Ec o l o g y a n d Ma n a g eme n t of Mule Deer and White-tailed Deer in Montana

Ecology and Management of Mule Deer and White-tailed Deer in Montana

By

Richard J. Mackie, David F. Pac, Kenneth L. Hamlin, and Gary L. Dusek

Montana Fish, Wildlife and Parks Wildlife Division Helena, Montana

Federal Aid Project W-120-R

1998 About the Authors

Richard J. Mackie is Professor Emeritus, Fish and Wildlife Program, Department of Biology, Montana State University, Bozeman. He conducted and supervised research on deer in Montana since 1960. During 1975-94, he served as coordinator of the statewide deer research program.

David F. Pac is Research Biologist, Wildlife Division, Montana Fish, Wildlife and Parks, Bozeman. He studied mule deer in the Bridger Mountains since 1974 and assisted in deer studies on other areas.

Kenneth L. Hamlin is Research Biologist, Wildlife Division, Montana Fish, Wildlife and Parks, Bozeman. His studies of mule deer began in 1972. During 1975-89, he conducted studies in the Missouri River Breaks, prairie-badlands, and prairie-agricultural environments.

Gary L. Dusek is Research Specialist, Wildlife Division, Montana Fish, Wildlife and Parks, Bozeman. His studies of deer began in 1969. During 1980-93, he conducted studies on white-tailed deer in plains riverbottom, prairie-agricultural, and northwest montane forest environments.

Illustration and Photo Credits

Illustrations: Diana Haker - Cover, pps. 11, 23, 27, 34, 40, 48, 67, 71, 75, 110, 117, 136, 141, 153, and 161; Media Works - Figs. 6, 8, 27, 28, 63, 64, 65, 66, 120; Robert Neaves - p. 58

Photos: Mike Aderhold - p. 156; Kevin Berner, courtesy North Dakota Game and Fish Dept. - p. 66; Brad Compton - p. 51; Kevin Dougherty, Glendive Ranger Review - p. 8 (lower right); Gary Dusek - Fig. 5A & B, pps. 8 (upper left), 79, 130, 149, and 160; Michael Francis - Fig. 9; Ken Hamlin - Fig. 3A, pps. 1, 102, 139, and 142; Terry N. Lonner - Fig. 2A & B, pps. 7, 69, 72, 101, 105, and 107; Rick Mace - Fig. 6B; Richard Mackie - Fig. 6A, pps. 3, 25, 50, 53, 77, and 159; MFWP - Fig. 19 and p. 94; Greg Pierson - p. 125; Craig Sharpe - p. 111; Frank Siroky Jr. - p. 147; USFWS, CMR - Fig. 3B; Dan Wesen - pp. 108, 124, 131, and 150; Tom Wesen - p. 122; Gene Wolfe - p. 119; Alan Wood - Fig. 4A & B

©1998 by Montana Fish, Wildlife and Parks. All rights reserved. Designed by Media Works, Bozeman, Montana. Printed in the United States of America by Color World Printers, Bozeman, Montana, on recycled paper.

Permission to reproduce or copy any portion of this bulletin is granted on condition of full credit to Montana Fish, Wildlife and Parks and the authors. Foreword

Mule deer and white-tailed and methods and their applications. deer are the most widely distributed By the early 1970s, the and abundant big game mammals in environments in which deer existed Montana. Although evolved to live and were changing rapidly. Some methods thrive in broadly different environments, for deer management became outdated the two species are remarkably adaptive. and it was evident that new information Both occur in a wide variety of habitats, and approaches were necessary. In under widely fluctuating environmental 1975, an important long-term and conditions, in the presence of numerous comprehensive statewide research other wild mammals and domestic effort was initiated employing new livestock, and in the wake of extensive and emerging technologies in research human development and disturbance. on both species and across a broad Managing deer across diverse spectrum of environments in Montana. habitats and conditions in Montana Numerous ancillary studies mostly in the begins with understanding both their form of 2-year graduate student research biology and behavior. It also requires projects were conducted in association effective methods for monitoring with this long-term investigation. populations and habitats as well as for This bulletin was prepared as a manipulating deer numbers or habitat comprehensive summary of results factors to meet diverse social and from all of these studies. Like earlier economic objectives. efforts, the results lend additional Montana has a long history of insight to understanding the behavior, research to provide basic information biology and ecology of the two about deer and their habitats and to species. However, unlike earlier develop and test new and improved investigations, this investigation focused methods and criteria for deer on formulating research results into management. Studies during the 1940s management recommendations. This and ’50s provided most of the first resulted in important advances to refine scientific data, laying the foundation management strategies and practices for “deer management based on facts.” to help reduce some of the uncertainty Later, studies evaluated and refined that always exists in dealing with wildlife some of the early management concepts resources in complex environments.

Donald A. Childress Administrator, Wildlife Division Montana Fish, Wildlife and Parks

Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a iii Preface

The research presented in this of habitat is described in terms of report was funded and sponsored by juxtaposition of all components and hunters and the Federal Aid in Wildlife their use by individual deer or family Restoration program. Together they have units of deer, does habitat become provided major funding for scientific truly meaningful. Deer habitat is game management since 1941, including multidimensional and must include not numerous short-term and comprehensive only the basic components for survival, long‑term field research investigations but also the social behavior of deer. as reported in this bulletin. Although Individual deer of both sexes the information presented herein is comprise the basis of a deer population. definitive, no reader should be deluded However, partitioning environments into thinking these are the “last words” into deer matriarchal units surrounded in our understanding of Montana’s two by nearby and overlapping younger most numerous large mammals. Our female units is integral to understanding knowledge can never be complete, both how populations operate and for nor will management be conducted management options of the two species. with certainty. As the environment and Male habitat selection and survival, society changes, so must our knowledge while necessary for species continuance, of deer-habitat relationships. Research is is peripheral to the importance of one means of obtaining the wherewithal matriarchal units for maintenance or to recognize and adapt management to increases in deer populations. Females those changes. establish the ultimate pattern of deer It is difficult to consider deer population distribution in both new separate from their environment. All habitats and in habitats recolonized after that deer are and all that deer do are population declines. biological and behavioral responses to Understanding population the environments in which they occur characteristics and dynamics, including as individuals, populations, and species. age structure of the female segment and This bulletin expands on that view in discussions of deer‑habitat relations and patterns and rates of fawn recruitment deer population ecology. For example, and adult mortality, are crucial to the term “habitat” has been defined and managing deer at the local level. The characterized by many but understood severe reduction or loss of one cohort by few. Ask anyone what constitutes due to environmental stress often linked “deer habitat” and most will describe a with predation may not be critical to landscape that usually includes a buck the population, but severe reductions and/or a doe, and often a fawn or two in two or more consecutive cohorts can in a picturesque outdoor setting. Such set the stage for a significant population images are designed for economic decline. Conversely, good survival of markets or artwork and not scientific consecutive cohorts can foretell an understanding. Only when the “concept” imminent population increase.

iv Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a The probability of survival is high for almost 50 years, only in the past for most deer after achieving adulthood. decade or so have biologists begun to The majority of adult deer mortality define, quantify, and understand the can be accounted for by legal hunting, interaction of all causes of mortality in terminal wounding losses and illegal population ecology. killing. This human‑induced mortality In terms of management, the replaces some natural mortality in results herein indicate that many adults, but those latter rates are existing theories or “principles of normally quite low. However, occasional deer management” are less applicable episodes of high natural mortality of than commonly believed. Traditional adult females also can trigger population interpretation of “carrying capacity” declines, especially in association with did not explain observed deer- low fawn recruitment. habitat interactions on or among the The authors have emphasized various study areas. The concept of a the importance of behavior in habitat consistent “limiting factor” influencing relationships and population dynamics. deer population dynamics statewide Social behavior, while hard to quantify could not be identified. Similarly, the and explain, is often the driving force concept of “compensatory” increases in selection of certain habitats and in fawn recruitment and deer numbers avoidance of other habitats. It also is or decreases in natural mortality with an important element that allows local increased hunter harvests and reduced subpopulations to use habitat most population density was discussed as efficiently, at optimal densities, and having limited application. To assume adapt to fluctuations and other changes the general existence and operation of in the environments deer occupy. these concepts in population dynamics Land use changes and hunting of both species, across all environments, regulations over the last two decades and over time will likely result in have led to a dramatic increase in misinterpretation of management distribution and abundance of white- opportunities. tailed deer in Montana (In 1996, and for Harvest rate recommendations the first time on record, the statewide presented in this bulletin, if based on white-tailed deer harvest was higher the required information, may cause than the mule deer harvest). This in turn some concern among law enforcement has led to a significant overlap in the personnel and the public. Why? Because distribution of the two species. Despite harvest regulations for deer populations this overlap and parallel strategies existing in close proximity may be for habitat selection, they remain subject to different population control two distinctly separate species that strategies. It may take considerable typically select and use very different time for hunters and the public to realize habitats within their range. Just as that populations or other groups of our knowledge about mule deer and deer in close proximity to one another white‑tailed deer has evolved, so too may not be influenced by the same must our philosophy about managing land management practices, hunting them as separate species as distinct influences, and environmental factors from one another as either is from elk or (in fact what is bad or has negative antelope, for example. effects in one place or at one time Humans are an integral part of the may be good or beneficial in another). ecology of deer. Their greatest long-term Recognizing this and applying it to impacts are on deer habitat (good and harvesting and overall management of bad). In effect a major predator, they are deer may complicate rather than simplify responsible for a significant amount of hunting regulations and assessment annual mortality of deer through hunting of environmental impacts. Monitoring and other means. Although hunter deer numbers and harvests and their harvests of deer have been monitored respective compositions will require time

Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a v and appropriate procedures along with the first during 1972-1977, the second administrative and political support. The beginning in 1995. The reaction less this occurs, the more speculative to the next “check” will reflect our the recommendations for hunting knowledge, skills and ability to detect seasons, and the lower the probability population fluctuations and respond of successfully achieving long‑term deer with appropriate adjustments in hunting management goals and objectives. regulations. Management of deer at virtually The research results and any population level may be possible management implications presented in many areas once desired deer in this bulletin provide some new numbers, density, and population‑units information about modeling the effects are delineated. Because of the vagaries of size (numbers of animals) and density of weather, however, population (number of animals per unit area) as goals should fall within numerical separate but interactive population ranges, not point estimates, based parameters. Based on biological on observed or reasonably expected parameters and ecological boundaries, values for population fluctuations in these models allow managers to test the a given environment over time. Other impacts of various mortality factors, important components in designing deer including harvest strategies, on existing management goals include agricultural deer populations and future trends. and forest economics, land development Another deer decline will doubtless activities, traffic safety, and social occur early in the 21st Century, but if the tolerances for hunters, hunting, and management regimes presented in this wildlife viewers. bulletin are accepted and implemented, Biologists, hunters, landowners and fluctuations may be better predicted the general public have experienced two and receive more timely management “reality checks” in deer management responses. in Montana during the past 25 years:

Terry N. Lonner Chief, Research and Technical Services Wildlife Division Montana Fish, Wildlife and Parks

New paradigms will replace old when the new can explain anomalies between observation and the old paradigms. Kuhn 1970

vi Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Acknowledgments

Our studies and this report could Sam Curtis, Bozeman, MT, read and not have been completed without the commented on the final draft. Graphics, assistance and support of many people. graphic design and layout were by These include colleagues from Montana Martha Lonner, Marcia Leritz, and Mary Fish, Wildlife and Parks and other state Myers of Media Works, Bozeman, MT. and federal agencies who were involved Special thanks to the following for in collection and analyses of data. their assistance: They also include faculty and students at Montana State University and the Montana Fish, Wildlife and Parks: University of Montana who supervised Eugene Allen and conducted graduate research on our Keith Aune study areas, skilled pilots who flew our Richard Bucsis aerial surveys and contributed greatly John Cada to our efforts to mark and monitor Bruce Campbell deer, landowners who provided access Donald Childress to their properties, and individuals James Cross who volunteered their time on various Joe Egan projects. We also are greatly indebted to Glenn Erickson our wives and families for their patience Charles Eustace and encouragement as well as assistance Arnold Foss (deceased) in various aspects of our studies. Wynn Freeman (deceased) Funding was primarily by Montana Kenneth Greer Fish, Wildlife and Parks Federal Tom Hay Aid Projects W-98-R and W-120-R. Bernie Hildebrand However, cooperating agencies and Terry Hill organizations including the USDI Bureau William Hoskins (deceased) of Land Management - Lewistown Reuel Janson and Miles City Districts, U.S. Fish and Marilyn Johnson Wildlife Service - Charles M. Russell Henry Jorgensen National Wildlife Refuge, USDA Forest Steve Knapp Craig Knowles Service - Flathead National Forest, Thomas Komberec the Montana University System and Terry Lonner the Montana Agricultural Experiment Allan Lovaas Station also provided financial and other Neil Martin support instrumental to the success of Pete Martin individual studies. Steve Martin The cover and other drawings were James Mitchell the artwork of Diana Haker, Stevensville, Ray Mulé Montana. Margaret Morelli, Montana John Mundinger Fish, Wildlife and Parks, Bozeman, typed Helga Ihsle Pac the final draft and assisted in proofing. Dan Palmisciano (deceased)

Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a vii Montana Fish, Wildlife and Parks Landowners: Other Contributors: (continued): Walt and Marlene Adams It is impossible to mention Duane Pyrah Mr. and Mrs. L. B. Chapman all individuals that contributed Merle Rognrud Gerry Devlin as members of supporting Michael Ross East Indian Butte Grazing agencies and organizations or as Philip Schladweiler District volunteers. Among those who Keith Seaburg Roy and Ethyl Gentry specifically merit recognition are: Shawn Stewart Don and Joan Hagler J. Barlow Rick Schoening Perry and Marge Kalal Carol Bittinger Carolyn Sime Larry and Sue Kalina Dan Bricco Thomas Stivers George and Hank Leffingwell Pat Caffrey Ron Stoneberg Elsie Longley Larry Calvert Dick Trueblood (deceased) Lloyd and Sandy Maher Justin Cross Ken Walcheck Dick Morgan Ann-Elaine Darling C. Robert Watts (deceased) Horace Morgan Doug Dusek Dick Weckworth (deceased) Roy Newton Larry Eichhorn John Weigand Walt, Art, and Bob Reukoff Steve Emerson Harold Wentland George Rice Robert Eng August Sobatka John Foster Montana University Harold Temple Mitchell Friedman Graduate Students: Ralph Fries Agencies and organizations Keith Boggs Glen Gray contributing field assistance or Bradley Compton Bill Haglan support: Arnold Dood Rodney Hay June Freedman Montana State University, Loren Hicks James Herriges Department of Biology, Fish Susan Hinkins and Wildlife Program Rosemary Leach Leonard Howke John Morgan University of Montana, School Lynn Irby Mary Morton of Forestry, Wildlife Biology Kevin McGovern Harvey Nyberg Program Don Quimby Shawn Riley Montana State University Harold Picton Alfred Rosgaard Jr. Agricultural Experiment John Roseland William Schwarzkopf Station John Rumely William Steery Montana Department of State Robert Short Robert Trout Lands, Swan River State Robert Staigmiller Alan Wood Forest Deanna Topp Heidi Youmans Montana Department of Theodore Weaver Pilots: Corrections, Swan River R.G. White Mark Duffy Forest Camp Dave Whitesitt Bruce Wilkins Murray Duffy USDA Agricultural Research Howard Hash Service, Fort Keogh Livestock James Innes and Range Station Finally, we especially acknowledge Research and Technology Services Keith Iverson USDA Forest Service, Flathead Keith Kinden Supervisors, Eugene O. Allen, National Forest John P. Weigand and Terry N. Don Newton USDI Bureau of Land Larry Schweitzer Lonner for their enduring support Management, Lewistown and and assistance through the Keith Stevens Miles City Districts Roger Stradley studies, as well as their patience David Stradley USDI Fish and Wildlife and assistance in completing this Service, Charles M. Russell James Stradley (deceased) bulletin. National Wildlife Refuge USDI Cooperative Wildlife Research Unit, Missoula, MT

viii Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Contents

Forward...... iii Preface...... iv Acknowledgments...... vii List of Tables...... xii List of Figures...... xiii Historical Perspective...... 1 Statewide Deer Research Studies...... 5 Methods...... 7 Environments Studied...... 11 Mountain-Foothill...... 13 Timbered Breaks...... 15 Prairie-Badlands and Prairie-Agricultural...... 17 Plains Riverbottom...... 19 Northwest Montane Forest...... 21 Habitat Relationships...... 23 Concept of Habitat...... 25 Reproductive Habitat...... 25 Maintenance Habitat...... 27 Summer...... 27 Winter...... 28 Unused Areas...... 31 Habitat Selection...... 32 Species Adaptations...... 32 Process of Habitat Selection...... 34 Behavior and Habitat Use...... 35 Influence of Social Structure...... 35 Interaction of Behavior and Resource Requirements...... 38 Patterns in Habitat Selection...... 40 Distribution, Movements, and Home Range...... 40 Fine-tuning the Home Range...... 44 Home Range Size...... 47 Selection and Use of Vegetation...... 49 Forage Selection and Use...... 53 Activity Patterns...... 56 Social Organization...... 61 Population Characteristics and Dynamics...... 67 Concept of Population...... 69 Colonization and Development of Populations...... 70 Population Organization...... 73

Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a ix Population Characteristics...... 76 Population Size and Density...... 76 Sex and Age Composition...... 80 Sex Composition...... 81 Age Composition...... 84 Age Structure...... 86 Population Dynamics...... 94 Reproduction and Recruitment...... 94 Fawn Mortality...... 97 Factors Affecting Fawn Mortality...... 100 Adult Mortality...... 103 Adult Female Mortality...... 103 Adult Male Mortality...... 106 Emigration and Immigration...... 108 Patterns of Population Growth and Fluctuation...... 110 Management Implications...... 117 Traditional Concepts...... 119 An Ecological Perspective...... 122 Deer Population Dynamics and Hunter Harvest...... 123 Deer Management in Specific Ecosystems...... 129 Mountain Ecosystem...... 130 Description of Deer Population Ecology...... 130 Mule Deer Vital Parameters and Harvest Recommendations...... 132 Special Mule Deer Population Management Issues...... 133 White-tailed Deer Vital Parameters and Preliminary Harvest Recommendations...... 134 Habitat Management in Mountain Ecosystems...... 135 Housing Developments...... 135 Timber Management...... 136 Road Management...... 137 Habitat Enhancement and Vegetation Manipulation...... 137 Livestock Grazing...... 138 Prairie Ecosystem...... 138 Mule Deer Population Ecology in Timbered Breaks Environments...... 138 Mule Deer Vital Parameters and Harvest Recommendations...... 140 Habitat Management for Mule Deer in Timbered Breaks Environments...... 141 Vegetation Management...... 141 Livestock Grazing...... 142 Mule Deer-Elk Interactions...... 143 Access Management...... 143 Mule Deer Population Ecology in Prairie-Badland Environments...... 143 Mule Deer Vital Parameters and Harvest Recommendations...... 144 White-tailed Deer Population Ecology in Prairie-Agricultural Environments...... 145 White-tailed Deer Vital Parameters and Harvest Recommendations...... 146 Special Deer Population Management Issues in Prairie Environments...... 147 Habitat Management in Prairie-Badland and Prairie-Agricultural Environments...... 148 Riverbottom Agricultural Ecosystem...... 148 White-tailed Deer Population Ecology in Plains Riverbottom Environments...... 149 White-tailed Deer Vital Parameters and Harvest Recommendations...... 150 Habitat Management for White-tailed Deer in Riverbottom Agricultural Ecosystems...... 152 Future Directions in Deer Management...... 154 Adaptive Deer Management...... 154

x Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Deer Population Objectives...... 154 Harvest Regulations...... 155 Deer Population Monitoring...... 156 Alternative Models of Population Dynamics...... 159 Literature Cited and Appendix...... 161 Literature Cited...... 163 Appendix - A list of Publications Resulting from Statewide Deer Research Studies...... 174

Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a xi List of Tables

1. Post-hunting sex and age ratios for mule deer populations on three Montana ...... study areas...... 81 2. Post-hunting sex and age ratios for white-tailed deer populations on two ...... Montana study areas...... 81 3. Variation in potential productivity and fawn survival to autumn between ...... two subpopulations of white-tailed deer on the lower Yellowstone River, ...... 1980-85 (after Dusek et al. 1989)...... 95 4. Mule deer and white-tailed deer fawn recruitment in Montana. Data are ...... observed fawn: adult and modeled fawn:female ratios in spring (ave. ±1 SD)...... 98 5. Average and range in total, hunting, and natural mortality rates for adult ...... (≥1 year) female mule deer in three Montana environments...... 103 6. Annual survival rates and cause-specific mortality rates for yearling and ...... older female white-tailed deer under different harvest regimes in ...... three Montana environments...... 104 7. Total and cause-specific annual mortality rates for adult ≥( 1 year) male ...... mule deer and white-tailed deer in four Montana environments...... 107 8. Major environment types occupied by mule deer and white-tailed deer in Montana...... 129 9. Harvest rates by sex and age class and their general effects on white-tailed ...... deer population trend on the lower Yellowstone (after Dusek et al. 1989)...... 151 10. Aerial observability indexes measured from samples of marked deer in ...... various environments in Montana...... 157

xii Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a List of Figures

1. Locations of major deer study areas and boundaries of Montana Fish, Wildlife ...... and Parks administrative regions...... 6 2. Mountain-foothill environment, the Bridger Mountains study area: looking north (A) ...... and east across a major west slope winter range (B)...... 14 3. Timbered breaks environment, Missouri River Breaks study area: upland breaks (A) ...... and bottomland and adjacent slopes along the Missouri River (B)...... 15 4. Prairie-badlands (A) and prairie-agricultural (B) environments on the Cherry Creek ...... study area...... 17 5. Plains riverbottom environment: Elk Island (A) and Intake (B) study areas...... 19 6. Northwest montane forest environment: Swan Valley (A) and Salish ...... Mountains (B) study areas...... 21 7. Conceptualized distribution of reproductive and maintenance habitat and unused area ...... on the mountain-foothill (A), timbered breaks (B), and prairie-badlands (C) study areas...... 26 8. Diorama showing distribution of important mule deer habitat components as influenced ...... by topography and local climate across two mountain ranges in southwestern Montana...... 29 9. A mule deer in a typical stott...... 32 10. A white-tailed deer in a typical bound...... 32 11. Home range boundaries and observation sites for a mule deer matriarch and her adult ...... daughter including a June-August fawning territory (after Hamlin and Mackie 1989)...... 35 12. Conceptualized home ranges and fawning territories of a matriarch and her two adult ...... daughters and a granddaughter during the fawn-rearing period...... 36 13. Fidelity of four female mule deer to individual branches of a drainage system in a ...... prairie-badland environment. Solid dots represent locations of 2 deer that resided in ...... the lower fork and open circles the locations of two deer that resided in the upper fork ...... of the drainage over a 27-month period (Wood el al. 1989)...... 36 14. Seasonal home ranges and interseasonal movement of three generations of related ...... female mule deer in a mountain-foothill environment. Individual relocations during ...... migration indicate a shared movement corridor (after Pac et al. 1991)...... 37 15. Distribution of home range boundaries for a mature male and eleven adult females ...... (after Hamlin and Mackie 1989)...... 38 16. Yearlong distribution of the observations of a typical resident deer (after Pac et al. 1991)...... 41

Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a xiii 17. Yearlong distribution of the observations of a typical deer with adjacent seasonal ...... home ranges (after Pac et al. 1991)...... 42 18. Yearlong distribution of the observations of a typical deer with distinct seasonal ...... home ranges (after Pac et al. 1991)...... 42 19. Conceptualized home range patterns and accessory areas showing increasing ...... specialization in the use of space by season. Individual telemetry locations are ...... labeled by season: W=winter, S=summer,. F=fall, Sp=spring. Accessory areas are shaded...... 44 20. Generalized yearlong use of forage classes of mule deer and white-tailed deer in ...... different environments...... 54 21. Use of forage classes by fawn, adult male, and adult female white-tailed deer on the ...... lower Yellowstone River during periods of vegetative growth and dormancy, 1980-86 ...... (Dusek et al. 1989)...... 54 22. Mean distances moved between successive locations of deer during 24-hour tracking ...... sessions through summer and winter on the Elk Island and Intake units along the lower ...... Yellowstone River (Dusek et al. 1989)...... 57 23. Nocturnal home ranges compared to daytime and total seasonal home ranges of one ...... radio-collared female mule deer during summer on open sagebrush-grassland habitat, ...... northeast Montana (after Jackson 1990)...... 59 24. Nocturnal home ranges compared to daytime and total seasonal home ranges of ...... three radio-collared female mule deer during summer on open sagebrush-grassland ...... habitat, northeast Montana (after Jackson 1990)...... 60 25. A comparison of mobility based on monthly average activity radii (km) for adult ...... female mule deer on the west and east slopes of the Bridger Mountains. Asterisk ...... indicates significant differences between areas P < 0.05, one-way ANOVA ...... (after Pac et al. 1991)...... 61 26. Monthly mean size of four social groups of mule deer on the west slope of the ...... Bridger Mountains (after Pac et al. 1991)...... 62 27. Spatial segregation of social groups of mule deer at fawning time (mid-June) in ...... the Bridger Mountains...... 64 28. Spatial segregation of social groups of white-tailed deer at fawning time (mid-June) ...... along the lower Yellowstone River...... 65 29. Five levels in a series of natural grouping of mule deer in the Bridger Mountains...... 74 30. Comparison of distribution of reproductive and maintenance habitats and unused ...... areas in Brackett Creek (A) and Battle Ridge (B) population habitat-units in the ...... Bridger Mountains (after Pac et al. 1991)...... 78 31. Trend in numbers of adult males and females, male:100 female ratio, and percent ...... males in the Missouri River Breaks mule deer population, early winter 1960-61 ...... through 1986-87...... 82 32. Trends in numbers of adult males and females, male:100 female ratios, and percent ...... males in a mule deer population on the northwest slope of the Bridger Mountains, ...... early winter 1973-74 through 1996-97...... 83 33. Relative trends in post-hunting male:100 female and 4-point-male:100 female ratios ...... for the South 16-Mile mule deer population, Bridger Mountains, 1986-96...... 84

xiv Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a 34. Trends in estimated total numbers of adults and fawns, fawn:100 adult ratios, and ...... percent fawns in the mule deer population on the Missouri River Breaks study area, ...... spring 1961-1987...... 85 35. Trends in estimated total numbers of adults and fawns, fawn:100 adult ratio, and ...... percent fawns in the mule deer population on the Northwest Slope, Bridger ...... Mountains, spring 1972-1997...... 85 36. Trends in estimated total numbers of adults and fawns, fawn:100 adult ratios, and ...... percent fawns in the mule deer population on the Cherry Creek study area, ...... spring 1976-1987...... 86 37. Comparison of trends in number of fawns:100 adults on Bridger Mountain, ...... Missouri River Breaks, and Cherry Creek study areas during spring 1976-1987...... 86 38. Conceptualized examples of four types of age structures observed in Montana ...... mule deer: Type 1 = pyramidal, Type 2 = flat or low pyramid, Type 3 = convex, ...... Type 4 = concave or “U” shaped...... 87 39. Comparative age structures for adult females in four mule deer population-habitat ...... units in the Bridger Mountains illustrating the mix of different age structures possible ...... in adjacent populations during a given year (after Pac et al. 1991)...... 89 40. Age structural dynamics of adult female mule deer in the Bridger Mountains and ...... Missouri River Breaks. A = prior to severe winter, B = one full year following ...... severe winter, C = following several years of population growth (after Pac et al...... 1991 and Hamlin and Mackie 1989)...... 90 41. Age structural dynamics of adult male mule deer in the Bridger Mountains and ...... Missouri River Breaks: A = during or prior to severe winter, B = year following ...... severe winter, C = at population low, and D = following several years of population ...... growth or recovery (after Pac et al. 1991 and Hamlin and Mackie 1989)...... 91 42. Comparison of representative adult male and adult female age structures for ...... white-tailed deer on the Swan Valley and Lower Yellowstone River study areas ...... (MFWP unpub., Dusek and Mackie 1988)...... 92 43. Age structure of female and male white-tailed deer on the lower Yellowstone River ...... during autumn 1980-85 (Dusek et al. 1989)...... 93 44. Age specific in-utero fetal rates for white-tailed deer on the Lower Yellowstone River ...... and Swan Valley study areas...... 95 45. Age-specific production and recruitment of fawns by female mule deer in the ...... Missouri River Breaks, Montana, 1976-1984 (after Hamlin and Mackie 1989, 1991)...... 96 46. Age specific production and recruitment of fawns by female mule deer on the west ...... slope of the Bridger Mountains (after Pac et al. 1991)...... 96 47. White-tailed deer fawn survival by female age class to autumn on the lower ...... Yellowstone River and early winter in the Swan Valley...... 97 48. Instantaneous monthly mortality rate for radio-collared mule deer fawns in the ...... Missouri River Breaks, 1976-1986 (after Hamlin and Mackie 1989)...... 98 49. Comparative trends in total annual mortality of mule deer fawns in mountain-foothill, ...... timbered breaks, and prairie-badlands environments, 1973-74 through 1986-87...... 99

Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a xv 50. Comparative trends in mortality of mule deer fawns from June 1 through ...... November 30 in mountain-foothill, timbered breaks, and prairie-badlands ...... environments, 1975-76 through 1986-87...... 99 51. Comparative trends in mortality of mule deer fawns from December 1 through ...... May 31 in mountain-foothill, timbered breaks, and prairie-badlands environments, ...... 1975-76 through 1986-87...... 100 52. Modeled trend in overwinter mortality rate for mule deer fawns in the Missouri River ...... Breaks, 1960-61 through 1986-87...... 100 53. Trends in total annual and overwinter mortality rates for adult female mule deer in ...... the Missouri River Breaks, 1960-61 through 1986-87...... 105 54. Estimated numbers of mule deer on the Missouri River Breaks study area ...... during early winter 1930-1995...... 112 55. Spring population trend in relation to fawn recruitment and adult mortality ...... for mule deer on the Missouri River Breaks study area, 1961-1987...... 112 56. Spring population trend in relation to fawn recruitment and adult mortality ...... for mule deer associated with the Armstrong Range, Northwest Slope, ...... Bridger Mountains, 1974-87...... 113 57. Spring population trend in relation to fawn recruitment and adult mortality ...... for mule deer on the Cherry Creek study area, 1976-1987...... 114 58. Comparative annual changes in mule deer numbers in three Montana ...... environments during spring, 1974-1987...... 115 59. Comparative population trends for white-tailed deer on Swan Valley, ...... Cherry Creek, and Lower Yellowstone River study areas during early winter, ...... 1975-76 through 1986-87...... 116 60. Comparative trends in total numbers of mule deer and white-tailed deer on the ...... Cherry Creek study area during early winter 1975-76 through 1986-87 ...... (after Wood 1987, Wood et al. 1989)...... 116 61. Relationships between annual natural mortality rate and annual hunting mortality ...... of adult female mule deer (A) and annual survival rate and annual hunting ...... mortality rate of adult female mule deer (B) in the Missouri Breaks ...... (after Hamlin and Mackie 1989)...... 127 62. Relationships between fawn recruitment and total annual mortality of adult ...... females that maintain stable numbers of does in the population (A) and fawn ...... recruitment and total annual mortality of adult males that maintain stable ...... buck:doe ratios in the population (B) (after Hamlin and Mackie 1989)...... 128 63. Recommended mule deer doe harvest rates across the span of expected variation in ...... fawn recruitment and natural mortality of adult females in a mountain-foothill environment...... 133 64. Recommended mule deer doe harvest rates across the span of expected variation in ...... fawn recruitment and natural mortality of adult females in a timbered breaks environment...... 140 65. Recommended mule deer doe harvest rates across the span of expected variation in fawn ...... recruitment and natural mortality of adult females in a prairie/badlands environment...... 145 66. Recommended white-tailed doe harvest rates across the span of expected variation in fawn ...... recruitment and natural mortality of adult females in a prairie/agricultural environment...... 146

xvi Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Historical Perspective

Deer management in Montana has traditionally attempted to satisfy the requirements of deer for survival and provide maximum hunting opportunity. Beginning in the 1940s, management was based on a conceptual model in which the key elements were winter range, the quantity and quality of forage (i.e., browse) available on primary wintering areas, and deer numbers and distribution relative to these resources. This model was developed from early knowledge of deer biology and theories about population ecology. It assumed that deer populations were highly productive, inherently irruptive, and capable of overpopulating and overbrowsing their ranges unless controlled by hunter deer. Deer browsing was the primary harvest. Forage, particularly the winter factor affecting forage plant abundance browse supply, was assumed to be the and productivity and deer could degrade primary factor limiting populations. their own habitat. “Carrying capacity” was the number of Based on these concepts, deer a range could sustain in balance many biologists defined deer habitat Forage, with the forage supply. As deer numbers synonymously with the browse supply on particularly the reached or exceeded carrying capacity, winter range. Healthy, productive winter winter browse the amount and quality of winter browse forage produced healthy productive deer supply, was available to individual deer declined, populations, and the way to sustain both assumed to be resulting in widespread malnutrition and was through sufficient hunter harvest the primary death from starvation. Malnutrition also to maintain the most favorable balance factor limiting adversely affected reproduction, body between population size and the “habitat.” populations. size, and antler growth. Early concepts identified other It also was assumed that heavy potential limiting factors like predation, browsing had another, perhaps even more disease, parasites, and severe weather insidious, effect. Overuse of important that could limit deer numbers. However, plants resulted in a hedged appearance; their effect was considered more an browse plants declined in size and expression of an underlying nutritional productivity until they died and were problem that weakened and predisposed replaced by less nutritious and palatable deer to those factors rather than direct plants. In this manner, carrying capacity limitation. presumably could be reduced to the point Regulated human predation, or where the range would support fewer hunting, was not considered limiting

Hi s t o r i ca l Pe r s pe c t i v e 1 because it was not believed additive to agriculture, and logging provided other mortality. Instead, hunting mortality increased habitat diversity and high was believed to be “compensatory”, energy forage for deer. In some areas i.e., it replaced natural mortality which of eastern Montana, dwindling human would otherwise occur. It also reduced populations reduced disturbance and population density to promote increased restored natural habitat. survival and reproduction among Growing deer populations brought remaining deer. new, unprecedented problems and Within this conceptual framework, conflicts. The first three decades of the deer were “deer,” i.e., mule deer, century witnessed low deer numbers and white-tailed deer, bucks and does, development of a protectionist mindset adults and young, were essentially the among hunters and landowners. By same organisms from a management the early 1950s, “overabundance” and perspective. Hunting and harvest were depredations on agricultural, range, and the primary tools for both population forest lands brought need for expanded and habitat management as well as the harvests for population control. To primary measure of success in deer conservation-minded sportsmen, however, management. The task of managers was the notion of liberalizing hunting to develop harvest strategies and sustain regulations from bucks-only to either sufficient hunter harvests to maintain sex or doe seasons was unthinkable. healthy, productive forage and deer Thus, the new “management model,” populations on winter range. though offering increased opportunity for The framework also simplified recreational hunting and greater hunting- harvest success, faced formidable social Regulated human management because biologists could focus surveys on a small portion (usually barriers. predation or <20 percent) of the total yearlong Theory and practice embodied hunting was not range of deer. Further, an elusive and in this winter forage limiting model considered limiting controversial estimate of the total provided biologists with an objective, because it was not numbers of deer on a range or in a scientific basis to address the problems believed additive population was not necessary. One of overabundance. It also gave new to other mortality. needed only measure utilization and opportunities for management. Instead, hunting condition of “key” browse plants on “key” Few wildlife management programs mortality was winter range areas to determine whether have been pursued as aggressively believed to be deer populations were too high, about or implemented so widely as “ deer “compensatory” right, or too low in relation to carrying management based on facts” (Cole 1958, capacity. Other databases (such as 1959, 1961; Newby 1958). The history trend in number of deer harvested and and results of the program from the fawn production), which were assumed early 1950s through the early 1970s are to be directly related to utilization and documented in deer hunting regulations condition of winter browse, also could and harvest records maintained since be sampled with reasonable ease and 1945, as well as in reviews of deer accuracy. management from 1941 through 1970 by By the 1940s and early 1950s, Allen (1971) and Egan (1971). concurrent with these developing Efforts to expand hunting concepts, deer populations were opportunity and reduce burgeoning deer expanding almost explosively. Restrictive populations through special either sex or hunting seasons severely limited deer “doe” hunting seasons began in western harvests locally and statewide. Major Montana in 1951 and 1952. Five years predators (wolves, mountain lions, and later, almost the entire state was opened coyotes) had been reduced or eliminated to general hunting of two deer of either from their natural ranges in Montana. species, either sex, during a one-month Favorable habitat conditions developed season. In addition, nonresident hunting in association with the end of the 1930s was expanded by offering $20 either-sex drought. Changes in livestock grazing, permits in many hunting districts. Areas

2 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a with severe overbrowsing or agricultural depredations were frequently delineated for special early or late seasons allowing harvest of additional antlerless animals. Under this management strategy, which represented an intensive statewide effort to apply “sustained yield harvest theory” to deer populations in Montana, harvests increased from less than 40,000 deer annually prior to 1952 to over 100,000 by 1955. Annual harvests exceeding 100,000 deer were sustained into the early 1970s with few exceptions. During much of this period, antlerless deer constituted 25 to 40 percent of the statewide harvest. By conservative estimate, total legal harvests probably removed no less than 10 percent of autumn populations statewide (a where ranges had deteriorated and deer statewide harvest of 100,000 deer would, numbers had declined. Yet, few deer at 10 percent, require an average density managers were skeptical of the simple, of about 3 deer per square kilometer cause-effect relationship between deer across the entire state, or an average of 4/ and their habitat. km2 on two-thirds of all land in the state). Research initiated in 1970 to Check station data and statewide harvest evaluate range surveys confirmed survey estimates for individual hunting numerous technical, analytical, and districts indicated much higher than conceptual problems with the program. Under this average harvest rates occurred locally, Range survey data had only limited management especially in areas of special management utility in expressing real trends in plant strategy, which concern. utilization and condition; in many cases probably represents Unfortunately, this aggressive they were inaccurate (Mackie 1975). The the best statewide practice of deer management through findings also raised questions about the effort ever liberal hunting seasons did not sustain general applicability of basic concepts attempted to apply deer populations or harvests. Change about deer-habitat interactions; especially “sustained yield was on the way as mule deer numbers the simple, direct relationship between harvest theory” to began to decline in Montana and across deer populations and the supply and deer populations, much of western North America during condition of key browse plants on key Montana deer the late 1960s and early 1970s. winter ranges. harvests increased As early as the mid-1960s, biologists Where data were available, browse from less than and others began questioning some utilization and condition trends were not 40,000 deer related to deer population trends. Winter of the concepts and practices that annually prior browse supply was only one of many were broadly applied when deer were to 1952 to over extremely abundant. Some questioned factors influencing deer populations, and 100,000 by 1955. the effectiveness of range survey methods deer use was but one of many factors and criteria for interpreting deer range influencing plant populations and forage conditions. Others believed that not supplies (Mackie 1973, 1975). Winter enough deer were killed to achieve a concentrations of deer and heavy use balance between deer populations and of browse probably revealed only available range. As a result, new studies where deer ended up and what they were requested to provide an ecological subsisted on under desperate conditions basis for determining range condition (Carpenter and Wallmo 1981). Winter and trend in the statewide management concentrations were not indicative of an program and to determine the degree of overriding importance of winter relative harvest necessary to maintain the balance to other seasonal ranges.

Hi s t o r i ca l Pe r s pe c t i v e 3 Widespread declines in mule deer Deer science lacked a broad, long-term populations during the early-mid 1970s population perspective that included came as a shock to most biologists. importance of the interaction among Theoretical concepts and principles of factors. Most deer management deer management could not explain the principles were derived from short-term decline (MCTWS 1975, Workman and studies on a few problem areas at a time Low 1976). Pengelly (1976) indicated when deer populations were abundant that the effects of nearly all limiting or from controlled experiments in pens. factors including hunting on deer Basic theory and management concepts populations were poorly understood. about population ecology and sustained Specific findings concerning limiting yield harvest were largely untested. factors were confusing or contradictory.

4 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Statewide Deer Research Studies

Studies on deer in Montana during and environments would provide basic the 1950s and 1960s were designed knowledge about deer and insight to primarily to identify problems and provide important management questions. For information on food habits and winter example: range use. A few evaluated the efficacy of • what constitutes deer habitat? management programs and concepts, as • how do deer select and use habitat and in the case of browse surveys. However, adapt to habitat variation? these studies were not sufficient to explain deer population phenomena • what constitutes habitat condition/ nor to answer concerns about existing habitat quality? management theory and practices. • how can condition trends be measured Because of this, comprehensive effectively? studies on the population ecology of both • what constitutes a deer “population”? species in representative Rocky Mountain • how are populations organized and and Great Plains habitats were initiated maintained over time? during 1975. Established upon a base • how do populations vary in space and of earlier deer research, the new studies time? were designed to: • how does the interaction of natural • provide more detailed knowledge and mortality and hunting influence an improved understanding of the population dynamics? biology and population ecology of mule The research embraced spatial deer and white-tailed deer in Montana and temporal scales previously avoided • develop new or improved methods for in studies on deer. Included were six managing mule deer and white-tailed intensive investigations in the major deer populations and habitats, and ecological types occupied by deer in • establish new guidelines for Montana (Fig. 1). All were full-time consideration of mule and white-tailed field studies conducted concurrently for deer in other wildlife, range, forest, and periods ranging from 7 to more than land management programs. 20 years. Additional, comparative data A long-term, comparative were available from earlier research and evaluation of deer-habitat interactions numerous less intensive, shorter term and population ecology among species studies throughout the state.

St a t e w i d e Dee r Re s e a r c h St u d i e s 5 Figure 1. Locations of major deer study areas and boundaries of Montana Fish, Wildlife and Parks administrative regions.

6 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Methods

Similar methods were employed on all areas, though some were modified in accord with local logistical constraints and individual study objectives. Technology available for field investigation and data analysis changed markedly through time, and we upgraded our methods during the work. On each of the primary study areas we endeavored to define: • characteristics of the habitat in terms of physical and biotic attributes • deer behavior or use of the area (distribution, movements, home range size and shape, use and selection for vegetation/cover types, activity patterns, and food habits) • biological attributes of deer (growth patterns, body size, condition, example, after studies in the Bridger longevity, reproduction, and mortality) Mountains indicated that mule deer • deer population characteristics were distributed in seven, relatively (size, sex and age structure) and discrete “population-habitat units,” dynamics (annual recruitment and habitat investigations were redirected to adult mortality rates, immigration, and comparative analysis of environmental emigration) over time and across the features and patterns of variation within range of environmental variation that and among units. Also, early analyses occurred. were relatively simple and utilized Details of methods used are given in very general measurements. Others final reports and publications for each of completed later in the studies were more the studies (e.g., Dusek et al. 1989, Wood complex and utilized detailed databases et al. 1989, Hamlin and Mackie 1989, Pac generated by computers and GIS mapping et al. 1991). Only a broad overview is technology. presented here. All of the studies relied on radio- All study areas were described in collared and other individually marked terms of geographic location, topography, deer to define habitat relationships and climate and weather, vegetation, major assess population characteristics and fauna, and land uses. Environmental dynamics. Deer were captured and descriptions, the system or intensity of marked with radio collars or neckbands mapping, and mode of analysis varied using various techniques that included among studies and over time. For bait trapping in corral and Clover traps,

Me t h o d s 7 Mackie (1989), and Pac et al. (1991), seasonal distributions and trends in population size and composition were determined over periods ranging from 7 to 35 years on major study areas. Our use of complete-coverage surveys eliminated possible bias resulting from sampling design in population estimates. Flown by pilots and observers experienced in aerial deer surveys, the counts and classifications always represented the minimum numbers of deer and sex/age classes on study areas. To develop reasonable total population estimates, we had only to determine the accuracy of our counts. However, in most cases this approach precluded opportunity to calculate confidence limits rocket netting, chemical immobilization, around annual population estimates. helicopter net gunning and drive netting, To account and adjust for visibility and hand capture of fawns. Collectively, bias, we developed observability indexes we captured and marked approximately (estimates of proportions of total deer 2,500 mule deer and 1,600 white-tailed observed) relative to study area/habitat, deer on the primary study areas from season, survey conditions, aircraft, 1975 through 1995. Of these, 880 mule and observer. These indexes, based on deer and 355 white-tailed deer were proportions of marked and radio-collared equipped with radio collars. Radio- deer observed, were generally consistent collared individuals were relocated by from year to year within study areas and periodic monitoring both from the ground seasons when the same pilot and observer and from the air; neck banded deer were were used. observed as opportunities permitted. To further strengthen population Most monitoring occurred during estimates, data on population daylight hours, but triangulation at hourly composition were applied in arithmetic intervals provided data on nighttime movements and activity over 24-hour periods in some areas. The development of highly reliable, long lasting radio transmitters enabled individuals and groups of deer to be monitored over several years. Some deer were recollared several times and followed for up to 13 years. Intensive aerial surveys, employing fixed-wing aircraft and helicopters, also were used in determining habitat use and population characteristics and dynamics of deer in all areas except the densely forested whitetail habitats in northwestern Montana. Following procedures outlined by Mackie et al. (1981), Dusek et al. (1989), Wood et al. (1989), Hamlin and

8 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a population models (Mackie et al. 1981) field. Ages were assigned in the field to reconcile any differences in estimates using tooth eruption and wear criteria between seasons and years. Deer harvest (Severinghaus 1949, Robinette et al. and mortality patterns and rates among 1957). When jaws or incisors were marked deer also were employed in obtained, dental cementum analysis (Low modeling and confirmation of population and Cowan 1963, Gilbert 1966) also was estimates. employed to ascertain age. Weights, Ground surveys supplemented aerial antler, and other measurements of surveys. They provided additional data on biological characteristics and condition all study areas and provided the primary (Riney 1955, Greer 1968, Verme and data on population characteristics and Holland 1973) were also obtained. dynamics of white-tailed deer in densely When possible, blood samples were forested northwestern Montana habitats. collected from deer handled in traps In recent years, camera surveys (Dusek or captured as fawns. During January- and Mace 1991, Dusek and Morgan June, reproductive tracts were removed 1991) were also employed successfully to from any female mortalities to evaluate evaluate whitetail population parameters reproductive performance. Measurement and habitat relationships in the Salish of serum concentration of progesterone Mountains. and pregnancy-specific protein B from Deer harvest rates were estimated peripheral blood (Wood et al. 1986) by marked and radio-collared deer. These provided additional estimates for data were corroborated by hunter check pregnancy rates. Rumen contents were stations, field checks, questionnaires, and sampled throughout the year for analysis the Montana Fish, Wildlife and Parks’ of food habits. statewide deer harvest survey. Special Analytical procedures and dead-deer surveys were conducted on methodology varied according to needs of some areas during spring. individual studies. Statistical procedures Sex, age, and various measurements generally followed Zar (1984). Most were recorded for all deer captured, analyses were conducted using Montana checked at stations, or examined in the State University computing services and field. Whenever possible, lower jaws a variety of computers, software, and or incisors were collected from deer statistical packages such as SAS (Ray examined at check stations or in the 1982) and MSUSTAT (Lund 1983).

Me t h o d s 9

Environments Studied

Mountain-Foothill

western footslopes to an estimated Studies in the mountain-foothill 127 cm along the Bridger Divide, then environment centered in the Bridger declines progressively to the east and Mountain Range. The Bridgers are north to 35-40 cm. a representative, semi-isolated range In aspect, the Bridger Range and located at 45˚53’ north latitude, 110˚53’ attending ridges and foothills comprise west longitude, on the eastern flank of an “island” of montane forest within a the Rocky Mountains in southwestern “sea” of lowland steppe (Pac et al. 1991) Montana (Fig. 1). Together with adjacent (Fig. 2A). The steppe, dominated by footslopes, the area encompasses about open grass and shrub-grass communities, 2,000 km2 and includes most of the covers approximately 60 percent of the topographic, climatic, and vegetational area. Montane forest, dominated by open variation characteristic of mountain- to dense stands of several conifer species, foothill environments in Montana (Pac et covers 38 percent of the area within an al. 1991). elevational range of 1,830 m to 2,700 m. The Bridger Mountain Range Highest elevations, above 2,400-2,700 m is dominated by a north-south along the main Bridger Range divide, are trending mountain divide that extends characterized by a subalpine-alpine zone approximately 40 km in an arcuate covering 2 percent of the total area. pattern along the west flank. Three Most of the Bridger Mountain study attending ridge formations extend area, including nearly all of the steppe easterly from the main divide to dominate zone and about one-half of the foothill the eastern flank and about two thirds area, is privately owned. Lands above the of the total area. Elevations vary from lower limit of forest are predominantly 1,365-1,630 m along lower footslopes to in public ownership administered by 2,400-2,947 m along the main Bridger the Gallatin National Forest. Along the Range and 2,100-2,400 m at highest east flank of the main Bridger Range points along the eastern ridges. Overall, ownership is in a checkerboard pattern the west flank is characterized by high with alternating sections of private timber topographic relief and short, steep-sided lands. Grazing, dryland grain farming, drainages; the east flank is lower and less hay production, timber harvest, and rural severe, with long drainages descending residential development are the primary gradually through open benchlands, uses of private lands. National forest timbered foothills, and gently rolling lands are managed for timber harvest, footslopes. livestock grazing, and recreation. The Bridger Mountains experience Early concerns about mule deer short, cool summers and long, cold populations and damage to agricultural winters, but local climates and weather products led to a study of food habits, patterns vary greatly. Average annual range use, and agricultural relationships precipitation increases sharply with on the west flank of the Bridger Range elevation from 40-45 cm along the (Fig. 2B) during 1955-1956 (Wilkins

Mo u n t a i n -Fo o t h i l l 13 1957). This was followed by special and range survey methods under the studies to develop methods for deer range statewide range research project. All of surveys during 1957-1959. During 1971- these provided background and baseline 76, five graduate thesis projects and other data for further, long-term research special studies were conducted to further on population ecology of mule deer evaluate mule deer habitat relationships beginning in 1975 (Pac et al., 1991).

Figure 2. Mountain-foothill A environment, the Bridger Mountains study area: looking north (A) and east across a major west slope winter range (B).

B

14 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Timbered Breaks

Timbered breaks occur within A a 10-50-km-wide by 300-km-long belt of rugged badlands along the Missouri River and its tributaries in northcentral Montana. The “breaks” are characterized by closely interspersed open ridges and sharply-cut drainageways or “coulees” that dissect the shale substrates of the area in a dendritic pattern and become progressively wider, deeper, and more steeply sloped as they approach the river. Our studies centered on a representative 275 km2 area located at 47˚ 30’ north latitude, 108˚ 30’ west longitude, about 40 km northeast of Roy in central Montana (Fig. 1). This area, described in detail by Mackie (1970) and Hamlin and Mackie (1989), extends B about 30 km in a 7-11-km-wide band along the south side of the Missouri River. Elevations range from about 945 m on rolling plains along the southern edge of the area to about 685 m on the Missouri River floodplain. The varied breaks topography and soils support a complex mosaic of open low shrub-grass and timbered vegetation types that impart a savannah-like aspect (Fig. 3A). Forested types cover about 50 percent of the area in scattered, open and medium density stands of coniferous trees and shrubs along the side-slopes of drainages. Riparian forest, dominated by deciduous trees and shrubs, is restricted to Missouri River bottomlands (Fig. 3B). Low shrub and grass dominated Figure 3. Timbered breaks environment, Missouri River Breaks study area: vegetation covers most of the remaining upland breaks (A) and bottomland and adjacent slopes along the area, including ridgetops, coulee bottoms, Missouri River (B). benches, and some steep south-facing slopes.

