EFFECTS OF RESOURCE EXTRACTION INDUSTRIES ON BEHAVIOUR AND POPULATION DYNAMICS OF GRIZZLY BEARS I N THE FLATHEAD DRAINAGE, BRITISH COLUMBIA AND MONTANA. By BRUCE NORMAN MCLELLAN B.Sc, The Univer s i t y of B r i t i s h Columbia, 1976 M.Sc, The Univer s i t y of B r i t i s h Columbia, 1983 A THESIS SUBMITTED I N PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGSEZ OF DOCTOR OF PHILOSOPHY i n THE.FACULTY OF GRADUATE STUDIES (Department of Animal Science) We accept t h i s t h e s i s as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1989 (c)Bruce Norman McLellan, 1989 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r a n a d v a n c e d d e g r e e a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l m a k e i t f r e e l y a v a i l a b l e f o r r e f e r e n c e a n d s t u d y . I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s m a y b e g r a n t e d b y t h e h e a d o f m y d e p a r t m e n t o r b y h i s o r h e r r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t b e a l l o w e d w i t h o u t m y w r i t t e n p e r m i s s i o n . D e p a r t m e n t T h e U n i v e r s i t y o f B r i t i s h C o l u m b i a V a n c o u v e r , C a n a d a D E - 6 (2/88) ABSTRACT i i The range and numbers of g r i z z l y bears (Ursus arctos L.) have been gre a t l y reduced since Europeans s e t t l e d North America and there i s concern that various human a c t i v i t i e s threaten many of the remaining bears. In t h i s t h e s i s I examine r e l a t i o n s h i p s between resource extraction i n d u s t r i e s and g r i z z l y bear ecology, behaviour, and population dynamics i a an area uridexg.oxng timber harvest and- gas exploration. The major hypothesis investigated was that the i n d u s t r i a l a c t i v i t i e s i n the study area would be detrimental to the g r i z z l y population and I predicted e i t h e r a low density of bears compared to other i n t e r i o r populations, or at l e a s t a negative rate of increase. To determine s p e c i f i c causes for any observed population response, I monitored behavioural reactions of g r i z z l y bears to i n d u s t r i a l a c t i v i t i e s and habitats modified by these a c t i v i t i e s . The s e l e c t i o n of seasonal home ranges and use of habitat components, elevation, and aspects by g r i z z l y bears from these ranges are presented. Based on 4872 relo c a t i o n s of 55 radio-c o l l a r e d bears over a 9 year period, g r i z z l i e s were found to generally follow 1 of 2 seasonal home range s e l e c t i o n s t r a t e g i e s . Mountain resident (MR) bears remained i n mountainous t e r r a i n for the e n t i r e year while e l e v a t i o n a l migrating (EM) bears moved down to the Flathead V a l l e y bottom twice each year. Habitats frequently used by bears were low elevation r i p a r i a n areas, snowchutes, high elevation burns, and i i i . low e l e v a t i o n timber and open timber. Cutting units were r a r e l y used. Most bears used habitats within 100 m of roads l e s s than expected, r e s u l t i n g i n an e f f e c t i v e habitat loss of 8.7%. Avoidance of roads was independent of t r a f f i c volume, suggesting that only a small number of vehicles i s s u f f i c i e n t to d isplace bears. However, roads and nearby areas avoided by bears during day were used at night. Yearlings and females with cubs used habitats near roads more than other bears. Five comparisons of bear use before, during, and a f t e r i n d u s t r i a l a c t i v i t y indicated l i t t l e displacement. Because the estimated average density of g r i z z l y bears was 2 6.4/100km , which was high f o r an i n t e r i o r population, and t h e i r estimated rate of increase was p o s i t i v e (r = 0.081), the major hypothesis was rejected. Resource extraction i n d u s t r i e s d i d contribute to g r i z z l y bear m o r t a l i t y i n d i r e c t l y , however, by making roads which provide easy access to hunters, poachers and s e t t l e r s . Of the 9 g r i z z l i e s which died when ra d i o - c o l l a r e d , 2 were k i l l e d by l e g a l hunting, 5 were i l l e g a l l y k i l l e d ; 3 of these i l l e g a l k i l l s occurred i n 4% of the study area with permanent human settlement. Vehicular access planning and post-operational control are major recommendations of t h i s study. TABLE OF CONTENTS Abstract : i i L i s t of Tables v i L i s t of Figures v i i Acknowledgements x CHAPTER ONE General Overview • 1 Study Area •; 3 Methods '• ; 6 CHAPTER TWO Seasonal home range and habitat s e l e c t i o n during a period of resource extraction 7 Methods Measuring s e l e c t i o n 8 Selection of seasonal ranges 10 Selection from within seasonal ranges 14 Comparisons among age-sex classes and reproductive.status of adult females 15 Results Selection of seasonal ranges from the study area - 15 Habitat s e l e c t i o n within seasonal ranges 21 Va r i a t i o n among age-sex classes and reproductive status of adult females 24 Discussion : '• 26 Management implications 29 CHAPTER THREE The behavioural response of g r i z z l y bears to resource extraction i n d u s t r i e s 31 Methods E f f e c t s of roads • 32 E f f e c t s of i n d u s t r i a l a c t i v i t i e s 35 Treatments 36 Seismic a c t i v i t y i n the Vaccinium area — 36 Bear-industry i n t e r a c t i o n s during spring 39 Results E f f e c t s of roads 41 Individual v a r i a t i o n 43 Va r i a t i o n among age-sex classes 43 V a r i a t i o n within age-sex classes 45 Female reproductive status and responses 45 Influence of t r a f f i c volume : 45 Nocturnal behaviour 47 Influence of habitat type 47 E f f e c t s of i n d u s t r i a l a c t i v i t i e s 49 Seismic a c t i v i t y i n the Vaccinium area 49 Bear-industry i n t e r a c t i o n s during spring 54 Long-term e f f e c t s on the bears 55 Discussion Loss of habitats near roads — — • 56 Demographic consequences of roads 58 Displacement of bears by resource industry a c t i v i t i e s — 6Q Overt versus covert responses 61 Management implications — — ;—• • 63 CHAPTER FOUR Dynamics of a g r i z z l y bear population during a period of i n d u s t r i a l resource extraction 64 Methods Density estimation 65 Age and sex composition 72 Survival rates and causes of death 72 N a t a l i t y and rate of increase 75 Results Densities 77 Age-sex composition 78 Survival rates 81 Causes of mortality • 83 Reproductive parameters • 86 Reproductive rates 87 Survival-fecundity rate of increase 89 Discussion Methods f o r estimating density • — 9 2 Population density and composition 93 Survival rates • 95 Causes of mortality • 95 N a t a l i t y • 97 Rate of increase • 100 E f f e c t s of industry. • 100 CHAPTER FIVE Conclusions — 102 LITERATURE CITED • 108 v i Table 1. Table 2. Table 3. LIST OF TABLES Habitat component descriptions f o r the Flathead study area a f t e r Servheen (1983) and Zager et a l . (1983). •--Page 12 Proportions of a l l relocations of 4 radio c o l l a r e d g r i z z l y bears i n each of 3 distance zones from active seismic operations i n a 12 km Vaccinium habitat. Proportions are shown fo r before, during and a f t e r the seismic operations. The l e v e l s of p r o b a b i l i t y of equal d i s t r i b u t i o n s among the zones (P) and sample sizes (n) are shown. 52 Extent of resource extraction i n d u s t r i e s and residences i n the Flathead study area as of 1986. 69 Table 4. Number of bears i n Trapping Area B each year determined by the proportion of 97.5% m u l t i -annual home ranges which overlapped t h i s area. '• 79 Table 5. Number of bears i n trapping area B determined " by the proportion of t h e i r multi-annual a e r i a l relocations (= time) which occurred i n t h i s area. :- 80 Table 6. The best estimates of annual s u r v i v a l rates of the r a d i o - c o l l a r e d g r i z z l y bears and t h e i r dependent o f f s p r i n g i n the Flathead Valley, 1979 to 1987. 82 Table 7. Seasonal s u r v i v a l rates of r a d i o - c o l l a r e d g r i z z l y bears i n the Flathead V a l l e y , 1979 to 1987. — 84 Table 8. Known and suspected g r i z z l y bear m o r t a l i t i e s i n the 2900 km study area, Flathead Valley, 1979 to 1987. • 85 Table 9. L i t t e r s i z e s t a t i s t i c s of g r i z z l y bears observed i n the Flathead Valley, 1979 to 1987. 88 Table 10. Values used i n the c a l c u l a t i o n of the rate of increase, r , by using the Lotka equation f o r the g r i z z l y bear population of the Flathead Valley, 1979 to 1987. — 90 v i i LIST OF FIGURES Figure Figure F- iguxe Figure Page 1. Map of the study area. The proportion of random points (coverage) located i n various habitat components (top), elevations (middle), and aspects (bottom) over the e n t i r e study area and early spring ranges of El e v a t i o n a l Migrant (EM; n=13) bears. Also shown i s the proportion of early spring radio-locations (use) of EM bears i n the habitat components, elevations, and aspects. 17 3-=. The proportion of random points (coverage) located i n various habitat components (top), elevations (middle), and aspects (bottom) over the en t i r e study area and spring ranges of E l e v a t i o n a l Migrant (EM; n=29) and Mountain Resident (MR; n=8) bears. Also shown i s the proportion of spring radio-locations (use) of EM and MR bears i n the habitat components, elevations, and aspects. 18 4. The proportion of random points (coverage) located i n various habitat components (top), elevations (middle), and aspects (bottom) over the en t i r e study area and summer ranges of E l e v a t i o n a l Migrant (EM; n=21) and Mountain Resident (MR; n=8) bears. Also shown i s the proportion of summer radio-locations (use) of EM and MR bears i n the habitat components, elevations, and aspects. 19 Figure 5. The proportion of random points (coverage) located i n various habitat components (top), elevations (middle), and aspects (bottom) over the en t i r e study area and autumn ranges of E l e v a t i o n a l Migrant (EM; n=24) and Mountain Resident (MR; n=8) bears. Also shown i s the proportion of autumn radio-locations (use) of EM and MR bears i n the habitat components, elevations, and aspects. 20 V l l l Figure 6. The paths of the 4 seismic lines-(broken, s t r a i g h t lines) across the 12 km Vaccinium spp. berry area and the dates they were cut are shown. Also i l l u s t r a t e d are the 8 minor drainages. ; 37 Figure 7. Observed versus expected use by 23 g r i z z l y bears of the 4 Distance-to-Road-Categories (DRC) i n the Flathead Valley, B.C. Data are combined f o r a l l road types, both seasons, ground and a e r i a l radio-locations, and a l l age-sex classes, between 1979 and 1985. Use d i f f e r i n g s i g n i f i c a n t l y (P < 0.05) from expected i s indicated by + or -. For differences i n n values see Methods i n t h i s chapter. -• 42 Figure 8. Relative use (proportion of radio-relocations) by 5 age-sex classes of g r i z z l y bears of the 4 Distance to Road Categories (DRC) i n the Flathead Valley, between spring and f a l l (1979 to 1985). S i g n i f i c a n t (P < 0.05) differences among classes i n t h e i r use of each DRC are indicated by d i f f e r e n t l e t t e r s . The number of radio l o c a t i o n s = n, number of bears i n parenthesis. 44 Figure 9. Relative use of the 4 Distance to Road Categories (DRC) by female g r i z z l y bears of d i f f e r e n t reproductive status during spring (top; n=5) and summer-fall (bottom n=2). Where use of a DRC d i f f e r e d s i g n i f i c a n t l y (P < 0.05) among females of d i f f e r e n t status, t h i s i s indicated by d i f f e r e n t l e t t e r s . The number of relocations = n. 46 Figure 10. Observed versus expected use by 23 g r i z z l y bears of each Distance to Road Category with d i f f e r e n t l e v e l s of t r a f f i c (Primary - top; Primary and Secondary - middle; T e r t i a r y -bottom) i n the Flathead Valley, B.C. Data represent a l l spring and summer-fall seasons between 1979 and 1985. Use d i f f e r i n g s i g n i f i c a n t l y (P < 0.05) from expected i s indicated by + or -. For dif f e r e n c e s i n n values see Methods i n t h i s chapter. 48 i x Figure 11. Proportions of combined radio locations of 2 adult females (23, 35), and 2 subadults (1 male 47, 1 female 36) g r i z z l y bears over a 3-year period, i n each of the 3 Distance Zones, before, during, and a f t e r seismic exploration. 51 Figure 12. The study area. Almost a l l trapping occurred i n the larger Area A, while the most intensive trapping e f f o r t was confined to a portion of Area A c a l l e d Area B. 67 Figure 13. Radio l o c a t i o n points and 97.5% multi-annual home range of a subadult female g r i z z l y bear. The proportion of t h i s range which, overlapped a trapping area i s the bear's contribution to the population density estimate. • 68 X ACKNOWLEDGEMENTS I am indebted to W. Carston, R. Demarchi, D. Eastman, C. Jonkel, J . Konkin, C. Servheen, and D. Shackleton f o r providing administrative, l o g i s t i c , and fund r a i s i n g assistance. I wish to acknowledge my advisory committee, Drs. R. B l a i r , F. Bunnell, A. Harestad, and A. S i n c l a i r f o r advice, guidance and t h e i r review of papers and t h i s t h e s i s . I thank the many journal reviewers f o r t h e i r suggestions on papers that form part of t h i s t h e s i s . D. Shackleton requires a s p e c i a l acknowledgement fo r h i s thorough review of a l l papers and the t h e s i s , plus the numerous discussions and continued encouragement over th_e many years that we have worked- together. This project- was supported by the B.C. F i s h and W i l d l i f e Branch, B.C. M i n i s t r y of Forests, Canadian W i l d l i f e Service Un i v e r s i t y Research Support Fund, U.S. F i s h and W i l d l i f e Service, East Kootenay Operators (7 B.C. f o r e s t r y companies), PlumCreek Timber Ltd., Crowsnest Resources Ltd., Sage Creek Coal, Westar Mining, S h e l l Canada Ltd., B.C. Guides and O u t f i t t e r s , S a f a r i International (B.C. Chapter), Boone and Crocket Club, National R i f l e Association, World W i l d l i f e Fund (Canada), the University of B r i t i s h Columbia, Fernie and Sparwood F i s h and W i l d l i f e Clubs, Margaret Bjorkman, and Mart Williams. Personal support f o r the author was provided through a B.C. Science Council GREAT Scholarship, Canadian W i l d l i f e Service Scholarship, a Klink Fellowship from the Uni v e r s i t y of B r i t i s h Columbia, a National W i l d l i f e Federation Fellowship and the National Fish and W i l d l i f e Foundation. I have been unable to give an adequate note of appreciation i n the papers published from t h i s t h e s i s to several s p e c i a l people, and t h i s I f i n d u n f a i r . I have been fortunate to have worked with very good f i e l d a s s i s t a n t s and very good friends, and I thank them a l l . Several re q u i r i n g s p e c i a l mention are Dan Carney, Dave Horning, Rick Mace, B i l l Noble, and Tim Thier. Most of a l l , however, I thank my wife Celine Doyon. For over 10 years she has l i v e d and r a i s e d a family i n a remote and p r i m i t i v e f i e l d camp. Over t h i s decade, she d i d everything from tracking bears high on snow covered mountains to being backup on a captured g r i z z l y cub. Without her help and support, there i s no doubt the study would have ended long ago. 1 CHAPTER 1: GENERAL OVERVIEW R e l a t i v e l y recent reductions i n g r i z z l y bears (Ursus arctos L.) throughout most of western North America have resulted i n bears being r e s t r i c t e d to remote and mountainous locations with l i m i t e d human i n t r u s i o n . Their apparent need f o r seclusion from people (Hamer 1974, Craighead 1976, Knight 1980) may be incompatible with e x p l o i t a t i o n and development of natural resources such as forests, o i l , gas, and minerals, over a large portion of t h e i r range. Although these i n d u s t r i a l a c t i v i t i e s are generally considered to be harmful to the g r i z z l y population (Mundy and Flook 1973, Pearson 1975, Schallenberger 1980, H o r e j s i 1986), developments are destined to occur because of the economic value of natural resources. Concurrent with resource extraction, most people desire the continued existence of remaining g r i z z l y bear populations (Hamer and Herrero 1983). I f we are to sustain both industry, and g r i z z l y bears, intensive management based on d e t a i l e d b i o l o g i c a l information i s required (Hamer and Herrero 1983). U n t i l recently, most g r i z z l y bear research focussed on population dynamics, behaviour, food habits, and habitat use, without s p e c i f i c a l l y addressing the influence of human i n d u s t r i a l a c t i v i t y (see review i n LeFranc et a l . 1987). This focus changed i n 1975 when the g r i z z l y bear i n the U.S.A. (excluding Alaska) was c l a s s i f i e d as threatened under the U.S.A. Endangered and Threatened Species Act and i t became 2 i l l e g a l to destroy or adversely modify c r i t i c a l g r i z z l y bear habitat (Jacobsen 1980). The Act highlighted the detrimental influence that human a c t i v i t i e s have had on g r i z z l y bears, and consequently, research began to focus on the r e l a t i o n s h i p between people and bears (Zager et a l . 1983, Aune et a l . 1986, Simpson et a l . 1985, Reynolds et a l . 1986, H o r e j s i 1986, B a l l a r d 1987, Mattson et a l . 1987, Archibald et a l . 1987). Although these studies were usually d i r e c t e d at behavioural responses of g r i z z l y bears to humans, the v a r i a b l e s of ultimate importance are those determining population dynamics (S-hank 1979) . I investigated both behavioural and population responses of g r i z z l y bears to timber harvest, gas exploration, and, to a l e s s e r degree, to human settlement and hunting i n the Flathead drainage between 1979 and 1987. Linking the two l e v e l s of responses, behavioural and demographic, allowed me to i d e n t i f y how the g r i z z l y population was influenced by people. I t also provided a focus for management pr e s c r i p t i o n s on the most important factors, while reducing emphasis on l e s s c r i t i c a l ones. This t h e s i s consists of four chapters following t h i s general introduction to the study, study area, and general methods used. The second chapter covers the s e l e c t i o n of seasonal home ranges from the study area, and habitats, elevations and aspects from within these ranges by g r i z z l y bears during a period of resource extraction. The t h i r d chapter consists of modifications of two papers 3 (McLellan and Shackleton 1988 a, McLellan and Shackleton i n press a) that address the behavioural responses of g r i z z l y bears to roads, seismic exploration, timber harvest, and road maintenance. Chapter 4 t r e a t s the population dynamics of the g r i z z l y population during the period of resource extraction. I t combines 3 papers; one on density and age-sex composition of bears, one on mortality rates and causes of death, and the t h i r d on n a t a l i t y rates and rate of population increase (McLellan i n press a, b, and c ) . The f i n a l chapter contains a summary/ general discussion and conclusions. STUDY AREA 2 The study area (Figure 1) was the 2900 km composite home range of g r i z z l y bears captured and r a d i o - c o l l a r e d i n the North Fork of the Flathead River drainage i n south-eastern B r i t i s h Columbia and adjacent Montana (114°85 /W; 49°l'N). This r i v e r flows through the area at an elevation ranging from 1310 to 1200 m. The 5 to 10 km wide, f l a t bottomed v a l l e y contains extensive r i p a r i a n areas which support several key spring and autumn bear foods. Large portions of the area were burned during the l a t e 1800's and ear l y 1900's (Zager et a l . 