1997 Annual Report

Gibeau, Michael L. and Stephen Herrero. 1998. Eastern slopes grizzly bear project: A progress report for 1997. Eastern Slopes Grizzly Bear Project, University of Calgary, Calgary, Alberta.
Note: The full text of this paper is displayed below. You also have the option of downloading a PDF version.
Prepared for the Eastern Slopes Grizzly Bear Steering Committee
This paper contains preliminary results of an on-going study and should not be cited without permission from the authors.
TABLE OF CONTENTS
INTRODUCTION
Project Background
Project Origins
Project Organization and Budget
ACKNOWLEDGMENTS
GOALS AND OBJECTIVES
STUDY AREA
METHODS
RESULTS
Population Studies
Capture
Telemetry Data Set
Population Demographics
Population Estimate
Critique
LITERATURE CITED
LIST OF TABLES
Table 1. Grizzly bear capture data in the Bow River Watershed, Alberta, 1997
Table 2. Grizzly bear trapping success in the Bow River Watershed, Alberta, 1997.
Table 3. Status of all grizzly bears captured in the Bow River Watershed, Alberta, as of December 1997.
Table 4. Unduplicated grizzly bear females with cubs of the year in the Bow River Watershed, Alberta, 1993 through 1997.
Table 5. Reproductive status of known female grizzly bears in the Bow River Watershed, Alberta, 1997.
Table 6. Summary of grizzly bear mortalities in the Bow River Watershed, Alberta, 1993 through 1997.
Table 7. Summary of grizzly bear translocations in the Bow River Watershed, Alberta, 1993 through 1997.
INTRODUCTION
Project Background
Several hundred years ago grizzly bears lived throughout much of what is now the province of Alberta. Today a historically estimated population of approximately 6000 individuals (Herrero unpublished data), has decreased to an estimated 600 grizzly bears on provincial land and another 200 within federal national parks (Nagy and Gunson 1990). Grizzly bears have declined in Alberta because of mortality in excess of recruitment and because people have occupied and developed land which once supported the bears and less industrialized people.
Grizzly bears are recognized in Alberta as one of the principle species that indicates wilderness –large scale landscapes in a relatively natural state, the raw material out of which our culture was, and is still being created. Now, however, the last remaining unprotected wildland areas in Alberta are being modified by industrial and recreational activity. Because Albertans value nature and wildlife in addition to economic development, there is an urgent need to understand the impacts of human-caused mortality and land use on grizzly bears, and to target mortality rates and habitat protection and management that will allow for grizzly bear persistence. This direction is supported by the Grizzly Bear Management Plan of Alberta which states that the provincial population will be increased to 1000 (Nagy and Gunson 1990). It is also consistent with National Park management objectives for ecological integrity as set by the National Parks Act and Policy (1988).
On the Eastern Slopes in Alberta grizzly bears occur at relatively low population densities, only one bear for each 60-100 km2. Male grizzlies have lifetime home ranges of approximately 1000-2000 km 2 (Russell et al. 1979, Carr 1989). Females do not begin breeding until they are 4-7 years old and then they produce significantly less than one cub per year. Because of these biological characteristics grizzly bears recover slowly if at all from population declines, and only if negative mortality factors have been brought under control (Mattson et al. 1996). These and other biological characteristics are part of the reason why human activities can have such a significant impact on grizzly bears.
Alberta has an expanding economy based significantly on the development of natural resources such as agriculture, oil and gas, forestry, and nature-based tourism. Individual grizzly bears, owing to their large home ranges, may come into contact with all of these activities. Research based in Yoho and Kootenay national parks showed that individual grizzly bears may enter four different management jurisdictions in a year (Raine and Riddell 1991). Whether land is managed as parks, commercial forests, or privately, management practices must respond to the grizzlies needs if these bears are to survive. There is an urgent need for scientific data on grizzly bears to help land managers better understand the affects of human activities on grizzly bears.
Project Origins (this section is quoted extensively from Herrero et al. (In Press)
The Eastern Slopes Grizzly Bear Project (ESGBP) formally began in May 1994. Neither the project, nor its membership, were formally designated by any group or agency. The Project and its members evolved from a number of different origins. An understanding of these helps in defining the nature of the Project.
First, were changes in legislation and policy at both the Federal and the Provincial levels. In 1988 the Government of Canada amended the National Parks Act. Changes included a recognition that ecological integrity was the primary objective of national park management. In this context, the grizzly bear is recognized as one of the most sensitive ecosystem elements, meaning they are difficult to maintain in landscapes that have a lot of human activities. Where grizzly bears exist, they are an indicator of ecological integrity. Parks Canada thus had new reason to be concerned about the status of grizzly bears, especially in national parks such as Banff which is part of one of the most developed landscapes where grizzly bears still survive. This legislative change was reflected in a re-written Parks Canada policy document that recognized the need for multi-agency approaches to parks management. Again, the grizzly bear with its wide-ranging movements across jurisdictional borders, became a focal species in trying to address multi-agency dimensions of parks management.