Ti m b e r e d Br e a k s 15 The climate is semiarid, Timbered breaks have long been characterized by moderately low and recognized as important habitat for mule variable precipitation, low to moderate deer. Although deer populations in the snowfall, low relative humidity, moderate vicinity of our study area declined to to strong winds, and great extremes in extreme scarcity during and following the temperature. Variation in all weather homestead era, they recovered during the factors is the rule; it influences wide late 1930s and 1940s. The increasing fluctuations in both growing season and populations focused attention on mule winter conditions. deer by the mid 1940s, and aerial and As a result of these environmental other surveys provided data on population factors, the breaks are primarily characteristics and trends from 1947 to rangeland. Approximately 75 percent 1952. Concern for possible competition of the area is in public ownership: 45 among mule deer, elk, and cattle led percent lies within the Charles M. Russell to an intensive study of interspecific National Wildlife Refuge, 25 percent relationships during 1960-64 (Mackie is administered by the Bureau of Land 1970). These studies initiated aerial Management, 5 percent is state land, surveys to determine early winter deer and 25 percent is privately owned. The and elk population characteristics and human population is low and use of the trends through 1974-75. area is largely related to livestock grazing and recreation.

16 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Prairie-Badlands and Prairie-Agricultural

Studies in the prairie badlands and A prairie-agricultural environments were conducted on a 543 km2 area centered at 47˚ north latitude, 106˚ west longitude, approximately 20 km northwest of Terry, in eastern Montana (Fig. 1). Described in detail by Wood et al. (1989), the Cherry Creek study area extended about 40 km east to west and 23 km north to south, spanning the drainage divide between the Yellowstone and Missouri Rivers. The most prominent topographic feature is Big Sheep Mountain which rises 90 m above the divide to an elevation of 1,096 m. The terrain slopes gradually downward from either side of the divide to the lowest elevation (771 m) on the southeast boundary. Drainages are relatively steep and narrow, resulting in a B badlands aspect near the divide, but they gradually widen and flatten along their length and develop distinct floodplains toward the perimeter of the area. The area is characterized by large tracts of open grassland that dominate flat or rolling terrain over 65 percent of the study area (Fig. 4A). Sparsely vegetated badlands and bunchgrass covered hills occur over about 25 percent, mostly in drainage heads. Sagebrush coverage in badlands and grasslands is generally sparse and individual shrubs are typically less than 50 cm tall. Deciduous trees and tall shrubs occur in linear stands along draws through grassland and other habitats. Patches of snowberry and rose occur as long narrow bands in draws Figure 4. above and below the hardwood stands. Prairie-badlands (A) and prairie-agricultural (B) environments These shrub and deciduous woodland on the Cherry Creek study area. communities collectively cover only

Pr a i r i e -Ba d l a n d s a n d Pr a i r i e -Agricultural 17 about 7 percent of the area. Agricultural that included the study area, while croplands (Fig. 4B), consisting of dryland white-tailed deer were locally abundant grain and legume hay fields, are small along some major river drainages and (<260 ha) and scattered over less than tributaries. Populations of both species 4 percent of the area. Livestock grazing declined with increasing human presence and crop production dominate human in eastern Montana from about 1870 use on the sparsely populated area that through the early 1920s, but expanded includes about 55 percent federal (BLM) again during the 1940s and 1950s. lands, 6 percent state lands, and 39 By 1950, mule deer were sufficiently percent private land. abundant to damage agricultural crops in The climate of the area, like that of the region, while whitetails first appeared the Missouri River Breaks, is semiarid on the study area around 1954 (Wood et and continental, marked by extreme al. 1989). Because the area was typical fluctuations in seasonal and annual of the plains environment occupied temperature and precipitation. Snowfall by both species and was considered also is variable, but normally moderate important for deer management in eastern with significant accumulations occurring Montana, it was selected for comparative only rarely. ecological studies of both mule deer and Historically, mule deer were common white-tailed deer in prairie environments in portions of the northern Great Plains beginning in 1975.

18 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Plains Riverbottom

The lower Yellowstone River A extends 350 km from the mouth of the Bighorn River east of Billings to the confluence of the Yellowstone and Missouri Rivers in western North Dakota. Our study spanned approximately 224 km2 of floodplain and islands between Glendive and Sidney, centered at approximately 47˚ 30' north latitude, 104˚ 30' west longitude, in eastern Montana (Fig. 1). Physiographic features of the area include rolling uplands, alluvial deposits, and a terraced floodplain to the northwest of the river (Fig. 5); and high benches and/or rugged badlands immediately adjacent to the river on the south side. Within the study area, the floodplain was relatively wide, varying from about 2 km B above Intake on the upper one-third of the area to 7 km at the lower boundary near Sidney. Elevations on the river varied from 625 m at Glendive to 577 m at Sidney. Similar to other major riverine environments in the northern plains, natural floodplain vegetation is dominated by the willow-cottonwood-shrubland- grassland sere. This sere originates with willow and cottonwood seedlings becoming established on newly formed sand or gravel bars deposited by annual flooding during May and June. The stands change progressively over some 100 years or more through willow, Figure 5. Plains riverbottom environment: Elk Island (A) and young cottonwood, mature cottonwood, Intake (B) study areas. decadent cottonwood-shrub, and shrub-dominated communities to relatively permanent grasslands. Where undisturbed by natural catastrophe such

Pl a i n s Ri v e r b o t t o m 19 as ice jams during spring runoff or land Cultivated lands on upland terraces and use for agriculture, complexes of these benches above the floodplain support communities dominate and provide a very dryland cereal crops. diverse vegetative aspect. Scattered small During the past century, riverine stands of green ash or other deciduous environments in the northern plains trees may provide further diversity. have come to represent habitats of high Land ownership was predominantly complexity, diversity, and stability within a private except for two parcels owned region characterized by relative simplicity and managed by the MFWP for wildlife and high variability in environmental habitat and recreation. On private lands, conditions (Dusek et al. 1989). High agriculture supported by irrigation of complexity and diversity are provided the floodplain and adjacent terraces by interspersion of small units of many downstream from Intake was the different vegetation types or communities, dominant land use. Croplands occurred land uses, and agricultural practices as variable-sized fields interspersed with along the floodplain. Because of this, or adjacent to stands of native floodplain areas such as the lower Yellowstone vegetation. Sugar beets, alfalfa, corn, and often support extremely high density small grains were the principal irrigated populations of white-tailed deer that crops. Untilled areas were generally are important in terms of hunting and grazed by livestock as were tilled areas agricultural relationships. during autumn and/or winter. Production As a representative bottomland of livestock, forage crops, and small habitat for which baseline data on deer- grains were principal land uses on non- habitat relations were required to evaluate irrigated portions of the floodplain and agricultural relationships and water above Intake. Most of these areas were resource allocation within the Yellowstone grazed during autumn and winter as well Basin (Swenson 1978), the area was as during abnormally dry summers when selected for intensive studies of white- forage production on uplands was poor. tailed deer beginning in 1979-80. Some Adjacent to the floodplain, background data for deer populations and rolling uplands are dominated by open habitats in the area were available from mixed grasslands, while badlands are studies conducted to evaluate impacts characterized by steep, sparsely vegetated of off-stream water impoundments on slopes with stands of juniper along wildlife on a portion of the area near some side drainages. More mesic draws Intake during 1976-77 (Swenson 1978). support linear stands of deciduous trees.

20 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Northwest Montane Forest

The Swan Valley extends from the A Swan-Clearwater divide north to Swan Lake and is bounded by the high and steep-sided Mission and Swan Mountain Ranges to west and east, respectively. Centered at approximately 47˚ 35' north latitude, 113˚ 45' west longitude, the valley is 64 km long and 10-16 km wide (Fig. 1). Elevations range from 900 to 1,300 m on the valley floor and from 2,100 to 3,100 m along the bordering mountain ridges. The Swan Valley is dominated by a subclimax conifer forest that extends the length and breadth of the area (Fig. 6A). Natural openings are few and associated mainly with marshy areas around lakes and ponds and along portions of the Swan River. Other openings occur as a result B of human developments on private land along the valley bottom. Collectively, openings comprised only about 6 percent of the area. Stand regenerating events, including logging, the major commercial land use, have occurred over 28 percent of the area since 1900. Most logging has occurred since 1960 and most wildfire occurred prior to 1920. Logging and other timber management patterns vary with the checkerboard land ownership pattern wherein sections of national forest land alternate with sections of private timber company land and state forest land throughout the valley. Harvest units are mostly small but include some up to 260 ha in size (Flathead National Forest Figure 6. 1994). Northwest montane forest environment: Swan Valley (A) and The climate of the Swan Valley is Salish Mountains (B) study areas. characterized by moderate summer and winter temperatures. Much of the annual

No r t h w e s t Mo n t a n e Fo r e s t 21 precipitation occurs as snow in winter of white-tailed deer in northwestern such that accumulations are greater and Montana expanded to densely forested endure longer than in eastern Montana or mountain foothill and valley habitats southwestern valleys. along the east face of the Salish Mountain The Swan Valley is historically Range northwest of Kalispell. Centered important white-tailed deer habitat and at approximately 48˚ 30' north latitude representative of the extensive conifer and 114˚ 30' west longitude, the primary forested mountain valley environment in study area included approximately 480 which the species occurs in northwestern km2 in and adjacent to the Tally Lake Montana. The area supports large Ranger District of the Flathead National populations of deer that provide Forest (Morgan 1993) (Fig 1.). significant recreational hunting and are Topographically, this area is an important consideration in silvicultural dominated by low to moderate elevation practices and other aspects of forest peaks and mountain ridges dissected by management. several major drainages (Fig. 6B). Slopes The importance of the deer resource are fairly moderate throughout within and possible impacts of deer on timber an elevational range of 915-1,935 m; 60 production focused interest on the percent of the area lies between 1,281 area in the early 1940s, when the Swan and 1,646 m. Valley Deer Study became one of the Similar to the Swan Valley, the first investigations developed under the climate is strongly influenced by moisture- Fish and Game Commission’s new policy laden air from the Pacific northwest to obtain scientific data as a basis for that imparts relatively mesic conditions wildlife management. A graduate thesis yearlong. Consistent with this moisture study (Hildebrand 1971) on the biology regime, over 90 percent of the area is of white-tailed deer on winter ranges in covered by conifer forest with only a the Swan during 1969 and 1970 also few natural grass and shrub openings. provided background for selection of this Current vegetation is a mixture of mature area for intensive studies on population conifer trees, cut-over areas in various ecology of whitetails in 1975 (Mundinger stages of regeneration, riparian areas, and 1981, 1984). natural willow/grass meadows. Most of The lower valley, from Condon north the open riparian shrub/grass vegetation to Goat Creek, provides critical winter is centered in two extensive wet meadow range for white-tailed deer throughout complexes in the central and northern the valley as well as for deer that migrate portions of the area. Timber production seasonally from the Clearwater River and recreation are the primary land uses drainage. Portions of the valley south on the area that consists almost entirely of Condon are more highly interspersed of national forest lands. Less than 10 with small lakes, ponds, marshes and percent is privately owned. other mesic/riparian complexes that The east slope of the Salish provide high quality summer habitats but Mountains is an important white-tailed accumulate excessive amounts of snow in deer habitat complex that, like the Swan late fall and winter. Because of patterns Valley, is subject to extensive logging of deer use, our intensive studies focused and silvicultural activity. Thus, it offered on winter habitat relationships; summer opportunity for comparative study habitat selection and use was evaluated of habitat relationships of whitetails under a supporting graduate thesis associated with somewhat more mesic research effort (Leach 1982). upland habitats on summer as well as During 1981-82, studies of habitat winter range. relationships and population ecology

22 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Habitat Relationships

The Concept of Habitat

Habitat can be defined as “. . .the lactation. Thus, they were generally resources and conditions present in an diverse, mesic environments dominated area that produce occupancy–including by a variety of forbs or agricultural survival and reproduction–by a given crops during late spring and summer, organism” (Hall et al. 1997). Our findings though succulent growth of shrubs and generally support this definition and its grasses also was important. Complex application to deer. topography that included a diversity In our perspective, deer habitat of microsites capable of providing ...reproductive is not simply a place with food, cover, alternative sources of some palatable habitat constituted water, and space; nor is it primarily green vegetation through summer also environments with vegetation or vegetation structure. was characteristic. Reproductive habitat resources required Rather, we recognize habitat as areas that, provided opportunity for isolation from for recovery of based on their physical and biological other deer, security from predation, and physical condition characteristics, provide for functions minimal competition with other wild and successful contributing to the survival of individuals, ungulates and domestic livestock. High reproduction by populations, and species. These functions quality reproductive habitat enabled deer. include reproduction and maintenance. reproducing females to maintain or regain Reproductive and maintenance habitats body condition and energy reserves prior may vary considerably in structure, form, to breeding and the onset of winter. and mix of components across species’ In mountain-foothill environments, distributions. They also vary seasonally reproductive habitat for mule deer within areas. They may be interspersed occurred most extensively in diverse, within a single yearlong range, overlap mesic montane forests at intermediate seasonally, or comprise discrete seasonal elevations (Fig. 7A). High elevation, ranges for deer. However, the fact that they provide for successful reproduction and/or maintenance is the common link for comparison of deer behavior and habitat use among areas.

Reproductive Habitat

On our study areas, reproductive habitat constituted environments with resources required for recovery of physical condition and successful reproduction by deer. Specific attributes varied, but all reproductive habitats provided dependable sources of succulent, high quality forage during fawning and

The Co n c ep t o f Ha b i t a t 25 subalpine-alpine habitats lacked some vegetationally diverse habitats that of the resources necessary to sustain provided succulent forage and visual/ adult females and young. Habitats that spatial isolation from other deer during consistently provided succulent forage and fawning to late summer. Moderately other resources essential to reproducing steep, northerly exposures dominated by females also were limited or patchy in dry diverse, mesic vegetation that sustained foothills and other low elevation habitats. some succulent forage through summer Reproductive habitat for mule deer were particularly important. In some was limited and patchy in timbered breaks prairie environments, alfalfa hayfields (Fig. 7B), prairie-badlands (Fig. 7C), in or adjacent to relatively steep terrain and prairie-agricultural environments. contributed to the occurrence of There, local areas used by adult females reproductive habitat where little or none and fawns were topographically and would otherwise have been available.

Figure 7. Conceptualized A distribution of reproductive and maintenance habitat

B ri and unused area on dger Divide the mountain-foothill (A), timbered breaks (B), and prairie- badlands (C) study areas.

B

i s s r i R M o u i v e r

Reproductive Maintenance - winter Maintenance - summer Unused C

C h e

r r y C r.

26 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a For white-tailed deer, reproductive Summer habitat typically included riparian features Summer maintenance habitat such as lakes, ponds, marshes, rivers, or provided opportunity for males and streams. These habitats usually were of nonproductive females to grow and low to moderate relief, but vegetationally recover physical condition. The supply diverse such that they provided of succulent, high quality forage was abundant succulent forage from spring inadequate to sustain the additional to late summer. Agricultural croplands, Maintenance demands of lactation and reproduction habitat consisted especially alfalfa fields, in close proximity in many years. Summer maintenance of environments to riparian cover were an important habitats tended to be drier, less diverse, component of reproductive habitat for and more variable environments than that provided whitetails in some plains environments. reproductive habitats. Also, risk of all resources predation and/or competition may have necessary for been comparatively greater than in adult survival, but Maintenance Habitat reproductive habitat. not necessarily Summer maintenance habitats recruitment of for mule deer in mountain-foothill young. Maintenance habitat consisted of environments included subalpine-alpine environments that provided all resources and shrub-grass steppe habitats above and necessary for adult survival, but not below the montane forest zone, as well as necessarily recruitment of young. It some dry interspersed ridges and slopes. included summer habitat suitable for In timbered breaks and prairie-badland sustaining males and nonproductive environments, they consisted of dry, females and winter habitat for all deer. open habitats of both low and extremely high relief, including gently sloping

The Co n c ep t o f Ha b i t a t 27 drainagehead and ridgetop areas subject were created by the interaction of to more intensive livestock grazing. geographic location, topography, climate, For white-tailed deer in northwest and vegetation. Local site characteristics montane forest environments, summer determined the specific location, size, and maintenance habitats characteristically shape of each winter range and patterns included higher, drier areas with shorter of deer dispersion within it. Vegetation growing seasons than reproductive structure and composition were typically habitats. In plains riverbottom only third order factors for mule deer. For environments, terrace rangeland white-tailed deer, vegetation augmented interspersed by hardwood draws and other site factors in minimizing snow mature cottonwood forests, often depth and determining the location, size, subjected to continuous livestock grazing, and shape of winter habitats. served as maintenance habitat. A dioramic cross section extending Although reproductive and west to east through the adjacent summer maintenance habitat together Bridger and Crazy Mountain Ranges constituted summer range for deer on (Fig. 8), illustrates how interaction our study areas, they are not conceptually among environmental factors influences equivalent. Summer range traditionally occurrence of winter maintenance has been perceived as the place deer live habitats for mule deer. Through from spring through autumn—essentially geologic processes, the Gallatin Valley a homogenous, unlimited summer floor dropped in elevation while the pasture. Its boundaries are fixed only adjacent surface uplifted and folded to form the Bridger Range. As a result, by the distance deer move from winter the west flank of the Bridger Mountains range, and may encompass both habitat consists of relatively steep west to south and areas unused by deer. In most areas, facing slopes at elevations between summer habitat is considered unlimited as approximately 1,525 m and 1,825 m. well as unlimiting to deer populations. Today, these steep, shrub-grass slopes In the context of our findings, Although provide a “window of opportunity” for reproductive however, reproductive and summer mule deer to survive the winter within and summer maintenance habitat are important a zone of increasing snowfall from the maintenance functional entities that are spatially and western footslopes to the Bridger Divide. habitat together temporally limited. Availability and use of Generally, on this wet, windward slope, constituted these habitats had direct consequences to snow accumulations below that zone summer range for population characteristics and dynamics. are sufficient to inhibit movement of deer on our study In variable environments such as the mule deer to lower valley areas, while areas, they are timbered breaks and prairie-badlands, snowfall above the zone precludes deer not conceptually some summer maintenance habitat may use at higher elevations during all but the equivalent. become reproductive habitat in wet mildest winters. years and some reproductive habitat In contrast to the Bridger Range, may become only maintenance habitat the Shields Valley attending the west flank in dry years. In mountain-foothill and of the Crazy Mountains rises gradually northwest montane forest environments from the Shields River eastward to the availability of summer reproductive approximately 1,825 m at the point of and maintenance habitats is more stable inflection to relatively moderate mountain from year to year. slopes above. The lack of steep, open west and south facing slopes combined Winter with increasing snowfall through the 1,525-1,825 m elevational zone precludes The broad distribution of winter opportunity for mule deer to winter maintenance habitat for both species along most of the west flank of the Crazy of deer in mountain environments was Mountains. Instead, most deer with associated with areas receiving minimal summer home ranges in montane forest snow accumulation. These conditions on the west flank migrate around or

28 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Figure 8. climate across two Diorama showing distribution of important mule deer habitat components as influenced by topography and local mountain ranges in southwestern Montana.

The Co n c ep t o f Ha b i t a t 29 across the mountains to winter habitat specialized use of habitat, particularly on the north, south, and east flanks in deep snow environments. Open (Simmons unpubl.). There, similar to environments with limited snowfall or the east slope of the Bridger Mountains, the presence of agricultural croplands reduced snowfall and other environmental fostered selective foraging and more conditions allow mule deer to distribute general use of habitat. themselves rather widely over a broad These relationships were illustrated area of rolling, open foothills. by differences in overwinter survival Within this general framework, strategies of both species among study specific physical and biotic features areas and habitats. In timbered winter of winter maintenance habitat varied maintenance habitats in the Swan Valley widely among, as well as across, habitats. and Salish Mountains of northwestern Vegetation typically included various Montana as well as the Long Pines in mixtures of trees, shrubs, forbs, and the southeast, white-tailed deer behavior grasses, with no particular plant species emphasized energy conservation. Deer or combination of species or forage localized on small home ranges in the ...most native classes appearing to be superior across most favorable thermal environments. forages available all, or even adjacent ranges. Individuals, On the lower Yellowstone River, in winter are too groups, and populations of deer of both white-tailed deer ranged widely from low in nutritional species wintered across a spectrum of riparian cover along the river to forage vegetation varying from essentially open selectively for crop residues in fields value to meet grassland/agricultural cropland to closed throughout the floodplain and adjacent the maintenance canopy coniferous forest and foraged on a benchlands. They also fed heavily on needs of deer. Deer wide variety of plant species. litterfall that was abundant in stands of survive primarily Drought and prior livestock grazing riparian cover. In prairie-agricultural by supplementing apparently did not reduce deer use of ecosystems north of the Yellowstone energy reserves winter range as it did on summer habitat. River, whitetails employed a combination accumulated prior Winter drought with reduced snowfall of strategies. Typically, they ranged to winter with typically increased the amount, if not the widely and foraged selectively in dryland energy intake from quality, of available winter maintenance grain fields. However, under severe submaintenance habitat. Most deer winter ranges were conditions they localized on topographic winter diets. grazed by livestock, often heavily. Also, sites that provided optimal protection deer of both species occasionally spent from wind chill (Dusek et al. 1988). part or all winter in or around human- Winter maintenance habitats along disturbed areas and utilized a variety of the west flank of the Bridger Mountains food and cover resources to survive. could be occupied successfully only Studies have shown that most when mule deer employed a specialized native forages available in winter are strategy. This involved utilizing specific too low in nutritional value to meet the microenvironments where snow maintenance needs of deer (Wallmo accumulation was reduced and elevation, et al. 1977). Deer survive primarily exposure, and timber cover provided by supplementing energy reserves favorable thermal conditions (Youmans accumulated prior to winter with energy 1979, Pac et al. 1991). It also involved intake from submaintenance winter diets. restricted movement and generalized This involves behavior that emphasizes foraging where energy expenditure was energy conservation. However, deer also minimal. selectively foraged under conditions Winter maintenance habitats that favor such a strategy, especially in along the east slope of the Bridgers agricultural areas with abundant, high are very different. With progressively quality forage. On our study areas, less snowfall at lower elevations, stands of coniferous timber and broken mule deer easily moved below the topography were important features level of restrictive snow depth to enhancing energy conservation through exploit extensive expanses of rolling

30 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a shrub-grassland (Pac et al. 1991). formations, alpine scree, and some low East slope winter ranges contain few elevation basins dominated by monotypic topographic sites offering protection grasslands or dryland grain. from cold temperatures and the strong Large, uniform areas of grasslands chilling winds that characterize eastern lacking inclusions of agricultural cropland foothills. Tree and tall-shrub cover and/or riparian tree/shrub vegetation was sparse or lacking. Opportunity for were not used by whitetails. Other energy conservation in this setting is suboptimal environments for this species limited. Instead, mule deer move widely were dry shrub-grasslands, breaks, and forage selectively within large badlands, and montane forest lacking a home ranges across expansive open significant riparian wetland or other mesic ridges. They focus on specific habitats component. and microenvironments only when The quantity, quality, and conditions are unusually severe. juxtaposition of reproductive habitat, maintenance habitat, and unused areas varied (Fig. 7). This in turn affected The quantity, Unused Areas spatial distribution of deer and the pattern of movement required to use all essential quality, and habitat elements. juxtaposition Environments lacking resources Areas inhabited by deer were of reproductive to sustain deer are largely unused somewhat more heterogeneous in time habitat, although deer traverse these areas during as well as space than unused areas. maintenance movement to other habitats. These areas Fluctuating environmental conditions habitat, and typically lack some component, such as influenced plant phenology and the kinds, unused areas rugged topography or diverse vegetation, amounts, and quality of forage and cover varied. This in and often consist of large blocks of available. This influenced fluctuation in turn affected uniform characteristics. the distribution, relative abundance, and spatial distribution Unused environments for mule deer quality of the two functional types of of deer and include large expanses of open rolling habitat. the pattern of grassland, shrub-grassland, or croplands. As noted earlier, the ratio of movement required However, these cover types may comprise reproductive to maintenance habitat can to use all essential occupied habitat where found in smaller fluctuate markedly in highly variable habitat elements. units, especially when interspersed with northern plains environments. Livestock other vegetation or diverse topography. grazing, other land uses, and predation Unused areas are relatively scarce in were other environmental variables that mountain-foothill environments occupied influenced the amount and effectiveness by mule deer (Fig. 7A). These areas of reproductive and maintenance habitat occur primarily as massive, steep rock available to deer.

The Co n c ep t o f Ha b i t a t 31 Habitat Selection

example, each species is known by Species Adaptations substantial differences in gaits. Mule deer are characterized by a unique four-footed All members of the deer family bound or stott (Fig. 9), and white-tailed are believed to have evolved as species deer by a long graceful bound with tail broadly adapted to woodland or forest flagging in movement (Fig. 10). The edge (Putman 1988). Thus, mule and structure and musculature associated white-tailed deer share many attributes with the stott by mule deer best fit life that link them to those habitats. Both and survival in steep, rocky, and relatively species are small bodied, short legged, open terrain. Conversely, the bounding and similar in head and jaw structure. gait of the whitetail can be recognized They have small rumens and a low ratio as an adaptation to rolling, brushy, or of rumen-reticular volume to body weight woodland habitats. that renders them concentrate selectors Other morphological attributes in feeding (Hoffman 1985). of white-tailed deer, including the fine- At the same time, however, the two lined body form and ears, brownish coat species also have evolved individual color, and large, showy tail might also be attributes that constrain each to its associated with life in dense deciduous own ecological niche and contribute woodland. The heavier build, large to differences in habitat selection. For mobile ears, excellent distance vision,

Figure 9. Figure 10. A mule deer in a typical stott. A white-tailed deer in a typical bound.

32 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a gray coat color with light rump patch, and temperatures, the two species respond small, inconspicuous tail of mule deer fit differently to decreasing temperatures. well with life in dry, open, rugged terrain. Studies by Mautz et al. (1985) showed Among males, the antler form in white- that mule deer were more tolerant of tailed deer may also be associated with cold, increasing their metabolic rate only life in dense cover; while the large, open, when temperatures dropped below -18° dichotomously-branched antlers of mule to -23° C. Whitetails responded sooner, deer may be an adaptation to living in at temperatures between -12° and -18° more open terrain. C. The difference is assumed related to Subtle differences in digestive mule deer being adapted to more open systems and physiology of the two species environments with greater extremes in also can be associated with habitat temperature and sharper wind chill than selection. Possessing small rumens and in the wooded cover typically selected gut length relative to body size, deer must by white-tailed deer. The overhead eat small volumes of easily digestible cover requirement of whitetails in winter ...the two species food compared with larger ruminants also may be related to snow and forage also have evolved or “bulk feeders” that can eat larger conditions locally. individual volumes of forage of lower nutritional In summer, different mechanisms attributes quality and more difficult to digest associated with cooling and water that constrain (Hoffman 1985). Within this general conservation may also encourage habitat each to its own framework, mule deer, which evolved in segregation. Mule deer that occupy open, ecological niche drier, more variable environments, seem dry environments lack sweat glands and and contribute slightly better adapted to handling larger rely on panting and possibly dense blood to differences in amounts of coarse forage. White-tailed vessels in their large ears to dissipate habitat selection. deer are restricted more to succulent, heat while conserving water. Similar higher quality foods. The adaptation studies have not been made on white- of whitetails to receive much of their tailed deer, but Parker et al. (1985) show nutrition from plant cell contents that elk, which also select moist habitats and other highly nutritious and easily with overhead cover in summer, rely on digestible plant parts (Hoffman 1985, sweating as a primary means of cooling. Klein 1985) is particularly evident in Although white-tailed deer don’t sweat the species’ close association with as elk do, their association with moist agriculture. habitat may be an indication of less heat Small differences in mouth, jaw, tolerance. and tooth features between species can Behavioral differences associated also be associated with differences in with habitat selection are exemplified feeding and habitat selection. Thus, the in the manner in which the two species smaller, more finely structured incisors in respond to disturbance and their predator whitetails relates to selective feeding and avoidance strategies (Lingle 1989, high use of succulent plant parts. Habitat Geist 1994, Wood et al. 1994). When and dietary differences between species threatened, whitetails typically attempt to also may partially explain differences in flee, using speed to put distance between patterns and timing of tooth replacement themselves and the disturbance. They (Dusek 1994). follow established trails to ensure swift Other physiological differences and sure passage through dense brush. influencing habitat selection in deer relate They may also try to hide or take to water to thermoregulation and adaptations to throw predators off their trail. Such associated with staying warm in winter behavior would be expected of animals and cool in summer. These differences that evolved in rolling, mesic woodland appear to stem from long-term environments. adaptation of each species to climates Mule deer, in contrast, often stand within their primary distributional range. their ground, assess the threat, and Although both mule and white-tailed may approach, confront, or attack the deer are exposed to similar cold winter predator when a threat is perceived.

Ha b i t a t Se l e c t i o n 33 When threatened, they rely on their four- and forages in an area are considered footed bounding gait to maneuver in any available and deer choose those which direction, uphill or down, through rough maximize foraging efficiency and terrain, over and around obstacles to reproductive success. Within this process, avoid and confuse the predator. The stott, other studies have suggested that females in which the legs are held in, close to the select habitat suitable for rearing young, body, also allows the mule deer to kick a while males select habitat primarily on predator while fleeing. The bounding gait the basis of foraging opportunities (Geist of the whitetail precludes kicking. 1981, Bowyer 1984, Clutton-Brock No studies have provided evidence et al. 1987, Jakimchuk et al. 1987). that factors such as social dominance This concept is based on differences in of one species over the other are energetic requirements and reproductive associated with habitat segregation. In strategies that exist between sexes in fact, studies have demonstrated that polygynous ungulates (Main and Coblentz interaction between species typically 1990). follows the same individual dominance Our findings agree that forage hierarchy commonly observed within and other resource requirements are Most deer, species (Anthony and Smith 1977). important, but habitat selection also must especially females, Direct interspecific interactions are be interpreted within the constraints were limited to infrequent and usually nonaggressive. of other requirements and traditional Because of this, we speculate that the behavior of the two species. Habitat selection and use species occupying a habitat acceptable selection does not involve random of habitat and to both may depend on which species encounter between individual and resources within first became established. This is followed environment. Most deer, especially or close to their by avoidance of that habitat by the females, were limited to selection and use mother’s home other species as long as the first is of habitat and resources within or close range. able to maintain itself under prevailing to their mother’s home range. Habitats environmental conditions. outside of this area were either occupied by other adult female/family groups or, if vacant, were unknown and potentially Process of Habitat available only to a limited number of dispersers. Thus, as concluded by Selection Schoen and Kirchoff (1985), “the composition of the

home range was Habitat selection in deer and other ungulates has been regarded as an optimization process. All habitat types

34 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a determined by the array of choices This behavior, by which the mature available to the individual, and habitat female maintains exclusive use of that selection was a function of the available portion of her home range most important choices.” to reproductive success, serves as the stimulus for her year-old offspring to begin the process of habitat selection and Behavior and Habitat establishment of their own home ranges. It also serves to allocate important Use reproductive habitat in an area among several generations of female descendants of the matriarch. Biologists have long recognized Among yearling females animal behavior as an integral part of life marked in the Missouri River and the process by which species adapt Breaks, 84 percent remained to living with one another and to diversity in the area of their natal home and variability in the environment range where they established (Wilson 1975). Geist (1981) observed individual home ranges that, at that habitat selection, food habits, least occasionally, overlapped reproduction, and population dynamics those of their mothers are all accomplished by deer through (Hamlin and Mackie 1989). behavior or closely linked to behavioral Typifying this relationship, adaptation. Our extensive marking and home range boundaries, radio-tracking studies confirmed that observation sites, and habitat selection and other deer-habitat fawning territories for interactions are rooted in social behavior one mule deer matriarch and family relationships among individual in relation to home deer. range boundaries and observations of one of Influence of Social Structure her adult daughters over a 4-year period are Both mule and white-tailed deer shown in Fig. 11. As exhibit social structures that are indicated, total home functionally organized around family ranges of the daughter groups consisting of two or more (age 1-5 years) and generations of related females and their mother (age 8-12 offspring. The dominant member or years) completely matriarch is a mature female with a overlapped. However, history of successful reproduction. At at no time during the fawning, she occupies a choice area of fawn-rearing period was reproductive habitat that Ozoga et al. the daughter ever observed (1982) called a parturition (fawning) within the fawning territory territory within her summer range. of the matriarch. This area comprises an optimal mix of terrain and vegetation that provides � isolation along with dependable sources Figure 11. 010.5 km of succulent, nutritious forage, hiding Home range cover, and opportunity to escape or evade boundaries and observation sites Mother predators and to avoid competitors. for a mule deer Daughter During this critical period of the year, the matriarch and her maternal female isolates herself from all adult daughter Relocation sites for mother and/or other deer and directs all of her energy including a June- daughter to successfully rearing young. Isolation August fawning Areas used by matriarch during is accomplished through habitat selection territory (after June - August but never used by and chasing her year-old offspring and all Hamlin and Mackie her daughter during that time after 1989). she was one-year-of age. other deer from the fawning territory.

Ha b i t a t Se l e c t i o n 35 Fig. 12 conceptually illustrates of reproductive habitat. Consequently, from observations across all studies resources within the home range of the how family groups formed clusters of reproductively successful matriarch, overlapping home ranges in local areas while somewhat diminished for her, were more likely to benefit her descendants than unrelated deer (Clutton-Brock et Figure 12. al. 1982, Dusek et al. 1989, Hamlin and Conceptualized Mackie 1989, Pac et al. 1991, Porter home ranges and 1991). When fawning territories near the fawning territories of matriarch became filled and unavailable a matriarch and her to other deer, yearling or young adult two adult daughters and a granddaughter daughters and granddaughters were either during the fawn- displaced to nearby summer maintenance rearing period. habitats or they dispersed permanently to establish home ranges and new matriarchal groups elsewhere. Although matriarchs and adult daughters are often located close to one another, their association in the same social group is influenced by reproductive success. When both matriarch and � daughter successfully rear fawns through 00.5 1 km weaning in late summer, a socially cohesive family group may not reform until late autumn or early winter if at all. Matriarch, age 12 years Association occurred earlier in summer Daughter, age 8 years when related females were barren or lost Daughter, age 6 years their fawns. Fidelity to home range and the Granddaughter, age 3 years long-lasting social bond between mother Exclusively-used fawning territory and most female offspring played a significant role, not only in forming, but also in perpetuating successful habitat use patterns. For example, Fig. 13 illustrates Figure 13. Fidelity of four the fidelity of four radio-collared adult female mule deer to female mule deer to individual branches individual branches of a drainage system on the Cherry Creek of a drainage system study area. Fig. 14 shows seasonal in a prairie-badland movement and home range fidelity of a environment. Solid mule deer matriarch, her daughter, dots represent and two granddaughters locations of 2 deer that resided in the monitored over a 13-year lower fork and open period in the Bridger circles the locations Mountains. of two deer that All four members resided in the upper of the family fork of the drainage group in Fig. 14 over a 27-month period (Wood el al. 1989). � 012 km

36 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a used overlapping winter home ranges spring. Instead, yearling males were connected by a commonly used movement social drifters that usually associated corridor to overlapping summer home with nonproductive adult females or ranges east of the precipitous Bridger mature bucks (Hamlin and Mackie 1989, Divide. Timing of movement across the Divide in relation to snow depth was critical to maintenance of this movement pattern and had to be learned or passed on from matriarch to offspring. The social system centered on the matriarch, her aggressive behavior, and use of a fawning territory during late spring and summer also influenced spatial distribution and habitat selection by yearling and mature males. Prior to fawning, yearling bucks left or were driven from the mother’s fawning territories. Yearling mule deer males appeared inherently more prone to leave their mother’s home ranges than yearling females and often dispersed before being chased by their mothers (Hamlin B and Mackie 1989). Like mature bucks, r e g d i r yearlings were relegated to maintenance habitats not occupied by productive females. At 11-14 months of age, 40 (70 percent) of 57 yearling male mule deer D marked in the Missouri River Breaks left d i v i their natal home ranges; 29 of those (51 percent of all yearling males) left the e study area (Hamlin and Mackie 1989). Among yearling male white-tailed deer marked on the lower Yellowstone River, 24 (46 percent) of 50 moved permanently from the vicinity of the mother’s home range (Dusek et al. 1989). Although a small percentage of yearling males � of both species remained on or near 012 km their mother’s home range, they were seldom included in their own mother’s Female # 1021, n = 38, 1973-75, age 2 1/2 - 4 1/2 social group. However, yearling males Female # 1062, n = 406, 1975-86, age 1/2 -12 1/2 commonly joined, and at least temporarily Female # 1166, n = 69, 1979-85, age 1/2 - 6 1/2 associated with groups of unrelated does and fawns during late summer, autumn, Female # 1284, n = 63, 1982-86, age 1/2 - 4 1/2 and winter. Movement corridor High mobility and variability in movements of yearling males of both Relocation of collared female along corridor species during June-November indicated Figure 14. Seasonal home ranges and interseasonal movement that most did not establish traditional of three generations of related female mule deer in a mountain- summer home ranges immediately foothill environment. Individual relocations during migration after breakup of family groups in indicate a shared movement corridor (after Pac et al. 1991).

Ha b i t a t Se l e c t i o n 37 Pac et al. 1991). As a result, most associated with lactation, security from yearling males tended to establish home predation for newborn fawns, and ranges and other habitat use patterns avoidance of competitors. by mimicking either mature males or By late summer and early autumn, tolerant, nonproductive females utilizing as fawns were weaned, resource maintenance habitats. requirements of maternal does changed. Upon reaching maturity, most Freedom from demands of lactation was bucks utilized large home ranges in followed by need to recover physical maintenance habitat where they range condition and develop fat reserves apart from productive females throughout while quality forage remained relatively the fawning season. During the breeding abundant. The shift in physiological season, mature males generally expand need coincided with a pronounced shift their movements and home range to in plant phenology. As native plants have access to as many adult females mature and produce fruit, nutritional as possible. As illustrated in Fig. 15, content and values also change, from the home range of one mature mule high protein in succulent, rapidly growing deer buck in the Missouri River Breaks plant materials to high carbohydrate in encompassed the home ranges of 11 fruits, seeds, and other plant parts. Thus, marked adult females and additional does selected habitats that allowed them unmarked females. to maximize energy gathering and intake. They often made specialized use of local Interaction of Behavior and habitats and resources that satisfied resource deficiencies experienced on Resource Requirements normal seasonal ranges. Movements In late spring and early summer, to agricultural fields with alfalfa, waste maternal does of both species on all grain and row crops, orchards, and other study areas restricted their movement areas represented subtle, yet critically and sought isolation from other deer important adjustments in habitat selection on fawning territories. This behavior that allowed individuals to continue was related to their needs for succulent, optimal use of habitat in spite of some high quality forage to sustain the seasonal deficiencies or changes in local high energetic and nutrient demands environments. Late autumn brought severe weather and resource scarcity for deer in Figure 15. mountain foothill environments. Adult Distribution of home does attempted to utilize summer-autumn range boundaries for a mature male and habitats as long as possible. This delayed eleven adult females their move to winter maintenance habitats (after Hamlin and which were often characterized by large Mackie 1989). concentrations of deer on small areas with low quality forage. In winter, two different patterns of behavior relative to habitat selection and resource use were observed. Deer occupying maintenance habitats dominated by natural vegetation reduced their movement and foraging activity, a strategy to conserve energy and fat � reserves. Where deer had access to 012 km abundant nutritious forage, especially agricultural crops, movements between Home range boundary - mature male bedding and feeding areas suggested a Home range boundaries - adult females strategy of selective foraging. Spatial Some locations of the male during the rut separation of females with fawns also

38 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a was relaxed as family groups reformed plentiful forage and selective feeding over and used the abundant resources. This relatively large home ranges compared to pattern was most prevalent among white- the small home ranges of adult females. tailed deer occupying plains riverbottom However, they often afforded only environments (Dusek et al. 1989) but limited security from predators. During could be followed by deer of either summer, bucks often associated in loosely species in any area where agriculture organized bachelor groups. Mobility of occurred in proximity to natural winter the groups varied, but often was limited, maintenance habitat (Kraft 1989). especially in early summer when habitat In spring, new growth of forbs selection and use centered on forage-rich and grasses strongly influenced habitat sites within the large home ranges. selection. Social barriers appeared even By early November, rutting more relaxed as deer congregated to dominated the behavior of mature males, forage intensively on local areas with and habitat use reflected their search for abundant early growth of green plant breeding females. In mule deer, dominant materials. These usually were located on males become particularly mobile and are or near winter maintenance habitats. As likely to accomplish most of the breeding. soon as snow melt and plant phenology Nonetheless, all bucks exhibit increased permitted, most adult females moved mobility as they actively seek to breed as off winter-early spring ranges to resume many receptive females as possible. In use of individual summer home ranges sharp contrast, dominant whitetail bucks located in timbered habitats offering both are usually associated with a definitive security and high quality forage. rutting territory established in dense The peak energetic The peak energetic demand for cover within the overall home range of demand for reproduction occurs during the rut for each individual. Each buck advertises reproduction occurs males and during late gestation and his presence by scent-marking and sign- during the rut for lactation for females. Therefore, habitat posting the landscape, and breeds only males and during selection unfolded differently between receptive females that occupy or enter the late gestation sexes. territory during the rut. If the dominant and lactation for During spring and summer, habitat buck is removed from the rutting females Therefore, selection of young adult males up to about territory, he usually is quickly replaced by habitat selection 4 years of age probably is motivated another buck. unfolded differently primarily by their need for resources to Following the rut, bucks of both between sexes. sustain body growth and secondarily for species reduce their mobility, shed antler development. By the fourth year, their antlers, and often move to winter as individuals achieve mature body size, maintenance habitats. Dominant the requirement shifts more to recovery breeding males tend to be in poor of body condition depleted during the condition following the energy-costly previous breeding and winter seasons rut. Because of this, and faced with a as well as to development of larger prolonged winter energy deficit, habitat antlers. Among mature bucks, antler selection and use reflect conservation of growth to achieve and maintain social remaining fat reserves. Some segregation dominance apparently is of equal or of the sexes usually occurs unless severe greater importance than body growth. winter conditions force concentration and However, in heavily hunted populations aggregation of all deer on limited habitat. with few mature males, younger males By spring, adult bucks selected may face the same increased requirement habitats that provided greatest to recover from the rut in addition to opportunity to recover physical condition. achieving mature body and antler size. Foraging activity focused on new plant In accord with their different growth that was available in large resource requirements, adult males quantities at low elevations. In contrast typically were segregated from productive to females with fawns, the lower security females during summer. The maintenance requirements of bucks allowed them habitats selected offered access to to make prolonged use of these open,

Ha b i t a t Se l e c t i o n 39 forage-rich areas. Because of this, bucks group to which they became attached. At often returned to summer home ranges the second or individual home range level, We found that somewhat later than adult females. each deer “fine-tuned” its own habitat as environment selection and use to the structure and varied patterns of resources of the local area it attempted distribution and Patterns in Habitat to exploit. At the third level, individuals movement, use of continually adjusted their habitat use vegetation/cover Selection patterns or behavior within individual types and forage, home ranges to fluctuations and activity regimes, ongoing changes in local environmental We found that as environment and sociality also conditions. Over time, effective varied, patterns of distribution and varied. exploitation of the diverse environments movement, use of vegetation/cover we studied involved numerous adaptations types and forage, activity regimes, and and adjustments in behavior. sociality also varied. Habitat selection in both species unfolded as a three-step process associated with establishment Distribution, Movements, and of individual home ranges. The first Home Range occurred at the landscape level and Spatial distribution, movement, and involved the broad behavioral strategy home range patterns of individual deer and habitat use necessary to settle into of both species centered in three general and successfully exploit a particular strategies that formed a continuum of environmental complex. It usually increasing specialization: residency; involved individuals adopting seasonal adjacent seasonal home ranges; and habitat use patterns that mimicked proven distinct seasonal home ranges. strategies of their mothers or a social

40 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a The first and most fundamental of locations typically occurred in one strategy that a deer can use in exploiting area during a season, deer following this its local environment is yearlong pattern tended to move back and forth residency or use of indistinct seasonal between the two seasonal home ranges. ranges. Deer used single yearlong home Deer used distinct seasonal home ranges where the basic requirements ranges when a relatively high degree of of maintenance and reproduction specialization was required to exploit a could be satisfied within one local area particular environment (Fig. 18). Winter and microsites important for seasonal and summer home ranges were distinct requirements were dispersed throughout entities separated by distances of a few the area. As a result, a deer could to as much as 130 km. As the distance generally occur in any portion of the between seasonal home ranges increased, resident home range at any time of the movements more closely resembled year (Fig. 16). However, certain portions true migration as described by Baker often received more use during some (1978). Intra-seasonal trips between the seasons. seasonal home ranges were the exception When resource requirements could and almost never occurred among mule not be met in one local area, specialized deer with winter and summer home movements and use of home range began ranges centered more than 5 km apart in to develop. Deer with adjacent seasonal mountain-foothill environments (Pac et home ranges used identifiable winter and al. 1991). summer home ranges that were usually The general type of movement separated by only a few kilometers (Fig. patterns employed by deer depended 17). Adjacent seasonal home ranges on the spatial arrangement of important represent a blend of both residency and habitat components that they attempted migration. Although a greater proportion to exploit. Among our study areas,

Figure 16. Yearlong distribution of the observations of a typical resident deer (after Pac et al. 1991).

� 0 0.5 1 km

Radio locations in summer Radio locations in winter

Ha b i t a t Se l e c t i o n 41 Summer home range

� Winter 00.5 1 km home range Radio locations in summer Radio locations in winter

Figure 17. Yearlong distribution of the observations of a typical deer with adjacent seasonal home ranges (after Pac et al. 1991).

Winter home range

Summer home range � 00.5 1 km

Radio locations in summer Radio locations in winter

Figure 18. Yearlong distribution of the observations of a typical deer with distinct seasonal home ranges (after Pac et al. 1991).

42 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a classification of entire populations of deer to higher elevations. Thus, most deer according to one of the three movement exhibited seasonal movement between patterns was rarely appropriate. Rather, adjacent or distinct seasonal home ranges. each environmental setting occupied Deer utilizing these two patterns made by deer usually produced a mosaic of up 70 percent of the mule deer inhabiting individual strategies that included all the west slope (Stansberry 1996) and 90 three patterns. percent of all whitetails along the east The general use The residency pattern was slope of the Salish Mountains (Morgan of movement common in environments where winter 1993). These patterns also were patterns employed maintenance and reproductive habitats employed by 85-90 percent of all mule by deer depended overlapped or were closely interspersed. deer in mountainous environments of on the spatial For example, the resident pattern southwest and northcentral Montana (Pac arrangement of described approximately 90 percent of et al. 1991, Kasworm 1981, Ihsle 1982). important habitat mule deer occupying reclaimed habitats Even in the more gentle topographic components that near the Colstrip mine in southeast settings of eastern Montana, use of they attempted to spatially separated seasonal home ranges Montana (Fritzen 1995) and 100 percent exploit. of mule deer on the Dog Creek study was common among deer in upland area in sagebrush grassland habitat of areas where reproductive and winter northeast Montana (Jackson 1990). maintenance habitats were patchy and About 80 percent of the white-tailed deer separated. These patterns typically inhabiting irrigated floodplains of eastern involved movement of deer from Montana (Dusek et al. 1989) displayed reproductive and summer maintenance this pattern as did 30-40 percent of mule habitats in areas of low topographic relief deer inhabiting prairie-badlands (Wood to winter maintenance areas in steeper et al. 1989) and 60 percent of those in terrain and lower snow accumulation. the Missouri River Breaks (Hamlin and About 40 percent of the deer studied Mackie 1989). in the Missouri River Breaks displayed The residency pattern accounted movements similar to either the adjacent for only 10-15 percent of mule deer seasonal home range or distinct seasonal inhabiting mountain-foothills of southwest home range patterns (Hamlin and Mackie Montana (Pac et al. 1991) and east 1989). On the Cherry Creek study front ranges in northcentral Montana area, about 45 percent of the mule deer (Kasworm 1981, Ihsle 1982). Similar followed these patterns; 25 percent were proportions of resident white-tailed described as “autumn migrants” (Wood et deer were documented in the northwest al. 1989). Each autumn migrant utilized coniferous forests (Morgan 1993). one home range most of the year, but However, the proportion of resident deer occupied a distinct autumn range from increased to approximately 30% in mule mid-August to mid-October. deer populations occupying a northwest On the upland prairie-agricultural montane forest environment (Stansberry habitats associated with the Cherry Creek 1996). study area, Wood et al. (1989) concluded Use of adjacent and distinct that female white-tailed deer exhibited seasonal home ranges was most prevalent individual movement patterns that were among deer of both species inhabiting difficult to categorize. Some used small, strongly seasonal mountain environments stable home ranges while others made in central and western Montana. In these erratic shifts among several seasonal settings, winter maintenance habitats activity centers within large home ranges. often were distributed in a narrow belt Our studies confirmed that more or in specific patches at lower elevations. specialized movement patterns among Winter ranges typically provided yearlong adult females probably developed from habitat for small numbers of deer. Most a requirement for an exclusive fawning reproductive and summer maintenance territory. Among females that shared habitat was located in relatively large a local area of winter maintenance expanses of montane forest at middle habitat, not all found suitable fawning

Ha b i t a t Se l e c t i o n 43 territories in that vicinity. Aggression by Knowledge of accessory areas established resident matriarchs forced was learned through exploration or by some female offspring to seek out and associating with other deer using these establish their own fawning territories at areas. Once individuals were monitored some distance from the established family for sufficient time to identify these subtle group. Females that developed these new specializations in habitat use, the factor or adjacent seasonal home range or distinct circumstance influencing the movement seasonal home range traditions would be was usually evident. Even though expected to pass on successful patterns to accessory areas were not occupied in the subsequent generations. traditional manner that seasonal home ranges were used, their use was relatively Fine-tuning the Home Range predictable based on environmental conditions. Use of seasonal accessory Individual deer “fine tuned” their areas was responsible for much of the home range to changes in the local variation observed among individual deer environment by incorporating subtle movement patterns. adjustments in their movements and Although any number of seasonal habitat use. Across the environments we accessory areas might occur, we identified studied, this process occasionally involved five different types used during winter, specialized use of accessory areas usually spring, early summer, late summer, and located outside of normal seasonal autumn in the Bridger Mountains (Pac home ranges. Deer used these areas to et al. 1991). Conceptual examples (Fig. satisfy temporary resource deficiencies 19) show how use of various types of experienced within the normal home accessory areas allowed deer to adjust range. their movements and home range to changes in the local environment.

Figure 19. Conceptualized home A. Resident range patterns and with winter accessory accessory areas area showing increasing specialization in the use of space by season. Individual telemetry locations B. Resident are labeled by with late season: W=winter, summer S=summer,. accessory F=fall, Sp=spring. area. Accessory areas are shaded.