1983). Lodgepole pine (Pinus contorta) i s the dominant tree species i n the v a l l e y . In several locations, buffalo berry (Shepherdia canadensis), an important summer bear food, grows i n abundance. During the mid 1970's, a mountain pine beetle (Dendroctonus ponderosae) epidemic k i l l e d many pines, and during the following decade a large proportion of the trees were harvested J Alberta a r British Columbia Montana I 1 I I 11 I O 10 km Figure 1. Map of the study area. 5 by c l e a r - c u t logging. On e i t h e r side of the wide v a l l e y , subranges of the Rocky Mountains r i s e to >2800 m. In the side v a l l e y s , sub-alpine f i r (Abies lasiocarpa) , spruce (Picea enqelmanni x P^ . glauca) and Douglas-fir (Pseudotsuqa menziesii) are common. Many of these higher timbered s i t e s suffered an i n f e s t a t i o n of spruce bark beetles (Dendroctonus pbesus) during the 1950's and early 1960's, and again during the early 1980's, which k i l l e d trees, many of these were salvaged by clear-cut logging. Extensive f o r e s t f i r e s burned some high elevation areas e a r l y i n the century. Some of these areas currently produce huckleberries (Vaccinium spp.), another important g r i z z l y bear food during l a t e summer and autumn. Rock outcrops, alpine and subalpine meadows, and snowchutes are common i n mountainous portions of the study area. The i n t e n s i t y of i n d u s t r i a l a c t i v i t y i n the study area varied g r e a t l y within and between years. At l e a s t some timber harvest occurred each year of the study, with a maximum of about 200 f o r e s t r y employees working at one time. Ongoing gas exploration began i n 1980 and a maximum of approximately 300 employees worked i n the study area at one time. The nearest settlement i n Canada i s 100 km from the study area, but settlement begins at the border i n the U.S.A. Many game animals, including g r i z z l y bears, are hunted i n the study area. Regulations f o r g r i z z l y hunting have become inc r e a s i n g l y r e s t r i c t i v e since 1968 when the use of b a i t for a t t r a c t i n g bears became i l l e g a l . P r i o r to 1971 when the autumn 6 season was closed, the average annual number of bears taken l e g a l l y i n the B r i t i s h Columbia Ministry of Environment Management Units i n which the study area i s located, 4-01 and 4-02, was 13.8. After 1975 when compulsory reporting of k i l l s began, t h i s number was reduced to 5.5. METHODS Capture and Radio-telemetry Using foot snares at approximately 150 trap loca t i o n s , 65 d i f f e r e n t g r i z z l y bears were captured, ear-tagged, l i p -tattooed, and had a v e s t i g i a l premolar removed f o r aging by counting cementum annuli; 55 of these bears were r a d i o - c o l l a r e d (Telonics Ltd; Mesa, Arizona; 164-166 MHz). When f i r s t captured, 3 g r i z z l i e s were cubs (0 to 1 years), 17 were yearlings (1 to 2 years), 11 were subadult males (2 to 5 years), 8 were subadult females, 16 were adult males (> 5 years) and 10 were adult females. A l l r a d i o - c o l l a r e d bears were relocated from fixed-wing a i r c r a f t approximately once each week. Because of the numerous roads i n the study area, ground tracking was r e l a t i v e l y e f f i c i e n t . Locations from the ground were us u a l l y obtained by recording a compass bearing from at l e a s t 3 separate map lo c a t i o n s . In steep t e r r a i n where signal bounce created problems, numerous approximate bearings were taken while walking at l e a s t 180° around the animal. Radio locations used to determine an i n d i v i d u a l ' s home range were obtained at lea s t one day apart. 7 CHAPTER 2: SEASONAL HOME RANGE AND HABITAT SELECTION DURING A PERIOD OF RESOURCE EXTRACTION Most g r i z z l y bears do not l i v e i n reserves or parks and many must co-exist with a v a r i e t y of human a c t i v i t i e s . Many a c t i v i t i e s such as timber harvesting, mining, ranching, gas and o i l developments, and h y d r o - e l e c t r i c r e s e r v o i r s can eliminate or at l e a s t a l t e r the physical and vegetative c h a r a c t e r i s t i c s of the bears' habitat. Even i n parks, where the env-ironment: has most protection, highways, r a i l r o a d s , and t o u r i s t f a c i l i t i e s can a f f e c t habitats and animals which depend on them. Understanding the r e l a t i o n s h i p between g r i z z l y bears and t h e i r habitat i s important i f we are to plan developments i n ways that reduce our impact on bears and minimize the p r o b a b i l i t y of dangerous bear-human encounters. Johnson (1980) suggested that resource s e l e c t i o n by animals i s h i e r a r c h i c a l . He defined f i r s t order s e l e c t i o n as that of a species' geographical range, from which an i n d i v i d u a l makes a second order s e l e c t i o n by choosing the l o c a t i o n of i t s home range. Higher orders of s e l e c t i o n include the habitats used from within a home range, choice of feeding and bedding s i t e s within a habitat, and the s e l e c t i o n of foods within a feeding s i t e . The degree of s e l e c t i o n possible and the way an animal se l e c t s probably depends upon the l e v e l considered. For example, an animal i s l e s s l i k e l y to be i n a general geographic area due to choice than due to circumstances of b i r t h . 8 Selection at t h i s l e v e l i s more a question of basic requirements of the species than of an act i v e choice by in d i v i d u a l s . When s e l e c t i n g a seasonal home range from a geographical area, an animal usually chooses a mosaic of hab i t a t s . Some of these habitats the animal w i l l require, others w i l l be l e s s important, while some may be of l i t t l e or no value but be included within the range simply because they are interspersed among those habitats that the animal does require. For an animal with large ranges due to s p e c i f i c behaviour or to habitat requirements, a preponderance of low q u a l i t y habitat may confuse an analysis of seasonal range s e l e c t i o n . Most g r i z z l y bear habitat research has been focused at the order of s e l e c t i o n of habitat components from an animal's range or study area (Zager et a l . 1983, Servheen 1983, Almack 1986). At t h i s l e v e l , the bears' a b i l i t y to know the d i s t r i b u t i o n of general habitats and r a p i d l y move about the seasonal range i s important i n determining how we measure habitat s e l e c t i o n (McLellan 1986) . In t h i s chapter I investigate 2 l e v e l s of s e l e c t i o n by various age and sex classes of g r i z z l y bears during a period of resource developments: (1) seasonal home ranges from a geographical area, and (2) habi t a t components, elevations, and aspects from within these seasonal ranges. METHODS MEASURING SELECTION Many methods have been proposed to estimate s e l e c t i o n at the l e v e l of habitat from within a study area by comparing habitats used to those a v a i l a b l e . A v a i l a b i l i t y has usually been assumed to equal the area a habitat covers (see reviews i n Alldredge and R a t t i 1986, McLellan 1986). A lack of s e l e c t i o n i s assumed i f an animal uses habitats i n the same proportion as t h e i r coverage (= a v a i l a b i l i t y ) . The method of Neu et a l . (1974), as modified by Marcum and Loftsgaarden (1980), has been used most aften i n g r i z z l y bear habitat research (Servheen 1983, Zager et a l . 1983, Almack 1986). McLellan (1986) i d e n t i f i e d shortcomings of t h i s and other methods f o r measuring habitat s e l e c t i o n . These shortcomings are p a r t i c u l a r l y prevalent when some habitats cover more area than others and are therefore assumed to be more a v a i l a b l e . This problem i s rooted i n the a b i l i t y of an animal to learn the d i s t r i b u t i o n of resources, whether these be p o t e n t i a l home ranges, habitat u n i t s or feeding s i t e s . Generally, i f an animal knows the l o c a t i o n of a l l resources and i f i t can choose any with equal e f f o r t , then a l l resources are equally a v a i l a b l e . I f the animal does not have t h i s knowledge or i f i t cannot obtain any resource at an equal cost, as i n the case of food s e l e c t i o n by a g e n e r a l i s t predator, then a measure of a v a i l a b i l i t y i s important. The process of s e l e c t i o n and how we measure i t w i l l vary among Johnson's (1980) orders of s e l e c t i o n because of differences i n what an animal can learn and because of differences i n costs required to obtain various resources. For the s e l e c t i o n order of habitat components from a seasonal home range, McLellan (1986) proposed that: (1) i f there i s r e l a t i v e l y l i t t l e cost i n moving among habitats, (2) i f resources i n the habitat are not depleted within the duration of the study, and (3) i f an i n d i v i d u a l has learned the l o c a t i o n of habitats within i t s range, then the a v a i l a b i l i t y of a habitat i s not equal to the proportion of the area i t covers, but a l l are equally a v a i l a b l e and s e l e c t i o n equals use. An omniscient animal w i l l always be at a s i t e that i t has selected at that p a r t i c u l a r time f o r a p a r t i c u l a r reason (McLellan 1986). I t i s recognized, however, that habitat s e l e c t i o n does not n e c e s s a r i l y equal habitat value and animals may " a c t i v e l y " s e l e c t habitats f o r some a c t i v i t i e s such as feeding while "passively" using other habitats. For example, an animal may use a c e r t a i n habitat to t r a v e l through only because i t l i e s between feeding s i t e s . To d i f f e r e n t i a t e among degrees of active s e l e c t i o n and passive or default use, higher orders of s e l e c t i o n must be studied and reasons for s e l e c t i o n ascertained. Selection of Seasonal Ranges Based on major changes i n foods and habitats used by g r i z z l y bears, I recognized 5 seasons: (1) Ear l y spring -emergence from dens u n t i l green-up or approximately between early A p r i l and 10 May. This season was short or non-existent fo r bears which emerged l a t e from dens. (2) Spring - from green-up u n t i l the ripening of berries or, approximately 10 May to 1 August. (3) Summer - when berries (Vaccinium spp., Shepherdia canadensis) became the dominant food or between approximately 1 August and 20 September. (4) Autumn - from when berry use decreased u n t i l denning, or approximately 20 September to mid-November. (5) Winter - when bears hibernated. Seasonal home ranges were p l o t t e d using the minimum convex polygon method (Mohr 1947). This chapter contains data c o l l e c t e d between A p r i l 1979 and December 1987. To permit comparisons among studies, my study area was; p a r t i t i o n e d into 10 habitat components (Table 1) s i m i l a r to those described by Servheen (1983), Zager et a l . (1983), and Almack (1986), and by management agencies i n Montana (Christensen and Madel 1982) and B r i t i s h Columbia (Harper 1988). Due to geology, vegetation, and disturbance h i s t o r y , habitat components were r e l a t i v e l y homogeneous, yet large enough to encompass most radio l o c a t i o n e rror polygons. The study area was also divided into 6 elevation groups (1 was < 1400 m; 2 to 5 were at 200 m i n t e r v a l s between 1400 and 2200 m; 6 was > 2200 m) , 4 aspects and f l a t (< 5% slope). To determine the c h a r a c t e r i s t i c s of the geographical area, 257 points were randomly located over the e n t i r e study area and at each point the habitat component, elevation and aspect were recorded (Marcum and Loftsgaarden 1980). The same information at an a d d i t i o n a l 1977 random points was recorded and those l y i n g within the seasonal range (x = 168 points/range) of each bear were used to estimate the c h a r a c t e r i s t i c s of t h i s range. The habitat component, elevation and aspect c h a r a c t e r i s t i c s of composite seasonal ranges of a l l bears were estimated by Table 1. Habitat component descriptions f o r the Flathead study area a f t e r Servheen (1983) and Zager (1983). Component Name Description c u t t i n g u n i t Open or p a r t i a l l y timbered s i t e where timber harvest has disturbed natural vegetation. meadow Naturally open s i t e s with r e l a t i v e l y f l a t topography. r i p a r i a n Hydrologically active with moving water that may be ephemeral. Generally timbered with dense understory. road Open, disturbed areas, cleared or graded. rocky Loose rock or s o l i d bedrock with various plant communities intermixed. burns Areas with conifers l e s s than 10 m i n height due to early succession a f t e r a w i l d f i r e . These s i t e s were often dominated by shrubs. Table 1. Continued Component Name ; Description • snowchutes Steep open areas where pe r i o d i c avalanches l i m i t s vegetation to forbs, graminoids, shrubs and stunted trees. open timber Sparsely timbered, r e l a t i v e l y f l a t areas sub-alpine generally above 1800 m i n elevation where sedge and forb f i e l d s are common. open timber A l l other areas except above with between 5 and 3 0% canopy cover. timbered A l l other areas except above with greater than 30% canopy cover. i n c l u d i n g a l l points from each i n d i v i d u a l ' s seasonal range. Because one point could be located i n the range of several bears, the sample s i z e of points representing the composite range was reduced to the number of d i f f e r e n t points used, but each point was weighted by how frequently i t was used. At the l e v e l of seasonal range s e l e c t i o n from the study area, I chose the method of Neu et a l . (1974), as modified by Marcum and Loftsgaarden:(1980), for 2 reasons. F i r s t , although a few bears may have learned the lo c a t i o n of a l l p o t e n t i a l seasonal ranges i n the study area, t h i s i s u n l i k e l y to be true for most, and so they cannot f r e e l y choose from a l l . Secondly, by t h e i r extensive nature, seasonal ranges would contain many habitat components, elevations or aspects, not because the bear desired them, but by default and so confuse analyses. To t e s t i f the seasonal ranges had an equal proportion of random points i n each habitat category as the e n t i r e area, the G t e s t was used (Sokal and Rohlf 1981). I f s i g n i f i c a n t l y d i f f e r e n t (P < 0.05), the Bonferroni z s t a t i s t i c (Marcum and Loftsgaarden 1980) was u t i l i z e d to determine which categories covered d i f f e r e n t proportions of the 2 areas. I used 90% simultaneous confidence i n t e r v a l s for habitat components (k=10), and 95% i n t e r v a l s f o r elevation (k=6) and aspect (k=5). Sel e c t i o n from within Seasonal Ranges Unlike s e l e c t i o n of seasonal ranges from the study area, bears on established ranges probably know the l o c a t i o n of most habitat u n i t s within a seasonal range at the component l e v e l and are free to choose any, independent of the area they cover. Therefore, bear s e l e c t i o n of habitat components, elevations, and aspects from within seasonal ranges w i l l equal t h e i r use. However, because of the p o s s i b i l i t y of s e l e c t i o n by default, the area covered by various habitat components, elevations and aspects i n seasonal ranges w i l l be mentioned. I w i l l address higher orders of s e l e c t i o n i n the future and t h i s should reveal the purposes various habitats are used f o r . Comparisons among Age-sex Classes and Reproductive Status of Adult Females To compare habitat components, elevations and aspects used by d i f f e r e n t age-sex classes and three reproductive states of adult females (with cubs, with yearlings or older o f f s p r i n g , and alone), the G t e s t was used (Sokal and Rohlf 1981). I f s i g n i f i c a n t l y d i f f e r e n t (P < 0.05), the Bonferroni z s t a t i s t i c was used to determine which categories the bears used d i f f e r e n t l y with the same simultaneous confidence i n t e r v a l s that were used f o r seasonal range s e l e c t i o n . RESULTS Selection of Seasonal Ranges from the Study Area G r i z z l y bears followed 1 of 2 general s t r a t e g i e s of seasonal range s e l e c t i o n . Some l i v e d i n the mountains on ei t h e r side of the Flathead V a l l e y and r a r e l y , i f ever, descended to the v a l l e y f l o o r . I c a l l e d t h i s group mountain resident (MR) bears. A second group, e l e v a t i o n a l migrating (EM) bears, moved to the v a l l e y soon a f t e r emerging from t h e i r dens and generally remained at lower elevations u n t i l b e r r i e s became r i p e i n high elevation burned-off areas. Members of both groups of bears usually moved to these burns, but most EM bears returned to the v a l l e y during the autumn. To reduce variance caused by these 2 s t r a t e g i e s , data from each group are analyzed separately i n t h i s chapter, and are only combined to investigate areas that both s t r a t e g i s t s use. Seasonal range s e l e c t i o n was determined by comparing the ranges of each group to the e n t i r e study area. During early spring, spring, and autumn, EM bears chose seasonal ranges that had d i f f e r e n t proportions of habitat components, elevations, and aspects than occurred over the e n t i r e study area ( a l l P < 0.001). In these three seasons, EM bears selected ranges with c h a r a c t e r i s t i c s t y p i c a l of the v a l l e y bottom, with s i g n i f i c a n t l y more low elevations and f l a t t e r r a i n than the e n t i r e study area and l e s s burns and open timbered subalpine habitat. The summer ranges of EM bears d i d not have s i g n i f i c a n t l y d i f f e r e n t proportions of habitat components or elevations than the study area but d i d have more west facing aspects (Figures 2, 3, 4, and 5). I n s u f f i c i e n t data were c o l l e c t e d to determine the s e l e c t i o n of e a r l y spring ranges or habitats from these ranges by MR bears. The spring, summer, and autumn ranges of MR bears, however, had s i g n i f i c a n t l y d i f f e r e n t proportions of habitat components, elevations, and aspects ( a l l P < 0.001) than the e n t i r e study area. A l l three seasonal ranges of MR 17 as-r?*3 SUB-ALPINE SNOWCHUTE ROCKY gg RIPARIAN 23 TOA0 I I MEADOW mm CUTTING UNIT 3URNS P~j OPEN TIMBER |~**| TIMBER USE EM COVERAGE AREA OS-53 22CO+ m g | | <1400m • W E S T ^ SOUTH £2 NORTH QFUAT EM EM STUDY MR MR USE COVERAGE AREA COVERAGE USE Figure 3 . The proportion of random points (coverage) located i n various h a b i t a t components (top), elevations (middle), and aspects (bottom) over the e n t i r e study area and spring ranges of El e v a t i o n a l Migrant (EM; n=29) and Mountain Resident (MR; n=8) bears. Also shown i s the proportion of spring r a d i o - l o c a t i o n s (use) of EM and MR bears i n the habitat components, elevations, and aspects. 