In 1992 the Federal government enacted the Canadain Environmental Assessment Act (CEAA) which broadened the scope of traditional environmental assessment to consider the cumulative effects of developments at a landscape scale. The following year (1993) the Alberta Environmental Protection and Enhancement Act (EPEA) passed which also included a provision for assessing the cumulative impacts of development proposals. The need to consider cumulative effects in evaluating development proposals has been highlighted in the review of several major project proposals for the Eastern Slopes of Alberta: the Energy Resources Conservation Board (ERCB)(now the Energy and Utilities Board – EUB) highlighted the need for cumulative effects assessment (CEA) in its review of AMOCO’s proposal to drill an exploratory well in the Whaleback region (ERCB 1994); the Natural Resources Conservation Board (NRCB) which was established to function similarly to the ERCB, but with regard to large, proposed recreational developments, indicated the need for CEA in its review of the Three Sisters Resort Proposal and the Westcastle Resort Proposal (NRCB 1993a, 1993b). In all these reviews grizzly bears, because of their regional movements and ecological relationships, and because of their sensitivity to development, became a focal species for cumulative effects assessment.
The second major element in the origin of the ESGBP was new information regarding the status of grizzly bears in the Canadian Rocky Mountains and elsewhere in Alberta. In 1990 the province of Alberta released its grizzly bear management plan (Nagy and Gunson 1990). This document clearly showed not only historic declines of grizzly bears in the province, but major over hunting, especially during 1980-1988. This launched the province into a limited entry system for managing hunting. It revealed how subject grizzly bear populations are to excessive mortality, not just from hunting but from all sources. This documented excessive mortality, combined with rapid expansion in resource harvesting activities in the province, was important in raising Alberta Fish and Wildlife’s concern for grizzly bears.
In the national parks new information also clearly documented the need for interagency management of grizzly bears. Research had shown that grizzly bears in the Canadian Rocky Mountain National Parks moved freely and extensively across park borders and that mortality outside of park borders was a significant issue (Russell et al. 1979, Raine and Riddell 1991). Herrero (1995) showed that Canadian National Park grizzly bear populations by themselves were probably all too small for a high probability of long term persistence, and therefore integrated management with surrounding provincial or territorial lands would be required. Within the boundaries of Banff , Yoho and Kootenay National Parks research by Gibeau (In press) showed that habitat effectiveness was significantly compromised by development. More recent research documents that grizzly bear populations in Banff Park have suffered exceptionally high mortality for a national park (Gibeau et al. 1996).
The third factor that led to formation of the ESGBP was growing awareness of the discipline of conservation biology. This is a discipline with the objective of using scientific information to help maintain biological diversity. Many of the principles of conservation biology focus on the design of systems of environmental reserves along ecological boundaries that most often cross jurisdictional divisions (Noss and Cooperrider 1994). Within conservation biology large-bodied mammalian carnivores such as the grizzly bear are often used as indicator and umbrella species (see August 1996 issue of the journal Conservation Biology). By maintaining the large carnivores we will also maintain a significant degree of terrestrial regional ecological integrity.
The ESGBP was a product of the foregoing series of societal level influences plus many others that have not been mentioned. Like most projects this one responded to a need perceived by many different individuals and institutions, sometimes for different reasons. By joining in a cooperative endeavor and pooling resources a major project was launched.
Project Organization and Budget
The ESBGP is an informal association of participants organized into a steering committee whose objectives are to: 1) review and suggest strategic direction for research and encourage a research-based understanding of grizzly bear biology and ecology in selected portions of the Eastern Slopes of the Rocky Mountains in Alberta, 2) help focus research efforts on the cumulative effects of regional land use and mortality factors on grizzly bears, 3) provide a forum for various stakeholders to discuss land use planning issues as they relate to grizzly bears, 4) help secure funding and other forms of agency support, 5) coordinate public outreach initiatives and 6) contribute to the conservation of grizzly bear populations and their habitat in the Eastern Slopes.
All steering committee participants contribute either money, time or both toward the objectives. The group, meets about 4 times a year. It has a chair who was elected from a core organizing group. Membership currently consists of a selection of representatives from various groups that have either jurisdiction, resource harvest activities or potential, or other interests regarding occupied grizzly bear habitat in the Eastern Slopes of Alberta’s Rocky Mountains. The principal participants are Parks Canada, the Province of Alberta (Energy and Utilities Board, Fish and Wildlife Division, Lands and Forest Service, and Kananaskis Country), the University of Calgary, conservation groups, the oil and gas industry, the forest products industry, the land development industry and the cattle industry. There are numerous minor supporters as well, but most do not have direct representation on the Steering Committee.
During meetings research findings and strategic directions are discussed along with budget needs to further the committees objectives. The group serves as a focal point for fund raising activities to support the Project. Significant development proposals and activities are discussed in light of their potential cumulative effects regarding grizzly bears and their habitat.
During the period of 1994-1997 the ESGBP was been successful in raising over $1,450,000 to support the research. Sources for this funding have been: Parks Canada 46%, oil and gas industry 34%, Alberta Government 11%, other research grants 4%, forest industry 3%, conservation groups 1%, and land development industry (Herrero and Herrero 1996). Contributions to this project are tax deductible because they go to support independent research by the University of Calgary.