C. Adjacent seasonal range pattern with early summer accessory area.

D. Distinct seasonal range pattern with spring and fall accessory areas.

44 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Resource requirements of resident females attempting to recover from the deer are generally satisfied within demands of lactation. Most deer located one home range (Fig. 19A). During alternative sources of succulent forage occasional severe winters, resources inside the boundaries of their home available within the normal home range of ranges. However, during very dry years, some deer became inadequate as a result some deer moved 1-14 km beyond normal of deep snow and cold winds. Under such home range boundaries to late summer conditions, these individuals temporarily accessory areas located near alfalfa fields, abandoned their normal home range and hay stacks, riparian communities, or to moved 1-20 km directly to site-specific areas of greater topographic complexity winter accessory areas that offered that offered moist sites and succulent, more tolerable snow depths and greater native forage. In the Missouri River topographic relief. These areas provided Breaks and the Bridger Mountains, more plentiful, steep south slopes with normal home ranges of resident and warm microsites that were often in close migrant mule deer using late summer proximity to conifer stands with dense accessory areas were generally located canopies. in areas of relatively low topographic Hamlin and Mackie (1989) noted complexity and were heavily grazed by that improved forage availability did not domestic livestock (Hamlin and Mackie appear to be the incentive for moving to 1989, Pac et al. 1991). Accessory winter accessory areas in the Missouri areas were invariably located in habitats River Breaks. There, these sites typically ungrazed by livestock at that time. Deer offered the poorest quantity and quality used late summer accessory areas more of forage among all vegetation types. Use frequently than winter accessory areas, of winter accessory areas favored energy but not every year. conservation, ease of mobility, and escape The autumn migrant pattern from predators. displayed by individual mule deer on During severe winters in the the Cherry Creek study area (Wood Missouri River Breaks and sagebrush et al. 1989) was similar to use of late grasslands in northeast Montana (Jackson summer accessory areas by resident deer. 1990), use of winter accessory areas However, autumn migrants moved outside outside of the normal home range was a of their normal home ranges each year common occurrence. Use of accessory to areas with deciduous shrubs. Habitats areas during severe winters in mountain used by these deer during most of the country was usually limited to resident summer contained less badlands than mule deer occupying home ranges with those used by resident deer. Autumn low topographic complexity, although migrants commonly experienced a few deer with adjacent and distinct shortages of succulent forage in early seasonal ranges used these areas (Pac et autumn within their home ranges. This al. 1991). Time spent on these areas was forced them to move to more suitable usually confined to the duration of severe habitats at lower elevations. conditions. Individual deer rarely used The search for high quality forage in winter accessory areas in consecutive late summer also influenced movements years unless severe conditions occurred in of whitetail does, even in riverbottom successive years. We suspect that white- environments. Herriges (1986) reported tailed deer employed similar strategies that the proportion of whitetails moving during severe winters. to agricultural fields increased in late Use of late summer accessory summer as fawns were able to travel areas by resident deer (Fig. 19B) usually with their mothers and native forages coincided with desiccation or killing became desiccated. These movements frosts which reduced the availability contrasted with the definition of a late of green, succulent forage during late summer accessory area because they summer. Succulent forage in late summer represented a strategy that occurred was particularly important for adult every year, usually within the confines of

Ha b i t a t Se l e c t i o n 45 the home range. However, the motivation high elevation habitats that were limited for patterns of habitat use in late summer in resources during early fawn-rearing. was essentially the same in both species. Use of such areas by deer in lower In rugged mountain habitat, adult elevation environments apparently is female mule deer often exploited high rare. Pac et al. (1991) reported that elevation summer home ranges in close one resident mule deer doe used an early proximity to their winter home ranges summer accessory area on the Brackett (Fig. 19C). Although the seasonal ranges Creek winter range. Fritzen (1995) may be only 1-5 km apart, the change in described a similar pattern for a resident elevation may exceed 435 m. Use of early mule deer doe on his study area near summer accessory areas was relatively Colstrip. common among mule deer in these Deer that migrate between distinct situations and appeared to be related to seasonal home ranges often incorporate resource requirements associated with the greatest degree of specialization into fawn-rearing (Pac et al. 1991). their annual movement patterns. These Beginning in late May and early individuals not only travel the longest June, as snow melt reached higher distances, but also exploit environments elevations, mule deer does following that require additional adaptations in adjacent seasonal home range movement movement and habitat use. Typically, use patterns moved directly to their summer of some type of spring and/or autumn home ranges. In some years, however, accessory area is involved (Fig. 19D). new plant growth was not sufficiently Pac et al. (1991) reported that use advanced at higher elevations to support of spring and autumn accessory areas the high demand for succulent forage was greatest among deer that crossed a associated with lactation. Under such major mountain divide or encountered conditions, does occupying these habitats substantial elevational relief along were forced at fawning to move down to movement corridors connecting their early summer accessory areas located seasonal home ranges. Both types of near their winter home ranges. While accessory areas were usually situated the more open sites provided adequate closer to the winter home range and at sources of succulent forage, they may elevations intermediate to winter and have offered limited security for newborn summer home ranges. Some individuals fawns. Choice sites offering both only used spring accessory areas, others security and succulent forage occurred only autumn, and some used both types. at intermediate elevations, but these sites Spring and autumn accessory areas were were usually occupied by other maternal used with greater regularity than other females that defended them against types of accessory area. intruding deer during the early fawn- Deer of both species occupying rearing period. mountain environments attempted to Pac et al. (1991) reported that leave winter concentration areas in spring early summer accessory areas were as snow melted and greenup began. In located an average of 2.3 km from most cases, only resident deer remained summer home ranges and almost 420 m on or near winter ranges. Some lower in elevation. Use of early summer individuals using adjacent seasonal ranges accessory areas along the west slope of with summer home ranges located at the Bridger Mountains generally occurred low-middle elevations moved early and during June 17-July 17. This period directly to summer range. Others, as well corresponded closely to the 30-day period as deer with distinct seasonal ranges in or around fawning when maternal does are across high elevation habitats, moved to aggressive toward other deer (Ozoga et spring accessory areas concomitant with al. 1982). early growth of herbaceous forage plants Early summer accessory areas at intermediate elevations. apparently represented a specialized Spring accessory areas tended to strategy that allowed females to exploit be site-specific for individuals or family

46 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a groups, though overlap often occurred open areas at a time when they were in areas of high deer density. Because gregarious. Bucks did not experience deer preferred to move to summer the same requirements for isolation and home ranges as soon as possible, spring security that caused adult does to move accessory areas were generally occupied as soon as possible to their summer for shorter periods than autumn areas. home ranges. Bucks were more “casual” Pac et al. (1991) reported that individual in their movements and tarried along mule deer used spring accessory areas their movement corridors as long as they from 15-40 days compared to 30-65 were able to find adequate succulent days for autumn areas. Spring snow forage. These differences in timing of storms and cold temperatures resulting in movements between the sexes provided persistent snowpack resulted in later or adult females with unrestricted choice of prolonged use of spring accessory areas summer habitats before males arrived to in some years. “fill in the holes” at a later date. Because The extent to which deer tolerated of this, increased population size resulted snow on their summer home ranges in relegation of adult males and younger ...differences varied widely among individuals adult females to summer maintenance in timing of depending on when the snow occurred, habitat. movements between the juxtaposition of their summer and During autumn, use of accessory the sexes provided winter ranges, and the topographic areas by mule deer bucks in mountain- adult females characteristics along their movement foothill environments was less site with unrestricted corridors. Deer with summer home specific because of greater mobility choice of summer associated with the breeding season. Rut- ranges on the opposite slope of a high habitats before mountain divide from their winter related movements often were tangential males arrived to ranges were the first to respond to to the normal, linear orientation of “fill in the holes” at autumn snowstorms and move to autumn their movement corridors. Some bucks a later date. accessory areas. Weather conditions on temporarily moved to higher elevations; the divide, rather than on the summer others crossed intervening ridges and home range, dictated when these deer moved into drainages not used during moved. Pac et al. (1991) reported that other seasons. During the rut, a few mule deer exploiting high elevation moved as much as 11 km from their summer ranges spent about 70 fewer normal ranges while others showed no days on summer home ranges than deer noticeable expansion of their home range. exploiting lower elevations in the same mountain range. Deer remained on Home Range Size autumn accessory areas as long as snow depths were tolerable. Deer that did Home range size was part of not use autumn accessory areas moved the habitat use strategy employed directly to winter ranges, usually later in by individual deer to exploit the autumn or early winter following snow environments they occupied. Extreme that precluded continued use of summer individual variability in home range ranges. Severe weather accelerated size was characteristic of home range autumn movements and resulted in earlier measurements in all environments. This concentration on winter ranges. indicated that mobility and home range Mule deer bucks exhibited a different size represent adaptations unique to pattern of use of spring and autumn individuals and the habitat they occupy. accessory areas than females (Pac et Home range parameters expressed as al. 1991). In spring, some yearling means for populations or habitats must be and mature males remained on winter viewed as generalizations and interpreted home ranges and spring accessory areas with caution. for up to one month longer than adult Home range sizes and relationships females. This appeared to be consistent in the various environments we studied with a male strategy of habitat use are described in detail in other reports that emphasized forage gathering in (Dusek and Mackie 1988, Dusek et al.

Ha b i t a t Se l e c t i o n 47 1989, Hamlin and Mackie 1989, Wood deer had home ranges averaging 2.1- et al. 1989, Pac et al. 1991). Generally, 2.9 km2 in summer and 2.3-3.4 km2 in however, males had larger home ranges winter, compared to an overall average than females. Among resident deer resident home range of 6.3 km2 (Wood et monitored in our studies, annual home al. 1989). White-tailed deer exhibited ranges of mule deer bucks in the Missouri individual movement patterns ranging River Breaks (Hamlin and Mackie 1989) from small, stable home ranges to erratic Extreme individual averaged 27 km2 compared to only 5.2 shifts within very large home ranges in variability in km2 for white-tailed bucks along the lower prairie-agricultural habitat. Thus, home 2 home range size Yellowstone River (Dusek and Mackie ranges averaged 3.3 km in summer and 2 was characteristic 1988). Annual home ranges of resident 6.3 km during winter within an overall 2 of home range females in those environments averaged average home range of 33.5 km . 2 2 measurements in 5.2 km and 1.1 km , respectively. Direct comparisons of average home all environments. Migratory deer exhibited smaller range sizes for deer in environments seasonal home ranges than resident deer we studied and other areas were which could use all portions of their hampered by differences in methods and annual home ranges during all seasons. sample sizes among studies. Despite Average summer home ranges of adult this limitation, our estimates for the female mule deer with distinct seasonal two species, both sexes, and various ranges in seven Bridger Mountain environments fell within the range of populations varied from 0.9 to 3.2 km2. home range sizes reported elsewhere Winter home ranges averaged 0.8- (Wood 1987). Largest and most 5.0 km2 (Pac et al. 1991). Migratory variable home range sizes occurred whitetail females inhabiting coniferous among mule deer and white-tailed deer in forest had home ranges averaging only prairie-badlands and prairie-agricultural 0.6-0.7 km2 in summer ( Leach 1982, environments. Smallest average sizes and Morgan 1993) and less than 0.3 km2 in least variability was evident in relatively winter (Mundinger 1982). In the prairie- diverse, stable riverbottom and montane badlands environment, forest environments. migratory female mule

48 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Selection and Use of Vegetation throughout the parturition-lactation period (Pac et al. 1991). Perhaps equally In conjunction with other habitat important, these stands also provided components, especially topography, isolation from other deer, security from vegetation helped structure the landscape predation, and avoidance of competition and provide the physical and biotic from other wild ungulates (e.g. elk) and environment deer required to inhabit livestock. Topographic diversity, which an area. Habitat diversity, influenced included variation in slope, exposure, by topographical and/or vegetational and elevation, also was important in diversity, appeared to be a good indicator providing diverse microenvironments and of intensity of deer use. In mule deer cover types that supplied a varied and habitats, vegetational diversity usually long-lasting source of quality forage. As followed topographic diversity, thus Klein (1985) also noted, maximization of topographic diversity may be the major or selective foraging is most likely to occur ultimate factor influencing mule deer use in mountainous regions where variability of an area (Hamlin and Mackie 1989). in exposure, slope, and altitude create a In white-tailed deer habitat, diversity of microclimatic influences. vegetational structure and other physical Structurally complex and diverse site factors promoted diversity. Overhead vegetation not only met all requirements cover and mesic areas were available for reproduction, but also met them better to some extent in all areas occupied by on more restricted areas. Thus, home white-tailed deer. Compton et al. (1988) Habitat diversity ranges were smaller, and greater numbers reported a direct relationship between the appeared to be a of individual parturition territories were amount of riparian cover and abundance good indicator of supported in any given area or unit of of white-tailed deer. Whitetails were more intensity of deer habitat. Conversely, extensive areas of commonly associated with agricultural use. gentle to moderate slopes dominated by lands than were mule deer. open shrub-grass vegetation or dense, Across the state, results indicated even-age conifer forests were avoided or that selection and use of vegetation was used as maintenance habitat by males and driven by habitat structure. At the local nonproductive females. level, specific vegetation types or plant In the Missouri River Breaks, species were important in use patterns. reproductive habitats were more However, specific plant types and species structurally diverse than maintenance were only part of the environmental or unused habitat (Hamlin and Mackie complex involved in providing for the 1989). Habitats used more heavily forage, spatial isolation, cover, and other by deer exhibited greater numbers of needs of deer. patches and diversity of cover types Where available, mule deer than areas that were unused or used selectively used structurally diverse forest only occasionally. Mule deer used areas vegetation, especially for reproductive containing forested cover types more habitat. Nonetheless, this apparent during all seasons than expected based preference was invariably tempered by on availability. Selection for diversity was selection for other site factors, especially also evident in their preference for areas topographic diversity. In non-timbered with four or more locally interspersed breaks, badlands, and other prairie cover types and at least moderate environments, topographic characteristics topographic relief. Patches of Douglas and variation appeared to become more fir-juniper and scattered-moderate density important, if not the driving force in ponderosa pine-juniper-grass cover were habitat selection. most strongly selected. Conclusions In the Bridger Mountains, about the overriding importance of multilayered, low- to medium-elevation any specific type(s), however, must be conifer stands in conjunction with tempered by the fact that Douglas fir topographic diversity provided a wide types were distributed only over the array of succulent, high quality forage western portion of the area; deer on the

Ha b i t a t Se l e c t i o n 49 eastern portion did not have the option to the physical conditions deer needed to select fir, and had to choose from pine- survive. juniper and other less preferred types. Habitat selection and use In timbered breaks, as in the of vegetation in prairie-badlands mountains, structurally diverse timbered environments also was strongly related types plus topographic variation provided to topographic characteristics. On the the highest quantity and quality forage, Cherry Creek study area, extensive open the greatest opportunity for extending rolling mixed-grass prairie was avoided selective foraging through the lactation (Wood et al. 1989). Badlands and period, spatial isolation for productive mesic hardwood draws were selected females, hiding cover for fawns, and and used heavily throughout the year. the least opportunity for interspecific Interspersion of hardwood draws and competition with elk and livestock. badlands provided the resources required Expansive ridgetop and coulee-bottom by mule deer throughout the year. areas dominated by open low shrub- Hardwood draws were the primary source grassland vegetation were avoided, of succulent forage and dense hiding especially by lactating females. cover during fawn rearing and comprised Selection of winter maintenance both reproductive and yearlong habitat habitat in the Missouri River Breaks was for adult females. Because of the lineal even more strongly related to topography nature of the draws, most females isolated and structure of vegetation. Douglas themselves on fawning territories spaced fir types received more use than during along the length of draws in suitably other seasons. Areas of low relief, even structured habitat. When badlands and when timbered, were vacated as snow hardwood draws were lacking, mule deer accumulated and weather conditions were dispersed widely and exhibited became severe. Under such conditions, greater variability in use of vegetation. moderately steep terrain that included On sagebrush grassland prairie both open or semi-open south-facing in northeast Montana where hardwood slopes and moderate to dense timber on draw vegetation was lacking, mule deer northerly exposures combined to provide occurred widely spaced as individuals

50 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a or small groups along narrow, sharply and reclamation as well as to forage and cut drainageways that dissected the area cover resources available within the city (Jackson 1990). These cuts provided (Fritzen 1995). Because of restricted some succulent vegetation as well as hunting, the deer were highly acclimated daytime cover. During summer, deer to the presence and activities of humans moved to adjacent uplands or creek and utilized all available habitat and bottoms where they foraged selectively vegetation resources present. In autumn through the night before returning to the and winter, deer made heavy use of fruits cuts at dawn. Use of all vegetation cover and other forages available from gardens types on the area, except for very limited and landscaping in the city. Some deer creek bottoms, generally corresponded to adapted a strongly nocturnal activity availability. This included the extensive regime while others acclimated to the big sagebrush-grassland type, which presence of people and were active covered 64 percent of the area and throughout the day. In adapting to the seasonally accounted for 38-59 percent of presence and activities of humans, deer the total use of vegetation cover types. benefitted not only from the resources Patterns of habitat/vegetation use available but also the reduced threat of documented that mule deer can adapt predation. successfully to most environments White-tailed deer occur almost in Montana. Occurrence of alfalfa exclusively in association with riparian and yellow sweetclover that provided or mesic upland vegetation that provide succulent, high quality forage during overhead cover or with agriculture. They summer and autumn or small grain also are associated with gently rolling and other croplands that provided high topography with slopes less than 30 energy forage in autumn and winter degrees. Dry, topographically diverse stimulated mule deer use of areas that upland environments that lacked riparian otherwise might be unused or used only vegetation or agriculture were selected occasionally. almost exclusively by mule deer. Mule deer near Colstrip responded In summer, vegetation selection by positively to vegetation and other habitat white-tailed deer, especially reproductive changes associated with strip-mining females, appears to be driven by needs

Ha b i t a t Se l e c t i o n 51 for high quality forage, isolation, and Riparian habitats were particularly security in association with overhead important in spring and early summer. cover. Environments that structurally However, as vegetation in these habitats provide for all three are selected matured, deer moved upward from wet irrespective of the specific vegetation bottomlands to mesic timbered types that cover types represented. Similar provided quality forage and cover during environments lacking in one or more late summer and autumn. serve as maintenance habitat for males The marked preferences of white- and/or nonproductive females. On the tailed deer for diverse forest cover lower Yellowstone River, productive associated with riparian habitats also was adult females used mature cottonwood apparent in winter. Winter maintenance stands more heavily than bucks and habitats selected by whitetails were unproductive females during spring characterized by interspersion of and summer. Although whitetails are timbered riparian areas and diverse, generally considered most adapted to subclimax coniferous forest habitat types. early successional vegetation, mature These habitat complexes, located in cottonwood exhibited the highest total foothill and lower valley areas, provide nutrient load among all seral communities overhead cover to maintain the snow on the area (Boggs 1984). and thermal conditions whitetails require At the landscape level, habitat while also providing opportunities for selection by whitetails in northwestern deer to forage. Open, logged, and Montana was constrained by elevation agricultural habitats that accumulated as well as by vegetation, roads, riparian deep snow were essentially unused except areas, slope, and aspect (Morgan 1993). under mild conditions or in proximity to Preferred habitats occurred below 1,525 human developments. m; habitats above about 1,650 m were In semiarid eastern Montana, white- avoided. The preferred elevational range tailed deer occur primarily in association spanned all major creek bottomlands with riparian vegetation along rivers, and adjacent drainages and slopes and streams, and other mesic drainageways. included the major riparian, meadow, and Locally, they also occur in association lower elevation forest habitats used by with agricultural and ponderosa pine white-tailed deer. When the influence of habitats (Allen 1971, Swenson et al. elevation was removed, vegetation was an 1983, Dusek 1987). important factor in habitat selection. A yearlong preference for mesic Perhaps the most distinctive feature ponderosa pine habitat types in the of vegetation use by whitetails across Long Pines in southeastern Montana all study areas was the preference was similar to use of conifer-dominated exhibited for diversity, both structural habitats in northwestern Montana. and vegetational, associated with riparian Selection of hardwood draw and habitat. Summer distribution in the Salish agricultural communities adjacent to Mountains (Morgan 1993) and Swan and the pine uplands during spring-autumn Clearwater River Valleys (Leach 1982) allowed females to isolate themselves and centered in riparian-meadow habitats maximize their intake of succulent forage along major river valleys, creek bottoms, during fawning. Strong selection of the and associated drainageways. Diverse pine community in winter, especially upland timbered types associated with severe winters, may have reflected the riparian habitats provided additional overriding value of thermal cover and resources and allowed individuals to energy conservation when agricultural meet their needs throughout spring, fields and other habitats were of limited summer, and autumn. Higher elevation value to deer. areas that whitetails avoided were further Whitetails on Yellowstone River from major riparian areas and lacked the bottomlands selected habitats with diverse vegetation associated with the relatively large amounts of riparian lower, preferred habitats (Morgan 1993). cover. Patterns of habitat selection and

52 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a reproductive output by adult females foraging adaptations has been underrated suggested that reproductive habitats on in evaluating deer-habitat relationships. the lower Yellowstone were characterized On most of our study areas, and by comparatively high diversity and especially in nonagricultural areas of interspersion of riparian communities eastern Montana, high quality forage and agriculture (Dusek et al. 1989). was relatively patchy while moderate Selection and use of agricultural fields by to low quality forage was typically does with fawns increased from mid-June abundant. Deer often were forced to to September. Selection of agricultural be opportunists, foraging selectively on lands at night increased from summer to the best of what was available when and Our studies winter. Summer habitats used by males where it was available. Forage selection suggested that and nonproductive females included varied temporally and spatially within the importance more stream bed, young riparian forest, and among study areas. Food habits of foraging and agriculture than habitats used by and foraging patterns also varied among adaptations has productive females. individuals, sex and age classes, and been underrated Deciduous riparian forest and shrub species (Figs. 20 and 21). in evaluating habitats that whitetails used in winter The composition of deer diets deer-habitat provided little thermal cover for energy reflects availability of vegetation types relationships. conservation. However, the availability of and plant species occurring within home agricultural forages allowed deer to range ranges. Because of variation in nutrient widely and forage selectively to maximize and other chemical characteristics, use of intake and maintain a favorable energy individual plants can vary greatly. What balance during cold periods even when is selected in one area at one time may be feeding at night in open environments. avoided or utilized at a different time in White-tailed deer in prairie- another area. agricultural environments also selected Collectively, our data on food habits riparian areas. However, habitat diversity supported the conclusions of Coblentz and interspersion of cropland and (1970) and Suring and Vohs (1979) that rangeland also influenced deer use (Dusek deer prefer green herbaceous forage. et al. 1988). On the Cherry Creek area, Browsing occurs extensively only in whitetails preferred hardwood draws absence of green herbaceous forage, throughout the year while agricultural and although leaves and fruits of numerous most rangeland types received minor use (Wood et al. 1989). Hardwood draws interspersed in badlands decreased winter home range size and provided the only shelter available during severe winter conditions. Only rough badlands and mixed-grass habitats were consistently avoided.

Forage Selection and Use As noted earlier, mule deer and white-tailed deer are adapted to selectively forage on plant materials that are low in cellulose and high in cell soluble proteins, carbohydrates, and fats (Hanley 1984, Hudson 1985, Putman 1988). These adaptations mandate that deer seek high quality, easily digested plant materials throughout the year. Our studies suggested that the importance of

Ha b i t a t Se l e c t i o n 53 Figure 20. Generalized yearlong use of forage classes of mule deer and white-tailed deer in different environments.

Figure 21. Use of forage classes by fawn, adult male, and adult female white-tailed deer on the lower Yellowstone River during periods of vegetative growth and dormancy, 1980-86 (Dusek et al. 1989).

54 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a shrubs and trees contribute importantly to forbs was moderate, typically peaking summer and autumn diets in most areas. in June and declining in relation to Extensive browsing on twigs, evergreen availability thereafter. Use of grasses leaves, and conifer needles occurs was minor and followed seasonal trends primarily in autumn, winter, and early similar to those in the timbered breaks spring when green forage is unavailable. environment. Summer diets of mule deer in the On the Colstrip area that included Bridger Range were dominated by forbs, ponderosa pine, disturbed and while browse and grasses received revegetated areas, local agricultural fields, moderate and minor use, respectively and urban habitats, herbaceous plant (Pac et al. 1991). During autumn, diet materials were relatively abundant and composition was highly variable, but available during all seasons. Use of forbs shifted toward browse. Use of grasses exceeded that on other areas, reaching and sedges increased. In winter browse 88 percent in summer, more than 50 predominated followed by grasses and percent during autumn and spring, and 16 forbs. During spring, mule deer made percent in winter (Fritzen 1995). Use of increased use of grasses and forbs. Use grass was minor, with peaks in spring and of agricultural crops and products was autumn. generally low and only local. White-tailed deer in northwestern Annual food habits of mule deer in Montana were primarily browsers timbered breaks paralleled those of deer during summer and autumn (Morgan in the mountain-foothill environment. 1993). Grass and grasslike plants Some seasonal differences in use of were selected in addition to browse in forage classes occurred as a result of wide spring, while forbs accounted for most fluctuations in forage availability between of the non-browse forage in summer and seasons and years in the Missouri River autumn. Grass, typically found in greater Breaks (Hamlin and Mackie 1989). abundance in riparian areas, made up Shrubs accounted for an average 36 a substantial portion of the diet only in percent of the diet from May through July spring when use of these habitats was and 50 percent or more of monthly diets greatest. Riparian meadows and adjacent during August-March. Peak use of shrubs open to diverse upland forest habitats occurred in December and January. Forbs also provided forbs and browse plants comprised one-third or more of the forage during late spring and early summer. As used from April through September, meadows and other low elevation sites with highest use during May-July, when began to dry and forage plants matured, they accounted for 60-70 percent of the deer moved to higher elevation forest diet. Lowest use of forbs occurred in habitats. Concurrently, deer shifted their December and January, coincident with diets from grasses and forbs to browse. greatest use of browse. Use of grasses Agriculture was limited throughout the was relatively minor, with peak use forested mountain valley region so crops from late March through April and, in were used only locally. some years with autumn greenup during Along the lower Yellowstone River October and November. Availability of in eastern Montana, food habits of white- agricultural crops was very limited and tailed deer were influenced by the relative local throughout the breaks. abundance and variety of agricultural Studies in prairie environments also crops as well as the natural forage indicated variability in forage selection available (Dusek et al. 1989). Browse in relation to local environmental (43 percent) and agricultural crops (39 characteristics and land uses (Dusek percent) dominated the yearlong diet 1975, Jackson 1990, Fritzen 1995). In (Figs. 20 and 21). Forbs were important prairie environments, shrubs comprised and received moderate use only in over 50 percent of the diet during all summer, while grasses received moderate seasons and up to 96 percent in winter use in spring. Use of agricultural crops (Dusek 1975, Jackson 1990). Use of was important during all seasons, but was

Ha b i t a t Se l e c t i o n 55 minimal in May and June and peaked in peaks of activity near sunrise and sunset autumn. Alfalfa was the primary species (Fig. 22). This bimodal pattern is used during summer while row crops generally characteristic of both species and small grains were selected in late of deer inhabiting a broad spectrum autumn and winter. However, differences of environments. However, deer are in availability influenced forage selection capable of considerable adjustment in spatially along the riverbottom. The activity patterns as they adapt to local importance of agricultural crops environmental conditions. surpassed browse where alfalfa or other The most distinct feature of daily crops were readily available. Availability activity patterns of white-tailed deer on of crop residues also varied in relation to the lower Yellowstone was movement irrigation and other cropping practices. between woody cover and agricultural Composition of the diet of white- fields. Daily activity of whitetails along tailed deer in the Long Pines reflected the lower Yellowstone River during their preference for ponderosa pine summer included very sedentary use of habitats. Pine habitats were characterized riparian cover during daytime, restricted by high diversity and abundance of low movement to nearby alfalfa fields before growing shrubs, and browse was strongly sunrise, and greater movement and more selected in all seasons. Forbs were most intense use of alfalfa fields after sunset abundant and received moderate use in (Fig. 22). spring and summer, with declining use Daily activities of adult females in autumn and winter. Grasses received during summer were influenced by their moderate use in spring and autumn. reproductive status. Non-producing Agricultural crops were used by some deer that moved daily from uplands to females made greater use of agricultural bordering agricultural fields. Alfalfa was fields during June-mid August while does selected in spring, wheat and barley in with fawns were largely sedentary in autumn. riparian cover. By early September, use of Like other attributes of habitat use, agricultural fields by productive does had ...food habits food habits reflected a high degree of risen to levels similar to nonproductive reflected a high selection of plant materials available to females. degree of selection individuals. Each deer selected the best During winter, daily patterns of plant materials of what was available within its home included greater activity and mobility available to range. Both behavior and resource associated with more intense nocturnal individuals. requirements motivated habitat and food use of agricultural fields (Fig. 22). Deer selection. often traveled 2-2.5 km to feed in fields located some distance from riparian cover. Whitetails usually returned to the Activity Patterns same daytime bedding and loafing areas We were able to monitor detailed even though other cover was available activity patterns by direct observations near the fields being used at a particular of deer and the activity and distances time. During winter, deer activity within traveled by radio-collared deer during cover patches used during daylight hours 24-hour tracking sessions. A more was much greater compared to summer generalized index to changes in activity and involved longer bouts of feeding on and mobility was gathered from native riparian forage. measuring the average distances traveled Ambient temperatures had a by radio-collared deer on a monthly basis. decided effect on daily activity, but this The two approaches helped gauge how relationship varied among the different deer were exploiting their habitat and environments occupied by deer. During why they were employing a particular summer on the lower Yellowstone combination of activities in various River, whitetail activity in midday was environmental settings. significantly lower at temperatures above The general pattern of deer activity than below 32˚ C. This relationship during a 24-hour period includes major reversed one hour after sunset with

56 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Figure 22. 600 SUMMER Mean distances moved Elk Island between successive Intake locations of deer 500 during 24-hour tracking sessions 400 through summer and winter on the Elk Island and Intake 300 units along the lower Yellowstone River 200 (Dusek et al. 1989).

100

Distance Between Hourly Locations (m) 0 -6-7-8 -5 -4 -3 -2 -1 12345678-9 -8 -7 -6 -5 -4 -3 -2 -1 12345 67 8 Hours from sunrise Hours from sunset

600 WINTER

500

400

300

200

100

Distance Between Hourly Locations (m) 0 -6-7-8 -5 -4 -3 -2 -1 12345678-9 -8 -7 -6 -5 -4 -3 -2 -1 12345 67 8 Hours from sunrise Hours from sunset greater movements observed at higher In contrast, Wood (1988) reported than lower temperatures. that ambient temperature and the effects During winter when whitetails had of wind chill significantly influenced access to crop residues and maintained winter habitat use and daily activity a high nutritional plane, they continued of mule deer occupying native prairie- to exploit agricultural fields even when badlands habitat. Wind was a constant wind chill temperatures dropped to -60˚ factor in this environment, averaging 21 C. With low temperatures and high km/hr during 57 random measurements. winds deer might be expected to select Over a normal winter ambient microsites with lower wind speeds within temperature range of -24° to 11° C, a field such as along edges or ditch banks, deer avoided windy sites at wind speeds but this was not observed for whitetails greater than 10 km/hr. By use of shelter along the lower Yellowstone (Herriges associated with badlands topography, 1986). mule deer reduced conductive heat loss

Ha b i t a t Se l e c t i o n 57 by 47 percent at feeding sites and by the river bottom or moved into adjacent 61 percent in bedding sites. Energy uplands during mid-late autumn. This gained from native forage offset the shift coincided with firearm hunting energy lost from increased exposure and seasons, introduction of cattle onto the mobility associated with feeding only river bottom, sugar beet harvesting, and when conditions were relatively mild social behavior of deer during the rut. (Wood 1988). Foraging was energetically During all seasons, whitetails inefficient during severe winter weather on the lower Yellowstone River made conditions. Bedding in protected sites disproportionately heavy use of areas was the favored strategy because it where cattle were absent, although conserved energy. use of these areas often involved only Adjustments in activity also were minor shifts in activity and distribution. made in response to different types Avoidance of cattle may reflect either of human disturbance. Vogel (1983) alteration of food and cover (Mackie concluded that deer activity during 1978) or a social intolerance (Lonner daylight hours decreased with increasing 1975). The latter was apparently of levels of human disturbance. Whitetails greater importance along the lower inhabiting more developed areas became Yellowstone River because deer vacated increasingly nocturnal and secretive an area as soon as cattle arrived and and made greater use of cover during returned to the grazed areas when cattle the day. Herriges (1986) indicated that departed (Compton 1986). agricultural fields in close proximity Telemetry tracking during 24-hour to human disturbance did not receive periods on some of our study areas diminished use by white-tailed deer. provided a better understanding of However, most of the deer activity in how nocturnal activity influenced our Whitetails those fields occurred during hours of interpretation of movement patterns inhabiting more darkness in both summer and winter. and home range size. Herriges (1986) developed areas Fritzen (1995) described nocturnal and Dusek et al. (1989) reported that became increasingly movements by mule deer that exploited nighttime activity associated with nocturnal and seasonally available forages associated movement out of large blocks of riparian secretive ad made with landscape plantings within the city cover to agricultural fields increased greater use of cover limits of Colstrip. Compton (1986) seasonal home range size 2-3 fold during the day. reported that whitetails made greater compared with daytime ranges. This use of larger tracts of riparian cover in dichotomy in habitat use tended to occur

58 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a in environments where preferred bedding that satisfied all resource requirements. and foraging sites were spatially distinct. However, these individuals represented a Jackson (1990) described day and small proportion of the total population in night differences in activity patterns for this rather simple, open environment. mule deer occupying native ranges in Dichotomy between areas used open sagebrush/grassland habitats. He through the day was less evident for deer reported that deer were highly active and occupying diverse timbered habitats. moved extensively at night. Summer Morgan (1993) reported that white- home ranges based on combined day tailed deer in northwest Montana made and night locations of radio-collared somewhat greater use of small, open deer were 105-934 percent larger than meadows at night. However, these those based only on daytime locations. meadows usually occurred interspersed One mule deer doe used totally different with other foraging sites, security cover, areas during each of three 24-hour and bedding areas within small seasonal tracking sessions; and none of these home ranges. Daily travel between areas overlapped the home range polygon bedding and foraging areas was minimal, plotted from daytime locations (Fig. 23). and other patterns of habitat use were These extended nocturnal movements more subtle than those in more open were associated with deer that spent environments. The luring influence of daytime hours in shale hills and at agricultural crops and proximity to high night traveled to drainage bottoms and levels of human disturbance was generally reservoir areas where succulent forage absent for whitetails in montane forests was available. of western Montana. Because of this, data Other female mule deer on Jackson’s collected only during daylight hours may (1990) study area showed much less be more representative of overall habitat dichotomy in day and night habitat use. relationships for deer in those areas Home ranges calculated from daytime than in open or patchy environments locations were similar in size to areas with significant human disturbance and used during both diurnal and nocturnal agriculture. periods (Fig. 24). Deer with relatively Changes in monthly activity and small home ranges occupied diverse sites mobility provide a generalized indicator

0 0.5 1.0 Km

7/23/86 - 7/24/86 Nocturnal 8/19/86 - 8/20/86 Home 9/11/86 - 9/12/86 } Ranges 6/1/86 - 9/30/86 (Daytime Seasonal Home Range) Total Seasonal Home Range

Figure 23. Nocturnal home ranges compared to daytime and total seasonal home ranges of one radio-collared female mule deer during summer on open sagebrush-grassland habitat, northeast Montana (after Jackson 1990).

Ha b i t a t Se l e c t i o n 59 of adjustments in the overall strategy of habitat use by deer during the biological year (Dusek et al. 1989, Hamlin and Mackie 1989, Pac et al. 1991). For example, average activity radii for radio- collared adult female mule deer in the Bridger Mountains show annual trends in mobility for mule deer in different environments (Fig. 25). Differences in physical and vegetative characteristics of east and west slope winter ranges resulted in divergent patterns of mobility and activity. During winter and spring mule deer on the east slope of the Bridger Range exhibited significantly higher mobility. In this open, dry environment deer remained widely distributed across large areas consisting of shrub-grassland and Deer No. 1187 some dryland agricultural fields. Such dispersion seemed to favor efficient Deer No. 1087 allocation of forage. Even under severe weather conditions, the overall winter strategy continued to emphasize high mobility to access available forage on widely separated, windblown ridges. On the west slope, very low mobility during the winter months (Fig. 25) was consistent with a strategy that favored energy conservation (Youmans 1979) through a winter period that averaged 40 days longer on the west as compared with the east slope. The spike in mobility that occurred for deer on the east slope during April and May was related to their earlier departure from winter ranges and greater distances traveled en route to summer ranges. On Deer No. 1287 0 0.5 1.0 km the west slope, deer left winter ranges later, and distances traveled to summer range were shorter. However, migratory 7/23/86 -7/24/86 Nocturnal movement extended into June because 8/19/86 -8/20/86 Home of the high elevation habitats that were 9/11/86 -9/12/86 } Ranges exploited. 6/1/86 -9/30/86 - Daytime Seaonal Home Ranges Trends in mobility among deer in the two environments converged during June- Total Seasonal Home Range August (Fig. 25). This was associated with fawn-rearing requirements and the Figure 24. Nocturnal home ranges compared to daytime and total seasonal sedentary behavior of deer on summer home ranges of three radio-collared female mule deer during home ranges. Such a convergence summer on open sagebrush-grassland habitat, northeast Montana in habitat use would likely occur (after Jackson 1990). when resource conditions were most favorable and deer preferences could be fully expressed even in very different environments.

60 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a 4.0 East slope West slope 3.5 = 1 SE 3.0

2.5

2.0 Kilometers 1.5

1.0

0.5

0.0 JanFeb MarApr MayJun JulAug SepOct Nov Dec ******MONTH *** Figure 25. Number of Activity Radii A comparison of mobility based on monthly average activity radii (km) for adult E 166 180 165 258 282 206 207 166 173 185 157 125 female mule deer on the west and east slopes of the Bridger Mountains. Asterisk indicates significantW 467 differences480 780 1079 between607 areas474 42P 7 < 360.05,0 33one-way6 369 28ANOVA9 28 9(after Pac et al. 1991).

Changes in mobility and strategies of environments and the specific behavioral habitat use occurred again in September and biological strategies required to when some west slope deer responded exploit them. to early autumn snowstorms by crossing the Bridger Divide in early movement Social Organization toward winter range. However, deer on the east slope exhibited greater mobility As discussed earlier, many attributes during October-December as increasing of social behavior are common to numbers of deer moved long distances both mule deer and white-tailed deer. toward winter ranges. West slope deer Interaction between behavior and completed movement toward winter resource requirements produced an array ranges during October and November, of strategies by which individuals of both and by December mobility was limited species exploited various environments. in the manner characteristic of winter. Social relationships are never static Annual differences in autumn movements because the environments deer inhabit and mobility appeared to be influenced and the manner in which deer use them primarily by weather patterns, though are constantly changing. In spite of this, rutting behavior and hunting probably members of established populations had some effect. function in an orderly manner. Social organization depends on Comparisons of data collected on individual recognition and established deer study areas across the state indicated patterns of communication between sex that adult females demonstrated very and age classes. It functions to minimize similar low levels of mobility during the tension among individuals, contributes summer fawn rearing period regardless to efficient allocation and use of habitat, of species and the type of habitat and enhances survival and reproductive occupied (Dusek et al. 1989, Wood et fitness among members of the population. al. 1989, Hamlin and Mackie 1989, Pac Much of this is accomplished through et al. 1991). In contrast, winter mobility adjustments in the size and composition patterns were more divergent, reflecting of social groups and their spatial differences in characteristics of local distribution across the landscape.

Ha b i t a t Se l e c t i o n 61 We distinguished four types of social types are of secondary importance in groups. Doe groups include at least one occurrence and persistence. In most adult (1 year or older) female but no environments, composition and size of mature (2 year and older) males. Buck mule deer doe groups was determined groups consist of at least one mature primarily by reproductive effort and male but no adult females. Mixed groups success. contain both adult does and mature During June and July, the majority of bucks. Yearling male groups include at does were either solitary or with newborn least one yearling male with no adult does fawns. This occurred on all study areas or mature bucks. Doe, buck, and mixed regardless of topographic and vegetative groups may include fawns and/or yearling structure. Fawns, the age class most Social organization males. However, in the Missouri River vulnerable to predation, occurred in the in both species Breaks, buck and most mixed groups smallest social groups widely distributed centers around of mule deer usually did not contain across all available reproductive habitat. maternally-related fawns (Hamlin and Mackie 1989). After Hamlin and Mackie (1989) reported doe groups. antler shedding in February, the identity that does without fawns joined other of the four social groups cannot be nonproductive does rather than remaining distinguished reliably until antler growth solitary during early summer. Female resumes in late-May and early June. groups were largest when reproductive Social organization in both species success was lowest, and vice versa. centers around maternally-related doe The size of mule deer doe groups in groups. The other three social group the Bridger Mountains gradually increased during August-October (Fig. 26) as fawns were weaned and began to travel routinely with the doe. Groups consisting of two or 13 Mixed group more does with their fawns also became 12 Adult female group more common as maternally-related Adult male group females regrouped. During the entire 11 summer-early autumn period fewer than 10 Yearling male group 5 percent of all adult females occurred in 9 groups with mature males (Hamlin and Mackie 1989, Pac et al. 1991). 8 Doe groups declined in size 7 somewhat in November with the onset 6 of the rut. Courtship and bucks chasing does in estrous apparently contributed Group Size 5 to temporary disruption of social 4 groups. Disturbance during the hunting 3 season may also have disrupted social 2 organization, resulting in smaller group size. 1 As autumn progressed toward winter, 0 forage desiccation and snow accumulation JunJul AugSep OctNov DecJan became primary influences on habitat Month use, overriding behavioral preferences for small group size and maximum Number of Groups dispersion. Size of doe groups increased Mixed group: 4 13 18 3 8 20 136 167 sharply from November through January Adult female group: 194 390 402 240 185 136 744 1127 as deer became restricted to winter Adult male group: 36 160 236 58 34 23 70 137 maintenance habitats (Fig. 26). Yearling male group: 4 44 33 19 10 10 20 26 In many populations, social Figure 26. affiliation and size of buck groups Monthly mean size of four social groups of mule deer on the west was undoubtedly minimized by high slope of the Bridger Mountains (after Pac et al. 1991).

62 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a hunting mortality. Mule deer buck group sizes during winter and spring. groups averaged about 2.0 during June- The largest groups tend to occur in September on the west slope of the mountainous habitat where snow Bridger Mountain Range (Fig. 26), and accumulation forces migratory deer to only about 20 percent of all buck groups congregate loosely on small, sparsely contained three or more males (Pac et al. timbered winter ranges. In other areas, 1991). deer may temporarily aggregate in Buck groups were smallest in large groups to exploit high quality food November during the rut when most resources. On the west slope of the mature bucks were alone or in mixed Bridger Mountain Range, group size was groups. Association between mature largest in spring, averaging 11 deer/group bucks and yearling bucks reached a with 27 percent of all groups consisting minimum during the rut (Pac et al. 1991, of 10 or more deer (Pac et al. 1991). Hamlin and Mackie 1989). Based on Group sizes also peaked during spring in observations of marked males, Hamlin the Missouri River Breaks but averaged and Mackie (1989) reported that size of only 4.5 deer/group (Hamlin and Mackie mature buck groups in winter-spring was 1989). equivalent to summer. Results of our studies showed Mixed groups were temporary that sex and age classes occurred not associations that occurred in relation only in different social groups, but to specific activities or in response to also in separate areas during much of environmental factors. Mixed groups in the year. Segregation of the sexes was the Bridger Mountains were uncommon apparently the end result of a number during June-October when does sought of interacting factors. As discussed isolation to raise fawns and bucks earlier, the peak energetic investment associated with one another (Fig. 26). in the reproductive effort by the two Observations of mixed groups increased sexes occurs at different times during in November with the onset of the the year; each with a corresponding set breeding season and peaked in December- of physiological consequences. Efforts January when weather conditions forced to meet these different reproductive deer to congregate on limited winter roles begin far in advance of the actual Results of our habitat. During all months mixed groups events when resource requirements studies showed were the largest of all group types. unique to each sex result in preference that sex and age Yearling male groups were for habitats with different characteristics. classes occurred temporary affiliations because individual Habitat partitioning among does and not only in members preferred to associate with bucks was most evident during the fawn- different social adult females or mature bucks whenever rearing period in summer when sexual groups, but also tolerated. On the west slope of the segregation was reinforced by aggression in separate areas Bridgers, yearling male groups averaged of maternal does toward other deer. during much of the 1-2 individuals in all months during Figs. 27 and 28 present two year. June-January (Fig. 26). Yearling males examples of social distribution at widely increasingly associated with adult females separated points along the environmental as the biological year advanced while spectrum in Montana. The high degree association between yearling and mature of topographic and vegetative diversity in bucks peaked during July-September and the Bridger Mountains (Fig. 27), provided declined thereafter (Hamlin and Mackie ample opportunity for habitat partitioning 1989, Pac et al. 1991). among the primary social groups during Relationships concerning group fawning. composition and size for mule deer in the Aggressive behavior by maternal Bridger Mountain Range and Missouri does resulted in efficient allocation of River Breaks generally apply to both diverse montane forest habitats among species in other environments (Wood the female segment. It also enforced the 1986, Dusek et al. 1989). Greatest segregation of mature bucks, yearling differences among populations involved males, and unproductive females into

Ha b i t a t Se l e c t i o n 63 Figure 27. Spatial segregation of social groups mule deer at fawning time (mid-June) in the Bridger Mountains.

64 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Figure 28. River. Spatial segregation of social groups white-tailed deer at fawning time (mid-June) along the lower Yellowstone

Ha b i t a t Se l e c t i o n 65 adjacent remaining habitats with different the greatest concentration of forage characteristics. Mature bucks could and cover. The greatest number of satisfy their requirements for large fawning territories used by maternal does quantities of forage in more open, less occurred in these communities situated secure habitats within the subalpine zone, adjacent to irrigated agricultural crops. along drier ridge tops within the montane Moderate numbers of maternal females forest, or on low elevation winter were associated with mature cottonwood maintenance habitats. Unproductive stands on islands and bottomlands lacking does and yearling males also utilized agricultural fields. Limited numbers of these habitats as well as portions of the does and fawns utilized open, upland montane forest that were not occupied by agricultural areas interspersed with maternal does. hardwood draws. The mosaic of topography and Groups of mature whitetail bucks vegetation along the lower Yellowstone were distributed along the interface River was far more subtle than that in the between mature cottonwood stands Bridger Mountains. Consequently, habitat and dry badland bluffs and in young partitioning by white-tailed deer occurred cottonwood stands along the river’s at a finer scale (Fig. 28), although its edge. Bucks also occurred in the functional significance in providing social interior of mature stands of cottonwood organization and allocation of habitat wherever use by maternal does was was similar. The distribution of all social limited. Unproductive does and yearling groups on the lower Yellowstone River males were generally distributed in at fawning time revolved around the willow thickets along the river, in open preference of maternal females for mature agricultural areas, and along hardwood cottonwood and green ash stands. These draws that traversed the dry terraces structurally diverse habitats provided adjacent to the riverbottom.

