the extent that is detrimental to their 
welfare (Rost and Bailey 1979).

Hunting

No poaching was observed during the study, although on several 
occasions there were signs of its occurrence. Local residents related 
cases of vandalism in which deer were shot and left and this sort of 
behavior was also documented. Siegler (1951) reported that a buck 
law resulted in a greater waste of usable deer and a greater kill 
outside of the season. Housing conditions in many parts of the 
Valley exclude hunting by rifle although hunting with a bow or a 
shotgun could be possible. Archers might wound more animals in the 
Valley than gun hunters since Stormer et al. (1979) reported that 
archers wounded 50% more deer per hunter than gun hunters. It would 
also seem quite difficult to control population numbers with a bow
season.



70

CONCLUSIONS AND RECOMMENDATIONS

Agriculture and wildlife are compatible with each other while 
housing developments are incompatible with both. Soils that are best 
suited for agriculture should not be used for housing developments. 
Every effort should be made to stop the deterioration or destruction 
of cover. Tax laws should be reviewed to insure that taxes on agri­
cultural property near developments do not exclude its use as farm­