ACKNOWLEDGMENTS
A very successful forth field season would not have been possible without the dedication of field biologists B. Benn, M. Jalkotzy, C. Mamo, C. Mueller, J. Paczkowski, I. Ross, J. Saher, S. Stevens, S. Stotyn, and M. Urquhart. Their efforts were augmented through the largely volunteer support of C. Campbell, P. Hoffer, and M. Morrow. Assistance in coordination of field staff was provided by A. Dibb, S. Donelon and T. Hurd. Trapping was conducted by R. Leblanc, C. Mamo, and I. Ross. Veterinary care was provided by Dr. Todd Shury. Several Alberta Fish and Wildlife Officers, Banff National Park Wardens and Peter Lougheed Park Rangers all provided invaluable safety backup and field assistance during trapping. The Banff Park Warden Service and Kananaskis Country Park Rangers provided logistical support through all stages of monitoring. Exemplary flying skills were provided by Alpine Helicopters of Canmore and fixed wing pilot M. Dupuis of Westpoint Aviation.
The Eastern Slopes Grizzly Bear Steering Committee helped implement and guided this research. All steering committee participants contribute either money, time or both toward the objectives. Through the Steering Committee, governments, industry, business and conservation groups work together to support this project. The supporters include:
Alberta Conservation Assoc.
Alberta Cattle Commission
Alberta Energy Utilities Board
Alberta Fish and Wildlife Service
Alberta Kananaskis Country
Alberta Lands and Forest Service
Alberta Provincial Parks
Alpine Helicopters
AMOCO Canada Petroleum Co. Ltd.
Bow Valley Naturalists
Calgary Zoological Society
Canadian Assoc. of Petroleum Producers
Canadian Parks & Wilderness Soc.
Canmore Collegiate High School
Human Resources Canada
Husky Oil
Parks Canada, BNP & AB Region
Shell Canada Ltd.
Sking Louise Ltd.
Spray Lakes Sawmills
Springbank Middle School
Three Sisters Resorts Inc.
University Of Calgary
Warner Guiding and Outfitting Ltd.
Wilderness Medical Society
World Wildlife Fund Canada
GOALS AND OBJECTIVES
Prior to the ESGBP, little research has focused on the effects of non-motorized tourism oriented activities on bears. One of our principle research questions is how do grizzly bear’s spatial and temporal use patterns differ in areas of high human presence compared to areas with low human presence in a landscape, some of which is dominated by tourism activities? Our situation is unique in that few other grizzly bear study areas in North America have both a high volume transcontentinal highway and railway bisecting occupied grizzly bear habitat along with intensive tourism. Analysis has never been done on the effects of such levels of human presence on grizzly bears. One important question is the extent to which the Bow River Valley continues to function as a major movement corridor for bears providing connectivity between habitats. Unregulated human access and development within bear habitat can contribute to increased bear mortality and affect grizzly bear use of existing habitat.
The overall goal of this research is to understand how developments and human-induced mortality impact grizzly bears. Specific research objectives include:
1.Determine the basic demographic parameters for the grizzly bear population within the study area.
2. Detect spatial and temporal activity patterns of bears given various levels of human influences.
3. Determine how the distribution of human influences affects a bear’s ability to use the landscape.
4. Determine if population connectivity is being impeded by major transportation corridors.
5. Determine what, if any adjustments to human activities would give bears better access to resources.
6. Suggest management alternatives for integrating land uses compatible with bear habitat needs for the study area.
STUDY AREA
The area of interest remains unchanged from year 1 with the Bow River Watershed, from its headwaters to approximately where it meets the prairies, as the core study area. The Bow River drainage system is approximately 11,400 km2. The greater study area defined by the movement of radio-collared bears is about 22,000 km2 or roughly twice the size of the core study area.
METHODS
Methods for both the capture and monitoring of bears remain unchanged from the detailed description found in the year 1 progress report (Gibeau and Herrero 1995). Approximately 25 grizzly bears per year have active radio-collars. These bears are monitored from air and ground wherever they go and our budget permits. Aerial monitoring gives infrequent, but relatively unbiased data regarding location. This facilitates understanding of home range, movements and habitat use. Ground-based research allows intensive monitoring of grizzly bear activities related to development features such as towns, highways, campgrounds and trails. Mortality is monitored using both aerial and ground-based telemetry, and by accessing and creating broader mortality data bases related to hunting and other human-induced mortality sources. The radio-telemetry monitoring area includes lands under several different jurisdictions. In the British Columbia portion of these lands, where some of our radio-collared grizzly bears are found, there is a Western Slopes Bear Research Project (Woods pers. comm.) which provides complementary data and will allow a broader ecosystem versus provincial boundary-based understanding of grizzly bears in what has been called the Central Rockies Ecosystem (Komex International 1995).
RESULTS
Population Studies
Capture
In 1997 emphasis was placed on recapturing bears whose transmitters were due for replacement. We recaptured six previously captured bears as well as 12 new bears (Table 1). Trap success was high averaging 7.2 site nights per capture (Table 2).
Since the Eastern Slopes Grizzly Bear Project began in May of 1994, we have captured 25 male and 28 female grizzly bears. Currently, 26 bears are radio-instrumented including 8 males, 18 females (Table 3). Of the remainder, 10 bears have died, 14 have dropped collars or stopped transmitting, 1 was removed from the system, and 2 were never collared initially.
Table 1. Grizzly bear capture data in the Bow River Watershed, Alberta, 1997.

