Waterbodies with Information on Mercury in Fish...discuss the evolution of fish mercury...

Waterbodies with Information on Mercury in Fish

RCEA Area 2

Government of Canada, Province of Manitoba, North/South Consultants andManitoba Hydro.

North/South Consultants

1.0

27-AUG-15

LegendFish Mercury Information

Focal Waterbody

RCEA Region of Interest

Generating Station (Existing)

Generating Station (Under Construction)

Generating Station (Potential)

Highway

First Nation Reserve

08-DEC-15

ButnauLake

HUDSONBAY

Gull Lake

Split Lake

River

Stephens

Lake

Assean Lake

Clark Lake

Haye

s

Moose NoseLake

KelseyG.S.

Long Spruce G.S.

Limestone G.S.Kettle

G.S.

KeeyaskG.S.

ConawapaG.S.

Gillam Fox Lake Cree NationA Kwis Ki Mahka Reserve

York Factory First NationYork Landing

(Kawechiwasik)

Fox LakeCree Nation

Fox Lake (Bird)

War Lake First NationIlford (NAC)

Tataskweyak Cree NationTataskweyak (Split Lake)

ShamattawaFirst Nation

Lake

Indian

FidlerLake

Aiken R.

War Lake

Nelson

River

Crying Lake

290280

Cre

ated

By:

cpa

rker

- B

Siz

e La

ndsc

ape

BTB

- M

AR

201

5

Sca

le: 1

:1,1

80,1

14

Regional Cumulative Effects Assessment

NAD 1983 UTM Zone 15N

0 10 20 Miles

0 10 20 Kilometres

File

Loc

atio

n: J

:\MY

P\M

H_R

CE

A\F

ish_

Qua

lity\

PH

AS

E_2

\Mxd

\Mer

cury

\201

5102

8_R

CE

A_P

hase

II_A

E_F

Q_E

E_W

ater

bodi

esW

ithIn

form

atio

nOnM

ercu

ryIn

Fish

RC

EA

Are

a2_c

p.m

xd

DATA SOURCE:

DATE CREATED:

CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:

COORDINATE SYSTEM:

RCEAArea 2

Hudson Bay

Thompson

Churchill

Map 5.5.3-1

REGIONAL CUMULATIVE EFFECTS ASSESSMENT – PHASE II WATER – MERCURY IN FISH

DECEMBER 2015 5.5-34

Seven focal waterbodies were selected for detailed presentation in the following discussion. These lakes and rivers were chosen because they experienced flooding (e.g., Stephens Lake, Long Spruce forebay; Limestone forebay) or other hydrological changes related to hydroelectric development (e.g., Nelson River downstream of Limestone), and where possible had fish mercury data with relatively good sample sizes and collection frequency to allow conclusions on the evolution of fish mercury concentrations over an extended time period (e.g., Split Lake). Assean Lake and the Hayes River were included as focal waterbodies because they are off-system and serve as CAMP reference waterbodies. The Aiken River was chosen as a focal waterbody because it is the only river in Area 2 (and in the entire ROI) for which fish mercury data were systematically collected from two separate sites, allowing for the analysis of inter-site differences in fish mercury concentrations for a fluvial system. More limited site-specific data were also available for the lower Nelson River.

5.5.3.3 Changes in Mercury Concentrations over Time This section provides a detailed description of the time series of fish mercury observed for each focal waterbody. Subsequent sections will compare mercury concentrations to standards and guidelines and discuss the evolution of fish mercury concentrations in Area 2 with respect to hydroelectric development.

5.5.3.3.1 Split Lake

Split Lake has one of the longest and the most complete record of fish mercury concentrations in the entire ROI. First commercial data are available for 1969 and over the following 44 years mercury concentrations have been measured for at least one of the three focal species in 29 years (Figure 5.5.3-1). Mean mercury concentrations in Northern Pike and Walleye from Split Lake have fluctuated greatly over the 20-year period from 1970–1990 (Figure 5.5.3-1) without showing any trends that could be attributed to the operation of either LWR (which affected flows into Split Lake by mid-1976) or CRD (which would have reduced flows into Split Lake from May 1974 to November 1975 and substantially increased flows by mid-1976). Maximum mean concentrations for survey samples of at least 10 fish were observed in 1982 for both pike (0.52 ppm) and Walleye (0.66 ppm). Starting in 1992, mercury concentrations began to decrease almost continuously and by 2005 reached all-time minimums of 0.18 ppm in pike and 0.12 ppm in Walleye. Lake Whitefish, for which non-commercial data are only available since 1983, showed a much smaller range in mercury concentrations than the two piscivorous species, but also decreased significantly from a maximum concentration of 0.11 ppm in 1986 to 0.03 ppm in 2005 (Figure 5.5.3-1). Since then, mercury levels have increased significantly in all three focal fish species and currently (2013) are 2–3 times higher than the 2005 minima.



Split Lake

Year1970

19751980

19851990

19952000

20052010

2015

Mus

cle

mer

cury

(pp

m)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Pike WalleyeWhitefish

*

**

*

***

*

*

*

* *

*

*

n=7 n=5

n=5

n=5

n=5

n=5

Start of flow changes due toLWR CRD

Means without CLs prior to 1980 are from commercial samples. An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was use; n represents sample size. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.3-1: Mean (95% Confidence Limits [CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Split Lake for 1969–2013

Fish mercury data for Split Lake also include several other species with records that span multiple years with means based on larger sample sizes than for most other lakes. These results illustrate that mercury concentrations of the mainly planktivorous Cisco (Coregonus artedi) were close to 0.1 ppm between 1983 and 1998 (Figure 5.5.3-2), and thus similar to those of the benthivorous Lake Whitefish. Concentrations in two common catostomids, White Sucker and Longnose Sucker, which diet is best described as benthivorous detritivors (Ahlgreen 1990a,b) were generally slightly below and slightly above 0.2 ppm, respectively, from 1983–1986 (Figure 5.5.3-2). These mercury levels are slightly (for White Sucker) or substantially (for Longnose Sucker) higher than concentrations in Cisco and whitefish from Split Lake and from any other waterbody in Area 2 for a comparable time-period. Concentrations in Longnose Sucker were similar compared to their conspecifics from the Nelson River downstream of the Limestone GS in 1992 and significantly lower than Longnose Sucker from Threepoint Lake in 1985 and 1992 (Figure 5.5.4-6). White Sucker had higher concentrations than their conspecifics from the Aiken River in 2002 and 2003 (see discussion in Section 5.5.3.3.6), but concentrations were often significantly lower compared to White Sucker from Threepoint Lake from 1981–1985 (Figure 5.5.4-6). Disregarding the



small sample of fish for 1976, Sauger from Split Lake had mercury concentrations of 0.44–0.67 ppm in years 1982 to 1989 (Figure 5.5.3-2). Although means were highly variable (despite sample sizes of mainly 50 or more fish), these results are similar than those for Walleye from Split Lake in comparable years and confirm that mercury concentrations in piscivorous species are elevated relative to species of lower trophic levels.

Split Lake

Year1975

19801985

19901995

2000

Mus

cle

mer

cury

(pp

m)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

White SuckerLongnose SuckerCiscoSauger

**

*

*

n=7

n=3

n=6


An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was use; n represents sample size. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.3-2: Mean (95% Confidence Limits) Length Standardized Muscle Mercury Concentrations of White Sucker, Longnose Sucker, Cisco, and Sauger From Split Lake for 1976–1998

5.5.3.3.2 Stephens Lake

Stephens Lake/reservoir was flooded by approximately 226% of its original area just prior to the in-service data of the first turbine unit at Kettle GS in December of 1970. Mercury concentrations in Lake Whitefish (0.19 ppm), Northern Pike (1.05 ppm), and Walleye (1.89 ppm) from Stephens Lake were highest when first sampled (second sample for whitefish) more than 10 years after impoundment in 1970 (Figure 5.5.3-3). These maxima represented an increase factor (IF; the ratio of the maximum mercury concentration observed post-flooding to the pre-flooding concentration) of 6.9 for Walleye, 3.5 for pike, and 3.1 for whitefish (Table 5.5.4-3 in Area 3). However, since mercury concentrations of the above



species typically peak within 3 to 8 years of flooding (Bodaly et al. 2007), maximum mercury concentrations (and the IFs) in fish from Stephens Lake were likely substantially higher than those first recorded in the early 1980s. Similar to all other lakes in the ROI that experienced substantial flooding, mercury concentrations in pike and Walleye from Stephens Lake decreased almost continuously after the first measurements, reaching 0.18 ppm in pike and 0.20 ppm in Walleye by 2005. Whitefish showed a similar, although less consistent declining trend; by 2005 mercury concentrations (0.03 ppm) were about 15% of those initially recorded. Since then, mercury levels have increased significantly in all three focal fish species and currently (2012) are approximately 30–80% higher than the 2005 minima (Figure 5.5.3-3).

Stephens Lake

Year1970

19801985

19901995

20002005

20102015

Mus

cle

mer

cury

(pp

m)

0.0

0.4

0.8

1.2

1.6

2.0

PikeWalleyeWhitefish

Year ofimpoundment

2.33n=5

*

***n=5

n=5n=7

n=4

n=9

n=6

An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used; a number denotes an upper CL not shown as a bar; n represents sample size. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.3-3: Mean (95% Confidence Limits [CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Stephens Lake for 1981–2013

5.5.3.3.3 Long Spruce Forebay

The Long Spruce forebay was created in 1977 by flooding the Nelson River valley increasing the water surface area by approximately 92%.



Mercury concentrations in Lake Whitefish (0.18 ppm), Northern Pike (0.70 ppm), and Walleye (0.64 ppm) from the Long Spruce forebay were highest when first sampled (second sample for Walleye) eight years after impoundment in 1977 (Figure 5.5.3-4). The maxima represented IFs of 2.8 for whitefish, and 2.3 for pike and Walleye (Table 5.5.4-3 in Area 3). However mercury concentrations of the above species typically peak within 3 to 8 years of flooding (Bodaly et al. 2007), and it can be assumed that maximum mercury concentrations (and their IFs) in fish from the Long Spruce forebay were somewhat higher than those first recorded in the mid-1980s.

Long Spruce Forebay

Year1980

19851990

19952000

20050.0

0.2

0.4

0.6

0.8

1.0


Year ofimpoundment

Mus

cle

mer

cury

(pp

m)

The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.3-4: Mean (95% Confidence Limits) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From the Long Spruce Forebay for 1985–2003

As is typical for newly created reservoirs in the ROI (and elsewhere), mercury concentrations in all three focal species from the Long Spruce forebay decreased almost continuously until 2003 after reaching (assumed) maxima in the early 1980s. Fish mercury monitoring at the Long Spruce forebay was discontinued after 2003 and it is unknown if concentrations were still declining at that time. Considering that most other on-system waterbodies in Area 2 (and in the entire ROI) with a suitable record of fish mercury data show a further decline in mercury concentrations of pike and Walleye between samples



taken in the early 2000s and 2005, it can be expected that concentrations in at least the two predatory species from the Long Spruce forebay had not yet reached a minimum by 2003.

5.5.3.3.4 Limestone Forebay

The Limestone forebay was created by flooding the Nelson River valley in 19909, increasing the water surface area by just over 8%.

Fish mercury concentrations increased rapidly in the Limestone forebay after impoundment and reached maximum levels within two years after flooding in Lake Whitefish (0.14 ppm) and within five years in Northern Pike (0.45 ppm), and Walleye (0.64 ppm; Figure 5.5.3-5). The maxima represented IFs of 2.3 for Walleye, 1.7 for whitefish, and 1.4 for pike (Table 5.5.4-3 in Area 3). Except for a transient increase in 2003, mercury concentrations in pike and Walleye have decreased continuously and significantly since 1994; reaching minima of 0.18 ppm in Walleye and 0.19 ppm in pike in 2005 (the same concentration was also recorded in 2001). The return times to the lowest post-flood concentrations were eight years for pike and 13 years for Walleye (Table 5.5.4-3) Mean standard concentrations in whitefish have also steadily decreased from 1991 to 2001, although not significantly (Figure 5.5.3-5). The high arithmetic mean of 0.14 ppm recorded in 2005 is for a relatively small sample of 10 very large (mean fork length = 494 mm) individuals. Only one and two whitefish were available for mercury analysis in 2010 and 2013, respectively and results are not shown in Figure 5.5.3-5).

Mercury concentrations in pike and Walleye from the Limestone forebay have increased significantly from 2005 to 2010. Concentrations have remained relatively high in 2013, although the standard mean for pike was statistically similar to that recorded in 2005.



Limestone Forebay

Year1990

19952000

20052010

2015

Mus

cle

mer

cury

(ppm

)

0.0

0.2

0.4

0.6

0.8

1.0


n=5

*

*

*

Start offlooding

An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used; n represents sample size. The mean of 0.62 ppm for three Walleye in 1989 is not shown. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.3-5: Mean (95% Confidence Limits) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From the Limestone Forebay for 1989-2013

5.5.3.3.5 Lower Nelson River

The Nelson River between the Limestone GS and the estuary at Port Nelson extends over approximately 111 km. Fish mercury data from this section of the lower Nelson River exist from the freshwater portion of the estuary near Port Nelson for the years 1975-1997 (commercial data only for 1975–1977) and from several reaches of the river mainstem starting just downstream of the Limestone GS to Seal Island, approximately 15 km upstream of Port Nelson (Map 5.5.3-1). As outlined in Section 5.5.3.2, conditions for mercury availability and bioaccumulation may vary over the length of the Nelson River mainstem, potentially resulting in differences in fish mercury concentrations among sites. To test this hypothesis, data collected for several species from six different river reaches during the same sampling year were compared for between-site differences in mercury concentrations. In all cases, the frequently observed differences in arithmetic means between sites were associated with differences in mean fish length, and standard means were statistically similar (for an example see Figure 5.5.3-6). Therefore, available data for the three focal species from all river reaches (including the estuary) were combined to evaluate



temporal trends in fish mercury for the lower Nelson River below the Limestone GS between the mid-1970s and 2013.

Arithmetic meanM

ercu

ry (

ppm

)

0.0

0.1

0.2

a

b,c b,c

c

a,bLe

ngth

(mm

)

0

100

200

300

400a

b,c b,c

b

a,b

1 2 3 4 5 6

Mer

cury

(pp

m)

0.0

0.1

0.2

n.s.

Reach

c

n.s.

