Grizzly Bear Population Linkage Zones in the Sea to Sky
Planning Area of Southwestern British Columbia
Clayton D. Apps, RPBio 1
Ministry of Water, Land and Air Protection
Surrey, British Columbia
Aspen Wildlife Research
2708 Cochrane Road NW • Calgary, Alberta, Canada • T2M 4H9 • 403-270-8663
Apps, C. D. 2001. Grizzly bear population linkage zones in the Sea to Sky Planning Area of
southwestern British Columbia. Prepared for Ministry of Water, Land and Air Protection,
Surrey, British Columbia. Aspen Wildlife Research, Calgary, Alberta, Canada.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 1
This project was funded by the Corporate Resource Inventory Initiative administered by the
British Columbia Land Use Coordination Office. Contract supervision was provided by Tom Plath
and Jim Roberts of the Ministry of Water, Land and Air Protection (WLAP) . Tom, Jim, Tony
Hamilton, and Steve Rochetta provided direction and advice for this work and extensive insight
into local grizzly bear habitat and other resource values that factored into analysis and
interpretation. Robert Hewlett and Gurdeep Singh of British Columbia Assets and Land
Corporation (BCAL) facilitated the acquisition and translation of required spatial data. Helpful
review and comments on an earlier draft of this report were kindly provided by Trevor Kinley, Jim
Roberts, Tony Hamilton, Matt Austin, and Tom Plath. Thanks to Al McEwan of the Pemberton
Wildlife Association (PWA) for sharing with me his insights and experiences and the views of his
Formerly Ministry of Environment, Lands and Parks (MELP).
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 2
Grizzly bear range has been dramatically reduced across North America since European
settlement, and British Columbia supports about one quarter of the remaining population. This
includes much of the species’ southern range extent, where populations are restricted to
mountains and high plateaus associated with restricted human access and settlement. In this
province, key components of a grizzly bear conservation strategy include the identification of core
population areas, within which human activities not compatible with grizzly bear conservation can
be controlled, and the provision of bear movement and genetic interchange among such areas.
Lands that provide connectivity among core population areas are termed “linkage zones”. The
identification of linkage zones is especially important within landscapes that currently are or may
be subject to at least moderate levels of human development. Within grizzly bear range, such
areas are typically associated with valley-bottom transportation corridors, where human impacts
are concentrated in a linear manner and where there is a trend toward increased and permanent
I conducted an analysis to define population linkage zones for grizzly bears within a 13,800
km study area defined by the Sea to Sky Planning Area (SSPA) in the southern Coast
Mountains of southwestern British Columbia. Within this area, the grizzly bear population
currently is considered “threatened”, while the species has been extirpated immediately to the
south. Restoring and maintaining a viable population into the future represents a tremendous
challenge. The SSPA is and will continue to be subject to land use pressures for settlement,
public and commercial recreation, and resort development from a growing human population
resident and within the nearby greater Vancouver area. Other land use demands include
resource extraction, agriculture, and the transportation infrastructure required by all land uses.
I modified and applied an existing GIS-based modeling approach to delineate grizzly bear
population linkage zones within the SSPA. A digital database was built from existing data
sources ranging from 1:20,000 to 1:250,000 in scale. Methods involved the derivation of 5
submodels: human features, linear disturbance, visual cover, food value, and terrain condition.
Classes within each submodel were rated relative to each other in terms of their contribution to
grizzly bear habitat “effectiveness”. Each submodel was then derived at both a broad and a fine
spatial scale within the GIS. At each scale, submodels were converted to standard scores across
the study area, a weighted average was obtained, and the result was again standardized across
the study area. Final values were converted to a relative scale of habitat effectiveness. At the
broad spatial scale, values reflected the average condition a female bear would encounter in
expected daily movements, and these results were used to delineate linkage zones. At the fine
Now the Sea to Sky Land and Resource Management Plan area.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 3
spatial scale, habitat effectiveness values portrayed potential sources of natural and human
fragmentation within and proximal to each linkage.
The analysis highlighted several potential linkage zones within landscapes that currently
are or may be subject to at least moderate levels of human influence. Between Squamish and
Whistler, an east-west linkage across Highway 99 was identified associated with Culliton Creek,
about 10 km north of Brackendale. Across the Lillooet River valley, northwest of Pemberton, a
significant linkage zone appears to connect the Ryan River valley in the southwest to drainages in
the vicinity of the upper Birkenhead and Hurley rivers in the northeast. Further down the Lillooet
River, a potential linkage was identified connecting Miller Creek in the southwest to Owl Creek in
the northeast and drainages further north. Between the towns of Pemberton and Mt. Currie, a
small connection may link drainages of Mount Currie and others within Garibaldi Provincial Park
with the Owl Creek drainage in the north. Further to the east, another minor linkage around the
north end of Lillooet Lake may connect the Ure Creek drainage and Garibaldi Park in the south
with side drainages of Joffre Creek in the north. Further up the Anderson Lake road, a significant
linkage appears to exist across the Gates River valley just south of the community of D’Arcy. It
connects the drainages of Haylmore and Spruce creeks in the southeast with the Blackwater
Creek drainage and tributaries of upper Birkenhead River in the northwest. Between Lillooet
Lake and Harrison Lake, in the southeast of the study area, there is little human development,
and virtually the entire Lillooet River valley bottom constitutes a potential east to west linkage
zone. This may facilitate bear movement between the eastern drainages Garibaldi Park and the
Stein and Nahatlatch drainages to the northwest.
Maintaining viable grizzly bear populations in the southern coastal mountains requires
special considerations in all landscapes, beyond those that may currently function as
core/security areas. Retaining or restoring the functional connectivity of population linkages
requires the management of various human influence sources at both broad and fine spatial
scales. The integrity of some linkages depends on commitment by private and First Nations
stewards. Over time, the viability of grizzly bears and perhaps other wide-ranging species within
the SSPA may depend on these zones facilitating successful dispersal and population
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 4
Table of Contents
Acknowledgements ............................................................................................................. 2
Table of Contents ................................................................................................................ 5
List of Tables ....................................................................................................................... 6
List of Figures...................................................................................................................... 7
List of Figures...................................................................................................................... 7
List of Acronyms Used........................................................................................................ 8
List of Acronyms Used........................................................................................................ 8
Introduction ......................................................................................................................... 9
Study Area ........................................................................................................................ 10
Grizzly Bear Population Status ......................................................................................... 12
Analysis Approach ............................................................................................................ 16
Geographic Data Sources ............................................................................................ 18
Model Structure............................................................................................................. 18
Results and Discussion .................................................................................................... 25
Management and Research Recommendations.............................................................. 37
Literature Cited.................................................................................................................. 39
Personal Communications................................................................................................ 42
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 5
List of Tables
Table 1. Current and potential population estimates (A. N. Hamilton, WLAP, personal
communication) for grizzly bear population units (GBPUs) comprising the
Sea to Sky Planning Area (SSPA) of southwestern British Columbia........... 12
Table 2. TRIM human feature categories and zones of influence assigned according
to the LZP model (Apps 1997)........................................................................ 19
Table 3. Human-influence scores assigned by proximity to zones of influence............. 20
Table 4. Linear disturbance feature types and relative weightings for density
calculations. .................................................................................................... 20
Table 5. Human-influence scores assigned to linear disturbance density classes........ 21
Table 6. Human-influence scores assigned to visual cover classes.............................. 22