19 MM SUB-ALPINE f2 SNOWCHUTE £3 ROCKY 55 RIPARIAN QROAO m MEAOOW UJIJ] CUTTING UNIT 5^ BURNS f~~l OPEN TIMBER I ! TIMBER e M 5M STUDY MR MR USE COVERAGE AREA COVERAGE USE CTl 22O0*50% of the time e i t h e r i n r i p a r i a n or snowchute habitats that cover only 7% and 2% of the study area. Between l a t e July and l a t e September, burns also received over h a l f the use but covered only 16% of the area. Proper management of these 3 habitat components i s c r i t i c a l f o r the long term conservation of the species i n the Flathead V a l l e y and other areas with s i m i l a r habitats. For r i p a r i a n and snowchutes, proper management may simply be l i m i t e d to protection from development and settlement, however, ac t i v e management should be investigated. For example, snow may r a r e l y s l i d e when conditions are optimum f o r large avalanches to generate bear habitat • The--use-, of explosives to s t a r t avalanches when conditions are.optimal could maintain avalanche chutes i n t h e i r most productive state f o r bear foods. Because burns are an early successional component, protection of t h i s habitat can only be a temporary management action. Prescribed f i r e s and/or a " l e t burn" p o l i c y i n appropriate high elevation s i t e s w i l l be required to maintain and improve the d i s t r i b u t i o n of t h i s habitat. Because timber was frequently used by bears i n the spring and autumn, and because c u t t i n g units were r a r e l y used i n any seasons by e i t h e r MR or EM bears, i t could be interpreted that timber harvest had a negative e f f e c t on g r i z z l y bear habitat. However, some bears chose to forage i n portions of some cuttin g units f o r b r i e f periods, and, although timber i s often used by bears, i t i s not c l e a r how much timber i s required by bears to maintain t h e i r current population density. When the s e l e c t i o n process i s studied at more d e t a i l e d l e v e l s , reasons, f o r the use of some cuts and not others may be better understood. 31 CHAPTER 3: THE BEHAVIOURAL RESPONSE OF GRIZZLY BEARS TO RESOURCE EXTRACTION INDUSTRIES There are many kinds of bear-industry i n t e r a c t i o n s , but the most immediate concern i s the extensive network of roads that i n d u s t r i e s depend on f o r access. Roads also increase access for hunters and poachers, the p r o b a b i l i t y of vehicle-bear c o l l i s i o n s , and the frequency of energy c o s t l y f l i g h t responses by the bears. Also, i n d i r e c t population constraints could r e s u l t i f long term displacement of bears from areas adjacent to roads occurs. Roads often follow v a l l e y bottoms and pass through r i p a r i a n areas which are frequently used by g r i z z l y bears. I f roads do displace bears, t h i s should lead to increased pressure on s i m i l a r habitats i n undisturbed regions and the " l o s s " of these e s s e n t i a l but l i m i t e d habitats. Some v a r i a t i o n i n bears' responses to roads has be predicted; adult females with young cubs may avoid areas near roads more than other bears (Zager 1980); darkness may acts as cover f o r bears (Servheen 1981) and t h e i r use of areas adjacent to roads should occur mostly at night. Although the i n t e r a c t i o n between g r i z z l y bears and the extensive road networks that resource extraction industries depend on has been of concern to land managers, the influence that the actual i n d u s t r i a l operations has had on these bears i s also important. This chapter f i r s t presents data on displacement of g r i z z l y bears by roads developed f o r resource-extraction i n d u s t r i e s . I tested the n u l l hypothesis that g r i z z l y bears use areas adjacent to roads as often as they use areas away from them. I also explored the e f f e c t s of age-sex cl a s s and reproductive status of bears, habitat type, amount of vehicular t r a f f i c , and time of day on t h i s pattern of use, and on demographic implications. F i n a l l y , I present data on a series of natural experiments i n which I was able to compare locations of i n d i v i d u a l bears before, during, and a f t e r periods of i n d u s t r i a l a c t i v i t y . Unlike Chapters 2 and 4, which use data c o l l e c t e d between 1979 and 1987, t h i s Chapter i s based only on information obtained before December 1985. METHODS EFFECTS OF ROADS Five distance-from-road categories (DRC) were delineated: 1) 0 to 100. m; 2) 101 to 250 m; 3) 251 to 500 m; 4) 501 to 1000 m; and 5) >1000 m. Category widths increased with distance from roads because radio relocations are " p r o b a b i l i t y areas" not points, so t h e i r p r e c i s i o n i s a function of the bear-observer distance. Most ground locations were made from roads, so narrower distance categories were acceptable f o r locations near roads. Roads were separated into 3 classes: primary roads (main roads leading into the v a l l e y ) , secondary roads ( f i r s t - o r d e r branches o f f primary roads), and t e r t i a r y roads ( a l l other roads ac c e s s i b l e by 2 wheel-drive v e h i c l e s ) . T r a f f i c frequency was sampled o p p o r t u n i s t i c a l l y by counting v e h i c l e s (primary roads: n = 241, 205.4 hours; secondary roads: n = 145, 97.9 hours). Vehicles were classed as small (passenger car or pickup truck) or large (> pick-up truck). Home ranges of a l l bears contained roads. Using the Marcum and Loftsgaarden (1980) method to estimate resource a v a i l a b i l i t y , 1928 random points were located over a map of the study area. For each point, the habitat component (ServheerL 1983, Zager 1980), elevation, and DRC of the nearest of each class of road were determined. A l l random points (x = 228 per seasonal range) occurring within the minimum convex polygon of each bear's seasonal range were used to determine the proportion i n each DRC. The observed frequency of radio locations within each DRC was.compared with the expected (available) using the G-test (Sokal and Rohlf 1981: 704-721). I f s i g n i f i c a n t l y d i f f e r e n t (P < 0.05), confidence i n t e r v a l s were calcul a t e d f o r each DRC using the Bonferroni approach with 90% simultaneous confidence l i m i t s (Marcum & Loftsgaarden 1980). Radio locations separated by at l e a s t 10 h were considered independent because a bear could t r a v e l between a l l DRC's i n t h i s time; most relocations (96%) were separated by at l e a s t 1 day. For many analyses, radio locations f o r more than 1 bear were grouped and compared to the sum of random points from each of these bear's seasonal home ranges. Because ranges overlapped, the same point could occur within the seasonal range of several bears and be used more than once. Therefore, t o t a l sample s i z e s of random points 34 were reduced to the number of d i f f e r e n t points used. Locating bears during the day often included walking through densely vegetated habitats lacking t r a i l s , while night locations could only be made from roads. Consequently> bears located at night were usually i n d i v i d u a l s that had been located e a r l i e r i n the day i n parts of the study area with high road d e n s i t i e s . I t was inappropriate to compare a l l day and night locations, so samples were paired within a 24 h period, and Wilcoxon's matched-pairs signed-ranks test: wa~s used. Day-night p a i r s were also grouped by DRC t o enable comparisons at d i f f e r e n t distances from roads. I f the paired locations f e l l i n d i f f e r e n t DRC's, the c l o s e r category was used f o r anal y s i s . Bear locations at or near den s i t e s were omitted because they were strongly affected by factors other than roads. For the same reason, locations of bears foraging f o r huckleberries i n high elevation burns during the summer were not used. Good huckleberry production i s l i m i t e d to these burns but, because of a lack of harvestable timber, these areas were 1.5 to 7 km away from roads, so both t h e i r use and a v a i l a b i l i t y data a l l f a l l i n the >1000 m DRC. Due to the short duration of the early spring season, both early spring and spring as described i n Chapter 2 were combined and c a l l e d spring for t h i s chapter. S i m i l a r l y , because most rel o c a t i o n s during the summer were i n high e l e v a t i o n burns and therefore not used i n t h i s analysis, locations during summer i n other habitats were combined with autumn l o c a t i o n s f o r t h i s chapter. Minimum sample s i z e s f o r comparisons among i n d i v i d u a l 35 bears were defined as 15 relocations per season per bear. Data from both EM and MR bears were grouped i n t h i s chapter. EFFECTS OF INDUSTRIAL ACTIVITIES Most data presented i n t h i s Chapter were obtained i n a 12 2 . . . . km portion of the study area c o n s i s t i n g of 8 minor side-drainages, each perpendicular to a major, steep mountain ridge. Each drainage had been burned, by wildfire^-in the_ 1920's (Zager et a l . 1983), and i s now i n various s e r a i stages. Generally, lodgepoie pine f o r e s t occurs at lower s i t e s (<1525 m), but with increasing elevations these trees, plus some subalpine f i r and spruce, become more sparsely d i s t r i b u t e d up to t r e e l i n e around 1975 m. Globe huckleberry (Vaccinium qlobulare) grows 2 abundantly i n the 12 km area, and t h e i r b e r r i e s dominate the g r i z z l y bears' d i e t from e a r l y August to l a t e September. The basic experimental design used to examine bear responses to i n d u s t r i a l treatments involved a comparison of radio locations of the same g r i z z l y bears recorded before, during, and a f t e r an i n d u s t r i a l a c t i v i t y took place. S p e c i f i c locations were determined e i t h e r from f i x e d wing a i r c r a f t , or by ground t r i a n g u l a t i o n using at l e a s t three compass bearings from a separate known map locations with e r r o r polygons less than 10 ha. Occasionally, i n steep t e r r a i n or remote areas, i t was impossible to use t r i a n g u l a t i o n . In these cases, s p e c i f i c locations (accurate to <10 ha) were determined by walking a s e m i - c i r c l e and often completely around the bears while taking numerous bearings. General locations (accurate to <50 ha) were often obtained i n unroaded areas by t r i a n g u l a t i o n from the 2 nearest road. In the case of the inte n s i v e l y studied 12 km berry area, the road was 2 to 3 km distant. Treatments There were 2 periods; 1 was the summer berry season when bears l i m i t t h e i r movements and generally r e s t r i c t themselves to 1 or 2 berry patches. This period i s important because most of the f a t reserves required for hibernation are l a i d down at t h i s time. The second period was spring when bears normally are much more mobile throughout t h e i r home ranges i n search of more widely dispersed foods or mates. Both periods provided contrasting conditions with p o t e n t i a l l y d i f f e r e n t implications fo r bear-industry i n t e r a c t i o n s . Displacement i n the summer berry season could have more s i g n i f i c a n t negative e f f e c t s on bears than during other seasons when more a l t e r n a t i v e foods were a v a i l a b l e and s u r v i v a l during hibernation was not at r i s k . Seismic A c t i v i t y i n the Vaccinium Area: The seismic a c t i v i t y involved hand-cut seismic l i n e s f o r gas exploration. These l i n e s were established across the 12 2 km berry area during the period when huckleberries were ripe i n 1981, 1984, and 1985 (Figure 6). For each of the l i n e s , 3 zones of i n d u s t r i a l a c t i v i t y were defined: Zone 1 was the seismic l i n e i t s e l f , or was the major h e l i c o p t e r f l i g h t path s e r v i c i n g the l i n e , together with a 500 m s t r i p on each side. Most helicopter a c t i v i t y , a l l 37 1 9 8 1 0 km Figure 6. The paths of the 4 seismic l i n e s (broken, s t r a i g h t l i n e s ) across the 12 tan* Vaccjnium spp. berry area and the dates they were cut are shown. Also i l l u s t r a t e d are the 8 minor drainages. 38 d r i l l i n g , s e t t i n g charges, s t r i n g i n g and c o l l e c t i n g sensing cable, and b l a s t i n g occurred along the l i n e . D r i l l i n g could require 5 days to cross the berry area, and each of the 5 portable d r i l l s , compressors, and equipment baskets were transported by e i t h e r B e l l 205 or 212 h e l i c o p t e r s . These machines made up to 20 return trips/day from a staging area to the l i n e , and spent up to 260 min/day a c t i v e l y moving equipment along a portion of the l i n e . D i f f i c u l t d r i l l i n g , poor weather, and mechanical problems occasionally reduced the i n t e n s i t y of a c t i v i t i e s . During the d r i l l i n g period, smaller h e l i c o p t e r s made up to 15 a d d i t i o n a l trips/day along the l i n e . The smaller h e l i c o p t e r s were used to s t r i n g sensing cables, but thesed were more ac t i v e than the large machines, t r a v e r s i n g the l i n e up to 30 times/day, and frequently landing or hovering to drop and pick up loads. Zone 2 was 501 to 2000 m from the seismic l i n e or major f l i g h t path, and without a mountain ridge between. I could hear and observe a l l h e l i c o p t e r t r a f f i c i n Zone 1 from Zone 2. Helicopters crossed Zone 2 up to 15 times/day. Zone 3 consisted of areas >2000 m from the seismic l i n e or major f l i g h t path, or areas within t h i s distance but with mountains between. Helicopter overpasses occurred l e s s frequently i n t h i s zone. I tested the n u l l hypothesis that the proportion of the bears' radio-locations i n the 3 zones were equal before, during and a f t e r the seismic program. Radio-locations f o r each c o l l a r e d bear using the area when huckleberries were r i p e during a year of seismic a c t i v i t y were used. A l l zones contained portions of the Vaccinium patch (Figure 6), but l o c a l berry production and consequent bear use can change s i g n i f i c a n t l y between years, so only within-year l o c a t i o n data were used. I believe that any confounding due to within-year differences i n berry a v a i l a b i l i t y r e s u l t i n g from v a r i a b l e ripening dates and bear consumption was minimal i n t h i s area l a r g e l y because the seismic l i n e s were independent of t e r r a i n , running/ s t r a i g h t across a l l aspects and elevations. The a c t i v i t i e s of the g r i z z l i e s were scan-sampled (Altmann 1974) at 15 s i n t e r v a l s . Sampling began when I f i r s t observed a bear and ended e i t h e r when i t moved out of sight or darkness f e l l . The "G" t e s t (Sokal and Rohlf 1981), with a l e v e l of s i g n i f i c a n c e of P < 0.05, was used to compare use between periods. Within the huckleberry area, radio locations separated by 10 h were c l a s s i f i e d as independent. G r i z z l y bears could e a s i l y traverse the burn i n much l e s s time than t h i s . Most locations were separated by at l e a s t 1 day. Bear-Industrv Interactions during Spring: Four bear-industry i n t e r a c t i o n s during the spring period studied i n s u f f i c i e n t d e t a i l , involved the following: 1. A crew of 6 people using 3 pick-up trucks was s t r i n g i n g sensing cable along a seismic l i n e . They were active i n t h i s area f o r 6 h. 40 2• A seismic sensing cable had been strung along a section of l i n e crossing another r i p a r i a n area the previous day and more cable was being transported along the l i n e by a small h e l i c o p t e r . This machine made 6 t r i p s along the l i n e at an estimated height of 150 m above ground l e v e l over 4 h. A recording truck carrying a small, operating generator, was stationed i n the r i p a r i a n area, and charges were detonated along t h i s portion of the l i n e . 3. On the f i r s t day of a c t i v i t y , a D7 C a t e r p i l l a r t r a c t o r was p i l i n g slash and b u i l d i n g skid roads. On the second day the D7, 4 skidders, 1 loader, 3 power saws, and 14 loggers were a c t i v e l y harvesting timber f o r 10 h. 4. A small C a t e r p i l l a r t r a c t o r and a grader operated almost continuously for 8 h to improve an old access road f o r timber extr a c t i o n . The working hypothesis used i n these 4 i n t e r a c t i o n s , was that bears within 250, 251 to 500, and 501 to 1000 m of an i n d u s t r i a l a c t i v i t y would move away immediately. For these t e s t s , i t was imperative that the bear be located p r i o r to the onset of the human a c t i v i t y and be monitored constantly at the onset of the a c t i v i t y . I t could then be sampled at decreasing frequencies. A pot e n t i a l d i f f i c u l t y of t h i s hypothesis i s that g r i z z l i e s normally are extremely mobile at t h i s time of year i n the absence of human s t i m u l i and so a bear which moves away from a human a c t i v i t y may have done so anyway. By continuously monitoring the bears and the human a c t i v i t y , I can have more confidence deciding i f they are displaced by a c t i v i t y . 41 RESULTS EFFECTS OF ROADS Of the 624 a e r i a l relocations, 32% were >1 km from roads compared to 15% of the 2196 made on the ground. This s i g n i f i c a n t d i f f e r e n c e (P < 0.01) was a r e s u l t of the radio receiver's l i m i t a t i o n s rather than a measure of bear use. Af t e r omitting data from the most dis t a n t DRC, there was no difference between the 422 a e r i a l and 1857 ground re l o c a t i o n s of the remaining four DRC's (P > 0.10). G r i z z l y bear use of the f i v e DRC's over both seasons combined, using a e r i a l r e locations, d i f f e r e d from that expected under the n u l l hypothesis (P < 0.01) . They used the 0 - 100 and 101 - 250 m DRC's s i g n i f i c a n t l y l e s s than expected, but none more than expected. Ground locations were biased only when >1 km from a road and, because displacement from roads occurred only i n the c l o s e s t DRC's, a e r i a l and ground relocations were combined f o r subsequent analyses, omitting data from the most dis t a n t DRC. Use of the 4 DRC's d i f f e r e d s i g n i f i c a n t l y from what was expected i n spring (P < 0.01), summer-fall (P < 0.01), and f o r both seasons combined (P < 0.01; Figure 7). In spring, habitats within 100 m of roads were used s i g n i f i c a n t l y l e s s than expected and the l e s s e r used area extended to 250 m during summer-autumn. CZ) OBSERVED n = 2278 1 2 3 4 DISTANCE-TO-ROAD-CATEGORY Figure 7. Observed versus expected use by 23 g r i z z l y bears of the 4 Distance-to-Road-Categories (m; DRC) i n the Flathoad Valley, B.C. Data are combined for a l l road types, both seasons, ground and a e r i a l radio-locations, and a l l age-sex classes, between 1979 and 19H5. Use d i f f e r i n g s i g n i f i c a n t l y (P < 0.05) from expected i s indicated by •»• or -.' For differences i n n values, and Distance-to-Road-Categorie3 see Methods i n t h i s Chapter. 43 Individual V a r i a t i o n Of the 23 g r i z z l i e s f o r which I had adequate spring information, 14 (61%) used the clos e s t DRC (0 - 100 m) s i g n i f i c a n t l y l e s s and 1 used i t s i g n i f i c a n t l y more than expected. The remaining bears' use of t h i s DRC was not s i g n i f i c a n t l y d i f f e r e n t from expected. No bears used the 101 -250 m d i f f e r e n t l y than expected, while 6 used e i t h e r the 251 -5.00 or 5-0.1 - 1000 m s i g n i f i c a n t l y more and 2 l e s s , i n spring. The 0 - 100 DRC was used l e s s than expected i n summer-autumn by 7 of 14 bears, while 3 also used the 101 - 250 m DRC also s i g n i f i c a n t l y l e s s . Greater use was recorded by 7 bears; each i n 1 of the DRC's between 101 and 1000 m. V a r i a t i o n Among Age-Sex Classes Independent yearlings used areas within 250 m of roads s i g n i f i c a n t l y more, and DRC's between 500 and 1000m s i g n i f i c a n t l y l e s s than a l l other classes of bears (Figure 8). Two of the 3 yearlings were o f f s p r i n g of ra d i o - c o l l a r e d females and they used the cl o s e s t DRC s i g n i f i c a n t l y more when alone than when they were cubs with t h e i r mothers. Contrary to what was predicted by Zager (1980), adult males used the clos e s t DRC s i g n i f i c a n t l y l e s s than d i d ei t h e r adult or subadult females (Figure 8). 