Standard mean

Fork length

Different letters indicate significant differences between means. Standard means were not calculated if the relationship between mercury concentration and fish length was not significant (n.s.).

Figure 5.5.3-6: Mean Arithmetic Standard Error and Standardized (Upper 95% Confidence Limit) Mercury Concentration, and Mean Standard Error Fork Length of Lake Whitefish Collected From Several Reaches of the Nelson River Mainstem in 1992



Mercury concentrations of pike (0.78–1.17 ppm) and Walleye (0.67–0.88 ppm) were highest for the commercial samples collected from 1974–1976 and the arithmetic mean for 15 very large (mean fork length = 733 mm) pike in 1982 (Figure 5.5.3-7). Likely a partial result of the often small sample sizes, particularly for walleye, concentrations in both predatory species fluctuated substantially in the 1980s and 1990s, but generally declined. Both pike and Walleye had a standard mean of 0.13 ppm in 2004. Thereafter, concentrations have increased significantly in both species and are currently (2013) more than twice as high as the 2004 minimum. Standard means for Lake Whitefish from 1986–2013 sometimes differed significantly between years, but not in a manner indicative of a time trend. Similar minimum concentrations of 0.07 ppm were recorded in 2004 and 2011. The arithmetic mean of 0.19 ppm for 2013 is for 10 very large (mean fork length = 429 mm) fish.

Nelson River

Year1975

19801985

19901995

20002005

20102015

Mus

cle

mer

cury

(ppm

)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4PikeWalleyeWhitefish

LWRLongSpruce Limestone

Kettle 1970

CRD

n=5

n=6n=6 n=5

*

*

*

**n=5

*n=6

*

*n=5

n=6

*

*

*

n=3

*

Means without CL prior to 1980 are from commercial samples. An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used. A number denotes an upper CL not shown as a bar; n represents sample size. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.3-7: Mean (95% Confidence Limits [CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From the Lower Nelson River Downstream of the Limestone Generating Station for 1975–2013



5.5.3.3.6 Aiken River

The Aiken River is a tributary to Split Lake, entering the lake from the southeast (Map 5.5.3-1). The two sampling sites on the river upstream of York Landing and a further 25 km upstream near Ilford did not experience any flooding.

Walleye and Northern Pike from the Aiken River at Ilford have been analyzed for mercury since 1978. However, until 2002 sample sizes have always been small (3–8 fish) for both species and only in two cases has the relationship between fish length and mercury concentration been significant during this time period. Except for the relatively small samples of pike and Walleye from the Aiken River at York Landing, information on fish mercury concentrations from this site only exist since 2003. Mercury data for Lake Whitefish are not available for the Aiken River, but some information exists for White Sucker.

Mercury concentrations in pike and Walleye from the Aiken River at Ilford have fluctuated substantially prior to 2002, likely a result of the generally small sample sizes. Concentrations in pike ranged from 0.20 ppm to 0.40 ppm whereas Walleye always had lower arithmetic mean concentrations than pike (sometimes significantly for the same year), ranging from 0.08 ppm to 0.24 ppm. Since 2002 standard means of Walleye have consistently and significantly increased and are currently (2012) approximately 60% higher (Figure 5.5.3-8). In contrast, standard means of pike have fluctuated without a significant change.

Northern Pike and Walleye from the Aiken River at York Landing had substantially higher concentrations (significantly for Walleye) than their conspecifics from the river at Ilford when first sampled in 1982 (Figure 5.5.3-8). However, when next analyzed in 2003 with larger fish sample sizes, standard means of each species at the two locations were similar and continued to be so until last sampled in 2012. Both pike (0.26 ppm) and Walleye (0.19 ppm) from the river at York Landing reached minimum concentrations in 2006 after which they significantly increased in 2009 and 2012 (Figure 5.5.3-8). Although having been sampled mostly one year apart since 2002, pike and Walleye from the Aiken River collected during the spawning season generally had very similar mercury concentrations to their conspecifics sampled several months later from several locations within Split Lake.

Fish mercury information for the Aiken River also exists for some larger samples of White Sucker in 2002 (from Ilford only) and 2003. All three samples have standard means of 0.09 ppm and are lower (although not significantly) than the standard means (0.13–0.17 ppm) of their conspecifics from Split Lake in 1983, 1984, and 1986 (see Section 5.5.3.3-1) As it can be assumed that White Sucker captured on the spawning grounds in the Aiken River have mainly been feeding and accumulating mercury within Split Lake, these results indicate that other species, and those of lower trophic levels, show a similar pattern of decline in mercury concentrations for the 1980–2005 period as has been repeatedly shown for Walleye and pike.



Aiken River

Year1975

19801985

19901995

20002005

20102015

Mus

cle

mer

cury

(ppm

)

0.0

0.2

0.4

0.6

0.8

1.0

Pike, YLPike, IFWalleye, YLWalleye, IF

*

*

*

*

*

**

**

n=5n=5

n=6

n=5

n=5 n=3

n=7

n=6

****

*

*

*

*

n=5n=5

n=5

n=5n=6 n=6

n=6

n=6*

n=7

An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used; n represents sample size. The stippled line indicates the 0.5 ppm Health Canada standard.

Figure 5.5.3-8: Mean (95% Confidence Limits) Length Standardized Muscle Mercury Concentrations of Northern Pike and Walleye From the Aiken River at York Landing and Ilford for Years 1978–2012

5.5.3.3.7 Assean Lake

Except for one commercial sample and two survey samples with small (n < 10) fish numbers, standard concentrations of all three focal species from Assean Lake, the off-system reference lake for the lower Nelson River Region of CAMP (CAMP 2014), were relatively stable and without an obvious time trend over the period of record 1981–2013. Concentrations ranged from 0.19–0.25 ppm in pike, 0.18–0.32 ppm in Walleye, and 0.04–0.06 ppm in whitefish (Figure 5.5.3-9). While no statistical difference existed between any of the yearly concentrations for whitefish, the standard mean for pike in 2010 was just significantly higher than the standard mean for 2001, and the standard mean for Walleye in 2013 was significantly higher than the means for 1985, 2001, 2002, and 2010.

When comparing the time-lines of mercury concentrations of fish from Assean Lake with those of on-system lakes in order to evaluate potential system effects in Area 2 (or beyond), it must be noted that the temporal resolution of fish mercury data is lower than for most on-system lakes. This is important because the absence of data for Assean Lake during times of dynamic change in fish mercury of on-system lakes limits our interpretation of the observed pattern. This is well exemplified by the time period 2001–2013, for



which no fish mercury data exist for Assean Lake between 2003 and 2009, while the results from several on-system waterbodies indicate that mercury concentrations in at least pike and Walleye reached a minimum in or within one year of 2005 and since have increased significantly (see Sections 5.5.3.3.1, 5.5.3.3.2, and 5.5.3.3.4).

Assean Lake

Year1970

19751980

19851990

19952000

20052010

2015

Mus

cle

mer

cury

(pp

m)

0.0

0.2

0.4

0.6

0.8

1.0



n=5

n=6

n=9

*

*

* *

n=6

Means without CL prior to 1980 are from commercial samples. An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used; n represents sample size. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.3-9: Mean (95% Confidence Limits [CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Assean Lake for 1978–2013

5.5.3.3.8 Hayes River Except for three commercial samples (mean of 0.02 ppm) for Arctic Char (Salvelinus alpinus) collected in 1970 (Derksen 1978a), fish mercury data from the Hayes River are available only since 2006. In 2006 and 2009 samples were taken at a location closer to the river mouth, whereas the more recent data come from the CAMP sampling site further upstream (Map 5.3.3-1). For the current analysis, data from both recent sampling locations have been combined. Standard means of Walleye have been statistically similar for 2006 and 2010, ranging from 0.35 ppm to 0.46 ppm, but were significantly lower (0.29 ppm) in 2013 compared to 2010 (Figure 5.5.3-10). Mercury concentrations in pike showed a similar temporal



pattern as Walleye with standard means at or just below 0.2 ppm. However, because of high confidence limits due to small sample sizes in 2010 and 2013, no significant differences existed between mean concentrations in pike. Arithmetic means indicated that mercury levels were consistently and often significantly lower in pike than in Walleye.

Hayes River

Year2000

20052010

2015

Mus

cle

mer

cury

(pp

m)

0.0

0.2

0.4

0.6Pike Walleye Whitefish

n=3

n=9n=5

*

*

n=8

An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.3-10: Mean (95% Confidence Limits) Length Standardized Muscle Mercury Concentrations of Northern Pike Walleye, and Lake Whitefish From the Hayes River for 2006–2013

Mercury concentrations in whitefish ranged from 0.07 to 0.09 ppm over the three available sampling years between 2006 and 2011 (Figure 5.5.3-10).

5.5.3.4 Cumulative Effects of Hydroelectric Development on Fish Mercury Concentrations in Area 2

5.5.3.4.1 Comparison to Benchmarks

With the exception of Northern Pike from the Limestone forebay, mercury concentrations in piscivorous fish species from all Area 2 waterbodies that were flooded in the course of the construction of a



hydroelectric generating station have at some point substantially exceeded the Health Canada standard of 0.5 ppm for the commercial sale of fish (e.g., Figures 5.5.3-4 to 5.5.3-6). Of the three generating stations in Area 2, the Limestone GS has created the least amount of areal flooding, and pike and Walleye from the Limestone forebay have experienced the smallest increase in mercury concentration, with the maximum for pike remaining just below and that of Walleye just above the standard. (For a detailed discussion of the relationship between the amount of flooding and fish mercury concentrations, see Section 5.5.4.4.3)

In addition to the fish in newly created forebays, pike, Walleye, and Sauger from other on-system waterbodies in Area 2 have routinely exceeded mean concentrations of 0.5 ppm during the first 2 decades of fish mercury monitoring in Manitoba (1969–1990). With few notable exceptions, such as Walleye (Figure 5.5.3-1) and Sauger (Figure 5.5.3-2) from Split Lake, the vast majority of these exceedances, which have generally been less severe than for the reservoirs, have been for commercial samples and small survey samples of larger fish (relative to the standard length of the species; Figures 5.5.3-2 and 5.5.3-8). The high concentrations observed in large (commercially caught) Walleye and pike from Split Lake when first sampled in 1970 did result in the partial (whitefish were still accepted for marketing) closure of the commercial fishery of the lake from July 1970–1976 (Derksen 1978b). These decisions were made at a time when not much was known about mercury concentrations in Manitoba fish and, as can be interpreted from the course of events in 1969–1970 described in Bligh (1971), the Province likely was under substantial public pressure to protect consumers in light of the recent reports from highly contaminated fish in the Saskatchewan River (see Section 5.5.1.2).

The mercury concentrations of 0.55 ppm (pike) and 0.76 ppm (Walleye) for the commercial samples in 1970 and of 0.47–0.75 ppm for survey samples of Walleye from 1972–1974 could not have been caused by LWR or CRD, which had not yet commenced at the time. A residual effect from the 1960 flooding of the Kelsey forebay and several upstream Lakes (e.g., Sipiwesk) is also unlikely. This because a potential downstream effect from the export of methylmercury in water and lower trophic level organisms into a large waterbody such as Split Lake must have been small, particularly almost 10 years after the initial increase in methylmercury availability from flooding. Even if some pike and Walleye which may have reached high (2.0 ppm) mercury concentrations 3–9 years after Kelsey flooding migrated downstream from waterbodies upstream of the generating station (no studies have been carried out to document such movements other than for Lake Sturgeon), their numbers would not have been large enough to measurably affect mean mercury concentrations of the (presumably) several hundred thousand adult conspecifics in Split Lake in the early 1970s. Instead the elevated (above 0.5 ppm) mercury concentrations of pike and Walleye from Split Lake during that time are most likely attributable to the large fish sampled for mercury from the commercial catches and for most of the (small) survey samples. A similar situation could occur even today. If only a few of the largest pike and Walleye from a current CAMP sample at Split Lake (or most other waterbodies) would be selected for mercury analysis, the likelihood of a mean arithmetic mercury concentration exceeding 0.5 ppm is very high (W. Jansen, North South Consultants Inc.[NSC], pers. observation).

The recent (post-2005) significant increases in mercury concentrations of mainly pike and Walleye observed for almost all on-system waterbodies in Area 2 have not resulted in concentrations that exceed the Health Canada 0.5 ppm standard.



More limited results for other fish species in Area 2 waterbodies also indicate that mercury concentrations mainly follow fish trophic level and size (age), and that they can exceed 0.5 ppm in piscivores but usually remain below 0.2 ppm in species feeding primarily on invertebrates and/or detritus. Mean concentrations in Sauger from Split Lake and the Limestone forebay, at least prior to 1994 (more recent data are not available) have exceeded the 0.5 ppm Health Canada standard. Cisco, for which recent data are also lacking, generally has mercury concentrations of ≤0.1 ppm and similar to Lake Whitefish. White Sucker from Split Lake (Figure 5.5.3-2) and the Aiken River (in 2003), also had concentrations close to 0.1 ppm. Longnose Sucker from Split Lake and the lower Nelson River had higher mercury levels than White Sucker in the past, sometimes clearly exceeding 0.2 ppm, but recent (>1992) data are not available. Concentrations in Burbot from the Limestone forebay were generally slightly higher than 0.2 ppm except for one sample of four fish in 1992 which had a concentration of 0.61 ppm and exceeded the Health Canada standard.

Together with the Churchill River in Area 4 (see Section 5.5.5.5.1), Area 2 has the most extensive information on Lake Sturgeon mercury concentrations in the ROI. Concentrations obtained from the Nelson and Hayes rivers indicate that individual sturgeon did not normally exceed 0.2 ppm for fork lengths up to 100 cm, but some larger fish reached concentrations that exceed the 0.5 ppm standard for human consumption (Table 5.5.3-1).