Table 7. Scores assigned habitat conditions relative to grizzly bear food value. ........... 23
Table 8. Scores assigned to slope classes..................................................................... 23
Table 9. Classification of cumulative habitat effectiveness (HE) scores for linkage
zone delineation. ............................................................................................. 24
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 6
List of Figures
Figure 1. Study area location relative to grizzly bear range in British Columbia
(McLellan 1998). ..............................................................................................13
Figure 2. Grizzly bear linkage zone study area encompassing the Sea to Sky planning
area in southwestern British Columbia............................................................14
Figure 3. Grizzly bear population units and current status within British Columbia, and
the study area defined by the Sea to Sky planning area. ...............................15
Figure 4. Thematic layers used in the grizzly bear Linkage Zone Prediction (LZP)
model, modified from Servheen and Sandstrom (1993). ................................17
Figure 5. Predicted grizzly bear population linkage across the Sea to Sky regional
planning area in southwestern British Columbia.............................................30
Figure 6. Identified linkage zones merged with Landsat 7 panchromatic satellite
imagery for the southwest focal area. .............................................................31
Figure 7. Level 2 habitat effectiveness values merged with hillshaded image for the
southwest focal area........................................................................................32
Figure 8. Identified linkage zones merged with Landsat 7 panchromatic satellite
imagery for the north focal area.......................................................................33
Figure 9. Level 2 habitat effectiveness values merged with hillshaded image for the
north focal area. ...............................................................................................34
Figure 10. Identified linkage zones merged with Landsat 7 panchromatic satellite
imagery for the southeast focal area. ..............................................................36
Figure 11. Level 2 habitat effectiveness values merged with hillshaded image for the
southeast focal area.........................................................................................36
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 7
List of Acronyms Used
BCAL British Columbia Assets and Land Corporation
BEC Biogeoclimatic Ecosystem Classification
BEI Broad Ecosystem Inventory
BEU Broad Ecosystem Unit
BTM Baseline Thematic Mapping
DEM Digital Elevation Model
FIP Forest Inventory Planning
GBCS British Columbia Grizzly Bear Conservation Strategy
GBPU Grizzly Bear Population Unit
GIS Geographic Information System
HE Habitat Effectiveness
LRMP Land and Resource Management Plan
LZP Linkage Zone Prediction
MELP Ministry of Environment, Lands and Parks
MOF Ministry of Forests
PWA Pemberton Wildlife Association
SSPA Sea to Sky Planning Area
TRIM Terrain Resource Information Management
WLAP Ministry of Water, Land and Air Protection
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 8
Grizzly bear populations were historically distributed across virtually all of western North
America. Over the past two centuries, however, a massive range contraction occurred as human
settlement and populations expanded. In the lower 48 United States, occupied grizzly bear range
is presently less than two percent of that prior to European settlement (Servheen 1990). Today,
British Columbia is home to one quarter of the North American and half of the Canadian grizzly
bear population (Banci et al. 1994). This includes much of the species’ southern range extent,
where populations are restricted to mountains and high plateaus associated with restricted
human access and settlement (McLellan 1998). Such lands are typically managed for multiple
resource values, presenting a major challenge to grizzly bear conservation as human populations
and land-use pressures mount. Aiming to moderate impacts of continued urban development
and multiple land uses on grizzly bears, the British Columbia Grizzly Bear Conservation Strategy
(GBCS) was released (Ministry of Environment, Lands and Parks 1995), a major goal of which is
to “maintain in perpetuity the diversity and abundance of grizzly bears and the ecosystems on
which they depend throughout British Columbia for future generations.”
Key components of the GBCS are to (1) delineate a network of areas associated with high-
quality grizzly bear habitat and where activities not compatible with grizzly bear conservation can
be controlled or prohibited, and (2) maintain population connectivity by allowing opportunities for
grizzly bear movement among such areas. This is directly analogous to the core/security area
and linkage zone concepts that have been promoted by grizzly bear biologists elsewhere.
Core/security areas are defined as landscapes where bears can meet their daily energetic
requirements while at the same time choosing to avoid people (Mattson 1993, Gibeau 2000).
Presumably, larger core areas with greater inherent habitat quality are more likely to function as a
regional population source. However, the conservation efficacy of the core/security area concept
largely depends on successful animal movement and gene flow among core areas. Insular
populations exhibit a diminished resilience to genetic, demographic, and environmental
stochasticity, and are thus prone to extirpation (Gilpin and Soulè 1986, Weaver et al. 1996).
Populations of wide-ranging species are particularly vulnerable to fracture and isolation where
current distribution and movement options are already limited by natural conditions. This is
especially true where human impacts are concentrated in a linear manner and where there is a
trend toward increased and permanent development. Such “fracture zones” have the potential to:
(1) inhibit the movement of grizzly bears and other wide-ranging species, and (2) cause
significant mortality as these animals move through or attempt to live within them (Servheen and
Sandstrom 1993). Fracture zones are typically associated with major valley-bottom
transportation routes preferred for human settlement and development. In mountainous
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 9
landscapes, such low elevation lands are of inherently high value to most wide-ranging species,
and are important for functional connectivity through movement and population interchange.
Such lands that span valleys, allowing individuals to not only move through freely but also to
remain resident for some period of time, are termed “linkage zones” (Servheen and Sandstrom
1993). Linkage zones are especially important for grizzly bears, given the gradual process and
relatively limited distance of known natal dispersals (McLellan and Hovey 2001).
The southern ranges of British Columbia’s Coast Mountains represent one of the greatest
challenges to grizzly bear conservation. Within the Highway 99 corridor and surrounding ranges,
heavy development pressures are fueled by a rapidly growing resident human population,
immediate access by a greater Vancouver population of over two million people, and international
recognition as an all-season resort destination. Beyond this, the area is subject to the same
issues and challenges as other provincial regions in balancing resource extraction,
recreation/tourism, agriculture, urban development and other non-consumptive values. In 1999,
an interagency planning project was initiated for the Squamish Forest District called the Sea to
Sky Public Land Strategy. In 2001, acknowledging mounting recreation and tourism demands,
high forest and wildlife values, rapid community growth, and transportation bottlenecks, the British
Columbia government initiated a process to develop a comprehensive Land and Resource
Management Plan (LRMP) for a 1.1 million hectare Sea to Sky Planning Area (SSPA).
Within the SSPA, the Ministry of Water, Land and Air Protection (WLAP) commissioned
me to assess landscape values for grizzly bear population linkage. To achieve this in the
absence of systematically collected empirical data, I modified and applied an existing GIS
modeling approach. My recommendations are based on the results of these analyses, combined
with cursory field assessments and local knowledge. The products derived are intended for
decision-support within the ongoing LRMP process. However, my intent was also to make this
work directly accessible to local communities, organizations, and individuals to support local
efforts to conserve grizzly bears and their associated natural community.
The study area encompassed the SSPA, delineated according to the administrative
boundaries of the Squamish Forest District. It occurred at a southern and western extent of
current grizzly bear range (Figure 1), comprised approximately 13,800 km , and included two
major watersheds (Figure 2). The Squamish River drains the western portion of the area and
empties into Howe Sound, and the Lillooet River drains the northern and eastern portion of the
Formerly Ministry of Environment, Lands and Parks (MELP).
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 10
area and empties through Harrison Lake into the Fraser River. The area falls within the Pacific
Ranges Ecoregion, the southernmost range in British Columbia’s Coast Mountains, and the
Interior Transition Ranges Ecoregion that have a coast-interior transition climate (Demarchi
1996). These ranges are steep and rugged, with glaciated terrain, and fertile river valleys and
floodplains. Biogeoclimatic zones (Meidinger and Pojar 1991) vary on an elevational gradient
and with proximity to the coast. At lower and higher elevations respectively, the relatively wet
Coastal Western Hemlock and Mountain Hemlock zones characterize most of the area.
However, in drier parts of the Lillooet River watershed, the Interior Douglas-Fir and Englemann
Spruce Subalpine Fir zones also occur. Alpine Tundra occurs at the highest elevations
throughout the area and includes several glaciers.
The study area includes the larger communities of Squamish, Whistler, and Pemberton,
and numerous smaller communities, including D’Arcy, Mt. Currie, Furry Creek, and Lions Bay.