0.6-(/) g 0.5 H5) S i g n i f i c a n t rp < 0 05) d i f f e r e n c e s among c l a s s e s i n t h e i r u s e o f e a c h DRC a r e i n d i c t e d by d i f f e r e n t l e t t e r s . The number o f r a d i o locations ; ^ ™ f e r o f b e a r s i n parenthesis . F o r D i s t a n c e - t o - R o a d - C a t e g o r l e * ; s e e Methods i n t h i s C h a p t e r . 45 Va r i a t i o n Within Aae-sex Classes There was l e s s v a r i a t i o n within than among some age-sex clas s e s ' use of areas near roads. A l l 3 independent yearlings used areas within 100 m of a road more than expected, the differ e n c e being s i g n i f i c a n t f o r 1 y e a r l i n g . Two of these yearlings used the 501 - 1000 m DRC s i g n i f i c a n t l y l e s s . A l l 6 adult and 3 subadult males used the c l o s e s t DRC s i g n i f i c a n t l y l e s s , and 1 i n d i v i d u a l i n each of these age-classes- also used the 101 - 250 m DRC les s than expected., Four of. 5 subadult and 4 of 6 adult females used areas i n the c l o s e s t DRC l e s s than expected. Female Reproductive Status and Responses In spring, females with cubs, with ye a r l i n g s , or alone, used the 4 DRC's d i f f e r e n t l y from each other (P < 0.05). When with cubs, they used the c l o s e s t DRC s i g n i f i c a n t l y more than when with older o f f s p r i n g or when alone (Figure 9). In summer-autumn, the 3 groups of females used a l l DRC's s i m i l a r l y (P > 0.10; Figure 9) . Influence of T r a f f i c Volume Primary roads were used on average by 1.3 large and 3.5 small v e h i c l e s per h; 0.4 large and 1.5 small v e h i c l e s per h used secondary roads. For most of the year, t e r t i a r y roads were used almost e x c l u s i v e l y by ourselves and a group studying wolves. Forestry and gas exploration personnel used them f o r short periods, and hunters used them frequently i n the autumn. DISTANCE—TO—ROAD-CATEGORY Figure 9. R e l a t i v e use of the 4 Distance to Road Categories (DRC; m) by female g r i z z l y bears of d i f f e r e n t reproductive status during s p r i n g (top; n=5) and summer-fall (bottom n=2). Where use of a DRC d i f f e r e d s i g n i f i c a n t l y (P < 0.05) among females of d i f f e r e n t s t a t u s , t h i s i s i n d i c a t e d by d i f f e r e n t l e t t e r s . The number of r e l o c a t i o n s = n. 47 I predicted bears would use areas adjacent to primary roads l e s s than to roads with l e s s t r a f f i c , but they d i d not. For both seasons combined, bears s i g n i f i c a n t l y reduced t h e i r use only of the 0 - 100 m DRC i n a l l road classes, and, i n general, used the 2 most d i s t a n t DRC's more than expected (Figure 10). Nocturnal Behaviour O v e r a l l , bears used areas near roads s i g n i f i c a n t l y more at night than during daylight (Wilcoxon's Z = 1.71; n = 121). When data were subdivided and re-analyzed f o r each DRC separately, the p r o b a b i l i t i e s of the 2 d i s t r i b u t i o n s being the same were 0.054, 0.066, 0.052, and 0.405 fo r DRC's 1 to 4 r e s p e c t i v e l y . With the smaller sample s i z e s , none were s i g n i f i c a n t , although night use of the 3 c l o s e s t DRC's was greatest. For the paired locations, bears were located on roads 12 times; 10 during the night. I f roads were used equally during day and night, the p r o b a b i l i t y of such an occurrence i s only 0.012. I also noted many more fresh bear tracks i n mud and snow on roads early i n the morning than l a t e r i n the day. Influence of Habitat Type Habitat components were not randomly d i s t r i b u t e d r e l a t i v e to the DRC's (P < 0.01). For example, most roads were b u i l t to remove timber, and c u t t i n g units comprised 33% of the habitats within 100 m of roads, but only 4% i n the >1000 m DRC. Conversely, r i p a r i a n areas made up only 5% of habitats i n the 0 48 0".»-i O OJ Ol £ 0-100 101-250 251-500 501-1000 DISTANCE-TO—ROAD-CATEGORY Figure 10. Observed versus expected use by 23 g r i z z l y bears of each Distance to Road Category (m) with d i f f e r e n t l e v e l s of t r a f f i c (Primary - top; Primary and Secondary - middle; T e r t i a r y - bottom) i n the Flathead Valley, B.C. Data represent a l l spring and summer-fall seasons between 1979 and 1985. Use d i f f e r i n g s i g n i f i c a n t l y (P < 0.05) from expected i s indicated by + or -. For differences i n n values see Methods i n t h i s Chapter. - 100 m DRC, and 17% by the 251 - 500m DRC. To overcome the lack of independence between habitats and roads, and to d i s t i n g u i s h which feature the bears were responding to, I compared the use of each habitat component with the use expected within each DRC. . I f roads caused bears to reduce t h e i r use of a given habitat component, use of such a habitat should have decreased the c l o s e r i t was to a road. But, i f bears responded primarily to habitats rather than to roads, I should f i n d no difference i n t h e i r use of a given habitat among the DRC's. When the habitat components were examined separately, there were adequate data only to analyze use of timber and r i p a r i a n types, which together comprised 72% of a l l data used. These habitats provided the densest cover, and I expected g r i z z l i e s to be displaced l e a s t while using them. However, they used timber and r i p a r i a n within 100 m of roads s i g n i f i c a n t l y l e s s than expected i n spring, and used timber i n t h i s c l o s e s t DRC s i g n i f i c a n t l y less i n summer-autumn. These r e s u l t s support the hypothesis that bears were displaced by roads and not simply avoiding the types of habitat found near roads. EFFECTS OF INDUSTRIAL ACTIVITIES Seismic A c t i v i t y i n the Vaccinium Area When the 226 s p e c i f i c radio locations of 2 adult females, 1 subadult female and 1 subadult male were combined over the 3 years, I found no difference (P > 0.20) i n these bears' use of the 3 zones before, during or a f t e r the seismic a c t i v i t y (Figure 11). When the 373 general radio l o c a t i o n s were also included, again I found no diffe r e n c e (P > 0.15) i n the bears' use of the 3 zones. Individual comparisons using both s p e c i f i c and general relocations f o r each bear each year that seismic work occurred, showed only 2 s i g n i f i c a n t differences i n zone use i n 11 cases; an adult female with 2 cubs i n 1984, and a subadult female i n 1985 (Table 2) . The adult female was not located i n Zone 1 during the seismic a c t i v i t y but was i n Zone 2 i n 11 of 15 loca t i o n s . On one occasion she and her cubs were observed f o r 1.5 h on an open h i l l s i d e overlooking the seismic a c t i v i t i e s which were 1400 m from them; During t h i s period, 10 h e l i c o p t e r f l i g h t s were made along the l i n e and 4 charges detonated. When hel i c o p t e r s flew approximately 14 00 m from her, the female ra i s e d her head f o r 3 to 15 s before resuming her previous behaviour. On 2 of the 10 occasions she d i d not even r a i s e her head from feeding. When the underground charges were detonated they were heard as f a i n t rumblings by humans about 2000 m away, but the bears gave no v i s i b l e response. Scan-sampling her behaviour revealed she fed 81% of the time (n = 168), compared to 85% (n = 416) i n 2 samples i n the 1984 berry season when seismic work was not i n progress (P > 0.20). This same bear was not displaced from Zone 1 by seismic a c t i v i t y when she was on her own i n 1985 (Table 2). During the seismic work, the subadult female changed (P < 0.01) her locati o n s , using Zone 2 less and Zone 3 more, than I I BEIORE n=78 ZONE 1 ZONE 2 ZONE 3 Figure 11. Proportions of combined radio locations of 2 adult females (23, 35), and 2 subadults (1 male 47, 1 female 36) g r i z z l y bears over a 3-year period, i n each of the 3 Distance Zones, before, during, and af t e r seismic exploration. Table 2. Proportions of a l l relocations of 4 radio c o l l a r e d g r i z z l y bears i n each of 3 distance zones from a c t i v e seismic 2 . . operations i n a 12 km Vaccmium habitat. Proportions are shown f o r before, during and a f t e r the seismic operations. The l e v e l s of p r o b a b i l i t y of equal d i s t r i b u t i o n s among the zones (P) and sample sizes (n) are shown. gegore Purina After 5°as Zone Zone £ Bear Year a 1 2 3 n 1 2 3 a 1 2 3 23 1981 19 .42 .05 .53 10 .40 .20 .40 6 .17 .33 .50 .41 23 1984 0 - - - 11 .09 .55 .36 34 .21 .24 .56 .16 23 1985* J7 .14 .24 .62 17 .06 .35 .59 18 .11 .17 .72 .70 23 1985 B 11 .09 .37 .54 11 .09 .27 .64 18 .17 .44 .39 .72 35 1984 7 .28 .57 .14 15 .00 .73 .27 34 .15 .29 .56 .01 35 1985 29 .34 .52 .14 19 .32 .63 .05 8 .38 .62 .00 .59 36 1984 0 - ' - - 10 .20 .60 .20 28 .43 .50 .07 .31 36 1985 25 .56 .36 .08 18 .33 .06 .61 11 .36 .45 .18 .00 47 1984 5 .40 .20 .40 10 .10 .70 .20 43 .12 .35 .54 .13 47 1985 A 31 .16 .19 •65 16 .06 .38 .56 9 .00 .33 .67 .32 47 1985 B 38 .08 .26 .66 9 .00 .22 .78 9 .00 .22 .78 .62 Means 234 .23 .31 .47 146 .16 .42 .42 218 .19 .34 .46 .18 Dates with a. and represent separate incidents. e i t h e r before or a f t e r the a c t i v i t y took place (Table 2). However, she used Zone 1 at a s i m i l a r frequency during a l l 3 periods. The relocations indicated that she used Zone 1 almost e x c l u s i v e l y during the seismic a c t i v i t y , even when helicopters passed only about 100 m above her. The bear was bedded when the intensive d r i l l i n g and a B e l l 205 transporting equipment gradually approached her. A f t e r 7 min of almost constant .helicopter work, within 100- m of her, she became active and 150 min l a t e r had t r a v e l l e d approximately 2400 m and out of the drainage. The following morning she was located about 10 km away i n a habitat r a r e l y used by g r i z z l y bears during t h i s season. She remained there 6 days before returning to the berry producing area and the seismic l i n e . This was the only case where the seismic a c t i v i t y appeared to cause a temporary range abandonment. One other adult female g r i z z l y bear was known to have l e f t t h i s berry area f o r b r i e f periods when seismic work was i n operation, but the i n t e n s i t y and location of the seismic work was such that i t i s doubtful she was aware of i t immediately p r i o r to her leaving. S i m i l a r l i m i t e d movements i n time and distance from the berry area occurred when seismic work was not i n progress. Both adult females moved to another berry area s h o r t l y a f t e r the 1984 seismic l i n e was completed. Although the a c t i v i t y may have influenced t h e i r movements, the changes were not immediately i n i t i a t e d by human a c t i v i t y . The poor berry crop i n 1984 also may have had influenced t h e i r movements. Bear-industry Interactions during Spring 1. An adult female and a subadult male were located together 500 m south of the l i n e when the seismic crew began s t r i n g i n g sensing cable across the r i p a r i a n area occupied by the bears. When the workers crossed the stream 4 h l a t e r , the 2 bears had moved to within 100 m of the l i n e . When the crew f i n i s h e d work 2 h l a t e r , the bears were about 350 m away from the l i n e and 1 h l a t e r 400 m away. The bears crossed the l i n e during the night and by daybreak had moved 5 km away. The bears c l e a r l y were not displaced from the 250-500 m areas around the a c t i v i t y , and although the bears moved to <100 m from the a c t i v i t y , i t i s unclear that they responded to i t . I f they did move away from the l i n e because of the seismic a c t i v i t y , they were not displaced by >400 m. 2. A subadult male was found p r i o r to the s t a r t of the day's seismic a c t i v i t i e s i n a r i p a r i a n area 80 m from a seismic road and a recording truck. The bear remained f o r 8 h <100 m from the l i n e u n t i l at le a s t 2 h a f t e r work ceased, and 6 h l a t e r he had moved. The r i p a r i a n area was investigated the following day and signs of feeding and a bed were found 90 m from the l i n e . Again the bear was not displaced from any of the distance categories by the i n d u s t r i a l a c t i v i t y . 3. An adult male was f i r s t located on an snowchute approximately 1000 m from an in a c t i v e logging s i t e . The following day he was found 400 m from t h i s i n a c t i v e s i t e , but about 600 m from an operating D7 C a t e r p i l l a r t r a c t o r . The t h i r d day, the logging crew began working and he remained i n a snowchute approximately 400 m away f o r at l e a s t 2 h. The following 4 days, he was located 320, 520, 500, and 2000 m from the logging operation. This bear was not displaced from the 250-500 and 501-1000 m distance classes. 4. An adult female and 2 cubs had occupied a small basin f o r 10 days when a small C a t e r p i l l a r t r a c t o r and a grader began work on a road into the basin. Immediately p r i o r to t h i s a c t i v i t y , the bears were located on an open ridge about 800 m from the road and remained there throughout the day despite the a c t i v i t y . They l e f t the basin several days l a t e r a f t e r road maintenance had f i n i s h e d . These bears were not displaced from the 501^1000 m distance category. Long-term E f f e c t s on the Bears None of the bears studied, and which were p o t e n t i a l l y disturbed by the various i n d u s t r i a l a c t i v i t i e s , died during the study. However, i t i s possible that the seismic a c t i v i t y had an e f f e c t on 1 bear. One of the adult females who d i d not 2 appear to respond immediately to the seismic work i n the 12 km berry area, gave b i r t h to at l e a s t 2 cubs. I saw her with one i n the ea r l y spring 1 km from her den and found the remains of another small cub i n a scat c o l l e c t e d at her den s i t e . Three weeks l a t e r she was alone. She had ra i s e d cubs s u c c e s s f u l l y before, and though possible, i t i s u n l i k e l y that her eating her cub and the loss of the other were rel a t e d to the 1981 seismic a c t i v i t y several months p r i o r . She had 2 cubs a f t e r the 1984 seismic program and both survived through the 1985 seismic 56 a c t i v i t y period. 2 Another adult female which used the 12 km area had 3 cubs with her during the 1981 seismic season, another 2 i n 1984 and then 3 i n 1986; she su c c e s s f u l l y weaned a l l 8 o f f s p r i n g . A subadult female which also used t h i s area had her f i r s t l i t t e r of 2 cubs i n 1986 at 5 years of age, the year a f t e r she was displaced from the area by the seismic work; she succe s s f u l l y weaned both of these. DISCUSSION Loss of Habitats Near Roads Most g r i z z l y bears used areas near open roads s i g n i f i c a n t l y l e s s than expected. This was equivalent to a habitat loss of 58% i n the 0 - 100 m DRC and 7% i n the 1 0 1 - 2 5 0 m DRC. For the whole Flathead study area, i t represents a lo s s of 8.7% of the area a v a i l a b l e to the bears. Further, the types of habitat often associated with roads (25% of r i p a r i a n h a b i t a t i s within 250 m of a road) are e s p e c i a l l y valuable to bears, because they contain high q u a l i t y foods i n spring and autumn. The bears' reduced use of areas within 100 m of primary, secondary or t e r t i a r y roads d i d not d i f f e r , suggesting that even a l i t t l e t r a f f i c i s s u f f i c i e n t to displace them. No bear's home range lacked roads or other human a c t i v i t i e s , so a l l should have had some opportunity to habituate or adapt to predictable road-related s t i m u l i . Certain aspects of t h e i r behaviour may have reduced the degree of habi t a t loss bears experienced. The f i r s t i s t h e i r use of roads and adjacent areas at night, which supports Servheen's (1981) p r e d i c t i o n that darkness o f f e r s cover to bears. Obviously, bears cannot use a l l areas a l l the time, but by a l t e r i n g t h e i r use of areas near roads from daylight to darkness, they may continue to use a large portion of valuable habitats located near roads. Although darkness probably provided sec u r i t y cover, t r a f f i c volume also would be reduced at night. V a r i a t i o n among age-sex classes i n the use of areas near roads also may have had an ameliorating e f f e c t . Adult males used, habitats near roads l e s s than other classes, while independent yearlings and some adult females with cubs used these areas more than any other cl a s s , contrary to Zager's (1980) p r e d i c t i o n . This d i f f e r e n t i a l response was also found for g r i z z l i e s i n Yellowstone National Park (Mattson et a l . 1987) and f o r black bears (U. americanus) i n Alberta ( T i e t j e and Ruff 1983). Adult males sometimes k i l l cubs and yearlings (Glenn et a l . 1976; Mundy & Flook 1973; Pearson 1975; Reynolds & Hechtel 1979; Troyer & Hensel 1962), so habitats near roads may have been r e l a t i v e l y safe f o r these vulnerable classes of g r i z z l i e s . Females with cubs generally avoid adult g r i z z l y males (Pearson 1975; Russell et a l . 1979), and an experimental manipulation suggested adult male black bears regulated population density i n northern Alberta (Kemp 1972, 1976). A d i f f e r e n t response to roads was found only 150 km south of my study area. There, 2 female g r i z z l i e s used habitats within 199 m of open roads l e s s and 2 males used such areas more frequently than expected (Zager 1980). The reason for the di f f e r e n c e between these studies could be due to the small sample of only 4 bears, but also to differences i n habitat a v a i l a b i l i t y . Highly productive, low elevation habitats i n Zager 7s (1980) study area were eliminated by water impoundment behind the Hungry Horse Dam, r e s t r i c t i n g bears to steeper mountain habitats and narrow side drainages where there were roads. Perhaps, as i n Jasper National Park, Alberta (Russell et a l . 1979), males frequently used the productive areas at low elevations i n the narrow side v a l l e y s where roads are usually located, while females used higher elevations. Most data on the displacement by roads i n my study area, were c o l l e c t e d on EM bears during spring and autumn, when a l l age-sex classes used the wide v a l l e y bottom and d i f f e r e n t i a l habitat use patterns was not affected by e l e v a t i o n to the degree i t was i n Jasper (Russell et a l . 