ID	Sex	Age Estimate	Weight (kg)	Area	Comments
10	M	16a*	170 est.	Bow Valley	recapture
13	M	8a	122.	Kananaskis	recapture
15	M	9a	150 est.	Cascade R.	recapture
30	F	12a	80	Lake Louise	recapture
35	F	5a	86	Kananaskis	recapture
42	M	9a	116	Spray R.	recapture
51	M	8a	147	Baker Cr.
52	M	7b	135 est.	Highwood R.
53	M	3a	80	Highwood R.
54	M	15a	190	Banff
55	F	6a	98	Cascade R.
56	F	3a	44	Lake Louise	cub of Bear #30
57	F	5a	130 est.	Highwood R.
58	M	9a	190 est.	Cascade R.
59	F	3a	39	Lake Louise	cub of Bear #30
60	F	3a	46	Lake Louise	cub of Bear #30
61	F	12a	103	Spray R.
62	F	8a	103	Cascade R.

* certainty code a= +/- 0 years, b= +/- 1-2 years, c= +/- 2-3 years
Table 2. Grizzly bear trapping success in the Bow River Watershed, Alberta, 1997.

Area	Total Site Nights	Total Grizzly	Total Black	Total Captures	Site Nights / Capture
Mid Bow R.	15	1	0	1	15.0
Kananaskis	65	2	5	7	9.3
Spray R.	49	4	4	8	6.1
Cascade R.	13	4	0	4	3.3
Lake Louise	25	5	1	6	4.2
Highwood R.	22	3	1	4	5.5
	189	18a	11	29	7.2

ID

Sex

Age at 1st capture

Date 1st captured

Current status

No of radio

relocations

10

M

13 a*

05/07/94

active

290

11

M

4 b

05/17/94

unknown – drop collar 07/97

49

12

M

13 b

05/19/94

dead – 10/04/94

14

13

M

5 a

05/25/94

active

363

14

M

9 a

05/29/94

active – no signal 05/97

49

15

M

6 a

05/20/94

active

225

16

M

5 a

08/16/93

removed to zoo 07/05/96

169

17

F

10 a

06/02/94

unknown – drop collar 07/96

103

18

F

6 a

05/30/94

active

84

19

M

6 b

05/13/94

dead – 05/14/94

1

20

M

11 a

05/14/94

unknown – drop collar 08/94

8

21

M

3 a

05/21/94

dead – 07/26/95

3

22

M

14 a

05/21/94

dead – 05/28/94

2

23

M

3 a

05/28/94

dead – 08/08/96

75

24

F

5 a

05/31/94

active

373

25

M

6 a

05/31/94

unknown – drop collar 09/94

15

26

F

18 a

06/08/94

active

345

27

F

2 a

06/13/94

unknown – no signal 04/96

35

28

F

22 a

06/08/94

dead – 08/24/96

72

29

M

2 a

06/13/94

unknown – never collared

1

30

F

9 a

09/28/94

active

751

31

F

7 c

06/25/94

unknown – drop collar 05/96

120

32

F

13 b

06/04/94

unknown – drop collar 10/97

156

33

F

19 a

06/14/94

active

219

34

M

6a

05/17/95

unknown – no signal 05/97

54

35

F

4a

05/17/96

dead – 09/20/97

186

36

F

8a

06/23/93

active

192

37

F

10 a

06/27/94

active

228

38

M

1 a

06/27/94

unknown – never collared

8

39

F

3a

05/10/95

unknown – no signal 08/96

105

40

F

15c

05/15/95

active

296

41

F

12a

05/28/95

active

54

42

M

7a

05/30/95

active

38

43

M

5a

05/24/96

dead – 10/10/96

11

44

M

4a

06/13/95

dead – 08/23/96

27

45

M

1a

06/15/95

active – no signal 05/97

2

46

F

11a

06/15/95

active

202

47

F

9a

06/02/96

active

137

48

F

2a

06/02/96

unknown – no signal 09/97

14

49

M

2a

06/02/96

unknown – no signal 08/97

17

50

M

4a

06/17/96

unknown – no signal 06/96

2

51

M

8a

05/23/97

active

16

52

M

7b

05/16/97

active

15

53

M

3a

05/15/97

active

23

54

M

15a

06/03/97

active

26

55

F

6a

06/07/97

active

12

56

F

3a

05/28/97

active

4

57

F

5a

05/17/97

active

17

58

M

9a

06/08/97

dead – 09/23/97

5

59

F

3a

05/28/97

active

4

60

F

3a

05/28/97

active

4

61

F

12a

06/11/97

active

72

62

F

8a

06/12/97

active

22

* certainty code a= +/- 0 years, b= +/- 1-2 years, c= +/- 2-3 years
Telemetry Data Set
Aerial and ground monitoring from the mid-March until the first week of December produced 1571 point locations for the 1997 field season. Of these 404 (25%) were from the air and 1167 (75%) from ground monitoring. Aerial locations were biased toward early morning hours. One hundred and seventy eight aerial locations were visual observations. Sightability was higher from the air (44%) than from the ground (9%). Table 3 outlines the number of telemetry points for each individual from project initiation to December 1997.
Population Demographics
Observations from the research team as well as records from Banff National Park, Kananaskis Country Rangers and Alberta Fish and Wildlife Services established a minimum unduplicated count of 2 females with cubs of the year for 1997 in the study area (Table 4). No radio-collared females had cubs of the year during 1997. Over time, a minimum count of sows with cubs can be established and used as a trend indicator (Knight et al. 1995).
Table 4. Unduplicated grizzly bear females with cubs of the year in the Bow River Watershed, Alberta, 1993 through 1997.

Family Identification	Most Cubs Observed	Location	# of Sightings
A – 1993	1	Bryant Creek	2
B – 1993	2	Fatigue Creek	1
C – 1993	2	Moraine Lake	1
D – 1993	2	Cascade River	1
E – 1993	2	Elbow R. / Nahahi Ridge	3
F – 1993	2	Kananaskis Lakes	4
A – 1994	2	Lower Cascade River	1
B – 1994	1	Moose Mtn. / Elbow R.	2
C – 1994	2	Mt. Indefatigable	4
D – 1994	1	Bryant Cr. / Mt. Nestor	2
Bear #28 1994	1	Upper Cascade River	2
Bear #30 1994	3	Baker Lake / Pipestone R.	5
Bear #36 1994	1	Upper Bow River	2
A – 1995	2	West Bragg Cr / Powderface	3
B – 1995	2	Skogan Pass / Wasootch	3
C – 1995	2	Upper Spray / Albert R.	3
Bear #17 1995	1	Cascade River	13
Bear #18 1995	3	Bryant Cr. / Assiniboine	10
Bear #26 1995	2	Nakiska / Evans Thomas	6
Bear #31 1995	2	Highwood River	3
Bear #32 1995	3	Forty Mile Cr. / Elk Lake	12
Bear #33 1995	3	Cascade River / Stoney Cr.	14
A – 1996	1	Cascade R. / Grassy Ridge	1
B – 1996	3	Mid Spray River	1
Bear #24 1996	2	Highwood Pass	25
Bear #36 1996	2	Upper Bow River	8
Bear #37 1996	2	Elbow / Sheep Rivers	3
A – 1997	2	Wind Valley	2
B – 1997	3	Elbow Lakes	2

Reproductive success of radio collared females was determined through visual observation during the spring and summer of 1997 (Table 5). Reproductive data from collared females will eventually be used to construct an estimate of whether the sample population is increasing or decreasing. Before this can be done, data on at least 100 female reproductive years will be needed. Cub mortalities for the year are recorded in the comments. Year to year cub survivorship can be tracked by referring to tables from previous years and the reproductive status of any given female.
Table 5. Reproductive status of known female grizzly bears in the Bow River Watershed, Alberta, 1997.