66 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Population Characteristics and Dynamics

The Concept of Population

The concept of “population” is connote a “deme” (Gilmour and Gregor fundamental to the field of ecology. 1939). However, no single definition is widely Nelson and Mech (1987) suggested accepted, and biologists often use the that white-tailed deer associated with term only vaguely to designate any group individual winter yards in northeastern of individuals of the same species. Minnesota comprised behaviorally Some ecologists conceptualize a separate “subpopulations” that constituted population as a discrete biological or genetic demes. They hypothesized ecological unit. For example, Nicholson that deer populations may consist of (1957) proposed that population be conglomerates of subpopulations or defined as “a group of interacting and demes. Groups of subpopulations interbreeding individuals that normally occupying discrete patches of habitat also has no contact with other groups of the have been termed “metapopulations” same species...that is to say...a discrete (Gilpin and Hanski 1991, Hunter 1996). dynamic unit.” Such populations may Generally, in all existing definitions, be selected for adaptation to the specific the essential criteria for conceptualizing environments in which they live (Mayr or designating a population are species, 1970), an interpretation that also can group, and space. More recently,

The Co n c ep t o f Po p u l a t i o n 69 interbreeding and other elements over time. Both environment and of genetics have become important. strategy influence deer demographics and Although Nelson and Mech (1987) dynamics. and others have implied a relationship Our concept of population as a between a deer population and its discrete behavioral, biological, and environment through social organization, ecological unit was supported by patterns no one directly links habitat and of population development and growth population in their definition. Thus, none observed or inferred for mule deer. We of the existing definitions for population speculate that development of deer completely described the deer population- populations follows a predictable pattern habitat relationship we observed. from initial colonization through the Based on behavioral responses various stages of habitat exploitation and associated with habitat selection and population growth. It also was possible use, we distinguished mule deer and to conceptualize additional categories or white-tailed deer populations as relatively levels of organization below and above discrete, dynamic units. However, we the population. also expanded this definition to imply behavioral and biological adaptation to the habitat(s) in which deer occur. Colonization and A similar view has been expressed by Lidicker (1994, 1995) with respect Development of to distribution and abundance of mammals in general. Ostfeld (1992) also Populations indicated that the community context of a population is an essential aspect of Development of a new deer its functioning and we cannot expect to population in any environmental complex understand populations independently of entails: this context. • colonization by pioneering individuals In our perspective, a deer population consists of an assemblage of individuals • population growth and spread through and family groups bonded together reproduction, dispersal of young females, and behavioral adaptation to In our perspective, by traditional distribution, movement, fill all available reproductive habitat a deer population and other habitat use patterns within a consists of an discrete unit of habitat. As described • behavioral fine-tuning, including assemblage of earlier, they are established and habitat partitioning and modification individuals and maintained through social organization of distribution, movements, and other and behavior inherent to a matriarchal family groups habitat use for optimal exploitation of society. Lifelong fidelity of individual bonded together the area. female deer and their offspring to by traditional Colonization of vacant habitat by seasonal ranges ensures some continuity females primarily involves young adults distribution, of basic distributional patterns through movement, and dispersing from natal home ranges and time. Further, it ensures continued emigrating from natal populations. other habitat use population unity as a group of partially patterns within Such movements typically occur during interacting and interbreeding individuals. late spring-early summer when adult a discrete unit of Based on this, the term “population” is habitat. females aggressively defend their fawning no longer a conceptual abstraction (Pac territories and force some yearlings to et al. 1991). Yearlong distribution of leave the maternal range. At this time, each population delineates the complete environmental conditions are most ecological unit required to sustain it. benign and provide dispersing young deer Physical and biotic characteristics of the with the best conditions for traversing habitat determine the strategies that can unfamiliar terrain and relocating outside be successfully employed by members of the maternal home range. the population to effectively occupy and Successful early colonists select maintain themselves in that environment home ranges that meet all of their

70 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a resource needs for survival and numbers of deer on limited winter reproduction. Because all habitats are maintenance habitat. It also gives potentially available, some emigrants will social organization to the developing/ settle in areas unsuitable for survival. expanding population as an array of Females that localize in areas which family groups comprised of related function both as winter maintenance females and offspring. and reproductive habitat would meet Exploitation of montane forest their needs and be afforded the greatest habitats, at least initially, requires only probability of survival. These deer moderate specialization. The distance and the habitat complexes they occupy separating seasonal ranges is slight and become the nuclei for expansion of migration involves only simple, up-down fledgling populations. Yearlong residency, movements. However, as reproductive which requires minimal specialization in habitat at middle elevations adjacent seasonal movement and habitat use, may to foothills becomes filled by adult be a fundamental strategy for successful females and their offspring, dispersing pioneering and colonization. juvenile females from both foothill Growth and development of and adjacent montane forest habitats Once pioneering colonizing populations requires continued must travel further and adopt more is accomplished, a recruitment and dispersal of young complex movement and habitat use successful colonist females to additional habitat. Although patterns. Habitats available at this probably employs foothills and other low elevation habitats time may be located primarily above a fundamental often provide sufficient resources for the montane forest, in more distant strategy of winter survival of large numbers of deer, forests across high mountain divides, or yearlong residency. local areas capable of providing succulent in patchy, low-elevation environments. forage and other resources essential for Use of such habitats requires greater successful reproduction by adult females specialization in movements and habitat may be limited. This minimizes the use. number of generations of females that This increasingly complex can establish home ranges near resident distribution and movement pattern matriarchs and results in dispersal of permits members some juvenile females to adjacent areas. In mountain ranges, young females seeking summer home ranges find extensive areas of mesic habitat in montane forests at middle elevations above the foothill zone (Pac et al. 1991). Although montane forests may provide plentiful reproductive habitat, snow limits use in late autumn and winter. This requires movement to lower elevations in autumn and the establishment of separate winter home ranges. If these winter ranges overlap the area on which the deer had ranged as fawns, they also serve to maintain the linkage between related females and fawns through the habitat they occupy. This seasonal association of maternally- related females on specific portions of winter range probably fosters behavioral accommodation of increasing

The Co n c ep t o f Po p u l a t i o n 71 of developing populations to exploit use lower quality, maintenance habitats all habitat capable of supporting available within that population-habitat reproduction in a particular location. unit. In some habitats this may also lead Ultimately population expansion and to other behavioral adaptations and more growth become limited as individuals refined use of space and resources by all encounter environmental barriers or deer. compete for space and other resources Colonization and selection of with deer from other developing habitats by males followed a different populations. Such barriers and areas of pattern. The social bond between overlap form the boundaries between young male mule deer and their mothers habitats used by deer from adjacent weakens earlier, and some leave family populations in the Bridger Mountains groups prior to the onset of fawning (Pac et al. 1991). Movement beyond the (Hamlin and Mackie 1989). In our boundaries usually results in emigration studies, mule deer dispersed from spring to different populations rather than through autumn. Most yearling male long-distance movement within the same whitetails spent summer on the periphery population. of natal home ranges and dispersed Successful patterns Successful patterns of habitat use during late autumn. Most yearling males of habitat use become traditional and transmitted from in both species dispersed from natal home become traditional generation to generation through the ranges and some left natal populations and transmitted long-lasting social bonds of matrilineal permanently. from generation to groups. This process provides for Young males of both species can generation through continued efficient exploitation of readily colonize vacant habitats because the long-lasting available habitat. It also assures that of their propensity to disperse from natal social bonds of boundaries between habitats used by home ranges. Most males of both species matrilineal groups. individual populations and other natural eventually established home ranges within groupings of deer are maintained. distributional perimeters previously High rates of recruitment, dispersal, established by females of the populations. and establishment of young females This concurred with Porter et al. (1991) in vacant habitat are associated with that males add little to spatial expansion eruptive population growth following or overall distribution of populations. colonization. When most or all Data for mule deer in the Missouri reproductive habitat is filled, habitat River Breaks suggest that recovery of partitioning becomes increasingly populations after periodic declines follows important, and young deer are forced to a pattern similar to initial colonization and development. Following severe population declines, spatial distribution of females centers in “core” areas that meet all requirements for survival under the most extreme environmental conditions. Core areas are local habitats with optimal mix and juxtaposition of both winter maintenance and reproductive habitat. They tend to be surrounded by areas of decreasing habitat quality and declining deer use. Deer occupying such core habitats are primarily residents. In the Missouri River Breaks, high recruitment and pre-saturation dispersal led to rapid filling of core habitat throughout the area (Hamlin and Mackie 1989). As these habitats filled with mature females, young does were increasingly forced to use lesser quality

72 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a reproductive and maintenance habitats. Under severe conditions and in early These usually were smaller patches spring, they often shared common range with less topographic and vegetational areas but avoided mixing with groups diversity than core habitats. They were of deer associated with adjacent winter also subject to greater fluctuation in habitat. In late spring, subpopulation environmental conditions, offered less identity became blurred as individuals security from predators, and increased and groups used overlapping movement opportunity for competition with livestock corridors and summer home ranges that and other wild ungulates. Use of these radiated outward from each segment of a habitats often resulted in larger home particular winter range. ranges, distributional shifts that included Overlapping subpopulations seasonal use of accessory areas, and associated with a definable unit of habitat migrational movements similar to patterns comprised population-habitat units (Fig. observed in mountain environments. 29C) (Pac et al. 1991). The existence Although we speculate recolonization of seven distinct population-habitat units proceeds similarly to initial colonization, in the Bridger Range was related to the social behavior of deer in river breaks and occurrence of discrete areas of winter prairie-badland environments displayed maintenance habitat around the perimeter greater plasticity than in mountain- of a long, narrow strip of diverse foothill habitats. Dispersal rate of reproductive and summer maintenance young females was greater than in the habitats. Each population-habitat unit mountains, perhaps because important comprised a discrete unit from the habitats occurred in smaller patches and/ standpoint of mule deer distribution, or average recruitment rates were higher. habitat use, demographics, and dynamics Further, it appeared that dispersal was (Pac et al. 1991). This concept does not adaptive and resulted in rapid fill and require total spatial separation or physical complete use of all available habitat in isolation of populations and some overlap these environments. Once populations of individuals from adjacent population- were established, high rates of dispersal habitat units commonly occurred along were less important in complex mountain the boundaries between units. environments. Higher levels of population organization in the Bridger Mountains were defined by genetic sampling of Population Organization mitochondrial DNA (Cronin et al. 1991) as well as other aspects of biology, behavior, and demographics (Pac et al. Mule deer in the Bridger Mountain 1991). Those data suggested that the Range provide a graphic example of four adjacent population-habitat units on population organization in the species, at the east slope of the Bridgers comprised least for mountain-foothill environments. a single deme (Fig. 29D), or the largest We recognized a series of five natural community of potentially interbreeding groupings of increasing complexity individuals within a generally similar and among mule deer inhabiting the Bridger continuous environment (Gilmour and Range (Fig. 29). Gregor 1939). A separate and distinct The most fundamental unit was deme included the three population- the matriarchal family group (Fig. habitat units along the western portion 29A). Aggregations or loose bands of of the Bridger Range (Fig. 29D). matrilineal groups and attendant buck Movements, including dispersal patterns groups associated with specific units of young deer, indicated interchange of habitat appeared to constitute local of animals occurred primarily among subpopulations (Fig. 29B) (Youmans adjacent population-habitat units within 1979, Pac et al. 1991). In winter, these demes. Dispersal across the Bridger groups used overlapping home ranges Divide to another deme was limited on restricted winter maintenance habitat. almost entirely to young males.

The Co n c ep t o f Po p u l a t i o n 73 B. Population Subunits

A. Matriarchal Family Group

+

+

B

r

e g d i +

r

+ d i v D i + +++ + e

� 0 1 2 km � 0 1 2 km

D. Demes C. Population-Habitat Units

� � 0 4 8 km 0 4 8 km

E. Metapopulation Winter range

Population-habitat unit boundary

West deme

East deme

� 0 4 8 km Seasonal movement vectors

Figure 29. Five levels in a series of natural grouping of mule deer in the Bridger Mountains.

74 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a At the highest level of organization, geographic portions of the area formed the two adjacent demes separated by more or less discrete subpopulations. a major landscape discontinuity may In these environments, intensive radio- represent a single Bridger Mountain tracking studies, similar to those on our “metapopulation” (Fig. 29E). Hanski and Bridger Mountain and Missouri River Gilpin (1991) and Hunter (1996) defined Breaks study areas, may be necessary to a metapopulation as a “population of define population-habitat units and higher populations.” levels of organization. We had confidence The same organizational in describing organization only through relationships may have characterized the subpopulation level. natural groupings of mule deer in In eastern Montana, white-tailed timbered breaks and prairie-badlands deer along the lower Yellowstone River environments. However, the broad apparently comprised a population expanse and subtle changes in distinct from whitetails occupying environmental characteristics across the adjacent prairie-agricultural and timbered northern plains made it difficult to define upland environments (Dusek 1987, Dusek units at the population and higher levels et al. 1988, Dusek et al. 1989, Wood of organization. Winter maintenance et al. 1989). Within the riverbottom, habitats were more diffuse or patchy subpopulations were associated with and often located within yearlong particular segments of floodplain (Dusek home ranges (Fig. 7). They did not et al. 1989). These exhibited differences necessarily form nuclei for development in demographics and dynamics similar and recognition of organizational units to differences exhibited by mule deer as in the mountains. Instead, population in different population-habitat units in structure centered in matriarchal groups the Bridger Mountains. However, the and subpopulations consisting of loose continuous yearlong distribution made it aggregations of those groups and impossible to clearly delineate discrete associated males. populations. Studies in other areas In prairie environments, spatial show that contiguous subpopulations distribution and movement patterns of white-tailed deer exhibiting different allowed recognition of matriarchal groups demographics and dynamics also are associated with core areas that included genetically separate units (Manlove et al. both reproductive and maintenance 1976, Ramsey et al. 1979, Rhodes and habitats for all members of the group. Smith 1992, Scribner 1993). Loose aggregations of deer within specific

The Co n c ep t o f Po p u l a t i o n 75 Population Characteristics

Our studies characterized all habitat complexes include some populations of both species across much proportion of unused area or rarely used of the broad spectrum of environments habitat which can be difficult to delineate they encounter in Montana. From a without intensive study. landscape perspective, each study area In our studies, the amount, quality, and deer population was characterized by and distribution of reproductive habitat demographics unique to the environment appeared to be the primary factor in which it occurred. The more alike influencing density distribution and the environments, the more similar potential total numbers of productive the populations were in demographic adult females in each population. The characteristics and dynamics. mix of food, cover, and water available in reproductive habitats determined the minimum home range size needed by Population Size and individual females. The total amount and quality of reproductive habitat Density determined how many home ranges or adult females each area could support. Population size and density have Complex, diverse environments capable of been used interchangeably. Population providing optimum resource availability allowed females to attain comparatively ...the amount, size refers to the total number or estimated number of deer in a population high densities. Such environments often quality, and included adjacent habitat for young adult distribution of at a given time; density is the number of deer per unit area. Population size females and adult males that resulted in reproductive is finite; density is relative, but it is the larger overall populations. Use of patchy, habitat appeared parameter most used when comparing variable environments characterized to be the primary deer numbers among areas of different by reduced or unpredictable resource factor influencing size. Deer density is far from uniform availability was associated with greater density distribution across environments. Density distribution movement, larger home ranges, and lower and potential varies widely between seasons in some overall densities. total numbers of environments, but it changes only slightly Winter maintenance habitat appeared productive adult in others. to exert less influence than the total females in each Measurement and comparison amount and quality of reproductive population. of deer densities also is hampered by and summer maintenance habitat on topographic variation in land surface per population size or overall density of unit map area. For example, a square deer. This does not negate the overall kilometer of steep, rugged mountainous importance of winter maintenance habitat terrain may include 10 percent more area to occurrence and abundance of deer. The than a square kilometer of moderately existence of deer populations depends dissected breaks and 20-30 percent on availability of areas that ameliorate more surface than the same area of level adverse effects of deep snow, provide to rolling prairie. In addition, nearly shelter from extreme temperatures, and

76 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Varying quantity provide escape cover from humans and Conversely, the Battle Ridge population and quality of other predators. Also, high quantity and occupied the largest area of any summer habitats quality of winter forage supplemented and population. It included a large amount of resulted in reduced the rate of utilization of energy winter range that sustained only a modest differences in reserves. However, deer are adapted to population (~850) at low average density numbers of deer 2 tolerate restricted distribution and high (1.4/km ). That low density reflected associated with densities during winter. Therefore, total large amounts of unused habitat and only winter ranges. numbers of deer in a population reach limited reproductive habitat distributed Those numbers levels dictated by the quantity and quality predominantly in patches around the could not be of available summer habitat, rather than periphery of the unit (Fig. 30B). directly correlated winter maintenance habitat. Hamlin and Mackie (1989) reported with kinds and an average population of about 1,000 mule Varying quantity and quality of amounts of wnter summer habitats resulted in differences deer on the 275 km2 Missouri River Breaks forage or other in numbers of deer associated with winter study area during winters 1960-1987. attributes of winter ranges. Those numbers could not be Annually, however, populations varied ranges. directly correlated with kinds and amounts widely from a low 390 (1.4/km2) in spring of winter forage or other attributes of 1976 to highs of about 1,700 (6.2/km2) winter ranges. in autumn 1983 and 1987. During spring, Comparisons of mule deer numbers density averaged 3.0 (1.4 to 4.5) mule and density among populations in the deer/km2 . Both the mean and range of Bridger Mountains exemplify these densities in the timbered breaks fell within relationships. Total deer numbers and the lower end of the range of densities density were not necessarily related to recorded for populations in the Bridger size of the population-habitat unit (Pac Mountains. et al. 1991). For example, the large Densities recorded for mule deer number (~2,100) and high average density populations on open prairie-badlands (6.7/km2) of deer in the Brackett Creek habitat were even lower, ranging from 0.6 population reflected occurrence of large to 2.0/km2 on the Cherry Creek study area units of high quality reproductive habitat (Wood et al. 1989) and 0.9 to 2.0/km2 on directly adjacent to a large area of winter- the Dog Creek study area in northeastern maintenance habitat (Fig. 30A). Montana (Jackson 1990). Densities in

Po p u l a t i o n Ch a r a c t e r i s t i c s 77 Figure 30. Comparison of Reproductive A distribution of reproductive and Maintenance - summer maintenance habitats and unused areas in Maintenance - winter Brackett Creek (A) Unused and Battle Ridge (B) population habitat- units in the Bridger Mountains (after Pac et al. 1991).

B

� 024 km

both of these areas fell within the range Relatively high densities were of densities calculated for mule deer in sometimes observed for mule deer in hunting districts across southeastern environments influenced by agriculture Montana (0.1-3.3/km2, Youmans and and other human activity. Winter densities Swenson 1982) and non-timbered breaks ranging from 5 to 10 /km2 (ave =7/km2) habitat adjacent to Tiber Reservoir in and 3 to 14/km2 (ave. = 8/km2) have northcentral Montana (0.7-4.4/km2, Olson been reported for non-timbered breaks- 1986). agricultural habitat along the Coffee

78 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Creek-Arrow Creek and Sage Creek- practices. Stability is inherent in the Indian Creek drainages, respectively, in sense that availability of vegetative central Montana (Stivers, pers comm.). resources is high and predictable over Similarly, Fritzen (1995) found an average time within small, local areas. Densities 7.5 mule deer/km2 in a plains population varied from <15/km2 to >50/km2, in inhabiting a surface mine-reclamation- accord with amount of riparian cover suburban habitat complex in the vicinity and other characteristics of bottomlands of Colstrip in southeastern Montana. (Compton et al. 1988, Dusek et For white-tailed deer, total numbers al.1989). By comparison Hamlin (1979, and densities in riverine environments 1980) reported minimum densities varied directly with amount of riparian of 6-12 whitetails/km2 on unfarmed forest and shrubland cover along bottomlands along the Missouri River. the bottomlands (Compton et al. Because tree and shrub components were 1988). Occurrence of riparian habitat similar to the lower Yellowstone, density interspersed with agriculture and differences between the two areas might rangeland also strongly influenced white- be attributed to the lack of agriculture tailed deer distribution, habitat use, and and interspersion of crop and riparian apparently their abundance on upland components along the Missouri River. prairie-agricultural habitats (Dusek et al. Lowest average density of whitetails For white-tailed 1988). was recorded in prairie-agricultural deer, total numbers Highest whitetail densities were habitat on the Cherry Creek study area, and densities recorded along the lower Yellowstone where densities peaked at <0.6/km2 in riverine River. Riverine environments like overall and 5/km2 on local areas (Wood et evironments bottomlands of the Yellowstone River al. 1989). Estimates based on modeling varied directly today represent habitats of relatively high of harvests and age structures in the Swan with amount of complexity, diversity, and stability (Dusek Valley (Riley pers. comm.) suggested riparian forest et al. 1989). Complexity and diversity maximum average densities of about 5-6 and shrubland 2 are provided by interspersion of relatively white-tailed deer/km of yearlong habitat. cover along the small units of many different vegetation Within study areas, density bottomlands. cover types, land uses, and agricultural distribution of deer varied seasonally in

Po p u l a t i o n Ch a r a c t e r i s t i c s 79 response to changes in environmental changes in deer numbers in areas outside conditions influencing quantity and core habitats (Hamlin and Mackie 1989). quality of habitat available and the annual While density increased or decreased cycle of deer behavior. In mountain- across broad areas, densities in core areas foothill habitats, deer were widely changed little, if at all, because young distributed in summer and early autumn females did not use the same sites as and became increasingly aggregated established matriarchs during summer during late autumn, winter, and early and early autumn. Similar to the findings spring. In the Bridger Mountains, mule of Van Horne (1983), at high population deer usually were restricted to 20 percent levels, deer density in lower quality or less of the total area during winter and habitat could equal or exceed that in high average deer densities were 2- to 12-fold quality core areas (Hamlin and Mackie (ave. 4.7-fold) higher than during summer 1989). (Pac et al. 1991). Under extreme midwinter conditions, densities on winter A similar relationship appeared to habitat along the west slope reached prevail among white-tailed deer on the While density nearly 190 mule deer/km2 (Mackie et al. lower Yellowstone River. Again, at low increased or 1976). Seasonal increases in density population levels all females were able decreased across tended to be highest where winter to occupy optimal reproductive habitat. broad areas, maintenance habitats comprised smaller As deer numbers increased, young adult females increasingly occupied low quality densities in core proportions of the total yearlong habitat habitats such that densities in optimal areas changed used by deer. The greatest increase 2 habitats did not increase until high little... (11.7-fold, 1.4-16.7/km ) occurred on the Battle Ridge unit where winter habitat population levels were attained (Dusek et comprised only 8.5 percent of the total al. 1989). unit area. Spatial distributions of mule deer in timbered breaks and prairie-badlands Sex and Age Composition environments followed the clumped or aggregated pattern typical of animals occupying patchy environments (Elseth Within the context of our studies, and Baumgardner 1981). Areas of sex and age composition refers to the high density were either bordered by relative abundance of adult males and areas of declining density (Hamlin and females, young and adults in a population. Mackie 1989) or interspersed among Measurement may be in terms of either low density and vacant areas (Wood absolute or relative numbers (i.e., ratios et al. 1989). In the Missouri River or percentages). Breaks, mule deer occurred on about 83 The numbers or proportions of percent of the area yearlong, 64 percent deer in different sex and age classes during summer, 60 percent in autumn, varied within and among populations 65 percent in winter, and 53 percent and habitats. Some of the variation in spring (Hamlin and Mackie 1989). appeared to occur within a “normal For Cherry Creek, averages were 37 range,” for a particular environment percent overall, 20 percent in autumn, 19 percent in winter, and 15 percent in (Tables 1 and 2). Because of this, sex spring (Wood et al. 1989). However, and age composition may be indicative of because both areas experienced highly basic deer-habitat relationships; that is, variable environmental conditions during how well each habitat meets the specific all seasons, spatial distribution and deer requirements of each sex and age class of densities were in a constant state of flux deer. However, variation in composition relative to availability of space, food, and influenced by environmental fluctuation, cover. hunting, sampling, and factor interaction Increases and decreases in often confounded analyses of factors population size in the Missouri River determining sex/age composition within Breaks were marked by disproportionate and among population units.

80 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Table 1. Post-hunting sex and age ratios for mule deer populations on three Montana study areas. Areas/Years Population Males:100 Fawns:100 Females Females Bridger Mountains 1971-87 NW Slope 26 (8-40)a 51 (9-85) 1974-87 Bracket Creek 11 (5-15) 61 (33-91) Missouri River Breaks 1960-87 – 31 (13-49) 65 (29-116) Cherry Creek 1975-87 – 18 (7-36) 72 (48-111) a average (range)

Table 2. the vegetationally and agriculturally more Post-hunting sex and age ratios for white- diverse Elk Island (ave. = 53 males:100 tailed deer populations on two Montana females; range 26-73:100) as compared study areas. with the Intake subpopulation unit (ave. Areas/Years Males:100 Fawns:100 = 38:100, range 20-49:100). Females Females Long-term trends in relative Cherry Cr. 28 (17-47)a 74 (34-120) abundance of male and female mule 1975-87 deer in the Missouri River Breaks and Lower Yellowstone River 25 (18-37) 83 (58-112) Bridger Mountains show the relative 1980-87 influence of different hunting strategies a average (range) on sex composition. The two populations were hunted under varying regulations Sex Composition designed to both limit and enhance hunting pressure and harvest of antlerless High proportions Among adults ≥1 year, females deer at different times. In the Breaks of maintenance outnumbered males by up to 6:1 pre- (Fig. 31), adult male:100 female ratios habitat hunting season and 20:1 post-hunting declined generally from the mid-1970s interspersed with season. Some of these differences were through the mid-1980s. During the reproductive influenced by selective hunting and 1960s and early 1970s, regulations habitat may be variation in hunting pressure; others emphasized increased hunting pressure necessary for an apparently were related to characteristics and harvest of antlerless deer through area to support of the environment occupied by deer. 2-deer either-sex hunts, unlimited low- many adult males We found no evidence that adult sex cost nonresidents licenses, and special or exhibit a high ratios attain equality in free-ranging deer early and late season hunts. At that sex ratio. populations even under differentially time, as hunting pressure and harvests heavy harvest of antlerless animals. increased, the proportion of antlerless Females may outnumber males by 2:1 or deer killed also increased from less than 3:1 even in unhunted populations of mule one-third to one-half or more of the total deer (Martinka 1978) and white-tailed harvest (unpubl. MFWP harvest data). deer (Gavin et al. 1984, Kie and White Post-season sex ratios averaging 40 adult 1985). Certain environments may provide males:100 females were recorded during opportunity for greater local occurrence 1960-74 (Hamlin and Mackie 1989). of adult males relative to females while The sharp decline in mule deer others may limit the occurrence of adult numbers in the Breaks during 1972-75 males. High proportions of maintenance (Fig. 31) resulted in restriction of harvest habitat interspersed with reproductive to “bucks only” from 1976 through habitat may be necessary for an area to 1980. This shift in harvest strategies support many adult males or exhibit a high increased mortality of adult males relative sex ratio (Hamlin and Mackie 1989). On to adult females. As a result, as the the lower Yellowstone River, adult males population recovered, the adult female were more abundant during autumn on segment increased by 111 percent from

Po p u l a t i o n Ch a r a c t e r i s t i c s 81 Figure 31. Trend in numbers 800 Number of Adult Males Male:100 Adult Female Ratio 60 of adult males and Number of Adult Females Percent Adult Males females, male:100 700 female ratio, and 50 percent males in Male:100 Adult Female Ratio or the Missouri River 600

Breaks mule deer Percent Adult Males 40 population, early 500 winter 1960-61 through 1986-87. 400 30

300 20

200 Number of Adult Males or Females 10 100

0 0

1960-611961-621962-631963-641964-651965-661966-671967-681968-691969-701970-711971-721972-731973-741974-751975-761976-771977-781978-791979-801980-811981-821982-831983-841984-851985-861986-87

1976 through 1982 while numbers of to harvest antlerless deer. Thus, the adult males increased only 47 percent low and decreasing ratios after the mid (Hamlin and Mackie 1989). Although 1970s were more the result of increased either-sex hunting resumed in 1981, numbers of adult females and factors hunter harvests of females did not reach influencing that increase than a reduction rates equivalent to those of the 1960s. in absolute numbers of adult males in the Numbers of adult females peaked during population. 1983-84 and fluctuated at relatively high Also, increases in numbers of levels through the late 1980s, while buck adult females reflected environmental numbers decreased from 1983 through ...in the Breaks conditions favorable to relatively high 1986. Perhaps as a result of hunter population, recruitment and adult survival during the selection for mature bucks, the mean sex ratio was late 1970s and early 1980s (Hamlin and post-season adult sex ratio declined from Mackie 1989). For example, changes influenced as 40 males:100 females during 1960-74 to much or more by in livestock grazing, favorable weather less than 25:100 females during 1975- conditions, and improved range condition numbers of females 87. Bucks ≥ 2 1/2 years comprised an as by numbers of beginning in the late 1970s may have average of 31 percent of males classified increased the quantity and/or quality of males. in early winters 1960-80, compared with reproductive and summer maintenance 20 percent during 1981-87. habitat that sustained progressively Overall, in the Breaks population, greater numbers of females on the area. sex ratio was influenced as much or more Decreased hunter selection for and by numbers of females as by numbers of males. Numbers of females were high harvest of adult females during years and gradually increasing through much of of favorable environmental conditions the period from 1977 to 1987; numbers and high recruitment could have led to of males were relatively low and stable, increased numbers of young females ranging from 100 to 200 post-season in settling in habitats that provided most years. This resulted at least partially resources necessary for survival but because of reduced harvest of females not reproduction as described for deer associated with changes in harvest populations in the prairie-badlands regulations and reduced desire of hunters environment (Wood et al. 1989).

82 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a In the Bridger Mountains (Fig. more closely tracked their abundance in 32), adult male:female ratios declined the population. similarly to those in the Missouri River Numbers of adult males and Breaks following onset of bucks-only male:female ratios for the South 16-Mile hunting in the mid 1970s, then increased population increased considerably after with population growth during the early hunting was greatly restricted on private 1980s and declined to 1987. A special lands that comprise much of that area. regulation limiting harvest of older males Most of the increase seemingly resulted during the last 2 weeks of the hunting from increased survival of mature season was implemented in 1989 (Pac males (Fig. 33). In contrast, total adult and Ross 1993). Subsequent studies male:female and 4-point-male:female (Pac unpubl. data) indicate that numbers ratios for heavily hunted populations and proportions of bucks increased occupying public land showed little again from 1987 to the early 1990s improvement or declined through the before declining. However, increases in period (Pac unpubl. data). buck numbers through the early 1990s Data sets that track variation were related primarily to above-average and trends in sex ratios (numbers of recruitment rather than any long-term males:100 females) were limited for reduction in hunting mortality. mule deer in other study areas and for In contrast to the Missouri River white-tailed deer generally. Buck:doe Breaks, numbers of females in the ratios were affected by recruitment and Bridger Mountains were relatively stable harvest strategies in all areas. As hunting throughout the period of study, and sex pressure, numbers of antlerless permits, ratios appeared to be influenced more and harvest rates changed, buck:doe by trends in the number of adult males ratios also changed. in the population. Also, harvest of On the Cherry Creek study area, females has always been light, at least numbers of female mule deer increased since the mid 1970s. The light harvests about 6-fold from 1975-76 through of females apparently contributed to 1983-84 before declining. Post-season maintenance of stable female numbers while selective harvest of adult males male:female ratios varied, but were

Figure 32. Number of Adult Males Male:100 Adult Female Ratio Trends in numbers of adult males and Percent Adult Males Number of Adult Females females, male:100 350 40 female ratios, and percent males in a 300 35 mule deer population Male:100 Adult Female Ratio or on the northwest 30 slope of the Bridger

250 P

e Mountains, early r 25 ce winter 1973-74 n

200 t through 1996-97. A

20 du l t

150 M

15 a l es 100 10 Number of Adult Males or Females 50 5

0 0

1973-741974-751975-761976-771977-781978-791979-801980-811981-821982-831983-841984-851985-861986-871987-881988-891989-901990-911991-921992-931993-941994-951995-961996-97

Po p u l a t i o n Ch a r a c t e r i s t i c s 83 30 9 ratios and percentages of males occurred Number of Males:100 Adult Females 4-Point Males:100 Adult Females during periods when total population 8 25 4-Point Males:100 Adult Females size was high (Figs. 31 and 32), but the 7 converse was also true. However, sex 20 6 ratios and percent bucks could decrease

5 because of increased numbers of adult 15 females. Differences in percentages 4 of bucks also reflected the influence of 10 3 variation in numbers of young in the

2 population. 5 Use and interpretation of ratios 1 as indices of population status or for Number of Males:100 Adult Females 0 0 management prescription requires knowledge of actual numbers of adult 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 Figure 33. males and females in a population. Ratios Relative trends in post-hunting male:100 female and 4-point- do not reflect the total numbers of bucks male:100 female ratios for the South 16-Mile mule deer available to hunters, only the relative population, Bridger Mountains, 1986-96. proportions of bucks and does that hunters might observe in the field. Where numbers were available, ratios were relatively high in 1976 and 1977 as deer less important. However, ratios indicate numbers began increasing and declined to relative observability of bucks versus 10-20 males:100 females during 1978- antlerless deer, and if ratios are low, many 1985 (Wood et al. 1989). As a result, hunters may perceive few bucks even sex ratios varied inversely with numbers when numbers are high. of adult females in the population. For the smaller whitetail population, numbers of adult males and adult females, as well Age Composition as sex ratios tracked one another from Both percent fawns and fawn:100 1975-76 through 1983-84. Thereafter, adult ratios generally tracked fawn heavy selective hunting of females numbers and trends in populations during under multiple antlerless permits led most years, though some discrepancies to disproportionately higher mortality occurred. The two parameters reflected of females such that post-season variation in numbers of adults as well as male:female ratios increased while fawns. Over time, similar percentages numbers of both sexes declined. and ratios were associated with large On the lower Yellowstone River, differences in numbers of fawns and/ white-tailed deer numbers increased with or adults in a population (Figs. 34, 35, increased numbers of adult females from and 36). Often, especially in prairie 1980-81 to 1983-84 and declined slightly deer populations, fawn percentages were to 1985-86. Numbers of males remained highest and ratios peaked 1-4 years prior relatively stable throughout the study, to peaks in total numbers of fawns and while post-hunting sex ratios inversely total deer in a population. Maximum tracked adult female numbers. fawn recruitment to spring did not exceed Although sex ratios and/or the 40-45 percent of total mule deer and 45- percentage of males are commonly 50 percent of whitetail populations on any taken as measures of the abundance of of the areas studied, suggesting this may adult males, we found highly variable be a “biological cap” for these species relationships between numbers of adult in Montana ecosystems. Comparison of males and females, male:female ratios, trends in spring fawn:100 adult ratios and percent males in the populations we for mule deer among populations and studied. In the Missouri River Breaks habitats (Fig. 37) shows that each and Bridger Mountains relatively high sex population exhibited trends characteristic

84 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a 1000 Fawn:100 Adult Ratio 90 Number of Adults 900

Number of Fawns Percent Fawns 80 Fawn:100 Adult Ratio or Percent Fawns

800 70

700 60

600 50 500 40 400

30 300 Number of Adults or Fawns 20 200

100 10

0 0

196119621963196419651966196719681969197019711972197319741975197619771978197919801981198219831984198519861987

Figure 34. Trends in estimated total numbers of adults and fawns, fawn:100 adult ratios, and percent fawns in the mule deer population on the Missouri River Breaks study area, Use and spring 1961-1987. interpretation of ratios as indices of population status or for management prescription requires knowledge of actual numbers Fawn:100 Adult Ratio 700 Number of Adults 70 of adult males Percent Fawns Number of Fawns Fawn:100 Adult Ratio or Percent Fawns and females in a 600 60 population.

500 50

400 40

300 30

200 20 Number of Adults or Fawns

100 10

0 0

1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Figure 35. Trends in estimated total numbers of adults and fawns, fawn:100 adult ratio, and percent fawns in the mule deer population on the Northwest Slope, Bridger Mountains, spring 1972-1997.

Po p u l a t i o n Ch a r a c t e r i s t i c s 85 Number of Adults Fawn:100 Adult Ratio

800 Number of Fawns Percent Fawns 80 Fawn:100 Adult Ratio or Percent Fawns 700 70

600 60

500 50

400 40

300 30

Number of Adults or Fawns 200 20

100 10

0 0

1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 Figure 36. Trends in estimated total numbers of adults and fawns, fawn:100 adult ratios, and percent fawns in the mule deer population on the Cherry Creek study area, spring 1976-1987.

of the environment in which it occurred. Bridger Mountains 90 Generalizations across populations and habitats based on “biological potential” 80 Missouri River Breaks rarely applied. Because of this and the failure of ratios and percentages to 70 Cherry Creek consistently track numerical changes, 60 interpretation of population trends and management opportunity from 50 age composition data alone could be 40 misleading. At least some knowledge or estimate of numbers of adult females or 30 adults at the time is required. 20 Number of Fawns:100 Adults

10 Age Structure

0 Age structure is the number or proportion of deer in successive age 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 classes from young of year through the Figure 37. oldest surviving animals. It represents Comparison of trends in number of fawns:100 adults on Bridger a combination of annual production/ Mountain, Missouri River Breaks, and Cherry Creek study areas during spring 1976-1987. recruitment of young, emigration/ immigration of yearlings or young adults, and adult survival between age classes. Thus, age structure reflects the cumulative effect of all factors influencing past gains and losses to a population.

86 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Age structure is most useful for “boom” and “bust years” resulting in determining what has happened to a shifting age structures. There was no population in the past (Hamlin and Mackie “typical” age structure for the species 1989) and may be misleading in predicting across habitats, between populations, or population status and trend. Like other over time for any one population. sex and age compositional parameters, We identified four different types age structure data are best applied in of age structure that occurred during conjunction with some measurement or some years within the female and male There was no estimate of population size and trend. segments of mule deer populations “typical” age Generalizations about age structure (Fig. 38): structure for the and relative survival rates by sex and age • Type 1 - a pyramidal age structure, species across class based on age structural data alone heavily skewed to young age (i.e., construction of life tables) should be classes, considered characteristic habitats, between viewed with caution. Data for mule deer of populations experiencing rapid populations, or increase as a result of high recruitment across a spectrum of habitats and for over time for any of young and/or relatively low annual different populations within the Bridger one population. adult survival (i.e., high annual Mountains indicated that “stable” age turnover); structures probably cannot be expected for this species. Persistence of a stable • Type 2 - a relatively flat or low pyramid age structure requires constant natality age structure, resulting from periods of low and stable recruitment and high and mortality. On our study areas, natality longevity among adults (i.e., low annual and mortality occurred in periods of both turnover);

Type 1 Type 2

N N

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Age Class Age Class

Type 3 Type 4

N N

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Age Class Age Class Figure 38. Conceptualized examples of four types of age structures observed in Montana mule deer: Type 1 = pyramidal, Type 2 = flat or low pyramid, Type 3 = convex, Type 4 = concave or “U” shaped.

Po p u l a t i o n Ch a r a c t e r i s t i c s 87 • Type 3 - a convex age structure, (Fig. 41) through periods of population dominated by middle or “prime” age decline and recovery in the Bridger classes and associated with sharply Mountains and Missouri River Breaks reduced or declining populations further illustrate the mix and succession (Elseth and Baumgardner 1981) and; of age structures that occur over time. • Type 4 - a concave or “U”-shaped age Prior to decline, female age structure structure containing more individuals on the northwest slope of the Bridger in young and old age classes than in Mountains included representation middle age classes and associated with of young, middle-age, and older deer. population recovery following a decline Losses of three successive cohorts and (Pac et al. 1991). substantial numbers of old females Populations in specific habitats may through the decline resulted in a strongly have spent more time in either type 1 or convex, Type 3 age structure 3 years later. type 2 age structures, thereby implying Thereafter, high survival of residual prime- characteristic age structures by habitat. age adult females combined with low but However, both type 1 and 2 structures stable recruitment maintained the female occasionally occurred in any population segment for several years until increased in response to various environmental recruitment provided opportunity for stimuli. Type 3 age structure occurred gradual population increase during only temporarily in delayed association the early 1980s. Cohorts were never with severe environmental events and sufficiently large, nor adult survival so low conditions resulting from low recruitment as to shift female age structure toward the in youngest cohorts for 1 or more pyramidal Type 1. years combined with high mortality in The age structure of male mule deer old age classes. High natural survival in the Bridgers exhibited a somewhat rate among residual, prime-aged deer different trend. Numbers of males were sometimes maintained this age structure reduced through the 1974-75 winter to some degree for several years until resulting in an age structure dominated increased recruitment began to skew the by prime-age bucks during 1975-76. age distribution toward the Type 4 age Harvests and low recruitment further structure, which was also temporary. reduced buck numbers to a few individuals At any given time, individual of various ages by winter-spring 1978. populations could exhibit very The increased cohort size in 1980 was different age structures resulting from sufficient to shift male age structure to the environmentally-induced variation and pyramidal form by 1981-82 (Fig. 41). hunting. For example, during two years Age structure of mule deer in the when comparable data were obtained, Missouri River Breaks, measured and age structures of female segments estimated through population modeling of four different populations in the for spring 1968-86 (Hamlin and Bridger Mountains varied from strongly Mackie 1989), was highly variable and pyramidal to somewhat “U”-shaped (Fig. shifted through two different periods 39). Male age structure was essentially of population increase, 1968-71 and pyramidal in all populations; apparently 1979-83 and two periods of decline, because hunters heavily selected for 1972-74 and 1984-86. Age structures older, large antlered males. Because of changed similarly through both periods of lower total numbers of males, even low population increase and decline, though annual recruitment provided sufficient the mid 1970s decline was more severe numbers of male fawns and yearlings to and resulted in a lower population. maintain the pyramidal structure during most years under all but the most severe All basic age structural types were environmental conditions. represented through the major population Age structural dynamics and trends fluctuation in the 1970s. For example, for mule deer females (Fig. 40) and males by 1970-71, following 4-5 years of

88 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a 30 30 Northwest slope - 1979 Northwest slope - 1982 27 27 N = 53 N = 57 24 24 21 21 18 18 15 15 Percent 12 Percent 12 9 9 6 6 3 3 0 0 12 3 456 718910 1121314 5 12345678910 112131415 Age Class Age Class

30 Brackett Creek - 1979 30 Livingston - 1982 At any given 27 N = 69 27 N = 40 time, individual 24 24 populations 21 21 18 18 could exhibit 15 15 very diffrent Percent Percent 12 12 age structures 9 9 resulting from 6 6 environmentally- 3 3 induced variation 0 0 123456789101 12 13 14 15 12345678910 112131415 and hunting. Age Class Age Class

30 South 16-Mile - 1982 27 24 N = 73 21 18 15 Percent 12 9 6 3 0 1 23456789101 12 13 14 15 Age Class Figure 39. Comparative age structures for adult females in four mule deer population-habitat units in the Bridger Mountains illustrating the mix of different age structures possible in adjacent populations during a given year (after Pac et al. 1991). population increase, both female and male with high survival of adults to older age segments exhibited distinct pyramidal age classes resulted in small populations structures. High winter mortality of fawns with relatively flat, Type 2 age structures and older adults during 1971-72 abruptly through spring 1978. Increased annual shifted age distributions to Type 3; few cohorts fostered rapid population growth deer of either sex remained that were less and shifted age distributions back toward than 2 or older than 6 years of age. Small pyramidal (Type 1), for the smaller male annual cohorts and light to moderate adult segment during 1979-83 and for females mortality further reduced deer numbers by during 1980-83. 1974-75, when low recruitment combined

Po p u l a t i o n Ch a r a c t e r i s t i c s 89 Northwest Slope Bridger MountainsMissouri River Breaks 32 A Winter A Spring 28 200 1973-74 1971 24 150 20 16 Number Number 100 12

8 50 4 0 0 1 2345678910 11 12 13 14 15 1 2 3 4 5 6 7 8 ≥9 Age Class Age Class

32 B Winter B Spring 28 1976-77 200 1973 24 20 150 16 100 Number 12 Number 8 50 4 0 0 12 34567891011112 31415 1 2 3 4 5 6 7 8 ≥9 Age Class Age Class

32 C Winter C Spring 28 1981-82 200 1983 24 20 150 16 100 12 Number Number 8 50 4 0 0 1 2345678910 11 12 13 14 15 1234 5678 ≥9 Age Class Age Class

Figure 40. Age structural dynamics of adult female mule deer in the Bridger Mountains and Missouri River Breaks. A = prior to severe winter, B = one full year following severe winter, C = following several years of population growth (after Pac et al. 1991 and Hamlin and Mackie 1989).

90 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Northwest Slope-Bridger Mountains Missouri River Breaks

24 A Winter A Spring 1974-75 200 20 1971

16 150

12

Number 100 Number 8 50 4

0 0 1 2345678≥9 1234 5678 ≥9 Age Class Age Class 24 B Winter B Spring 1975-76 200 20 1972

16 150

12 Number

Number 100 8 50 4

0 0 1 2345678≥9 1234 5678 ≥9 Age Class Age Class 24 C Winter C Spring 1977-78 200 1976 20

16 150

12 Number

Number 100 8 50 4

0 ≥ 0 12345678 9 1234 5678 ≥9 Age Class Age Class 24 D Winter D Spring 1981-82 200 20 1983

16 150 12 Number Number 100 8

4 50

0 0 1 2345678≥9 1234 5678 ≥9 Age Class Age Class

Figure 41. Age structural dynamics of adult male mule deer in the Bridger Mountains and Missouri River Breaks: A = during or prior to severe winter, B = year following severe winter, C = at population low, and D = following several years of population growth or recovery (after Pac et al. 1991 and Hamlin and Mackie 1989).

Po p u l a t i o n Ch a r a c t e r i s t i c s 91 Recurrent low recruitment and small River were pyramidal and similar (Fig. cohorts in the Breaks during 1984 and 42). Younger age classes were better 1985 shifted age distributions for both represented in the productive riparian- sexes sharply toward the type 3 again agricultural habitat where an average during 1984-86. Adult mortality remained of 75 percent of autumn populations low and the female population did not during 1980-85 was 2 years-old or less decline, but more or less continued to (Dusek et al. 1989) compared with 60 increase following the trend established in percent in the Swan (MFWP unpubl 1977-78. With continued hunter harvests, data). Individuals 8 years and younger numbers of adult males declined as in the comprised more than 90 percent of all mid 1970s when a similar age structure does on both areas, while few males lived prevailed. longer than 4 years. Age structures for white-tailed deer Age structures for the Swan inhabiting conifer forest habitat in the and lower Yellowstone River were for Swan Valley and riparian-agricultural populations at or near peaks following habitat along the lower Yellowstone several years of increase. Comparative

Figure 42. 70 Comparison of representative adult Adult Males male and adult 60 female age structures Swan Valley for white-tailed 50 deer on the Swan Lower Yellowstone River Valley and Lower 40 Yellowstone River study areas (MFWP unpub., Dusek and 30 Mackie 1988). 20

10

0 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Age Class 30 Adult Females 25 Swan Valley

20 Lower Yellowstone River

15

10

5

0 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Age Class

92 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a age structures for the lower Yellowstone In populations of both species, River during 1980-85 (Fig. 43) (Dusek adult females greatly outnumber adult et al. 1989) and age structure of white- males in winter and spring. Because of tailed deer examined at hunter checking this, the size of each annual cohort has stations in the Swan Valley during 1974- a proportionately greater effect on total 1989 (Coates 1996) indicate that age numbers and age distribution of males structures were more stable than those than females. Differential heavy harvests recorded for mule deer. This probably of males also reduces the influence of occurred because of the more stable strong year classes of young sooner in environments occupied by these white- the male segment. Combined with hunter tailed deer. Large fluctuations in age selection for the largest available antlered composition of whitetails on the Cherry males, this results in male age structures Creek study area reflected boom and bust retaining a more pyramidal shape across natality and mortality similar to mule deer a wider range of habitats and population in plains environments. size than for females.

Figure 43. ≥ 8 1980 Female ≥ 8 1981 Age structure of 7 Male 7 female and male white-tailed deer on 6 6 the lower Yellowstone 5 5 River during autumn 4 4 1980-85 (Dusek et al. 3 3

Age Class 1989). Age Class 2 2 1 1 0 0 25 20 15 10 50510152025 25 20 15 10 50510152025 Percent of Population Percent of Population

≥ 8 1982 ≥ 8 1983 7 7 6 6 5 5 4 4 3 3 Age Class Age Class 2 2 1 1 0 0 25 20 15 10 50510152025 25 20 15 10 50510152025 Percent of Population Percent of Population

≥ 8 1984 ≥ 8 1985 7 7 6 6 5 5 4 4 3 3 Age Class Age Class 2 2 1 1 0 0 25 20 15 10 50510152025 25 20 15 10 50510152025 Percent of Population Percent of Population

Po p u l a t i o n Ch a r a c t e r i s t i c s 93 Population Dynamics

Understanding population Reproduction and dynamics requires knowledge of factors Recruitment influencing fawn recruitment and adult mortality and how they interact to change deer numbers over time. We Reproductive rate (i.e. the number ...factors affecting use recruitment to mean reproductive of fawns produced per doe at birth) varies among populations of both species mortality potential less all fawn mortality through the year following birth. (Cheatum and Severinghaus 1950, rates of fawns The influence of fawn mortality Robinette and Gashwiler 1950, Robinette generally have differs from adult mortality; factors et al. 1955, 1977). It also varies over greater impact on affecting fawn mortality determine annual time within populations. Although recruitment than recruitment, whereas those affecting adult reproductive potential can vary, post- those affecting mortality determine population losses. partum mortality is usually more variable. initial production. Also, different factors may operate at Thus, factors affecting mortality rates of different intensities on fawn and adult fawns generally have greater impact on mortality. Immigration and emigration are recruitment than those affecting initial special cases of recruitment and mortality production. and are difficult to quantify. Fawns per productive female for mule deer in the Missouri River Breaks varied among years from 1.25 to 1.76 fawns:doe, with a mean of 1.58 fawns:doe, during 1975-1986 (Hamlin and Mackie 1989). Studies in other eastern Montana environments have indicated potential reproductive rates ranging from 1.52 (Jackson 1990) to 1.67/1.68 (Dusek 1971, Fritzen 1995) and 1.90 fetuses per mule deer female (Eustace 1971). Data from collections of reproductive tracts in the Bridger Mountains indicated lower potential reproduction for mule deer in that mountain-foothill environment (1.27 fetuses:female > 1 year of age) than for other Montana habitats (Pac et al. 1991). Like other studies that have shown wide spatial and temporal variation in productivity of both mule deer and white-tailed deer (Beasom and Wiggers 1984), we found no consistent differences

94 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a in reproductive potential between the of fawn rearing to 6 months paralleled two species in Montana. Collections initial production; 4- and 5-year old of white-tailed deer reproductive tracts females were most successful in in the Swan Valley (Mundinger 1981) recruiting fawns to 12 months. and along the lower Yellowstone River Table 3. (Dusek et al. 1989) revealed fetal rates ...spatial variation Variation in potential productivity and averaging 1.50 and 1.68 fetuses:female, in reproductive respectively. Smaller samples from other fawn survival to autumn between two subpopulations of white-tailed deer on potential that study areas suggested even higher rates, the lower Yellowstone River, 1980-85 occurred appeared up to 2.0 fetuses:female on Missouri (after Dusek et al. 1989). related more to River bottomlands (Allen 1965) and 1.75 environment and fetuses:female in the Salish Mountains Age (yrs) Above Intake Below Intake population-habitat (Morgan 1993). Within this framework, n % or ratio n % or ratio spatial variation in reproductive potential relationships than Pregnancy rates to species. that occurred appeared related more Yearling 12 83% 23 91% to environment and population-habitat ≥ 2 45 91% 52 98% relationships than to species. Thus, Fetal rates whitetail females in the conifer forested Yearling 7 100:100 11 155:100 Swan Valley exhibited lower reproductive ≥ 2 16 156:100 26 200:100 potential than both their counterparts in Fawn-rearing success riparian-agricultural habitat on the lower 2 14 43% 27 78% Yellowstone River (Fig. 44) and mule 3 16 44% 25 92% deer in some eastern Montana habitats. 4-7 39 79% 51 90% Subpopulations associated with different ≥ 8 9 22% 12 92% habitat complexes along the Yellowstone Survival to autumn (fawns:females) River also exhibited different reproductive 2 14 50:100 27 104:100 potentials (Table 3). 3 16 63:100 25 144:100 4-7 39 108:100 51 149:100 Both in-utero and observed age- ≥ 8 9 22:100 12 133:100 specific reproduction among marked Percent multiple births deer indicate significant difference in 2 6 17% 21 33% productivity of females by age class. 3 7 43% 23 56% Fawns of both species rarely conceive 4-7 31 35% 46 67% and do not contribute to reproduction ≥ 8 11 45% in Montana. This may reflect inability of fawns to achieve the size and physiological state necessary for sexual maturity or the combined effects of photoperiod and early winter weather in northern latitudes (Dusek et al.

1989). Among adult females, potential 2.5 productivity typically was lowest and most Figure 44. Age specific in-utero variable in 2-year-old females, increased 2.0 fetal rates for white- progressively to a maximum among 3-6- tailed deer on the year olds, and then declined somewhat 1.5 Lower Yellowstone with age. Age-specific reproduction River and Swan varied like other parameters, among 1.0 Valley study areas. populations and habitats (Figs. 44-46, Lower Yellowstone River Table 3). 0.5 Swan Valley

Initial fawn production among Ave. No. of Fetuses/Female mule deer in the Missouri River Breaks 0.0 increased steadily with age to a peak 0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 among 6-year-old females, declined Age Class of Females (years) sharply at age 7, then stabilized at a level similar to 3-year olds (Fig. 45). Rates

Po p u l a t i o n Dy n a m i c s 95 2.0

1.8

1.6 Fawns : Pregnant Female

1.4 Initial Production

1.2 Survival to 6 Mo. 1.0 Survival to 12 Mo. 0.8

Number of Fawns Per Female Per Number of Fawns 0.6

0.4

0.2

0.0

12345678&9 ≥10 Female age at Parturition Figure 45. Age-specific production and recruitment of fawns by female mule deer in the Missouri River Breaks, Montana, 1976-1984 (after Hamlin and Mackie 1989, 1991).

Initial Production

1.8 Survival to Summer-Autumn 1.6

1.4 Survival to 6 Mo. 1.2 Survival to 12 Mo. 1.0 0.8

0.6 0.4

Number of Fawns per Female 0.2 0.0 12345678910 11≥ 12 Female Age at Parturition Figure 46. Age specific production and recruitment of fawns by female mule deer on the west slope of the Bridger Mountains (after Pac et al. 1991).