Table 5.5.3-1: Arithmetic Mean (range) Mercury Concentrations, Parts-per-Million, and Fork Length for Lake Sturgeon from Two Reaches of the Nelson River (downstream of Limestone Generating Station to Gillam Island, and the estuary), the Hayes River, and Split, Gull, and Stephens Lakes

Year n Hg, Mean Hg, Range Length, Mean Length, Range

Nelson River, downstream Limestone GS 2014 2 0.190 0.15–0.23 766.5 703–830 2013 1 0.035 - 425 - 2011 3 0.166 0.14–0.21 693.7 654–715 2010 6 0.313 0.18–0.38 1086.7 690–1,345 2008 7 0.159 0.08–0.28 672.3 537–800 2003 7 0.185 0.13–0.34 841.4 725–1,200

1970 a 4 0.11 0.09–0.13 - - Nelson River, estuary

1982 a 5 0.220 0.10–0.60 - - 1983 a 1 0.12 - - - 1982 a 1 0.12 - - - 1978 a 1 0.10 - - - 1977 a 1 0.06 - - - 1976 a 1 0.16 - - - 1975 a 1 0.06 - - - 1970 a 5 0.10 - - -

Hayes River 2013 1 0.176 - - 720

2011 1 0.213 - - 771

2010 1 0.194 - - 664

2009 2 0.098 0.07–0.13 550.5 543–558

Split Lake 1970 a 1 0.014 - - -

Gull Lake 2014 1 0.166 - - - 2013 1 0.052 - - - 2006 1 0.039 - - - 2004 10 0.207 0.04–0.67 1158.8 1,035–1,286 2002 3 0.166 0.10–0.21 1162.5 1,050–1,275

Stephens Lake 2014 1 0.100 - - 598

2012 1 0.174 - - 802

2008 1 0.099 - - 587 a Commercial samples



5.5.3.4.2 Evolution of Fish Mercury Concentrations in Relation to Hydroelectric Development

CHURCHILL RIVER DIVERSION AND LAKE WINNIPEG REGULATION

Lake Winnipeg Regulation affected inflows to Split Lake via the Nelson River. Inflows into the lake were reduced during summer and fall 1973 due to the construction of cofferdams in the west channel of the Nelson River. Two Mile Channel was completed in mid-1976, at which time flow regulation related to LWR, impoundment of the Jenpeg forebay, and changes to inflows to Split Lake were fully implemented. The monthly average discharge at Kelsey GS pre- and post-LWR shows that LWR increased winter flows at Kelsey, and, on average, decreased open water flow (Water Regime, Section 4.3.2.4.3). Operation of the Kelsey GS does not cause substantial water level changes to Split Lake. Higher flows (approximately 7-fold increase) in the Burntwood River as a result of CRD entered Split Lake starting in mid-1976. Post-CRD/LWR water levels have been 0.4 m higher on average than pre-CRD/LWR water levels. Also, there was a seasonal reversal in water levels because average winter levels are typically higher than average summer levels (see Sections 5.5.3 and 5.5.3.3.1).

The above changes in hydraulic conditions due to CRD/LWR apparently had little effect on mercury concentrations in whitefish, pike, and Walleye from Split Lake. Any increase in concentrations that may have occurred a few years after the initial post-CRD increase in Split Lake water levels were too small to be detected within the large variability of mean concentrations in the 1970s and 1980s. Mercury concentrations in pike and Walleye from Split Lake have fluctuated greatly over this period (Figure 5.5.3-1) and the timeline of concentration changes does not correlate with any of the key hydrological events of CRD/LWR.

RESERVOIRS AND OTHER ON-SYTEM WATERBODIES

The time-lines of fish mercury concentrations in all three Area 2 impoundments (Stephens Lake, Long Spruce forebay, Limestone forebay) conform to the well-established pattern for boreal reservoirs (Schetagne et al. 2003; Bodaly et al. 2007; see Figure 5.5.4-11). This conformity is apparent despite the fact that concentrations for the first 8–11 years after reservoir creation are lacking for Stephens Lake and the Long Spruce forebay, and the exact extent and timing of potential maximum concentrations remains unknown.

Mercury concentrations in pike and Walleye from all Area 2 reservoirs continued to decline (Stephens Lake and Long Spruce forebay) or started to decline (Limestone forebay) after 1994. Similar declining trends were also observed for other fluvial lakes on the Nelson River, such as Split Lake (Section 5.5.3.3.1) and Gull Lake, and for the lower Nelson River itself (Section 5.5.3.3.5). Mean concentrations in pike and Walleye from all these waterbodies were at or mostly below 0.2 ppm by approximately 2005. A similar temporal pattern of mercury concentrations exists for Lake Whitefish although it is less pronounced and the number of waterbodies with sample frequencies and fish sample sizes comparable to the two piscivorous species is lower. Similar to pike and Walleye, whitefish in on-system waterbodies of Area 2 also reached minimum concentrations of 0.03-0.08 ppm in 2004/2005.



The decline in mercury concentrations for pike and Walleye from Split and Gull lakes, and the lower Nelson River is particularly remarkable because these waterbodies were not directly affected by flooding, and changes in concentration cannot be explained by small fish sample sizes and/or differences in fish length for arithmetic means. It is possible that starting in the mid-1990s the lower concentrations in pike and Walleye from all waterbodies on the Nelson River were at least partially driven by an ecological event unrelated to hydrological development. Rainbow Smelt, an invasive fish species first recorded in Lake Winnipeg in 1990, expanded its distribution into Split Lake and the Area 2 forebays somewhere between 1994 and 1996 (Remnant et al. 1997). The species rapidly increased in abundance and numerically dominated all forage fish species in the diet of pike and Walleye from Split and Gull lakes and the three reservoirs (NSC 2012). Smelt mercury concentrations, measured for maximally five years in several waterbodies, were substantially (up to 3 times) lower than those of Spottail Shiner (Notropis hudsonius) and Emerald Shiner (Notropis atherinoides), common forage species in the diet of pike and Walleye prior to the arrival of smelt (KHLP 2012). It seems that the two piscivores switched their diet from fish species with relatively high mercury concentrations to a species with a relatively low concentration in response to changes in relative abundance. As a possible result, mercury concentrations of pike and Walleye in the on-system waterbodies invaded by smelt declined and the rate of decline in the three reservoirs increased. There are no off-system lakes with known smelt populations for comparative purposes.

Current (2009–2014) standard means of pike and Walleye from reservoirs and other on-system waterbodies (Split and Gull lakes, lower Nelson and Aiken rivers) range from 0.24 ppm in pike from the Limestone forebay to 0.41 ppm in Walleye from Split Lake and are mostly significantly higher compared to 2005. With the exception of Split Lake, concentrations in whitefish have not followed the post-2005 increase observed for the two piscivores. Again, this difference is partially a result of small fish sample sizes (Stephens Lake) and/or lack of post-2005 samples (Long Spruce forebay). Current (2009–2013) standard means for whitefish, available only for Split and Stephens lakes and the lower Nelson River, range from 0.046 ppm to 0.102 ppm.

The current means for Walleye from the on-system waterbodies are mostly statistically similar to those from their conspecifics from off-system Assean Lake, whereas means of pike and whitefish are significantly higher for approximately half of the available yearly comparisons. These differences among the three species are partially caused by the significant difference between the mean concentrations of Walleye from Assean Lake in 2010 and 2013 (Figure 5.5.3-9 and Table 5.5.4-2). To minimize the effect of annual variability in mercury concentrations and other stochastic effects associated with results for a single waterbody (i.e., Assean Lake), all recent data for four off-system CAMP reference lakes within the ROI were used to obtain an estimate of current background concentrations (see Section 5.5.4.4.2).

DOWNSTREAM EFFECTS OF HYDROELECTRIC DEVELOPMENT ON FISH MERCURY CONCENTRATIONS IN AREA 2

Based on the data presented in Section 5.5.3.3.1 and the discussion in Section 5.5.3.4.2, there is no clear evidence that potential exports of mercury and/or methylmercury, either suspended in water or incorporated into biota from flooded waterbodies upstream of the Kelsey GS (i.e., LWR) or on the Rat/Burntwood River (i.e., CRD) had a measurable effect on fish mercury concentrations in Split Lake. Because large waterbodies on a diversion route act as methylmercury sinks and greatly reduce further export downstream (Schetagne et al. 2003; for more detail see Section 5.5.4.4.3), it can be assumed that



fish mercury concentrations in Area 2 on-system waterbodies downstream of Split Lake were not affected by the potential export of methylmercury in water, particular organic matter, and plankton from CRD/LWR.

There also is little evidence for a downstream effect of reservoir flooding on fish mercury concentrations within Area 2. At least none of the detailed temporal changes in concentrations of the three focal species from any of the three reservoirs can be traced in the respective concentrations in fish from the waterbody immediately downstream. For example, the increase in concentration of whitefish from the Long Spruce forebay between 1993 and 1996 was not observed in their conspecifics from the Limestone forebay over the same time period. Also, the increase in mercury concentrations between 1990/1991 and 1994 and the downward trend in concentrations between 1994 and 2001 in fish from the Limestone forebay was not reflected in the results from the lower Nelson River over the same time periods, although the available data for comparisons are limited. These patterns (or lack thereof) do not exclude the possibility of a downstream effect, but if it exists, it must be relatively small.



5.5.4 Area 3: Southern Indian Lake to Split Lake Inlet As part of CRD, a control structure was placed at Missi Falls on the northern outlet of SIL, a large (1,977 km2) multibasin lake through which the Churchill River flowed (Rosenberg et al. 1987). Upon completion on October 15, 1976, the Missi Falls Control Structure (CS) raised the water level of SIL by approximately 2.7 m above its long-term mean (see Section 4.3.3.2.3), increasing the lake area by approximately 136 km2, and diverting the majority of the upper Churchill River flow into the Rat/Burntwood River system. Water was diverted via a channel constructed between South Bay in the southwestern part of SIL and Isset Lake at the headwaters of the Rat River in the Nelson River basin. A CS was built at Notigi Lake to regulate flows down the Rat River valley.

Construction activities for the Notigi CS were initiated in winter 1972/1973. Discharge from Notigi Lake was stopped over the period from May 1974 to November 1975 to retain local runoff until water levels in the Rat River valley and SIL were similar. Following impoundment, approximately 453 km2 were flooded which caused an approximate doubling of the combined surface area of lakes in this reach, including Issett, Karsakuwigamak, Pemichigamau, Rat, Mynarski, and Notigi lakes. The greatest flooding and change in surface area occurred in the Issett Lake area, which increased from approximately 4.6 to 119 km2 and Rat Lake which increased from approximately 93.6 to 241 km2. Diversion of the Churchill River was initiated in June 1976, and the cofferdam at the Notigi CS was removed in September 1976.

Due to the extensive amount of flooding from CRD (approximately 293 sq mi or 759 km2 in Areas 3 and 4), fish mercury concentrations in some lakes along the diversion route had the most substantial increases in the ROI. All of the lakes sampled for fish mercury in Area 3 (both on-system and off-system) are shown in Map 5.5.4-1

5.5.4.1 Key Published Information Mercury concentrations in several fish species from primarily commercial catches from Area 3 waterbodies between 1970–1972 were first compiled by Bligh (1970, 1971) and Derksen (1978a, b, 1979). Mercury data for fish sampled commercially, including analyses of individuals from diversion route lakes through to 1979 were summarized in Bodaly and Hecky (1979) and McGregor (1980). Commercial data from composite fish samples through to 1985 were compiled as part of the Canada-Manitoba Agreement on the Study and Monitoring of Mercury in the Churchill River Diversion (DFO 1987; Rannie and Punter 1987 (data to 1983); see Section 5.5.1.2). Bodaly and Hecky (1979) also reported on mercury concentrations of Lake Whitefish at Southern Indian and Issett lakes in 1975 and 1978.

Several publications include mercury data from individual fish based on survey samples collected at SIL and diversion route lakes from 1982–1985 under the auspices of the 1983 Canada-Manitoba Agreement (Bodaly et al. 1984a; Strange 1985; Green 1986; Bodaly et al. 1987; Derksen and Green 1987; Hecky et al. 1987). In addition to the data determined from the 1983-85 fish collections, Derksen and Green (1987) analysed all mercury concentrations existing at the time for pike, Walleye, and Lake Whitefish from 11 Area 3 lakes.



Bodaly et al. (1987) reported on mercury concentrations in two forage fish species (Spottail Shiner and Yellow Perch) sampled from several lakes (SIL Issett, Granville, Notigi, Mynarski, Footprint) in 1981 and 1982. In addition to FEMP studies outlined in Section 5.5.1.2, Ramsey (1990) provided mercury concentrations in Yellow Perch captured in Granville Lake and transferred to holding pens in Methyl Bay (SIL) to measure methylation rates in flooded zones. Also as part of FEMP, Baker and Davies (1991) compiled information on fish mercury concentrations in lakes harvested by NFA communities. Fish mercury concentrations continued to be measured for Rat and Threepoint lakes between 1986 and 1989 by Manitoba Natural Resources (Green 1990).

Fish mercury data from SIL and several diversion route lakes were collected on multiple occasions between 1992 and 2005 in conjunction with the MMMR Program and successor programs (see Section 5.5.1.2; Strange 1993, 1995; Strange and Bodaly 1997, 1999; Jansen and Strange 2007a). In 2007/2008, monitoring of mercury in fish from lakes on the CRD route (SIL, Issett, Leftrook, Notigi, Threepoint, and Wuskwatim lakes) was conducted as part of Manitoba Hydro’s environmental studies (Jansen 2010a).

Fish mercury concentrations continue to be monitored in Area 3 at SIL (2 areas) and Rat Lake on a three-year rotational basis, and annually at Threepoint and Leftrook lakes under Manitoba/Manitoba Hydro’s CAMP. Results for sampling conducted in 2010, when monitoring was initiated in this area under CAMP, are presented in CAMP (2014).

Summaries of fish mercury data from northern Manitoba often focus on waterbodies on the CRD route (Area 3) and include assessments of the effects of hydroelectric development on mercury concentrations and associated fisheries impacts (Bodaly and Hecky 1979; McGregor 1980; Bodaly et al. 1984a, b; Strange et al. 1991; Strange and Bodaly 1999; Bodaly et al. 2007; Jansen and Strange 2007a). The temporal scope of these reviews was limited to the data available at the time and the statistical analysis chosen for determining the time of return to background mercury concentrations in one earlier review (Bodaly et al. 2007) resulted in conclusions that partially differ from the current report (for details, see Section 5.5.4.4.3).

Johnston et al. (1991) developed and tested predictive models relating reservoir flooding to fish mercury concentrations for reservoirs using fish mercury data from SIL and several of the diversion route lakes. More recently, Bodaly et al. (2007) used similar, updated data as part of an analysis of post-impoundment trends in fish mercury concentrations in boreal Manitoba reservoirs.

Leaf Rapids

Thompson

O-Pipon-Na-Piwin Cree NationSouth Indian Lake

Nisichawayasihk Cree NationNelson House (NAC)

KelseyG.S.

Wuskwatim G.S.

493

280

6

391

CousinsLake

WuskwatimL.