Bear Identification	# of Cubs	Comments
18	2 two year olds
24	2 yearlings
26	0	1 yearling in 96
30	3 three year olds	family still together in Nov.
32	3 two year olds	family broke up early June
33	2 two year olds	family broke up early June
35	0	died Sept 97
36	1 yearling
37	1 yearling	2 yoy in fall of 96
40	0
41	0
46	2 three year olds	family still together in Nov.
47	2 three year olds	family broke up end of May
48	0	subadult
55	0	new capture
56	0	3 year old cub of bear #30
57	0	new capture
59	0	3 year old cub of bear #30
60	0	3 year old cub of bear #30
61	0	new capture
62	0	new capture

There were 3 known mortalities within the study area in 1997 (Table 6). One was an unmarked bear that was found many months after death, while the other 2 were radio collared study bears.
Table 6. Summary of grizzly bear mortalities in the Bow River Watershed, Alberta, 1993 through 1997.

Bear Identification	Date	Location	Sex	Age	Kill Type
AFWS #21055a	08/19/93	West Spray-408b	M	3	PWc
Research #19	05/13/94	Kananaskis-648	M	6	AC
Research #22	05/28/94	Albert R.-B.C.	M	14	LH
AFWS #25161	09/29/94	Fortress Mt-408	M	Subadult	IL
Research #12	10/04/94	Simpson R.-B.C.	M	13	SD
Research #21	07/26/95	Elkford B.C.	M	4	PW
AFWS #25722	08/20/95	Sarcee Reserve	M	unkn	TI
investigate	fall 95	3 Point Cr.-406	?	unkn	IL
BNP L952104	09/25/95	Lake Louise	F	adult	PW
BNP L952104	09/25/95	Lake Louise	F	yly	PW
AFWS #34990	06/04/96	Morley	M	adult	TI
Research #44	08/23/96	Stoney Reserve	M	5	TI
Research #28	08/24/96	Cascade River	F	24	NA
Research #23	08/08/96	James River	M	5	PW
Research #43	10/10/96	Grease Creek	M	5	IL
BNP97-1567	fall 1996	Spray Lake	?	Subadult	?
Research #35	09/20/97	Evan Thomas Cr.	F	5	SD?
Research #58	09/23/97	James River	M	9	PW

a Registration or file number
b Wildlife Management Unit
c PW=problem wildlife, AC=accidental, LH=legal hunter, SD=self defense, NA=natural, TI=treaty Indian, IL=Illegal
Translocations accounted for removal of 1 bear from the study area in 1997 (Table 7). This bear was removed as a result of human food conditioning and associated public safety concerns. Shortly after translocation the bear began frequenting a campground in Jasper National Park and was subsequently destroyed.
Table 7. Summary of grizzly bear translocations in the Bow River Watershed, Alberta, 1993 through 1997.

Bear Identification	Date	Translocation	Sex	Age
		From To
AFWS #407801a	09/04/93	Canmore-410b Owl Crk-339	M	Subadult
Research #23	10/21/94	Sundre-318 Mitsue-350	M	3
B.C. GF75	09/26/95	Lake Louise Kinbasket L	F	9 & 1yly
Research #50	06/17/96	Canmore-410 Highwood-404	M	4
Research #16	07/05/96	Banff Calgary Zoo	M	8
AFWS#	0729/97	PLPP-648 Nordegg – 428	M	Subadult