In west slope populations in the age 12 and older continued to rear fawns Bridger Mountains, survival to summer- at a higher rate than 2- and 3-year olds autumn increased progressively to a peak and a similar rate to 4-year olds. Survival among 5-year-old females and then declined of fawns to winter declined in all but the in a fluctuating pattern (Fig. 46). Females youngest age classes of does; recruitment

96 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a in spring declined even further and badlands habitats. The difference became similar for females of all age reflected variation in initial production classes 3 through 12 and older. and winter mortality of fawns among Survival of white-tailed deer fawns environments. Mean recruitment for to autumn on the lower Yellowstone River white-tailed deer was similar among areas increased with female age to 4 years, (Table 4) but was most variable on prairie- declined slightly to age 5, and increased agricultural habitat along Cherry Creek. again to a peak among 6-7-year olds before declining among females 8 years Fawn Mortality and older to about the same level as 2-year olds (Fig. 47). Age specific rates Average annual fawn losses up of fawn survival to autumn also differed to three-fourths or more of potential between subpopulations (Dusek et al. reproduction were common in populations 1989). Although all females reared fewer we studied. Mortality of fawns was fawns on river bottoms above Intake as high during early summer, dropped compared to the area below, fawn survival progressively through late summer and was especially low among youngest and autumn to early winter, and then increased oldest age classes above Intake (Table 3). to mid-late winter (Fig. 48). About two- Survival of whitetail fawns to autumn thirds of the total fawn mortality in mule and winter in the Swan Valley (Fig. 47) deer populations occurred in summer showed the same general increase with and one-fourth during winter, but specific age as observed for deer elsewhere. seasonal mortality patterns and rates Some variation in annual varied widely among years. Two-thirds or reproduction appeared to occur among more of annual mortality in fawn white- Recruitment rates mule deer females ≥8 years of age in tailed deer also occurred in summer; for mule deer were winter mortality of fawns was light and the Missouri River Breaks (Hamlin and lower and more averaged about 10 percent and 3 percent Mackie 1989) and among young and old variable than for on the lower Yellowstone River and Cherry females in the Bridger Mountains (Pac et white-tailed deer. al. 1991). Irregular reproduction may Creek areas, respectively. also have been characteristic of youngest Annual fawn mortality rates for and oldest reproducing female white- individual populations ranged from a low tailed deer in the Swan Valley (Mundinger average 32 (10-48) percent for white- 1981). Reproductive patterns among tailed deer on the lower Yellowstone marked deer indicated that females in the Missouri River Breaks and other variable, unpredictable environments 1.6 may experience physiological exhaustion leading to both declining productivity 1.4 and increased mortality in their fifth 1.2 reproductive year, especially after periods 1 when reproductive success has been high 0.8 for several years (Hamlin and Mackie 1991). 0.6 LYR Fawn recruitment rates varied across 0.4 the spectrum of environments we studied 0.2 SV (Table 4). Recruitment rates for mule Number of Fawns per Female deer were lower and more variable than 0 for white-tailed deer. Based on both 1234567≥ 8 observed fawn:100 adult and modeled Female Age Class fawn:100 female ratios for mule deer, Figure 47. recruitment was lowest in the Bridger White-tailed deer fawn survival by Mountains, followed by the Missouri female age class to autumn on the lower Yellowstone River and early winter in the River Breaks, and Cherry Creek prairie- Swan Valley.

Po p u l a t i o n Dy n a m i c s 97 Table 4. Mule deer and white-tailed deer fawn recruitment in Montana. Data are observed fawn:adult and modeled fawn:female ratios in spring (ave. ±1 SD).

Environment Observed Fawn:100 Adult Modeled Fawn:100 Female Study area Ratio Ratioa MULE DEER Mountain-Foothill 29 ± 13 33 ± 13 Bridger Mountains range 6-63 range 4-53 (Northwest Slope) CVb = 0.45 CV = 0.39 N = 25 years N = 14 years Timbered Breaks 40 ± 23 51 ± 27 Missouri River Breaks range 5-82 range 6-103 CV = 0.58 CV = 0.53 N = 21 years N = 28 years Prairie-Badlands 48 ± 21 56 ± 24 Cherry Creek range 11-76 range 13-90 CV = 0.44 CV = 0.43 N = 12 years N = 12 years White-Tailed DEER Northwest Montane Forest 54 ± 13 67 ± 18 Salish Mountains range 26-71 range 30-96 CV = 0.23 CV = 0.27 N = 11 cases, 7 years N = 11 cases, 7 years Riverbottom Agricultural 58 ± 18 75 ± 20 Lower Yellowstone range 37-81 range 52-101 CV = 0.31 CV = 0.27 N = 6 years N = 6 years Prairie-Agricultural 53 ± 27 66 ± 30 Cherry Creek range 17-92 range 24-110 CV = 0.51 CV = 0.45 N = 11 years N = 11 years

a Assumes buck:doe ratio same as early winter. b CV = coefficient of variation (CV = s/ x)

0.25

awns 0.20

0.15 onth For F For onth

0.10 Instantaneous Mortality Instantaneous

Rate Per M Per Rate 0.05

0.00 Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Month Figure 48. Instantaneous monthly mortality rate for radio-collared mule deer fawns in the Missouri River Breaks, 1976-1986 (after Hamlin and Mackie 1989).

98 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a during 1980-86 to a high average 82 (35- to 73 (55-97) percent on the northwest 91) percent for Cherry Creek whitetails slope of the Bridger Mountains during between 1975 and 1987 (Dusek et al. 1973-87. 1989, Wood et al. 1989). An estimated Total annual fawn mortality rates 59 percent of potential white-tailed deer varied among the three mule deer study fawns was lost annually in the Swan Valley areas (Fig. 49). Summer-autumn fawn during 1976-79 (Mundinger 1981). mortality rates were similar in level Although some extremes in average and trend among populations (Fig. annual fawn mortality rates were 50); therefore variation in annual rates observed in white-tailed deer, mule deer among areas resulted primarily from populations appeared to experience differences in overwinter mortality (Fig. higher average and overall greater annual 51). Average winter-spring mortality variation in fawn mortality. Mule deer rates were lower than rates for summer- fawn mortality ranged from a low average autumn and varied from 27 percent in 60 (34-95) percent on Cherry Creek the prairie-badlands to 33 percent in during 1976-86 to 65 (17-96) percent in timbered breaks and 35 percent in the the Missouri River Breaks during 1960-87 mountain-foothill environments. Years of

100 Figure 49. 90 Comparative trends in total annual 80 mortality of mule 70 deer fawns in mountain-foothill, 60 timbered breaks, and 50 prairie-badlands environments, 1973- 40 Mountain-Foothill 74 through 1986-87. 30

Percent Fawn Mortality Timbered Breaks 20 Prairie-Badlands 10

0

1973-741974-751975-761976-771977-781978-791979-801980-811981-821982-831983-841984-851985-861986-87 Biological Years

90 Figure 50. 80 Comparative trends in mortality of mule 70 deer fawns from June 1 through 60 November 30 in mountain-foothill, 50 timbered breaks, and 40 prairie-badlands environments, 1975- 30 76 through 1986-87. Mountain-Foothill Percent Fawn Mortality 20 Timbered Breaks 10 Prairie-Badlands 0

1975-76 1976-77 1977-78 1978-79 1979-80 1980-81 1981-82 1982-83 1983-84 1984-85 1985-86 1986-87 Biological Years

Po p u l a t i o n Dy n a m i c s 99 Both 1978 and 1979 winters were 90 Mountain-Foothill severe across the entire region, yet 80 Timbered Breaks fawn mortality during 1979 was light to moderate in the prairie-badlands and 70 Prairie-Badlands timbered breaks populations but high 60 in the mountain-foothill environment. Conversely, under moderate winter 50 conditions following several years of 40 drought, fawns experienced very high winter mortality on the plains during 30 1984-85, but only average mortality in

Percent Fawn Mortality 20 the mountains. White-tailed deer fawns also experienced high mortality on Cherry 10 Creek and relatively high mortality on 0 the lower Yellowstone River during those drought years when adult female densities 1975-76 1976-77 1977-78 1978-79 1979-80 1980-81 1981-82 1982-83 1983-84 1984-85 1985-86 1986-87 also were high (Dusek et al. 1989). Biological Years Moderate to high rates of overwinter fawn mortality were more Figure 51. common among mule deer in mountain Comparative trends in mortality of mule deer fawns from environments. Populations in the December 1 through May 31 in mountain-foothill, timbered Missouri River Breaks and other eastern breaks, and prairie-badlands environments, 1975-76 through Montana prairie-badlands habitats 1986-87. occasionally experienced episodes of

90 extremely high mortality interspersed 80 with several years of low mortality. This 70 is clearly evident in trends in overwinter 60 mortality of mule deer fawns in the 50 Missouri River Breaks (Fig. 52). Over 27 40 years, beginning in 1960-61, there were 30 20 5 years in which overwinter mortality Percent Mortality 10 exceeded 77 percent. These included two 0 “back-to-back” years of high mortality in the mid-1980s. More typically, however, 1960-611963-641966-671969-701972-731975-761978-791981-821984-851986-87 years of high mortality were interspersed Figure 52. by 3-7 years of low to average mortality. Modeled trend in overwinter mortality Winters of intermediate fawn mortality rate for mule deer fawns in the Missouri were rare, suggesting that most often River Breaks, 1960-61 through 1986-87. conditions predisposed entire cohorts either to survival or death. Subsequent population surveys, during 1988-1997 high winter mortality in mountain foothill (Stivers unpubl.) indicate overwinter environments sometimes coincided with fawn mortality exceeding 80 percent years of comparatively low fawn losses also occurred during 1988-89 and 1994- in prairie environments and vice versa. 95, continuing the long-term pattern of Apparently, summer-autumn mortality severe fawn losses at 5-8 year intervals. was influenced more by region-wide environmental conditions than was winter Factors Affecting Fawn Mortality mortality. Winter mortality appeared to be influenced by complex interactions In our studies, factors influencing between specific populations and their fawn mortality varied widely such that the local environments. concept of a consistent “limiting factor” rarely applied. Differences between reproductive potential and recruitment

100 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a typically accrued from several sources and involved different patterns of fawn mortality. In the Missouri River Breaks, predation was the single most important proximal mortality factor for fawns (Hamlin and Mackie 1989). During the 12-year period, 1975-87, predation by coyotes accounted for 88 percent of all deaths of radio-marked fawns and 85 percent of all known losses of fawns during summer and autumn. Coyote predation was also the proximate cause of death of almost all (95 percent) fawn mortality during winters 1976-86, though other factors were involved and predisposed some individual fawns to forage supply, fawn mortality rate is predation. also expected to increase. We did not Despite the high incidence of find evidence to support this concept, predation, fawn mortality rates in especially in terms of a compensatory the Missouri River Breaks were not relationship between winter population significantly correlated with coyote size or density and overwinter mortality ...fawn recruitment density. Other factors, including and spring recruitment rates. Among was not related population levels of alternative prey, mule deer, recruitment rate was to relative deer yearlong habitat and forage conditions, significantly related to the number of density or the and winter severity, influenced predation adults in the population during spring amount of forage rates. This indicated that coyotes and only on the Cherry Creek area. However, per capita in the predation, though important, were not it was also related to drought. For white- population. Rather tailed deer, the number of mature females always the overriding or ultimate factors the relationship in the lower Yellowstone River population affecting fawn survival. Rather, the appeared to be during spring also was significantly importance of predation was tied to other with forage quality biological and environmental factors that correlated with number of fawns recruited to the following spring. The limitation or the length simultaneously influenced mortality. of time green, Productivity and fawn survival was behavioral and involved resource partitioning during summer rather than succulent forage to early winter in the Missouri River was available. Breaks and elsewhere also were amount of forage available per capita closely correlated with summer forage (Dusek et al. 1989). There were no production and conditions during most significant, positive linear relationships years (Hamlin and Mackie 1989, Wood et between numbers of mule deer and fawn al. 1989). Forage quality, as determined mortality rate during summer-autumn or by the succulence of vegetation, and the winter-spring on any area, nor was such a timing and length of the period when relationship evident for white-tailed deer green, succulent vegetation was available, on Cherry Creek. In the Missouri River appeared to be most important. Forage Breaks where forage production was production and quality in the Missouri measured over 11 years, fawn recruitment River Breaks were positively related to was not related to relative deer density precipitation prior to the growing season or the amount of forage per capita in the and negatively to temperature during the population (Hamlin and Mackie 1989). early growing season (Hamlin and Mackie Rather the relationship appeared to be 1989). with forage quality or the length of time The relationship between forage green, succulent forage was available. supply and fawn mortality in deer has Our findings concerning density been assumed to be density dependent; and fawn survival do not imply that as deer numbers increase relative to deer density is not involved or cannot

Po p u l a t i o n Dy n a m i c s 101 Bridgers (Pac et al. 1991). This variation indicated that winter mortality involved factors and interactions beyond simple winter severity. Prior forage conditions, especially in the Missouri River Breaks, helped explain some of the anomalies in the relationship between winter severity and fawn mortality. In the Breaks, drought that resulted in below average forage conditions and early desiccation of vegetation during one or more years prior to a moderate or severe winter appeared to cause significant mortality; above-average forage conditions greatly tempered overwinter losses even in severe winters. Poor summer forage conditions prior to winter accentuated mortality through the winters of l964-65 and 1971- 72 and apparently was the major factor in be a factor of some importance in fawn the heavy mortality that occurred through mortality. However, we documented the winters of average severity in 1983-84 no consistent density dependent and 1984-85. Conversely, above average relationships within limits of the variation forage conditions preceding very severe in deer numbers and fawn mortality/ winters in 1968-69, 1977-78, and 1978- recruitment rates observed. The only 79 put deer in very good condition and direct correlation we observed between greatly reduced mortality through those deer numbers and mortality appeared to winters. involve social rather than forage factors. Other studies have identified or The effects of winter severity on implicated other factors that interact fawn mortality appeared to be related to a with predation, forage production, complex of factors that influenced animal population density, and weather to condition prior to, during, and following influence overwinter mortality of fawns. winter. In the Missouri River Breaks and For example, birth dates and weights of Bridger Mountains, winter severity and fawns can vary considerably from year fawn mortality during the same winter to year. Later birth dates and/or reduced were positively related, but variation birth weight may, in turn, result in smaller was high. Also, there were numerous fawns in poor condition or with limited exceptions such that predictability was body reserves in autumn. However, limited (Hamlin and Mackie 1989, Pac et these variables that may predispose al. 1991). For example, fawn mortality fawns to winter mortality are effects of during two of the most severe winters environmental fluctuation rather than on record in the Breaks, 1977-78 and direct determinants of mortality rates. 1978-79, was less than half that recorded Hunting mortality of fawns varied in other severe winters (Fig. 52). widely with regulations governing timing Conversely, winter severity was below the and take of antlerless deer, hunter mean in 1983-84 and only slightly above selectivity, habitat security, and hunting average during 1984-85, when fawn pressure. Hunters usually selected mortality rates were among the highest against fawns during antlerless seasons. recorded. Similar variation between Harvest rate of fawns averaged 59 winter severity and mortality occurred in percent (43-78 percent) of the harvest the Bridger Mountains. There also were rate of adult female mule deer in the differences in winter severity-mortality Missouri River Breaks (Hamlin and relationships between populations in the Mackie 1989) and 60 percent (27-83

102 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a percent) of the harvest rate of adult high and some consider mature deer female white-tailed deer on the lower relatively immune to all but the most Yellowstone River (Dusek et al. 1989). extreme natural conditions, especially For mule deer, an average 12 percent where large predators are not a threat. (1-35 percent) of the estimated number of Yet the patterns and rates of adult deaths fawns on the Missouri River Breaks study (or survival) can be equally as critical as area disappeared during autumn, about recruitment to population characteristics, one-third (4 percent) of this mortality was dynamics, and trends over time. Adult associated with hunting. On the Cherry females and males have energy demands Creek area during 1982-86, a slightly that are unique to their roles in the higher 20 percent (0-45 percent) of the reproductive effort and result in different estimated number of mule deer fawns survival rates. Understanding these present in early autumn was lost through sex-related differences in mortality rates the hunting season. Average autumn is essential in management to sustain mortality due to hunting was 8 percent. quality populations and diverse hunting Hunters removed an average 12 percent opportunities. Harvest management (3-22 percent ) of all white-tailed deer often assumes that hunting or harvest- fawns on the lower Yellowstone River related mortality replaces other forms of during autumn, 1980-85. Fawn harvest adult mortality, but both harvest and other rates increased with hunting pressure, mortality events and rates may operate especially in association with availability and vary independently. of multiple antlerless permits. Fawn deaths due to disease, Adult Female Mortality abandonment, accidents, and traffic always occur and contribute to mortality Natural mortality rates were low and patterns and rates. Because of low generally similar across environments incidence, their individual and collective (Tables 5 and 6). Total annual mortality contributions to fawn mortality have in the Bridger Mountains for adult rarely been monitored. female mule deer was 12 percent during 1973-86 (Pac et al. 1991). During summer-autumn, adult female mortality Adult Mortality from all causes in this mountain-foothill Harvest environment averaged only about 5 management often Once established as adults in a percent (range = 0-11 percent) with assumes that population, probability of survival is natural and hunting mortality each hunting or harvest- related mortality Table 5. replaces other Average and range in total, hunting, and natural mortality ratesa for adult (≥1 year) forms of adult female mule deer in three Montana environments. mortality, but both harvest and other Environment total Annual hunting Natural (Non-hunting) mortality events Mortality Mortality Mortality and rates may Mountain-Foothillb 0.121 (0.083)c 0.048 (0.031) 0.073 (0.065) operate and vary N = 14 years range 0.000-0.286 range 0.000-0.111 range 0.000-0.236 independently. Timbered Breaksd 0.172 (0.090) 0.110 (0.075) 0.062 (0.047) N = 26 years range 0.022-0.430 range 0.000-0.298 range 0.014-0.248 Prairie-Badlandse 0.254 (0.166) 0.210 (0.153) 0.050 (0.045) N = 5 years range 0.003-0.420 range 0.001-0.370 range 0.000-0.120

a Calculated from arithmetric models that reconciled successive population estimates with data on population composition, recruitment, and marked deer mortality rate. b Pac et al. (1991). c Average (std. dev.) d Hamlin and Mackie (1989). e Wood et al. (1989).

Po p u l a t i o n Dy n a m i c s 103 Table 6. Annual survival rates and cause-specific mortality rates for yearling and older female white-tailed deer under different harvest regimes in three Montana environments.

Mortality Survival hunting Natural(Non-hunting) Age Rate a 95% C.I. Rate a Rate a Northwest Montane Forest b 1989-1995 (low harvest) ≥ 1 yr. 0.843 0.809-0.878 0.058 0.099

Plains Riverbottom c 1980-1984 (low to moderate harvest) 1 yr. 0.830 0.690-0.990 0.050 0.120 ≥2 yrs. 0.680 0.590-0.790 0.200 0.110 1984-1988 (high harvest) 1 yr. 0.430 0.230-0.800 0.330 0.240 ≥2 yrs. 0.450 0.360-0.570 0.390 0.160 Prairie-agricultural c 1984 (moderate harvest) ≥2 yrs. 0.800 0.660-0.970 0.160 0.040 1984-1987 (high harvest) ≥2 yrs. 0.660 0.530-0.830 0.290 0.050 a Calculated using Micromort program (Heisey and Fuller 1985) b Sime, unpublished data c Dusek et al. (1992)

accounting for about half of the losses. Hunting was the major known During winter-spring, the average female cause of death among adult female mule mortality rate was 7 percent (range = deer in all but the Bridger Mountains 0-24 percent). where natural mortality accounted for Females in the Missouri River Breaks about three-fourths of the total. Hunting had an average annual mortality rate mortality varied with season type and bag of about 17 percent (2-43 percent), 11 limits for antlerless deer. percent occurred in summer-autumn The lowest average annual mortality and 6 percent over winter, during 1960- rate for adult females occurred in the 86 (Hamlin and Mackie 1989). The Bridger Mountains where relatively difference in annual mortality between conservative hunting regulations and these two environments was due to other factors limited harvest to an average greater hunter harvest of females in the 4.8 percent of preseason numbers (Table Missouri River Breaks. Natural mortality 5). In the Missouri River Breaks total was very similar in the two environments annual mortality of females averaged (Table 5). 17 percent over a 26-year period. This included annual mortality averaging 22 Although coyote predation on adult percent during 13 years of 2 deer-either females was low, it was the major known sex seasons, 17 percent during 6 years natural mortality factor and the second with 1 deer-either sex hunting with some leading cause of death overall in the additional antlerless licenses, and 7 Breaks; predation was also suspected percent for 8 years in which only males to be at least the proximal factor in could be taken on general licenses. On most losses of unknown cause. Coyote the Cherry Creek area, annual hunting predation on adults was highest during mortality rates for adult females averaged winter and spring and included individuals 21 percent during 1982-86, but increased in good as well as poor condition (Hamlin progressively from 1 percent during a 1 and Mackie 1989). Although coyotes and/ deer-either sex season to 9 percent with or other predators occurred on all study 1-deer either sex plus limited antlerless areas, we did not determine the incidence permits and 32 percent (27-37 percent) of predation relative to other natural when multiple licenses for antlerless deer mortality factors on adult females in other were available to hunters (Wood et al. populations during the studies. 1989).

104 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Hunting was the major known cause of death among adult female mule deer in all but the Bridger Mountains where natural mortality accounted for about three-fourths of the total.

Long-term trends in overwinter and females experienced average hunting total annual mortality of adult females mortality of 6 percent in the northwestern in the Missouri River Breaks (Fig. 53) montane forest environment where illustrate the variability in female mortality relatively conservative hunting occurred. rates and the downward trend in total This compared with harvest rates of 16- annual mortality associated with reduced 39 percent in the prairie-agricultural and hunting mortality after 1975. These data plains riverbottom environments (Table 6). also document unusually high incidences of mortality among adult females during 1961-62 and 1971-72. Similar high total and overwinter mortality exceeding 40 percent apparently occurred during 1994- 45

95 (Stivers unpubl.). 40 Total Annual Mortality Rate The high annual mortality rate for adult females during 1961-62 was not 35 Overwinter Mortality Rate associated with overwinter mortality, 30 but with heavy harvest rates of females 25 promoted by special harvest regulations and environmental conditions that 20 increased vulnerability of deer to harvest 15 that year. The extreme mortality during Percent Mortality winter 1971-72 occurred at a high female 10 population when dry conditions reduced 5 the quantity and quality of forage produced 0 during the previous summer. This event occurred after a long series of years of 1960-611962-631964-651966-671968-691970-711972-731974-751976-771978-791980-811982-831984-851986-87 relatively heavy harvest of adult females (Fig. 53). Figure 53. Among whitetail populations, non- Trends in total annual and overwinter mortality rates for adult female mule deer in the Missouri River Breaks, 1960-61 through hunting mortality rates of females were 1986-87. low to moderate (≤ 16 percent). Mature

Po p u l a t i o n Dy n a m i c s 105 In both species and across all summer and/or severe winter conditions study areas hunting mortality had little, such as 1971-72 and 1994-95. Relatively if any, measurable effect on natural few adult females remained in areas of In both species and mortality rates of adult females. Thus, low relief following episodes of significant across all study our findings do not support the concept overwinter mortality (Hamlin and Mackie areas hunting that mortality from hunting would 1989). replace that from other causes. Perhaps mortality had little, the low levels of natural mortality that if any, measureable Adult Male Mortality characterized adult females in all of our effect on natural studied populations may have represented As noted earlier, adult males mortality rates of base levels of mortality that could not were outnumbered by females in all adult females. be substantially reduced (Dusek et al. populations. This reflects higher 1992). We conclude that opportunities mortality of males associated with hunter to substitute harvest for individuals that selection for males as well as differences might otherwise die from other factors in biology and behavior of the sexes. is limited. Consequently, the death of We found that males rarely lived one adult female would not increase the more than 7-8 years. Hunters selected probability that others would survive, for older, larger antlered males over and hunting losses became additive to younger, smaller individuals. Also, where overwinter and other natural mortality. they occurred, males over six years of Where hunting mortality of adult age appeared particularly vulnerable females is low (Pac et al. 1991), relatively to mortality during severe winters. As large numbers of adult females may occur dominant breeders, most of these older in older, more vulnerable age classes. males entered winter in poor physical When severe winters occur in such areas, condition. Yearling males were also higher than usual overwinter mortality vulnerable to mortality during severe of adult females may result. Such winters. As with fawns, yearling males mortality is not necessarily related to apparently emphasize body growth over forage availability, but rather to decreased accumulation of fat reserves, and many energy reserves and physical condition are unable to survive the energy deficits associated with old age. Although heavier associated with long periods of cold and hunting may have removed some of these utilization of low quality forage. vulnerable deer, hunting mortality tended In our studies, adult males to be biased toward younger rather than experienced mean annual mortality rates old age classes. ranging from 41 to 61 percent, more We also determined differences than twice the rates experienced by adult in mortality rates among adult females females (Table 7). Average rates and related to habitat-use strategy and variation were similar for the two species. home range characteristics. Females Annual mortality rates varied widely establishing yearlong residency in areas within and among individual populations of low topographic relief in the Missouri from a low 18 percent to a high 80 Hunting was River Breaks experienced significantly percent. Both the highest and the lowest the major cause higher mortality than either females annual rates were recorded for mule deer of mortality for resident to home ranges in steep terrain in the Bridger Mountains. adult males of with north- and south-facing slopes or Hunting was the major cause of both species in all females that migrated annually to winter mortality for adult males of both species environments. in steeper terrain (Hamlin and Mackie in all environments. Among populations, 1989). As numbers of adult females mean annual hunting mortality rate increased in the population, greater ranged from 33-58 percent of estimated numbers inhabited drainage heads and pre-hunting season populations. Within along ridgetops on home ranges with low individual populations, annual variation topographic relief. This may have served in hunting mortality was greatest in the to predispose these females to very high Bridger Mountains where rates varied mortality during years of extremely dry from 14-70 percent (Table 7).

106 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Despite greater vulnerability and more uniform distribution across populations and habitats, hunting mortality rates for yearling males were generally lower than rates for older males, and smaller antlered (spike) yearling males were harvested at a lower rate than larger antlered (2-point) yearlings. In the Missouri River Breaks yearling males were harvested at an average rate of 36 percent compared to 53 percent for older adults (Hamlin and Mackie 1989). On the lower Yellowstone River, hunters harvested males ≥ 2 years in greater proportion than their occurrence in the population, while the relative proportion of yearling males taken decreased with smaller antler size (Dusek et al. 1989). Spike antlered yearling males apparently were selected against or undetected by hunters in all areas. The relatively low rates of overwinter among adult males than it does among mortality for adult males suggested females. With the high harvest rates for that hunting mortality may substitute adult males in most populations, a greater for natural mortality to a greater degree proportion of these animals (e.g.. older,

Table 7. Total and cause-specific annual mortality ratesa for adult (≥1 year) male mule deer and white-tailed deer in four Montana environments.

Environment total Annual Mortality hunting Mortality Natural (Non-hunting) Mortality MULE DEER Mountain-Foothillb 0.460 (0.145)c 0.404 (0.160) 0.056 (0.081) N = 14 years range 0.182-0.800 range 0.143-0.700 range 0.000-0.257 Timbered Breaksd 0.414 (0.109) 0.375 (0.112) 0.038 (0.024) N = 26 years range 0.217-0.609 range = 0.150-0.580 range = 0.000=0.114 Prairie-Badlandse 0.612 (0.065) 0.584 (0.064) 0.036 (0.023) N = 5 years range 0.520-0.700 range 0.480-0.650 range 0.000-0.060 White-Tailed DEER Plains Riverbottomf 0.600 (0.070) 0.550 (0.070) 0.060 (0.020) N = 6 years Northwest Montane Forestg 0.419 (0.019) 0.328 (0.018) 0.091 (0.032) N = 6 years a Calculated from arithmetic models that reconciled successive population estimates with data on population composition, recruitment, and marked deer mortality rates on all areas except the Northwest Montane where rates were calculated using Micromort (Heisey and Fuller, 1985). b Pac et al. (1991). c Average (std. dev.). d Hamlin and Mackie (1989). e Wood et al. (1989). f Dusek et al. (1989). g Sime, unpubl., representing mortality rates calculated across the span of years.

Po p u l a t i o n Dy n a m i c s 107 predation by coyotes, mountain lions, and possibly bears. Adult females, conversely, selected home ranges that provided for isolation and security from predation. Evidence from recent studies in the Bridger Mountains (Pac, unpubl. data) suggests that predation by both coyotes and mountain lions may be greater than previously believed or has increased recently and is sufficient to influence survival of males of all age classes.

Emigration and Immigration

Dispersal is the movement of (young) animals from their natal or a dominant breeding males destined to die subsequently established home range. over winter) are harvested. Lower harvest Dispersal movement culminating rates during many years, especially those beyond boundaries of a population following extremely dry summers and/or we call emigration. Movement of preceding severe winters might not result deer from adjacent populations into a in appreciably lower total annual mortality population we studied was immigration. rates. Gavin et al. (1984) reported that Immigration and emigration represent adult male Columbian white-tailed deer special gains and losses (i.e., recruitment experienced 40 percent annual mortality and mortality) of adults in population even in the absence of hunting. dynamics. Immigration and emigration Data for white-tailed deer on the also may be viewed as a mechanism for lower Yellowstone River also indicated that redistribution of individuals, especially hunting mortality compensated for natural young males, for genetic interchange The relatively low mortality because few males >4 years of and resource partitioning (Dusek et al. rates of overwinter age survived to be predisposed to natural 1989, Hamlin and Mackie 1989, Pac et al. mortality for adult mortality. Based on annual mortality rates 1991). In some mammals, redistribution males suggested observed, only 26 (3 percent) of 1000 of individuals between established that hunting yearling males would be expected to live populations has been reported to mortality may to 5 years when the cumulative energetic influence reproductive potential through substitute for cost of breeding could lead to overwinter disruption of social structure and stress natural mortality mortality. This compared with a similar effects (Bailey 1969). to a greater degree cohort of 1000 yearling females, from Most deer populations, including among adult which 370 (37 percent) would be alive all we studied, were surrounded at least males than among after 5 years, and 175 (18 percent) would in part by habitat to or from which deer be expected to live to 8 years. females. could move. In most areas, we were only Our intensive studies through 1987 able to measure emigration. Immigration did not directly detect predation as an was rarely detected directly and was important mortality factor on adult males evident only because population data did of either species in any environment. not indicate annual declines associated However, resource partitioning and with emigration. habitat segregation between the sexes Emigration and immigration rates often resulted in male home ranges varied within and among populations and located in local environments frequented habitats. Emigration rates were strongly by and possibly conducive to successful sex and age related. Relatively high

108 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a rates of emigration were recorded among females was ≤10 percent during most young mule deer only on the Missouri other years. However, net emigration River Breaks and Cherry Creek areas and involving possibly 25 percent or more for white-tailed deer on Cherry Creek. of the yearling females from the 1986 Emigration was uncommon among mule cohort apparently occurred between deer in Bridger Mountain populations, spring and December 1987 (Hamlin and especially among females on the west Mackie 1989). This followed the highest slope. Emigration was recorded but adult female density ever recorded on the appeared rare among whitetails marked in area, and may have represented a form the Swan Valley (Mundinger 1980). of “saturation” dispersal (Lidicker 1985) Where it occurred, emigration was following the filling of most suitable greatest and most consistent among reproductive habitat in the area by the young males; emigration of females was increasing numbers of females during the comparatively low and more variable. late 1970s-mid 1980s. Emigration centered in yearlings, with Emigration of yearling, 2- and 3-year only occasional incidence among 2- and old female mule deer was very low in the 3-year old males and females or >8-year Cherry Creek population and must have old females. been balanced by immigration (Wood Relatively high rates of emigration et al. 1989). In the Bridger Mountains, averaging 51 percent (27-78 percent emigration of females was very rare and annually during 1976-85) among yearling unimportant in dynamics of west slope male mule deer in the Missouri River populations. Greater numbers of young Breaks apparently were balanced by females emigrated from populations nearly equal immigration by young on the east slope, but this also was males from adjacent areas (Hamlin and unimportant in population dynamics. Mackie 1989). Even higher rates (ave. Data for white-tailed deer in 67 percent) of emigration among yearling river bottom and prairie-agricultural mule deer males in the prairie-badlands environments indicated relatively high environment also appeared to be largely rates of dispersal of yearlings from compensated by immigration; though natal home ranges, but variable rates slight declines in numbers of yearling of emigration from natal populations males on the area through summer (Dusek et al. 1989, Wood et al. 1989). suggested a possible net loss of adults On the lower Yellowstone River, nearly Overall, emigraton from these movements (Wood et al. half (46 percent) of all males marked as did not appear to 1989). fawns and monitored for 18 months or occur in either a The more variable rates of more dispersed from natal home ranges, manner or to an emigration by young females may have primarily to establish new ranges up or extent that it would been influenced by matriarchal aggression down the river bottom. However, very singly control deer and population density, and apparently few left the river bottom. A similar numbers in any were not balanced by immigration. In pattern of dispersal was evident for young population. the Missouri River Breaks, most female females of which 17 percent dispersed to emigration (ave. = 27 percent) from the establish home ranges elsewhere on the study area occurred at low population river bottom (Dusek et al. 1989). Limited levels, among yearlings in the first two, data from marked white-tailed deer and numerically large cohorts recruited a high net loss of adults during summer following the population low in the suggested a high rate of emigration for mid 1970s (Hamlin and Mackie 1989). young females as well as males from the This was similar to the “pre-saturation” Cherry Creek population (Wood et al. dispersal proposed by Lidicker (1978), 1989). and it served to promote rapid filling of Overall, emigration did not appear to vacant habitat and irruptive population occur in either a manner or to an extent growth in the patchy environment. that it would singly control deer numbers Annual emigration rate of yearling in any population. Rather, as in the

Po p u l a t i o n Dy n a m i c s 109 Missouri River Breaks, it was one of many Once established, all deer factors that at times limited population populations fluctuate over time. Such growth below numbers that may have fluctuation varies among populations in occurred based on fawn recruitment response to the specific conditions that (Hamlin and Mackie 1989). There, exist in each area. Although, at times, and perhaps in other areas, the most populations of both species may appear important role of dispersal and emigration to fluctuate in synchrony, we found may be as a mechanism to maintain deer deer numbers and dynamics to vary in available habitat, allocate resources by independently between species and study sex and age, and promote the genetic and areas. The more similar the environments behavioral diversity necessary to sustain and conditions, the less the difference populations over time. in the pattern, timing, and magnitude of population fluctuation. We studied established populations, Patterns of Population so we had only limited opportunity to study early development and Growth and Fluctuation growth phases of populations. Data on population growth were obtained primarily during periods of recovery from Deer experience two distinct low population levels. Other long-term phenomena in population dynamics: deer population studies (e.g., McCullough (1) a period of population growth 1979) that provide the basis for current associated with colonization or concepts about population dynamics reestablishment in new or vacant have focused primarily on the process of habitats, and (2) post-establishment population growth. fluctuations and changes in population We recognized four types of size and composition influenced by population change over time: general, population-habitat relationships, long-term changes in distribution and environmental variation, and demographics; periodic fluctuations management. Where minimum spanning several years; year-to-year adaptation is required and recruitment fluctuations; and season-to-season and dispersal are high, initial population changes within years. growth may be rapid, in an eruptive Long-term changes in abundance pattern, at least during the early stages can occur at landscape, ecosystem, or of development. Where substantial population levels. Mule deer increased adaptation and specialized movements from scarcity in most areas of Montana are necessary to access and effectively to abundance throughout the state during utilize all habitat, or recruitment and the 1940s-1960s (Egan 1971). Since the female dispersal are limited, deer will colonize slowly and population growth may be relatively gradual.

110 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a mid to late 1960s populations have declined in many areas, especially in western and southern mountain-foothill environments, while numbers have increased in plains habitats (Mackie et al. 1985, Eustace 1988). White-tailed deer have gradually increased and spread throughout most of Montana since the 1930s and 1940s (Allen 1971). Present whitetail populations may represent all-time highs in both distribution and abundance for the species in the northern Great Plains and Rocky Mountains (Mackie et al. 1985, Dusek et al. 1989). We found no evidence of direct competition leading to replacement of one species by the other. However, regional buck harvest trends, especially for southern and southwestern Montana, suggest continuous spread and increases in numbers of whitetails since the mid 1970s, while mule deer have declined from 1960-1975 levels (MFWP Wildl. Div. 1995, Hamlin and Erickson and Mackie 1989). Aerial surveys 1996). Such trends may be indicative during winter and/or early spring, 1988- of changes in relative abundance of the 1995 (Stivers unpubl.) provided data two species in areas where whitetail to generally estimate population trends populations have developed or expanded following termination of intensive studies in the wake of environmental changes in 1987. Collectively, these data provided and conditions favorable to the species estimation of population trend from the (Eustace 1988, Dusek et al. 1989), 1930s through 1995 (Fig. 54). while mule deer numbers have declined, These data, together with various especially in their stronghold, mountain- reports (Hamlin and Mackie 1989), foothill habitats. suggested that the present population Long-term trends in distribution on the Missouri River Breaks study area and abundance in specific habitats developed from a small number of mule include deer responses to changes in deer remaining on the area during the land use. Long-term trends also reflect 1930s (Fig. 54). Mule deer numbers subtle responses to changes in climate apparently increased progressively from and weather conditions, the distribution, 1938 or 1939 through 1947-1948 in a abundance, and behavior of predators pattern similar to irruptive or introduced and other wild animals, and the presence populations (Caughley 1970, McCullough and activities of man. Thus, trends may 1979, 1983). After 1948, the population be difficult to detect, and cause-effect fluctuated frequently and often widely, may be impossible to isolate or quantify. between years and over periods of years For our study areas, data were in relation to patterns of recruitment available to quantify population and adult mortality. Trends in spring fluctuations over periods of 6 to 28 years. populations during the 27-year period of In the Missouri River Breaks, additional intensive studies from 1960-61 through data were available to generally estimate 1986-87 (Fig. 55) indicate the relative early population growth and subsequent contributions of recruitment and adult fluctuations in numbers of mule deer mortality. from the late 1930s to 1960 when more Even in these long-term data intensive studies were initiated (Hamlin sets, interpretation of pattern in long-

Po p u l a t i o n Dy n a m i c s 111 term trends can vary depending upon size. A second could suggest periodic perspective and any additional population fluctuations within a long-term trend data available and applied to analysis. of generally increasing deer numbers One perspective might interpret the long- from the early 1960s through the early term trend to reflect annual and periodic 1990s. A third perspective might point fluctuations of variable amplitude around to fluctuating, but generally decreasing a long-term mean, or average population numbers from the initial peak in the

1800 Modeled Estimates

1600 Observed Estimates

1400

1200

1000

800

600 Number of Mule Deer 400 ...long-term 200 trends in mule deer numbers 0 in the Missouri 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 River Breaks Figure 54. do not support Estimated numbers of mule deer on the Missouri River Breaks study area during early the view that a winter 1930-1995. population, after declining from an initial peak, never recovers to that 600 1400 high. Fawn Recruitment Number of Mule Deer in Spring Population Adult Mortality 1200 500 Deer Numbers

1000 400

800 300 600

200 400

100 200

Number of Fawns Recruited or Adults Removed 0 0

196119621963196419651966196719681969197019711972197319741975197619771978197919801981198219831984198519861987 Figure 55. Spring population trend in relation to fawn recruitment and adult mortality for mule deer on the Missouri River Breaks study area, 1961-1987.

112 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a late 1940s through 1976-77, followed documented fluctuations and trends only by generally increasing numbers from after populations had become established 1977 through the early 1990s. Still a and declined from early “peaks.” fourth view might suggest that trends In the Bridger Mountains, mule deer reflect periodic, cycle-like fluctuations, “…were quite scarce and infrequently influenced by population and proximal seen” as late as 1948, while a “large environmental factors. Within the long- population” was evident by the mid- term data set, trends for any combination 1950s (Wilkins 1957). Data obtained of two years or more could provide during 1972-75 compared with 1955-56 numerous interpretations or conclusions suggested a possible population decline depending upon such chance events as the of 50-67 percent by the time our intensive particular year selected to begin and the population studies were initiated (Hamlin total number of years involved (Hamlin 1977). Our data provided detailed and Mackie 1989). It should also be estimates of population size and trend noted that regardless of perspective relative to recruitment and adult mortality and interpretation, long-term trends in for mule deer on the Northwest Slope mule deer numbers in the Missouri River from 1972 through 1987 (Fig. 56) (Pac Breaks do not support the view that a et al. 1991) and more general estimates population, after declining from an initial during 1988-97 (Pac unpubl. data). peak, never recovers to that high. There was no evidence, either from Current mule deer populations in trends in estimated mule deer numbers the Bridger Mountains and on the Cherry in individual populations or from more Creek area also developed during the general population surveys through 1940s and 1950s (Wilkins 1957, Wood et 1995-96, of long-term increase or al. 1989), either as a result of increases decrease in populations in the Bridger among small numbers of deer that Mountains since 1974. Although periodic existed in these areas or by colonization fluctuations spanning several years from other habitats. Because data on appeared large to hunters and public early growth were lacking, our findings perceptions of deer abundance, they

80 250 Fawn Recruitment Deer Numbers Number of Mule Deer in Spring Population 70 Adult Mortality 200 60

50 150

40

100 30

20 50

10 Number of Fawns Recruited or Adults Removed

0 0

1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 Figure 56. Spring population trend in relation to fawn recruitment and adult mortality for mule deer associated with the Armstrong Range, Northwest Slope, Bridger Mountains, 1974-87.

Po p u l a t i o n Dy n a m i c s 113 seldom exceeded a 65 percent change in Relative changes in populations of numbers. It appeared that the Bridger mule deer in the three broadly different Mountain populations, dominated by environments were measured only during long-lived adult females and low annual 1975-1987 (Fig. 58). Again, although recruitment (Pac et al. 1991), were less some synchrony was evident in pattern likely to experience the wide fluctuations and timing of changes in the three observed in mule deer populations in the populations, each fluctuated in a different Missouri River Breaks and other habitats. manner and at a slightly different rate. No data were available concerning Greatest fluctuation in annual specific population characteristics or population levels occurred on the Cherry trends on the Cherry Creek area prior to Creek Area, where mule deer numbers 1975. However, if mule deer trends on at their peak in 1983 were 600 percent this area were similar to southeastern higher than the low in 1976, then declined Montana as a whole, harvest data (MFWP by more than 75 percent to another low in Annual Hunting and Harvest Reports, 1987. Conversely, population size varied Wildl. Div., Helena.) would suggest that only slightly in populations in mountain- the population increased to an initial peak foothill habitat where mule deer numbers during the early 1960s. It subsequently on the northwest slope also increased, but fluctuated through a low in the mid 1960s were only 68 percent higher in 1984 than to a second, possibly higher, peak during at the low in the mid-1970s. Further, they the early 1970s, and declined sharply did not decline significantly during the to another low in 1975 (Wood 1987). mid 1980s, and they increased to an even During this study, surveys were completed higher peak in 1991. and population estimates relative to The fluctuation in numbers in recruitment and adult mortality were timbered breaks habitat was intermediate calculated only during 1975-1987 when between mountain-foothill and prairie- mule deer numbers again cycled from a badlands environments. Mule deer low through a peak in 1983-84 to still numbers on the Missouri River Breaks another low by 1986-87 (Fig. 57). study area increased about 250 percent from 1977 to 1983, declined slightly during 1984-86, then increased again in 1987 (Hamlin and Mackie 1989). The Cherry Creek and perhaps other 600 1200

Number of Mule Deer in Spring Population mule deer populations in southeastern Deer Numbers Montana plains habitats may be

500 Fawn Recruitment 1000 characterized by extreme “boom and bust” dynamics and trends over time. Adult Mortality Locally, however, trends may also be 400 800 influenced by specific environmental conditions, deer population structure, 300 600 and harvest strategies. Deer surveys on lightly-hunted, surface-mined, and urban habitats in the vicinity of Colstrip 200 400 indicated more or less steady population growth from 1979-80 to the early 1990s 100 200 (Fritzen 1995); there was no evidence of severe decline similar to that on the Cherry Creek area during 1984-87. Number of Fawns Recruited or Adults Removed 0 0 Long-term census data to quantify

1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 growth and development of individual Figure 57. populations were lacking for white- Spring population trend in relation to fawn recruitment and tailed deer. Historical populations in adult mortality for mule deer on the Cherry Creek study area, eastern Montana were largely extirpated 1976-1987. during the late 1800s, and whitetails

114 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a 1.0

0.8 Mountain-Foothills

0.6 Timbered Breaks

0.4 Prairie-Badlands

0.2

0.0

-0.2

-0.4 (deviation from mean log) Changes in Deer Numbers -0.6

-0.8

-1.0 For both

1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 species, greatest Figure 58. fluctuations were Comparative annual changes in mule deer numbers in three Montana environments in open, patchy, during spring, 1974-1987. and variable environments; the least fluctuations percent in white-tailed deer numbers from became reestablished on bottomlands were associated along the Yellowstone River and upland 1975-76 to 1984-85 followed by a slight with more diverse, prairie-agricultural environments only decline to 1986-87. Maximum estimates relatively stable during 1940s and 1950s (Dusek et al. suggested continuous population increase habitats. 1989, Wood et al. 1989). On the other to an approximate doubling of the hand, the species apparently was well population between 1975 and 1987. By established in the Swan Valley and other comparison, white-tailed deer numbers on coniferous forest habitats in northwestern the lower Yellowstone River populations Montana prior to and following increased by 54 percent from 1980 to settlement. Reports indicated abundant 1983-84 before decreasing to near-1980 populations there in the early 1900s, levels by 1986-87. increases to early peaks in the mid-1950s, Numbers of whitetails on the Cherry and general declines by the early 1970s Creek study area were relatively stable (Arno et al. 1987) when our studies during 1975-78, approximately doubled began. Thereafter, harvest records from from 1979 to 1983, then declined by questionnaire surveys and population 91 percent to about one-fourth of the estimates (MFWP unpubl.) indicate that previous low in 1986-87. Although white-tailed deer numbers gradually white-tailed deer and mule deer fluctuated increased through the mid 1990s. somewhat similarly, whitetail numbers Comparative post-season numbers peaked and began to decline one year and trends in the three whitetail earlier than mule deer numbers (Fig. 60). populations during years for which For both species greatest fluctuations population estimates are available showed were in open, patchy, and variable that each population changed differently environments; the least fluctuations were (Fig. 59). For the Swan Valley, minimum associated with more diverse, relatively population estimates (constructed using stable habitats. harvest-based models, Riley, pers comm.) indicated a gradual increase of about 44

Po p u l a t i o n Dy n a m i c s 115 Swan Valley 8000 400 Lower Yellowstone River Number of Deer on Cherry Creek Area 7000 350 Cherry Creek

6000 300

5000 250

4000 200

3000 150

No. of Deer on Swan Valley or 2000 100 Lower Yellowstone River Areas

1000 50

0 0

1975-76 1976-77 1977-78 1978-79 1979-80 1980-81 1981-82 1982-83 1983-84 1984-85 1985-86 1986-87

Figure 59. Comparative population trends for white-tailed deer on Swan Valley, Cherry Creek, and Lower Yellowstone River study areas during early winter, 1975-76 through 1986-87.

400 Number of White-tailed Deer 1400

Number of Mule Deer 350 1200 Number of Mule Deer 300 1000 250 800 200 600 150 400 100 Number of White-tailed Deer 50 200

0 0

1975-761976-771977-781978-791979-801980-811981-821982-831983-841984-851985-861986-87 Figure 60. Comparative trends in total numbers of mule deer and white-tailed deer on the Cherry Creek study area during early winter 1975-76 through 1986-87 (after Wood 1987, Wood et al. 1989).

