Trends in Distribution and Abundance of Westslope Cutthroat Trout and Sedimentation in the Upper Oldman River Watershed, 2015-2016
Author(s)
Jason Blackburn, Kevin Fitzsimmons, Brad Hurkett, and Logan Redman
Summary
Westslope cutthroat trout (WSCT) is considered Threatened in Alberta under Canada’s Species at
Risk Act. Long-term survival of the species requires identification, protection and restoration of
strongholds where genetically pure populations remain. One of the last remaining strongholds
for the species is in the upper Oldman River watershed, which has undergone varying intensities
of landscape disturbance. Fine sediment deposition from surrounding land-use activity has been
identified as a key threat and major limiting factor affecting recovery of Alberta’s WSCT
populations. We completed a comprehensive, two-year study to document abundance,
population structure and distribution of genetically pure WSCT relative to trends in
sedimentation.
We surveyed more than 25 km of stream at 73 randomly selected sampling sites along the
mainstems of 18 streams, collecting fish, sediment, habitat, stream channel and pool information.
We collected fish size, abundance and distribution information using backpack and totebarge
electrofishing methods; measured sediment quantity and composition; and performed pool
counts to determine pool frequency. We constructed generalized additive models of sediment
quantity, and longitudinal WSCT abundance by maturity class, using single-pass electrofishing
data corrected with capture-mark-recapture derived capture-efficiencies. In all, we captured
3,824 WSCT and collected 1,151 tissue samples for genetic (DNA) analysis.
We recorded the highest catch rates of WSCT (all fish ≥70 mm fork length; FL) in Vicary Creek
(355 fish/km), juveniles (≥70 – 149 mm FL) in Pasque Creek (321 fish/km) and adults (≥150 mm)
in Ridge Creek (247 fish/km). Rearing streams, where catches were dominated by juvenile fish
and where fish sizes were the smallest, were Pasque (91% juveniles, 83 mm FL), Oyster (91%
juveniles, 93 mm FL), Speers (84% juveniles, 110 mm FL), and Beaver (79% juveniles, 115 mm FL)
creeks. Streams where catches were dominated by adult fish and median fish size was the largest
were Deep (75% adults, 189 mm FL), Ridge (71% adults, 172 mm FL) and Lyall (71% adults,
160 mm FL) creeks.
Of the watersheds suitable for modelling longitudinal abundance of WSCT, the highest mainstem
abundance was in Vicary Creek for both total fish (n = 20,930) and juveniles (n = 14,344), exceeding
that of Racehorse, South Racehorse, North Racehorse, Dutch and Hidden creeks combined. The
highest adult abundance occurred in White Creek (n = 9,012), exceeding that of Racehorse, South
Racehorse, North Racehorse and Dutch creeks combined.
Streams among those with both the highest proportion of fine sediment fractions <6 and <2 mm
and median sediment volumes included Pasque, Speers, Oyster, Deep and Ridge creeks. Streams
among those with the lowest fine sediment proportions and volumes included Daisy, Mean,
Racehorse, North Racehorse, Dutch, Beehive and Hidden creeks. Scour-pool frequency was
variable relative to sediment quantity; however, it was highest in many of the streams where
median sediment volume was also highest. The highest scour-pool frequencies occurred in White
Creek (36 pools/km), followed by Ridge and Oyster creeks (33 pools/km) and Pasque Creek
(22 pools/km), which were also some of the streams with the highest WSCT catch rates.
The relationship between sediment quantity and fish population structure within the watersheds
in the study area was unclear. Pasque, Speers and Oyster creeks had both the highest proportions
of fine sediment as well as juvenile fish. Conversely, Ridge and Deep creeks had the highest
proportion of adult-sized fish but were still among the streams with the highest sediment levels.
Through concurrent modelling of longitudinal sediment quantity, and WSCT abundance, we
revealed that watersheds with the highest catch rates had a trend of decreasing deposited
sediment quantity in a downstream direction, whereas those with the lowest catch rates had a
trend of increasing deposited sediment quantity in a downstream direction. Variables such as
reach-scale channel morphology, stream gradient and elevation may have confounded
interpretation of relationships between sediment quantity and fish abundance by differentially
altering sediment transport, retention and/or settling rates. For example, deep bedrock pools and
high-gradient step-pool sequences retained fish but did not create scour-pools from which to
measure transported sediment.
Pool availability may also have confounded interpretation of maturity-class composition relative
to sediment. Both Vicary and White creeks had similar measures of fine deposited sediment;
however, White Creek, which had the most scour-pools per kilometre, had an inverse
longitudinal relationship between adult and juvenile abundances, whereas Vicary Creek, which
had fewer scour-pools, had a disproportionate abundance of juveniles.
Proximity of disturbances to the stream channel may be a key variable influencing WSCT
longitudinal population structure, given adult abundance in South Racehorse Creek plummeted
sharply where the stream closely parallels a main road and access is increased.
The interactions between fine deposited sediment, stream morphology, WSCT abundance and
population structure were complex and will require considerable further analysis to better
understand underlying mechanisms that impact WSCT populations in the upper Oldman River
watershed.