The resident human population is currently estimated at 31,000 with a 20 year annual growth rate
of 3.9%, but the area attracts large numbers of visitors from the nearby greater Vancouver
population, estimated at 2,011,000, and from around the world (Holman and Nicol 2001). First
Nations include the Squamish, Lil’wat, Tsleil-Waututh, N’Quat’qua, and In-SHUCK-ch peoples.
Resource use is varied and includes forestry, public and commercial recreation, mineral
exploration and development, small hydro development, range use, small-scale agriculture, and
water supply. In addition to the major resort of Whistler/Blackcomb, there are several new
ski/resort developments proposed including event sites associated with a bid for the 2010 Winter
Olympics. The cumulative effects of these projects are expected to result in additions to
transportation infrastructure, urban development, population growth, and greater demand for
outdoor recreation throughout the SSPA (Holman and Nicol 2001). Protected areas in the form of
provincial parks and environmental reserves comprise 22% of the analysis area, and include
Garibaldi, Birkenhead Lake, Joffre Lake, Upper Lillooet, Clendinning, Tantalus, Callaghan Lake,
Golden Ears, Indian Arm, and Pinecone Burke provincial parks. Private lands and Indian
reserves comprise 2.6% of the SSPA and are clustered along transportation corridors within the
In addition to grizzly bears, the analysis area contains significant values for several wildlife
species of concern. These include several anadromous fish species. Birds include the
mature/old-growth forest dependent spotted owl, marbled murrelet, and northern goshawk,
raptors such as the peregrine falcon and bald eagle, and others such as American bittern, great
blue heron, green heron, trumpeter swan and harlequin duck. Amphibians and reptiles include
the tailed frog and rubber boa. Ungulates include mule deer, mountain goat and moose. Bats
include Keen’s long-eared myotis and Townsend’s big-eared bat. Other wide-ranging carnivores
include wolverine, cougar, fisher and bobcat.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 11
Grizzly Bear Population Status
Grizzly bears are listed as “vulnerable” in Canada (Banci et al. 1994) and “threatened” in
British Columbia (Cannings et al. 1999). The province has been split into 65 grizzly bear
population units (GBPUs), which delineate current and/or historic populations for management
purposes. Each GBPU is rated according to the current estimated population relative to the
estimated inherent capability to support grizzly bears under ideal conditions. On this basis,
populations are assigned a conservation status of “viable”, “threatened” or “extirpated”. GBPUs
considered to be threatened (i.e., current population = 1 – 50% of potential) are subject to the
development of a comprehensive recovery plan to provide direction for population recovery and
long-term viability (e.g., North Cascades Grizzly Bear Recovery Team 2001).
Occurring at a southern and western extent of grizzly bear range, the SSPA encompasses
parts of four GBPUs, all of which are considered threatened: Squamish-Lillooet, Garibaldi-Pitt,
Stein-Nahatlatch, and South Chilcotin Ranges (Figure 3). Population recovery plans have yet to
be completed for any of these. Grizzly bears have not been legally hunted within the SSPA since
1970 (R. D. Forbes, WLAP, personal communication), with the exception of the South Chilcotin
Ranges GBPU, including the Gates River watershed, which supported an open hunt until 1989
and a limited entry hunt until 1999 (D. Jury, WLAP, personal communication), Using a qualitative
approach to population estimation (Fuhr and Demarchi 1990) with modification, A. N. Hamilton
(WLAP, personal communication) recently estimated grizzly bear populations within GBPUs that
comprise the SSPA (Table 1). This translates to a density of 5 bears / 1000 km of “useable”
habitat, or 44 bears out of a potential of 169 (24%), although grizzly bear densities are clearly
expected to be uneven across the SSPA.
Table 1. Current and potential population estimates (A. N. Hamilton, WLAP, personal
communication) for grizzly bear population units (GBPUs) comprising the Sea to Sky Planning Area
(SSPA) of southwestern British Columbia.
GBPU Current Potential Realized Current / 1000 km
Squamish-Lillooet 27 133 0.2 5
Garibaldi-Pitt 19 167 0.11 3
Stein-Nahatlatch 60 183 0.33 8
South Chilcotin Ranges 104 230 0.45 7
Combined for SSPA 44 169 0.24 5
Calculation based on “useable” area.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 12
Expected Grizzly Bear Range
Medium to High Densities
WASHINGTON IDAHO MONTANA
Figure 1. SSPA study area location relative to grizzly bear range in British Columbia (McLellan 1998).
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 13
Figure 2. Grizzly bear linkage zone study area encompassing the Sea to Sky planning area in southwestern
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 14
Grizzly Bear Population Units
and Current Status
within British Columbia
Sea to Sky Regional
Figure 3. Grizzly bear population units and current status within British Columbia, and the study
area defined by the Sea to Sky Regional Planning Area.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 15
To delineate potential grizzly bear population linkage zones within the study area, I
modified and applied existing GIS-based modeling methods. This approach is objective,
consistent, and repeatable, and the assumptions are explicit. Model output can also be easily
integrated with the LRMP process and local planning and public outreach initiatives. Although the
assumptions reflect general relationships that have been quantified through field research
elsewhere, the influences of habitat quality and human activity on bear behaviour and mortality
(e.g., Archibald et al. 1986, Mattson 1990, McLellan 1990) and associated interactions are
complex and have yet to be empirically modeled. Therefore, I recommend that every opportunity
be taken to test and refine model parameters with field data derived through appropriate sampling
My methods were based on the Linkage Zone Prediction (LZP) model developed by
Servheen and Sandstrom (1993) and modified for application within British Columbia (Apps
1997). This model has been applied to evaluate the likelihood of grizzly bear population
connectivity through landscapes with a variety of human activities, ownerships, and management
objectives. It evolved from earlier cumulative effects modeling (Weaver et al. 1986), but is largely
driven by human influence and disturbance factors. It is assumed that in areas of high human
development, human influence will limit bear persistence and movement potential rather than
seasonal food availability. However, outside zones of high human influence, population
connectivity may be largely determined by food availability and terrain conditions. Because I
intended to apply the model over a variety of landscape conditions across the study area, I
modified the LZP model to better account for food value and terrain conditions. Also, because
spatial scale may greatly influence any delineation of habitat connectivity or population linkage, I
incorporated methods that allowed for model application within a specific context of spatial scale,
and I applied the model at both a broad and fine spatial scale.
There are four submodels within the original LZP model, each of which yields a GIS layer
scored according to influence on bears: developed human features, linear disturbance, visual
cover, and riparian areas. In my revision, the riparian component was encompassed within a
food value submodel that accounted for several other factors that are expected to influence
grizzly bear foraging success. I also included a terrain submodel that accounted for the influence
of slope condition on grizzly bear movements. Within each submodel, classes were rated relative
to each other, and a standardized scored map was produced for the study area. The submodels
were then weighted relative to each other and summed (Figure 4), with raw output again
standardized for the study area. Because food value and terrain conditions were accounted for, I
refer to the final model output as “habitat effectiveness”, as it is expected to reflect inherent
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 16
habitat value downgraded according to human influence. A functional response of grizzly bears
to higher human influence levels is assumed to occur through direct displacement and/or higher
incidence of mortality and translocation. A direct positive relationship between cumulative habitat
effectiveness within a landscape and population linkage potential is expected.
Although some bears (e.g., females with cubs) may habituate to human activities, they are
also much more likely to cross the threshold of human tolerance and be removed from the
population (Mattson et al. 1992, McLellan et al. 1999, Gibeau 2000). This issue is amplified in the
southern Coast Mountains where valley bottoms are of extremely high value to bears during
spring and fall, but are highly limited in area and typically subject to human activity and
development (A. N. Hamilton, WLAP, personal communication).