aOccurance number
bWildlife Management Unit
Population Estimate
Mike Proctor undertook the analysis and reporting of our 1996 DNA hair sampling for population estimation. His report follows:
In an effort to estimate the density of grizzly bears in the East Slopes study area a DNA-based population estimate was undertaken in 1996. Details of this effort may be found in Sherry (1996) and entailed using microsatellite genotyping of individual bears as “marks” in a mark-recapture population estimate. Radio-collared bears, with known genotypes and known to be in the DNA study area, were considered marked animals and DNA sampling sessions were undertaken in an attempt to “recapture” these animals. Individual genotypes were generated from hair collected at remote sampling stations using barbed wire. During June and July 1996, the East Slopes team systematically collected bear hairs at 40 sampling stations over 4000km2 (2000km2 north of Hwy 1 in Banff National Park and 2000 km2 south of Hwy 1 in Kananaskis Country). Sampling stations were set 1 per cell (1 cell = 10km x 10km). Three consecutive sampling sessions were run for approximately 10 days each and the hair trap sites were moved each session within each cell. In 1296 active days of sampling 250 samples were collected of which 75 were grizzly bear. Fifty of these samples produced a usable DNA genotype identifying 15 different grizzlies. Twelve of these bears were new individuals and 3 were known as radio-collared bears. Of the 15 “DNA captured ” bears 5 were male, 5 female and 5 undetermined sex. Dr. Curtis Strobeck’s population genetics laboratory at the University of Alberta carried out the DNA analysis. While data was too sparse to generate a reliable precise estimate, a density of 1.4 grizzly bears/100 km2 was estimated based on Program NOREMARK (White 1995).
One of the biggest problems in mark-recapture experiments applied to bear studies is most population estimators require the assumption of closure during the sampling. Meeting this assumption is difficult because bears have relatively large movement patterns and will move in and out of the sampling area. Important in any mark-recapture effort is that animals have an equal probability of being captured and recaptured in subsequent sampling sessions. When a “marked” individual leaves the area during one of the sessions, it has no chance of being recaptured, thus introducing a positive bias or an overestimate of the population. This is because the estimators use the ratio of marked to unmarked animals caught in any session as the basis for the population estimate.
The use of radio-telemetry has potential to compensate for this lack of closure. Several methods have been described in the literature (Eberhardt 1990, Garshelis 1992, Miller et al. 1997). These methods require larger sample sizes than attained in this effort in the number of bears marked and recaptured. One estimator, the joint hypergeometric estimator (JHE) within program NOREMARK, is designed to incorporate telemetry data to compensate for lack of geographic closure. The JHE is a maximum likelihood estimator designed for mark – resight population estimates. A number of animals are “marked”, released, and several “resight” sessions follow. In the East Slopes case, the “marked” animals are the radio-collared animals known to be in the sampling area during each of 3 “resight” sessions. The subsequent hair-DNA capture sessions constitute these “resight” sessions.
The value of using the radio-collared bears for the marked sample is that only those bears known to be in the sampling area during each of 3 sessions are counted as the marked sample. This greatly reduces the positive bias associated with “edge effects” due to closure violations as only those marked animals available for resight are included in the calculations of the population estimate. Each of the 3 DNA sampling sessions has a set of marked animals even though the 3 sets may contain different combinations of radio-collared animals.
While 16 radio-collared bears frequented the area during this period, radio-locations determined that 12 individuals were in the sampling area during sessions 1 and 2, and 11 during session 3. The DNA “capture” results are summarized in Table 1. The NOREMARK population estimate is 57 animals in the 4000 km2 area which is a density of 1.4 bears per 100 km2. (Table 2).
Table 1. Summary of grizzly bear hair-DNA captures for 1996.

Session	Radio-Collared available	Marked captured	Unmarked captured
1	12	0	2
2	12	0	2
3	11	3	9

Program CAPTURE, an alternative method of estimating population and density was also used. Program CAPTURE has similar closure assumptions but no direct method of compensation for closure violations. Compensatory methods exist but require larger sample sizes to be effective (see comments above). Program Capture is a set of sophisticated population estimators designed to uncover violations of equal catchability assumptions, again requiring sufficient data not attained in this study. It was necessary to group the East Slopes data into 2 sessions with the radio-collared individuals as the marked animals in session 1 and all 3 DNA collections grouped together as session 2. To then adjust Program CAPTURE’s estimate for closure, the estimate was multiplied by the proportion of time the radio-collared animals spent in the study area over the collection period. This naive telemetry-adjusted estimate yields 46 animals with a density of 1.2 animals per 100 km2. (Table 2)
Table 2. Population and density estimates for the East Slopes 1996 grizzly bear population.

	Estimate	95% CI	90%CI
NOREMARK population	57	29 – 201	31 -156
NOREMARK density	1.4 / 100 km2	0.7 – 5.0 / 100 km2	0.8 – 9/100 km2
CAPTURE (Mt Chao1)	67	40 – 154
Telemetry adjusted CAPTURE ( x 0.69	46	28 – 106
Adjusted density		1.2 /100 km2	0.7 – 2.7/100 km2

1 Mt Chao is one of 10 population estimator models within program CAPTURE. Data was too insufficient for proper model selection within the program. Mt Chao was chosen subjectively as it is the simpler Lincoln-Petersen estimator adjusted by Chao (1989) for sparse data sets (see Boulanger 1997a)
Several considerations are important when interpreting these results. First, is that the number of hair-DNA captures and resights was too low to inspire confidence in the estimates. The lack of precision is primarily due to low capture and recapture rates. Only 3 radio-collared bears were DNA captured in all 3 resight sessions. Sessions 1 and 2 caught only 2 bears each. These sparse data yield confidence limits of 300% to 400%. While the biological significance of these limits may be questionable, they indicate that to generate reliable results more captures and resights are required.
For a reasonably precise result White (1980) recommends:

a capture probability	p = 0.3	for populations under 100
	p = 0.4 or 0.5	for populations near 50
	p = 0.2	for populations over 200