Notigi Lake

Issett

Lake

OpachuanauLake

Lake

KarsakuwigamakLake

MynarskiLakes

Southern

Indian

Osik L.

Leftrook

Lake

PemichigamauLake

MysteryL.

BaldockLake

ThreepointLake

WapisuLake

Rat Lake

Moss Lake

Ospwagan

Lake

Barrington

Lake

GranvilleLake

Gauer Lake

Le ClairLake

FootprintL.

SILArea 4

SILArea 6

SILArea 5

SILArea 2

Notigi Control Structure

Missi Falls Control Structure

1.0

22-OCT-15


Province of Manitoba, Government of Canada, North/South Consultants andManitoba Hydro.

05-DEC-15

Hudson Bay

Thompson

Churchill



0 10 20 Kilometers

0 10 20 Miles

DATA SOURCE:

DATE CREATED:

CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:

COORDINATE SYSTEM:

RCEAArea 3

Legend


RCEA Area 3

NOTE: (NAC) Northern Affairs Community

Fish Mercury InformationFocal WaterbodyRCEA Region of InterestSouthern Indian Lake (SIL) Historical Geographic Area

Control StructureGenerating Station (Existing)HighwayFirst Nation Reserve

Map 5.5.4-1



5.5.4.2 New Information and/or Re-analysis of Existing Information The current assessment is based on the updated analysis of all available data from 1969 to 2014. With mercury concentrations for almost 27,000 fish of 18 species, Area 3 represents more than half of all mercury analyses for individual fish in the entire ROI (Table 5.5.1.-1). In addition to this, more than 300 analyses for composite, commercial fish samples exist for Area 3. The commercial data were used mainly for years predating-CRD and when no results based on individual fish were available. In these cases, some of the published yearly means were recalculated for years with multiple composite samples that were reported separately. Almost 90% of all individual mercury data for Area 3 are for Cisco, Lake Whitefish, pike, and Walleye, and almost 60% are just for pike and Walleye.

Six on-system lakes and one off-system lake were selected as focal waterbodies for detailed presentation in the following discussion due to the presence of data in a time-series with sufficient resolution to allow conclusions on the evolution of fish mercury concentrations during and following flooding (i.e., Southern Indian, Issett, Rat, Notigi, Threepoint and Wuskwatim lakes). In addition, data are available from several off-system lakes both prior to and after CRD to provide a record of among-lake variability in fish mercury concentrations independent of the effects of hydroelectric development. Leftrook Lake is the off-system reference lake with the longest and most complete record of post-CRD fish mercury concentrations in the entire ROI and it was included as a focal lake. The waterbody with the most extensive data set and on which most of the early studies on fish mercury were focused is SIL. Southern Indian Lake also is the only waterbody in Area 3 for which there are fish mercury data from different areas of the lake, illustrating among-area differences in fish mercury concentrations.

5.5.4.3 Changes in Mercury Concentrations over Time

PRE-CRD BASELINE MERCURY CONCENTRATIONS

The majority of the fish mercury monitoring in Area 3 that pre-dates CRD (i.e., 1969–1974) relates to commercially caught fish. This includes mercury data from both on-system and off-system lakes, such as Southern Indian, Gauer, Baldock, Barrington, Opachuanau, and Wuskwatim lakes (Derksen 1978a, 1979; Bodaly and Hecky 1979; McGregor 1980; Bodaly et al. 1984; DFO 1987; Table 5.5.4-1). In addition to the mercury concentrations from commercial catches, pre-CRD mercury data exist for five samples of 24-25 Lake Whitefish from Cousins and Issett lakes and Areas 2, 4, and 6 of SIL collected in 1975. Although the time of collection falls after the start of impoundment of SIL and Issett lakes, and the flooding at Cousins Lake, it predates the start of the major ponding of SIL in June 1976. Furthermore, the mercury concentration in the relatively large whitefish sampled in 1975 would not have been affected (or only marginally) by any ‘new’ mercury potentially accumulated during the early flooding stage. This conclusion is supported by findings that mercury concentrations in the muscle of large fish respond more slowly to increased mercury availability, as in the case of reservoir flooding, than small fish (Harris and Hutchinson 2009).

Based on data from commercial fish, pre-CRD mercury concentrations were similar between on-system and off-system waterbodies with means ranging from 0.06–0.13 ppm in Lake Whitefish, 0.17–0.36 ppm in



Northern Pike and 0.19–0.36 ppm in Walleye (Table 5.5.4-1). Because of the composite nature of commercial samples (see Section 5.5.1.4), the typically small sample size and unknown length of the fish included in the sample, statistical comparisons among commercial samples are rarely feasible. Comparisons to means based on individual fish are also problematic. Therefore, the conclusion of equality in fish mercury concentrations between on-system and off-system waterbodies is tentative for comparisons of individual lakes, and is mainly based on the weight of evidence for the entire dataset.

Table 5.5.4-1: Mean (range) Mercury Concentrations, Parts-per-Million, for Northern Pike, Walleye, and Lake Whitefish from On-System and Off-System Waterbodies in Area 3 for the Years Predating Churchill River Diversion (1969–1974, except Wuskwatim Lake, 1970–1976)

Off-system Location n Mean Range On-system

Location n Mean Range

Northern Pike

Baldock Lake 1 0.17 - Opachuanau Lake 2 0.31 0.31–0.31

Barrington Lake 3 0.36 0.33–0.39 SIL, Area 6 3 0.29 0.26–0.32

Gauer Lake 1 0.31 -

Granville Lake1 4 0.35 0.22–0.43

Walleye

Baldock Lake 2 0.23 0.18–0.27 Opachuanau Lake 3 0.19 0.17–0.21

Barrington Lake 3 0.21 0.13–0.26 SIL, Area 6 3 0.23 0.19–0.28

Gauer Lake 2 0.22 0.18–0.25 Wuskwatim Lake 6 0.34 0.25–0.44

Granville Lake 5 0.23 0.10–0.30

Lake Whitefish

Baldock Lake 1 0.12 - Opachuanau Lake 1 0.06 -

Barrington Lake 1 0.07 - SIL, Area 6 4 0.06 0.02–0.11

Gauer Lake 1 0.13 - Wuskwatim Lake 1 0.08 -

Granville Lake 1 0.07 -

Note: All data are for fish from composite samples of commercial catches; n indicates the number of years with data; occasionally the concentration for a particular year represents a mean based on two or more composite samples.

1. Granville Lake is located just west of Area 3 and is included for comparative purposes; it was included because of its long-term data set on fish mercury concentrations, including pre-CRD data (see Figure 5.5.4-9).

Considering the problems with comparing mercury concentrations from composite samples and those based on individual fish, it is noteworthy that the arithmetic means measured in individual whitefish from areas 2, 4, and 6 of SIL in 1975 (range of 0.046 to 0.070 ppm) are almost identical to the mean of 0.06 ppm of the six commercial samples analysed prior to flooding (1969–1973). The standard mean of



0.103 ppm for individual whitefish caught in 1975 from Cousins Lake, an off-system lake, also closely matches the mean for commercial samples measured in four off-system waterbodies (Table 5.5.4-1). The survey sample mean of 0.146 ppm from Issett Lake in 1975 is slightly outside the range of concentrations from commercial samples of whitefish collected from the lakes listed in Table 5.5.4-1 and is significantly higher than the mean for each of the four survey samples from SIL in 1975. Perhaps by the late summer of 1975 the rate of mercury accumulation in whitefish from Issett Lake had started to increase due to the flooding that occurred slowly from May 1974 to November 1975 (see Section 5.5.4.3.2). The data for whitefish from Issett Lake in 1975 were not used in any calculations of pre-CRD mercury baseline concentrations.

POST-CRD CHANGES IN MERCURY CONCENTRATIONS

Post-CRD mercury data have been collected for approximately 27,000 fish from almost 100 waterbodies in Area 3. As illustrated by the results for pike, most of the data are from approximately 20 lakes and for years between 1978 and 1994 (Appendix 5.5.4A). Mean mercury concentrations of pike in several lakes other than the focal lakes, such as Apussigamasi, Karsakuwigamak, Moss, East Mynarski, West Mynarski, Mystery, Opachuanau, Osik, Ospwagon, Pemichigamau, and Wapisu, have exceeded 1.0 ppm in at least one year post-CRD (Figure 5.5.4A-11). All of these lakes have been flooded to various degrees by CRD.

Subsequent sections will first provide a detailed description of the time series of fish mercury observed for each focal lake. This will be followed by a general discussion of the evolution of fish mercury concentrations in Area 3, including comparisons to standards and guidelines.

5.5.4.3.1 Southern Indian Lake

Southern Indian Lake was flooded by approximately 7% of its original area starting in August 1973 but the majority of flooding took place between June and October 1976. Year 1976 was considered as year 1 post-flooding. During studies conducted by DFO in the 1970s, the lake was divided into six areas based in part on pre and post flooding and diversion differences in water flow (Bodaly and Hecky 1979). Mercury concentrations for pike, Walleye, Lake Whitefish, Burbot and Longnose Sucker (Catostomus catostomus) from different areas of SIL (see Map 5.5.4-1) are provided in Appendix 5.5.4B. The following discussion focuses on Area 6 of SIL because of its relatively good overall data record and the presence of pre-CRD data. Area 6 is also referred to as South Bay and represents the furthest downstream part of the lake prior to the construction of the diversion channel.

Mercury concentrations in in Lake Whitefish and Walleye from Area 6 increased relatively rapidly after CRD flooding, but took longer to reach maximum values in pike. Whitefish reached a maximum mean of

1 Tables, figures, and maps with a letter in their numbers (e.g., A) can be found in the appendices for this chapter.



0.24 ppm in 1979, four years after flooding (Figure 5.5.4-1). The maximum concentration was four times higher than the average pre-CRD concentration (i.e., IF = 4.0; Table 5.5.4-3). Concentrations decreased after 1979 to reach a minimum of 0.026 ppm by 2005, resulting in a return time of 26 years (Table 5.5.4-3). Concentrations have remained similar since that time (Figure 5.5.4-1). The post-CRD minimum was less than half of the mean pre-CRD concentrations of 0.06 ppm (Table 5.5.4-3).

Mean concentrations in Walleye from Area 6 reached a maximum of 0.80 ppm three years after flooding and, in contrast to whitefish, fluctuated substantially until reaching a secondary peak of 0.78 ppm in 1982 (Figure 5.5.4-1). Thereafter, mercury concentrations declined, again with many ups and downs to reach a minimum of 0.33 ppm in 2005. The corresponding return time was 27 years and the IF was 3.5 (Table 5.5.4-3). The post-CRD minimum was substantially higher than the pre-CRD mean (from commercial samples) of 0.23 ppm. Mean concentrations increased significantly from 2005 to 2007 and remained similar thereafter.

SIL, Area 6

Year1970

19751980

19851990

19952000

20052010

2015

Mus

cle

mer

cury

(pp

m)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

PikeWalleyeWhitefish

*

**

*

*

*

**

**

**

*

* *

*

* ***

CRDflooding

Means without CL prior to 1980 are from commercial samples. An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.4-1: Mean (95% Confidence Limits [CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Area 6 (South Bay) of Southern Indian Lake for 1969–2013



Northern Pike had the highest post-CRD mercury concentration of all fish species analyzed from SIL. After a steady increase from a pre-CRD mean of 0.29 ppm, pike from Area 6 reached a maximum concentration of 1.1 ppm in 1984, nine years after flooding (Figure 5.5.4-1). Thereafter, concentrations decreased rapidly within the next year and then more steadily until reaching a minimum of 0.32 ppm in 2005, which was just slightly higher than the pre-CRD mean. The return time was 24 years and the IF was 3.8 (Table 5.5.4-3). Similar to Walleye, concentrations increased after 2005 and were significantly higher for all three sampling years until 2013 (Figure 5.5.4-1).

Although the general post-CRD pattern of rise and fall in fish mercury concentrations was similar for the five different geographical areas sampled in SIL, and for the samples representing a mix of fish from different areas (designated ‘No area’ in all respective figures), sometimes substantial differences existed when comparing year-specific concentrations and rates of increase and/or decline in concentrations. For example, maximum mercury concentrations in pike from other areas of SIL were often significantly lower than from Area 6, never exceeding 0.8 ppm except for fish from ‘No area’ (Figure 5.5.4B-1).

The decline of concentrations in pike and Walleye prior to 2005 in other areas of SIL was much less consistent compared to the pattern observed for Area 6 (see Figures 5.5.4B-1 and 5.5.4B-2). This may partially have been a result of the generally smaller sample sizes in areas other than 4 and 6. Because of the lack of pre-CRD data for all areas except Area 6, the proportional increase (i.e., IF) could not be calculated for any other area and was estimated only for Area 4 by using pre-CRD concentrations from Area 6 (Table 5.5.4-2). For Lake Whitefish, standard means recorded in 1975 from Areas 2, 4, and 5 were all similar and only marginally lower than the 0.07 ppm mean for whitefish from Area 6 (Figure 5.5.4B-3). Whitefish in Areas 4 and 5 reached similar maximum concentrations of 0.22 and 0.26 ppm, representing increases of approximately four times from pre-flood levels, which was similar to Area 6. In contrast, whitefish from Area 2 reached a lower maximum concentration of 0.10 ppm five years post-flooding.

One other notable difference in mercury concentrations of fish from the five areas of SIL existed for Walleye in Areas 4 and 6. For all sampling years (starting in 1979 for Area 4), mean concentrations in fish from Area 6 were higher than those from Area 4, and since 1998/1999 the difference has been significant (Figure 5.5.4B-2). Furthermore, mercury levels in Walleye from Area 5 and ‘No area’ (the only other areas with a comparable dataset to Areas 4 and 6) have also been lower than those from Area 6. This has been observed since 1989, and concentrations in Area 4 and ‘No area’ have been at or below pre-CRD concentrations for Area 6 since 1996 (Figure 5.5.4B-2).

Species other than Lake Whitefish, Northern Pike, and Walleye have been less frequently analyzed for mercury, and interpretation of their mercury concentrations and the impact of CRD is more difficult. Concentrations in Burbot, a piscivorous species that is occasionally consumed domestically, remained consistently lower than those in pike and Walleye from SIL (Figure 5.5.4B-4). There is some indication that concentrations increased post-CRD to reach maximums of 0.27–0.48 ppm between 1982 and 1986 and then declined. Mean concentrations in Burbot were low (0.15 ppm) when last sampled in Areas 5 and 6 in 1998 (Figure 5.5.4B-4). Longnose Sucker did not show an obvious increase in mercury concentrations post-CRD when first measured in 1982, and the standard means of fish from Areas 2, 4, and 5 ranged from 0.07-0.11 ppm (Figure 5.5.4B-5).