The LZP modeling approach has been applied in conservation planning for several areas
including Montana’s Swan-Clearwater valleys (Servheen and Sandstrom 1993) and Evaro Hill
area (Meitz 1994), Alberta’s Bow Valley (Gibeau et al. 1996), and the Highway 3 corridor of
southeastern BC and southwestern Alberta (Apps 1997). In the Flathead valley, Kehoe (1995)
verified the basic model assumptions. Within identified linkage zones, model output has also
been used to promote community-based initiatives consistent with grizzly bear conservation (e.g.,
Figure 4. Thematic layers used in the grizzly bear Linkage Zone Prediction (LZP) model, modified
from Servheen and Sandstrom (1993).
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 17
Geographic Data Sources
Geographic data were obtained from several sources. Planimetric data of hydrography and
point and linear human features were derived from 1:20,000 Terrain Resource Information
Management (TRIM) files (Surveys and Resource Mapping Branch 1992). Additional road data
were acquired from the Squamish district forest inventory database. Landsat 7 multispectral and
panchromatic data were obtained for the analysis area from Geographic Data BC, Victoria. I
geocorrected and merged each image, and I adjusted reflectance values to produce a seamless
coverage across the study area. Terrain coverages were derived from a 1:20,000 digital
elevation model (DEM; Geographic Data BC 1996). A 1:20,000 forest cover inventory was
acquired for the Squamish Forest District, Garibaldi Provincial Park, and Tree Farm License #38
(FIP; Resources Inventory Branch 1995). Other habitat data included 1:250,000 biogeoclimatic
ecosystem classification (BEC; Meidinger and Pojar 1991), broad ecosystem inventory data (BEI;
Resources Inventory Committee 1997), and baseline thematic mapping (BTM; Surveys and
Resource Mapping Branch 1995). Through the Sea to Sky LRMP spatial database, I obtained
land use and resource management designations, landscape unit polygons, ownership and
administration, and habitat values for select species. I acquired GBPU data through the WLAP
provincial grizzly bear specialist. All data were rasterized at 100 m for broad-scale analysis and
25 m for fine-scale analysis.
A spatially referenced database of grizzly bear occurrences was assembled by Steve
Rochetta (WLAP, Squamish) reflecting the locations of direct observations, confirmed sign, or
control actions recorded since 1975. I acquired this database in its form at the time of this
analysis, prior to an exhaustive survey of all potential observers. It is expected to be continually
updated as past and future observations are reported (J. Roberts, WLAP, personal
communication). I expected that these data were largely biased by grizzly bear sightability and
human access, and thus could not be treated as a random sample of grizzly bear occurrence
derived from systematic sampling. However, they provided a cursory evaluation of population
linkages relative to landscapes that have been known to support grizzly bears.
I modified and applied the LZP modeling approach (Servheen and Sandstrom 1993, Meitz
1994, Apps 1997) to assess grizzly bear habitat effectiveness and population linkage potential. I
first applied the model at a broad scale (level 1), to evaluate population linkage and fracture
across the entire SSPA. On the basis of these results, I defined key focal areas for maintaining
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 18
or restoring population connectivity within landscapes that are or may eventually be subject to at
least moderate levels of human development. Within each focal area, I applied the model at a
much finer scale to evaluate current habitat effectiveness and to provide more detail with respect
to potential linkage zones.
Human Features Layer
Zones of influence were assigned to each of several human feature categories to derive
influence scores associated with human developments (Table 1). The size of each buffer zone is
based on assumptions of the distance at which each type of disturbance will influence bears,
through either attraction or avoidance, leading to direct displacement or increased mortality risk.
For this analysis, I adopted buffer zones described by Meitz (1994) consistent with those
recommended for cumulative effects modeling. A scored map was then derived by assigning
values to each of four influence proximity classes reflecting the expected relative contribution to
habitat effectiveness (Table 2). The assigned scores reflect decreased human influence at
greater distances to point and area features of human development.
Table 2. TRIM human feature categories and zones of influence assigned according to the LZP
model (Apps 1997).
50 m influence 100 m influence 200 m influence
Park-Picnic_Area Fish_Hatchery Yard Dump
Ski_Area Electrical_Substation_Complex Lumber_Yard Sewage_Treatment
Barn Pile Stock_Yard Airport
Greenhouse Pit_Sand-Gravel Mine Air_Strip
Fire_Station Quarry_Dry Military_Establishment Helipad
Weigh_Scale Tailing_Pond Campground-Campsite Railway_Yard
Historic_Site-Pnt_Interest Air_Field Exhibition_Ground Built_Up_Area
Ski_Jump Building Golf_Course Designated_Area
Ski_Lift Burner Sports_Field College
Aerial_Cableway Oil_Well Driving_Range Library
Dock_Dry Gas_Well Rifle_Range School
Dock_Ferry Nursery Trailer_Park Penitentiary
Pier Hospital Police_Station
Wharf Oil-Gas_Facilities Post_Office
Higher scores reflect a greater positive influence on habitat effectiveness. Lower scores reflect
a greater negative impact on habitat effectiveness.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 19
Table 3. Human-influence scores assigned by proximity to zones of influence.
Proximity Class Assigned Score
> 200 m from influence zone 8
100 - 200 m from influence zone 5
< 100 m from influence zone 3
within influence zone 2
Linear Disturbance Layer
Densities of roads and other linear disturbance features were calculated using the TRIM
linear features data and the forest inventory road database. The density calculations carried out
to derive this layer are greatly influenced by decisions about the type of disturbances to be
included, dramatically affecting model results. I therefore adopted a subjective weighting scheme
(Apps 1997) to account for the relative influence of each of four disturbance classes (Table 3).
Features corresponding to secondary roads were weighted as 1, whereas all other types were
subjectively weighted relative to this class. This effectively caused roads weighted as 2 to be
treated as two roads in the density calculation, and those weighted as 0.5 to be treated as one-
half of a road. Because road access through private land leading to the Ryan Creek drainage is
restricted, I removed this road network from density calculations.
Table 4. Linear disturbance feature types and relative weightings for density calculations.
Disturbance Class Specific Feature Types Weight
primary all paved roads 2
secondary all gravel and 2 lane loose roads 1
all rail lines
tertiary all rough, loose dry weather and 4-wheel 0.5
quaternary “cart-tracks”, seismic lines, and all above 0.25
ground transmission and pipe lines
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 20
On the weighted raster map of linear disturbance, a GIS moving window procedure was
performed to determine, for each pixel, the focal mean density in a surrounding circle with an 846
m radius. This equates with 0.87 mi (2.25 km ) and was selected as the analysis window size to
remain consistent with previous applications of the LZP model. The result was converted to a
value depicting linear disturbance in terms of km/2.25 km (mi/mi ). This map was subsequently
reclassified and scored according to four disturbance density classes (Table 4).
Table 5. Human-influence scores assigned to linear disturbance density classes.
Linear Disturbance Density Assigned Score
0 km/2.25 km (0 mi/mi ) 8
0 - 1.6 km/2.25 km (0 - 1 mi/mi ) 5
1.6 - 3.2 km/2.25 km (1 - 2 mi/mi ) 3
> 3.2 km/2.25 km (> 2 mi/mi ) 2
Density was calculated as previous applications of the LZP model (miles/miles ),
however, units are converted to km here.
Visual Cover Layer
I applied 1:20,000 FIP data to derive the model’s visual cover layer. The model identifies
three classes of visual cover for bears: cover, edge, and non-cover. Servheen and Sandstrom
(1993) define grizzly bear hiding cover as “vegetation at sufficient density to hide 90% of an adult
grizzly bear at 200 feet.” This was interpreted from the BC forest inventory as stands at least
three meters in height and with > 30% canopy closure (A. N. Hamilton, WLAP, personal
communication). Edge was defined as areas within 50 meters of cover, allowing for quick escape
to hiding cover, and non-cover was > 50 meters from hiding cover. Reflecting assumed mortality
risk, the model scores cover as 5, edge areas as 3 and non-cover as 2. However, it is widely
held among researchers that the relationship between mortality risk and visual cover is
dependent on human access. For this analysis, I therefore adopted the revision applied by Meitz
(1994). This assumes that lack of visual cover will not influence habitat use by bears beyond 500
meters of roads, and thus assigns a value of 5 to all non-cover areas beyond this limit (Table 5).