1979). In the Flathead, as i n T i e t j e ' s and Ruff's (1983) f l a t study area, human-use areas probably had a more d i r e c t e f f e c t on age-sex cla s s segregation of g r i z z l i e s than did elevation. When adequate information i s obtained on MR bears, perhaps r e s u l t s s i m i l a r to those of Zager (1980) w i l l be found. Demographic Consequences of Roads To be of major concern to w i l d l i f e managers, behavioural responses to disturbance must have demographic consequences (Shank 1979). The population was at a r e l a t i v e l y high density during my study, and the survival-fecundity rate of increase (Caughley 1977:55), was p o s i t i v e (see Chapter 4) i n d i c a t i n g roads were not having a severe impact during my study. Three c h a r a c t e r i s t i c s of the study area may have reduced the p o t e n t i a l impact of roads on the g r i z z l y bear population. F i r s t , bear s u r v i v a l and reproduction depends gre a t l y on f a t reserves obtained p r i m a r i l y i n l a t e summer by foraging on huckleberries. In my study area, these grow best i n high elevation, p o s t - f i r e shrubfields that do not contain roads and are therefore t o t a l l y a v a i l a b l e to the bear population. Second, i n the B.C. portion of the study area, resource-industry employees make up most of the resident human population. Potential road-related impacts on the bears were probably lessened because of some i n d u s t r i e s 7 p o l i c i e s (e.g. S h e l l Canada: no firearms, r e s t r i c t e d use of p r i v a t e v e h i c l e s , d a i l y garbage i n c i n e r a t i o n ) . Third, hunting regulations have generally become more r e s t r i c t i v e i n the study area, and the annual l e g a l harvest of g r i z z l i e s i s c l o s e l y monitored. Roads did increase the bears 7 v u l n e r a b i l i t y to l e g a l hunters and to poachers by providing ready access. A l l but l known and suspected adult and subadult g r i z z l y deaths (n = 28) since 1979 have been caused by people, e i t h e r l e g a l l y or i l l e g a l l y ; most shot from roads (see Chapter 4). Although the Flathead g r i z z l y population seems to have withstood the e f f e c t s of human access during my short study, the roads 7 p o t e n t i a l as a s i g n i f i c a n t negative demographic factor remains high. Once roads are developed i n any g r i z z l y habitat, the population i s placed i n a more precarious p o s i t i o n and management of bears must be changed accordingly. 60 Displacement of Bears by Resource Industry A c t i v i t i e s From the analysis of the pooled samples from the 4 bears over the 3 years of intensive study I cannot r e j e c t the n u l l hypothesis that the bears were not displaced from an important habitat associated with seismic operation a c t i v i t i e s . When the behaviour of i n d i v i d u a l bears were examined year by year, i n only 2 of 11 cases were there s i g n i f i c a n t d i f f e r e n c e s . Results of bears' reactions to various types of i n d u s t r i a l a c t i v i t y during spring, when the animals normally move throughout t h e i r ranges, also suggested minimal displacement from i n d u s t r i a l a c t i v i t i e s . These r e s u l t s are s i m i l a r to those described by Simpson et a l . (1985), who studied the e f f e c t s of geological exploration on the movements of g r i z z l y bears i n the S e l k i r k Mountains north of my study area. Unlike bear responses i n t h i s and Simpson et a l . ' s (1985) study, Mace and Jonkel (1980) i n f e r r e d s i g n i f i c a n t displacement of bears because they never found 3 c o l l a r e d g r i z z l y bears i n a small drainage where active logging occurred, but relocated them frequently i n adjacent drainages. The di f f e r e n c e s i n bear responses i n these studies may be p a r t i a l l y a function of bear density r e l a t i v e to carrying capacity. Most bear populations appear to be l i m i t e d by human predation (Cowan 1972, Bunnell and T a i t 1985), but when not, other factors must eventually be l i m i t i n g . Though habitat q u a l i t y appears to be the ultimate fac t o r l i m i t i n g bear density, there i s evidence that i t i s proximately regulated by s o c i a l l y induced-dispersal (Kemp 1972, 1976, Knight and Eberhardt 1985). A bear subjected to a disturbance can choose to remain or move away from i t , both choices have t h e i r costs. By staying, i t s metabolic costs can r i s e due to stress and i t may even be k i l l e d i f i t comes across a human, whereas i f i t moves, i t expends energy i n t r a v e l l i n g and searching f o r an a l t e r n a t i v e , perhaps l e s s productive area. In a population at carrying capacity, a bear's cost of moving (displaced) may be high due to s o c i a l intolerance of other bears, or due to competition for some l i m i t i n g resource. In my study area, there i s a high density of g r i z z l y bears (see Chapter 4), so here, the best strategy may have been be f o r a bear to remain unless the population was below carrying capacity, i n which case s o c i a l l y a v a i l a b l e habitats would be r e l a t i v e l y abundant and moving from the disturbance would incur l e s s cost. Overt versus Covert Responses Although the 12 instances I observed of bears' reactions to the use of helicopters f o r seismic exploration i s a small sample, the r e s u l t s are consistent with those reported i n other studies (McCourt et a l . 1974, Quimbey 1974, McLellan and Shackleton i n press b). In general, i t appears that a movement response from a single h e l i c o p t e r o v e r f l i g h t i s most l i k e l y when the bear i s i n the open. Whether the a i r c r a f t i s merely f l y i n g at a tangent to the bear or d i r e c t l y towards i t , w i l l also be important i n open habitats. In timbered habitats, I found that an overt avoidance or displacement response required high i n t e n s i t y h e l i c o p t e r 62 a c t i v i t y such as carrying equipment within 200 m of the bear. This does not mean that bears are not stressed by these a c t i v i t i e s . MacArthur et a l . (1982) found that although mountain sheep (Ovis canadensis) showed no overt behavioural response to disturbance, t h e i r heart rate increased s i g n i f i c a n t l y , and Stemp (1983) found that the duration of elevated heart rate can l a s t several hours. As increased heart rate i s a t y p i c a l f i r s t l e v e l response of mammals to stress, there i s no reason to suspect g r i z z l y bears to be d i f f e r e n t . Heart rate telemetry studies should be the next step i n i n v e s t i g a t i n g the f u l l p o t e n t i a l e f f e c t s of i n d u s t r i a l a c t i v i t y on bears i n cover. However, I found that most of g r i z z l y bears i n cover remained i n the general l o c a t i o n f o r several hours or even days a f t e r a bout of intensive h e l i c o p t e r a c t i v i t y . The few which were displaced, moved o f f slowly, i n d i c a t i n g a low l e v e l response. The generally moderate response of c o l l a r e d g r i z z l y bears to i n d u s t r i a l a c t i v i t y i n the study area may be a function of habituation. Not only have these bears been exposed to some machinery and s i m i l a r human a c t i v i t i e s throughout t h e i r l i v e s , but they have been often radio-located from a i r c r a f t as well. Mech (1966) found that wolves (Canis lupus) habituated to a i r c r a f t a f t e r 1 day of monitoring, and s i m i l a r behaviour may be expected with bears. MANAGEMENT IMPLICATIONS G r i z z l y bear food species vary g r e a t l y among seasons and i t appears that q u a l i t y and abundance of foods do as w e l l (McLellan unpublished). Consequently, some seasonal habi ta ts may be very r e s t r i c t e d and others abundant, so even at c a r r y i n g capac i ty , displacement may have r e l a t i v e l y l i t t l e cost for bears during seasons of food abundance ( e . g . , s p r i n g ) . I f economically and l o g i s t i c a l l y p o s s i b l e , i n d u s t r i a l a c t i v i t i e s should be conducted during t h i s season. The moderate to low response of c o l l a r e d bears to i n d u s t r i a l a c t i v i t y does not n e c e s s a r i l y suggest that t h i s a c t i v i t y has no detr imental e f f e c t s . Q u a n t i t a t i v e l y or q u a l i t a t i v e l y d i f f e r e n t i n d u s t r i a l a c t i v i t i e s may show greater or longer l a s t i n g displacement of bears. Even at the l e v e l of i n d u s t r i a l a c t i v i t y i n my study, p o t e n t i a l problems remain. There i s always a r i s k of confrontat ions between seismic company employees and bears because construct ing seismic l i n e s often d i d not d i s p l a c e the bears. In a d d i t i o n , the seismic l i n e increases the ease of human access to remote areas, i n c l u d i n g those where bears are p a r t i c u l a r l y v i s i b l e and vu lnerable . I t i s therefore important to make o l d seismic l i n e s impassible to a l l - t e r r a i n v e h i c l e s , p a r t i c u l a r l y those l i n e s passing through berry patches to which bears show high annual f i d e l i t y . 64 CHAPTER 4. DYNAMICS OF A GRIZZLY BEAR POPULATION DURING A PERIOD OF INDUSTRIAL RESOURCE EXTRACTION The behavioural responses of g r i z z l y bears to i n d u s t r i a l a c t i v i t i e s and the r e s u l t i n g habitat a l t e r a t i o n s , i n d icate where s i g n i f i c a n t overt c o n f l i c t s l i k e l y occurred, but the response v a r i a b l e s of ultimate importance are those measuring the dynamics of the population (Shank 1979). By combining information about the .bears 7 behavioural, habitat use, and population responses to i n d u s t r i a l a c t i v i t y , I expected to gain a more complete understanding of the r e l a t i o n s h i p between industry and g r i z z l i e s than i f I l i m i t e d my research to only one or two of these aspects. I hypothesized that the i n d u s t r i a l a c t i v i t y i n the study area would be harmful to g r i z z l y bears, and predicted that i t would r e s u l t i n a low density of bears compared to other i n t e r i o r areas without i n d u s t r i a l a c t i v i t y , and/or that t h e i r s u r v ival-fecundity (Caughley 1977:54) exponential rate of population increase would be negative. To t e s t t h i s hypothesis, I f i r s t present data on g r i z z l y bear density and age-sex structure of the population during a period of resource extraction. Then mortality rates, causes of death, and estimates of reproductive v a r i a b l e s are reported. F i n a l l y , I estimate the rate of increase of the population, and investigate the s e n s i t i v i t y of t h i s estimated rate of increase to changes i n parameter values. METHODS Density Estimation Estimating g r i z z l y bear density i s d i f f i c u l t i n a heavily timbered area without the presence of seasonal concentration s i t e s such as garbage dumps or salmon spawning streams (Knight and Eberhardt 1985), p a r t i c u l a r l y where the system i s open to immigration and emigration. Standard capture-recapture methods were thought to be inappropriate f o r t h i s study because: 1) the assumption of equal c a t c h a b i l i t y could not be met because some in d i v i d u a l s were c l e a r l y more d i f f i c u l t to capture than others, and almost a l l were much more d i f f i c u l t to recapture, and 2) the study area was not a closed system and so I was sampling bears from a much larger area than the trapping areas alone, which would lead to an i n f l a t e d density estimate. Adding one home range radius around the study area (Caughley 1977:140) may decrease t h i s bias, but v a r i a t i o n i n home range s i z e and shape reduces confidence with small samples. A commonly used density estimation technique f o r g r i z z l y bears i s to record a l l bears seen or trapped i n a study area of know s i z e , and then derive the density by simple d i v i s i o n (Mundy and Flook 1973, Martinka 1974, Pearson 1975, Russell et a l . 1979, Servheen 1983). Using saturation trapping, observations, or both, p a r t i a l l y a l l e v i a t e s the problem of unequal c a t c h a b i l i t y because bears only have to be captured once, or not at a l l i f only observation i s required. Capturing or observing a l l of the g r i z z l i e s i n a study area i s unl i k e l y , so t h i s estimate i s usually regarded as a minimum or a conservative estimate (Mundy and Flook 1973, Russell et a l . 1979). However, i f a l l bears trapped or seen i n a study area are included i n the density estimate, we are faced with problem (2) mentioned above, and the "minimum" estimate may be a c t u a l l y higher than the true- density. I devised two methods to estimate density. The f i r s t i s based on home range information determined by radio tracking. F i r s t , I decided on a core study area (trapping Area A; Figure 12) where I would estimate the density of g r i z z l y bears and concentrated almost a l l trapping there. However, with a maximum f i e l d crew of 3 available, even t h i s area was too large to trap e f f i c i e n t l y , so I delineated a more intensive trapping area (trapping area B; Figure 12) within area A where most timber harvest and gas exploration occurred (Table 3). A f t e r the trapping was completed, trapping areas A and B were plo t t e d as minimum convex polygons by enclosing a i l successful trap locations f o r the entire study period. Trapping area A covered 264 km2 and B 130 km2. Each bear's 97.5 percent multi-year, minimum home range convex polygon (Harestad 1981) was then p l o t t e d using ground and a e r i a l relocations. Each bear's contribution to the density estimate was the proportion of t h i s home range that f e l l within the trapping area boundaries (Figure 13). Thus, only bears with a 97.5% multi-annual home range located e n t i r e l y within a trapping area were counted as a "whole" bear. In Chapter 2, I compared random points from within e n t i r e Figure 12. The study area. Almost a l l trapping occurred i n the la r g e r Area A, while the most intensive trapping e f f o r t was confined to a portion of Area A c a l l e d Area B. 68 Figure 13. Radio l o c a t i o n points and 97.5% multi-annual home range of a subadult female g r i z z l y bear. The proportion of t h i s range which overlapped a trapping area i s the bear's co n t r i b u t i o n to the population density estimate. Table 3. Extent of Resource Extraction Industries and residences i n the Flathead study area as of 1986. Human A c t i v i t y Area A Area B Area (km ) 264 130 2 Open roads (km/100 km ) 69 93 2 2 Cutting u n i t s (km /100 km ) 16 18 2 Seismic l i n e s (km/100 km ) 63 86 D r i l l r i g s (per 100 km2) 0.04 0 2 I n d u s t r i a l camps (per 100 km ) 1.1 0.8 2 Permanent residences (per 100 km ) 2.3 4.6 seasonal ranges to points o v e r l a i d on the e n t i r e study area because I was interested i n s e l e c t i o n of e n t i r e seasonal ranges. In t h i s chapter, however, extreme o u t l i e r locations (2.5%) from multi-year home ranges were omitted because they could lead to a small underestimate of density due to the nature of bears' ranges. T y p i c a l l y , ranges included several c l u s t e r s of locations representing high-use areas, with a few scattered locations i n between and some extreme o u t l i e r s . A f t e r examining each range, an average of 2.5% of the locations were estimated to be o u t l i e r s . Excluding these outlying locations of bears whose high-use areas were within the trapping areas, reduced the bias of not catching, and therefore of not counting bears whose high-use areas were elsewhere and only entered the trapping area occasionally. This f i r s t density estimate i s based on the assumption that I captured a l l , but only, those bears which had at le a s t part of t h e i r 97.5% home range within the trapping area. I f 100% home ranges had been used, I would have had to assume that I had trapped every bear that even j u s t entered the trapping area during the study period. The f i r s t assumption may be f a l s e , and therefore the density estimate i s conservative, but the second assumption i s known to be wrong because several unmarked g r i z z l i e s were seen i n the study area. The second method I used resembles the f i r s t , but instead of using the proportion of each bear's range that f e l l i n the trapping areas as t h e i r contribution to the density estimate, the proportion of time each bear spent i n the trapping area was used as t h e i r contribution to the estimate. This proportion was assumed to equal the proportion of each bear's radio locations, obtained from a i r c r a f t , that was i n the trapping area. A e r i a l relocation eliminates any bias due to d i f f e r e n t i a l road access used f o r ground tracking. Both methods r e l y e x c l u s i v e l y on functioning radio c o l l a r s , but b a t t e r i e s f a i l and c o l l a r s are pul l e d o f f , so not a l l bears c a r r i e d functioning c o l l a r s f or the en t i r e study period. Consequently, I was forced to extrapolate information from when ind i v i d u a l s were c o l l a r e d to when they were not. When an adult bear's c o l l a r ceased functioning, but the i n d i v i d u a l was l a t e r recaptured or harvested i n the study area, I assumed i t had remained i n the same multi-annual home range i t had occupied when radio tracked (average uncollared i n t e r v a l = 1.7 years). In t h i s case, i t s contribution to the population estimate was the same each year whether i t was c o l l a r e d or not. Unlike adults, which have r e l a t i v e l y stable range locations between years, subadults were not included i n the density estimate a f t e r t h e i r radios ceased unless they were recaptured and t h e i r new range determined. Bears which e i t h e r shed t h e i r c o l l a r s or were shot before they were relocated 20 times, were not included i n the density estimate. The census date was i n early May, which was when a l l bears had emerged from t h e i r dens, but was p r i o r to when they were hunted. 72 Age and Sex Composition Some c h a r a c t e r i s t i c s of a g r i z z l y bear population's age and sex composition are l i k e l y biased, i f based on hunter harvest or trapped animals (Troyer and Hensel 1964, Bunnell and T a i t 1980). However, when der i v i n g one estimate of bear density, I assumed that every bear whose 97.5% home range overlapped the trapping area was captured. I f t h i s assumption i s acceptable, then the population's age-sex structure can be estimated based on each i n d i v i d u a l bear's contribution, as was dene to ^ P t i m a f the density. Survival Rates and Causes of Death Bear mortality was divided into: 1) natural and not d i r e c t l y due to people, 2) l e g a l harvest, 3) i l l e g a l harvest, 4) l e g a l control losses (the agency responsible e i t h e r k i l l e d or removed a bear causing a problem to people), 5) i l l e g a l nuisance (a c i t i z e n i l l e g a l y k i l l e d a bear causing a problem), and 6) research losses. A l l recorded m o r t a l i t i e s were c l a s s i f i e d e i t h e r as known when the carcass was examined, or suspect when the carcass was not recovered but there was s u f f i c i e n t circumstantial evidence to indicate a death had occurred (e.g. finding a c o l l a r cut o f f , rumors of a bear being k i l l e d a f t e r the radio s i g n a l of one that was frequently located near human settlement stopped being received, or when a cub was no longer seen with i t s mother). Shooting r a d i o - c o l l a r e d g r i z z l i e s i s l e g a l i n B r i t i s h Columbia, Alberta, and Montana, though B r i t i s h Columbia hunting regulations have requested hunters to avoid shooting c o l l a r e d w i l d l i f e since 1985. The bias against k i l l i n g c o l l a r e d bears i s unknown, but probably minimal. Between 12 and 20 hunters were permitted to hunt g r i z z l y bears i n the study area"each spring, and when I encountered hunters i n the f i e l d they were informed that they were allowed to shoot a c o l l a r e d bear i f they desired. Also, c o l l a r s were black and d i f f i c u l t f o r a hunter or poacher to see i n most f i e l d conditions. I was not aware of any hunter who refrained, from shooting a bear because i t was c o l l a r e d . Hunters i n B r i t i s h Columbia, Alberta, and Montana must report g r i z z l y bears they k i l l , so information on l e g a l k i l l s i s r e a d i l y a v a i l a b l e f o r both r a d i o - c o l l a r e d and unmarked bears. Accurately documenting i l l e g a l and natural m o r t a l i t i e s was more d i f f i c u l t . A l l c o l l a r e d bears were relocated from a i r c r a f t approximately once each week as well as from the ground, and the motion-sensitive radio transmitters used ale r t e d us to the death of the animal or removal of a c o l l a r . When t h i s happened, the l o c a t i o n was inspected and the cause of sig n a l change determined. To f i n d unmarked bears k i l l e d f o r reasons other than l e g a l hunting, I r e l i e d on information from people reporting the incident. Seasonal and annual s u r v i v a l rates of r a d i o - c o l l a r e d bears were estimated using the number of days that animals were tracked and by the number of deaths (Heisey and F u l l e r 1985). The method assumes equal mortality rates within seasons, so i n t h i s chapter, seasons were selected d i f f e r e n t l y than i n Chapters 2 and 3. For estimating mortality rates, seasons were: 1) spring - May and June and included the g r i z z l y bear hunting season i n B r i t i s h Columbia; 2) summer - July and August; 3) autumn - September to November, and included the general, b i g game hunting season; and 4) winter - December to A p r i l when most bears were hibernating. Deaths related to research were not included i n c a l c u l a t i n g s u r v i v a l rates. Survival rates for cubs and yeaxlings- of c o l l a r e d females were ca l c u l a t e d from the number of days between my f i r s t observation of them to when they were l a s t seen. I f a cub disappeared, i t was assumed to be dead. S i m i l a r l y , missing yearlings were also assumed dead unless the e n t i r e l i t t e r separated from the mother during the mating season (May and June), which i s the most common time of family break-up. However, some cubs can survive i f separated from t h e i r mother (Russell et a l . 