In addition to the empirical data for fish captured pre-CRD and for several decades post-CRD from SIL, the increase in mercury availability following lake flooding has also been demonstrated experimentally for Yellow Perch. Mean muscle mercury concentrations of age 1+ fish (54–65 mm total length at the end of the experiment) from Granville Lake almost doubled from 0.043 to 0.083 ppm within 57 days after transfer into enclosures within the flooded zone of SIL Area 6 in the summer of 1989 (Ramsey 1990). The post-transfer values were similar to mean mercury concentrations of 0.06–0.11 ppm measured in feral 1-year-old Yellow Perch (58–72 mm total length) from SIL Areas 4, 5, and 6 in 1982 (Bodaly et al. 1987).

5.5.4.3.2 Issett Lake

The Issett Lake area was flooded by approximately 2,470% of its original area, greatly expanding the footprint of the reservoir to the west of the original lake location. Impoundment occurred in stages between May 1974 and November 1975, and from June to October 1976. Year 1975 was considered as year 1 post-flooding.

Mercury concentrations in Lake Whitefish increased significantly from 0.15 ppm for the one available pre-CRD survey sample in 1975 to a maximum of 0.32 ppm in 1978 (Figure 5.5.4-2). However, caution is warranted regarding the results for 1975 and 1978. The standard mean in 1975 was potentially elevated because of initial flooding effects (see Section 5.5.4.3). The arithmetic mean for 1978 was based on only five small (< 310 mm) fish, and may have underestimated or been not representative of the average concentration of the population at the time. For these reasons the IF of 2.2 (Table 5.5.4-3) is also questionable. Concentrations decreased after 1978 to 0.21 and 0.17 ppm in years 1982 and 1983 (Figure 5.5.4-2), the latter mean being significantly different from the maximum.

Concentrations in whitefish declined steadily each year from 1983 to 1986, remaining similar until 1988, and were significantly lower when sampled again in 1996 and 1998 (approximately 0.04 ppm). A minimum concentration of 0.023 ppm was observed in 2005, significantly lower than any of the previous means and the mean for the last sample in 2008. The return time was 27 years (Table 5.5.4-3).

Walleye had a maximum mercury concentration of 1.52 ppm based on five fish collected in 1978, four years after flooding (Figure 5.5.4-2). Concentrations decreased rapidly to 0.60 ppm by 1985, fluctuated substantially until 1996, then declined steadily to reach a minimum of 0.34 ppm in 2005. Concentrations were statistically similar to those in 2005 when last sampled in 2008. The peak concentration corresponded to an increase factor of 5.5 and the return time was 20 years (Table 5.5.4-3).

Mercury concentrations of pike increased from 1978 to 1983, reaching a maximum of 0.85 ppm nine years post-flooding (Figure 5.5.4-2). The maximum corresponds to an IF of 2.8 (Table 5.5.4-3). Concentrations steadily declined thereafter, except for a transient increase in 1992, and reached a minimum of 0.36 ppm in 2005. Identical to the pattern for Walleye, concentrations were statistically similar to those recorded in 2005 when last sampled in 2008. The return time was 23 years (Table 5.5.4-3).



Issett Lake

Year1975

19801985

19901995

20002005

Mus

cle

mer

cury

(pp

m)

0.0

0.5

1.0

1.5

2.0

Pike Walleye Whitefish

CRDflooding

*

*

*

**

*

*

*

*

*


Figure 5.5.4-2: Mean (95% Confidence limits) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Issett Lake for 1975–2008

5.5.4.3.3 Rat Lake

Rat Lake was flooded by approximately 158% of its original area. Impoundment occurred in stages between May 1974 and November 1975, and from June to October 1976. Year 1975 was considered as year 1 post-flooding.

Mercury concentrations in Lake Whitefish were highest (0.40 ppm) for the first post-CRD sample in 1978 (Figure 5.5.4-3); however, it is likely that the sample of five very large (mean fork length = 439 mm) fish did not well represent the average concentration of the population. Therefore, the second highest concentration of 0.25 ppm observed in 1984 was used as the peak concentration, increasing the time to reach the maximum from four to 10 years and resulting in an IF of 4.1 instead of 6.6 (Table 5.5.4-3). Mercury concentration fluctuated between 0.18 and 0.22 ppm from 1985 to1989, then rapidly decreased to a minimum of 0.03 ppm in 1994. Subsequently concentrations have significantly increased to approximately 0.07 ppm in 1996 and 1999. More recent concentrations are only available for samples of 2-3 fish.



Rat Lake

Year1975

19801985

19901995

20002005

20102015

Mus

cle

mer

cury

(ppm

)

0.0

0.5

1.0

1.5

2.0

2.5

3.0


CRDflooding

*

*

*


Figure 5.5.4-3: Mean (95% Confidence Limits) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Rat Lake for Years 1978–2005

Mercury concentrations in pike and Walleye from Rat Lake were almost identical both in absolute concentrations and in the post-CRD pattern of decline for most years of the available record (Figure 5.5.4-3). Only for the first few years, when sample sizes were small (22–26 fish) and included a commercial sample for pike, did the timeline of mercury levels diverge between the two species. Concentrations in Walleye were highest (2.4 ppm) when first sampled in 1978 four years after flooding, resulting in an IF of 8.7 (Table 5.5.4-3). The maximum concentration used for pike was the standard mean of 1.69 ppm recorded four years post-flooding and representing an IF of 5.7 (Table 5.5.4-3). A mean of 2.14 for five commercial samples for 1977 and a value of 2.32 for a single commercial sample in 1980 suggest that the maximum concentration in pike may have been higher. The decline in mercury concentrations from 1985 onwards was almost identical for both species (Figure 5.5.4-3), including a transient increase from 1992–1994, a minimum of 0.48 ppm (pike) and 0.41 ppm (Walleye) in 2005, and the subsequent increase, which was significant for Walleye in 2010 and 2013. The return time was 21 years for pike and 26 years for Walleye.



5.5.4.3.4 Notigi Lake

Notigi Lake was flooded by approximately 324% of the original area of lakes and creeks that became part of the reservoir. Impoundment occurred in stages between May 1974 and November 1975, and from June to October 1976. Year 1975 was considered as year 1 post-flooding.

Mercury concentrations in Lake Whitefish were first available in 1980 (i.e., six years after flooding) and the mean (0.12 ppm), which was based on six relatively large (mean fork length = 421 mm) fish, may not well represent the population average. It is likely that the maximum concentration of whitefish in Notigi Lake was not captured in monitoring programs. This is corroborated by the absence of an obvious peak in concentrations and the fact that the observed maximum of 0.27 ppm was measured 11 years post-flooding (Figure 5.5.4-4). Both these patterns are unusual for the evolution of mercury concentrations in boreal reservoirs (see Section 5.5.4.4.3), Whitefish mercury concentrations decreased significantly from 1985 to the next sample in 1999 (0.07 ppm), and then remained statistically similar until 2007, when last measured. Recognizing the shortcomings in the data record associated with the observed maximum concentration (see above), the IF was 4.3 and the return time was 18 years (Table 5.5.4-3).

Walleye had a maximum mean of 2.90 ppm recorded for a sample of four fish in 1980, six years after flooding (Figure 5.5.4-4). Further support for such a high concentration comes from a commercial sample for the same year with a mean of 2.58 ppm (DFO 1987). However, both these concentrations come from relatively large fish (mean fork length of the survey sample = 453 mm) and to obtain a more realistic and comparable mean concentration, data for 1980 were combined with the 29 fish sampled in 1981, resulting in a standard mean of 1.86 ppm. Concentrations decreased rapidly thereafter to 0.63 ppm in 1988, fluctuated between 0.55 ppm and 1.01 ppm for the following four samples until 2002, and reached a minimum of 0.45 ppm when last sampled in 2007. The minimum was significantly lower than the means for 2001 and 2002. The IF was 6.7 and the return time was 18 years (Table 5.5.4-3).

Mercury concentrations of pike reached a maximum of 1.95 ppm in 1980 (Figure 5.5.4-4), corresponding to an IF of 6.5 (Table 5.5.4-3). Concentrations decreased relatively slowly and included a transient increase to 1.69 ppm in 1988. Thereafter, the decrease in concentration was more steady, reaching a minimum of 0.59 ppm in 2002. The minimum was not significantly different from the preceding mean for 2001 and the last recorded mean in 2007. The return time was 21 years (Table 5.5.4-3).



Notigi Lake

Year1975

19801985

19901995

20002005

Mus

cle

mer

cury

(pp

m)

0.0

0.5

1.0

1.5

2.0

2.5

3.0


3.8 CRDflooding

* *

*

*

*

*

*

*

*

*

*

*

An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used. A number denotes an upper CL not shown as a bar. The mean for pike in 1977 is from commercial samples ; the stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.4-4: Mean (95% Confidence Limits [CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Notigi Lake for 1977–2007

5.5.4.3.5 Threepoint Lake

Threepoint Lake was flooded by approximately 44% of its original area between September 1976 and August 1977; 1977 was considered as year 1 post-flooding.

Mercury concentrations in Lake Whitefish were highest (0.39 ppm) for the first sample available post-CRD in 1980 (Figure 5.5.4-5). Subsequent means have been almost half of this values and it is likely that the only five relatively large (mean fork length = 429 mm) did not adequately represent the average concentration of the population in 1980. Similar to the approach for Walleye from Notigi Lake (see Section 5.5.4.3.4), data for fish collected in 1980 and 1981 (which represented the second highest concentration) were combined to estimate a maximum concentration of 0.251 ppm which was assigned to 1981. This resulted in an IF of 4.1 (Table 5.5.4-3). Mercury concentration decreased slowly until 1994, then more rapidly to less than 0.1 ppm in 1996 and 1998. Subsequently, slightly higher means were based on small (< 8) fish sample sizes until yearly sampling from 2011–2014 indicated stable



concentrations of approximately 0.06 ppm (Figure 5.5.4-5). These four means were not significantly different from the mean of 0.13 ppm for 2005, which represented only four fish and was associated with large confidence limits. The return time was 30 years (Table 5.5.4-3).

In contrast to SIL and similar to Rat Lake, mercury levels in Walleye and Northern Pike from Threepoint Lake were very similar both in absolute concentrations and in the post-CRD pattern of increase and decline. Both species reached similar maximum concentrations of 1.76 ppm (pike) and 1.65 ppm (Walleye) in 1983, seven years after flooding (Figure 5.5.4-5), representing IFs of 5.9 and 6.0, respectively. Concentrations in both species consistently declined to reach nearly identical minimums of 0.38 ppm (pike) and 0.35 ppm (Walleye) in 2005, resulting in return times of 22 years for pike and 18 years for Walleye (Table 5.5.4-3). Thereafter, concentrations increased and became significantly higher than the 2005 minimum by 2007 (Walleye) and 2010 (pike), and remained elevated (mostly significantly for Walleye) until last measured in 2014.

Threepoint Lake

Year1975

19801985

19901995

20002005

20102015

Mus

cle

mer

cury

(pp

m)

0.0

0.5

1.0

1.5

2.0

Pike

Whitefish Walleye

0.52*

*

*

*

* *

**

CRDflooding

An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used. A number denotes a CL not shown as a bar. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.4-5: Mean (95% Confidence Limits[CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Threepoint Lake for 1980–2014



Fish mercury data for Threepoint Lake also include several other species with mercury data that span multiple years and with means based on larger sample sizes than for most other lakes. These results illustrate that mercury concentrations of the mainly planktivorous Cisco are slightly higher, but of similar magnitude and show a similar pattern of post-CRD decline compared to the benthivorous Lake Whitefish (Figure 5.5.4-6). Two common catostomids, White Sucker and Longnose Sucker, which diet is best described as benthivorous detritivors (Ahlgreen 1990a, b) had mercury concentrations generally above 0.4 ppm from 1981–1992 (Figure 5.5.4-6). These values are consistently higher than those measured in Cisco and whitefish from Threepoint Lake and any other waterbody in Area 3 for a comparable time-period. Longnose Sucker concentrations were higher (generally significantly) when compared to their conspecifics from SIL (Figure 5.5.4B-5). Sauger from Threepoint Lake had mercury concentrations of 1.4–2.3 ppm for four samples measured from 1981–1985 (Figure 5.5.4-6), indicating that this species can reach similarly high concentrations as the two focal piscivores, pike and Walleye.

Threepoint Lake

Year1975

19801985

19901995

20002005

Mus

cle

mer

cury

(pp

m)

0.0

0.5

1.0

1.5

2.0

2.5

SaugerCisco White SuckerLongnose Sucker

**

*

*

*

CRDflooding

*


Figure 5.5.4-6: Mean (95% Confidence Limits) Length Standardized Muscle Mercury Concentrations of Sauger, Cisco, White Sucker, and Longnose Sucker From Threepoint Lake for Years 1981-2005



5.5.4.3.6 Wuskwatim Lake

Wuskwatim Lake was flooded due to CRD by approximately 44% of its original area between September 1976 and August 1977; 1977 was considered as year 1 post-flooding. Starting in 2007, a generating station was built just past the outlet of Wuskwatim Lake into the Burntwood River. Impoundment occurred from October to November 2011, flooding approximately 60 ha in the forebay for a Flood-1% of 0.7% (Manitoba Hydro and Nisichawayasihk Cree Nation 2003). If the approximately 400 ha of peat islands that will be flooded within Wuskwatim Lake are included, Flood-1% increases to 5.2%.

Mercury concentrations in Lake Whitefish were highest (0.25 ppm) for the first available sample in 1981, five years after flooding (Figure 5.5.4-7). Considering the temporal pattern in mercury concentrations of pike and Walleye, which have a more extensive record during the period immediately following flooding (see below), it is questionable if the 1981 mean for whitefish adequately represents the maximum concentration. Nevertheless, it was 3.1 times higher than the only available pre-CRD concentration of 0.08 ppm measured for commercial fish (Table 5.5.4-3). Standard means decreased gradually from 1981 onwards to a minimum of 0.05 ppm in 2005 and then significantly increased in 2007 and 2014 (Figure 5.5.4-7). The return time was 24 years (Table 5.5.4-3).