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 21
Table 6. Human-influence scores assigned to visual cover classes.
Visual Cover Class Assigned Score
cover > 500 m from roads 5
edge > 500 m from roads 5
non-cover > 500 m from roads 5
cover < 500 m from roads 5
edge < 500 m from roads 3
non-cover < 500 m from roads 2
Food Value Layer
The food values layer is included in the model as a coarse filter to identify lands that are
more likely to support higher or lower concentrations of bear foods, influencing bear movements,
habitat use, and reproductive potential. Food value associated with a given habitat is also
expected to mediate grizzly bear displacement by various human activities (Mace and Waller
1997:120). Consistent with Meitz (1994), all riparian zones of intermittent streams were depicted
by a 25 m buffer, and those of perennial streams were depicted by a 75 m buffer. These riparian
areas were scored as 6 relative to other classes influencing food value.
For this application of the LZP model, I accounted for several other habitat conditions that
are expected to be important in determining food value within coastal ecosystems. Salmon
spawning streams are a major factor in the movements and density of coastal grizzly bears, and
this concentrated food source may moderate the displacement effect of some human activities. I
digitized a 1:250,000 hard-copy map of salmon concentration polygons provided by WLAP, and I
used this to classify streams and water body edges. Sites within 50 m of a watercourse that may
support spawning salmon were scored as 9 if they were associated with “highest” salmon
concentrations and 7 if salmon were considered to be simply “present”. In mountainous
environments, avalanche chutes are also typically highly productive for bear foods due to a
combination of high solar exposure and soil moisture. I identified avalanche chutes using
1:250,000 BTM data. These data did not differentiate between shrub- and forb-dominated
features, and I scored all avalanche chutes as 7. I derived icefields and lakes from the same
dataset and scored them as 1 and 0 respectively, to reflect the very low food value of these
conditions to bears. Within the food value layer, I scored all other habitat conditions as 4 (Table
6). In cases where food value classes overlapped, such as riparian areas and salmon presence,
the higher score was adopted.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 22
Table 7. Scores assigned habitat conditions relative to grizzly bear food value.
Class Assigned Score
Stream reaches with salmon present
in high concentrations 9
Stream reaches with salmon present 7
Avalanche chutes 7
Riparian areas 6
Other habitats 4
Terrain Condition Layer
Previous applications of the LZP model did not consider terrain constraints within the model
structure. However, in a highly mountainous region like the SSPA, slope condition may be an
extremely important factor determining habitat and population connectivity within a landscape.
Therefore, I incorporated into the model a layer accounting for slope condition. I assumed that all
slopes <40% do not hinder bear movements and are equally preferred, and I scored these as 7. I
assumed that there would be a constant increase in energetic cost to bear movement on steeper
slopes, and I scored slopes 40 – 50% as 7, slopes 50 – 60% as 6, and slopes 60 – 70% as 4. I
assumed that slopes steeper than 70% would be much less conducive to bear movement, and I
scored these as 1 (Table 7).
Table 8. Scores assigned to slope classes.
Slope Class Assigned Score
< 50% 7
50 – 60% 6
60 – 70% 4
> 70% 1
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 23
Spatial Scale and Standardization
I depicted each of the above layers at 2 spatial scales. At each scale, I adjusted image
resolution by aggregating data using a GIS moving window routine (Bian 1997). At the broad
scale (level 1), I generalized layers using a circular moving window with a 2.4 km radius. This is
the mean daily movement rate for grizzly bears within the Flathead Valley of the Rocky Mountains
(B. McLellan, MOF, personal communication), and may approximate the minimum scale that
must be considered when defining functional population linkages. At the fine scale (level 2), I
applied a window with a 0.6 km radius, one quarter that of level 1. I then standardized each
layer, such that the mean score across the study area was 0, with a standard deviation of 1.
Habitat Effectiveness and Linkage Zones
At each spatial scale, images reflecting standard scores for each layer were summed to
produce a map of cumulative habitat effectiveness across the study area. In doing so, I weighted
all layers equally, with the exception of terrain condition, which I weighted as 2 to reflect the
greater expected influence of this submodel in defining connectivity. I standardized and
reclassified the results to define relative habitat effectiveness and potential population linkage
across the study area (Table 8). The final result represents a combination of natural and human
factors that may influence grizzly bear movement, displacement and persistence. Potential
linkage zones were defined from level 1 results only, and land ownership, jurisdiction, or zoning
status played no direct role in their identification. Given expected grizzly bear daily movements,
linkage zones may allow bears to traverse and persist for some period of time within landscapes
dominated by potentially high human influence and/or natural constraints.
Table 9. Classification of cumulative habitat effectiveness (HE) scores for linkage zone delineation.
HE Score HE Rating Scale Linkage Zone Potential
<-10 Very low Low
-10 to -1 Low Low
-1 to +1 Moderate Moderate
+1 to +10 High High
>+10 Very high High
Linkage zones defined from level 1 results only.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 24
Results and Discussion
Overview of Linkage Zones
Output from the modified LZP model applied in this analysis is intended to provide decision-
support to direct conservation planning. However, there are several caveats. The model was
developed to predict grizzly bear population linkages and may not necessarily predict core
population areas. Therefore, my interpretation of results is focused on landscapes currently or
potentially subject to at least moderate levels of human development, under the assumption that
surrounding ranges support core areas. Output from this model should not be interpreted to
define core areas. Furthermore, model output is based is on the best available information of
grizzly bear – habitat – human relationships and local conditions within the SSPA. Imprecision
associated with model parameters, model structure, and GIS data can be expected. Therefore,
output is unlikely to be completely veracious and is not intended to provide a reasonable
substitute for intensive field research. Although local knowledge may not everywhere concur with
model output, it must be remembered that known grizzly bear occurrences and movements are
themselves likely to be somewhat biased by human access and bear “sightability”.
Level 1 results highlighted several potential population linkages that may provide
connectivity through human-dominated landscapes now or in the future (Figure 5). Between
Squamish and Whistler, an east-west linkage across Highway 99 was identified associated with
Culliton Creek, about 10 km north of Brackendale. Model output showed no further linkages
associated with the Highway 99 transportation corridor to the town of Pemberton. Given terrain
conditions, results suggest that human development and activity associated with and proximal to
the community of Whistler has fractured the grizzly bear population in that section of the study
area. Across the Lillooet River valley, northwest of Pemberton, a significant linkage zone
appears to connect the Ryan River valley in the southwest to drainages in the vicinity of the upper
Birkenhead and Hurley rivers in the northeast. Further upstream, the valley bottom associated
with the upper Lillooet River appears to provide some potential for population linkage, largely a
result of the relatively low levels of human influence in that part of the SSPA. More extensive
linkage appears to exist further up the Lillooet River valley to the northwest. Further down the
Lillooet River, a potential linkage was identified connecting Miller Creek in the southwest with Owl
Creek in the northeast and drainages further north. Between the towns of Pemberton and Mt.
Currie, a small connection may exist linking drainages of Mount Currie and others within Garibaldi
Provincial Park with the Owl Creek drainage in the north. Further to the east, another minor
linkage around the north end of Lillooet Lake may connect the Ure Creek drainage and Garibaldi
Park in the south with side drainages of Joffre Creek in the north. Further up the Anderson Lake
road, a significant linkage appears to exist across the Gates River valley just south of the
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 25
community of D’Arcy. It connects the drainages of Haylmore and Spruce creeks in the southeast
with the Blackwater Creek drainage and tributaries of upper Birkenhead River in the northwest.