1979), and separation i s common for yearlings; thus s u r v i v a l estimates f o r these two classes may be low. The time of family break-up could be a time of high mortality f o r young bears because of aggressive male s u i t o r s . A p o t e n t i a l bias r e s u l t s i f s u r v i v a l rates of uncollared young approaching t h i s period are included i n the c a l c u l a t i o n s . I was able to record s u r v i v a l information on only 10 bears during t h i s separation period, 9 of which were known to have survived, but the fate of the other i s unknown. To reduce possible bias, the number of days between the l a s t observation of a young bear when i t was s t i l l with i t s r a d i o - c o l l a r e d mother and when i t was f i r s t captured a f t e r family break-up, were not counted as days survived. Complete annual s u r v i v a l rates of cubs were not determined because I d i d not enter maternal dens to count cubs at b i r t h ! Assuming a 1 February birthdate, t h e i r average age when f i r s t observed was 145 days, and cub s u r v i v a l was calculated only for the following 220 days (365-145). N a t a l i t y and Rate of Increase Information on reproductive rates, including the age of f i r s t p a r t u r i t i o n , i n t e r v a l between l i t t e r s and the number of cubs per l i t t e r , was c o l l e c t e d by radio-tracking i n d i v i d u a l female bears over several years. Additional data on l i t t e r s i z e were recorded from unmarked females observed i n , or adjacent to, the study area. Calculating the annual s u r v i v a l rate of cubs f o r only the l a s t 220 days of the year eliminates the bias of f i r s t observing l i t t e r s when the cubs were an average of 145 days old. I f one cub died before I observed the l i t t e r , the recorded l i t t e r s i z e would be one l e s s than the true value, but the recorded s u r v i v a l rate would be higher. S i m i l a r l y , i f a whole l i t t e r was l o s t p r i o r to observation, these m o r t a l i t i e s again would go unrecorded, but the breeding i n t e r v a l estimate would be extended, or i f the l i t t e r was the female's f i r s t , her recorded age of f i r s t p a r t u r i t i o n would be i n f l a t e d . I used 3 methods of estimating reproductive rates. The f i r s t and second included only information from female bears that were radio-tracked through at l e a s t one i n t e r b i r t h i n t e r v a l . For method 1, the reproductive rate was the t o t a l number of cubs observed, divided by the t o t a l number of bear-years required to produce them. This method gives more weight to i n d i v i d u a l s that were tracked f o r more than one i n t e r b i r t h i n t e r v a l and had more l i t t e r s recorded. For method 2, the reproductive rate was the average of each i n d i v i d u a l ' s rate. Method 2 gives equal weight to each bear regardless of how many of her l i t t e r s were recorded. For method 3, the reproductive rate was the average l i t t e r s i z e divided by the average i n t e r b i r t h i n t e r v a l . More information was a v a i l a b l e f o r estimating the reproductive rate using method 3 because a l l l i t t e r s i z e data could be used. Of the several d i s t i n c t measures of rate of increase described by Caughley (1977:53), I chose to use the s u r v i v a l -fecundity rate, r g , which i s the exponential rate at which a population would change i f i t had a stable age d i s t r i b u t i o n appropriate to i t s current schedules of age-specific s u r v i v a l and fecundity. Caughley c a l l e d t h i s measure of increase "demographic vigour", because i t measures how well populations contend with t h e i r current s i t u a t i o n s . I estimated t h i s rate of increase by i t e r a t i n g the Lotka equation (Caughley 1977:110). The s e n s i t i v i t y of r to changes i n parameter values was tested by a l t e r i n g the value of each parameter alone, while leaving the others at the best estimate from the a v a i l a b l e data, u n t i l the rate of increase equalled 0.0. I also changed a l l s u r v i v a l rates equally u n t i l a zero rate of increase was obtained. 77 RESULTS Densities Of the 65 d i f f e r e n t g r i z z l y bears that I captured, 54 were caught i n the trapping areas and of these, 47 were equipped with radios. The multi-annual (x = 2.1 years, range = 0.9 to 5.0) home range information used f o r the density estimates was based on 4482 radio locations- of 13 male-.and 14 female c o l l a r e d i n d i v i d u a l s (x = 166 r e l o c a t i o n s / i n d i v i d u a l , range = 21 to 551). Five c o l l a r e d bears t r a v e l l e d with t h e i r c o l l a r e d mothers, so t h e i r telemetry data were redundant. F i f t e e n other bears were captured i n the trapping areas but were not used i n the density estimates. These included 3 males and 2 females caught at the extremity of t h e i r home ranges and on the periphery of the trapping area, leaving no overlap, 2 males which died shortly a f t e r capture, 4 males and 1 female that shed t h e i r c o l l a r s within 1 month of capture and were never recaptured or harvested and therefore the l o c a t i o n of t h e i r range was unknown, and 1 female cub and 2 adult males that were captured and ra d i o - c o l l a r e d i n the autumn of 1986 and whose home range locations were also unknown. The average range of the 27 bears used i n estimating density had 42% overlap with trapping area A and 28% overlap with trapping area B. On average, 41% of these bears' a e r i a l locations were i n trapping area A and 28% i n area B. The estimated average minimum density of bears i n the 2 larger trapping area A was 4.5/100 km using both methods. 78 This i s known to be a low estimate because unmarked bears were often observed i n the west side of t h i s area, but i t was too large to trap i n t e n s i v e l y . In area B, average de n s i t i e s of 6.0 2 . and 6.1/100 km were recorded over 8 years using the home range and a e r i a l l o c a t i o n methods respectively (Tables 4 and 5). The de n s i t i e s calculated f o r 1979 and 1980 may be low because few of the subadults captured then were ra d i o - c o l l a r e d , thus violating- the assumption that a l l bears were counted. The average density of g r i z z l y bears estimate f o r the l a s t 6 years 2 of study was 6.4/100 km , and the highest annual estimate was 8/100 km2. Age-sex Composition Of the 65 g r i z z l y bears captured, only 3 were cubs and a l l of these were female. Seven females and 10 males were yearlings when f i r s t caught, f o r a 50:50 sex r a t i o i n the f i r s t 2 age classes. Of the bears which were subadults when f i r s t captured, 11 were male and 8 female, which i s not s i g n i f i c a n t l y d i f f e r e n t from an equal r a t i o (P = 0.64). S i m i l a r l y , the r a t i o of captured adult bears, 16 male to 10 female, was not s i g n i f i c a n t l y d i f f e r e n t from equality (P = 0.32). Based on the density of bears i n the study area, I estimated that 21.5% of the population were cubs, 17.5% were yearlings, 26.5% were subadults and 34.5% were adults (Tables 4 and 5). Though there were more adult males captured than adult females, the density estimates indicate that there are more adult females than males i n the trapping areas. 79 Table 4. Numbers and dens i t i e s (bears/100 km ) of bears i n Trapping Area B each year determined by the proportion of 97.5% multi-annual home ranges which overlapped t h i s area. Acre--sex Classes of Bears Year- Sub- Adult Adult Year Cubs l i n a s Adults Males Females Total Density 1979 2.45 0.31 0.40 1.12 2.00 6.28 4.83 1980 0.0 2.45 0.34 1.19 2.00 5.98 4.60 1981 1.94 0.25 2.28 1.12 1.69 7.28 5.60 1982 1.33 2.19 1.10 0.98 1.69 7.29 5.61 1983 1.28 1.15 2.50 0.98 1.69 7.60 5.85 1984 1.33 0.61 2.27 1.08 1.69 6.98 5.37 1985 2.40 2.00 3.25 1.08 1.69 10.42 8.02 1986 2.20 2 .54 1.99 1.29 2.18 10.20 7.85 Overall Mean 1.62 1.44 1.77 1.11 1.83 7.75 5.96 % 23 18 22 14 23 100 _____ 1981-•86 Mean 1.75 1.46 2.23 1.09 1.77 8.30 6.38 % 21 18 27 13 21 100 80 Table 5. Number and dens i t i e s (bears/100 km ) of bears i n Trapping area B determined by the proportion of t h e i r multi-annual a e r i a l relocations (= time) which occurred i n t h i s area. Age-sex Classes of Bears Year- Sub- Adult Adult Year Cubs li n g s Adults Males Females Total Density 1979 2.20 0.67 0.27 1.04 2.33 6.51 5.01 1980 0.0 2.20 0.78 1.16 2.33 6.47 4.98 1981 2.07 0.25 2.56 1.04 1.70 7.62 5.86 1982 0.99 2.24 0.95 0.97 1.70 6.85 5.27 1983 1.36 0.99 2.42 0.97 1.70 7 .44 5.72 1984 1.38 0.59 2.77 0.97 1.70 7.41 5.70 1985 2.20 2.00 3.27 0.97 1.70 10.14 7.80 1986 3 .16 2.34 1.31 1.33 2.56 10.70 8.23 Overall. Mean 1.67 1.41 1.79 1.06 1.97 7.89 6.07 % 21 18 23 13 25 100 1981-•86 Mean 1.86 1.40 2.21 1.04 1.84 8.36 6.43 % 22 17 26 13 22 100 81 Survival Rates Sampling i n t e n s i t y of the 55 r a d i o - c o l l a r e d bears was s i m i l a r among seasons with an average of 7.3, 7.3, 7.4 and 7.5 bear-years of functioning c o l l a r s per month fo r spring to winter re s p e c t i v e l y . Radio-collared bears and t h e i r dependent o f f s p r i n g were monitored f o r a t o t a l of 110 bear-years. Excluding the 2 deaths r e l a t e d to trapping, 7 m o r t a l i t i e s were known or suspected to have occurred to bears with active c o l l a r s , plus 4 deaths of t h e i r dependent o f f s p r i n g (Table 6). Adult bears were tracked f o r 57.6 bear-years, during which 4 known or suspected deaths occurred for an average su r v i v a l rate of 0.93. Of these deaths, 2 were females and 2 were males, f o r estimated s e x - s p e c i f i c s u r v i v a l rates of 0.94 and 0.92 r e s p e c t i v e l y (Table 6). Radio contact was l o s t p r i o r to the l i f e expectancy of the c o l l a r on one 9-year-old male and h i s fate remains unknown. This bear was l a s t located during the mating season when other c o l l a r s have f a i l e d because they were damaged by mates. I t i s also possible that he dispersed out of the study area, but i f t h i s male was i l l e g a l l y k i l l e d and the c o l l a r s destroyed or removed, the annual s u r v i v a l rate of adult males would have been 0.88. A decrease i n s u r v i v a l of older bears was indicated. Both adult female m o r t a l i t i e s , 1 natural and 1 i l l e g a l k i l l i n g , were of bears 18 years of age and older. Subadult females had an annual s u r v i v a l rate of 0.94 and Table 6. The best.estimates of annual s u r v i v a l rates of the r a d i o - c o l l a r e d g r i z z l y bears and t h e i r dependent o f f s p r i n g i n the Flathead Valley, 1979 to 1987. Age/sex Bear-years S u r v i v a l rate class of tracking M o r t a l i t i e s (95% CL) Adult Females 34. .5 2 a 0. .94 (• . 87* -1, ,0) Adult Males 23 . . 1 2 . o. 92 (• .81--1. ;0) Subadult Females 16. .5 1 0. ,94 (• .84-- i . .0) Subadult Males 11. .0 l b 0. ,91 (• .76-- l . ,0) Yearlings 15. .3 2 0. .88 (• .73-- l . ,0) Cubs 9. .3 3 0. .82 (• . 66-- l . •0)< does not include 1 i l l e g a l l y k i l l e d when recovering from drugging. k does not include 1 k i l l e d by another bear while trapped. c c a l c u l a t e d for 220 days. 83 subadult males 0.92 a f t e r a t o t a l 27.5 bear-years of active c o l l a r s . Yearlings were tracked f o r 3.0 bear-years by c o l l a r i n g them and f o r 12.3 bear-years by tracking t h e i r mothers. With one suspected i l l e g a l m ortality of a c o l l a r e d y e a r l i n g , the estimate of the annual s u r v i v a l rate was 0.94. A second y e a r l i n g mortality may have occurred when a l i t t e r changed from 3 to 2, however, a young g r i z z l y was observed sho r t l y a f t e r i n a lo c a t i o n favoured by t h i s family group. I f t h i s young bear d i d die, y e a r l i n g annual s u r v i v a l would be 0.88. Cubs had a 220 day s u r v i v a l rate of 0.82. I f cub su r v i v a l rate was the same before they were f i r s t observed as af t e r , t h e i r annual rate would have been 0.73. Three bears with operating c o l l a r s died i n the spring and 4 died i n the autumn (Table 7). Although the m o r t a l i t i e s were seasonal, the average annual s u r v i v a l f o r a l l c o l l a r e d bears was calculated to be 0.92 by both the product of the seasonal rates and using a 1-year i n t e r v a l , r e f l e c t i n g the even sampling i n t e n s i t y among seasons. Causes of Mo r t a l i t y Between 1979 and 1987, 30 g r i z z l y bears are known to have died or been removed from the study area, and there i s evidence that 8 ad d i t i o n a l m o r t a l i t i e s probably occurred. These 38 instances included 9 bears with operating radio c o l l a r s and 7 marked bears (Table 8). Three m o r t a l i t i e s were re l a t e d to trapping. An uncollared y e a r l i n g was k i l l e d by i t s mother when both were captured at 84 Table 7. Seasonal s u r v i v a l rates o f r a d i o - c o l l a r e d g r i z z l y bears i n the Flathead Valley, 1979 to 1987. N days i n N bear- Interval Season Interv a l i n t e r v a l years M o r t a l i t i e s Rate Spring May—Jun 14-. 7 3 0.966 Summer Jul-Aug 62 14.5 0 1.000 Autumn Sep-Nov 91 22.1 4 a 0.956 Winter Dec-Apr 151 37.8 0 1.000 a does not included 1 i l l e g a l l y k i l l e d when recovering from drugging or 1 k i l l e d by another bear while trapped. 85 Table 8. Known and suspected g r i z z l y bear m o r t a l i t i e s i n the 2900 km2 study area, Flathead Valley, 1979 to 1987. Source of Adult Subadult mortality Male Female Male Female Yearling Cub Unknown Total Natural death. 0 K D a 0 0 1 4 0 6(1) Legal harvest 6(1) 5 7(1) 1 0 0 0 19(2) I l l e g a l harvest 0 2( 2 ) b 1 1(1) 2(1) 1 1 8(4) Legal control 0 0 1 0 0 0 0 1 I l l e g a l nuisance 2(1) 0 0 0 0 0 0 2(1) Research 0 0 I d ) 0 1 0 0 2(1) Total 8(2) 8(3) 10(2) 2(1) 4(1) 5 1 38 (9) Numbers i n parentheses represent m o r t a l i t i e s of bears with active radio c o l l a r s . includes 1 i l l e g a l l y k i l l e d when recovering from drugging. 86 the same location, and a ra d i o - c o l l a r e d subadult male was k i l l e d by an unknown bear. In the t h i r d case, a r a d i o - c o l l a r e d adult female was shot out of hunting season by a poacher before she had f u l l y recovered from being immobilized. This mortality was c l a s s i f i e d as i l l e g a l harvest (Table 8), but, because she would not have been k i l l e d without the research, i t was not included i n c a l c u l a t i n g mortality rates (Tables 6 and 7). Of a l l known m o r t a l i t i e s , l e g a l harvest was the most commonly recorded cause, accounting f o r 19 known deaths. Seven of 13 successful hunters whom.I interviewed s a i d that they had been on a road when they shot t h e i r bear. Probably, I did not record a l l g r i z z l y bears k i l l e d i l l e g a l l y i n the study area. Of the 9 m o r t a l i t i e s of bears wearing functioning radio c o l l a r s , 2 were a t t r i b u t e d to l e g a l hunting while 5 were i l l e g a l (Table 8). Six of the 10 known or suspected i l l e g a l k i l l s of marked and unmarked bears were i n 2 the 119 km portion (4%) of the study area where the only permanent human settlement i s concentrated. No known or suspected g r i z z l y bear m o r t a l i t i e s occurred at any i n d u s t r i a l camps, though some black bears were k i l l e d and others trapped and removed. One 18 year-old female died i n the autumn of what I suspect was a natural cause, although i t i s possible that she was i l l e g a l l y shot. Reproductive Parameters A t o t a l of 27 female g r i z z l y bears was captured and marked, and 23 f i t t e d with radio c o l l a r s . The sample contained 11 ra d i o - c o l l a r e d adults and 5 of these were radio-tracked f o r at le a s t 1 complete breeding cycle. Two bears had t h e i r f i r s t l i t t e r when known to be 5 and 6 years of age. Three others were estimated by cementum annuli to be 5, 6, and 8 years of age when they had t h e i r f i r s t l i t t e r . The average age of f i r s t p a r t u r i t i o n of these 5 females was 6.0 years. Nine i n t e r b i r t h i n t e r v a l s were recorded f o r 5 females; 1 was 1 year, 2 were 2 years, 5 were 3 years, and 1 was 4 years, for an average of 2.67 years. A l l cubs i n the l i t t e r of a 3-year and of the 1-year i n t e r b i r t h i n t e r v a l died. At f i r s t observation, r a d i o - c o l l a r e d females had 4 0 cubs i n 17 l i t t e r s f o r an average l i t t e r s i z e of 2.35 cubs per female (Table 9). The average s i z e of 21 l i t t e r s of yearlings and 2 year-olds was 2.33. Adding l i t t e r s of uncollared females reduces the average l i t t e r s i z e of cubs to 2.26 and that comprised of 1 or 2-year-olds to 2.29. Of the 40 cubs of ra d i o - c o l l a r e d females, 21 were eventually captured and 11 were males and 10 were females. Most of these were captured as yearlings, and I do not know i f behavioural differences between the sexes influenced the p r o b a b i l i t y of capture. Reproductive Rates Five females were radio-tracked for at l e a s t one i n t e r b i r t h i n t e r v a l and produced 24 cubs i n 21 bear-years f o r an estimated 88 Table 9. L i t t e r s i z e s t a t i s t i c s of g r i z z l y bears observed i n the Flathead Valley, 1979 to 1987. L i t t e r s Frequencies of Number of Total number Average from: l i t t e r s i z e s l i t t e r s of young l i t t e r s i z e 1 2 3 Collared Females Cub a 1 9 7 17 40 2.35 1 y r 1 7 6 14 33 2.36 2 y r l_ 3. 