Mean mercury concentrations in Walleye increased to a maximum of 0.94 ppm within four years after flooding (Figure 5.5.4-7). The maximum was almost three times the average concentration of 0.34 ppm, determined from six commercial samples collected pre-CRD (Table 5.5.4-3). Concentrations declined almost continuously from 1981 onwards and reached a minimum of 0.24 ppm in 1996, remained steady until 2005 and then significantly increased in 2007 (Figure 5.5.4-7). The concentration in 2014 was still elevated but not significantly higher than in 2005. The return time was 17 years (Table 5.5.4-3).

Mercury concentrations in Northern Pike concentrations showed a similar increase and decline compared to Walleye for three consecutive years starting with the first measurement in 1979, but then increased to a maximum of 1.42 ppm in 1984, seven years after the end of flooding (Figure 5.5.4-7). The maximum corresponded to an IF of 4.8 (Table 5.5.4-3). Thereafter, concentrations declined rapidly to 0.34 ppm in 1996, remained statistically similar until 2002, and reached a minimum of 0.31 ppm in 2005 (Figure 5.5.4-7). The minimum was significantly lower than the previous mean for 2002 but statistically similar to the means in 1996 and 1998. Concentrations significantly increased in 2007 compared to the 2005 minimum and were just slightly lower when last sampled in 2014 (Figure 5.5.4-7). However, the last mean was based on only 12 fish and the associated large confidence limits resulted in a non-significant differences compared to the 2005 minimum. The return time was 21 years (Table 5.5.4-3).

Mercury concentrations in all three focal species for 2014 were slightly lower and statistically similar compared to 2007, indicating that the 2011 flooding by the GS has not (yet) measurable affected mercury bioaccumulation rates in fish from Wuskwatim Lake.



Wuskwatim Lake

Year1970

19751980

19851990

19952000

20052010

2015

Mus

cle

mer

cury

(pp

m)

0.0

0.4

0.8

1.2

1.6PikeWalleyeWhitefish

CRDflooding

*

*

*

*

*

*

*

Means without CL prior to 1980 are from commercial samples. An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.4-7: Mean (95% Confidence Limits) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Wuskwatim Lake for 1970–2014

5.5.4.3.7 Leftrook Lake

Leftrook Lake is located north of the CRD route and was not affected by flooding. Leftrook Lake is a CAMP reference lake and the off-system waterbody with the most complete historic record of fish mercury concentrations in the ROI.

Mercury concentrations in Lake Whitefish were highest (0.05 ppm) when first measured for 36 large (mean fork length = 424 mm) fish in 1992 (Figure 5.5.4-8). Because the relationship between mercury concentration and fish length was not significant, only the arithmetic mean is available for the 1992 sample and a direct comparison to the subsequent nine standard means cannot be made. From 1994 onwards until the last sample in 2014, concentrations ranged from 0.018 to 0.037 ppm. Although a few of the means for this period were significantly different, no consistent temporal pattern in concentrations was evident, and it is likely that the observed fluctuations in mercury concentrations were part of natural temporal variability.



Walleye and pike were first sampled for mercury from 1981–1983 by the DFO Inspections Branch. The small number of Walleye analysed were relatively large (> 412 mm) headless dressed fish or small (352 mm) dressed fish. The actual (fork) length of a headless dressed fish is unknown and has been estimated using a conversion factor. Only arithmetic means could be calculated for the above three samples. The mean of 0.73 ppm for 1981 likely represents an outlier, although the standard mean for the first sample of 36 survey fish in 1992 indicates that concentrations have exceeded 0.4 ppm in the past (Figure 5.5.4-8). From 1994–2014, standard means of Walleye ranged from 0.21–0.29 ppm, have been associated with very small confidence limits, and have been statistically similar except for the high and low means observed in 2001 and 2014, respectively.

Leftrook Lake

Year1975

19801985

19901995

20002005

20102015

Mus

cle

mer

cury

(pp

m)

0.0

0.2

0.4

0.6

0.8

Pike Walleye Whitefish

1.0*

*

*

*

*

An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used. The data point without a CL for 1977 is the mean of two commercial samples (0.25 and 0.33 ppm) from 1977 and 1981 which cannot be assigned to a specific year (see Rannie and Punter 1987). A number denotes a CL not shown as a bar. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.4-8: Mean (95% Confidence Limits [CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Leftrook Lake for 1981–2014

Except for a mean of 0.39 ppm in 2010, mercury concentrations in pike have ranged from 0.19 ppm to 0.27 ppm over the entire record from 1981–2014. Excluding the anomalous result for 2010, standard means have been statistically similar except for the minimum recorded in 2014, which was significantly different from the means for 1992 and 2007.





Mercury concentrations in piscivorous species (i.e., pike and Walleye) from years pre-dating hydroelectric development in Area 3 were mostly considerably lower than the Health Canada standard of 0.5 ppm for the commercial sale of fish (Table 5.5.4-1). These relatively low concentrations are even more remarkable when considering that these data were obtained from commercial catches of fish that likely included a bias towards larger individuals than most of the fish analysed from post-CRD survey samples. Mean pre-CRD mercury concentrations of the benthivorous focal species, Lake Whitefish, from commercial samples (Table 5.5.4-1) but also from the samples of individual fish from SIL, Issett, and Cousins lakes in 1975 (see Section 5.5.4.3) were always much lower than the 0.5 ppm standard and often less than 0.1 ppm.

Following CRD, the concentrations of mercury in Walleye and Northern Pike in all focal lakes and several other waterbodies in Area 3 increased to (substantially) above 0.5 ppm (see Figure 5.5.4A-1 for pike), and as discussed in more detail for other areas (see Section 5.5.2.3.1), were often no longer accepted for marketing in a commercial fishery (Green 1986; Derksen and Green 1987). The period during which concentrations remained above 0.5 ppm in pike and Walleye varied among lakes, but typically lasted for two or more decades (see discussion on return times in Section 5.5.4.4.3). Based on the concentrations of 1.4–2.3 ppm measured in Sauger from Threepoint Lake for 1981–1985, this species may have experienced increases in post-CRD mercury concentrations at least similar to those observed in the two focal piscivores, pike and Walleye.

The recent (post-2005) significant increases in mercury concentrations of mainly pike and Walleye observed for almost all on-system waterbodies in Area 3 have resulted in concentrations that exceed the Health Canada 0.5 ppm standard for some lakes (e.g., Rat and Threepoint).

The post-CRD increases in mercury concentrations of Lake Whitefish never approached the 0.5 ppm limit for the commercial sale of fish, reaching maximums of 0.20–0.25 ppm in some lakes. Although quite variable, the three standard means for 1983–1985 and the arithmetic mean in 1992 for Longnose Sucker from Threepoint Lake at least suggest that concentrations in non-piscivorous species may have approached or even exceeded 0.5 ppm in some of the more extensively flooded lakes (Figure 5.5.4B-5).

5.5.4.4.2 Current Fish Mercury Concentrations in Reference Lakes

For an assessment of current fish mercury concentrations of on-system waterbodies in the ROI it is important to establish a benchmark of “natural” concentrations against which a comparison can be made. Off-system lakes, unaffected by hydrological changes and other human impacts that may alter the local conditions of methylmercury availability to fish, were assumed to represent natural conditions. Relatively few off-system lakes with recent fish mercury data exist in the ROI. To obtain a broader geographical representation of “natural background concentrations” for each of the three focal species, results for off-system lakes from other areas were used in addition to those for Leftrook Lake, the CAMP off-system reference lake for the CRD Region. This also included Granville Lake which is located just west of Area 3.



Granville Lake is the CAMP off-system reference lake for the Upper Churchill River Region and has a long-term data set on fish mercury concentrations, including from recent sampling. Furthermore, to base the estimate of background concentrations on a broad data base, and because there was no temporal trend in the yearly means, mercury concentrations for years 2002–2014 were considered “current” and included in the calculation of the overall mean for each species (Table 5.5.4-2).

With one exception (pike from Leftrook Lake in 2010), mercury concentrations of Walleye and pike from the off-system Leftrook, Gauer and Assean lakes have ranged from slightly below 0.2 ppm to slightly above 0.3 ppm since 2002, with overall means for each lake at or marginally above 0.2 ppm (Table 5.5.4-2). Pike from Setting Lake and pike and Walleye from Granville Lake had mean mercury concentrations between 0.32 and 0.38 ppm since 2002 (Table 5.5.4-2). Gauer Lake generally had the lowest and Granville Lake generally had the highest mercury concentrations in pike and Walleye of all the off-system lakes. For whitefish the pattern differed, as mean concentrations in fish from Assean Lake were twice as high in those from Setting Lake (Table 5.5.4-2).

Granville Lake is considered unaffected by hydroelectric development (Water Regime, Section 4.3.3.2.3) and the cause of the elevated mercury concentrations in pike, that were fairly consistently observed from 1979-1992 (Figure 5.5.4-9) remains unknown. These relatively high concentrations were all based on commercial samples and, primarily, arithmetic means for relatively few, large fish (as was the entire record for pike from 1970–2004). Granville Lake was excluded from the calculation of background concentrations based on recent data from off-system lakes (Table 5.5.4-2).



Table 5.5.4-2: Standard Mean Mercury Concentrations, Part-per-Million, for Northern Pike, Walleye, and Lake Whitefish from Five Coordinated Aquatic Monitoring Program Off-System Reference Lakes for Years 2002–2014

Lake 2002-07 2010 2011 2012 2013 2014 Mean

Northern Pike

Leftrook 0.277 0.392 2 0.242 0.248 0.229 0.187 0.252

Gauer 0.182 0.202 - - 0.195 - 0.194

Granville 0.400 1 0.441 - - 0.309 2 - 0.370

Setting - 0.392 - - 0.314 - 0.352

Assean 0.194 0.248 - - 0.233 - 0.226

All lakes * - - - - - - 0.238

Walleye

Leftrook 0.266 0.255 0.245 0.273 0.234 0.293 0.260

Gauer 0.193 0.246 - - 0.180 - 0.211

Granville 0.229 1 0.441 - - 0.278 2 - 0.340

Setting - 0.277 - - 0.206 2 - 0.241

Assean 0.203 0.235 - - 0.317 2 0.243

All lakes * - - - - - - 0.239

Lake Whitefish

Leftrook 0.037 4 0.026 0.026 0.021 0.027 0.031 0.024

Gauer - 0.036 - - 0.034 - 0.035

Granville - 0.052 - - 0.046 - 0.049

Setting - 0.025 1 - - 0.025 - 0.025 1

Assean 0.057 0.039 2 - - 0.060 1 - 0.046

All lakes * - - - - - - 0.033

Note: Mean represents the standard means calculated from the raw data for all fish by species and lake across years.

1. Arithmetic mean (relationship between fish length and mercury concentration was not significant). 2. Significantly different from other standard means for the lake. 3. Significantly different from 2011 and 2012. * Excluding Granville Lake.



Granville Lake

Year1970

19751980

19851990

19952000

20052010

2015

Mus

cle

mer

cury

(pp

m)

0.0

0.2

0.4

0.6

0.8Pike Walleye Whitefish

*

*

*

**

*

*

*

*

*

***

*

*

*

*

***

CRDflooding

* *

Means without CL prior to 1980 are from commercial samples. An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used. Except for 1981 and 1999, all arithmetic means were for 10 or less fish. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.4-9: Mean (95% Confidence Limits [CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Granville Lake Lake for 1970–2014

The recent mercury concentrations observed in the three focal species from the ROI off-system lakes listed in Table 5.5.4-2 are remarkably similar to the concentrations obtained from Area 3 off-system lakes (which, with the exception of Granville Lake represent a different set of lakes) during years pre-dating hydroelectric development (Table 5.5.4-1). Combined with the relatively stable mercury concentrations observed in fish from Leftrook Lake between 1976 and 2014 (see Section 5.5.4.3.7), this indicates that baseline mercury concentrations in the three best-studied fish species have changed little in Area 3 off-system waterbodies over the past 30 years.



5.5.4.4.3 Evolution Fish Mercury Concentrations in Relation to Hydroelectric Development

CHURCHILL RIVER DIVERSION

This section will summarize results from Area 3 for the time course of fish mercury concentrations in response to flooding from CRD. To establish common generalities in the response but also to document the range in responses, results from the three reservoirs in Area 2 (Stephens Lake, Long Spruce and Limestone forebays) were included to more fully document temporal change in mercury concentrations of the three focal species in the entire ROI.

Table 5.5.4-3 summarizes several parameters that characterize the evolution in mercury concentrations for Lake Whitefish, Northern Pike and Walleye from reservoirs and lakes that experienced flooding due to hydroelectric development. The percent flooding for each lake/reservoir is indicated, as well as mercury concentrations prior to flooding (Hg pre-flood; if available) and recorded maximum (Hg max) and minimum (Hg min) post-flood concentrations. The number of years to reach Hg min after reaching Hg max (return time) is also provided. To obtain a measure of the relative increase in mercury concentrations due to flooding, an increase factor (IF) was calculated as the ratio of Hg max to Hg pre-flood. Given the absence of pre-Project data for most of the lakes, pre-flooding mercury concentrations were mostly estimated based on the average (weighted by sample size) concentration from three on-system lakes with existing pre-flooding data: 0.060 ppm for whitefish, 0.298 ppm for pike, and 0.275 ppm for Walleye (see Table 5.5.4-1). These estimates are for composite samples of fish from commercial catches. Although pre-CRD (1975) samples for whitefish from several areas of SIL and from Cauchon Lake indicate that samples based on individual fish may provide similar results than the commercial samples (see Section 5.5.4.3), as documented on multiple previous occasions, mercury concentrations from commercially caught fish, particularly pike and Walleye for a specific waterbody and year are often higher than standard means and likely overestimate the population as a whole. However, based on current mercury concentrations in fish from off-system lakes (Table 5.5.4-2), this difference may be small for the estimates of Hg pre-flood used in Table 5.5.4-3. Nevertheless, the IF values presented in Table 5.5.4-3 likely represent underestimates, particularly for fish from waterbodies for which sampling started 10 or more years after flooding and where Hg max was recorded in the first sampling year (e.g., Stephens lake, Long Spruce forebay).