In the southeast of the study area, between Lillooet Lake and Harrison Lake, there is little human
development, and virtually the entire Lillooet Valley bottom constitutes a potential east to west
linkage zone. This may facilitate bear movement between the eastern drainages of Garibaldi
Park and the Stein and Nahatlatch drainages to the northwest.
Culliton Creek Linkage
The Culliton Creek linkage may provide connectivity across the Highway 99 transportation
corridor between Squamish and Whistler. It provides a direct connection between the lower
reaches of the Squamish River, known for supporting high concentrations of spawning salmon,
and habitats in the vicinity of Garibaldi Lake and the upper Pitt River drainage (Figure 6). Grizzly
bears have been known to occur in both areas, although there have been no reports in the vicinity
of Highway 99 within or near the identified linkage. Terrain conditions appear to be highly
conducive to bear movement through this linkage, and effective habitats are quite close on either
side of the Highway corridor (Figure 7). Because the middle portion of the Culliton Creek valley is
relatively narrow and has been logged, human access management in this section is extremely
important to the integrity of this linkage. Similarly, access management on the west side of the
linkage in the vicinity of Pillchuck Creek is important. The linkage takes in part of the Squamish
municipal zone near the junction of Culliton Creek and the Cheakamus River and its persistence
will largely depend on land use decisions by the associated governing body.
Ryan River Linkage
The Ryan River linkage connects drainages of the Ryan River and those further south with
drainages of the upper Birkenhead and Hurley rivers and beyond (Figure 8). It includes part of
the ridge in the Lillooet River valley known as the Camel’s Back. Private lands and associated
agricultural operations within the valley bottom are a source of artificial food attractants for grizzly
bears that has and may continue to result in human-bear conflict and control actions. The main
attractants have been cow carcasses and carrot/parsnip plantations (A. McEwan, PWA, personal
communication). The removal of bears due to inevitable conflict with humans threatens the
integrity of this linkage. The restricted motorized access currently in place within the Ryan River
valley due to private land at the base of this drainage may be extremely important in allowing
bear movements to funnel down this valley. This may be especially true given the restriction of
effective habitat imposed by the very narrow width of the valley and extreme slope grade in the
lower third of the valley between Wasp and Petersen creeks. This is apparent in the discontinuity
of the linkage up the Ryan Valley (Figure 8) and the relatively low levels of effective habitat
associated with the break (Figure 9). Moreover, consistent motorized disturbance within this
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 26
narrow valley may displace resident bears to surrounding valleys, including the Lillooet where
direct bear-human conflict is more likely (A. McEwan, PWA, personal communication). Terrain
conditions are also a factor on the northern side of the Lillooet Valley, and control of motorized
access to road networks near the base of the Hurley Creek road may provide greater movement
options for grizzly bears and other wildlife traversing this linkage. A likely connection is by way of
Wolverine Creek and either side of Tenquille Mountain (Figure 9). There may also be some
movement directly up and down the upper Lillooet Valley. Because there is strong evidence that
drainages on either side of the Ryan River linkage represent core grizzly bear habitat, this
connection likely is extremely important in a regional context. It is highly feasible that bears may
traverse at least one part of the Pemberton Icefield between the Ryan and upper Elaho rivers.
From the Elaho drainage, there are several passes and/or short glacier traverses that link to Toba
and Jervis inlets, which may represent a source for the regional population (A. N. Hamilton,
WLAP, personal communication).
Miller Creek Linkage
This is a small but potentially important linkage spanning the Lillooet River near the junction
of Miller Creek about 3 km northwest of Pemberton. It connects the drainages of Miller and
Pemberton Creeks in the southwest with Owl Creek and additional drainages to the northeast
(Figure 8). The upper reaches of both Miller and Pemberton creeks are associated with
moderate to high levels of effective habitat, and the Pemberton Creek drainage is subject to an
Integrated Watershed Management Plan. In the late 60s and early 70s, grizzly bears consistently
occurred in the headwaters of Miller Creek, but have not been observed in recent years (A.
McEwan, PWA, personal communication). Lowest habitat effectiveness levels occur near the
base of Miller Creek and on the southwest side of the Lillooet Valley road, while habitat
effectiveness is relatively high on the northeast side of the road (Figure 9). Much of the valley
bottom within this linkage is in private ownership. Access to the Miller Creek watershed is
controlled by a gate on private land near the base of the creek in the Lillooet Valley. However, an
improved road and some increased traffic into the drainage can be expected from a small-scale
hydroelectric project that has recently been approved (J. Roberts, WLAP, personal
Mount Currie Linkage
This is a minor linkage that connects drainages of Mount Currie in the south with those of
Owl Creek and beyond in the north. It constricts to a fairly narrow corridor to the north of the
highway (Figure 8), within which a 5 year old male grizzly bear was removed in July of 1989.
Much of this linkage traverses private land, which again poses a challenge to maintaining or
enhancing the connection.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 27
Lillooet Lake Linkage
This linkage connects Ure Creek in Garibaldi Park and tributaries of Gravel Creek to the
southwest with drainages of Joffre Creek in the northeast (Figure 8). The valley-bottom
connection is around the northwest end of Lillooet Lake. This connection is somewhat restricted
by human dwellings near the lake itself and steeper slopes directly north of Lillooet Lake (Figure
9). At the end of Lillooet Lake, the linkage spans both private and Indian reserve land, and
maintaining the connection will require coordination and cooperation with the Mt. Currie band and
other landowners. Joffre Creek is also associated with the well-traveled Duffy Lake Road
(Highway 99), and access management of associated spur roads should be a management
consideration for this linkage. This is especially important given that northwest and southeast
bear movements among important habitats within the Lillooet Ranges will span the Joffre Creek
This linkage spans the Gates River Valley about 2 km southwest of the town of D’Arcy. It
connects the Haylmore and Spruce Creek drainages in the southeast with the Blackwater
drainage and tributaries of upper Birkenhead River in the northwest (Figure 8). This appears to
be a substantial linkage and may be extremely important for grizzly bears in this region. Grizzly
bear occurrences have been recorded within the linkage and within drainages on either side,
particularly in the upper Haylmore Creek (Figure 5). Although there is little road access southeast
of the zone, Haylmore Creek is currently accessible by 4-wheel drive and the upper reaches can
be accessed by ATV. The timber licensee that operates within this drainage is currently
developing forestry roads into all 3 main reaches of Haylmore Creek (J. Roberts, WLAP, personal
communication). To maintain the effectiveness of highly important grizzly bear habitat there, it is
critical that motorized public access to these roads be strictly controlled using the gate installed at
Common Johnny Creek. Northeast of the linkage, there is a 10 km all weather access road to
Birkenhead Lake Provincial Park. Camping facilities are located at Birkenhead Lake and at
Blackwater Lake about 5 km up (Figure 9).
Maintaining the integrity of this linkage will require that dispersed motorized access be
controlled within all associated drainages, and that human development and bear attractants be
minimized within the Gates Valley bottom, from the Gates railway siding to D’Arcy. This is of
particular importance for the small hamlet of Devine which falls directly within the linkage.
Special measures are already being taken to protect the designated community watershed
defined by the Spruce Creek drainage, and the Spotted Owl Special Resource Management
Zone within the Blackwater and Birkenhead valleys. This will result in reduced forestry
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 28
development resulting in less road development, and restricted motorized access to roads that
are in place.