116 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Management Implications

Traditional Concepts

Management of mule deer and populations was limited. Technology for white-tailed deer in Montana during monitoring deer populations included the 1930s through the mid-1970s was binoculars, spotting scope, and horse based on the theory of range “carrying or motor vehicle. As a result, early capacity”, which related deer numbers observers concluded that deer could to the amount of forage available in not be counted accurately and placed winter. This concept originated primarily little emphasis on monitoring population in animal science and husbandry of characteristics and dynamics in most domestic livestock in a simple, controlled areas. management system. Although defined in Early deer science embraced essentially the same way, it was applied to only limited understanding of the free-ranging deer in complex ecological complexities of natural ecological systems systems. and deer-habitat relationships. The It seems easy to critique and knowledge accumulated was based on criticize early wildlife management interactions that were most apparent or concepts applied to deer. In restrospect, discernible within existing technology at the time early theory and “principles” and management objectives. Improved were developed, knowledge about deer understanding came later as new

Tr a d i t i o n a l Co n c ep t s 119 concerns and technology paved the way condition of important browse plants, for more intricate studies of wildlife harvest surveys, and indices of fawn populations, their environments, and survival. Fawn:female ratios lower than animal-habitat interactions. 100:100 were considered indicative of Pioneering studies in the western deteriorating forage conditions on winter United States were focused on mountain- ranges. foothill habitats where mule deer had Deer populations were presumed to become abundant and predators were be inherently eruptive, and if unharvested relatively scarce. Deer congregated at they would increase until they exceeded high densities on limited winter ranges range carrying capacity and overused the where an apparent direct correlation browse supply. Overuse of the browse between deer numbers, winter forage supply resulted in long-term declines supply, and over-winter mortality provided in range carrying capacity and deer a clearly focused management problem. numbers. As populations numerically In Montana and other states, biologists expanded toward carrying capacity, Pioneering studies extended this interpretation to essentially fawn recruitment declined and age in the western all environments occupied by both species structure of the population became older. United States of deer. Consequently, the uniform goal of deer were focused on In Montana, game biologists management in Montana emphasized mountain-foothill understood carrying capacity as the the use of hunter harvest to bring deer habitats where number of deer a local area could numbers into balance with forage supplies mule deer had sustain without long-term damage to (Mussehl and Howell 1970). browse plants supporting the population become abundant This management theory assumed during winter. The concept of deer and predators were that each incremental decrease in deer habitat primarily considered only the relatively scarce. density below carrying capacity would winter forage supply, and it was believed result in corresponding incremental that browsing pressure by deer was increases in fawn production and the main factor that influenced forage recruitment (Cole 1961). Increased condition and productivity of browse recruitment resulted from improved plants. Changes in the browse supply quantity and quality of food available detected through monitoring were to surviving individuals. Such a believed to reflect changes in condition density-dependent response is often and productivity of deer populations. The called compensatory reproduction. It primary techniques employed in deer was thought to reach an optimum at management were estimates of use and a population density approximately half the number of deer occurring at “K” carrying capacity. Harvesting the population to this theoretical level each year would result in “Maximum Sustained Yield” (MSY) (McCullough 1984) which represented the maximum number of deer that may be harvested while maintaining a stable population. Other related principles evolved including compensatory mortality. That is, “In a resilient population, severe loss rates [from many mortality factors] may in effect substitute for each other without mounting up excessively high in total. . . .The death of one individual may mean little more than improving the chances for living of another one.” (Errington 1943). This implied that hunting mortality substituted for natural losses

120 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a on nearly a 1:1 basis. A deer population in this system were dictated by only a experiencing increasing rates of hunting few influential components such as deer mortality would experience corresponding numbers and food. Biologists assumed decreases in natural mortality from that a population would readily return starvation. to a previous state of equilibrium once a These relationships are embodied perturbing force such as a hunter harvest in what is technically called the “density- was removed. dependent logistic model.” Obviously, The theories underlying most biologists don’t inform the public these concepts were developed to Deer populations that management decisions are based mathematically explain growth of were presumed on such a technical-sounding concept laboratory populations of bacteria and to be inherently nor that its application involves many insects that grew at a constant rate and eruptive, and if assumptions that may not be met in all reached carrying capacity in very simple unharvested, they areas under all environmental conditions. closed systems. As implied by Botkin would increase However, this conceptual view guided (1990), this theoretical and experimental until they exceeded deer management in Montana for more approach removes the very essence of range carrying than 40 years. It subtly gave biologists ecological systems which exhibit variation and the public the false perception at all levels of organization and function. capacity and that deer populations functioned in a Botkin (1990) concluded that it is overused the dependable, almost mechanical manner. impossible to overestimate the influence browse supply. Until the 1970s, application of these theories on twentieth century of these concepts was logical and population biology. Watt (1962) noted convenient because it was assumed that that ecologists believed that their science few data were required to assess deer lacked theory when, in fact, it had too population status. Population trend, fawn much theory, often borrowed from other recruitment, and adult survival did not sciences, and not carefully connected need to be monitored. By manipulating to field observations. Often, theory and hunter harvest, monitoring browse conceptual models are not considered supplies, and occasionally gauging important to the practicing deer biologist. ...ecologists success by monitoring fawn:doe ratios, However, as described in previous believed that their managers believed the equilibrium pages, theory has played a dominant science lacked between deer and vegetation could be role in shaping traditional concepts of theory when, in deer populations and their habitats. We adjusted to increase yield. fact, it had too For this conceptual view to be can improve the theory and art of deer much theory, often applicable, certain assumptions must be management if they are implemented in borrowed from made concerning the deer population and the context of an experimental design other sciences, its habitat. A logistic population occupies (Romesburg 1981, Walters 1986). With a steady-state environment that provides these management experiments, we can and not carefully resources at a constant rate where each continually contribute to knowledge about connected to field individual deer consistently decreased the dynamics of deer populations across observations. the availability of food for other members the ecological spectrum in Montana. of the population. Therefore, changes

Tr a d i t i o n a l Co n c ep t s 121 An Ecological Perspective

Our profession has traditionally a characteristic faunal array along with approached management of deer the distribution of human land uses. All populations and their habitats based on this creates the arrangement of habitats concepts that minimized the importance that satisfy the different requirements of variation in all aspects of the system. of the two species of deer. Therefore, Management, however, must function the spectrum of deer densities among within the realm of ecological variation Montana environments is determined to be successful. Here we present a by ecological factors largely beyond the conceptual framework that explains the control of management. We observed dynamics of deer populations within the high deer density where resources context of the environmental variation for fawn-rearing were of high quality, that occurs across Montana. plentiful, and in close proximity to large Our attempts to explain fluctuations areas that optimized deer survival over in abundance of mule deer and white- winter. Low density occurred where tailed deer required an expansion of the resources supporting reproduction and/or concept of habitat to include more than survival were scarce or widely distributed. food, cover, and water upon which most Superimposed over these landscape traditional interpretations were based. level differences were other more Deer populations function within complex localized environmental factors that ecosystems. The interface between varied on a daily, seasonal, and annual topography and local climate results in basis. Fluctuations in the environment a mosaic of vegetation communities and resulted in dynamic, short-term changes in conditions favorable and unfavorable for deer (Mautz 1978). Variation in precipitation and temperature prior to and during the growing season determined forage quantity and quality. Consequently, the physical condition of deer as they entered winter varied from year to year as did winter severity. This relationship between environmental variation and the physiological response by deer can be visualized in terms of an energy balance that has positive and negative periods. During positive periods, energy in the form of high quality forage is plentiful and cost of obtaining it is minimal. The physiological demands of reproduction and recovery of body condition are not only met, but ultimately more energy is

122 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a acquired than expended and the excess is sex, age, reproductive status, and social “stored” as fat reserves. During negative position. Behavioral adaptations such periods, energy is obtained in the form as sexual segregation, home range of poor quality forage which may be fidelity, and female territoriality during scarce and is obtained at high cost. Less fawning insulate individual deer from energy is taken in than expended, and fat direct competition for forage with other Our profession reserves are depleted by basic activities members of the population. Social has traditionally related to survival. Energetics varied behavior serves as a mediating influence approached between species, among sex and age between deer and their environment. management of classes, and from one environment to Rather than the traditional view of a deer populations another. Annual changes in the energy balance between a few components (such and their habitats budget experienced by specific deer as deer density and forage quantity), we based on concepts populations were influenced by variation hypothesize that deer populations vary that minimized in these environmental factors and by in accordance with a balance of the total the importance social interactions that relegated some environment (Hamlin and Mackie 1989). of variation members of the population to lower Over time, favorable conditions that in all aspects quality maintenance habitats. Caughley result in population growth are balanced of the system. and Gunn (1993) also concluded that by unfavorable conditions that result Management, herbivore nutritional status could change in decline. Environmental variation is however, must independently of herbivore numbers or measurable, and population fluctuations function within the density in environments with high annual usually remain within bounds which are realm of ecological weather variations. characteristic of a particular environment. variation to be Across many of the environments If the environment fluctuates too widely occupied by deer in Montana, annual or changes its character, populations may successful. variation in precipitation, duration of the disappear on a local scale. This has and growing season, and changes in winter will continue to happen. severity would violate an important Viable deer populations exist in some assumption of the logistic model requiring places and not in others, and density a steady-state environment. Caughley varies across the species’ distribution. (1987) indicated that concepts pertaining The dynamics of deer populations to ecological carrying capacity were constantly adjust as the environment largely an abstraction in environments fluctuates within predictable bounds that experienced substantial variation whether it’s the wide swing in variation Rather than the in annual rainfall. At a minimum, in fawn recruitment and population size traditional view assessment of variation in northern experienced in the prairie-badlands or of a balance environments must include factors the narrow swing that occurs in the between a few influencing both summer growing seasons Salish Mountains of northwest Montana. components (such and winter severity. These two variables The outcome is a characteristic set of as deer density and are instrumental in defining the periods population parameters that reflect the forage quantity), of energy surplus and deficit experienced environmental variation. we hypothesize by deer. We contend that variation in that the deer many of the environments occupied by populations vary in deer is of much greater importance to Deer Population accordance with a understanding their population dynamics than previously believed. Caughley Dynamics and Hunter balance of the total (1987) concluded that environmental environment. variability “can no longer be ignored as Harvest noise, it is now the signal.” In Montana, most environments Deer managers cannot avoid the are not stable, and deer are not equal process of obtaining vital population in their resource requirements. Deer parameters (fawn recruitment rate, respond to their local environments as natural mortality rate of adult males and individuals. Each has its own unique females) as well as monitoring population strategy of habitat use depending on

An Ec o l o g i c a l Pe r s pe c t i v e 123 size if the management goal is to optimize adult bucks are shot before they attain harvest. Interaction between fawn an age when they are most vulnerable recruitment (population gain) and natural to natural mortality. Bucks experienced mortality of adults (all non-hunting high annual average harvest rates (33- losses to the population) determines 58 percent) in all populations studied. the “window of opportunity” for hunter Preliminary results from recent studies in harvest. The effect of a harvest rate the Bridger Mountains indicate that the on population trend will depend on annual natural mortality rate of adult mule year to year interactions among these deer bucks can approach 20 percent in parameters. Also, harvest rate is lightly hunted buck populations (Pac and influenced by weather conditions that Ross 1993, Pac et al. 1995). affect animal vulnerability and by social A complicated interplay occurs changes in hunting pressure and hunter between different types of natural distribution. Our challenge is to monitor mortality that unfold in a deer population biological and social parameters that during critically important environmental define opportunity for a corresponding stress periods. For example, drought harvest rate that can direct the population during a growing season followed by a toward a defined management goal. long, severe winter often results in several For most populations, annual factors impacting deer populations fluctuations in recruitment rate were simultaneously. Deer experience a decline much more influential on harvest in physical condition because of a shorter opportunity than fluctuations in natural period of energy surplus and a prolonged For most mortality rates of adults. Recruitment period of energy deficit. Small mammal populations, rates for mule deer were generally lower populations often plummet to low levels annual fluctuations and more variable than those of white- during drought conditions because of in recruitment tailed deer, but recruitment varied across poor vegetation production. Scarcity of rate were much the diverse environments in Montana. mice, voles, and rabbits generally cause more influential on In most areas, environmental variation coyote predation to shift toward deer. harvest opportunity and its effect on energetics determine than fluctuations in fawn recruitment to a greater extent than natural mortality density of deer. Fawn recruitment was dependably influenced by deer density rates of adults. in only the most stable environments augmented by irrigated agriculture. Average annual rates of natural mortality (malnutrition, predation, disease, accidents, etc.) of adult female mule deer were low and similar (5-7 percent) among the study areas (Table 5). In variable environments, unfavorable conditions can temporarily increase these rates 3- to 4-fold or more. High mortality of adult female mule deer often occurs with the same environmental conditions contributing to low fawn recruitment. Natural mortality rates of female white- tailed deer (≥2 years) varied between 5 and 16 percent among environments (Table 6). Average annual rates of natural mortality of adult bucks of both species also were quite low (4-9 percent) but increased 5- to 6-fold under certain conditions (Table 7). However, many

124 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a The effects of predation on deer population size can only Predation loss in a deer population However, predators are capable of be understood has been traditionally considered a killing fawns regardless of their physical in the ecological compensatory form of mortality. As condition. When high predation loss in context of other inferred previously, this means that summer is followed by continuing high important factors predation would substitute for other losses to predation and malnutrition that influence types of natural mortality, and therefore during winter and spring, numbers of populations. it does not result in further decline in surviving fawns will be insufficient to population size. These conclusions were replace adult losses. based on the assumption that predators The effects of predation on deer kill only the weak, sick, and old that are population size can only be understood in part of the “doomed surplus.” However, the ecological context of other important recent studies (Gasaway et al. 1983, factors that influence populations. These Hamlin and Mackie 1989, Pac and Ross factors include annual and seasonal 1993, and Kunkel 1997) demonstrate weather patterns that affect relative that predators such as mountain lions, suitability of important habitats, forage coyotes, and wolves can also kill healthy, production, forage quality, winter prime-aged ungulates that are unlikely severity, deer physical condition, type to succumb to other forms of natural and abundance of alternate prey, type and mortality in the short term. These abundance of predators, rates of other predation losses are likely to be additive types of mortality, and relative size of the to other forms of mortality. Therefore, deer population. the potential for mortality to be additive Some authors conclude that in varies among different sex and age areas where deer and predators coexist, classes of deer and across a spectrum of the ratio between number of effective environmental conditions. Fawn losses predators and the number of deer in the to predators during severe winters would population is crucial in determining the more likely be compensatory because the degree to which predators may exert an probability is high that many fawns are effect on the deer population (Connelly in poor condition and may subsequently 1978, Keith 1983, Mech 1984, Kunkel die of malnutrition (White et al. 1987). 1997).

An Ec o l o g i c a l Pe r s pe c t i v e 125 In an excellent study on predator- that time of year. Harvest of adult does prey relationships, Gasaway et al. has a comparatively lower probability of (1983) cautioned against traditional substitution because young and prime- interpretations of a balance between aged females experience very low rates vegetation and ungulates where the of natural mortality (Hamlin and Mackie system includes effective predators: 1991, Dusek et al. 1992). Consequently, “Managers should not use survival of harvest of prime-aged females would young ungulates as an indicator of the more likely be additive to natural losses. vegetation-ungulate relationship because The rate of adult female natural predation on young obscures this mortality in harvested populations did relationship.” In this and other studies not decline with significant increases (Mech and Karns 1977, Hamlin and in harvest rate. In other words, adult Mackie 1989, Ballard et al. 1991, Boertje female harvest during autumn did not et al. 1996), predation perpetuated increase survival of remaining adult population declines that were initiated females during the subsequent winter. by other factors causing ungulates to This relationship is graphically portrayed reach or remain at low densities even in Fig. 61A for mule deer does in the after a return of favorable environmental Missouri River Breaks. The low and conditions. narrow span of natural mortality rates Our results also indicated that of adult does occurred across a wide reducing population density through spectrum of hunter harvest rates ranging hunter harvest will not dependably from near zero to 30 percent. Therefore, stimulate higher rates of fawn it is not surprising that annual survival recruitment. As previously discussed, this rates of adult females declined in a linear The buck segment is most apparent in variable environments relationship as hunting mortality rate of most deer where deer experienced drought and increased in this population (Fig. 61B). populations is winter energy deficits and coexisted with However, the influence of any change in harvested at rates effective deer predators. However, in female survival rate on population trend up to 10 times that riverbottom-agricultural environments, depended on the recruitment rate of of adult females. hunter harvest can stimulate socially- incoming one-year-old females. based, density-dependent increases in The proportion of hunting mortality fawn recruitment where populations exist of adult females that is compensatory on a high nutritional plane yearlong and (substitutes for other losses) would have coexist with low numbers of effective deer little impact on population trend. In predators. contrast, the additive component of the Biologists traditionally believed that harvest has the potential to influence deer killed by hunters would substitute trends in population size. This potential for deer that would otherwise die of can only be evaluated in relation to rates natural causes. In practice, traditional of natural mortality and fawn recruitment theory implied that substitution operated in the population at the same time. For on nearly a 1:1 basis because of intense example, even though a substantial competition for limited food on winter proportion of adult female harvest may be ranges. Obviously, some proportion of additive to other mortality, a population hunting mortality will replace natural will not decline if fawn recruitment mortality. The question is how much? is adequate to replace the combined Our results indicated that opportunity losses. Under conditions of very low for substituting hunting mortality for fawn recruitment, any hunting mortality natural mortality varies among species, of adult females in combination with populations, and sex and age classes. In other losses could initiate or intensify a most deer populations, hunter harvest population decline. of fawns has the highest probability of The buck segment of most deer substituting for mortality experienced populations is harvested at rates up to 10 over winter because fawns experience times greater than adult females (Tables higher natural losses than adults during 6 and 7). We did not observe high rates

126 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Figure 61. 1.00 A Relationships 0.90 between annual natural mortality rate and annual 0.80 hunting mortality of adult female 0.70 mule deer (A) and annual survival 0.60 rate and annual hunting mortality 0.50 rate of adult female mule deer (B) in 0.40 the Missouri Breaks (after Hamlin and

Natural Mortality Rate 0.30 Mackie 1989).

0.20

0.10

0.00 .10 .20 .30 .40

1.00 B

0.90

0.80

0.70

0.60

0.50

0.40

Annual Survival Rate Annual Survival 0.30

0.20

0.10

0.00 .10 .20 .30 .40 Hunting Mortality Rate of natural mortality among bucks in the higher natural mortality rates than adult populations we studied because high females. Pac et al. (1991) and Pac and harvest rates precluded opportunity for Ross (1993) reported that predation loss most adult males to die of natural causes. was apparently higher among all age In the absence of substantial harvests, classes of bucks. Also, bucks became however, adult bucks may experience vulnerable to winter mortality at an

An Ec o l o g i c a l Pe r s pe c t i v e 127 earlier age than females. Consequently, numerically stable while experiencing a compared to those rates listed in Table 7 higher mortality rate than adult females. for heavily exploited buck populations, The degree to which males can be natural mortality rates of adult bucks harvested more heavily than females would likely increase under special buck depends on the fawn recruitment rate and management strategies that expand age the post-season male:female ratio the deer diversity. manager would like to maintain. Fawn Montana deer populations contain recruitment and adult mortality rates fewer bucks than does. Therefore, are not stable, and averages are seldom recruitment of equal numbers of male and applicable in variable environments. female fawns adds proportionately greater However, it is important to understand numbers to the male segment. This how variation in fawn recruitment and means that the male segment can remain adult mortality can influence trends in availability of the two sexes. Relationships between recruitment and adult mortality are modeled in Fig. 100 A 62. The importance of these relationships 90 is not in their absolute precision, but in

80 s providing insight to the probable outcome le a m of harvest rates relative to recruitment in 70 fe lt u a particular year. Values of recruitment 60 ad f o and mortality that intersect to the left of rs 50 e the female curve (Fig. 62A) correspond to b m u 40 n increases in the female population. Values le b to the right of the female curve initiate 30 a St a decline. For example, if 40 fawns:100 20 Recruited fawns : 100 Females females are recruited and total mortality 10 of adult females from all causes exceeds

0 17 percent, numbers of adult females 100 20 30 40 50 60 70 80 90 will decline. In most Montana mule and % Total Annual Adult Female Mortality white-tailed deer populations, recruitment is almost never high enough to maintain stability in the female segment when total 100 s s s s e e B e l l le al a a a adult female mortality, including harvest, m m m e e em 90 f f f fe 0 0 exceeds 33 percent. 0 0 0 00 0 1 1 1 80 : : 1 At 30 fawns:100 females recruited in s s es: s: le le l e a a a l spring, a sex ratio of about 15 bucks:100 70 m m m ma 0 0 0 6 4 2 0 n 1 does can be maintained when annual buck 60 in i in a ta a in t n t n i in ta mortality is about 50 percent (Fig. 62B). i a a n a M i 50 M M a M At the same recruitment rate, a buck:doe 40 ratio of 40:100 could be maintained if

30 buck mortality was reduced to about 30 percent. This graph is most applicable to 20 populations experiencing low adult female Recruited Fawns : 100 Females 10 harvest rates or stable female populations 0 because buck:doe ratios are influenced 100 20 30 40 50 60 70 80 90 by relative differences in rate of harvest % Total Annual Adult Male Mortality of the sexes. Also, relationships in the graphs are based on the assumption that Figure 62. Relationships between fawn recruitment and total annual density dependence and compensation mortality of adult females that maintain stable numbers of does are not significant influences on fawn in the population (A) and fawn recruitment and total annual recruitment and adult mortality. mortality of adult males that maintain stable buck:doe ratios in the population (B) (after Hamlin and Mackie 1989).

128 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Deer Management in Specific Ecosystems

All wildlife populations are specific deer populations into the general products of the land, inextricably tied to management categories we describe in the habitat that supports them (Leopold subsequent sections. 1933). Mule deer and white-tailed deer Deer population response to fill different niches, even locally. Mule fluctuation in different environments is deer tend to occupy rugged, broken portrayed in pendulum graphs in the topography with more variable climatic following section (Figs. 63-66) . In conditions, and white-tailed deer tend to Montana deer periodically experience occupy more gentle topography. East unfavorable conditions that correspond of the Continental Divide, whitetails to a long energy deficit (severe winter) are often associated with riparian and short energy surplus (poor growing communities and agricultural croplands. season). This results in declining West of the Continental Divide, white- physical condition of deer and increases tailed deer occur in association with in predation and winter mortality. Under riparian components of coniferous these conditions populations enter a low forests. output phase characterized by minimal We realize that the environmental fawn recruitment and relatively high settings occupied by deer are endlessly natural losses of adults. During this variable, and any attempt to categorize phase opportunity for hunter harvest is them will include numerous deficiencies. limited unless reduction in population size Nonetheless, we believe the potential is desired. benefits of aligning deer management Short periods of energy deficits and goals and strategies with differences in long periods of energy surplus result in population dynamics across the ecological improved physical condition. With this spectrum will outweigh potential the population enters a high output phase problems (Table 8). Biologists can fit characterized by high fawn recruitment and low natural mortality of adults. Table 8. Higher rates of hunter harvest are then Major environment types occupied by required to stabilize population size. mule deer and white-tailed deer in We assimilated information from Montana. Tables 4-6 and Fig. 62 to describe Mountain Ecosystem interrelationships between fawn Mountain-foothill Environment recruitment, adult female natural Northwest Montane Forest Environment mortality, hunter harvest of does, and Prairie Ecosystem their combined effect on adult female Timbered Breaks Environment population trend. The amplitude of the Prairie-Badlands Environment pendulum swing (Figs. 63-66) represents Prairie-Agricultural Environment the outer limits of expected variation Riverbottom Agricultural Ecosystem in these parameters as the population Plains Riverbottom Environment responds to fluctuations in a given Intermountain Valley Environment environment. The solid pendulum bars

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 129 White-tailed deer predominate in densely-timbered river valleys and maritime-influenced mountain ranges in western and northwestern Montana. Examples include the Salish Mountains, and the mountains and valley bottoms associated with the Swan, Fisher, and Blackfoot Rivers. White-tailed deer also are locally abundant in foothills and along drainages into some mountain ranges east of the Continental Divide where deer have access to agriculture on a seasonal or yearlong basis. Examples include the Big and Little Snowy Mountains, Bears Paw Mountains, Rosebud Plateau, and portions of the Moccasin and Judith Mountains. However, little is known about the population dynamics of white-tailed deer in this ecological setting. Additional investigations are needed to adequately delimit the ave. ± 1 std. deviation in describe population parameters and population parameters while the dashed harvest relationships of whitetails in the pendulum bars represent highest and mountain-foothill environment east of the lowest recorded values. Adult female Divide. harvest rates indicated at each level of output would stabilize numerical growth in the female population. The graphs Description of Deer Population provide general guidelines for selecting Ecology adult female harvest rates that may Both species congregate on their direct populations in various ecosystems respective winter maintenance habitats in and environments toward desired the lower foothills and river valleys where management goals. deer occur at densities of 6-60 deer/km2. In summer, deer exploit both adjacent and distantly located reproductive habitats Mountain Ecosystem at higher elevations where densities vary from 1-15 deer/km2. In mountainous terrain, both species The mountain ecosystem contains of deer contend with an energy deficit populations of both species of deer that during winter that is often of longer collectively occupy most of the western duration than experienced by deer in third of Montana including essentially other environments. During winter, the all of MFWP Regions 1, 2, and 3, and primary survival strategy emphasizes mountain ranges in Regions 4, 5, and 6 specialized use of habitat to conserve (Fig. 1). fat reserves. Duration and severity of Mule deer predominate in drier the deficit varies from year to year. It mountain-foothill environments such can also vary locally depending on as the Garnet Mountains west of the elevation and relief of the local landform Continental Divide and most mountain and climate, especially the relative ranges east of the Divide. Mule deer are tendency to accumulate snow. East of also found in the higher elevation portions the Continental Divide, some mountain of mountain ranges in the northwest ranges occupied by mule deer occur in montane forest environment. the “chinook zone” where snow cover periodically melts during the winter

130 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a months. Other mountain ranges outside coming winter during a comparatively this zone retain their low-elevation short summer growing season. To snowpack throughout the late autumn- be successful, most deer in mountain spring period. environments employ specialized Delayed snowmelt and plant growth migratory movements that are passed are typical of some mountain areas where from one generation to the next by social high elevation ranges comprise much of bonds within matrilineal groups. In mountainous the summer habitat. In these situations, The relationship between habitat and terrain, both mule deer populations experience deer abundance is far more complex in species of deer “bottlenecks” in seasonal distribution. mountain environments than previously contend with an Compared to lower elevation mountain understood. Specialized behavioral energy deficit ranges, deer exploiting high elevations responses to severe environmental during winter spend more time confined to winter range conditions contrasted with the traditional that is often of and as much as 70 fewer days on their concept of predicting deer numbers longer duration summer home ranges (Pac et al. 1991). simply from the amount of forage than experienced White-tailed deer occupying the maritime available in winter. We also were unable by deer in other influenced mountain ranges in northwest to document a relationship between adult environments. Montana usually experience less variation density and fawn recruitment (Mackie et in precipitation and temperature during al. 1990, Pac et al. 1991). While critical summer and winter. to the occurrence of deer in mountain In all types of mountain environments, winter range was not the environments in Montana, deer must primary factor determining deer numbers recover physical condition, reproduce, and dynamics. and accumulate fat reserves for the

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 131 In most mountain ecosystems, deer deer in the presence of predators. persist under the potential influence of However, the increasing trend could end several effective deer predators. These in the face of a catastrophic winter. If commonly include coyotes, mountain adult female harvest rates were high in lions, black bears, and bobcats. A few conjunction with high predation and poor mountain environments also include fawn recruitment, a significantly lower grizzly bear and/or recolonizing wolf population could persist for a time even populations. after a return to favorable environmental Each of these predators tends to conditions. exploit prey in a characteristic manner Mule Deer Vital Parameters and and each is most effective in a particular Harvest Recommendations —Mule deer environmental setting. The mountain fawn recruitment rates (ave. = 33 ± 13 lion and wolf occur at relatively low fawns:100 females) in mountain-foothill abundance and are capable of killing environments are low compared to any sex or age class of deer regardless populations occupying other ecosystems of size or condition (Kunkel 1997). (Table 4). Somewhat higher recruitment Comparatively coyotes may occur in rates may be observed in some mountain higher densities in some mountain ranges where the period of energy deficit environments. However, coyotes are is shorter or deer have seasonal access to more opportunistic feeders such that agricultural crops. However, recruitment predation on deer may vary substantially rates from the Bridger Mountains and unpredictably from year to year. Less incorporated estimates of fawn mortality is known about the potential impacts of through mid-May. On other areas, black bears and bobcats on deer. They classification data often miss significant were not significant in killing mule pulses of fawn mortality that periodically deer older than six months of age in occur after March. the Bridger Mountains (Pac and Ross Natural mortality rate of adult 1993). Their effectiveness at killing females (7 ± 6 percent, Table 5) was deer fawns between birth and 6 months similar to that of mule deer in timbered of age has not been studied in mountain upland breaks or prairie-badlands. The environments. However, black bear have interaction of fawn recruitment and The combined been documented as substantial predators natural mortality of adult females presents effects of multiple of newborn elk calves in mountain a rather narrow opportunity for hunter predators could environments (Schlegel 1976). harvest of adult females (Fig. 63). The exert greater and Potential exists for predation low output phase corresponded to more consistent to influence population trend in the recruitment rates ≤ 20 fawns:100 females predation pressure mountains of western Montana. The and natural mortality rates ≥ 14 percent compared to other combined effects of multiple predators among females. A goal of maintaining environments with could exert greater and more consistent reasonable stability in the doe population fewer effective deer predation pressure (MacNab 1985) leaves little opportunity to harvest adult predators. compared to other environments with females during lower output phases. fewer effective deer predators. In the Harvest opportunity is restricted because North Fork of the Flathead River, Kunkel of the low number of incoming recruits. (1997) indicated that the white-tailed deer If the goal is to reduce the population, population was declining because of low however, the low output phase is an fawn recruitment caused by the additive excellent time to harvest adult does. effects of multiple predators. When the population enters a White-tailed deer populations in high output phase during favorable northwest Montana increased to an environmental conditions, female apparent record high level during the harvest rates of 15-21 percent are last 20 years. This probably resulted required to keep it from increasing. from favorable habitat changes, mild This corresponds to recruitment of ≥ 45 winters, low hunter harvest rates, and fawns:100 females and natural losses possibly a numerical advantage favoring of does ≤ 3 percent. The population

132 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Figure 63. Recommended mule deer doe harvest rates across the span of expected variation in fawn recruitment and natural mortality of adult females in a mountain-foothill environment.

varies across a span of moderate output improved population monitoring and a during about 6 of every 10 years. During more integrated process for implementing these years, it can support a doe harvest appropriate regulations in response of 1-15 percent. However, mule deer to significant changes in population populations can exhibit considerable status. The margin of error is small variation in the length of time they remain in managing female mule deer harvest ...the public within a particular output phase. Also, on vast expanses of accessible public demands improved other situations can occur where fawn land where high numbers of elk hunters detection of recruitment and adult female mortality result in increased hunting pressure population declines rates on the pendulum graph do not align on deer (Hamlin and Erickson 1996). and expects a more with each other. Conservative harvests should be applied timely harvest Special Mule Deer Population to these areas where female harvest is not management Management Issues—In recent years, required to solve game damage problems. response. public controversy concerning mule A large population of productive females deer management has centered in the is required to maximize buck fawn yield to mountain-foothills where hunting is sustain the heavy buck harvest rates that focused on public lands. Two issues are prevail in these settings. Recolonization of primary importance. will occur slowly in areas that have been First, the public demands improved heavily harvested because recruitment detection of population declines and emigration rates are low. This and expects a more timely harvest problem becomes critical when it involves management response. This will require population segments with specialized

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 133 migratory movements to distantly located In northwest montane forest summer ranges (Kasworm 1981, Pac et environments, the natural (non-hunting) al. 1988). Management effectiveness will mortality rate of whitetail does is 0.10 (95 be enhanced if hunting district boundaries percent C.I. 0.07 - 0.13) (Sime unpubl.). are aligned to the ecological boundaries This estimate represents a cumulative rate of individual populations. using the analysis technique MICROMORT Second, creative strategies are (Heisey and Fuller 1985) for all radio- needed to improve hunter opportunity to collared females monitored between 1988 harvest older, larger-antlered mule deer and 1995. bucks. Many mule deer populations in Similar to the relationships shown Regions 2 and 3 have few bucks older for other populations, white-tailed deer than 3 years. Any buck management in northwest montane environments strategy that strives to increase age probably fluctuate between low, moderate, diversity will require reducing annual and high output phases. To make harvest hunting mortality from the prevailing 50- recommendations prior to a thorough Management 70 percent to about 35 percent or less. analysis of research data, we suggest effectiveness will This will probably require a substantial that the cumulative natural mortality be enhanced if reduction in the number of hunters afield. rate of 10 percent could be interpreted hunting district To produce bucks with large antlers, as the natural mortality rate when the population is in a moderate output phase. bondaries are the age structure must include bucks ≥4 aligned to years. However, natural mortality rates The 95 percent confidence interval would probably represent upper and lower the ecological rise with age and bucks seldom survive bounds of variation in natural mortality boundaries beyond their seventh year. while the population is operating in high of individual Some habitats may provide greater or low output phases. opportunity than others for sustaining populations. When northwest montane white- older mule deer bucks. Thus, increasing tailed deer populations are in the the availability of mature bucks will be moderate output phase, defined by a most efficient when directed at specific recruitment rate of 67 fawns:100 females populations or hunting units. Further and a natural mortality of 10 percent, a discussion of buck population trends and 15 percent harvest rate of adult females review of some management strategies would be necessary to prevent increases is provided in the Draft Environmental in numbers of females. During the high Analysis for establishing the 1997 Deer output phase with recruitment of 85 Hunting Season (Hamlin and Erickson fawns:100 females and 7 percent natural 1996). Buck:doe ratios and antler mortality, a 23 percent adult female characteristics by age class in different harvest rate would prevent population environments are also discussed in that growth. When populations are in the document. low output phase with recruitment of 49 White-tailed Deer Vital fawns:100 females and 13 percent natural Parameters and Preliminary Harvest mortality, a 7 percent harvest rate of adult Recommendations —The northwest females would be sufficient to prevent Montana white-tailed deer project population growth. is ongoing, and data have not been Annual mortality of whitetail completely analyzed in a manner allowing bucks on the Salish Mountains study construction of a pendulum graph. area averaged 42 percent compared Nonetheless, data are available to make to 61 percent in a prairie-agricultural preliminary harvest recommendations. environment (Table 7). In the northwest Sime (1996) reported that the white- montane forest, annual mortality tailed deer fawn recruitment rate in attributed to hunting averaged 33 percent the Salish Mountains (ave. = 67 ± 18 and natural mortality accounted for 9 fawns:100 females, Table 4) was higher percent. Relationships portrayed in than the recruitment rate for mule deer in Fig. 62 indicated that buck:doe ratios mountain-foothill environments. approaching 50:100 could be maintained

134 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a when recruitment averaged about 65 is necessary to stabilize numbers of fawns:100 females. adult females in whitetail populations in The severe winter of 1996-97 montane environments. represents the lowest recruitment rate ever observed for this population. At Habitat Management in Mountain the time of measurement in March 1997, Ecosystems the recruitment rate was 30 fawns:100 females. However, actual recruitment Management recommendations likely fell below that value because concerning land use issues will be deer were confined to winter ranges for most effective when they are based on another two months. Migrant deer did knowledge of yearlong deer habitat not begin leaving the winter range until use, seasonal distribution, and specific mid-May and some stayed until late May. resource requirements of an individual Incidental observations gathered in April population. Requirements of both species and May indicated that the fawn:adult in winter emphasize conservation of fat ratio continued to decline. reserves. White-tailed deer may use Because of the severity and duration agricultural crops where available, and of the 1996-97 winter, a separate use varies from year to year depending on accounting of adult female natural winter severity (Kamps 1969). Whitetails mortality was made between the close usually rely on an energy conservation of the general hunting season December strategy during harsh winters and are 2, 1996, and June 1, 1997, for those more opportunistic during mild winters. females marked previously and whose In summer, reproductive requirements radio collars were still functioning June are met by using a diversity of vegetation 1. In that 6 month interval, 26 percent of communities centered in riparian areas. the radio-collared adult females died. On Housing Developments—Many an annual basis, natural mortality during western Montana deer habitats occurring 1996-97 must have exceeded 26 percent on private land have been subdivided, and was probably at least 31 percent, particularly in MFWP Regions 1, 2, and 3. assuming that at least 5 percent died We recommend that biologists be involved during the remainder of the year. as much as possible in city and county When recruitment falls to 30 planning efforts. Accurate information fawns:100 females or less and adult on location of seasonal ranges, movement female natural mortality is 26 percent patterns, and population size can or greater, all adult female harvest is significantly influence outcomes in the precluded if the management goal is to planning process. Cluster development maintain the population at the current can preserve open space and minimize level. Given these values, the population human disturbance. Conservation is already in decline, regardless of the easements could be pursued on priority antlerless harvest rate, because natural areas under private ownership. mortality exceeds recruitment. For mule deer winter ranges east of At the opposite end of the spectrum, the Continental Divide, housing densities some years appear to have been very of <2 homes/km2 caused relatively little favorable for whitetail populations in conflict between deer and people other northwest montane forest environments. than periodic game damage complaints The highest observed recruitment rate in these rural agricultural settings. At was 96 fawns:100 females. In the best of housing densities of 3-8 homes/km2, the years, adult female natural mortality could winter range becomes a complicated reasonably be estimated at 5 percent. mix of urban and agricultural land uses. This value is similar to other study Segments of the local deer population populations with actual annual mortality adapt to the new cover and forage sources estimates for the most favorable years. associated with housing development Under such highly favorable conditions, and become somewhat conditioned a 27 percent adult female harvest rate to human presence. Deer damage

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 135 complaints accelerate, and effective providing easier access to foraging solutions such as sport hunting decline. sites (Youmans 1979). White-tailed Deer are considered a nuisance in these deer consistently preferred overstory situations by some individuals and a canopy coverages > 50 percent and resource by others. This creates conflict showed no preference for any overstory between neighbors and between housing canopies < 41 percent during severe development residents and the statewide winter conditions (Baty 1995). During wildlife agency (Baker and Fritsch 1997). storm-free winter conditions, mule Vogel (1989) indicated that densities deer preferred open Douglas fir stands of >12 homes/km2 virtually preclude often associated with steep shrub fields use by deer in an intermountain valley and rock outcrops that acted as “solar environment. radiators” (Steerey 1979, Youmans 1979). Timber Management—Timber Our studies and those of Baty management on winter maintenance (1995), Stansberry (1996), and Baty habitat should emphasize retention of et al. (1996) found no evidence that conifer forest stands because of their commercial thinning of conifer canopies importance when deer habitat use or reduction in understory conifer strategies emphasize conservation of density was desirable on winter ranges energy reserves. Mature conifer forest where deer frequently contend with deep is often preferred by both species during snow conditions. Baty et al. (1996) deep snow conditions and severe weather hypothesized that white-tailed deer in the northwest montane forests and population productivity could decrease some mountain-foothill ranges (Pac et due to reduced cover and increased al. 1991, Baty 1995). In some areas, energy expenditures and recommended availability of conifer cover in winter may against broadscale silvicultural practices be essential to the occurrence of deer. that create open savannah-like stands of Conifer cover ameliorates temperature mature ponderosa pine on white-tailed extremes, and reduces wind velocity and deer winter ranges west of the Continental radiant heat loss. Snow depth under Divide. However, he indicated that such the conifer canopy is changes my be neutral for mule deer also minimized, and beneficial for elk. Arno et al. (1987) recommended coupling prescribed fire with small partial cutting units (0.1-0.6 ha) to perpetuate the overstory canopy and stimulate forage production on whitetail winter ranges in the northwest montane forest. Timber management to optimize deer habitat and maintain or increase deer numbers on summer ranges should emphasize perpetuation or enhancement of habitat diversity. For mule deer in mountain- foothill environments, mature (150-300 years) Douglas fir stands provide critical fawn- rearing habitats (Pac et al. 1991). These often uneven- aged stands have irregular

136 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a canopies and structure that promote a Road management guidelines must patchwork of understory diversity. High be tailored to specific objectives for quality forage resources and security managing deer in a particular area. If for fawn rearing occur in optimal maximum deer harvest is appropriate, combination in topographically complex extensive open road networks may forest communities. be beneficial. Restrictions on use of For mule deer in mountain ranges open roads and limitations on building east of the Divide, timber management new roads could be incorporated in should avoid large scale removal of these strategies for improving or maintaining stands to ensure that regenerating stands buck deer age diversity. Otherwise, are allowed to reach the age where these deer population objectives will continue attributes are replaced. Cutting units to be accomplished through significant that focus on large, even-aged stands of reductions in hunting opportunity. lodgepole pine could benefit mule deer. Morgan (1993) reported that Small, irregular shaped cutting units roads in habitat used by white-tailed (0.5-2.0 ha) will maximize edge effect deer during summer-early autumn in and minimize reduction in habitat security northwest Montana, did not negatively when loosely distributed across large affect deer distribution and use except stands. immediately adjacent to roads. Closing For white-tailed deer occupying roads in the most preferred habitats could northwest montane forests, Morgan be beneficial to deer in the immediate (1993) concluded that riparian areas vicinity. and adjacent uplands containing pole/ Habitat Enhancement and immature timber were very important Vegetation Manipulation—During the as centers of deer use in summer. Many last 50 years, extensive effort has been of these habitat complexes were found directed at vegetation manipulation to between 1,100 and 1,700 m in elevation increase the amount of forage available on east to southeast aspects on the Tally on winter ranges in the western United Lake Ranger District. Cutting units would States. Numerous studies show positive have minimal effect on white-tailed deer if responses in ungulate distribution and located on northerly and westerly aspects, use related to habitat manipulation ridgetops, and other sites more than 750 (Anderson et al. 1974, Riggs and m from riparian habitat. Peek 1980, Bentz and Woodard 1988, Road Management—Road building Klinger et al. 1989). Stansberry (1996) proliferated in many western Montana evaluated habitat enhancements on mule mountain ranges with the rapid expansion deer and bighorn sheep winter range of timber harvest on public and private adjacent to and above the area inundated lands during the last 50 years. Road by the formation of Lake Koocanusa building represents a two-sided issue for in northwest Montana. He concluded deer management in mountain ranges that forage production was increased east and west of the Continental Divide. by slashing, burning, and fertilization. Access is necessary for even distribution Although animal use apparently increased of the deer harvest. However, too much following the treatments, increases of access on public lands contributes numbers, productivity, and survival of significantly to low escapement by bucks deer or bighorn sheep were not observed. These results generally agreed with other during the hunting season. Unfortunately, studies (Cook et al. 1989, Klinger et al. most road building was conducted 1989, Stussy 1993). We recommend that without considering security needs of deer habitat improvements in mountain deer during the hunting season. This environments focus on acquisition or problem is acute in areas where mule deer easements that protect the ability of deer and elk distribution overlaps and hunting to use winter ranges in an undeveloped pressure is high (Hamlin and Erickson and undisturbed condition. Land use 1996). practices that eliminate important

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 137 vegetation communities such as multi- Montana, recreational hunting of deer is aged stands of conifer forest (Baty an activity of significant economic value. et al. 1996) or sagebrush/grassland communities (Peterson 1996) can be Mule Deer Population Ecology in expected to have negative consequences. Timbered Breaks Environments Livestock Grazing—Scientific studies of the relationships between Our knowledge base concerning livestock grazing and deer populations mule deer population ecology in timbered have not been conducted in mountain breaks environments (Hamlin and Mackie environments in Montana. Specifically, 1989) may be more complete than for relationships between cattle grazing, either species in other parts of Montana. small mammal abundance, and coyote Mule deer density on the Missouri River predation on mule deer require further Breaks study area varied between 1.4 study. and 6.2 deer/km2 during a 27-year period Cattle grazing on summer ranges (Hamlin and Mackie 1989). Elsewhere, occurred primarily on drainage bottoms higher densities occurred locally where and areas of low topographic relief. timbered breaks included agriculture. These areas were often selected by Mule deer density in an upland pine/ white-tailed deer, resulting in potential agricultural area in the lower Stillwater for competitive interaction. Spatial River drainage of southcentral Montana overlap with mule deer was limited by varied from 7-19 deer/km2 during 1979- their preference for more rugged terrain. 97 (Stewart, personal communication However, intense summer-long livestock 1997). grazing over large areas of diverse Individual deer displayed either Douglas fir communities also could resident movement patterns or short decrease habitat quality for mule deer. distance seasonal migrations. Fawn- rearing habitats closely interspersed with small, scattered inclusions of winter Prairie Ecosystem habitat served the needs of family groups of resident deer. Seasonal migrants made short distance connections between Prairie extends across much of summer habitat in upper coulees with the eastern two thirds of Montana, wintering areas in rough topography at dominating significant portions of the lower end of coulees. Regions 4 and 5 and the majority of In the timbered breaks, significant Regions 6 and 7 (Fig. 1). Included in variation in annual precipitation and the prairie ecosystem are three distinct temperature results in dramatic changes environments: timbered breaks, prairie- in forage production and the duration of badlands, and prairie-agricultural. succulence during the growing season. Landforms vary from flat to highly The effects of these environmental factors eroded, rugged badlands. The climate on forage quantity and quality were much is semi-arid and volatile fluctuations more influential than browsing by deer. occur during all seasons. Vegetation An equilibrium relationship (carrying is characterized by grassland, shrub/ capacity) between deer and forage grassland communities with hardwood supplies seldom unfolded in a definable draws or other woody cover along or predictable manner. However, Hamlin some drainages, and pine-dominated and Mackie (1989) indicated that fawn dendritic “breaks.” Land ownership is survival to December could be predicted predominantly public in the timbered from the same factors that significantly breaks and mostly private in prairie- predicted forage quality. These factors badland and prairie-agricultural were precipitation during July-April prior environments. Dryland agriculture and to the growing season and temperatures livestock grazing represent the primary during May. economic land uses. As in all parts of

138 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a In the timbered breaks...An equlibrium relationship (carrying capacity) between deer and forage supplies seldom unfolded in a definable or predictable manner.

Although weather conditions are This occurred because coyotes displayed highly variable, winter is usually over by flexible and opportunistic food habits mid-March. Length of the winter deficit (Schladweiler 1980). Increases in coyote period is about one month shorter than predation of mule deer often occurred experienced by mule deer in mountain during drought conditions or other environments. periods when small mammal populations The classic logistic relationship were low. Some of the predation loss between population density and fawn was additive and significant in its effect, recruitment rate did not predictably particularly when the deer population operate in this variable environment. was low as a result of a combination of During favorable environmental mortality factors. conditions, fawn recruitment could be Mule deer populations in the excellent at high population density. It timbered breaks are capable of quickly ...mortality rates could also be poor at minimal population rebounding from population lows. Their can be nutrition density if conditions were unfavorable numbers have doubled in 2 years when related but not for fawn survival. Hamlin and Mackie environmental conditions were favorable. necessarily related (1989) indicated that mortality rates can Lush summer growing seasons, abundant to density of deer be nutrition related but not necessarily small mammal populations that serve as related to density of deer in the alternative prey for coyotes, and mild in the population. population. winters contribute to population recovery. In timbered breaks, mule deer Deer numbers can also decline by 50 coexisted with the coyote as the only percent or more in years of unfavorable effective deer predator. Coyote density conditions. was apparently higher in the timbered White-tailed deer populations of breaks compared to other parts of eastern management importance occur locally Montana (Pyrah 1984). Hamlin and where inclusions of river bottom- Mackie (1989) reported that predation agricultural or prairie-agricultural rate on mule deer varied significantly environments occur within the timbered from year to year even though coyote breaks. Habitat issues concerning these population size was relatively constant. populations are similar to those discussed

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 139 in later sections concerning whitetails is limited during the low output phase in riverbottom-agricultural and prairie- unless further reduction in the population agricultural environments. is desired. Mule Deer Vital Parameters and During most years, the population Harvest Recommendations—Populations fluctuated across a wide but moderate in the timbered breaks displayed greater output phase indexed by recruitment fluctuation compared to populations in rates of 30-75 fawns:100 adult the mountains. In the former, recruitment females. Natural mortality rates of averaged 51±27 fawns:100 adult adult females during this phase are females compared to 33±13 fawns:100 generally low, varying from 1-10 percent. Populations in the adult females in the mountains (Table Corresponding adult female harvest rates timbered breaks 4). During unfavorable environmental vary from 3 to 24 percent during the displayed greater conditions in the timbered breaks, moderate output phase. fluctuations recruitment varied from 5-25 fawns:100 At high output, the population compared to adult females (Fig. 64). A significant experiences recruitment ranging between populations in the pulse of high natural mortality of adult 75 and 103 fawns:100 adult females. mountains. females often, but not always, occurred in A female harvest of 26-33 percent concert with low fawn recruitment. The is required to stabilize adult female magnitude of this pulse, which varied numbers. from 11-24 percent of adult females, Emigration is a form of population could initiate a population decline even loss that is hard to document and to in the absence of any measurable female incorporate into harvest prescriptions. harvest. Opportunity for female harvest Emigration of yearling females was

Figure 64. Recommended mule deer doe harvest rates across the span of expected variation in fawn recruitment and natural mortality of adult females in a timbered breaks environment.

140 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a significant and apparently not balanced with rough topography provided forage by immigration at the lowest and highest resources and extended the period of female densities recorded on the Missouri succulent forage availability during late River Breaks study area (Hamlin and summer and autumn. Rough terrain also Mackie 1989). Substantial emigration provided hiding and escape cover from usually occurred among yearling females natural predators and hunters. Areas with in the first two large cohorts following a a diversity of topography and vegetation low population phase. It also accounted were associated with more fawning for about 50 percent of a yearling territories and consequently a higher female cohort recruited during a high population density. population. However, emigration did not Vegetation Management—We do increase proportionally with population not recommend large-scale vegetation density. The harvest rates we indicate manipulations (chaining, cutting, on the pendulum graph (Fig. 64) would spraying, plowing, burning) that be greater than necessary to stabilize traditionally focused on increasing the female segment in core populations quantity of forage on treated areas. Our during years of high yearling female interpretation of the relative importance emigration. of habitat components is different from It is not possible Annual mortality rate of mule deer the “key species-key area” concept. A to isolate any bucks on the Missouri River Breaks study variety of topographic settings, vegetation single habitat area averaged 41 percent compared types, and other habitat attributes component that to 46 percent recorded in the Bridger were critical to mule deer survival. It was consistently Mountains and 61 percent on Cherry is not possible to isolate any single important enough Creek (Table 7). In the timbered habitat component that was consistently to warrant direct breaks environment, annual mortality important enough to warrant direct attributed to hunting averaged 37 manipulation or enhancement programs. manipulation percent and natural mortality accounted Mule deer in the Missouri River or enhancement for the remaining 4 percent. However, Breaks used all vegetation types on programs. hunting mortality rates fluctuated from the study area for some purpose at 15-58 percent and natural mortality some time (Hamlin and Mackie 1989). rates varied from 0-11 percent among Forage production and quality across individual years. Relationships portrayed all vegetation types was determined in Fig. 62 indicated that a buck:doe primarily by soil type and variation in ratio approaching 40:100 could temperature and precipitation. be maintained if total annual male mortality approximated 40 percent and recruitment averaged 50 fawns:100 females.

Habitat Management for Mule Deer in Timbered Breaks Environments Hamlin and Mackie (1989) concluded that mule deer distribution in timbered breaks was positively associated with topographic and vegetative diversity. Rough topography was of critical importance to mule deer in winter because south-facing slopes were relatively snow free, and timbered north-facing slopes provided thermal cover. Diversity of vegetation types and microclimatic conditions associated

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 141 deer. Utilization of this resource often involves local movement of mule deer from public to private land, causing game damage problems for landowners. Where this problem is chronic, a practical application of alfalfa fields on public lands could serve as a “lure crop” and reduce deer damage to private agricultural fields and haystacks, particularly during dry years. Livestock Grazing—Livestock grazing is the most common land use in timbered breaks environments. Relationships between livestock grazing and mule deer have been studied to a greater degree in this environment than in other parts of Montana, and Hamlin and Mackie (1989) reported they are complex. Deriving general that forage quantity was not limiting recommendations from these studies and during any year or season during their applying them to broad geographic areas study. Rather, variation in forage quality is problematic. (nutritional content) among years Hamlin and Mackie (1989) could not and the length of time nutritionally- detect any cause and effect differences adequate forage was available in any in mule deer fawn survival or population year was of major importance to mule trend between season-long and rest- deer physical condition and population rotation grazing systems during 1976- dynamics. Density-independent, rather 1987. Annual changes in fawn survival than density-dependent, factors exerted and recruitment that occurred under the primary influence on deer-nutritional those grazing systems were apparently relationships. synchronized to broad climatic influences In the timbered breaks, mule simultaneously affecting both areas. ...the length of deer significantly preferred Douglas fir Overlap in distribution, habitat use, time nutritionally- communities during all seasons. These and food habits of mule deer and cattle adequate forage were relatively moist types within a was minimal (Mackie 1970, Knowles was available in drought-prone environment. Douglas 1975, Komberec 1976). Cattle primarily any year was of fir types provided succulent forage and use the adjacent prairie grasslands, open major importance hiding cover during summer-autumn and flat ridgetops, and larger coulee bottoms. to mule deer thermal cover and more tolerable snow Mule deer confined their primary use to physical condition conditions during winter. All of these diverse, steep, timbered terrain located between areas used by cattle. Hamlin and and population contribute to a positive energy balance on Mackie (1989) reported that opportunity dynamics. an annual basis. In portions of the timbered breaks for competition was greatest in spring and throughout the prairie-badlands, (April-May) when distribution and food mule deer successfully occupy many habits exhibited the greatest overlap. areas without Douglas fir communities. Deferral of spring cattle grazing until late Other communities (e.g., ponderosa May would give mule deer exclusive use pine/juniper) obviously provide a similar of sites that experience the earliest green function. Mule deer will shift their habitat up. use to agricultural areas when succulent To minimize cattle use of areas forage becomes scarce in native plant important to mule deer, we recommend communities (Knapp 1972, Griffiths against development of stock water 1990, and Olenicki 1993) . Agricultural sources at the terminal portions of large crops such as alfalfa increase the time ridges or on smaller ridges within the high quality forage is available to mule timbered breaks.