This study realized a capture probability of 0.22 with an estimated population of approximately 50. The particularly low capture rates in Sessions 1 and 2 may have been due to unusually cold weather and therefore less attractive bait, as well as public safety constraints in placement of sample station locations may have contributed to the low result.
Closure violations generally result in overestimates, large variation in capture probabilities and imprecise estimates. While this effort used radio-collared bears to compensate for lack of closure, several problems remained unsolved. Do radio-collared bears visit hair traps with the same frequency as non radio-collared animals? In 1296 trap nights only 3 radio-collared animals were captured. While data is too sparse to adequately address this question, Boulanger (1997b) looked at a similar problem in the 1996 West Slopes Bear Research Project’s data where only 2 radio collared bears were DNA capture in 2653 trap nights. Boulanger concluded preliminarily that it is likely that capture probabilities of radio-collared bears is lower than non collared animals and that more definitive answers require better data potentially supplied by West Slope’s 1997 data. If radio-collared animals do indeed have a lower capture probability than non collared animals then the resight numbers would be low, resulting in an overestimate of the population.
Considering the program CAPTURE telemetry adjustment, using the radio-collared bears to estimate the proportion of time all bears spend in the study area is also problematic. Consider when 2 radio locations of 1 animal are separated by 7 days and the animal is located within the area but near the edge, the assumption is made that the bear was in the study area for the entire 7 days but it is possible that the animal left the area for some amount of time during this interval. The opposite is true for an edge bear found outside of the study area on 2 consecutive locations. In the East Slopes case the trapping of radio-collared animals occurred inside and outside of the DNA study area, diminishing the bias associated with collared animals over-representing the animals within the DNA study area.
Another concern is that the cell sizes may be too large to provide an equal opportunity for capture of some female bears with small seasonal home ranges. A test of this problem is being carried out in the West Slopes 1997 inventory effort. Cell sizes have been reduced to 25 km2 in an area that was sampled at 64 km2 in 1996. Results comparing the 2 years may indicate how much of a problem large cell sizes posed for the 1996 efforts in the East and West Slopes DNA inventories.
Furthermore, the habitat of the DNA sampling area is high quality for bears so extrapolation of densities must consider variations in habitat quality across the ecosystem.
Conclusions
The most important limitation of the DNA based population estimate is the low sample sizes. An estimate based on 3 resights of radio-collared animals should be considered unreliable.
While the use of radio telemetry holds great promise as a tool used in conjunction with DNA sampling, many questions remain unanswered. As many varied DNA-based grizzly bear inventories are being carried out in BC in 1996 and 1997, answers may be found within several years.
The East Slopes has a valuable bank of microsatellite genotyped individual grizzly bears. With the trapped radio-collared bears in the larger East Slope’s ecosystem and the DNA bears of 1996, approximately 74 bears have been identified. This genetic data will contribute to planned future work looking at the grizzly bear meta-population in SW Alberta and SE British Columbia, helping identify dispersal and gene flow patterns, population fragmentation, and linkage corridors.
References
Boulanger, J. 1997a. DNA mark-recapture methods for inventory of grizzly bear populations in British Columbia: Elk Valley (1996) case study. Ministry of Environment, Lands, and Parks, Wildlife Research Branch, Victoria, BC. 21 pp.
Boulanger, J. 1997b Preliminary report: Inference from bears radio-collared during the West Slopes 1996 DNA mark-recapture inventory project. West Slopes Bear Research Project. Revelstoke, BC.
Eberhardt, L. L. 1990. Mark recapture estimation for mark-recapture studies with edge effects. Journal of Applied Ecology, 27:259-271.
Garshelis, D. L. 1992. Mark recapture estimation for animals with large home ranges. In Wildlife 2001: Populations (eds D. R. McMullough and R. H. Barret), pp1098-1109. Elsevier, New York.
Miller, S.D., G.C. White, R.A. Sellers, H.V. Reynolds, J.W. Schoen, K. Titus, V.G.J. Barnes, R.B. Smith, R.R. Nelson, W.W. Ballard and C.C. Brown. 1997. Brown and black bear density estimation in Alaska using radiotelemetry and replicated mark-resight techniques. Wildlife Monograph No. 133. 55 pp.
Sherry, E. E. 1996. An Analysis of Methodologies for assessing Grizzly bear (Ursus arctos horribilis) population. M.Sc. Thesis University of Kent. Canterbury. 155 pp.
White, G. C. 1995. Program NOREMARK Software Reference manual. Colorado State University. Fort Collins. 31 pp.
Critique
Population estimation is a complex enough subject that we felt a critique of Mike Proctor’s work was necessary. That report follows:
Review by John Boulanger, Integrated Ecological Research, Box 5360, Squamish BC. (604-892-2624)
The analysis of the East Slopes estimates by Mike Proctor provides a good coverage of issues concerning this data set and potential problems with population estimates.
I agree with Mike that the estimate of population size for the East Slopes is not completely reliable primarily because of low sample sizes and issues concerning the trappability of radio collared bears.
Two methods of estimation using CAPTURE (Otis et al., 1978) and NOREMARK (White, 1996) were proposed by Mike Proctor for population estimates. I will comment about the East Slopes estimates using these two methods in terms of the three primary design issues (White et al., 1982).
1) Closure:
a) DNA/CAPTURE estimate:
The frequency of occasions in which radio collared bears were in and out of the trapping area was used to adjust the estimate for closure bias (by .69). This adjustment factor of .69 is also an estimate based on assumptions (as Mike Proctor outlined) regarding the movement pattern of radio collared bears. Therefore, this adjustment factor also has a variance. I attempted to use a bootstrap technique to get at the potential variance of the adjustment factor with the West Slopes 1996 data and found the confidence interval on the West Slopes correction factor estimate of 73% to be 55.23% and 90.92%. Given that sample sizes were about equal between the East Slopes and West Slopes we can expect a similar confidence interval width. Therefore, the utility of this correction factor is not great given that the population estimate also has a large confidence interval width.
Also, this adjustment factor only accounts for the geographic bias caused by the grid sampling a larger area due to edge effects. It does not account for capture probability bias caused by closure violation.