Lillooet to Harrison Lakes Linkage
A large linkage zone occurs between Lillooet Lake and Harrison Lake across the Lillooet
River valley (Figure 10). It may provide functional connectivity between the eastern drainages of
Garibaldi Provincial Park and those of the Lillooet Ranges to the northeast. Central to this
linkage is the reach of the Lillooet River between Rogers and Sloquet creeks that supports high
concentrations of spawning salmon. There are several key watersheds of which this larger
linkage is comprised. These include Billygoat, Tuwasus, Snowcap, and Sloquet creeks in the
southwest, and Rogers, Gowan, and Douglas creeks to the northeast. Rogers and Gowan
creeks may provide very important connections to the upper reaches of the Nahatlatch River and
Stein Valley Provincial Park. However, the upper reaches of these drainages are associated with
open road densities, which, within relatively narrow, steep-sided valleys, may substantially
decrease habitat effectiveness and linkage potential (Figure 11). On the southwest side of the
linkage, the lower reaches of Tuwasus and Snowcap creeks are within a designated Special
Resource Management Zone for spotted owls, which will limit forestry development and
associated road access.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 29
Figure 5. Predicted grizzly bear population linkage and focal areas across the Sea to Sky Regional Planning
Area in southwestern British Columbia. Grizzly bear occurrences correspond to locations of confirmed
sightings, sign, or control actions dating from 1975; these were derived from a preliminary databased
assembled by S. Rochetta (WLAP, Squamish) that was not based on exhaustive survey and will continually
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 30
Figure 6. Identified linkage zones merged with Landsat 7 panchromatic satellite imagery for the southwest focal area.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 31
Figure 7. Level 2 habitat effectiveness values merged with hillshaded image for the southwest focal area.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 32
Figure 8. Identified linkage zones merged with Landsat 7 panchromatic satellite imagery for the north focal area.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 33
Figure 9. Level 2 habitat effectiveness values merged with hillshaded image for the north focal area.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 34
Figure 10. Identified linkage zones merged with Landsat 7 panchromatic satellite imagery for the
southeast focal area.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 35
Figure 11. Level 2 habitat effectiveness values merged with hillshaded image for the southeast focal
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 36
Management and Research Recommendations
Based on female home range data from western Canada and the Northwestern United
States, Woodroffe and Ginsberg (1998) predicted that grizzly bear conservation requires a
minimum reserve size of ≥3900 km . Clearly, maintaining viable grizzly bear populations in the
southern coastal mountains requires special considerations in all landscapes, beyond those that
may currently function as core/security areas. At the regional level, one report has suggested
that existing logging road densities, coupled with high natural fragmentation, are well beyond
levels at which normal grizzly bear populations can persist (McTavish and McCrory 1998).
Notwithstanding inherent habitat value, watersheds in which motorized access is or can be
restricted can provide necessary security areas for bears and other species. In some cases,
such as the upper Elaho Valley, these landscapes may also represent important nodes facilitating
population flow within the SSPA and with external core population areas, such as Toba Inlet.
Retaining or restoring the functional connectivity of identified grizzly bear population
linkages requires that human influence be managed at both broad and fine spatial scales. Based
on modeling criteria and assumptions, human activity within a 2.4 km radius of a defined linkage
zone can affect the integrity of the linkage itself, with impacts being potentially greatest toward the
center of the zone. In several cases, identified linkages are associated with private or First
Nations reserve land. Within grizzly bear range, habitats on or adjacent to private lands are
generally associated with higher densities of humans and roads, poor sanitation, and artificial
attractants. Such lands are typically considered population “sinks” because they are associated
with a much higher incidence of bear-human conflict resulting in bear mortality or translocation
(Mace et al. 1996). In the SSPA, it is therefore imperative that landowners and native bands
responsible for the stewardship of lands within or proximal to identified linkage zones recognize
the implications of their activities and development to the regional conservation of grizzly bears
and other wide-ranging species within the larger region. A functioning linkage requires that
grizzly bears are actually able to persist within it. This requires the management of artificial
attractants, human settlement, habitat conditions, and motorized access within and proximal to
each zone. Natural conditions, including terrain, limit movement options for wildlife. Where such
constrictions exist, such as in narrow, steep-sided valleys, permanent or seasonal closure of road
networks to motorized access may be key to facilitating movement by wide-ranging species.
Although defined linkages represent obvious priorities for conservation action, the measures to
conserve grizzly bears will be of greatest benefit to this and many other species if they are not
restricted to specific linkage zones but applied over larger landscapes.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 37
Unfortunately, I was not able to extend my analysis into the Hurley River, Anderson Lake,
or Cayoosh drainages. Inherent habitat quality appears to be relatively high there, as indicated
by habitat effectiveness results within my study area, the distribution of known grizzly bear
occurrences, and field assessment (A. N. Hamilton, WLAP, personal communication). These
landscapes may represent a core population area within the region, and any development
proposal in associated watersheds should be carefully scrutinized to ensure that effective habitat
and the integrity of population core/security areas or linkages is not compromised. Moreover,
given the human population within greater Vancouver, and projections for its growth, any
development in and adjacent to the SSPA will inevitably result in “spin-off” development, greatly
increasing human access and settlement. Therefore, it is imperative that new developments be
evaluated in the context of cumulative effects at the regional scale.
Despite mounting land use pressures, relatively little is known of grizzly bear occurrence,
density, distribution, or movements within the SSPA. Although based on demonstrated
relationships of grizzly bear occurrence with habitat conditions and human activity, the approach I
have adopted for this analysis was based on model parameters and structure that is largely
subjective. Given the “threatened” status of grizzly bears within the SSPA and the current and
projected multiple land use pressures, I recommend systematic field sampling of grizzly bear
occurrence within the study area to feed into land management and planning decisions and to
monitor population and distribution trends. For this, a non-invasive hair-snag sampling approach
(Woods et al. 1999) would be appropriate. Using these detection and DNA fingerprinting
methods, I recommend a nested sampling design that is extensive across the entire study area,
but progressively more intensive within potential core population areas and landscapes within
which functional linkages may occur. This research design will facilitate grizzly bear population
estimation and trend monitoring (Boulanger et al. 2001), empirical spatial modeling of the regional
population density and distribution, including core, periphery, and linkage zones (Apps et al.
2001), and the degree of genetic mixing within the population (Proctor et al. 2001).
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 38
Apps, C. 1997. Identification of grizzly bear linkage zones along the Highway 3 corridor of
southeastern British Columbia and southwestern Alberta. Prepared for BC Ministry of
Environment, Lands and Parks, and World Wildlife Fund Canada and USA. Aspen Wildlife
Research, Calgary, Alberta, Canada.
Apps, C. D., B. N. McLellan, J. G. Woods, and M. F. Proctor. 2001. Distribution of a grizzly bear
population relative to habitat and human influence in southeastern British Columbia.
Journal of Wildlife Management. Submitted.
Archibald, W. R., R. Ellis, and A. N. Hamilton. 1986. Responses of grizzly bears to logging truck
traffic in the Kimsquit River valley, British Columbia. International Conference on Bear
Research and Management 7:251-257.
Banci, V., D. A. Demarchi, and W. R. Archibald. 1994. Evaluation of the population status of
grizzly bears in Canada. International Conference on Bear Research and Management
Bian, L. 1997. The multiscale nature of spatial data in scaling up environmental models. Pages
13-26 in D. A. Quattrochi and M. F. Goodchild, editors. Scale in remote sensing and GIS.
Lewis Publishers, New York, New York, USA.
Boulanger, J., J. Woods, B. N. McLellan, and M. F. Proctor. 2001. An investigation of sampling
design and capture probability bias in DNA based mark-recapture estimates of grizzly bear
populations. Journal of Wildlife Management. Submitted.
Cannings, S.G., Ramsay, L.R., Fraser, D.F., and Fraker, M.A. 1999. Rare amphibians, reptiles
and mammals of British Columbia. Ministry of Environment, Lands and Parks, Victoria,
British Columbia, Canada.
Demarchi, D. A. 1996. An introduction to the ecoregions of British Columbia. Wildlife Branch,
Ministry of Environment, Lands and Parks, Victoria, British Columbia, Canada.