142 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Mule Deer-Elk Interactions—Elk western Montana. In most areas, access populations have increased significantly to public lands in timbered breaks has in portions of the timbered breaks been quite compatible with mule deer environment. Only limited information harvest management strategies. Hunting is available to evaluate the effect of pressure can be expected to increase in increasing elk numbers and distribution the future on these lands as more hunters on mule deer. Hamlin and Mackie (1989) turn away from congested areas and concluded that the mule deer population restrictive seasons associated with the reached two all-time peaks despite a mule deer hunting experience on public three-fold increase in elk numbers during lands in the western part of the state. the previous 20 years. A third, apparently We recommend that travel higher peak in mule deer numbers management plans be creatively devised occurred in 1993-94. It is unknown for important areas in the timbered whether further increases in elk numbers breaks before problems related to heavy will negatively influence mule deer hunting pressure become chronic. A populations. series of areas on public lands in eastern Significant differences in habitat Montana could be strategically selected preferences for the two species enabled for maintaining and improving age both to use the Missouri River Breaks diversity in populations of mule deer study area with minimal conflict. Use of bucks. Accomplishing deer management vegetation cover types differed during all objectives with carefully conceived hunter seasons. Mule deer used areas of greater access plans may be preferable to making topographic relief than elk (Hamlin and major changes in the hunting season Mackie 1989). structure once the problem becomes Different grazing systems and cattle chronic. stocking rates could influence habitat Progress in maintaining and relationships between mule deer and expanding hunter access to privately elk. Campbell and Knowles (1978) owned portions of the timbered breaks indicated that elk selected rested pastures and prairie-badlands environments compared to grazed pastures in a rest- continues under MFWP Block rotation system in the Nichols Coulee area Management and Habitat Montana of the Missouri River Breaks. Elk were programs. These efforts represent able to make these spatial adjustments in important steps in expanding hunting their habitat use because of larger home opportunity and help buffer the increase ranges and greater mobility than deer. in hunting pressure on public land. The shifting of elk away from areas used by cattle could result in greater spatial overlap between elk and mule deer, Mule Deer Population Ecology in particularly if elk made greater use of Prairie-Badland Environments rough terrain. The more limited mobility Prairies represent patchy of mule deer made them more vulnerable environments for mule deer. Preferred to situations where other ungulates (cattle habitats occur as relatively small or elk) used their home ranges (Knowles inclusions within large areas receiving 1975). little or no use. Therefore, population Access Management—Large density is generally low overall. Although blocks of public land occur within the Wood et al. (1989) documented densities timbered breaks environment, particularly of 0.3-3.0 deer/km2 on the Cherry Creek along the Missouri River. Much of this study area during 1975-1987, higher expansive area is generally accessible densities were recorded in small areas of to public hunting by a network of preferred habitat often associated with unimproved roads and trails. The distant agriculture. location of these areas from large urban Most mule deer populations in centers has not resulted in the intense prairie-badland environments are closely hunting pressure experienced in parts of associated with rugged badlands or

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 143 non-timbered breaks. Availability of appeared similar to other non-agricultural badlands in units larger than 4 km2 during prairie environments. winter provided minimum snow depth, Studies of predator-prey shelter from wind, preferred forage, and relationships between mule deer and security from predation (Dusek 1975, coyotes, the most effective deer predator Geist 1981, Wood 1987). In many in this system, have not been conducted. prairie-badlands environments, mule Coyote densities apparently were lower in deer depend on native forage species shrub-grasslands compared to timbered during winter. Energy available in native breaks (Pyrah 1984). forage is often less than daily energy When fawn recruitment was high requirements. Consequently, a strategy and hunting mortality of adult females emphasizing energy conservation during was low, Wood et al. (1989) documented severe weather and foraging during a 600 percent increase in mule deer mild conditions was vital to overwinter population size in 8 years. When the survival. In the absence of conifer converse was true, deer numbers declined vegetation, mule deer used topographic by 70 percent in 4 years. features to help conserve energy (Wood Mule Deer Vital Parameters and 1988). Harvest Recommendations —Variation in Winter severity and associated fawn recruitment and natural mortality restrictions on habitat availability and of adult females in prairie-badlands use did not consistently limit fawn environments was similar to that recruitment or adult survival (Wood et al. described for mule deer populations in 1989). Rather, environmental conditions timbered breaks. Consequently, the range prior to and during the growing season of doe harvest rates required to stabilize appeared to exert primary influence on adult female numbers in the prairie- mule deer population dynamics. badlands was also comparable (Fig. During summer, springs, swales, 65). Net recruitment in prairie-badlands and creek bottoms preferred by adult environments was described from data females during summer provided spanning 12 years; mortality rates for succulent forage and other resources females was based on only 5 years of important to fawn rearing. Annual data. variation in precipitation and temperature Unfavorable environmental resulted in wide fluctuations in forage conditions in prairie-badlands were production and the period it remained associated with recruitment rates of 10-30 succulent and nutritious. Wood et al. fawns:100 adult does and natural adult (1989) reported a positive correlation female losses between 10-12 percent. between fawn recruitment rate and total During low output phases, doe harvest precipitation from July-April prior to rates compatible with stable numbers in fawning. Mackie et al. (1990) reported the female segment varied from less than Where prairie a statistically significant relationship 1 to not more than 3 percent. mule deer had between increasing adult density and In most years, recruitment varied yearlong access decreased fawn recruitment the following between 35-75 fawns:100 adult females to agricultural year. However, this low fawn recruitment while natural losses of does were 2-9 crops, apparently also occurred coincident with a decrease percent. Associated with this moderate higher population in precipitation from July through April. output phase, female harvest rates of 4-26 densities were Where prairie mule deer had percent were possible while maintaining sustained than yearlong access to agricultural crops, it stable numbers of does in the population. in the absence of appeared that higher population densities Mule deer populations in prairie cropland. were sustained than in the absence of environments are capable of high output cropland. Stivers (pers. comm. 1997) when lush growing seasons are followed recorded average densities of 8 deer/km2 by mild-moderate winters. Recruitment (range 3-14 deer km2) during 1971-1996 as high as 75-90 or more fawns:100 adult along Sage and Indian Creeks in HD 419. females, and negligible natural losses of Annual variation in fawn recruitment rate adult females required doe harvest rates

144 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Figure 65. Recommended mule deer doe harvest rates across the span of expected variation in fawn recruitment and natural mortality of adult females in a prairie/badlands environment. of 27-30 percent to stabilize numbers of Extensive movements within large adult females during those periods. resident home ranges allowed whitetails Annual mortality rate of adult bucks to satisfy resource requirements. in the Cherry Creek mule deer population Whitetail density in prairie-agricultural 2 averaged about 60 percent, most of environments varied from 1 to 5 deer/km which resulted from hunting (Table (Dusek et al. 1988, Wood et al. 1989). 7). Fig. 62 correctly predicted that sex However, density is often aggregated in ratios of 18-20 bucks:100 does would be preferred habitats interspersed among maintained when recruitment averaged large areas receiving little or no deer use. about 55 fawns:100 females. However, Agricultural crops such as alfalfa very few bucks older than 3 years would and grain were utilized in varying be expected when mortality was ≥ 60 amounts during all seasons. Croplands percent. were important not only in satisfying nutritional requirements but also in creating the niche occupied by whitetails White-tailed Deer Population in this environment. Ecology in Prairie-Agricultural We speculated that whitetails usually Environments opted for a strategy of winter survival Habitat for white-tailed deer that involved increasing energy intake by in upland prairie centers on the selectively foraging on agricultural crops interspersion of woody draws, dryland (Dusek et al. 1988). Use of nutritious agricultural fields, and adjacent shrub- forage apparently compensated for much grass rangelands (Swenson et al. 1983, of the energy loss experienced in these Dusek et al. 1988, Wood et al. 1989). open habitats during winter. Although

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 145 whitetails utilized more gentle terrain, most common in Montana east of an they conserved energy during severe approximate north-south line extending conditions by increased use of rough through Lewistown (Walcheck 1978, topography and woody draws similar to Swenson 1979, Feldner and Smith 1981). mule deer (Wood et al. 1989). Periodic unpredictable losses of 33 Despite their dependence on percent or more of a population can occur agriculture, whitetails occupying variable (Swenson 1979). semi-arid environments were influenced Influences of coyote predation on by conditions during the growing season white-tailed deer population dynamics in similar to the relationships described prairie-agricultural areas have not been for mule deer in timbered breaks and studied in Montana. prairie-badlands environments. Total White-tailed Deer Vital Parameters precipitation received during a 10-month and Harvest Recommendations—Wood period (July-April) prior to fawning was et al. (1989) reported fawn recruitment positively correlated with the percentage rates during an 11-year period for a of whitetail fawns recruited to spring white-tailed deer population in a prairie- populations. Consequently in populations agricultural environment. Only 4 years of of both species increased and decreased data on mortality rates of adult females similarly across the span of environmental and males were available. variation that occurred. Low, moderate, and high output Epizootic hemorrhagic disease phases for prairie-agricultural whitetail (EHD) occurs periodically and results in populations (Fig. 66) were similar to mule significant mortality of all sex and age deer populations occupying the prairie- classes in some whitetail populations badlands (Fig. 65). Whitetail recruitment during late summer. The disease is rates were higher (ave. = 66±30, coef.

Figure 66. Recommended white-tailed doe harvest rates across the span of expected variation in fawn recruitment and natural mortality of adult females in a prairie/agricultural environment.

146 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a var. = 45 percent) than those for mule of equal importance because this is when deer (ave. = 56±24, coef. var. = 43 managers have the best opportunity to percent), but similarly variable (Table bring population size closer to the goal 4). Natural mortality rates among adult for a particular area. Consequently, females of both species in prairie-badland population changes experienced during and prairie-agricultural environments ensuing decline or increase phases may Management were similar. During a low output phase, be of lower magnitude and the harvest credibility is whitetail doe harvest rates of 1-6 percent regulation changes may not be as drastic, tested by private would stabilize numbers of adult females. resulting in fewer repercussions in the landowner The stabilizing harvest rate varied from 7 public arena. tolerance during to 29 percent across a span of moderate Management goals for deer deer population output and was 29-35 percent during population size should be stratified highs and by periods of high output. according to land ownership. On The annual mortality rate of adult private lands, control of deer numbers hunters and the whitetail bucks was approximately 55 is a priority concern and requires close general public percent (Wood et al. 1989). Observed cooperation between wildlife managers, during population postseason buck:doe ratios averaging landowners, and hunters. A particularly lows. 28:100 could be maintained with that liberal set of regulation packages should mortality rate because recruitment rates be focused on those areas where private averaged 65 fawns:100 adult females. land occurs in large blocks. In general, whitetail harvest Special Deer Population objectives on private lands should be Management Issues in Prairie given priority because of their integral Environments association with agriculture. When deer populations are at moderate-high levels Management of white-tailed deer in eastern Montana, hunter numbers in in prairie-agricultural environments local communities are often inadequate to and mule deer in prairie-badland meet harvest objectives. It is necessary to environments constitutes a challenging attract additional hunters from other parts “balancing act” between different land of the state, or issue multiple licenses. To ownership concerns and changes in deer solve important and often chronic game numbers on those lands. Decisive action damage situations, available hunting is particular important to successful pressure must be directed to private lands deer management. Management designated as high priority. This may responsiveness begins with a well- organized monitoring system that can detect major changes in deer populations. This information must initiate an administrative process that can rapidly implement the appropriate regulation package to accomplish a distinct change in harvest rate. Management credibility is tested by private landowner tolerance during deer population highs and by hunters and the general public during population lows. Harvest management during moderate population outputs draws far less public attention. Interest in improving management tends to focus only at the two extremes of change. However, the capability to design regulations that accomplish appropriate harvest rates during the “quiet”, intervening years is

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 147 require greater specificity in identifying more whitetails in the prairies of today where additional antlerless tags are than during pre-settlement times when valid. In some cases, smaller hunting it was native rangeland. Relationships units may be necessary. Entry into Block between agriculture and mule deer are Management Programs could also be less clear. Most croplands occur on prioritized to achieve greater access in flat terrain that probably would receive high priority game damage areas. limited use by mule deer in the absence On large blocks of public land, of agriculture. Large scale elimination a more conservative set of harvest of shrub/grasslands may have been regulations may apply. This could require detrimental to mule deer in some areas greater refinement in hunting district directly adjacent to rough topography. boundaries or designated subunits within Introduction of alfalfa fields in areas existing large hunting districts. It could adjacent to rugged topography can be also require changes in the way over-the- expected to benefit mule deer, but these counter antlerless tags are issued. In the situations often generate chronic deer past, liberal harvest regulations designed damage complaints. to reduce populations on private land For purposes of soil stabilization, have been applied broadly to encompass substantial areas of cropland have been all land ownership. Harvest pressure converted under the Federal Conservation applied to antlerless deer on public lands Reserve Program (CRP) to vegetation competes with opportunity for controlling types including substantial amounts of excessive deer numbers on private lands. alfalfa and sweet clover that appear to be Most of the time, a general either-sex preferred by deer, particularly whitetails. license would be appropriate for both We speculate that CRP contributes to deer There probably land ownerships. Additional antlerless population increases in some intensively- are many more licenses could be limited to private lands farmed areas where it enhances habitat whitetails in the or specific areas where ownership is diversity. A possible unintended outcome prairies of today intermixed. has been an increase in white-tailed deer than during pre- numbers in areas of private land where settlement times they are least welcome. Habitat Management in Prairie- when it was native Protecting wooded draws, creek Badland and Prairie-Agricultural rangeland. bottoms, and other mesic areas from Environments heavy livestock grazing should be Habitat components of fundamental beneficial to both species (Kraft 1989, importance to mule deer in prairie Wood et al. 1989, Jackson 1990). badlands include rugged topography This would be applicable to mule deer and springs, seeps, or creek bottoms when these areas occur within rugged that provide vegetative cover and topography, and for white-tailed deer high quality forage. White-tailed deer when applied to mesic areas within or in prairie-agricultural environments adjacent to agricultural croplands. require interspersed brushy draws and agricultural cropland. Prairie environments encompass higher Riverbottom Agricultural percentages of private land than most mountain and timbered breaks Ecosystem environments. Habitat management programs and expectations for either The riverbottom agricultural species must be aligned to these ecosystem includes irrigated and non- differences in land ownership. irrigated croplands associated with fertile Agricultural croplands positively valleys and bottomlands along major affect white-tailed deer distribution and rivers of Montana. We delineated two density when fields remain small and environmental subcategories; plains native habitats are maintained in close riverbottoms and intermountain valleys. proximity. There are probably many The first includes the mainstem of the

148 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Yellowstone River and portions of its principal tributaries such as the Clark Fork, Bighorn, Tongue, and Powder Rivers. It also includes portions of the mainstem Missouri River and its tributaries wherever narrow, fertile floodplains provide a mixture of riparian forest and irrigated croplands. The regional climate prevailing over the eastern plains riverbottoms is semi-arid. Irrigated floodplains represent a veritable oasis in an otherwise dry, variable, prairie environment. Important vegetation communities include cottonwood (Populus sp.), green ash (Fraxinus pennsylvanica), willow (Salix sp.), shrublands, and grasslands. Crops include alfalfa, sugar beets, pinto beans, corn, and grain. Livestock grazing is lower Yellowstone River (Herriges 1986, prevalent. Compton 1986, Compton et al. 1988, Intermountain valley environments Dusek et al. 1989). The highest deer are found closer to the headwater basins densities recorded for either species in of major river drainages in western Montana occurred in this area. Overall Montana. They provide widely distributed white-tailed deer density on the lower riparian plant communities, irrigated and Yellowstone varied between 14-27 deer/ non-irrigated croplands, grazed areas, km2 during 1981-1987 (Dusek et al. and urban development. Examples east 1989). Densities as high as 50/km2 were of the Continental Divide include the recorded on smaller sections of river Gallatin, Jefferson, Shields, and Paradise bottom during years of peak populations. Valleys. The lower Flathead, Bitterroot, Abundance of deer varied according Clark Fork, Blackfoot, and portions of to the characteristics of the river channel. the Kootenai Valleys are examples west The highest density occurred where the of the Divide. Compared to the plains river channel meandered and formed riverbottoms, annual precipitation and bottomlands that supported large tracts average snowfall are greater in the of riparian forest (Boggs 1984, Compton intermountain valleys though deep, et al. 1988). Lower deer densities were persistent snow cover is infrequent. found along straight sections of river Riverbottom agricultural channel where bottomlands and riparian environments support abundant, cover were limited. productive white-tailed deer populations. Riparian vegetation provided cover Today, mule deer are more incidental in and natural forage yearlong and was their occurrence or make seasonal forays particularly important to adult females into major riverbottoms. Land ownership during fawning. These habitats were is essentially private. Hunting access allocated among maternal does through varies significantly but is generally more territorial behavior associated with limited in the intermountain valleys. fawning. At low deer density, all females occupied optimal habitats and fawn White-tailed Deer Population recruitment was high. At high densities, many young does move to lower quality Ecology in Plains Riverbottom habitat where they lost one or both Environments fawns. Increased social strife at high Knowledge concerning the ecology density caused more frequent aggressive of white-tailed deer in this environment encounters as evidenced by greater is based primarily on studies along the mobility among females. This resulted

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 149 in lower fawn survival among prime- and Annual fawn mortality (ave = older-aged does (Dusek et al. 1989). 50 percent) was the lowest of any Thus, density-dependent changes in fawn deer population studied in Montana. survival were caused by social behavior Magnitude, timing, and cause of mortality ...riverbottom rather than competition for winter among fawns and adults occurred in a environments forage. Whitetail populations in plains pattern different from deer populations in demonstrate riverbottom environments demonstrate other environments. Eighty-six percent resiliency resiliency to harvest exploitation of 179 documented deaths of collared to harvest because fawn recruitment rates display a deer >4 months of age occurred during exploitation compensatory response to reduction in September-November, 7 percent during because fawn female density. winter, 5 percent during spring, and 2 recruitment Availability of high quality food percent during summer (Dusek et al. rates display a was abundant yearlong. Irrigated crops 1989). Predation and winter kill were compensatory supplemented the summer diet and minor causes of death among riverbottom response to may have buffered the effect of drought whitetails while hunter-kill, road-kill, and reduction in female conditions that reduced forage quality periodic losses to epizootic hemorrhagic density and are and availability elsewhere. Whitetails disease were important. Dusek et al. comparatively high consistently foraged on high-energy foods (1989) estimated that 26 percent and 2 over time. such as alfalfa, sugar beets, and grain in percent of autumn populations were killed winter, thus avoiding significant energy annually by hunting and automobile-deer deficits. Seasonal trends in their physical collisions, respectively. Swenson (1979) condition indicated that fawns and adult estimated losses of approximately 33 females maintained their body weight to percent of the whitetail population along late winter similar to deer supplementally the lower Yellowstone to EHD in the late fed high quality forage in penned studies. 1970s. White-tailed Deer Vital Parameters and Harvest Recommendations—In contrast to other deer populations in Montana, environmental variation had only minor influence on population dynamics of whitetails in riverbottom agricultural environments. Here, resources of importance to deer were comparatively more stable in their yearlong availability than in other environments. Population output phases were influenced primarily by adult female density. This important difference precluded use of a pendulum graph to describe whitetail population dynamics in this environment. Recruitment averaged 75±20 fawns:100 adult females, the highest recorded among all deer populations studied in Montana (Table 4). Variation in the fawn recruitment rate along the lower Yellowstone (coefficient of variation =27 percent) was much lower than other deer populations east of the Continental Divide. Mortality rates of adult females along the lower Yellowstone (Dusek et al. 1992) were calculated using the Micromort Software Program. Natural

150 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a mortality rates in plains riverbottom (1989) indicated this approach resulted environments for females 2 years in a harvest of 30-33 percent of adult and older were somewhat higher (11- females which initiated a population 16 percent) compared to other deer decline in 1984 and 1985. In contrast to other populations (Table 6) and appeared to If the management goal is to deer populations display relatively little annual variation. maintain certain whitetail populations at in Montana, However, Dusek et al. (1992) concluded moderate levels in the plains riverbottom environmental that opportunity was limited for hunting ecosystem, a sustained total annual variation had only mortality to substitute for non-hunting mortality rate (harvest and natural minor influence mortality of adult females. mortality) of about 25 percent of adult on population On the lower Yellowstone, females may be necessary. These rates dynamics of recruitment of 50-55 fawns:100 adult are subject to change depending on local whitetails in females was associated with high densities management goals, rates of recruitment, riverbottom of adult females. Under a prevailing and natural mortality that apply to a agricultural particular whitetail population. Whitetails natural mortality rate of 14 percent of environments. adult females at this low output phase, occupying other plains riverbottom a harvest of only 6 percent would keep (Allen 1968, Hamlin 1979, 1980) or the population from increasing. The intermountain valley environments with management prescription for this area less intensive water development and less usually requires a significant harvest of irrigated cropland may experience lower females to reduce the population to a and more variable fawn recruitment rates level within landowner tolerances at these and a lower spectrum of female harvest high deer densities. This is particularly rates. Buck hunting mortality rate averaged challenging because harvests must 58 ± 11 percent annually. However, overcome the compensatory increases a relatively high, average postseason in fawn recruitment that will occur as buck:doe ratio of 25:100 was maintained population density is reduced. Adult because fawn recruitment was high and female natural mortality rates could have stable. varied from 5-14 percent depending For general reference, Table 9 upon the method of estimation. If the provides the estimated harvest rates lower natural mortality estimate was achieved for each sex and age class and most nearly correct, then a 25-28 percent the corresponding effects on population hunter harvest of adult females would trend during a 6-year period on the lower accomplish a decline in that segment. A Yellowstone (Dusek et al. 1989). 16-19 percent harvest would be required Essentially no data have been if the higher estimate of natural mortality collected on recruitment, natural mortality was in effect. High levels of female rates, or population size for whitetails harvest in this population were only in intermountain valley environments. accomplished by issuing each hunter We speculate that dynamics of these multiple antlerless tags in addition to the populations may be similar to whitetails either-sex general license. Dusek et al.

Table 9. Harvest rates by sex and age class and their general effects on white-tailed deer population trend on the lower Yellowstone (after Dusek et al. 1989). Percent Harvest by Sex and Age Class Year total Population Adult Females Adult Bucks Fawns Effect on Population Trend 1980 13 11 38 3 Increase 1981 14 6 60 5 Increase 1982 25 14 68 10 Increase 1983 32 26 69 14 Stable 1984 35 30 53 17 Decline 1985 34 33 57 22 Decline

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 151 in plains river bottoms, although effects • Large scale land clearing, burning, of winter severity may be somewhat more or logging of cottonwood stands removes influential and require more conservative existing riparian forests and rapidly harvest rates. advances succession toward grasslands Whitetails in riverbottom-agricultural which could result in reduced white-tailed areas have the greatest potential for deer densities. causing crop damage because of high • The linear relationship between deer density and productivity per unit deer density and amount of riparian of occupied habitat. Hunter access to cover (Compton et al. 1988) may private land is of critical importance to provide a useful tool for indexing relative whitetail harvest management in these abundance of white-tailed deer along areas. Compared to other environments, major rivers in the agricultural areas of more consistent and intense hunting Montana. pressure must be focused on riverbottom • Maintenance of large tracts whitetails to keep them within levels of dense, woody, riparian vegetation tolerable to landowners. Evenly provides important security cover for distributed, dependable, long term hunter white-tailed deer occupying areas of access agreements are a key element in intense human activity (Vogel 1983, consistently directing hunting pressure to Herriges 1986). these lands. These areas require priority Whitetails in • Housing developments in the status when implementing hunter access western intermountain valleys were least riverbottom- programs. Alternatively, multiple licenses agricultural compatible with use by white-tailed deer issued to limited numbers of hunters that when housing was evenly distributed. areas have the can gain access to private lands may also Vogel (1989) recommended a strategy greatest potential be used to harvest these populations. for causing crop of clustering housing density on or near damage because of already developing areas, especially those high deer density Habitat Management for White- of little value to agriculture or wildlife, and productvity tailed Deer in Riverbottom- rather than develop new areas. per unit of occupied Agricultural Ecosystems • In areas where chronic problems habitat. In plains riverbottoms and occur with high density white-tailed deer intermountain valleys, white-tailed deer populations, elimination of riparian cover rely on the interspersion of riparian in combination with heavy hunter harvest vegetation and agricultural croplands for or other types of removal may offer the all their resource requirements. Following only long-term solution. are our primary recommendations: • White-tailed deer distribution was • Maintenance of successional negatively influenced by the presence of relationships in riparian vegetation cattle on the river bottom (Compton et communities is fundamentally related al. 1988). When cattle were introduced, to periodic flooding which also causes deer moved immediately to the nearest bank erosion and deposition of sediments cattle-free area. Deer resumed use of to form bottomlands. If flooding is the area when cattle were removed. minimized by water impoundments This suggested that, as observed for elk or diversions, bank cutting is reduced and other cervids (Lonner and Mackie and successionally advanced plant 1983), avoidance of cattle may reflect communities (grasslands) become more social intolerance rather than forage common (Boggs 1984). White-tailed competition. deer selected for mid-late successional • Alfalfa fields associated with communities such as mature cottonwood, riparian cover were used to a greater shrub, green ash, and mature willow with extent than fields without nearby cover. tall dense cover (Herriges 1986). In Deer use of alfalfa fields during summer the absence of periodic flooding, there increased with distance from human is minimal replacement of these plant settlement, amounts of adjacent cropland communities in the riparian system. other than alfalfa, and amount of nearby

152 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a breaks topography (Rausher 1995). during the summer growing season. Densities of 19 deer/hectare/day for 30 Large numbers of deer used early green days or more were required for deer up in some winter wheat fields, though foraging to significantly reduce alfalfa the large size of these fields probably yields during the growing season. These reduced potential for substantial densities were seldom reached along the economic damage. lower Yellowstone River (Herriges 1986, • Development of small fields of Rauscher 1995). alfalfa or grain in areas of riparian cover • White-tailed deer use of sugar on state-owned lands could reduce deer beets and winter wheat did not appear use of adjacent private lands. Only two to be heavy enough to cause damage of cuttings of alfalfa should be harvested economic significance (Herriges 1986). so stands are maintained in vigorous Beet fields did not receive heavy use condition (Herriges 1986).

Dee r Ma n a g eme n t i n Spe c i f i c Ec o s y s t em s 153 Future Directions in Deer Management

The impetus for progressive Our understanding of the change often occurs in the aftermath of relationships between deer population declines in abundance of an important dynamics and hunter harvest still has resource. Response to declining deer serious limitations. This uncertainty populations in Montana during the results in differences of opinion mid-1970s emphasized the need for concerning harvest management ecologically-based knowledge of deer within MFWP and the general public. populations and their habitats. Montana Walters (1986) concluded that we FWP initiated a series of long-term learn about responses of natural studies that ultimately resulted in the populations to harvest strategies description of habitat relationships mainly through experience. This can and population dynamics summarized only be accomplished if management in this document. The decline in deer is conducted within the context of populations during the mid-1990s systematically measuring the effects has recently provided the opportunity of harvest regulations and other to merge that knowledge of deer important environmental factors on deer population biology with a more innovative populations. This measurement and its management process. feedback to management represents the essence of adaptive deer management and sets it apart from traditional approaches. Adaptive Deer MFWP officially began development of an Adaptive Deer Management (ADM) Management program in 1996. The basic components that are linked together in this process include specific deer population The concept of adaptive objectives, distinct sets of harvest management of natural resources was regulations, a monitoring program, initially defined almost 20 years ago by and alternative models of population Walters and Hilborn (1978). More recent dynamics. The following comments are applications have been described by preliminary because there is much to Walters (1986), Johnson et al. (1993), be learned about application of these and Williams and Johnson (1995). It concepts to management of deer in is a process that functionally integrates Montana. the basic components of management to positively influence administrative decisions concerning harvest regulations. Deer Population Objectives The primary outcome is to recommend Deer population objectives are an harvest management practices that integral component of ADM because optimize hunter harvests of populations they influence the outcome of the occupying dynamic ecological and social entire management program. Specific environments. objectives for population and harvest

154 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a management are required to define strategies. Variation in habitat security, direction and measure progress over proximity to urban areas, land ownership, time. However, determining objectives and hunter access can create broad is often the most difficult task. We differences in the rate of harvest recommend that the ecological spectrum achieved by a particular regulation. of deer population dynamics described Regional managers can integrate the in this document should frame the population parameters and harvest rates key elements of management goals, summarized in this document with these objectives, and strategies. Differences other considerations to tailor regulations in biology and socioeconomic concerns appropriate to their areas. warrant separate management programs Harvest regulations can be designed specific sets of objectives for each to fit each species and ecological setting. species. This approach would provide Including a restrictive, moderate, greater management resolution and and liberal regulation will provide an responsiveness. appropriate array of harvest rates. The Specific objectives should be intent is to achieve the proper harvest We recommend directed toward important, individual according to the status of the population that the ecological populations or groups of populations and its current position relative to spectrum of that occupy a similar ecological system. management objectives. When a change deer population Objectives focusing on management in harvest strategy is warranted, the of the total population in a particular adjustment in harvest rate should be large dynamics described area should clearly describe the desired enough to detect a population response in ths document numerical size and acceptable limits through the monitoring program. should frame the of fluctuation based on biological and Regulations that are easily understood by key elements of social (landowner and sportsman) the hunter will be most acceptable, and management goals, considerations. they should be easily enforced. However, objectives, and Earlier in this document, we goals to both simplify regulations and strategies. described phases of variation in vital increase the diversity of buck hunting population parameters and corresponding opportunity may conflict. The targeted harvest rates that would decrease, rate of harvest to be achieved by a stabilize, or increase population size. particular regulation should be specific Objectives should include quantified to adult females and/or adult bucks, threshold levels in these parameters or depending on population management “triggers” that cause a change to a more objectives. conservative or liberal set of regulations. In variable environments occupied Parameter levels used in the objectives by deer in Montana, populations may would apply to those deer populations not respond in a predictable or desired that are censused in the monitoring direction even when a prescribed harvest program. rate is achieved. In these situations, Objectives for increasing buck knowledge of deer population dynamics availability and age diversity could be and effectiveness of harvest regulations described in a similar manner. However, would be significantly increased if hunting descriptive parameters and triggering seasons were evaluated as management mechanisms might then include numbers experiments. Use of treatment and of bucks and buck:doe ratios in post control areas within the same ecological season populations as well as age system could potentially sort out effects structure and antler size in the harvest. of harvest from environmental variation on trends in population size. Harvest Regulations Different sets of harvest regulations should be applied to mule We do not describe specific harvest deer populations occupying mountain regulations because a variety of social and and prairie ecosystems. It appeared economic considerations exert influence that dynamics of populations in the on local and regional deer management timbered breaks environment were

Fu t u r e Di r e c t i o n s i n Dee r Ma n a g eme n t 155 generally similar to populations in the document. For mule deer, these include prairie-badlands and may not warrant mountain-foothill, northwest montane, separate regulations. However, spatial timbered breaks, and prairie-badlands. and temporal variation in population Census areas for white-tailed deer should dynamics may require that more than be distributed within mountain-foothill, one component of the regulation package northwest montane, riverbottom- could be applied to parts of this vast agricultural, and prairie-agricultural area within the same year. White-tailed environments. Although separate sets deer populations occupying northwest of regulations may not be necessary for montane forest, prairie-agricultural, and each of these combinations, we believe river bottom-agricultural environments that some monitoring sites should occur may require three distinct sets of in all types to account for spatial and regulations. Additional regulations may temporal differences in population trends result from new information collected as well as socioeconomic considerations. on deer populations in other ecological Discussions with local biologists are settings or as required by differences in required before the number and location land ownership, game damage, or hunting of census areas are determined. Funding pressure. and data collection on these selected census areas would receive priority over other deer trend areas. Deer Population Monitoring Subsequent discussions of A clearly focused monitoring monitoring techniques will be confined program is essential to implementing to environments where it is feasible to A clearly focused innovative approaches in deer conduct aerial surveys with helicopters monitoring management. Informed decisions must be or Piper Supercubs. Fortunately this program is based on systematic and consistent data includes about 75 percent of the land essential to on population status and trend, achieved area in Montana. Other ground-based techniques such as camera surveys implementing harvest rates, and how these relate to (Dusek and Mace 1991, Sime 1996) innovative important environmental factors. As a may be appropriate in densely timbered approaches in deer starting point, we recommend a statewide habitats. Specific recommendations network of census areas for each species management. include: of deer distributed across important • Where possible, “full coverage” ecological categories described in this flights over the census area be conducted during each aerial survey (Mackie et al. 1981). Consistency in this effort will maximize the number of deer counted in relation to the total number present on the area. Using these methods, consistent percentages of deer have been observed in various environments with helicopters and Supercubs (Table 10). These aerial observability indices could be used to convert total numbers of deer counted to an estimate of population size. Each census area should be flown twice each year, post-hunting season and spring. • Primary population parameters measured in post-season census are total numbers, fawn:doe and buck:doe ratios, and antler point classes. Post-season flights should occur between December 1 and January 15 for mule deer and no later than January 1 for whitetails because of earlier antler shedding.

156 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Table 10. Aerial observability indexes measured from samples of marked deer in various environments in Montana.

Environmental Type Study Area Aircraft % Observability Number of Index ( ±1sd) Surveys Mule Deer - Post Season Surveys Mountain-foothills Bridger Mts. (PHU1) Jet Ranger, Bell 47G 46±13 7 (timbered) Mountain-foothills Bridgers Mts. (PHU4) Jet Ranger, Bell 47G 41±12 5 (shrub/grassland) Timbered breaks Missouri River Breaks Bell Soloy, Bell 47G 75±7 10 Prairie-badlands Cherry Creek Supercub 70±5 4 Mule Deer - Spring Surveys Mountain-foothills Bridgers Mts. (PHU1) Jet Ranger, Bell 47G 57±7 10 (timbered) Mountain-foothills Bridgers Mts. (PHU4) Jet Ranger, Bell 47G 68±12 9 (shrub/grassland) Mountain-foothills Rocky Mt. Front Bell 47G 57±7 5 Timbered breaks Missouri River Breaks Bell 47G 74 2 Timbered breaks Missouri River Breaks Supercub 68±5 7 Timbered breaks Lower Stillwater Bell 47G 58±11 4 (with agriculture) Prairie-badlands Cherry Creek Supercub 65±6 5 Prairie-badlands Hammond Area Supercub 76±5 6 White-tailed Deer - Post Season Surveys Prairie-agricultural Cherry Cr. Supercub 51±1 3 White-tailed Deer - Spring Surveys Prairie-agricultural Cherry Cr. Supercub 57±6 4 Plains riverbottom Lower Yellowstone-Elk Island Supercub 25±2 15 Plains riverbottom Lower Yellowstone-Intake Supercub 35±3 15

• Primary parameters measured the aircraft should turn the group back during spring census are total numbers toward the area already counted; this and fawn:adult ratios. Census flights no is most important on areas with high later than 2 weeks after the beginning of deer density. Survey time and weather spring greenup will optimize observability. conditions such as cloud cover, wind, This period varies from year to year relative snow depth and coverage, and depending on plant phenology but usually temperature should be recorded as well occurs between March 15 and April 30. as deer behavior relative to use of open During both census periods, deer or timbered sites. Ideally, surveys are observations recorded by social group conducted with either total snow cover or size and composition will provide the total bare ground as opposed to patchy greatest amount of information. Location snow conditions. However, completing of deer groups can be recorded by the census is most important, regardless drainage name and elevation or global of snow conditions. More details on positioning system (GPS) methods. aerial survey methodology are provided Classification of sex and age classes is in Unsworth et al. (1994). most efficient when the pilot presents the Selecting the size and boundaries observer with a low-level “broad-side” of census areas is critical in minimizing view of all members of the group. After a ingress and egress of deer during and group of deer is classified and tabulated, between sampling periods. This will

Fu t u r e Di r e c t i o n s i n Dee r Ma n a g eme n t 157 ensure that numbers of deer counted during peak populations. This also over time will reflect actual changes in will help minimize movement of deer population size. Populations that have in and out of the census area. This been the subject of telemetry studies of arrangement will provide improved seasonal and yearlong distribution should ability to detect population increases receive first priority in selecting census and decreases based on differential use areas. of these flatter areas. Inclusion of only An ecologically complete unit rugged terrain in the census area may of winter range should represent the indicate changes in population size of a census area in mountain habitats with much smaller magnitude than across the migratory deer. Use of helicopters for broader area. The census area should census flights will enhance observability include a minimum of 100 adult females in rugged, partially timbered terrain during population lows. This may 2 and offer greater observer efficiency require an area of up to 250 km . The at high deer density. Upper and lower increased knowledge of deer use gained by recording locations of deer groups Selecting the size elevational boundaries within which the census is conducted will vary from year during flights can enable a large census and boundaries to year depending on snow depth and area to later be reduced in size without of census areas deer distribution. During each census losing past years’ information. However, is critical in we recommend that aerial coverage if a census area is found to be too small minimizing ingress extend or adjust to “run out” of deer at and must be expanded, the end result is and egress of both high and low elevations. This will equivalent to starting over. deer during and account for differences in distribution Supercub surveys have been demonstrated to be effective in estimating between sampling between years and between bucks and trends in population size and composition periods. does. Lateral boundaries to the census for white-tailed deer in the riverbottom- unit should represent areas that are agricultural environment (Dusek et essentially devoid of deer in winter and al. 1989). Habitats are essentially represent discontinuities between winter linear and easy to define along major range units. Size of the census area 2 river valleys. Because deer density is may vary from 25-125 km . Because exceptionally high and often associated deer on mountain-foothill winter ranges with dense riparian cover, a linear census usually occur at high density, sample area along approximately 25 km of river size will almost always exceed minimal channel should be adequate. Highest requirements (>100 females and deer densities are often associated with associated bucks and fawns). Virtually meanders in the river channel and would all of our experience in aerial census on constitute the core of the census area. mountain-foothill winter ranges is with End points can be selected along straight mule deer, so modifications in approach sections of river channel where riparian may be necessary for white-tailed deer. cover and deer habitat diminish. Deer in prairie or timbered breaks Complete snow cover will optimize environments primarily display resident survey conditions during the post-season movement patterns and are distributed census by maximizing the background at relatively low density across large contrast of deer located in riparian cover. expanses of habitat. Surveys can be We recommend partitioning the search efficiently conducted with a Piper effort according to cover density. Large Supercub, although a helicopter would pieces of riparian forest are systematically enhance classification of bucks in searched before moving to the next timbered breaks. It is important to cover piece. Deer in agricultural fields include a large piece of rugged terrain are easily observed but have a greater including springs, seeps and other tendency to run for cover when hazed moist sites in the central portion of the by the aircraft. They should be turned census area. The perimeter should be toward cover patches already surveyed. situated in relatively flat terrain with Reliable classification of whitetail fawns some creek bottoms that are used only will require practice.

158 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Census flights conducted during regulations are poorly understood. By morning and evening hours will coincide connecting a check station effort with a with periods when deer are most active. priority census area, numbers of females Spring census flights should focus on harvested from the censused population the early spring greenup when the could be more accurately estimated. Sex highest proportion of deer may be using specific rates of harvest could also be agricultural fields. estimated by employing samples of radio- White-tailed deer census areas in collared deer in priority populations. prairie-agricultural environments should 2 approximate 200-250 km in size. A Alternative Models of Population relatively high density of wooded cover Dynamics patches in association with agriculture should occur in the central portion of Population modeling is a tool to the census area. Perimeter boundaries increase our understanding of how can occur in flat expanses of agricultural and why populations change. Recent croplands with limited cover patches. advances in modeling capabilities make it Techniques for conducting aerial surveys possible to more accurately describe the are similar to those described above. dynamic changes that wildlife populations Other important components in the experience in variable environments monitoring program for both species (Lubow 1995, Lubow et al. 1996). include the Statewide Hunter and Harvest When a feedback loop is established Survey and local check stations. Both between monitoring and modeling, sources provide important information more responsive management can be concerning trends in hunter effort, advanced. Modeling assists in deciding percent success, the magnitude of harvest which harvest management option is achieved under different regulations, appropriate for the current population and whether harvest objectives are met. status. Monitoring the managed Detailed data on specific deer populations populations measures the outcome of can be collected at check stations. Rates that decision. We caution that modeling of female harvest achieved by particular can be ineffective in its application

Fu t u r e Di r e c t i o n s i n Dee r Ma n a g eme n t 159 to management when it substitutes After the hunting season, aerial for monitoring actual changes in real monitoring of deer populations will populations. provide information to compare the Models not only describe the factor status of the population in year t + 1 interactions that affect population size with the predicted outcome of the various and composition, they can include models. The model that did the best job different assumptions about additive or of predicting the observed response by Models not only compensatory effects of hunting mortality. the actual population in year t + 1 is describe the Competing models are constructed using given more weight in the decision-making factor interactions these different assumptions to test an process for prescribing an optimal that affect array of hunting season regulations. regulation in the next year (t + 2). population size Each model is initiated with accurate Over time, repetition of this and compoistion, data describing the current status of modeling-monitoring feedback loop will they can include representative populations and conditions improve management performance by different of importance in the environment. In reducing the amount of uncertainty in each year (t), the modeling exercise our knowledge concerning the effects of assumptions identifies a specific hunting regulation harvest on population dynamics. This will about additive which optimizes harvest opportunity enhance our ability to detect and respond or compensatory in relation to management goals while to significant changes in deer populations effects of hunting minimizing negative effects on the and more closely meet objectives for mortality. population in year t + 1. A hunting population size and composition. With regulation is selected through the this approach, MFWP can move forward modeling process and then applied to with an innovative and progressive actual populations that fit the ecological deer management program to meet the parameters upon which the model was challenges of the twenty-first century. designed.

...this modeling- monitoring feedback loop... will enhance our ability to detect and respond to significant changes in deer populations and more closely meet objectives for population size and composition.

160 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Literature Cited and Appendix

Literature Cited

Allen, E. 0. 1965. Food and range Baker, R. R. 1978. The evolutionary use habits of white-tailed deer on ecology of animal migration. Missouri River bottomlands in Holmes and Meier Publ., Inc. New northcentral Montana. M.S. Thesis, York. 1012 pp. Mont. State Univ., Bozeman. 41 pp. Baker, S. V., and J. Fritsch. 1997. New _____. 1968. Range use, foods, territory for deer management: condition, and productivity of white- human conflicts on the suburban tailed deer in Montana. J Wildl. frontier. Wildl. Soc. Bull. 25:404- Manage. 32:130-141. 407. _____. 1971. White-tailed deer. Pages Ballard, W. B., J. S. Whitman, and D. J. 69-79 in T. W. Mussehl and F. W. Reed. 1991. Population dynamics Howell, eds. Game Management of moose in southcentral Alaska. in Montana. Mont. Dept. Fish and Wildl. Monogr. No. 114. 49 pp. Game, Helena. 238 pp. Baty, G. R. 1995. Resource partitioning Anderson, B. L., R. D. Pieper, and V. V. and browse use by sympatric elk, Howard, Jr. 1974. Growth response mule deer and white-tailed deer on and deer utilization of fertilized a winter range in western Montana. browse. J. Wildl. Manage. 38:525- M.S. Thesis. Univ. of Montana, 530. Missoula. 228 pp. Anthony, R. G., and N. S. Smith. 1977. _____, C. L. Marcum, M. J. Thompson, Ecological relationships between and J. M. Hillis. 1996. Potential mule deer and white-tailed deer in effects of ecosystem management on southeastern Arizona. Ecol. Monogr. cervids wintering in ponderosa pine 47:255-277. habitats. Intermountain J. Sci. 2:1-7. Arno, S. F., G. E. Gruell, J. G. Beasom, S. L., and E. P. Wiggers. 1984. Mundinger, and W. C. Schmidt. A critical assessment of white-tailed 1987. Developing silvicultural and mule deer productivity. Pages prescriptions to provide both deer 68-79 in Proc. Workshop on Deer winter habitat and timber. Western in the Southwest. New Mexico State Wildlands. Winter:19-24. Univ., Las Cruces. Bailey, E. D. 1969. Immigration Bentz, J. A., and P. M. Woodard. 1988. and emigration as contributory Vegetation characteristics and regulators of populations through bighorn sheep use on burned and social disruption. Can. J. Zool. unburned areas in Alberta. Wildl. 47:1213-1215. Soc. Bull. 16: 186-193.

Li t e r a t u r e Ci t e d 163 Boertje, R. D., P. Valkenburg, and M. E. conditions. Trans. North Am. Wildl. McNay. 1996. Increases in moose, Conf. 15:170-189. caribou, and wolves following wolf control in Alaska. J. Wildl. Manage. Clutton-Brock, T. H., F. E. Guinness, 60:474-489. and S. D. Albon. 1982. Red deer- behavior and ecology of two sexes. Boggs, K. W. 1984. Succession in University of Chicago Press. 378 pp. riparian communities of the lower Yellowstone River. M. S. Thesis. _____, G. R. Iason, and F. E. Guinness. Montana State University, Bozeman. 1987. Sexual segregation and 107 pp. density-related changes in habitat use in male and female red deer Botkin, D. B. 1990. Discordant (Cervus elaphus). J. Zool. London harmonies: a new ecology for 211:275-289. the twenty-first century. Oxford University Press. 241 pp. Coates, K. 1996. Region one white-tailed deer 1995 harvest report. Mont. Bowyer, R. T. 1984. Sexual segregation Fish, Wildl. and Parks, Kalispell. 39 in southern mule deer. J. Mammal. pp. 65:410-417. Coblentz, B. E. 1970. Food habits of Campbell, R. B., and C. J. Knowles. 1978. George Reserve Deer. J. Wildl. Elk-hunter-livestock interactions in Manage. 34:535-540. the Missouri River Breaks. Mont. Dept. Fish and Game. Job Final Cole, G. F. 1958. Range survey guide. Rept., Proj. W-130-R-9. 153 pp. Mont. Dept. Fish and Game. Booklet. 18 pp. Carpenter, L. H., and O. C. Wallmo. 1981. Rocky Mountain and intermountain _____. 1959. Key browse survey method. habitats, part 2: habitat evaluation Proc. West. Assoc. Game and Fish and management. Pages 399-421 Comm. 39:181-186. in O.C. Wallmo, ed., Mule and _____. 1961. Big game population black-tailed deer of North America. dynamics and season setting. University of Nebraska Press, Information Bull. No. 14. Mont. Lincoln. 605 pp. Dept. Fish and Game, Helena. 7 pp. Caughley, G. 1970. Eruption of ungulate Compton, B. B. 1986. Distribution of populations with emphasis on white-tailed deer along the lower Himalayan thar in New Zealand. Yellowstone River. M.S. Thesis. Ecology. 51:53-70. Mont. State Univ., Bozeman. 74 pp. _____. 1987. Ecological relationships. _____, R. J. Mackie, and G. L. Dusek. Pages 159-187 in G. Caughley, 1988. Factors influencing density N. Shepard, and J. Short, distribution of white-tailed deer in eds., Kangaroos: their ecology and riparian habitats. J. Wildl. Manage. management in the sheep rangelands 52:544-548. of Australia. Cambridge University Press, London. 253 pp. Connelly, G. E. 1978. Predators and predator control. Pages 369-394 _____, and A. Gunn. 1993. Dynamics in J. L. Schmidt and D. L. Gilbert, of large herbivores in deserts: eds. Big Game of North America. kangaroos and caribou. Oikos Stackpole Books, Harrisburg, PA. 67:47-55. 494 pp. Cheatum, E. L., and C. W. Severinghaus. Cook, J. G., E. B. Arnett, L. L. Irwin, 1950. Variations in fertility of and F. G. Lindzey. 1989. Ecology white-tailed deer related to range and population dynamics of two

164 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a transplanted bighorn sheep herds in relationships of deer. Proc. Third south-central Wyoming. Univ. Wyo., International Congress on Natural Laramie. 234 pp. Resources and Wildlife. Guadalajara, Mexico. Cronin, M. A., M. E. Nelson, and D. F. Pac. 1991. Spatial heterogeneity of _____, A. K. Wood, and R. J. Mackie. mitochondrial DNA and allozymes 1988. Habitat use by white-tailed among populations of white-tailed deer in prairie-agricultural habitat in deer and mule deer. J. Heredity 82: Montana. Prairie Nat. 20:135-142. 118-127. _____, A. K. Wood, and S. T. Stewart. Dusek, G. L. 1971. Range relationships 1992. Spatial and temporal patterns of mule deer in the prairie habitat, of mortality among female white- northcentral Montana. M. S. Thesis. tailed deer. J. Wildl. Manage. Mont. State Univ., Bozeman. 63 pp. 56:645-650. _____. 1975. Range relationships of Egan, J. L. 1971. Mule deer. Pages mule deer and cattle in prairie 53-67 in T. W. Mussehl and F. W. habitat. J. Wildl. Manage. 39:605- Howell, eds., Game management 616. in Montana. Mont. Dept. Fish and Game, Helena. 238 pp. _____. 1987. Ecology of white-tailed deer in upland ponderosa pine Elseth, G. D., and K. D. Baumgardner. habitats in south-eastern Montana. 1981. Population biology. D. Van Prairie Nat. 19:1-17. Nostrand Co., NY. 623 pp. _____. 1994. Dentition of deer. Pages Errington, P. L. 1943. An analysis of 71-75 in D. Gerlach, S. Atwater, mink predation upon muskrats in and J. Schnell, eds. Deer. Stackpole northcentral United States. Res. Books, Mechanicsburg, PA. 384 pp. Bull. Iowa Agric. Exp. Sta., 320:797- 924. _____, and R. D. Mace. 1991. Application of remote photography Eustace, C. D. 1971. Mule deer food surveys to ungulate research and habits and browse use study. Final management. Pages 240-243 in Rep., Fed. Aid Proj. W-130-R-1 and A. G. Christensen, L. J. Lyon, and 2, Job 1-7.1. Mont. Dept. Fish and T. N. Lonner, comps., Proc. Elk Game, Helena. 25 pp. Vulnerability Symp. Montana State University, Bozeman. 330 pp. _____. 1988. How’re the deer doin’? Montana Outdoors 19(5):11-15. _____, and R. J. Mackie. 1988. Population ecology and habitat Feldner, T. J., and M. H. Smith. 1981. relationships of white-tailed deer Epizootic Hemorrhagic disease virus in river bottom habitat in eastern in Montana: isolation and serologic Montana. Final Rep., Fed. Aid Proj. survey. Am. J. Vet. Res. 42:1198- W-120-R-12-18. Montana Dept. 1202. Fish, Wildlife and Parks, Helena. Flathead National Forest. 1994. Wildlife 116 pp. habitat evaluation, Swan Valley. _____, R. J. Mackie, J. D. Herriges, and USDA, Forest Service, Flathead B. B. Compton. 1989. Population National Forest, Kalispell, MT. 94 pp ecology of white-tailed deer along + Appendices. the lower Yellowstone River. Wildl. Fritzen, D. E. 1995. Ecology and Monogr. 104:1-68. behavior of mule deer on the _____., and J. T. Morgan. 1991. A Rosebud Coal Mine, Montana. Ph.D camera system to evaluate Thesis. Montana State University, population parameters and spatial Bozeman. 143 pp.