b) Mark/resight estimate.
This technique uses the radio collared bears as the primary sample unit. Closure bias can be more exactly corrected because it is known whether these bears are in or out of the study area during the recapture process. Furthermore, the correction accounts more directly for capture probability bias due to closure violation because the capture frequencies of radioed bears are adjusted each sample session based on whether they were in or out of the grid. Therefore this method is superior to the DNA capture method.
2) Sample size: Both methods are compromised by low sample size. As a result the confidence intervals are quite large. In addition, it is not possible to discern potential capture probability variation. A value of using mark-recapture as opposed to census based methods is that an estimate of uncertainty can be calculated (confidence interval). While this may not give the most favorable results it is still better than blindly trusting a census or count estimate which has no estimate of confidence.
3) Capture probability variation. Low sample sizes made it difficult to determine if there is capture probability variation in this data set. Some general comments.
a) As mentioned by Mike Proctor, there is some concern that radio collared bears may show lower probabilities of capture then other bears and therefore estimates using radio collared bears may show a positive bias.
b) I pose the following questions: Can it be assumed that the radio collared bears are a random sample of the population of DNA bears?. Can it be assumed that the radio collaring efforts that produced the first radio sample targeted the same population as the DNA census? In terms of each method of estimation the following problems are possible if this assumption is not met:
i) CAPTURE-Program capture assumes you are sampling a closed population. The best analogy is a ball and urn used in the lottery in which each ball is a bear. Balls or bears can have variable capture probabilities with capture but it is assumed that they are all present in the urn. If there is a different urn, or different population was being sampled when bears were radio collared then it is invalid to use the radio collared bears as a first mark and DNA bears as a second mark.
ii) NOREMARK-The assumptions outlined above for program CAPTURE are relaxed. It is assumed that after collaring the radio collared bears intermix with the non-radio collared bears so at the time of sampling they are a random sample (in terms of spatial distribution, demographics etc) of the targeted DNA census population. This is why Miller et al.,(1997.) recommends that mark-resight censuses for bears up north occur at least a few months after the bears are radio marked. It seems like this assumption would be met in the case of the East Slopes.
Given the above outlined points I recommend that the NOREMARK estimate be used for the population estimate of the East Slopes. The large confidence interval width on this estimate should be emphasized.
In conclusion, the report provided by Mike Proctor does an excellent job of covering the bases in terms of mark-recapture issues and possible analysis techniques. It should provide a valuable interpretation of the results of the East Slopes mark-recapture effort.
References
Miller, S. D., White, G. C., Sellers, R. A., Reynolds, H. V., Schoen, J. W., Titus, K., Barnes, V. G. J., Smith, R. B., Nelson, R. R., Ballard, W. W. & C.C., S. (1997.). Brown and black bear density estimation in Alaska using radiotelemetry and replicated mark+resight techniques. Wildl. Monogr. No. 133., 55pp.
Otis, D. L., Burnham, K. P., White, G. C. & Anderson, D. R. (1978). Statistical inference from capture data on closed animal populations. Wildlife Monographs 62, 1-135.
White, G. C. (1996). NOREMARK: Population estimation from mark-resighting surveys. Wildlife Society Bulletin 24, 50-52.
White, G. C., Anderson, D. R., Burnham, K. P. & Otis, D. L. (1982). Capture-recapture and removal methods for sampling closed populations. Los Alamos National Laboratory .
LITERATURE CITED
Carr, H.D. 1989. Distribution, numbers and mortality of grizzly bears in and around Kananaskis Country, Alberta. Fish and Wildl. Div. Wildl. Manage. Branch Wildl. Res. Series 3. 49 pp.
ERCB. 1994. Decision report D 94-8, Application for an exploratory well, Amoco Canada Petroleum Company Ltd., Whaleback Ridge Area. Energy Resources Conservation Board, Edmonton.
Gibeau, M.L. In press. Grizzly bear habitat effectiveness model for Banff, Yoho and Kootenay National Parks, Canada. Int. Conf. Bear Res. And Manage. 10: 000-000.
Gibeau, M. and S. Herrero. 1995. Eastern Slopes Grizzly Bear Project: 1994 Progress Report. University of Calgary, AB. 26 pp.
Gibeau, M. and S. Herrero. 1996. Eastern Slopes Grizzly Bear Project: 1995 Progress Report. University of Calgary, AB. 46pp.
Gibeau, M., S. Herrero, J. Kansas and B. Benn. 1996. Grizzly bear population and habitat status in Banff National Park: A report to the Banff Bow Valey Task Force. University of Calgary, AB. 62 pp.
Herrero, S. 1995. The Canadian National Parks and grizzly bear ecosystems: The need for interagency management. Int. Conf. Bear. Res. and Manage. 9:7-21.
Herrero, S. and J. Herrero. 1996. Cheviot Mine: A proposed carnivore compensation program. BIOS Environmental Research Ltd., Calgary, AB. 38 pp.
Herrero, S., D. Poll, M. Gibeau, J. Kansas, and B. Worbets. In Press. The eastern slopes grizzly bear project: Origins, organization, and direction. Proceedings; Canadian Council on Ecological Areas, Annual meeting, Calgary, AB. November 1995.
Komex Intl. 1995. Atlas of the Central Rockies Ecosystem. Komex Intl., Calgary, A.B.
Mattson, D.J., S. Herrero, R.G. Wright and C.M. Pease. 1996. Science and management of Rocky Mountain grizzly bears. Consv. Biol. 10(4): 1013-1025.
Nagy, J.A. and J.R. Gunson. 1990. Management plan for grizzly bears in Alberta. Alberta Fish and Wildlife, Edmonton. 164 pp. plus appendicies.
Noss, R. and Cooperrider. 1994. Saving natures legacy: Projecting and restoring biodiversity. Defenders of Wildlife and Island Press, Washington, D.C.
NRCB. 1993a. Decision report. Application to construct a recreational and tourist resort Project in the Town of Canmore, Alberta. Application 9103- Three Sisters Golf Resorts Inc. Natural Resources Conservation Board. Edmonton.
NRCB. 1993b. Decision report, Application #9201, Vacation Alberta corporation application to construct recreation and tourism facilities in the West Castle Valley, Near Pincher Creek, Alberta. Natural Resources Conservation Board, Edmonton.
Raine, M. and R. Riddell. 1991. Grizzly bear research in Yoho and Kootenay National Parks. Canadian Parks Service Report, Calgary, Alberta.
Russell, R.H., J.W. Nolan, N.G. Woody, and G.H. Anderson. 1979. A study of the grizzly bear in Jasper National Park. Canadian Wildlife Service, Edmonton. 102 pp. plus 10 appendicies.
PERSONAL COMMUNICATIONS
Woods, John. Research ecologist. Glacier/Revelstoke National Parks, Revelstoke, B.C.