3 7 16 2.29 Uncollared Females Cub 3 6 5 14 30 2.14 1-2 1 6 3 10 22 2.20 A l l Females Cub 4 15 12 31 70 2.26 1-2 3 16 12 31 71 2.29 Totals 7 31 24 62 141 2.27 Cubs = l i t t e r s O f cubs; 1 yr = l i t t e r s of yearlings ; 2 yr = l i t t e r s of 2 year olds; 1-2 = l i t t e r s of both yearlings and 2 year olds. 89 reproductive rate of 0.875. These f i v e females had i n d i v i d u a l reproductive rates of 0.67, 1.0, 0.83, 0.75, and 1.0 cubs per year f o r an average of 0.85. The t h i r d estimate of the reproductive rate i s the average l i t t e r s i z e divided by the average i n t e r b i r t h i n t e r v a l ; 2.26/2.67 = 0.85. Survival-fecundity Rate of Increase For estimating the survival-fecundity rate of increase using the Lotka equation, the age of f i r s t p a r t u r i t i o n was 6 . years, with one t h i r d of the females having an average sized l i t t e r at 5 years, one t h i r d at 6 years and one t h i r d at 7 years. By averaging the 3 estimates of reproductive rate and assuming a 50:50 sex r a t i o , females, 8 years and older, had reproductive rates, mx, df 0.43. The values f o r s u r v i v a l rates were obtained from Table 6, except those f o r adults 15 years of age and older. The mortality rate for these older bears was set at 0.82, as was used by Knight and Eberhardt (1985), because an increased rate i n older classes i s expected (Caughley 1966) and my l i m i t e d data suggested i t . Using these data (Table 10), the population's rate of increase was estimated to be 0.081. Although the samples represented almost the e n t i r e study population, actual numbers were small, and therefore I tested the s e n s i t i v i t y of the rate of change to errors i n the estimates. To have a rate of increase of 0.0, the most s e n s i t i v e parameter, prime adult survival rate, would have to be reduced from 0.94 to 0.76, while, the subadult s u r v i v a l rate Table 10. Values used to c a l c u l a t e of the rate of increase, using the Lotka equation f o r the g r i z z l y bear population of the Flathead Valley, 1979 to 1987. Reproductive Proportion Annual Stable Rate Surviving s u r v i v a l age Age (mx) ( l x ) rate d i s t r i b u t i o n cub 0 1.000 .82 .166 i - i . 0 .820 .88 . 126 2 0 .722 .94 . 102 3 0 .678 .94 .088 4 0 .638 .94 .077 5 .38 .599 .94 .067 6 .38 .563 • 94 .058 7 .38 .530 .94 . 050 8 .43 .498 .94 .043 9 .43 .468 .94 .038 1° .43 .440 .94 .033 11 .43 .413 .94 .028 12 .43 .389 .94 . 024 13 .43 .365 .94 .021 14 .43 .343 .94 . 018 15 •43 .323 .82 .016 16 .43 .265 .82 .012 17 • 43 .217 .82 .009 18 .43 .178 .82 .007 19 .43 .146 .82 .005 Table 10. Continued. 20 .43 .120 .82 .004 21 .43 .099 .82 .003 22 .43 .080 .82 .002 23 .43 .066 .82 .002 of increase. A y e a r l i n g s u r v i v a l rate of 0.35 or a cub s u r v i v a l . o f 0.33, would both decrease r to 0.0, i f a l l the mortality occurred a f t e r the mating season. Greater mortality of these young bears could occur i f deaths were p r i o r to the mating season, because the mother's i n t e r v a l length could shorten as a r e s u l t . An 8% decrease i n a l l s u r v i v a l rates would r e s u l t i n a zero rate of increase, as would a 7% decrease i n a l l s u r v i v a l rates and i n the reproduction rate. Although, jny data concerning the age at f i r s t p a r t u r i t i o n are meager, females would have to be over 13 years old for r to s t a b i l i z e at 0.0. S i m i l a r l y , the other 2 reproduction parameters would require u n r e a l i s t i c values to obtain a zero rate of increase. At the stable age d i s t r i b u t i o n appropriate to the population parameter estimates, 44% of the females would be adults while only 40% of the l i v e population's were reported to be. I t was also estimated that the l i v e population of females consisted of 20% cubs, 16% yearlings and 24% subadults, while at a stable age d i s t r i b u t i o n , the proportions would be 17%, 13%, and 27% res p e c t i v e l y . DISCUSSION Methods fo r Estimating Density Caution must be exercised when comparing my density estimates with those from other areas because of the d i f f e r e n t methods employed. For example, i f I had used the method of recording a l l the d i f f e r e n t bears trapped and observed i n the would have to be reduced from 0.94 to 0.69 to get a zero rate study area as was done by others (Mundy and Flook 1973, Martinka 1974, Pearson 1975, Russell et a l . 1979), the highest 2 annual density would have been 20 bears/100 km , which i s 2.5 times greater than the highest annual density estimated using my methods. S i m i l a r l y , Russell et a l . (1979) estimated that they had captured a l l but 3 bears i n t h e i r study area. From t h e i r home range data, I made- an approximate density estimate using the proportion that each c o l l a r e d bear's home range overlapped , t h e i r study area, and gave the 3 uncollared bears the average value. This resulted i n a minimum density f o r the Jasper 2 study area of 0.56 bears/100 km , which i s l e s s than h a l f the 2 . 1.25/100 km estimated when Russell et a l . (1979) included the 3 unmarked bears. In making t h i s re-analysis, I wish only to demonstrate p o t e n t i a l differences between methods, not to give an a l t e r n a t i v e density estimate f o r Jasper National Park. The open system bias w i l l decline as the s i z e of the study areas increases and/or the ranges of the bears decrease. Population Density and Composition LeFranc et a l . (1987:52-53) tabulated density estimates of g r i z z l y bears from 20 i n t e r i o r study areas i n North America. More than 1 estimate was presented i n several of these lo c a t i o n s , but the average of the highest estimate from each 2 area was 1.68 bears/100 km . Glacier National Park, B r i t i s h Columbia, had the highest recorded densities of the 20 areas, 2 . with 3.6 to 5.6 bears/100 km . Even with the r e l a t i v e l y conservative methods I used, density estimates i n my Flathead study area are higher than any of these. Higher d e n s i t i e s of brown bears have been reported only i n coastal Alaska (LeFranc et a l . 1987; J . Schoen, pers. comm.) and i n Eurasia (Zunino and Herrero 1972, K i s t c h i n s k i 1972). Annual v a r i a t i o n i n the density of bears i n my study area appears due to 3 factors, 1) synchronous weaning of o f f s p r i n g among females, 2) my i n a b i l i t y to capture and therefore include a l l weaned o f f s p r i n g , and 3) the enlarged- ranges cxt those dispersing subadults I did capture. Consequently, years with high d e n s i t i e s occurred were when most females were accompanied, e i t h e r by cubs or by yearlings, and low density years are when these same females were s o l i t a r y . The d i f f i c u l t y of keeping track of independent subadult bears probably resulted i n them being underestimated. Because the large home ranges of adult males (Bunnell and T a i t 1980) reduces the p r o b a b i l i t y of capturing a l l whose 97.5% range overlapped the study area, t h i s c l a s s i s probably underestimated. Adult males a l s o removed t h e i r c o l l a r s more frequently than did other age/sex classes. Four adult males, known to have been near the center of trapping area B, were not included i n the age/sex composition or the density estimate because I was not sure of t h e i r range l o c a t i o n s . For these reasons, the actual density of bears i n my study area i s probably higher than estimated here. The standing age d i s t r i b u t i o n s , derived by both my methods (Tables 4 and 5), are s i m i l a r to those that Aune et a l . (1986) reported for a g r i z z l y bear population on the eastern edge of the Rocky Mountains. Both populations appear to have s l i g h t l y higher proportions of cubs and yearlings, and fewer adults than reported i n G l a c i e r and Yellowstone National Parks (Martinka 1974, Craighead et a l . 1974). Survival Rates The estimated s u r v i v a l rates of adult g r i z z l y bears (0.93) and cubs (0.82) i n the Flathead are s i m i l a r to those recorded for bears between 5 and 15 years of age (0.91 and 0.93) and for cubs (0.80 and 0.89) i n Yellowstone National Park by Craighead et a l . (1974) and by Knight and Eberhardt (1985) res p e c t i v e l y . The s u r v i v a l rates of Flathead yearlings (0.94) and subadults (0.93) was higher than those reported i n Yellowstone. Craighead et a l . (1974) estimated s u r v i v a l rates of 0.95 f o r y e a r l i n g females, 0.68 fo r y e a r l i n g males, 0.80 for subadult females, and 0.73 f o r subadult males during the period when garbage dumps were i n use i n the Park. Knight and Eberhardt (1985) noted the d i f f i c u l t y i n obtaining an accurate estimate of these values, but suggested s u r v i v a l rates of 0.73 for yearlings and 0.75 f o r subadults i n Yellowstone a f t e r the dumps were closed. Causes of Mo r t a l i t y I suspected that only 1 recorded m o r t a l i t y was due to natural causes during 85 bear-years that adult and subadult g r i z z l y bears were radio-tracked, and I found no evidence of other unmarked adult or subadult bears dying from natural causes. S i m i l a r l y , no natural m o r t a l i t i e s were reported i n approximately 27 bear-years of radio-tracking adult and subadult g r i z z l y bears i n other studies i n the mid Rocky Mountains (Hamer et a l . 1983, Hamer and Herrero 1983, Russell et a l . 1979, Servheen 1981, Sizemore 1980). Instead, 6 known or suspected deaths of bears with, functioning radio c o l l a r s were caused by people i n my study area, excluding the 2 research r e l a t e d losses. In the other studies mentioned above, another 7 bears with functioning radio c o l l a r s were ei t h e r k i l l e d by people, or were removed from the study area as part of bear-people management programs. Although most recorded deaths of g r i z z l y bears were due to l e g a l hunting, the le s s biased sample of r a d i o - c o l l a r e d bears indicated that i l l e g a l k i l l i n g was perhaps even more common, p a r t i c u l a r l y near human settlement. No recorded or suspected g r i z z l y bear deaths were d i r e c t l y r e l a t e d to the i n d u s t r i a l a c t i v i t y i n the study area. S h e l l Canada Ltd. does not permit firearms i n i t s camps and discourages the use of p r i v a t e vehicles i n the area. A f t e r 1980, garbage was incinerated at least once each day. Although g r i z z l i e s were occasionally reported near camps, they did not remain f o r long. The logging camps d i d not have in c i n e r a t o r s , but the l a r g e s t camp moved i t s garbage d a i l y to a dump approximately 6 km away. Some smaller camps had poor garbage management, with dumps as close as 100 m from the cook-house. Black bears were known to have been shot i n these 97 camps, but I was not aware i f g r i z z l i e s were also k i l l e d . There was no evidence that any adult or subadult g r i z z l y bears died i n d i r e c t l y because of the i n d u s t r i a l a c t i v i t y . I f stress or displacement caused by the ind u s t r i e s had impaired foraging e f f i c i e n c y , a l t e r e d the animal's physiology, or caused an energetic drain to the extent that bears eventually died because of i t , they would have been recorded as natural m o r t a l i t i e s . Because the only natural m o r t a l i t y recorded was of an old bear who's range was i n a portion of the study area with r e l a t i v e l y l i t t l e i n d u s t r i a l a c t i v i t y , I conclude that the l e v e l of human a c t i v i t y experienced by the study bears was not s u f f i c i e n t to cause the death of these age classes of bears. One cub was eaten by i t s mother j u s t p r i o r to leaving the maternity den, 8 months a f t e r the female had seismic a c t i v i t y i n her seasonal range, but i t i s doubtful that the 2 events were re l a t e d (see Chapter 3) . N a t a l i t y Mean l i t t e r s i z e i s the parameter estimate based on the lar g e s t sample i n t h i s and other studies, and so i s l i k e l y the most usefu l f o r comparisons. The average l i t t e r s i z e of 2.26 i n the Flathead i s s i m i l a r to 2.29 reported on the Rocky Mountain East Front (Aune and Brannon 1987), and 2.24 i n Yellowstone National Park p r i o r to the c l o s i n g of the garbage dumps (Craighead et a l . 1974). I t i s higher, however, than the average i n other i n t e r i o r areas: 1.7 i n both G l a c i e r National Park, Montana (Martinka 1974, Kendall 1985) and southern Yukon (Pearson 1975), 1.9 i n Yellowstone Park a f t e r the dumps were closed (Knight and Eberhardt 1985), and 2.0 i n Canadian Rocky Mountain National Parks (Mundy and Flook 1973). Data on i n t e r b i r t h i n t e r v a l s are l e s s numerous than those on l i t t e r s i z e i n the l i t e r a t u r e , and d i f f e r e n t studies have reported i t i n s l i g h t l y d i f f e r e n t ways (Knight and Eberhardt, 1985). In Yellowstone Park, Knight and Eberhardt (1985) calcula t e d an average of 3.0 years between l i t t e r s , and suggested that 3.2 years would be a comparable- frequency f o r pre-dump closure periods using Craighead et a l . ' s (1974) data. At McNeil River, Alaska, Glenn et a l . (1976) recorded an average of 3.6 years f o r 12 i n t e r b i r t h i n t e r v a l s . A l l these estimates are longer than the 2.67 years estimated i n the Flathead study area. In t h e i r review, Bunnell and T a i t (1981) suggested that the n a t a l i t y rate of bears was c o n t r o l l e d by n u t r i t i o n a l factors, generally i n a density independent fashion. However, McCullough's (1981) and Stringham's (1983) analysis of g r i z z l y bear data c o l l e c t e d i n Yellowstone National Park by Craighead et a l . (1974), showed a negative c o r r e l a t i o n between recruitment and the number of adult males. They did not d i f f e r e n t i a t e the cause of the apparent density dependence between changes i n n a t a l i t y rate, cub mortality rate, nor examine the possible confounding (Stringham 1983) introduced by the a v a i l a b i l i t y of natural foods. G r i z z l y bear l i t t e r s i z e s within the unhunted and unsettled G l a c i e r National Park, Montana, are among the smallest recorded on the continent (Martinka 1974, Kendall 1985). Immediately outside the park, g r i z z l i e s are hunted, and as about twice as many males as females were shot, t h i s should have affected the sex r a t i o of l i v i n g population of adult bears (McLellan and Shackleton 1988). In t h i s unprotected area, both Aune and Brannon (1987) and I found g r i z z l y bear l i t t e r s i z e s to be among the l a r g e s t i n North America. This information, combined with the rare occurrence of natural mortality of subadult and older bears i n a l l telemetry studies made i n the mid-Rocky Mountains, supports the hypothesis that density-dependent population regulation acts on both reproductive and mortality rates. The density-dependent mechanism that a f f e c t s cub production may not be a s i g n i f i c a n t reduction i n food at high bear d e n s i t i e s , but rather a reduction i n high q u a l i t y habitats a v a i l a b l e to females accompanied by o f f s p r i n g and to subadults, due to the presence of adult males (Pearson 1975, Russell et a l . 1979, t h i s t h e s i s Chapters 2 and 3). In unhunted areas such as parks and reserves, adult males should be r e l a t i v e l y numerous and female displacement common (Pearson 1975, Russell et a l . 1979), whereas hunted populations contain fewer males and segregation i s not as pronounced. This could be more adequately investigated by a comparative population, habitat and behaviour study, focusing on age/sex cl a s s d ifferences, between G l a c i e r National Park, Montana, and adjacent study areas. 100 Rate of Increase Although few rates of increase have been reported, the value of 0.081 I estimated appears high f o r g r i z z l y bears, r e f l e c t i n g both high n a t a l i t y and low mortality rates. My samples on some reproductive parameters such as the age of f i r s t p a r t u r i t i o n are l i m i t e d , but u n r e a l i s t i c values would be required to reduce the rate of increase to zero. This i s not the case f o r s u r v i v a l rate values. I f subadult female su r v i v a l rates were as low as those given by Craighead et a l . (1974) or by Knight and Eberhardt (1985), the rate of increase would decrease to 0.037 and 0.015 respectively. Although the rate of change i s even more s e n s i t i v e to changes i n adult mortality, the l a r g e r sample of bear-years recorded f o r t h i s age c l a s s , and the s i m i l a r i t y of the rates reported between studies, increases confidence i n my estimate. E f f e c t s of Industry I f the l e v e l of a c t i v i t y of the resource extraction i n d u s t r i e s operating during my study had caused female g r i z z l y bears to expend a d d i t i o n a l energy, reduce foraging e f f i c i e n c y , or use sub-optimal habitats to a degree that the population's demography was s i g n i f i c a n t l y affected, I expected to f i n d some i n d i c a t i o n of impaired reproduction. One female l o s t her e n t i r e l i t t e r of 4 month old cubs approximately 8 months a f t e r a seismic l i n e traversed her seasonal range (see Chapter 3). Although the 2 events were probably not related, i t was the only reduction i n reproduction remotely a t t r i b u t a b l e to the 101 a c t i v i t i e s of resource i n d u s t r i e s . The l i t t e r s i z e s I recorded were as high or higher than most other g r i z z l y bear populations with or without resource i n d u s t r i e s . S i m i l a r l y , the survival-fecundity rate of increase or "demographic vigor" (Caughley 1977:55) of the g r i z z l y bear population was high during the period of resource extraction, i n d i c a t i n g a productive population. "Because I found both a high density of g r i z z l y bears and a p o s i t i v e rate of increase during a period of i n d u s t r i a l a c t i v i t y , I r e j e c t my main hypothesis. CHAPTER 5; CONCLUSIONS It i s very d i f f i c u l t to conduct large scale experiments with a wide-ranging and long - l i v e d species. This i s p a r t i c u l a r l y true when the species occurs at low d e n s i t i e s t y p i c a l of g r i z z l y bears, without some confounding and consequent d i f f i c u l t i e s f o r i n t e r p r e t a t i o n . The design of my study does not allow me to conclude that: tire Indus t r i a l a c t i v i t y i n the study area had no e f f e c t cn the demography of the g r i z z l y bear population. Reducing the amount of l e g a l k i l l by changes i n hunting regulations between 1968 and 1982 confounds the influence of industry. The p o s i t i v e rate of increase may be the population's response to r e l a x i n g a factor (hunting) which kept i t below carrying capacity. Although g r i z z l y bears used cutt i n g u n i t s l e s s than expected, i t i s also possible that timber harvest improved the bear's habitat over the short term by increasing ungulate numbers which are an important bear food. Unfortunately, there are no data on ungulate response to the logging, but my subjective estimates indicate that both w h i t e - t a i l deer (Odocoileus virqinianusV and moose (Alces alces) populations increased during the study period, while e l k (Cervus elaphus) decreased i n numbers. This possible ben e f i t may be only short term because ungulate numbers may decrease a f t e r second growth timber i s established. There was l i t t l e evidence that the l e v e l and type of habitat a l t e r a t i o n i n the study area had a serious e f f e c t on 103 the g r i z z l y population. Although there were s t r i k i n g exceptions, the i n t e g r i t y of the 2 most important spring habitats, r i p a r i a n and snowchutes, was maintained i n most loc a t i o n s . S i m i l a r l y , burns were by f a r the most important summer habitat, and these had not been a l t e r e d by development. I f these important habitats had been c a r e l e s s l y modified by the various i n d u s t r i e s , i t i s probable that negative population responses would have been evident. There was l i t t l e evidence that the disturbance and displacement caused by the l e v e l and types of human a c t i v i t i e s that occurred i n the study area had a s i g n i f i c a n t impact on the g r i z z l y population. The habitat loss equivalence due to the high road density was l i k e l y o f f s e t by increased nocturnal use of areas near roads and d i f f e r e n t i a l use of these areas by age-sex c l a s s . The dense cover i n most of the study area and the a b i l i t y of bears to habituate to human a c t i v i t i e s l i k e l y reduced the impact of p o t e n t i a l human induced disturbances (McLellan and Shackleton i n press b). I had no evidence that i n d u s t r i a l personnel k i l l e d any g r i z z l y bears during the period of study. In a study such as t h i s , there i s a p o t e n t i a l bias due to i n d u s t r i a l personnel keeping t h e i r camps cleaner than normal because I was studying the bears i n the area. However, I doubt that they did modify t h e i r behaviour because my base was 10 km away from the c l o s e s t i n d u s t r i a l camp and I never "inspected" a camp for proper garbage management. Nor did I notice differences between camps i n my study area and one located outside of i t . The construction of roads was the most notable impact of the resource industries i n the Flathead study area because roads permitted easy access f o r hunters, poachers and s e t t l e r s . In my study area, 8 of the 9 deaths of bears with functioning radio c o l l a r s were caused by people and 5 of these were i l l e g a l . Most of the recorded i l l e g a l k i l l i n g s of marked and unmarked bears occurred i n 4% of the study area with human settlement. Most bears were shot, both l e g a l l y and i l l e g a l l y , from roads. Fortunately, the high reproductive rate of the Flathead g r i z z l y bears appears to have compensated thus f a r fo r t h i s l e v e l of mortality. On the East Front of the Rockies i n Montana, c a t t l e ranching, human settlement, and o i l and gas exploration occur. There, 30 of 32 recorded g r i z z l y bear m o r t a l i t i e s were due to people; 10 by l e g a l hunting and 20 for other reasons (Aune and Brannon 1987). As i n my study area, the g r i z z l i e s on the East Front also have a very high reproductive rate (Aune et a l . 