Fuhr, B. L., and D. A. Demarchi. 1990. A methodology for grizzly bear habitat assessment in
British Columbia. Habitat Inventory Section, Wildlife Branch, Ministry of Environment,
Victoria, British Columbia, Canada.
Geographic Data BC. 1996. Gridded DEM specification, release 1.1. Ministry of Environment,
Lands and Parks, Victoria, British Columbia, Canada.
Gibeau, M. L. 2000. A conservation biology approach to management of grizzly bears in Banff
National Park, Alberta. Ph.D. Dissertation. Resources and the Environment Program,
University of Calgary, Calgary, Alberta, Canada.
Gibeau, M., S. Herrero, J. L. Kansas, and B. Benn. 1996. Grizzly bear population and habitat
status, Banff National Park. Chapter 6 in J. Green, C. Pacas, S. Bayley, and L. Cornwell,
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 39
editors. Ecological outlooks project: a cumulative effects assessment and futures outlook
of the Banff Bow Valley. Prepared for the Banff Bow Valley Study, Banff, Department of
Canadian Heritage, Ottawa, Ontario.
Gilpin, M. E., and M. E. Soulè. 1986. Minimum viable populations: processes of species
extinction. Pages 13-34 in M. E. Soulè, editor. Conservation biology: the science of
scarcity and diversity. Sinauer Associates Inc., Sunderland, Massachusetts, USA.
Holman, G., and S. Nicol. 2001. Socio-economic base case draft report for discussion. Sea to
Sky Land and Resource Management Plan, Government of British Columbia, Victoria,
British Columbia, Canada.
Kehoe, N.M. 1995. Grizzly bear distribution in the north fork of the Flathead River valley: a test
of the linkage zone prediction model. M.S. thesis. University of Montana, Missoula,
Mace, R. D., and J. S. Waller, editors. 1997. Final report: grizzly bear ecology in the Swan
Mountains. Montana Fish, Wildlife, and Parks, Helena, Montana, USA.
Mace, R. D., J. S. Waller, T. L. Manley, L. J. Lyon, and H. Zuuring. 1996. Relationships among
grizzly bears, roads, and habitat in the Swan Mountains, Montana. Journal of Applied
Mattson, D. J. 1990. Human impacts on bear habitat use. International Conference on Bear
Research and Management 8:57-64.
Mattson, D. J., 1993. Background and proposed standards for managing grizzly bear habitat
security in the Yellowstone ecosystem. Cooperative Park Studies Unit report. University of
Idaho, Moscow, Idaho, USA.
Mattson, D. J., B. M. Blanchard, and R. R. Knight. 1992. Yellowstone grizzly bear mortality,
human habituation, and whitebark seed crops. Journal of Wildlife Management 56:432-
McLellan, B. N. 1990. Relationships between human industrial activity and grizzly bears.
International Conference on Bear Research and Management 8:57-64.
McLellan, B. N. 1998. Maintaining viability of brown bears along the southern fringe of their
distribution. Ursus 10:607-611.
McLellan, B. N., and F. W. Hovey. 2001. Natal dispersal of grizzly bears. Canadian Journal of
McLellan, B. N., F. W. Hovey, R. D. Mace, J. G. Woods, D. W. Carney, M. L. Gibeau, W. L.
Wakkinen, and W. F. Kasworm. 1999. Rates and causes of grizzly bear mortality in the
interior mountains of British Columbia, Alberta, Montana, Washington, and Idaho. Journal
of Wildlife Management 63:911-920.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 40
McTavish, C., and W. McCrory. 1998. Grizzly/black bear habitat assessment, Stoltmann
Wilderness, Elaho River drainage, British Columbia, Canada. Submitted to Western
Canada Wilderness Committee, Vancouver, British Columbia, Canada.
Meidinger, D. V., and J. Pojar. 1991. Ecosystems of British Columbia. Special Report Series
No. 4. Ministry of Forests, Victoria, British Columbia, Canada.
Meitz, S. N. 1994. Linkage zone identification and evaluation of management options for grizzly
bears in the Evaro Hill area. M.S. thesis. University of Montana, Missoula, Montana, USA.
Ministry of Environment, Lands and Parks. 1995. British Columbia grizzly bear conservation
strategy. British Columbia Ministry of Environment, Lands and Parks, Victoria, British
North Cascades Grizzly Bear Recovery Team. 2001. Recovery plan for grizzly bears in the
North Cascades of British Columbia. Consensus recommendation draft for consultation.
Ministry of Environment, Lands and Parks, Victoria, British Columbia, Canada.
Pelletier, K. 1995. Managing private land in Swan Valley linkage zones for grizzly bears and
other wildlife. Swan Linkage Zone Working Group, U.S. Fish and Wildlife Service,
Missoula, Montana, USA.
Proctor, M. L., B. N. McLellan, and C. Strobeck. 2001. Population fragmentation of grizzly bears
in southeastern British Columbia, Canada. Ursus. Submitted.
Resources Inventory Branch. 1995. Relational data dictionary (RDD) 2.0. Ministry of Forests,
Victoria, British Columbia, Canada.
Resources Inventory Committee. 1997. Standards for broad terrestrial ecosystem classification
and mapping for British Columbia. Ministry of Environment, Lands and Parks, Victoria,
British Columbia, Canda.
Servheen, C. 1990. The status and conservation of bears of the world. International Conference
on Bear Research and Management, Monograph Series 2.
Servheen, C., and P. Sandstrom. 1993. Ecosystem management and linkage zones for grizzly
bears and other large carnivores in the northern Rocky Mountains of Montana and Idaho.
Endangered Species Technical Bulletin 18:10-13.
Surveys and Resource Mapping Branch. 1992. British Columbia specifications and guidelines
for geomatics. Content Series Volume 3: digital baseline mapping at 1:20,000 release 2.0.
Ministry of Environment, Lands and Parks, Victoria, British Columbia, Canada.
Surveys and Resource Mapping Branch. 1995. Baseline thematic mapping; present land use
mapping at 1:250,000, release 1.0. Ministry of Environment, Lands and Parks, Victoria,
British Columbia, Canada.
Weaver, J. L., P. C. Paquet, and L. S. Ruggiero. 1996. Resilience and conservation of large
carnivores in the Rocky Mountains. Conservation Biology 10:964-976.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 41
Woodroffe, R., and J. R. Ginsberg. 1998. Edge effects and the extinction of populations inside
protected areas. Science 280:2126-2128.
Woods, J. G., D. Paetkau, D. Lewis, B. N. McLellan, M. Proctor, and C. Strobeck. 1999. Genetic
tagging of free-ranging black and brown bears. Wildlife Society Bulletin 27:616-627.
Forbes, R. D. (Bob). Regional Wildlife Section Head. Fish and Wildlife. Ministry of Water, Land
and Air Protection. Cranbrook, British Columbia.
Hamilton, Anthony E. Wildlife Research Biologist. Research and Conservation Section. Ministry
of Water, Land and Air Protection. Victoria, British Columbia.
Jury, Doug. Wildlife Biologist. Fish, Wildlife and Habitat. Ministry of Water, Land and Air
Protection. Kamloops, British Columbia.
McEwan, Al. President. Pemberton Wildlife Association. Pemberton, British Columbia.
McLellan, Bruce. Senior Wildlife Habitat Ecologist. Wildlife Habitat and Range. Ministry of
Forests. Revelstoke, British Columbia.
Roberts, Jim. Forest Ecosystem Specialist. Conservation Officer Service. Ministry of Water,
Land and Air Protection. Squamish, British Columbia.
Rochetta, Steve. Habitat Protection Officer. Conservation Officer Service. Ministry of Water,
Land and Air Protection. Squamish, British Columbia.
Grizzly Bear Linkage Zones in the Sea to Sky Planning Area • C. Apps • August 2001 42