fir (Abies lasiocarpa) (Holland and Coen 1983). Vegetation within the CPR
right-of-way (ROW) was kept in an early successional stage by mechanical treatments
and consisted primarily of herbaceous vegetation (grasses, sedges and rushes), dandelions (Taraxacum spp.), bearberry (Arctostaphylos uva-ursi) and horsetail
(Equisetum spp.). Numerous species of shrub including buffaloberry (Sheperdia
canadensis) and tree species including lodgepole pine and white fir were present at
reduced heights (<2 m). Vegetation between the rails was almost non-existent due to
herbicide treatments and the rocky substrate. Where there was vegetation between the
rails there was small herbaceous growth of native and non-native grasses or young
sprouts of train-spilled agricultural products (cereal grains).
Data Collected
Wildlife Strike Records
Counts of train-struck ungulates and bears were obtained from Parks Canada for a
21-year period (1989–2009) (Parks Canada, unpublished data, 2010). These records
included all strikes reported by CPR and Parks Canada personnel. Strikes were
visited and removed only by Parks Canada personnel to avoid double counting.
Records after 2001 were spatially referenced with a Garmin
® global positioning
system (GPS), but before 2001, records were referenced to permanent mile-marker
posts or to geographic features along the CPR. As a result some strike records were
spatially inaccurate by as much as 1.6 km (1.0 mile); thus, an error of plus or minus
1.6 km was assumed for all records. A spatial data layer representing the railroad
was acquired from CPR and then imported into a geographic information system
(GIS). The spatial layer was separated into 30 roughly equal segments 4.86 km
(3 miles) in long. The segment length was based on a conservative estimate of the
minimum mapping accuracy of the wildlife strike records, which were tallied to the
appropriate 4.86 km (3 mile) segment in a GIS.
Relative Abundance Data
Relative abundance data were collected by sampling transects oriented parallel and
perpendicular to the CPR (hereinafter called “rail bed transects” and “perpendicular
transects,” respectively). Both methods were binned to 4.86 km long rail segments,
ensuring that data were analyzed at the same spatial scale. Sampling was carried
during the summers of 2008–2009 by the same observer. A total of 313 km of rail
bed transects and 129 perpendicular transects were surveyed in 2008 and 2009.
The relative abundance of bears was sampled using both methods. However,
ungulates were sampled only using perpendicular transects that allowed for an
accurate representation of ungulate distribution throughout the year.
Parallel transects were centered directly on the rail bed. Segments were treated as
a strip transect 3 m wide and 4.86 km long. Locations of bear scats were recorded
9 Relative Risk and Variables Associated with Bear and Ungulate …
139
right-of-way (ROW) was kept in an early successional stage by mechanical treatments
and consisted primarily of herbaceous vegetation (grasses, sedges and rushes), dandelions (Taraxacum spp.), bearberry (Arctostaphylos uva-ursi) and horsetail
(Equisetum spp.). Numerous species of shrub including buffaloberry (Sheperdia
canadensis) and tree species including lodgepole pine and white fir were present at
reduced heights (<2 m). Vegetation between the rails was almost non-existent due to
herbicide treatments and the rocky substrate. Where there was vegetation between the
rails there was small herbaceous growth of native and non-native grasses or young
sprouts of train-spilled agricultural products (cereal grains).
Data Collected
Wildlife Strike Records
Counts of train-struck ungulates and bears were obtained from Parks Canada for a
21-year period (1989–2009) (Parks Canada, unpublished data, 2010). These records
included all strikes reported by CPR and Parks Canada personnel. Strikes were
visited and removed only by Parks Canada personnel to avoid double counting.
Records after 2001 were spatially referenced with a Garmin
® global positioning
system (GPS), but before 2001, records were referenced to permanent mile-marker
posts or to geographic features along the CPR. As a result some strike records were
spatially inaccurate by as much as 1.6 km (1.0 mile); thus, an error of plus or minus
1.6 km was assumed for all records. A spatial data layer representing the railroad
was acquired from CPR and then imported into a geographic information system
(GIS). The spatial layer was separated into 30 roughly equal segments 4.86 km
(3 miles) in long. The segment length was based on a conservative estimate of the
minimum mapping accuracy of the wildlife strike records, which were tallied to the
appropriate 4.86 km (3 mile) segment in a GIS.
Relative Abundance Data
Relative abundance data were collected by sampling transects oriented parallel and
perpendicular to the CPR (hereinafter called “rail bed transects” and “perpendicular
transects,” respectively). Both methods were binned to 4.86 km long rail segments,
ensuring that data were analyzed at the same spatial scale. Sampling was carried
during the summers of 2008–2009 by the same observer. A total of 313 km of rail
bed transects and 129 perpendicular transects were surveyed in 2008 and 2009.
The relative abundance of bears was sampled using both methods. However,
ungulates were sampled only using perpendicular transects that allowed for an
accurate representation of ungulate distribution throughout the year.
Parallel transects were centered directly on the rail bed. Segments were treated as
a strip transect 3 m wide and 4.86 km long. Locations of bear scats were recorded
9 Relative Risk and Variables Associated with Bear and Ungulate …
139