1986), and the population trend may be stable (K. Aune, pers. comm.). In other areas, t h i s may not be the case. In 2 areas of western Alberta, where i n d u s t r i a l roads provide easy access, the number of g r i z z l y bears being k i l l e d by people appears to be causing population declines. Horejsi (1986) recorded 5 l e g a l hunter k i l l s and 3 i l l e g a l k i l l s of 35 marked bears i n 4 years, r e s u l t i n g i n an annual mortality rate of 6%. Ho r e j s i (1986) reported that 4 of 9 marked adult females were k i l l e d . Although he did not present any reproductive information, he suggested that the population was d e c l i n i n g . In the second well accessed study area i n Alberta, 11 hunter k i l l s , 5 research r e l a t e d deaths and 2 natural m o r t a l i t i e s from a t o t a l of 38 marked bears i n 5 years was thought to be causing a decrease i n a g r i z z l y bear population with a low reproductive rate (Nagy et a l . 1988). I f w i l d l i f e managers consider the mortality rate of g r i z z l i e s to be excessive, and the user groups (hunters) can be convinced of t h i s , i t i s r e l a t i v e l y easy to reduce the number of l e g a l k i l l s . However, once an area i s accessible and s e t t l e d to the degree where i l l e g a l and c o n t r o l k i l l i n g alone are excessive, reducing the number of human induced m o r t a l i t i e s becomes extremely d i f f i c u l t and expensive. This i s one of the problems facing the g r i z z l y bears i n Yellowstone National Park and adjacent areas (Knight and Eberhardt 1984). The reduction i n g r i z z l y bear numbers and the eventual elimination of the species appears to follow a consistent pattern i n the i n t e r i o r of North America, with i n d u s t r i a l a c t i v i t y playing a very important r o l e . Wilderness i s f i r s t explored by hunters, trappers, foresters and geologists. Then industry b u i l d s roads to remove resources, and these roads, i n turn, make i t much easier f o r hunters and poachers. G r i z z l y numbers may begin to decline at t h i s stage i f hunting seasons are too l i b e r a l f o r the p r o d u c t i v i t y of the area. Also, new access i s often followed by ranchers and s e t t l e r s , and soon the g r i z z l i e s are gone. Stopping resource development to maintain g r i z z l y populations (and other wilderness values) i s a v i a b l e option 106 f o r some r e l a t i v e l y small areas. However, due to the economic value and continued demand f o r natural resources, i n d u s t r i e s are destined to extract resources from most of the remaining, "unprotected" g r i z z l y bear habitat. A l t e r n a t i v e l y , maintaining g r i z z l i e s requires having, 1) conservative hunting regulations based on the area's p r o d u c t i v i t y , 2) stringent regulations f o r habitat protection and f o r camp and personnel management, 3) access plans to ensureroads are closed a f t e r developments are complete, and 4) keeping human habitations from being scattered i n many drainages. Regulations such as these w i l l l i k e l y be unpopular with some administrators and c i t i z e n s who want to continue to " s e t t l e the west". Land and w i l d l i f e managers, and user groups must co-operate with the resource i n d u s t r i e s to educate the p u b l i c and advertise the value of long term co-ordinated land use plans. In t h i s way i t may be possible to extract natural resources from an area while sustaining v i a b l e g r i z z l y bear populations, and maintain the option of regaining wilderness. The Flathead study should be viewed as one, r e l a t i v e l y short term, experimental u n i t . U n t i l r e p l i c a t e s of t h i s study are conducted, t e s t i n g both population and behavioural responses of g r i z z l y bears to i n d u s t r i a l a c t i v i t i e s i n d i f f e r e n t regions, i t i s d i f f i c u l t to extrapolate r e s u l t s with confidence. The c h a r a c t e r i s t i c s of the Flathead Drainage that made industry and g r i z z l i e s compatible during the short duration of my study may have l i m i t e d d i s t r i b u t i o n . Research i s continuing i n my study area to further evaluate long term impacts of resource i n d u s t r i e s . In addition, the research i s being expanded to include the influence of ungulates discussed e a r l i e r , and p o t e n t i a l i n t e r s p e c i f i c competition between black and g r i z z l y bears. LITERATURE CITED Alldredge J.R. and R a t t i , J.T. 1986. Comparison of some s t a t i s t i c a l techniques f o r analysis of resource s e l e c t i o n . J . W i l d l . Manage. 50:157-165. Almack, J.A. 1986. G r i z z l y bear habitat use, food habits, and movements i n the S e l k i r k Mountains, northern Idaho. Pages 150-157 i n Contreras, G.P. and Evans, K.E. (eds). Proceedings: grisrzly bear habitat symposium. Gen. Tech. Rep. INT-207. USDA For. Serv., Intermountain Res. Sta., Ogden, Utah. Altmann, J . 1974. Observational study of behaviour: sampling methods. Behaviour 49:227-267. Archibald, W.R., E l l i s , R. and Hamilton, A.N. 1987. Responses of g r i z z l y bears to logging truck t r a f f i c i n the Kimsquit River Valley, B r i t i s h Columbia. Int. Conf. Bear Res. Manage. 7:251-257. Aune, K. and Brannon, B. 1987. East Front G r i z z l y Studies. Montana Dep. Fish, W i l d l . Parks, Helena, MT. 195pp. Aune, K., Madel, M. and Hunt, C. 1986. Rocky Mountain front g r i z z l y bear monitoring and inves t i g a t i o n . Montana Dep. Fi s h , W i l d l . Parks, Helena. MT. 175pp Ba l l a r d , W.B. 1987. Demography of Noatak g r i z z l y bears i n r e l a t i o n to human ex p l o i t a t i o n and mining development. Alaska Dep. Fish and Game, Fed. Aid i n W i l d l . Rest. Prog. Rep. Pro j . W-22-5 and W-22-6. Jobs 4.20R. Juneau. AK Bunnell, F.L., and T a i t , D.E.N. 1980. Bears i n models and i n r e a l i t y - implications to management. Int. Conf. Bear Res. Manage. 4:15-23. Bunnell, F.L. and T a i t , D.E.N. 1981. population dynamics of bears - implications. Pages 75-98 i n Fowler, CW. and Smith, T.D. (eds). Dynamics of large mammal populations. John Wiley and Sons, New York, New York. Bunnell, F.L. and T a i t , D.E.N. 1985. M o r t a l i t y rates of North American bears. A r c t i c 38:316-323. Caughley, G. 1966. Mo r t a l i t y patterns i n mammals. Ecology 47:906-918. Caughley, G. 1977. Analysis of vertebrate populations. John Wiley and Sons, New York, New York. 232pp. Cowan, I. McT. 1972. The status and conservation of bears (Ursidae) of the world - 1970. Int. Conf. Bear Res. Manage. 2:343-367. Craighead, J . J . 1976. Studying g r i z z l y habitat by s a t e l l i t e . National Geographic 150:148-158. Craighead, J . J . , Varney, J.R. and Craighead, F.C J r . 1974. A population analysis of the Yellowstone g r i z z l y bears. Montana For. and Cons. Exp. Stn. B u l l . 40. Univ. Montana, Missoula, MT. 20pp. Christensen, A. and Madel, M. 1982. Cumulative e f f e c t s analysis process and g r i z z l y habitat component mapping. Unpublished ms. Kootenai National Forest. Libby MT. Glenn, L.P., Lentfer, J.W., Faro, J.B. and M i l l e r , L.H. 1976. Reproductive biology of female brown bears, (Ursus arctos), McNeil River, Alaska. Int. Conf. Bear Res. Manage. 3:381-390. Hamer, J.D.W. 1974. D i s t r i b u t i o n , abundance and management implications of the g r i z z l y bear and mountain caribou i n the Mountain Creek watershed of G l a c i e r National Park, B r i t i s h Columbia. M.Sc. th e s i s , Univ. Calgary. 164pp. Hamer, D. and Hererro, S. 1983. Ec o l o g i c a l studies of the g r i z z l y bear i n Banff National Park. F i n a l Rep. prep, for Parks Canada. Univ. Calgary, Alberta. 303pp. Hamer, D., Hererro, S. and Brady, K. 1983. E c o l o g i c a l studies of the g r i z z l y bear i n Watert onal Park, a progress report for 1982. Parks Canada Contract WR 70-82. Waterton Park, Alberta. 59pp. Harestad, A.S. 1981. Computer analysis of home range data. B.C. Mi n i s t r y of Environment., F i s h and W i l d l . B u l l . No. B - l l . 25pp. Harper, F.E.1988. W i l d l i f e management plan: Peace subregion. B. C. Ministry of Environment., W i l d l i f e Branch, V i c t o r i a . B.C. 212 pp. Heisey, D.M., and F u l l e r , T.K. 1985. Evaluation of s u r v i v a l and cause-specific mortality rates using telemetry data. J . W i l d l . Manage. 49:668-674. Horejsi, B.L. 1986. I n d u s t r i a l and a g r i c u l t u r e incursion into g r i z z l y bear habitat: the Alberta story. Pages 116-123 i n : Contreras, G.P. and Evans, K.E. (eds). Proceedings: g r i z z l y bear habitat symposium. Gen. Tech. Rep. INT-207. U.S.D.A. For. Serv., Intermountain Res. Sta., Ogden, Utah. 252pp. Jacobsen, R.D. 1980. Legal aspects of c r i t i c a l habitat I l l determinations. Int. Conf. Bear Res. Manage. 4:5-8. Johnson, D.H. 1980. The comparison of usage and a v a i l a b i l i t y measurements f o r evaluating resource preference. Ecology 61:65-71. Kemp, G.A. 1972. Black bear population dynamics at Cold Lake, Alberta, 1968-70. Int. Conf. Bear Res. Manage. 2:26-31. Kemp, G.A. 1976. The dynamics and regulation of black bear (Ursus' enaerxcarrus) populations i n northern Alberta. Int. Conf. Bear Res. Manage. 3:191-197. Kendall, K.C. 1985. G r i z z l y bear population trend studies, Apgar Mountains, G l a c i e r National Park. National Park Service Progress Report f o r 1984. G l a c i e r National Park, MT. 2 3pp. K i s t c h i n s k i , A.A. 1972. L i f e h i s t o r y of the brown bear (Ursus arctos) i n northeastern S i b e r i a . Int. Conf. Bear Res. Manage. 2:67-73. Knight, R.R. 1980. B i o l o g i c a l considerations i n the del i n e a t i o n of c r i t i c a l habitat. Int. Conf. Bear Res. Manage. 4:1-3. Knight, R.R. and Eberhardt, L.L. 1984. Projected future abundance of the Yellowstone g r i z z l y bear. J . Wi l d l . Manage. 48:1434-1438. Knight, R.R. and Eberhardt, L.L. 1985. Population dynamics of Yellowstone g r i z z l y bears. Ecology 66:324-334. ' LeFranc, J r . , M.N., Moss, M.B., Patnode, K.A. and Sugg, III W.C. (eds). 1987. G r i z z l y Bear Compendium. Interagency G r i z z l y Bear Committee, Washington, D.C. 540pp. MacArthur, R.A., Geist, V., and Johnson, R.H. 1982. Cardiac and behavioral responses of mountain sheep to human disturbance. J . W i l d l . Manage. 46:351-358. Mace, R.D., and Jonkel, C. 1980. The e f f e c t s of logging a c t i v i t y on g r i z z l y bear movements. Border G r i z z l y Proj., Univ. Montana., Missoula. Spec. Rep. No. 38. l l p p . Marcum, C.L. and Loftsgaarden, D.O. 1980. A nonmapping, technique f o r studying habitat preferences. J . W i l d l . Manage. 44:963-968. Martinka, C.J. 1974. Population c h a r a c t e r i s t i c s of g r i z z l y bears i n Glacier National Park, Montana. J . Mammal. 55:21-29. Mattson, D.J., Knight, R.R. and Blanchard, B.M. 1987. The e f f e c t s of developments and primary roads on g r i z z l y bear habitat use i n Yellowstone National Park, Wyoming. Int. Conf. Bear Res. Manage. 7:259-273. McCourt, K.H., F e i s t , J.D., D o l l , D., and R u s s e l l , J . J . 1974. Disturbance studies of caribou and other mammals i n the Yukon and Alaska, 1972. Renewable Resources Consulting Services Ltd., Canadian A r c t i c Gas Study Ltd., B i o l . Rep. Ser. 5. 245pp. McCullough, D.R. 1981. Population dynamics of the Yellowstone g r i z z l y . Pages 173-196 i n Fowler, CW. and Smith, T.D. (eds). Dynamics of large mammal populations.. John Wiley and Sons, New York, New York.. McLellan, B.N. 1986. U s e - a v a i l a b i l i t y analysis and timber s e l e c t i o n by g r i z z l y bears. Pages 163-166 i n Contreras, G.P. and Evans, K.E. (eds). Proceedings: g r i z z l y bear habitat symposium. Gen. Tech. Rep. INT-207. USDA For. Serv., Intermountain Res. Sta., Ogden, Utah. McLellan, B.N. (in press a.) Population dynamics of g r i z z l y bears during a period of resource ex t r a c t i o n development. I. Density and age/sex structure. Can. J . Zool. Accepted February 1989. McLellan, B.N. (in press b) . Population dynamics of grizzly* bears during a period of resource ex t r a c t i o n development. I I . Morality rates and causes. Can. J . Zool. Accepted February 1989. McLellan, B.N. (in press c ) . Population dynamics of g r i z z l y bears during a period of resource extraction development. I I I . N a t a l i t y and rate of change. Can. J . Zool. Accepted February 1989. McLellan, B.N. and Shackleton, D. M. 1988 a. G r i z z l y bears and resource extraction i n d u s t r i e s : e f f e c t s of roads on behaviour, habitat use and demography. J . Appl. E c o l . 25:451-460. McLellan, B.N. and Shackleton, D.M. 1988 b. A comparison of g r i z z l y bear harvest data from Montana and southeastern B r i t i s h Columbia. W i l d l . Soc. B u l l . 16:371-375. McLellan, B.N. and Shackleton, D.M. (in press a). G r i z z l y bears and resource extraction i n d u s t r i e s : habitat displacement i n response to seismic exploration, timber harvesting and road maintenance. J . Appl. E c o l . Accepted March 1989. McLellan, B.N. and Shackleton, D.M. ( i n press b). Immediate reactions of g r i z z l y bears to human a c t i v i t i e s . W i l d l . Soc. B u l l . Accepted June 1989. Mech, L.D. 1966. The Wolves of I s l e Royale. Fauna of the Nat. Parks of the U.S. Faunal Series No. 7, U.S. Gov. P r i n t . Off., Washington D.C. 210pp. Mohr, CO. 1947. Tables of equivalent populations of North American small mammals. Am. Midi. Nat. 37:223-249. Mundy, K.R.D., and Flock. D.R. 1973. Background f o r managing g r i z z l y bears i n the National Parks of Caiiarla. Can. W i l d l . Serv. Rep. Ser. No. 22. 35pp. Nagy, J.A., Hawley, A.W.L., Barrett, M.W. and Nolan, J.W. 1988. Population c h a r a c t e r i s t i c s of g r i z z l y and black bears i n west c e n t r a l Alberta. Alberta Environmental Centre, V e g r e v i l l e , A l t a . 34pp. Neu, C.W., Byers, CR. and Peek. J.M. 1974. Techniques f o r analysis of u t i l i z a t i o n - a v a i l a b i l i t y data. J . W i l d l . Manage. 38:541-545 Pearson, A.M. 1975. The northern i n t e r i o r g r i z z l y bear. Can. W i l d l . Serv. Rep. Ser. No. 34. 86pp. Quimbey, R. 1974. G r i z z l y bear. Pages 1-85 i n Jakimchuk, R.D. (ed). Mammal studies i n northeastern Alaska with emphasis within the Canning River drainage. Can. A r c t i c Gas Study Ltd., B i o l . Rep. Ser. 24. Reynolds, H.V. and Hechtel, J . 1979. Structure, status, reproductive biology, movement, d i s t r i b u t i o n , and habitat u t i l i z a t i o n of a g r i z z l y bear population. Alaska Dep. F i s h and Game, Fed. Aid i n W i l d l . Rest. Prog. Rep. Proj. W-17-11. Juneau. Reynolds, P.E., Reynolds, H.V. and Follmann, E.H. 1986. Responses of g r i z z l y bears to seismic surveys i n northern Alaska. Int. Conf. Bear Res. Manage. 6:169-175. Russell, R.H., Nolan, J.W, Woody, N.G. and Anderson, G.H. 1979. A study of the g r i z z l y bear i n Jasper National Park, 1975 to 1978. Can. W i l d l . Serv. Edmonton, Alberta. 136pp. Schallenberger, A. 1980. Review of o i l and gas e x p l o i t a t i o n impacts on g r i z z l y bears,, Int. Conf. Bear Res. Manage. 4:271-276. Schoen, J.W., Lentfer, J.W. and Beier, L. 1986. D i f f e r e n t i a l d i s t r i b u t i o n of brown bears on Admiralty Island, southeast Alaska: a preliminary assessment. Int. Conf. Bear Res. Manage. 6:1-5. Servheen, C. 1981. G r i z z l y bear ecology and management i n the Mission Mountains, Montana. Ph.D. t h e s i s , Univ. Montana, Missoula. MT 139pp. Servheen, C. 1983. G r i z z l y bear food habitats, movements, and habitat s e l e c t i o n i n the Mission Mountains, Montana. J . W i l d l . Manage. 47:1026-1035. Shank, C.C. 1979. Human-related behavioural disturbance to northern large mammals: a bibliography and review. F o o t h i l l s Pipelines (South Yukon) Ltd. Calgary. Simpson, K., Hebert, K.B. and Woods, G.P. 1985. Habitats and management of the g r i z z l y bear (Ursus arctos) i n the Columbia Mountains of B r i t i s h Columbia. B.C. Ministry 116 of Environment, Nelson, B.C. 47 pp. Sizemore, D.L. 1980. Foraging strategies of the g r i z z l y bear as r e l a t e d to i t s e c o l o g i c a l energetics. M.Sc. Thesis. Univ. Montana, Missoula. 67pp. Sokal, R.R. and Rohlf, J.F. (1981). Biometry. 2nd e d i t i o n . W.H. Freeman and Co., San Francisco. 859pp. Stemp, G.E. 1983. Heart rate responses of bighorn sheep to environmental factors and harassment. M.Sc. Thesis, Univ. Calgary, Calgary. 314pp. Stringham, S.F. 1983. Roles of adult males i n g r i z z l y bear population biology. Int. Conf. Bear Res. Manage. 5:140-151. T i e t j e , W.D. and Ruff, R.L. 1983. Responses of black bears to o i l development i n Alberta. Wildl..Soc. B u l l . 11:99-112. Troyer, W.A. and Hensel, R.J. 1962. Cannibalism i n brown bear. Anim. Behav. 10:231. Troyer, W.A. and Hensel, R.J. 1964. Structure and d i s t r i b u t i o n of a kodiak bear population. J . W i l d l . Manage. 29:769-772. Zager, P.E. 1980. The influence of logging and w i l d f i r e on g r i z z l y bear habitat i n northwestern Montana. Ph.D. t h e s i s , Univ. Montana, Missoula. 131pp. Zager, P.E., Jonkel, C. and Habeck, J . 1983. Logging and w i l d f i r e influence on g r i z z l y bear habitat i n northwestern Montana. Int. Conf. Bear Res. Manage. 5:271-276. Zunino, F., and Herrero, S. 1972. The status of the brown bear (Ursus arctos) i n Abruzzo National Park, I t a l y , 1971. B i o l . Conserv. 4:263-272.