Cisco distributed freshwater fish across all of North

Cisco (Coregonus
artedi) in the family Salmonidae, are midwater to benthic feeding fish. Cisco
play an important ecological role in the food web as they are a major secondary
consumer and feed mainly on plankton, insects and mollusks in their adult life
(book). Lake Trout, Yellow Walleye and Northern Pike are examples of species
that prey upon Cisco. This species also has significant economic importance to local
commercial fisheries throughout Manitoba. Cisco’s are one of the most widely
distributed freshwater fish across all of North America (Scott and Crossman,
1973). Similar to other fish communities, Cisco’s are affected by energy inputs
into the lake, habitat and environmental factors, nutrients and interactions
between other species and communities (Evans et al., 1987).

 

Lake Winnipeg has three distinct
geographic regions including the channel and the north and south basin, which have
differing morphology including characteristics such as depths, surface
temperature and water clarity. The north basin is more remote and less subject
to anthropogenic influences which heavily affect the opposing south basin.
Increasing human populations and agriculture use in the water catchment area
has led to eutrophication in the lake (Schindler et al., 2002). The water in Lake
Winnipeg is well mixed vertically most of the time and considered isothermal (Brunskill
et al., 1980) due to large amounts of shallow water and frequent strong wind energy
on the lake. Rare stratification of certain areas in the summer and winter has
been reported by Wassenaar (2012) which occurs when oxygen levels decline.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

Lake
Winnipeg has gone through large changes in its fish composition in the past several
decades with the invasions of Rainbow Smelt (Osmerus mordax) and White Bass in the early 1990’s (Campbell et al., 1991). Rainbow smelt dominated the
north basin for many years (Remnant, 1991), but their population has severely
crashed, and in 2017 not a single smelt was caught during the three sampling surveys
onboard the Lake Winnipeg Research Consortium Vessel the M/V Namao. Rainbow
smelt have additionally had large negative effects on Cisco populations in
other northern lakes also (Hrabik
et al. 1998). Zebra mussels (Dreissena
polymorpha) invaded Lake Winnipeg in 2013 and have since
become widely abundant and well established. Zebra mussels alter the nutrient
and energy distribution and contribute to benthification which redistributes
energy from pelagic to benthic pathways (Miehls et al 2009). As Zebra mussels
filter the water and consume plankton, this directly increases the water clarity
(MacIsaac, 1996). In the north basin the water clarity has also increased due
to Cedar Lake Grand Rapids impoundment which is a hydroelectric dam, directly
reducing the amount of sediments made available to the lake (Patalas and Salki,
1992). Lake Winnipeg has also had high increases of nitrogen and phosphorus creating nutrient overloading. Between
1990 and 2000 phosphorus doubled (McCullough et al., 2012), leading to frequent algal
blooms.

Long term monitoring programs on
Lake Winnipeg are critical, since having data from before, during and after the
introduction of invasive species can help demonstrate the impacts on the
ecosystem and fish communities. The main objective of this paper is to evaluate
the environmental factors and their relation to the growth of Cisco as a
measured by average fork length in Lake Winnipeg between 2002-2017.

 

 

Methods:

 

Study Area: 

Lake Winnipeg is located between N
50º 23′ to N 53º 50′ in Manitoba, Canada (Brunskill et al., 1980), and with a
surface area of 23, 759km2 is considered the tenth-largest
freshwater lake in the world. There are three distinct areas in Lake Winnipeg, the
south basin, channel, and north basin, all containing different morphology and limnology
characteristics.  Lake Winnipeg is a shallow
lake with the mean depth of 9.7 m in the south basin, 7.2 m in the channel, and
13.3 m in the north basin (Lumb et al., 2011).