Similarly, most of the “return” times listed in Table 5.5.4-3 represent underestimates of the time it takes for fish to return to a concentration assumed to have existed pre-flooding (i.e., Hg pre-flood) or that represents a new baseline concentration (also referred to as “current background concentrations” as measured from off-system lakes for 2002–2014; Table 5.5.4-2). This applies in particular to those waterbodies and species for which the year Hg min was recorded represented the last year of sampling (e.g., Walleye in Notigi Lake) and the concentrations were (still) substantially higher than current background. Assuming that the overall mean (0.24 ppm for pike and Walleye) for the four off-system lakes is a reasonable approximation of current background concentrations, Hg min was still significantly higher for pike and Walleye from all lakes on the CRD route, except for Walleye from SIL Area 4. Conversely, concentrations representing Hg min for pike and Walleye from the three reservoirs on the lower Nelson River in Area 2 (see Table 5.5.4-3) were similar to or even significantly lower (e.g., pike and



Walleye from Limestone forebay) than current background concentrations. Such a difference between ROI Areas 2 and 3 was not observed for whitefish. For this species, Hg min was mostly higher than background in both areas (0.033 ppm; see Table 5.5.4-2), except for whitefish from SIL Area 6, and Stephens and Issett lakes (Table 5.5.4-3).

Irrespective of the limitations of the available data, the evidence from the many flooded lakes and reservoirs in the ROI clearly shows that inundation of soils and vegetation resulted in substantially elevated fish mercury concentrations compared to those existing pre-flood or for waterbodies not affected by hydroelectric development. The underlying causes for this phenomenon have been extensively studied (particularly within the ROI) and are well understood. Bodaly and Hecky (1979) were the first to attribute the increase in fish mercury concentrations after flooding to a greater supply of inorganic mercury from the leaching of humic rich soils and an increase in mercury bioavailability due to bacterial mercury methylation. This conclusion was based mainly on the mercury results for whitefish from Southern Indian Lake (SIL) and Issett Lake, along with accompanying limnological research. These authors further postulated that mercury concentrations would be highest in top-level predators such as Northern Pike and Walleye due to the biomagnification of methylmercury through the food chain and that the increase in fish mercury concentrations would be in some proportion to the enhanced supply and increased methylation rate of mercury. These hypotheses have been confirmed and the knowledge base expanded over the last three decades (Bodaly et al. 1984a b, 1997, 2007; Hecky et al. 1991; Lucotte et al. 1999; Schetagne et al. 2003). This includes studies on experimental reservoirs that have greatly enhanced the understanding of the processes affecting rates and persistence of mercury methylation and the bioavailability of methylmercury (Kelly et al. 1997; St Louis et al. 2004, Bodaly et al. 2004; Hall et al. 2005). Predictive models also have been developed to quantify expected mercury levels in fish following flooding (e.g., Johnston et al. 1991; Harris and Hutchinson 2009).

The results presented in Table 5.5.4-3 differ to some degree from similar data previously published in Bodaly et al. (2007) for several reasons:

• Additional data on fish mercury concentrations have been obtained from previously unavailable sources and in a few cases this has changed the value of Hg max and the time to reach Hg max.

• For the current assessment, standard means were only used in those cases where the regression equation for the relationship between mercury concentration and fish length was significant. If this was not the case, the arithmetic mean was used. Also, if (arithmetic) means from two adjacent years represented the highest and second highest concentrations for a species in a particular lake and were based on small sample sizes, data from both years were combined to calculate a standard mean (e.g., Walleye from Notigi Lake; pike and Walleye from Osik Lake).

• The return times listed in Table 5.5.4-3 were estimated based on the changes in fish mercury concentrations within each lake. In contrast, Bodaly et al. (2007) assigned return times for fish from each lake by using the first year for which the lower limit of the 95% confidence interval of an observed mean concentration overlapped with the upper limit of the 95% confidence interval for the mean of a large number (> 60) of “pristine” lakes. However, these lakes included waterbodies that were impacted by human activities (e.g., Saskatchewan River), had relatively high mercury concentrations in at least some resident fish species for unknown reasons (e.g., LeClair, Drunken, and Granville lakes) and/or the mean concentrations for a particular year was based on few, large



fish. The resulting “background mean concentrations” were quite high (approximately 0.41 ppm for pike and Walleye) and were associated with relatively large confidence intervals.

• Many of the estimates of lake areas prior to and post flooding, and the resulting percentage of flooding differed between the current data set and those previously published (e.g., Newbury et al. 1984, Bodaly et al. 2007). Pre- and post-flooding areas, particularly for Area 3, have recently been re-calculated (see Appendix 5.5.4C) and for some waterbodies the percentage of flooding as calculated in Section 5.5.1.4 is substantially smaller or larger than the existing estimates. This obviously will affect any analysis relating Hg max to Flood (%). Except for Issett Lake, for which estimates of the percentage of flooding increased from 455% to 2,467% (Table 5.5.4-3), all other waterbodies were included with their recent area estimates and percentage of flooding in the respective analysis (see Figure 5.5.4-10 and Table 5.5.4-3).

The current assessment indicates that maximum mercury concentrations in pike from waterbodies in the ROI ranged from 0.35 to 1.95 ppm, corresponding to IFs of 1.1-6.5 (Table 5.5.4-3). Considering the only few available samples based on relatively few fish for Opachuanau Lake for 1978–1988, the Hg max of 0.35 ppm likely does not represent the actual Hg max of the pike population. Maximum concentrations in pike from both regions of SIL were 1.5 times to more than two times higher than those for fish from the Limestone forebay, which, according to the most recent calculations was flooded to a similar degree (7-8%). Despite such large differences in fish mercury concentrations for similar amounts of flooding between some waterbodies, a significant linear relationship existed between (log) Hg max and (log) Flood(%), with percent flooding explaining almost half of the variability in Hg max (Figure 5.5.4-10). For the 10 waterbodies with sufficient data to allow an estimate, return times ranged from 8 (Limestone forebay) to 30 (SIL, Area 4) years.

Maximum mercury concentrations in Walleye, excluding Opachuanau Lake, ranged from 0.54 to 2.38 ppm, corresponding to IFs of 2.0–8.7 (Table 5.5.4-3). Unlike for pike, Hg max for Walleye from both regions of SIL were similar than those for their conspecifics from the Limestone forebay. There was a significant linear relationship between (log) Hg max and (log) Flood(%), the latter explaining 64% of the variability in Hg max (Figure 5.5.4-10). For the 10 waterbodies with sufficient data to allow an estimate, return times ranged from 13 (Limestone forebay) to 30 (SIL, Area 4) years.

The number of waterbodies with sufficient data to estimate parameters that characterize the response of mercury concentrations to lake flooding was smaller for whitefish than for the two piscivorous species (Table 5.5.4-3). Excluding Isset Lake (see Section 5.5.4.3.2), maximum mercury concentrations in whitefish ranged from 0.14 to 0.27 ppm. However, the corresponding IFs of 1.7 to 4.3 were similar or even higher (e.g., Long Spruce forebay, SIL) compared to pike and Walleye from lakes with little or moderately flooding (i.e., < 100%), whereas the differences in IF between the benthivore and the two piscivores were larger for the extensively flooded lakes (Table 5.5.4-3). Similar to pike, Hg max and, particularly the IF for whitefish from both regions of SIL were substantially higher than those for their conspecifics from the Limestone forebay. However, unlike pike and Walleye, the relationship between Hg max and percentage flooding was not significant for whitefish. For the nine waterbodies with sufficient data to allow an estimate, return times ranged from 0 (i.e. Hg max was never significantly different from any of the following standard means) for the Limestone forebay to 30 years for Threepoint Lake (Table 5.5.4-3).



Table 5.5.4-3: Average mercury concentration prior to flooding (Hg pre-flood), maximum post-Churchill River Diversion mercury concentration (Hg Max), the number of years to reach Hg Max after the end of flooding, minimum mercury concentration post-Churchill River Diversion (Hg Min), the year Hg Min was observed, the increase factor, and the return time for Lake Whitefish, Northern Pike, and Walleye from fifteen Area 3 lakes flooded by Churchill River Diversion and three reservoirs in Area 2

Lake / Reservoir Flood (%)* Species Hg pre-flood (ppm) Hg Max (ppm) Years to Hg Max Increase

Factor** Hg Min (ppm) Year of Hg Min (post Hg max) Return time (years) Increase post Hg Min

SIL, Area 6 7 Whitefish 0.06 0.238 4 (79) 4.0 0.026 2005 (26) 26 (79-05) no Pike 0.29 1.095 9 (84) 3.8 0.317 2005 (21) 24 (82-05) yes Walleye 0.23 0.803 3 (78) 3.5 0.334 2005 (27) 27 (78-05) yes

SIL, Area 4 7 Whitefish 0.053 0.218 3 (78) 4.1 0.067 1 1986 1 n/a n/a Pike - 0.704 9 (84) 2.4 0.371 1 2010 1 <30 no Walleye - 0.607 5 (80) 2.6 0.193 2013 10 (33) 30 (80-10) n/a

Issett 2467 Whitefish 0.146 12 0.324 12 3 (78) 2.2 0.023 2005 (27) 27 (78-05) yes Pike - 0.846 9 (83) 2.8 0.361 2005 (22) 23 (82-05) no 11 Walleye - 1.520 2 4 (78) 5.5 0.342 2005 (27) 20 (78-98) no 11

Rat 158 Whitefish - 0.250 3 10 (84) 4.1 0.031 1994 (10) 10 (84-94) yes Pike - 1.688 4 4 (78) 5.7 0.475 2005 (25) 21 (78-99) no Walleye - 2.380 4 (78) 8.7 0.409 2005 (27) 26 (78-05) yes

Notigi 324 Whitefish - 0.267 11 (85) 4.3 0.066 2001 (16) 18 (81-99) yes Pike - 1.948 6 (80) 6.5 0.591 2002 (22) 21 (80-01) no Walleye - 1.857 5 7 (81) 6.7 0.452 2007 10 (26) 18 (81-99) n/a

Threepoint 44 Whitefish - 0.251 6 5 (81) 4.1 0.054 2014 10 (33) 30 (81-11) n/a Pike - 1.758 7 (83) 5.9 0.377 2005 (22) 22 (83-05) yes Walleye - 1.654 7 (83) 6.0 0.345 2005 (22) 19 (83-02) yes

Wuskwatim 50 Whitefish 0.08 0.248 5 (81) 3.1 0.049 2005 (24) 24 (81-05) yes Pike - 1.418 8 (84) 4.8 0.305 2005 (21) 21 (84-05) yes Walleye 0.34 0.944 4 (80) 2.8 0.242 1996 (16) 17 (79-96) yes

Footprint 87 Whitefish - 0.225 4 (80) 3.6 - - - n/a Pike - 1.479 8 (84) 5.0 - - - n/a Walleye - 1.229 7 (83) 4.5 - - - n/a

Osik 55 Pike - 1.457 7 7 (83) 4.9 - - - n/a Walleye - 0.917 8 7 (83) 3.3 - - - n/a

Apussigamasi 69 Pike - 1.177 8 (84) 3.9 - - - n/a Walleye - 1.528 8 (84) 5.6 - - - n/a

Karsakuwigamak 105 Whitefish - 0.216 - # (84) 3.5 - - - n/a Pike - 1.405 - # (84) 4.7 - - - n/a Walleye - 1.972 - # (84) 7.2 - - - n/a

Mynarski, West 38 Pike - 0.805 10 (84) 2.9 - - - n/a Walleye - 0.922 10 (84) 3.4 - - - n/a

Mynarski, East 37 Walleye - 0.567 9 ≤7 (81) 2.1 - - - n/a

Pemichigamau 118 Pike - 1.191 9 - # (85) 4.0 - - - n/a Walleye - 1.471 9 - # (85) 5.3 - - - n/a



Table 5.5.4-3: Average mercury concentration prior to flooding (Hg pre-flood), maximum post-Churchill River Diversion mercury concentration (Hg Max), the number of years to reach Hg Max after the end of flooding, minimum mercury concentration post-Churchill River Diversion (Hg Min), the year Hg Min was observed, the increase factor, and the return time for Lake Whitefish, Northern Pike, and Walleye from fifteen Area 3 lakes flooded by Churchill River Diversion and three reservoirs in Area 2

Lake / Reservoir Flood (%)* Species Hg pre-flood (ppm) Hg Max (ppm) Years to Hg Max Increase

Factor** Hg Min (ppm) Year of Hg Min (post Hg max) Return time (years) Increase post Hg Min

Moss 52 Pike - 0.793 - # (85) 2.7 - - - n/a Walleye - 1.075 9 ≤7 (82) 3.9 - - - n/a

Opachuanau 7 Pike - 0.349 - # (81) 1.1 - - - n/a Walleye - 0.303 - # (81) 1.1 - - - n/a

McLeod 14 Pike 0.559 - # (83) 1.9 - - - n/a Walleye 0.540 - # (85) 2.0 - - - n/a

Stephens 226 Whitefish - 0.191 <14 (84) 3.2 0.029 2005 (21) 22 (83–05) yes Pike - 1.045 9 ≤13 (83) 3.5 0.180 2005 (≥22) ≥22 (83–05) yes Walleye - 1.890 9 ≤11 (81) 6.9 0.204 2005 (≥24) ≥24 (81–05) yes

Long Spruce 92 Whitefish - 0.175 9 ≤8 (85) 2.9 0.073 1993 (≥8 ) ≥7 (85–92) n/a Pike - 0.697 9 ≤8 (85) 2.3 0.231 10 2003 (≥18) ≥18 (85–03) n/a Walleye - 0.641 9 (86) 2.3 0.239 10 2003 (≥17) ≥17 (86–03) n/a

Limestone 8 Whitefish 0.082 0.136 2 (91) 1.7 0.079 2003 (≥12) 0 yes Pike 0.324 0.453 5 (94) 1.4 0.188 2001 ( 7) 8 (93–01) yes Walleye - 0.639 5 (94) 2.3 0.184 2005 (11) 13 (92–05) yes

Notes: parts-per-million (ppm) A significant increase in mercury concentrations after reaching Hg Min is indicated (Increase post Hg Min). The return time is the Number of years from the earliest year mercury concentrations were not significantly different from Hg Max to the earliest year concentrations were statistically similar to Hg Min. * Flood (%) = (flooded area/pre-flooding area) x 100.** Ratio of Hg Max to Hg pre-flood; IF for lakes without pre-CRD data was estimated as the weighted average of the means for Opachuanau, SIL Area 6, and Wuskwatim lakes (Table 5.5.4-1): 0.062 ppm for Lake Whitefish, 0.298 ppm for Northern Pike, and 0.275 ppm for Walleye.# = Insufficient data to assign.1. No data exist for years 1989-2009; the results for 1986 or 2010 likely don’t represent Hg min.2. Arithmetic mean is for 5 fish; next highest Hg max is a standard mean of 1.13 ppm for 1986.3. An arithmetic mean of 0.396 ppm for 5 large (mean fork length = 439 mm) for 1978 exist.4. Standard mean; a mean of 2.14 ppm exists for 5 commercial samples in 1977.5. Mean is the standard mean for 4 fish from 1980 and 29 fish from 1981.6. Mean is the standard mean for 5 fish from 1980 and 16 fish from 1981.7. Mean is the arithmetic mean for 5 fish from 1982 and 5 fish from 1983.8. Mean is the arithmetic mean for 6 fish from 1982 and 12 fish from 1983.9. When measured for the first time.10. Last year of sampling.11. Only one sample for 2008 is available.12. Likely represents an overestimate of the actual population means at the time (see Section 5.5.4.3.2)



Northern Pike

5 10 20 50 100 200 300

Mus

cle

Mer

cury

(pp

m)

0.2

0.3

0.5

1

2

3

SIL6

SIL4

Opach

LSt

MCL

My-W

3-Pt

Wus

Moss

Osik

Apus

FPr

LSp

Kar

Pemi

Rat

Stp

Not

log (Hg) = -0.461 + 0.279 (log Flood%) ; r2=0.49

Walleye

Flooding (%)5 10 20 50 100 200 300

Mus

cle

Mer

cury

(pp

m)

0.2

0.3

0.5

1

2

3

SIL6

SIL4

Opach

LSt

MCLMy-E

My-W

3-Pt

Wus

Moss

Osik

Apus

FPr

LSp

Kar

Pemi

Rat

Stp Not

log (Hg) = -0.605 + 0.368 (log Flood%) ; r2=0.64

Regression equation (log Hg) = a + b (log Flood%). The Best-fit regression and the corresponding regression lines are shown.