Li t e r a t u r e Ci t e d 165 Gasaway, W. C., R. O. Stephenson, J. L. Hamlin, K. L. 1977. Population dynamics Davis, P. E. K. Shepherd, and O. E. and habitat relationships of mule Burris. 1983. Interrelationships deer in the Bridger Mountains, of wolves, prey and man in interior Montana, 1955-1976. Abstr. Ann. Alaska. Wildl. Monogr. 84:1-50. Conf. N. W. Section, The Wildlife Society, Kalispell, MT. Gavin, T. A., L. H. Suring, P. A. Vohs, and C. E. Meslow. 1984. Population _____. 1979. Population ecology and characteristics, spatial organization, habitat relationships of white-tailed and natural mortality in the deer of bottomlands of the Missouri Columbian white-tailed deer. Wildl. River in northcentral Montana, Monogr. 91:1-41. Pages 167-174 in Montana Deer Studies. Job Progr. Rep., Fed. Aid Geist, V. 1981. Behavior: adaptive Proj. W-120-R-10. Mont. Dept. Fish, strategies in mule deer. Pages 157- Wildl. and Parks, Bozeman. 182 pp. 223 in O. C. Wallmo, ed. Mule and black-tailed deer of North America. _____. 1980. Population ecology and University of Nebraska Press, habitat relationships of white-tailed Lincoln. 605 pp. deer of bottomlands of the Missouri River in northcentral Montana, _____. 1994. Origin of the species. Pages 176-181 in Montana Deer Pages 2-15 in D. Gerlach, S. Atwater, Studies. Job Progr. Rep., Fed. Aid and J. Schnell, eds. Deer. Stackpole Proj. W-120-R-11. Mont. Dept. Fish, Books, Mechanicsburg, PA. 384 pp. Wildl. and Parks, Bozeman. 182 pp. Gilbert, F. F. 1966. Aging white-tailed _____, and G. L. Erickson. 1996. deer by annuli in the cementum of Environmental assessment of the first incisor. J. Wildl. Manage. proposed changes in hunting 30:200-202. season structure for mule deer in Gilmour, J. S. L., and J. W. Gregor. southwestern Montana. Wildlife 1939. Demes: a suggested new Division, Mont. Dept. Fish, Wildl. terminology. Nature 144:333-334. and Parks, Helena. 124 pp. Gilpin, M., and I. Hanski. 1991. _____, and R. J. Mackie. 1989. Mule Metapopulation dynamics: empirical deer in the Missouri River Breaks, and theoretical investigations. Montana: a study of population Academic Press. 336 pp. dynamics in a fluctuating environment. Final Rep., Fed. Aid Greer, K. R. 1968. A compression Proj. W-120-R. Mont. Dept. Fish, method indicates fat content of elk Wildl. and Parks, Bozeman. 401 pp. (wapiti) femur marrows. J. Wildl. Manage. 32:747-751. _____, and _____. 1991. Age-specific reproduction and mortality in Griffiths, D. E. 1990. Green vegetation female mule deer: implications to and fecal protein relationships in population dynamics. Pages 569- two southeastern Montana mule deer 573 in B. Bobek, K. Perzanowski, populations. M. S. Thesis. Mont. and W. Reglin, eds., Global trends in State Univ., Bozeman. 99 pp. wildlife management. Proc. XVIII Halls, L. S., P. R. Krausman, and M. Congr. Int. Union Game Biology., L. Morrison. 1997. The habitat Krakow, Poland. 1987. Vol. 1 Swiat concept and a plea for standard Press, Krakow. 660 pp. terminology. Wildl. Soc. Bull. Hanley, T. A. 1984. Relationships between 25:173-182. Sitka black-tailed deer and their habitat. USDA Forest Service, General Technical Rep. PNW-168. 21 pp.

166 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Hanski, I., and M. Gilpin. 1991. Jakimchuk, R. D., S. H. Ferguson, and Metapopulation dynamics: brief L. G. Sopuck. 1987. Differential history and conceptual domain. habitat use and sexual segregation in Pages 3-16 in M. Gilpin and I the central arctic caribou herd. Can. Hanski, eds., Metapopulation J. Zool. 65:534-541. dynamics: empirical and theoretical investigations. Academic Press, NY. Johnson, F. A., B. K. Williams, J. D. 336 pp. Nichols, J. Hinds, W. L. Kendall, G. W. Smith, and D. F. Caithamer. Heisey, D. M., and T. K. Fuller. 1985. 1993. Developing an adaptive Evaluation of survival and cause- management strategy for waterfowl specific mortality rates using in North America. Trans. North Am. telemetry data. J. Wildl. Manage. Wildl. Nat. Resour. Conf. 58:565- 47:668-674. 583. Herriges, J. D., Jr. 1986. Movement, Kamps, G. F. 1969. Whitetail and mule activity, and habitat use of whit-tailed deer relationships in the Snowy deer along the lower Yellowstone Mountains of central Montana. M. S. River. M.S. Thesis. Mont. State Thesis. Mont. State Univ., Bozeman. Univ., Bozeman. 131 pp. 59 pp. Hildebrand, P. R. 1971. Biology of Kasworm, W. F. 1981. Distribution and white-tailed deer on winter ranges population characteristics of mule in the Swan Valley, Montana. M. S. deer along the East Front, north- Thesis. Univ. Mont., Missoula. 91 pp. central Montana. M. S. Thesis. Mont. State Univ., Bozeman. 73 pp. Hoffman, 1985. Digestive physiology of deer-their morphophysiological Keith, L. B. 1983. Population dynamics specialization and adaptation. Pages of wolves. Can. Wildl. Serv. Rep. 393-407 in P. K. Fennessy and Ser. 45: 66-77. K. R. Drew, eds., Biology of deer production. Bulletin 22, The Royal Kie, J. G., and M. White. 1985. Society of New Zealand, Wellington. Population dynamics of white-tailed 482 pp. deer (Odocoileus virginianus) on the Welder Wildlife Refuge, Hudson, R. J. 1985. Body size, Texas. The Southwestern Naturalist. energetics, and adaptive radiations. 30:105-118. Pages 1-24 in R. J. Hudson and R. G. White, eds. Bioenergetics of wild Klein, D. R. 1985. Population ecology: herbivores. CRC Press, Boca Raton. the interaction between deer and their food supply. Biology of Deer Hunter, M. L., Jr. 1996. Fundamentals Production. Bulletin 22, The Royal of conservation biology. Blackwell Society of New Zealand, Wellington. Science. 435 pp. 482 pp. Ihsle, H. B. 1982. Population ecology Klinger, R. C., M. J. Kutilek, and H. S. of mule deer with emphasis on Shellhammer. 1989. Population potential impacts of oil and gas responses of black-tailed deer development along the east slope of to prescribed burning. J. Wildl. the Rocky Mountains, northcentral Manage. 53:863-871. Montana. M. S. Thesis. Mont. State Univ., Bozeman. 85 pp. Knapp, S. J. 1972. Range use of mule deer prior to initiation of rest- Jackson, S. D. 1990. Ecology of mule deer rotation grazing for cattle on the on a sagebrush-grassland habitat in Fort Howes Range District, Custer northeastern Montana. M. S. Thesis. Natl. Forest, Montana. M. S. Thesis. Mont. State Univ., Bozeman. 111 pp. Mont. State Univ., Bozeman. 50 pp.

Li t e r a t u r e Ci t e d 167 Knowles, C. J. 1975. Range relationships in mammalian ecology and of mule deer, elk, and cattle in a conservation. University of rest-rotation grazing system during Minnesota Press, Minneapolis. 215 summer and fall. M. S. Thesis. pp. Mont. State Univ., Bozeman. 111 pp. Lingle, S. 1989. Limb coordination and Komberec, T J. 1976. Range body configuration in the fast gaits relationships of mule deer, elk, and of white-tailed deer, mule deer, and cattle in a rest-rotation grazing their hybrids: adaptive significance system during winter and spring. and management implications. M. S. M. S. Thesis. Mont. State Univ., Thesis. Univ. Calgary. Bozeman. 79 pp. Lonner, T. N. 1975. Long Tom Creek Kraft, S. K. 1989. Ecology of mule deer Study pp. 60-72 in Montana in the upper Missouri River Breaks, cooperative elk-logging study. Job Montana. M. S. Thesis. Univ. Mont., Prog. Rept. Mont. Dept. Fish and Missoula. 83 pp. Game, Helena. 146 pp. Kunkel, K. E., 1997. Predation by _____, and R. J. Mackie. 1983. On the wolves and other large carnivores nature of competition between in northwestern Montana and big game and livestock. Pages southeastern British Columbia. 53-58 in B. F. Roche, Jr. and D. Ph.D. Thesis. Univ. Mont., Missoula. M. Baumgartner, eds., Forestland 272 pp. Grazing. Proc of a Symposium. Coop. Ext. Serv., Wash. State Univ., Leach, R. H. 1982. Summer range Pullman. 114 pp. ecology of white-tailed deer in the coniferous forests of northwestern Low, W. A., and I. M. Cowan. 1963. Age Montana. M. S. Thesis. Univ. Mont., determination in deer by annular Missoula. 80 pp. structure of dental cementum. J. Wildl. Manage. 27:466-471. Leopold, A. 1933. Game management. Scribner’s, NY. 481 pp. Lubow, B. C. 1995. SDP: Generalized software for solving stochastic Lidicker, W. Z., Jr. 1978. Regulation dynamic optimization problems. of numbers in small mammal Wildl. Soc. Bull. 23:738-742. populations-historical reflections and a synthesis. Pages 122-141 in D. _____, G. C. White, and D. R. Anderson. P. Snyder, ed., Populations of small 1996. Evaluation of a linked- mammals under natural conditions. sex harvest strategy for cervid Spec. Publ. Ser., Pymatuning Lab. populations. J. Wildl. Manage. Ecol., University of Pittsburgh. 5:1- 60:787-796. 237. Lund, R. E. 1983. A user’s guide to _____. 1985. Dispersal. Pages 420- MSUSTAT-an interactive statistical 454 in R. H. Tamarin, ed., Biology analysis package, CP/M version. of new world Microtus. Amer. Soc. Research and Development Mamm., Spec. Publ. 81-893. Institute, Mont. State Univ., Bozeman. _____. 1994. A spatially explicit approach to vole population Mackie, R. J. 1970. Range ecology and processes. Polish Ecological Studies relations of mule deer, elk, and 20:215-225. cattle in the Missouri River Breaks, Montana. Wildl. Monogr. 20:1-79. _____. 1995. The landscape concept: something old, something new. _____. 1973. Responses of four browse Pages 3-19 in W. Z. Lidicker, species to protection. Job Progr. Jr., ed. Landscape approaches Rep., Fed. Aid Proj. W-120-R-4.

168 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Mont. Dept. Fish and Game, Helena. Ann. Conf. Southeastern Assoc. Fish 30 pp. and Wildl. Agencies 30:487-492. _____. 1975. Evaluation of the key Martinka, C. J. 1978. Ungulate browse survey method. Fifth Mule populations in relation to wilderness Deer Workshop, Albuquerque, NM. in Glacier National Park, Montana. 17 pp (multililth). Trans. N. Am. Wildl. Nat. Resour. Conf. 43:351-357. _____. 1978. Impacts of livestock grazing on wild ungulates. Trans. Mautz, W. W. 1978. Sledding on a brushy N. Amer. Wildl. Nat. Resour. Conf. hillside: the fat cycle in deer. Wildl. 43:462-476. Soc. Bull. 6:88-90. _____, G. L. Dusek, K. L. Hamlin, H. _____., P. J. Pekins, and J. A. Warren. E. Jorgensen, D. F. Pac, and A. 1985. Cold temperature effects on K. Wood. 1985. Montana deer metabolic rate of white-tailed, mule, studies. Job Progr. Rep., Fed. Aid and black-tailed deer in winter coat. Proj. W-120-R-16. Mont. Dept. Pages 453-457 in P. K. Fennessy Fish, Wildl. and Parks, Helena. and K. R. Drew, eds., Biology of deer 132 pp. production. Bulletin 22, The Royal Society of New Zealand, Wellington. _____, K. L. Hamlin, and J. G. Mundinger. 482 pp. 1976. Habitat relationships of mule deer in the Bridger Mountains, Mayr, E. 1970. Populations, species, and Montana. Job Progr. Rep., Fed. Aid evolution. Harvard University Press, Proj. W-120-R-6. Mont. Dept. Fish, Cambridge, MA. 453 pp. Wildl. and Parks, Bozeman. 44 pp. McCullough, D. R. 1979. The George _____, _____, and D. F. Pac. 1981. Reserve deer herd: population ecology of a k-selected species. Census methods for mule deer. University of Michigan Press, Ann Pages 97-106 in F. L. Miller and Arbor. 271 pp. A. Gunn, eds., Symposium on census and inventory methods for _____. 1983. Rate of increase on the populations and habitats. Univ. George Reserve: A response. J. Idaho, For., Wildl., and Range Exp. Wildl. Manage. 47:1248-1250. Sta., Contribution No. 217. 220 pp. _____. 1984. Lessons from the George _____, _____, _____, G. L. Dusek, and Reserve, Michigan. Pages 211-242 A. K. Wood. 1990. Compensation in L.K. Halls, ed., white-tailed deer in free-ranging deer populations. ecology and management. Stackpole Trans. N. Am. Wildl. Natur. Resour. Books, Harrisburg, PA. 870 pp. Conf. 55:518-526. Mech, L. D. 1984. Predators and MacNab, J. 1985. Carrying capacity and predation. Pages 189-202 in L.K. related slippery shibboleths. Wildl. Halls, ed., white-tailed deer ecology Soc. Bull. 13:403-410. and management. Stackpole Books, Main, M. B., and B. E. Coblentz. 1990. Harrisburg, PA. 870 pp. Sexual segregation among ungulates: _____, and P. Karns. 1977. Role of the a critique. Wildl. Soc. Bull. 18:204- wolf in a deer decline in the Superior 210. National Forest. USDA Forest Manlove, M. N., M. H. Smith, H. O. Service Research Paper NC-148. 23 Hillestad, S. E. Fuller, P. E. Johns, pp. and D. O. Straney. 1976. Genetic MCTWS. 1975. Proc. Ann. Meeting subdivision in a herd of white-tailed Montana Chapter The Wildlife deer as demonstrated by spatial Society, Bozeman, February 20-22, shifts in gene frequencies. Proc. 1975. 102 pp.

Li t e r a t u r e Ci t e d 169 MFWP Wildl. Div. 1995. Montana deer Newby, F. 1958. How big game seasons and deer hunting; a management are set. Montana Dep. Fish and analysis. Mont. Fish, Wildl. and Game, Information Bull. No. 12. 7 pp. Parks, Helena. 55 pp. Nicholson, A. J. 1957. The self- Morgan, J. T. 1993. Summer habitat use adjustment of populations to change. of white-tailed deer on the Tally Lake Cold Spring Harbor Symp. On Ranger District, Flathead National Quantitative Biol. 22:153-173. Forest. Ph.D. Thesis. Mont. State Univ., Bozeman. 103 pp. Olenicki, T. J. 1993. Seasonal use of fecal nitrogen and forage succulence Mundinger, J. G. 1980. Population to assess condition and movements ecology and habitat relationships of of two southeastern Montana mule white-tailed deer in coniferous forest deer populations. M. S. Thesis. habitat of northwestern Montana. Mont. State Univ., Bozeman. 118 pp. Pages 8-92 in Montana deer studies. Job Progr. Rep., Fed. Aid Proj. Olson, G. R. 1986. Tiber Reservoir mule W-120-R-11. Mont. Dept. Fish, deer monitoring and investigation. Wildl. and Parks, Helena. 205 pp. Appendix 1 in Big game survey and inventory-Region Four. Job Progr. _____. 1981. White-tailed deer Rep. Fed. Aid Proj. W-130-R-17. reproductive biology in the Swan Mont. Dept. Fish, Wildl. and Parks, Valley, Montana. J. Wildl. Manage. Helena. 33 pp. 45:132-139. Ostfeld, R. S. 1992. Effects of habitat _____. 1982. Biology of the white-tailed patchiness on population dynamics: deer in the coniferous forest of a modeling approach. Pages 851- northwestern Montana. Pp 275- 863 in D. R. McCullough and R. 284 in W.R. Meehan, T. R. Merrill, H. Barrett, eds. Wildlife 2001: Jr., and T. A. Hanley, eds., Fish and populations. Elsevier Appl. Sci., Wildlife Relationships in Old Growth London. Forest, Proceedings of a Symposium American Institute of Fishery Ozoga, J. J., L. J. Verme, and C. S. Bienz. Research Biologists, 425 pp. 1982. Parturition behavior and territoriality in white-tailed deer: _____. 1984. Population ecology and impact on neonatal mortality. J. habitat relationships of white-tailed Wildl. Manage. 46:1-11. deer in coniferous forest habitat of northwestern Montana. Pages 54-63 Pac, D. F., K. L. Hamlin, and R. M. in Montana deer studies. Job Progr. Desimone. 1995. Mortality of Rep., Fed. Aid Proj. W-120-R-15. bull elk and buck mule deer in Mont. Dept. Fish, Wildl. and Parks, southwestern Montana: comparative Helena. 80 pp. implications for management. Proc. Western States and Provinces 1995 Mussehl, T. W., and F. W. Howell. 1970. Game management in Montana. Joint Deer and Elk Workshop. Idaho Mont. Fish and Game Dept., Helena. Dep. Fish and Game, Boise. p. 135 238 pp. (Abstract only). Nelson, M., and L. D. Mech. 1987. Demes _____, R. J. Mackie, and H. E. Jorgensen. within a northeastern Minnesota deer 1991. Mule deer population population. Pages 27-40 in B. D. organization, behavior, and dynamics Chepko-Sade and A. Halpin, eds., in a northern Rocky Mountain Symposium on patterns of dispersal environment. Final Rep., Fed. Aid among mammals and their effects on Proj. W-120-R-7-18. Mont. Dept. Fish, the genetic structure of populations. Wildl. and Parks, Helena. 316 pp. University of Chicago Press.

170 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a _____, and M. S. Ross. 1993. Production heterogeneity in South Carolina and survival of mule deer bucks deer populations. J. Wildl. Manage. in mountain-foothill habitats. Job 43:136-142. Progr. Rep., Fed. Aid Prog. W-100-R. Mont. Dept. Fish, Wildl. and Parks, Rausher, R. L. 1995. Deer use Helena. 27 pp. of irrigated alfalfa along the Yellowstone River, Custer County, Pac, H. I., W. F. Kasworm, L. R. Irby, and Montana. M. S. Thesis. Mont. State R. J. Mackie. 1988. Ecology of the Univ., Bozeman. 50 pp. mule deer, Odocoileus hemionus, along the east front of the Rocky Ray, A. A. 1982. SAS User’s Guide. SAS Mountains, Montana. Can. Field- Institute. Cary, NC. 921 pp. Nat. 102:227-236. Rhodes, O. E., and M. H. Smith. 1992. Parker, K. L., C. T. Robbins, T. A. Hanley, Genetic perspectives in wildlife and J. W. Thomas. 1985. The management: the case of large energetic basis of habitat selection herbivores. Pages 985-996 in D. by deer (Odocoileus hemionus) and McCullough and R. Barrett, eds. elk (Cervus elaphus). Page 458 in Wildlife 2000: Populations. Elsevier P. K. Fennessy and K. R. Drew, eds., Science Publishers Ltd., NY. Biology of deer production. Bulletin Riggs, R. A., and J. M. Peek. 1980. 22, The Royal Society of New Mountain sheep habitat-use patterns Zealand, Wellington. 482 pp. related to post-fire succession. J. Pengelly, W. L. 1976. Probable causes Wildl. Manage. 53: 197-206. of the recent decline of mule deer-a Riney, T. 1955. Evaluating condition summary. Pages 129-134 in G. of free-ranging red deer (Cervus W. Workman and J. W. Low, eds., elaphus), with special reference to Mule deer decline in the west, a New Zealand. New Zealand J. Sci. symposium. Utah St. Univ. Agr. Exp. and Technol. 36:429-463. Sta., Logan. 134 pp. Robinette, W. L., and J. S. Gashwiler. Peterson, J. G. 1996. Ecological 1950. Breeding season, productivity, implications of sagebrush and fawning period of the mule deer manipulation.: a literature review. in Utah. J. Wildl. Manage. 14:457- Mont. Dept. Fish ,Wildl. and Parks, 469. Helena. 49 pp. _____, D. A. Jones, and H. S. Crane. Porter, W. F., N. E. Mathews, H. B. 1955. Fertility of mule deer in Utah. Underwood, R. W. Sage, and D. R. J. Wildl. Manage. 19:115-136. Behrend. 1991. Social organization in deer: implications for localized _____, _____, G. Rogers, and J. S. management. Envir. Manage. Gashwiler. 1957. Notes on tooth 15:809-814. development and wear for Rocky Mountain mule deer. J. Wildl. Putman, R. 1988. The natural history Manage. 21:134-153. of deer. Cornell University Press, Ithaca, NY. 191 pp. _____, N. V. Hancock, and D. A. Jones. 1977. The Oak Creek mule deer Pyrah, D. 1984. Social distribution and herd in Utah. Publ. No. 77-15, Utah population estimates of coyotes in St. Div. Wildl. Res., Salt Lake City. north-central Montana. J. Wildl. 148 pp. Manage. 48:679-690. Romesburg, H. C. 1981. Wildlife Ramsey, P. R., J. C. Avise, M. H. science: gaining reliable knowledge. Smith, and D. F. Urbston. 1979. J. Wildl. Manage. 45:293-313. Biochemical variation and genetic

Li t e r a t u r e Ci t e d 171 Schladweiler, P. 1980. Coyote densities, Stussy, R. J. 1993.The effects of forage small mammal population indices, improvement practices on Roosevelt and big game fawn production and elk in the Oregon Coast Range. survival in the various study areas. M. S. Thesis. Oregon State Univ., Final Research Rep., Fed. Aid Proj. Corvallis. 77pp. W-120-R-11, Study NG-47.1. Mont. Dept. Fish, Wildl. and Parks, Helena. Suring, L. H., and P. A. Vohs, Jr. 1979. 78 pp. Habitat use by Columbian white- tailed deer. J. Wildl. Manage. Schlegel, M. 1976. Factors affecting 43:610-619. calf elk survival in north-central Idaho—a progress report. Proc. Swenson, J. E. 1978. Intake terrestrial West. Assoc. State Game and Fish wildlife study. Final Rep., Ecol. Serv. Comm. 56:342-355. Div., Mont. Dept. Fish and Game, Helena. 72 pp. Schoen, J. W., and M. D. Kirchoff. 1985. Seasonal distribution and home _____. 1979. Effects of a hemorrhagic range patterns of Sitka black-tailed disease epizootic on a white-tailed deer on Admiralty Island, southeast deer population in eastern Montana. Alaska. J. Wildl. Manage. 49:96- Proc. Montana Acad. Sci. 38:25-32. 103. _____. 1982. Effects of hunting on Scribner, K. T. 1993. Conservation habitat use by mule deer on mixed- genetics of managed ungulate grass prairie in Montana. Wildl. Soc. populations. Acta Theriologica 38, Bull. 10:115-120. Suppl. 2:89-101. _____, S. J. Knapp, and H. J. Wentland. Severinghaus, C. W. 1949. Tooth 1983. Winter distribution and development and wear as criteria habitat use by mule deer and white- of age in white-tailed deer. J. Wildl. tailed deer in southeastern Montana. Manage. 13:195-216. Prairie Nat. 15:97-112. Sime, C. A. 1996. Population ecology of Unsworth, J. W., F. A. Leban, D. J. white-tailed deer in northwestern Leptich, E. Garton, and P. Zager. Montana. Job Progr. Rep., Fed. Aid 1994. Aerial survey users manual, Proj. W-100-R-4. Mont. Dept. Fish, second edition. Idaho Dept. Fish Wildl. and Parks, Helena. 75 pp. and Game, Boise. 84 pp. Stansberry, B. J. 1991. Distribution, Van Horne, B. 1983. Density as a movements, and habitat use during misleading indicator of habitat spring, summer, and fall by mule quality. J. Wildl. Manage. 47: 893- deer in the north Salish Mountains, 901. Montana. M. S. Thesis. Mont. State Verme, L. J., and J. C. Holland. 1973. Univ., Bozeman. 64 pp. Reagent-dry assay of marrow fat in _____. 1996. Evaluation of bighorn white-tailed deer. J. Wildl. Manage. sheep and mule deer habitat 37:103-105. enhancements along Koocanusa Vogel, W. O. 1983. The relationship of Reservoir. Final Rept. Mont. Fish, white-tailed deer and mule deer to Wildl. and Parks. 76 pp. agriculture in the Gallatin Valley. Steerey, W. F. 1979. Distribution, range M. S. Thesis. Mont. State Univ., use and population characteristics Bozeman. 86 pp. of mule deer associated with the _____. 1989. Response of deer to Schafer Creek winter range, Bridger density and distribution of housing Mountains, Montana. M.S. Thesis, in Montana. Wildl. Soc. Bul. 17:406- Mont. State Univ., Bozeman, 119 pp. 413.

172 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Walchek, K. 1978. Epizootics: mystery Wood, A. K. 1987. Ecology of a prairie deer diseases. Montana Outdoors mule deer population. Ph.D. Thesis, 9:30-35. Montana State University, Bozeman. 205 pp. Wallmo, O. C., L. H. Carpenter, W. L. Reglin, R. B. Gill, and D. L. Baker. _____. 1988. Use of shelter by mule 1977. Evaluation of deer habitat deer. Prairie Nat. 20:15-22. on a nutritional basis. J. Range Manage. 30:122-127. _____, R. J. Mackie, and K. L. Hamlin. 1989. Ecology of sympatric Walters, C. 1986. Adaptive management populations of mule deer and white- of renewable resources. Macmillan tailed deer in a prairie environment. Publ., New York, NY. 374 pp. Mont. Dept. Fish, Wildl. and Parks, Helena. 97 pp. _____, and Hilborn. 1978. Ecological optimization and adaptive _____, _____, and G. L. Dusek. 1994. management. Ann. Rev. Ecol. Syst. Where the species come together. 9:157-188. Pages 344-350 in D. Gerlach, S. Atwater, and J. Schnell, eds. Deer. Watt, K. E. F. 1962. Use of mathematics Stackpole Books, Mechanicsburg, in population ecology. Ann. Rev. PA. 384 pp. Ent. 7:243-252. _____., R. E. Short, A. E. Darling, G. L. White, G. C., R. A. Garrott, R. M. Dusek, R. G. Sasser, and C. A. Ruder. Bartmann, L. H. Carpenter, and A. 1986. Serum assays for detecting W. Alldredge. 1987. Survival of pregnancy in mule and white-tailed mule deer in northwest Colorado. J. deer. J. Wildl. Manage. 50: 648- Wildl. Manage. 51:852-859. 687. Wilkins, B. T. 1957. Range use, food Workman, G. W., and J. B. Low. 1976. habits, and agricultural relationships Mule deer decline in the west: a of mule deer, Bridger Mountains, symposium. Utah State University, Montana. J. Wildl. Manage. 21:159- Agr. Exp. Sta., Logan. 134 pp. 169. Youmans, H. B. 1979. Habitat use by Williams, B. K., and F. A. Johnson. 1995. mule deer of the Armstrong winter Adaptive management and the range. M. S. Thesis, Mont. State regulation of waterfowl harvests. Univ., Bozeman. 66 pp. Wildl. Soc. Bull. 23:430-436. _____, and J. E. Swenson. 1982. Big _____. 1956. Range use, food habits, and game survey and inventory (deer) agricultural relationships of mule Region Seven. Prog. Rept. Fed. Aid deer, Bridger Mountains, Montana. J. Proj. W-130-R-13, Job I-7. Mont. Wildl. Manage. 21:159-169. Dept. Fish, Wildl. and Parks, Helena. Wilson, E. O. 1975. Sociobiology: the 47 pp. new synthesis. Harvard University Zar, J. H. 1984. Biostatistical Analysis. Press, Cambridge, MA. 697 pp. Second Ed. Prentice-Hall, Englewood Cliffs, NJ. 718 pp.

Li t e r a t u r e Ci t e d 173 Appendix

A List of Publications Resulting from Statewide Deer Research Studies

Arno, F., G. E. Gruell, J. G. Mundinger, Dusek, G. L. 1984. Some relationships and W. C. Schmidt. 1987. Developing between white-tailed deer and silvicultural prescriptions to provide agriculture on the lower Yellowstone both winter deer habitat and timber. River. Pages 27-33 in Agriculture Western Wildlands. Winter 1987:19-24. and Wildlife. Proc. Montana Chapt. of The Wildlife Society. Boggs, K. W. 1984. Succession in riparian communities of the lower Yellowstone _____. 1987. Ecology of white-tailed deer River. M.S. Thesis, Montana State in upland ponderosa pine habitats in Univ., Bozeman. 107 pp. southeastern Montana. Prairie Nat. 19:1-17. Compton, B. B. 1986. Use of agricultural crop types by white-tailed deer. Proc. _____. 1990. Displacement of white-tailed Montana Acad. Sci. 46:5-18. deer by flooding. Prairie Nat. 22:57- 58. _____. 1986. Distribution of white-tailed deer along the lower Yellowstone _____. 1990. Two worlds of the whitetail. River. M.S. Thesis. Montana State Montana Outdoors 21(6):2-7. Univ., Bozeman. 73 pp. _____. 1994. Dentition of deer. Pages 71- _____, R. J. Mackie, and G. L. Dusek. 75 in D. Gerlach, S. Atwater, and J. 1988. Factors influencing Schnell, eds. Deer. Stackpole Books, distribution of white-tailed deer in Mechanicsburg PA. 384 pp. riparian habitats. J. Wildl. Manage. 52:544-548. _____, B. B. Compton, and R. J. Mackie. 1986. Relationships between white- Cronin, M. A. 1989. Molecular tailed deer and a free-flowing river in evolutionary genetics and phylogeny eastern Montana. Abstr. 31st Annual of cervids. Ph.D. Thesis, Yale Univ., Meeting Central Mountains and Great New Haven, CT. Plains Section of The Wildlife Society. _____, M. E. Nelson, and D. F. Pac. _____, and R. D. Mace. 1991. Application 1991. Spatial heterogeneity of of remote photography surveys to mitochondrial DNA and allozymes ungulate research and management. among populations of white-tailed Pages 240-243 in A. G. Christensen, deer and mule deer. J. Heredity. L. J. Lyon, and T. N. Lonner, comps., 82:118-127. Proc. Elk Vulnerability Symp., Montana State Univ., Bozeman. 330 Dood, A. R. 1978. Summer movements, pp. habitat use, and mortality of mule deer fawns in the Missouri River Breaks, _____, and R. J. Mackie. 1988. Factors Montana. M.S. Thesis, Montana State influencing distribution and Univ., Bozeman. 55 pp. abundance of white-tailed deer in a

174 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a prairie riverine environment. Abstr. management of major wildlife Ann. Meeting of the N.W. Section of species on rangeland in northcentral The Wildlife Society. Montana. Proc. West. Assoc. Fish and Wildlife Agencies 62:112-119. _____, and _____. 1988. Population ecology and habitat relationships Fisher, J. W., Jr., and D. F. Pac. 1994. of white-tailed deer in river bottom Age determination of Rocky habitat in eastern Montana. Final Mountain mule deer (Odocoileus Report, Research Project No. hemionus hemionus) based on W-120-R-12-18. Montana Fish, mandibular dentition. Abstr. 7th Wildl. and Parks Dept., Helena. 116 Cong. of the International Council for pp. Archaeozoology, Konstanz, Germany. _____, _____, J. D. Herriges, and B. B. Fritzen, D. E. 1995. Ecology and behavior Compton. 1989. Population ecology of mule deer on the Rosebud Coal of white-tailed deer along the lower Mine, Montana. Ph.D. Thesis, Yellowstone River. Wildl. Monogr. Montana State Univ., Bozeman. 143 No. 104. 68 pp. pp. _____, and J. T. Morgan. 1991. A camera _____, and R. J. Mackie. 1996. Living system to evaluate population with mule deer in Colstrip, Montana. parameters and spatial relationships Booklet. Western Energy Co., of deer. Proc. Third International Colstrip, MT. 11 pp. Cong. on Natural Resources and Wildlife. Guadalajara, Mexico. Hamlin, K. L. 1977. Population dynamics and habitat relationships of mule deer _____, S. J. Riley, J. G. Mundinger, and in the Bridger Mountains, Montana. R. J. Mackie. 1987. The implication 1955-1976. Abstr. Ann. Conf. of reproductive strategy to body N.W. Section, The Wildlife Society, weight in young adult female white- Kalispell, MT. tailed deer. Pages 555-558 in B. Bobek, K. Perzanowski, and W. L. _____. 1978. Deer, coyote and alternate Reglin, eds. Global Trends in Wildlife prey relations in the Missouri River Management Trans. XVIII Congress Breaks, Montana. Abstr. Ann. Conf. Int. Union of Game Biolo., Krakow, N.W. Section of The Wildlife Society. Poland, 1987. Vol. 1. Swiat Press. _____. 1985. Observability indexes for Krakow-Warszowa. 660 pp. aerial surveys of deer in Montana. _____, and A. K. Wood. 1986. An Paper presented at Western Deer evaluation of the average activity Workshop. Bozeman, MT. March 4-6, radius as an estimator of monthly 1985. movements of deer. Proc. Montana _____. 1997. Survival and home Acad. Sci. 46:19-26. range fidelity of coyotes in _____, _____, and R. J. Mackie. 1988. Montana: implications for control. Habitat use by white-tailed deer Intermountain J. of Sciences 3:62-72. in prairie-agricultural habitat in _____ , and G. L. Erickson. 1996. Montana. Prairie Nat. 20:135-142. Draft environmental assessment _____, _____, and S. T. Stewart. 1992. of proposed changes in hunting Spatial and temporal patterns of season structure for mule deer in mortality among female white-tailed southwestern Montana. Mont. Dept. deer. J. Wildl. Manage. 56:645-650. Fish, Wildl. And Parks, Helena. 124 pp. Eng, R. L., and R. J. Mackie. 1982. Integrating grazing-range ______, and R. J. Mackie. 1989. Mule management practices with deer in the Missouri River Breaks, Montana: A study of population

Appe n d i x 175 dynamics in a fluctuating MS thesis, Montana State Univ., environment. Final Rep., Fed. Aid Bozeman. 131 pp. in Wildlife Restor. Proj. W-120-R. Mont. Dept. Fish, Wildl. and Parks, Irby, L. R., R. J. Mackie, H. I. Pac, and W. Helena. 401 pp. F. Kasworm. 1987. Management of mule deer in relation to oil and gas _____, and _____. 1991. Age-specific development in Montana’s overthrust reproduction and mortality in belt. Pages 113-121 in J. Emerick et female mule deer: implications to al. eds., Proceedings III: Issues and population dynamics. Pages 569- Technology in the Management of 573 in B. Bobek, K. Perzanowski, Impacted Wildlife. Thorne Ecological and W. L. Reglin, eds. Global Trends Institute, Boulder, CO. in Wildlife Management Trans. XVIII Congress Int. Union of Game Biol., Ihsle, H. B. 1982. Population ecology Krakow, Poland, 1987. Vol. 1 Swiat of mule deer with emphasis on Press. Krakow-Warszowa. 660 pp. potential impacts of gas and oil development along the east slope of _____, _____, J. G. Mundinger, and D. the Rocky Mountains, northcentral F. Pac. 1982. Effects of capture Montana. M.S. Thesis, Montana State and marking on subsequent fawn Univ., Bozeman. 85 pp. production in deer. J. Wildl. Manage. 46:1086-1089. Jackson, S. D. 1990. Ecology of mule deer on a sagebrush-grassland _____, D. F. Pac, and R. M. DeSimone. habitat in northeastern Montana. 1995. Effect of elk population M.S. Thesis. Montana State Univ., increase on deer in Montana. Abstr. Bozeman. 111 pp. No. 9. Proc. Western States and Provinces 1995 Joint Deer and Elk Janke, D. M. 1977. White-tailed deer Workshop. Idaho Dept. of Fish and population characteristics, Game, Boise. 139 pp. movements, and winter site selection in western Montana. M.S. Thesis, _____, D. F. Pac, and R. J. Mackie. 1987. Univ. of Montana, Missoula. 92 pp. Food for thought on the trophy- quality hunting issue. A special Kasworm, W. F. 1981. Distribution research report reviewing factors and population characteristics of influencing the availability of mule deer along the East Front, trophy bucks and opportunities for northcentral Montana. M.S. Thesis, establishing special, quality-type Montana State Univ., Bozeman. 73 hunting area in Montana. Special pp. Research Rep., Mont. Dept. Fish, Leach, R. H. 1982. Summer range ecology Wildl. and Parks, Helena. 26 pp. of white-tailed deer in the coniferous _____, S. J. Riley, D. B. Pyrah, A. R. forests of northwestern Montana. Dood, and R. J. Mackie. 1984. M.S. Thesis, Univ. of Montana, Relationships among mule deer fawn Missoula. 80 pp. mortality, coyote, and alternate prey Longhurst, W. M., A. L. Lesperanse, M. species. J. Wildl. Manage. 48:488- Morse, R. J. Mackie, O. L. Neal, H. 499. Salwasser, D. Swickland, P. J. Urness, _____ and L. Schweitzer. 1979. and J. O. Yoakum. 1983. Livestock Cooperation by coyote pairs and wild ungulates. Pages 42-64 in attacking mule deer fawns. J. Proc. Workshop on livestock and Mammal. 78:849-850. wildlife-fisheries relationships in the Great Basin. Spl. Publ. 3301. Herriges, J. D. 1986. Movement, activity, Univ. of California, Div. of Agr. Sci., and habitat use of white-tailed deer Berkeley. 173 pp. along the lower Yellowstone River.

176 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a Lonner, T. N., and R. J. Mackie. 1983. ed. Symposium on Western Elk Interactions between big game and Management. Utah State Univ., Logan. livestock. Pages 53-58 in B. F. 213 pp. Roche, Jr. and D. M. Baumgartner, eds. Forestland Grazing: Proc. of a _____. 1987. Mule Deer. Pages 265-271 in Symposium, Coop. Ext. Serv., Wash. H. Kallman ed., Restoring America’s State Univ., Pullman. 114 pp. Wildlife 1937-1987. U.S.D.I., Fish and Wildlife Service, Washington, D.C. _____, and D. F. Pac. 1990. Elk and 294 pp. mule deer use of whitebark pine forests in southwestern Montana: _____. 1988. Mule Deer. North American An ecological perspective. Pages Deer Foundation 1:10-17. 237-244 in Proc. of Sympos. _____. 1988. What’s in an trend? Abstr. on Whitebark Pine ecosystems: Ann. Meeting of the Montana Chapt. Ecology and management of a high- of The Wildlife Society, Lewistown, mountain resource. USDA Forest M T. Service, Intermount. Research Sta. Gen. Tech. Rept. INT. 270. 386 pp. _____. 1991. Big game historical perspective. Abstr. Livestock- Mackie, R. J. 1976. Interspecific Big Game Symposium, Reno, NV. competition between mule deer, other September 18-20, 1991. game animals and livestock. Pages 49-54 in G. W. Workman and J. B. _____. 1994. Mule deer habitat. Pages Low, eds. Symposium on mule deer 286-296 in D. Gerlach, S. Atwater, decline in the west. Utah State Univ., and J. Schnell, eds. Deer. Stackpole Coll. of Nat. Resour., Agr. Exp. Sta., Books, Mechanicsburg, PA. 384 pp. Logan. 134 pp. _____. 1994. Reacting to weather. Pages _____. 1978. Impacts of livestock grazing 297-300 in D. Gerlach, S. Atwater, on wild ungulates. Trans. N. Amer. and J. Schnell, eds. Deer. Stackpole Wildl. Nat. Resour. Conf. 43:462-476. Books, Mechanicsburg, PA. 384 pp. _____. 1979. “Competition” and the _____. 1994. Populations and habitat. future of wild ungulates on Montana Pages 322-328 in D. Gerlach, S. rangelands. Abstr. Joint Meeting of Atwater, and J. Schnell, eds. Deer. the Montana Chapters Soil Cons. Stackpole Books, Mechanicsburg, PA. Soc. of Amer., Amer. Fisheries Soc., 384 pp. Soc. Amer. Foresters and the Wildlife _____. 1994. Longevity. Pages 328-329 Society, Missoula, MT. in D. Gerlach, S. Atwater, and J. _____. 1981. Interspecific relations. Schnell, eds. Deer. Stackpole Books, Pages 487-507 in O. C. Wallmo, ed., Mechanicsburg, PA. 384 pp. Mule and Black-tailed Deer of North _____. 1994. The subspecies of mule America. Univ. of Nebraska Press. deer. Pages 346-350 in D. Gerlach, 605 pp. S. Atwater, and J. Schnell, eds. Deer. _____. 1983. Natural regulation of mule Stackpole Books, Mechanicsburg, PA. deer population. Pages 112-125 in 384 pp. D. S. Eastman, F. L. Bannnull and J. _____. 1995. Mule deer, elk, and white- M. Peck, eds., Natural regulation of tails: recent trends and future wildlife. Proc. No. 14, Univ. of Idaho management in an ecosystem context. For., Wildl. and Range Exp. Sta., Abstr. No. 4 in Proc. Western States Moscow. 225 pp. and Provinces 1995 Joint Deer and _____. 1985. The deer-elk-cattle triangle. Elk Workshop. Idaho Dept. Fish and Pages 51-56 in G. W. Workman, Game, Boise. 139 pp.

Appe n d i x 177 _____, and G. Bujalska. 1991. Convenor’s _____, and G. L. Dusek. 1992. Trapping report: wildlife population dynamics white-tailed deer on summer range. and regulation. Pages 657-658 in Wildl. Soc. Bull. 20:39-41. B. Bobek, K. Perzanowski, and W. L. Reglin, eds. Global Trends in Morton, M. A. 1976. Nutritional values Wildlife Management. Trans. XVIII of major mule deer winter forage Congress Int. Union of Game Biol., species in the Bridger Mountains, Krakow, Poland, 1987. Vol. 1. Swiat Montana. M.S. Thesis. Montana Press. Krakow-Warszowa. 660 pp. State Univ., Bozeman. 104 p. _____, and G. L. Dusek. 1992. Mundinger, J. G. 1979. The Swan Valley: Deer habitat relationships and White-tail Country. Montana management in the northern Rocky Outdoors 10(5):35-41. Mountains and Great Plains. Western _____. 1981. White-tailed deer Wildlands 18:14-19. reproductive biology in the Swan _____, K. L. Hamlin, and D. F. Pac. 1981. Valley, Montana. J. Wildl. Manage. Census methods for mule deer. Pages 45:132-139. 97-106 in F. L. Miller and A. Gunn, _____. 1981. Impacts of timber harvest on eds. Symposium on census and white-tailed deer in the coniferous inventory methods for population forests of northwestern Montana. and habitats. Univ. of Idaho. Forest, Abstr. Ann. Conf. of the N.W. Section Wildlife and Range Experiment of The Wildlife Society. Station, Contribution No. 217. 220 pp. _____. 1984. Biology of the white-tailed deer in the coniferous forests of _____, _____, and _____. 1982. Mule northwestern Montana. Pages 275- deer. Chapter 44, pages 862- 284 in W. R. Meehan, T. R. Merrell, 867 in J. A. Chapman and G. A. Jr., and T. A. Hanley eds., Fish and Feldhammer, eds. Wild Mammals wildlife relationships in old-growth of North America. J. Hopkins Univ. forests. Proc. of a Symposium. Press. 1147 pp. Amer. Inst. of Fishery Research _____, _____, _____, G. L. Dusek, and Biologists. 425 pp. A. K. Wood. 1990. Compensation Nyberg, H. E. 1980. Distribution, in free-ranging deer populations. movements and habitat use of mule Trans. N. Amer. Wildl. Natur. Resour. deer associated with the Brackett Conf. 55:518-526. Creek winter range, Bridger _____, and D. F. Pac. 1980. Deer Mountains, Montana. M.S. Thesis. and subdivisions in the Bridger Montana State Univ., Bozeman. 106 Mountains, Montana. Proc. West pp. Assoc. of Fish and Wildlife Agencies O’Conner, K. S. 1987. Ecology of 60:517-526. white-tailed deer and mule deer in _____, A. K. Wood, and G. L. Dusek. 1994. agricultural lands in the Gallatin Deer and other ungulates. Pages 351- Valley, Montana. M.S. Thesis. 358 in D. Gerlach, S. Atwater, and J. Montana State Univ., Bozeman. 67 Schnell, eds. Deer. Stackpole Books, pp. Mechanicsburg, PA. 384 pp. Pac, D. F. 1976. Distribution, movements, Morgan, J. T. 1993. Summer habitat use of and habitat use during spring, white-tailed deer on the Talley Lake summer, and fall by mule deer Ranger District, Flathead National associated with the Armstrong Forest. Ph.D. Thesis, Montana State Winter Range, Bridger Mountains, Univ., Bozeman. 103 pp. Montana. M.S. Thesis. Montana State Univ., Bozeman. 120 pp.

178 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a _____. 1982. When you’re on top, its hard Pyrah, D. B. 1984. Social distribution to remember the bottom. Pages 40- and population estimates of coyotes 42 in Practical Application of Recent in northcentral Montana. J. Wildl. Research. Proc. of Montana Chapter Manage. 48:679-690. of The Wildlife Society. Riley, S. J. 1982. Survival and behavior of _____, K. L. Hamlin, and R. M. DeSimone. radio-collared mule deer fawns during 1995. Mortality of bull elk and summers, 1978-1980, in the Missouri buck mule deer in southwestern River Breaks, Montana. M.S. Thesis. Montana: comparative implications Montana State Univ., Bozeman. 59 for management. Abstr. No. 42 in pp. Proc. Western States and Provinces 1995 Joint Deer and Elk Workshop. _____, and A. R. Dood. 1984. Summer Idaho Dept. Fish and Game, Boise. movements, home range, habitat use, 139 pp. and behavior of mule deer fawns. J. Wildl. Manage. 48:1302-1310. _____, and R. J. Mackie. 1981. Conflict for space. Montana Outdoors Rosgaard, A. I., Jr. 1981. Ecology of 12(2):13-16. the mule deer associated with the Brackett Creek winter range in _____, _____, and H. E. Jorgensen. 1984. the Bridger Mountains, Montana. Relationships between mule deer M.S. Thesis. Montana State Univ., and forest in southwestern Montana Bozeman. 76 pp. - some precautionary observations. Pages 321-328 in W. R. Meehan, Sasser, R. G., C. A. Ruder, A. K. Wood, R. T. R. Merrell, Jr., and T. A. Hanley E. Short, C. Robbins, D. Houston, eds., Fish and wildlife relationships and D. Palmisciano. 1985. Detection in old growth forests. Proc. of a of pregnancy in wild ruminants by Symposium. Amer. Inst. of Fishery measurement of a pregnancy-specific Research Biologists. 425 pp. protein in serum. Proc. Amer. Assoc. Zool. Vet. _____, _____, and _____. 1991. Mule deer population organization, behavior, _____, _____, K. A. Ivani, R. E. Short, and and dynamics in a northern Rocky A. K. Wood. 1985. Pregnancy-specific Mountain environment. Final Rep., Protein B in serum of various species. Fed. Aid in Wildlife Restor. Proj. Pages 161-163 in F. Ellendorff and E. W-120-R. Mont. Dept. Fish, Wildl. Koch, eds., Early Pregnancy Factors. and Parks, Helena. 316 pp. Proceedings of an international workshop in Mariensee, West Pac, H. I., W. F. Kasworm, L. R. Irby, and Germany. Perinatology Press. R. J. Mackie. 1988. Ecology of the mule deer, Odocoileus hemionus, Schladweiler, P. 1980. The effects of along the east front of the Rocky coyotes on big game populations in Mountains, Montana. Canadian Montana. Fed. Aid Job Final Rept., Field-Naturalist 102:227-236. Proj. W-120-R. Mont. Dept. Fish, Wildl. and Parks, Helena. 78 pp. Peek, J. M., R. J. Mackie, and G. L. Dusek. 1990. Over-winter survival strategies _____. 1981. Coyotes and big game. of North American Cervidae. Montana Outdoors 12(2):17-20, 29. Proc. 3rd Int. Moose Symposium. _____, K. L. Hamlin, and D. B. Pyrah. Syktyvkar, USSR. Alces Supplement 1984. Coyote food habits, prey 7:156-161. availability, and relationships to Picton, H. D. and R. J. Mackie. 1980. predation on deer in the Missouri Single species island biogeography River Breaks, MT. Paper presented at and Montana mule deer. Biological Third Predator Symposium, Missoula, Conservation. 19:41-49. M T.

Appe n d i x 179 Slott, B. J. 1979. White-tailed deer in Montana. Spl. Research Rept., movements, minimal and population Montana Dept. of Fish, Wildl. and characters in the Clearwater River Parks, Helena. 26 pp. Drainage, Montana. MS Thesis, Univ. of Montana, Missoula. 62 pp. _____, and _____. 1987. What’s working and what’s not: an overview of Stansberry, B. J. 1991. Distribution, approaches to management for quality movements, and habitat use during hunting. Proc. West. Assoc. Fish and spring, summer, and fall by mule Wildlife Agencies 62:112-119. deer in the North Salish Mountains, Montana. M.S. Thesis, Montana State Wood, A. K. 1987. Ecology of a prairie Univ., Bozeman. 64 pp. mule deer population. Ph.D. Thesis. Montana State Univ., Bozeman. 205 Steerey, W. F. 1979. Distribution, range use pp. and population characteristics of mule deer associated with the Schafer Creek _____. 1988. Use of shelter by mule deer winter range, Bridger Mountains, during winter. Prairie Nat. 20:15-22. Montana. M.S. Thesis. Montana State Univ., Bozeman. 119 pp. _____. 1989. Comparative distribution and habitat use by pronghorn antelope and Trout, R. G. 1978. Small mammal mule deer. J. Mammal. 70:335-340. abundance and distribution in the Missouri River Breaks, Montana. M.S. _____, G. L. Dusek, and R. E. Short. 1986. thesis. Montana State Univ., Bozeman. Effects of capture method on stress, 64 pp. survival, and pregnancy diagnosis in deer. Proc. Montana Acad. Sci. 46:75- Unsworth, J. W., D. F. Pac, G. C. White, 80. and R. M. Bartmann. In prep. Mule deer survival in Colorado, Idaho, and _____, R. J. Mackie, and G. L. Dusek. 1990. Montana. Submitted J. Wildl. Manage. Statewide trends on white-tailed deer Jan. 1998, 23 pp. distribution, fawn recruitment and Vogel, W. 1983. The effects of housing harvest. Proc. Montana Chapter of developments and agriculture on the the Wildlife Society. ecology of white-tailed deer and mule _____, _____, and _____. 1994. Where deer in the Gallatin Valley, Montana. species come together. Pages 344- M.S. Thesis, Montana State Univ., 350 in D. Gerlach, S. Atwater, and J. Bozeman. 100 pp. Schnell, eds. Deer. Stackpole Books, _____. 1984. Deer, agriculture, and housing Mechanicsburg, PA. 384 pp. developments in the Gallatin Valley, _____, _____, and K. L. Hamlin. 1989. Montana. in Agriculture and Wildlife. Proc. Montana Chapter of the Wildlife Ecology of sympatric mule and white- Society. Pp. 52-56. tailed deer populations in a prairie environment. Tech. Bull., Montana _____. 1989. Response of deer to density Dept. of Fish, Wildlife and Parks, and distribution of housing in Helena. 98 pp. Montana. Wildl. Soc. Bull. 17:406- 413. _____, R. E. Short, A. E. Darling, G. L. Dusek, R. G. Sasser, and C. A. Ruder. Watts, C. R., L. C. Eichhorn, and R. J. 1986. Serum assays for detecting Mackie. 1987. Vegetation trends pregnancy in mule and white-tailed associated with season-long and rest- deer. J. Wild. Manage. 50:648-687. rotation grazing on breaks-type range. J. Range Manag. 40:393-396. Youmans, H. B. 1979. Habitat use by mule deer of the Armstrong Winter Range, Weigand, J. P., and R. J. Mackie. 1985. A Bridger Mountains, Montana. M.S. review of winter feeding of big game Thesis, Montana State University, animals and potential applications Bozeman. 66 pp.

180 Ec o l o g y a n d Ma n a g eme n t o f Dee r i n Mo n t a n a