 

Fish Collection:

Fish
collections took place during the spring (May- June), summer (July August) and
fall (September-October) from 2012-2017. 
Lake wide midwater trawls were conducted by fisheries personnel onboard
the M/V Namao (Lake Winnipeg Research Consortium, LWRC) at existing long term
water quality monitoring stations. The existing stations collected various
parameters including dissolved oxygen, turbidity, total suspended solids,
nitrogen and phosphorus concentrations. Each trawl was deployed while trawling
into the station at a towing speed of around 3.9km/h for 30 minutes. While the Namao
was over the station, other parameters such as, latitude and longitude, wind
speed, air and water temperature(ºC), lake depth (m), secchi depth (m), along with
the flowmeter and odometer readings were additionally recorded for the trawl. The
dimensions of the trawl were 3 m2 and 10.8 m long with mesh sizes decreasing
from 76.2 mm at the mouth end to 19.1 mm at the opposing cod end. The cod end was
made of 1.22m PVC pipe, 114.3 mm in diameter with a screw out plug for easy
fish access. Once the trawl was pulled up all fish species over 300 mm were
promptly identified, measured, weighed, and released back into the lake.
Walleye and Sauger under 300 mm were put in a water mixture containing MS-222
and baking soda for humane euthanization and later preserved in formalin. All
the remaining fish were sorted by species then stored in whirlpack bags and promptly
frozen. After two 42 ounce (1.25L) whirlpack bags were full of the same
species, the remaining fish were bulk weighed and released.

 

Trawl Depth:

Trawl depth
was randomly selected from three possibilities: surface, midwater, or
deepwater.  Surface trawls sample the
upper 2.5 m of the water column, where the upper trawl beam is lowered just
below the surface of the water. Midwater trawls in the south basin and channel
sample 2.7-5.3 m below the water, whereas in the north basin they sample 6.4-9m
below the surface. Deepwater trawls roughly sample 9.8-12.4 m below the and as
such could not be used in the south basin, due to depth restrictions. Certain
other stations throughout the lake had other depths restrictions which were
monitored by the captain. Pressure sensors are attached to the top and bottom
beam of the trawls and record depth and temperature every three minutes. A
flowmeter was attached to the mouth of the allowing for the calculation of water
volume flowing through the trawl. A GPS was used to record the coordinates of
the start and end position on the trawl.

 

Laboratory Fish Processing

After
the fish were slowly thawed in water, species identification were confirmed,
and the fish were sorted into appropriate size classes (young of year, small,
medium, large) which varied based upon species size classes. A subsample of 100
fish from each of the same size classes was fully sampled by measuring the fork
length (mm) and weight (g). The remaining fish were enumerated, bulk weighed
and recorded. Before sampling the formalin preserved Yellow Walleye and Sauger,
the fish were soaked in water for three days and finally preserved in 70% ethanol
until processing. The identification of Yellow Walleye and Sauger was mainly
based upon pyloric ceacae length and numbers. Pigmentation on the dorsal surface
of the head was also recorded, as many Yellow Walleye have dark melanophores in
this location where Sauger in contrast are rather pale and unpigmented. The
remaining fish species were identified by following an identification key found
in the Freshwater Fishes of Manitoba book.

 

Statistical Analysis:

Data
including station latitude and longitude, season, year, secchi depth (m),
surface temperature (ºC), trawl type, average fish length (mm) by station and
the count of fish at each station for all Cisco from 2002-2017 was compiled on
Microsoft Access. Data was explored by plotting histograms to get a better
understanding of the frequency and range of fish fork lengths.  Using the program R, certain data categories
such as secchi depth and surface temperature were transformed by taking the log
to meet normality requirements. Data was then selected based on the fish count
per station, and only samples with 30 or more individual fish were used
accounting for 641/1558 of the observations. The remaining 641 data points were
scattered throughout the three distinct areas in the lake, ensuring full lake
coverage was present in the data. Other categories such as surface temperature,
latitude, secchi, and year were scaled to ensure they all had a reasonable
reference point. The data was then put into a General Linear Model (GLM), and
the function Step was used to create a better combination of variables leading
to the greatest effect on average Cisco length.