Figure 5.5.4-10: Maximum Observed Mercury Concentration in Northern Pike and Walleye From Lakes and Reservoirs in Areas 2 and 3



Based on the large amount of data on fish mercury concentrations from almost 20 lakes spanning more than four decades, a generalized timeline for the rise and fall of concentrations after lake flooding can be obtained for the ROI (Figure 5.5.4-11).

3-9 years

~10 - >30 years

Time (years)0 5 10 15 20 25 30 35 40 45

Mus

cle

mer

cury

(ppm

)

0.0

0.5

1.0

1.5

2.0

Start offlooding

The time to reach maximum concentrations (blue font) and the return time to pre-flood or current background concentrations (green font) is indicated.

Figure 5.5.4-11: Generalized Timeline (Red Line) of Changes in Fish Mercury Concentrations Based on Results From Reservoirs and Flooded Lakes Due to Hydroelectric Development in the Region Of Interest

Figure 5.5.4-11 shows that fish mercury concentrations increased relatively rapidly after flooding to reach a very transient maximum, after which concentrations declined over an extended period of time, relatively quickly at first, more gradually later. Maximum concentrations in piscivorous species from the more extensively flooded lakes reached approximately 2 ppm, whereas maxima in large bodied species of lower trophic levels generally did not exceed 0.5 ppm and remained below 0.3 ppm in Lake Whitefish. Maximum concentrations were usually reached within 3–9 years post-flood in pike, Walleye, and whitefish, with the latter species often being at the lower end of the range. The estimated time to reach minimum concentrations after peaking was quite variable between lakes and species, spanning a range from slightly less than 10 years to 30 years. In almost all cases minima were either measured in the last year that sampling was conducted in a given waterbody or occurred in 2005 and were followed by significantly higher concentrations in subsequent years.



DOWNSTREAM EFFECTS OF HYDROELECTRIC DEVELOPMENT ON FISH MERCURY CONCENTRATIONS IN AREA 3

Increases in fish mercury concentrations in downstream environments have been attributed to the export of methylmercury from reservoirs in Manitoba (Johnson et al. 1991; Bodaly et al. 2007), Québec (Schetagne et al. 2003) and Newfoundland and Labrador (Harris and Hutchinson 2007). It is difficult to quantify the effect of downstream transport on fish mercury concentrations because of the paucity of similar assessments in the literature and the potential for site and species specificity of the downstream effect (Bodaly et al. 2007; Harris and Hutchinson 2007). There is little direct evidence for a downstream effect of reservoir flooding on fish mercury concentrations within Area 3. Large waterbodies on a diversion route (such as CRD) greatly reduce further export downstream because they act as methylmercury sinks via sedimentation of suspended particular matter and local predation of fish food organisms with elevated mercury concentrations (Schetagne and Verdon 1999a; Schetagne et al. 2003). Thus, the magnification of a local increase in fish mercury concentration from a flooded lake located upstream (i.e., a downstream effect) should conceptually only happen for pairs of adjacent lakes on the CRD route if at all. However, no empirical studies exist that would allow discrimination of any upstream contributions to the environmental methylmercury pool and its subsequent uptake through the food chain into the focal fish species in a downstream lake or reservoir.

The general pattern of rise and fall in fish mercury concentrations is similar for all flooded lakes and it is difficult to find a reflection of some of the detailed temporal changes in concentrations from any of the lakes in the immediate downstream lake. These patterns (or lack thereof) do not exclude the possibility of a downstream effect, but if one existed, the effect was too small to be detected.



5.5.5 Area 4: Missi Falls Control Structure to the Churchill River Estuary

The Churchill River downstream of SIL has experienced substantial alterations in its hydrological regime as a result of CRD. With the completion of the Missi Falls CS and the South Bay diversion out of SIL in the fall of 1976, most of the flow of the Churchill River was diverted into the Nelson River basin. According to Section 4.3.3.3.3 the diversion reduced Churchill River discharge by an average of 780 m3, representing approximately 79% of the annual mean natural flow at Missi Falls CS. In addition to changes in flows, there have been variations in the timing of peak flows. Pre-CRD, mean monthly flows peaked in July at the outlet of Fidler Lake and in August at Red Head Rapids. Post-CRD, mean monthly flows are shown to peak during June at the outlet of Fidler Lake and at Red Head Rapids (see Section 4.3.3.3.3). The first peak at Red Head Rapids is likely due to the influence of inputs from major tributaries (e.g., Little Churchill River and Little Beaver River).

Churchill River flows have also fluctuated substantially in response to natural causes (e.g., high/low runoff and/or high/low precipitation in the drainage basin) and enhanced by Hydro regulation (e.g., reduced discharge at the Notigi CS). As a result, the Churchill River discharge has fluctuated widely since 1978 (first year of records). Recent flood events, where post-CRD 95%ile flows at Missi Falls have been exceeded for a period of more than two weeks, happened during July to November 2005, August 2006, July to August 2009, and August to September 2011 (based on data introduced in Section 4.3.3.3). Reductions in flow and water level have resulted in an increase in the amount of non-wetted river channel. As elaborated in more detail in Erosion and Sedimentation, Section 4.4.2.5, channel width has been reduced by approximately 30% and an area of about 203.6 km2 has been dewatered between Missi Falls CS and Mosquito Point.

In October 1999, as part of the Manitoba Hydro and the Town of Churchill the Lower Churchill River Water Level Enhancement Weir Project, a weir (“the Churchill Weir”) was constructed across the Churchill River just upstream of Mosquito Point (see Section 4.3.3.3.3). The river was deepened by increasing the water level by approximately 2.0 m along a 10 km reach between the weir and the Town. Construction of the weir and ancillary components was carried out to improve boating, provide access to a source of drinking water and increase the amount and productivity of fish habitat, and therefore fish, in this reach of the river.

5.5.5.1 Key Published Information Mercury concentrations in mainly Northern Pike, Walleye, Lake Whitefish, and Lake Trout (Salvelinus namaycush) from commercial catches from 13 Area 4 waterbodies between 1970–1972 were first compiled by Derksen (1978a, b, 1979) and McGregor (1980). Commercial samples up to 1981–1983 from Ashley, Dysart, Holmes, Cygnet, Limestone, Northern Indian, Partridge Breast, Thornsteinson, and Wood lakes and the Churchill River mouth were reported by Rannie and Punter (1987) as part of a data compilation funded by the Canada-Manitoba Agreement on the Study and Monitoring of Mercury in the Churchill River Diversion (see Section 5.5.1.2). As part of the Agreement, mercury levels were also



measured in individual fish from survey samples conducted from 1978–1983 at selected lakes, including Northern Indian and Partridge Breast lakes (data summarized across all sampling years in Rannie and Punter 1987). Data from these two lakes were used to test predictive models relating reservoir flooding to fish mercury concentrations developed by Johnston et al. (1991).

Mercury concentrations (including standard means) in pike and Walleye from Waskaiowaka and Limestone lakes from 1978–1982 were reported in Green (1986) as part of a summary and statistical analysis of fish mercury data collected under the Canada-Manitoba Agreement for 1983–1985 that also included a compilation of available data for lakes in the Churchill and Nelson River drainages not flooded by CRD.

Fish mercury concentrations were collected as part of the environmental assessment for the Churchill River weir (Manitoba Hydro and the Town of Churchill 1997). To address concerns associated with a potential for increases in fish mercury levels in response to re-watering and flooding of terrestrial habitat resulting from the project, baseline mercury data was collected annually from 1993 to 1996 (Remnant and Bernhardt 1994; Remnant 1995; Remnant and Kitch 1996; Fazakas and Remnant 1997), and compared to post-project data collected from 1999 to 2013 (Bernhardt 2000, 2001, 2002, 2003; Bernhardt and Holm 2003, 2007; Bernhardt and Pisiak 2006; Bernhardt and Caskey 2009; Hertam et al. 2014 ).

Whitefish, pike, and Walleye from Caldwell, Christie, Kiask, Limestone, Pelletier, Recluse, Thomas, and Waskaiowaka lakes were analyzed for mercury in 2005 as part of the Adverse Effects Agreement for the Keeyask Generation Project and detailed results have been reported in the Environmental Impact Statement (EIS) (KHLP 2012).

Fish mercury concentrations continue to be monitored in Northern Indian Lake (NIL) and the Churchill River at its confluence with the Little Churchill River and upstream of the Churchill Weir on a three-year rotational basis under CAMP. Results for the 2010 sampling at NIL and the Churchill River/Little Churchill River are presented in CAMP (2014).

5.5.5.2 New Information and/or Re-analysis of Existing Information The current assessment is based on the updated analysis of all fish mercury data available for Area 4 from 1970 to 2014. Mercury concentrations for approximately 2,400 fish of 11 species have been collected in Area 4 (Table 5.5.1-1), representing 4.5% of all mercury data for individual fish in the ROI. In addition, 67 analyses of composite, commercial fish samples exist for Area 4 that can be associated with a particular year (i.e., not including data presented in Rannie and Punter 1987; see Section 5.5.5.3). These commercial data were used in instances where no data for individual fish were available. In these cases, some of the published yearly means were recalculated for years with multiple composite samples that were only reported separately. The vast majority (92%) of all individual fish mercury data collected in Area 4 are for Lake Whitefish (24.7%), Northern Pike (38.9%), and Walleye (28.4%). All of the sites sampled for fish mercury in Area 4 are shown in Map 5.5.5-1.

Two focal waterbodies located along the lower Churchill River were selected for detailed presentation in the following discussion. Northern Indian Lake and the Churchill River experienced hydrological changes related to hydroelectric development, notably a reduction in flows due to diversion of the Churchill River at



SIL, and both have fish mercury data with relatively good sample size and collection frequency to allow conclusions on the evolution of fish mercury concentrations over an extended time period. Fish mercury data for the Churchill River are available for three locations: the confluence of the Little Churchill River, upstream of the Churchill Weir, and the river mouth (Map 5.5.5-1), with the latter two sites having extensive data sets from two largely separate time periods.

Churchill River Weir

Limestone Lake

HolmesLake

AshleyLake

BissettLake

CygnetLake

PelletierLake

RecluseLakeChristie

Lake

DysartLake

WaskaiowakaLake

CampbellLake

Billard Lake

PartridgeBreast Lake

Wood Lake

ThorsteinsonLake

Church

ill Rive

r

HUDSONBAY

Northern Indian Lake

Kiask Lake Caldwell

Lake

ThomasLake

Little Churchill

River


Churchill

1.0

28-OCT-15


08-DEC-15

Province of Manitoba, Government of Canada and Manitoba Hydro.


RCEA Area 4

LegendFish Mercury InformationFocal WaterbodyRCEA Region of Interest

Hudson Bay

Thompson

Churchill


DATA SOURCE:

COORDINATE SYSTEM: DATE CREATED:

CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:


0 10 20 Miles

0 10 20 Kilometres

Cre

ated

By:

cpa

rker

- B

Siz

e La

ndsc

ape

STD

- M

AR

201

5

Sca

le: 1

:13,

000,

000

RCEAArea 4

InfrastructureControl Structure

Map 5.5.5-1



5.5.5.3 Pre-CRD Baseline Mercury Concentrations Area 4 contains a relative large number (compared to the total number of waterbodies with fish mercury data) of off- and on-system waterbodies with fish mercury data from composite commercial samples that predate CRD (i.e., 1976). Although it is often difficult to precisely assign some of the information presented in Derksen (1978 a, b) and Derksen (1979) to a specific year over the period 1970–1972, a comparison of location-specific sample sizes and mercury concentrations among the three publications by Derksen in combination with information provided in Rannie and Punter (1987) and McGregor (1980) allows such an assignment with considerable confidence (Table 5.5.5-1). Only the several samples that exist for Northern Pike and Walleye from the mouth of the Churchill River for 1973–1976 (Rannie and Punter 1987) cannot be dated to a specific year because the authors only provide an average concentration and range for the entire sampling period on record.

Based on the results presented in Table 5.5.5-1, pre-CRD mercury concentrations were similar between on-system and off-system waterbodies with means ranging from 0.07–0.16 ppm in Lake Whitefish, 0.27-0.51 ppm in Northern Pike, and 0.14–0.37 ppm in Walleye. The range of mercury concentrations in pike and Walleye from off-system lakes encompassed the range of concentrations in their conspecifics from on-system waterbodies. With a few exceptions (e.g., pike from Limestone Lake, whitefish from NIL) concentrations are also similar to those for fish sampled commercially from Area 3 lakes during years pre-dating CRD (Table 5.5.4-1). As noted in Section 5.5.4.3 (Area 3), the composite nature of commercial samples (see Section 5.5.1.4), and their typically small sample size and unknown length of the fish included in the sample precludes statistical comparisons among commercial samples. Therefore, the conclusion of equality in fish mercury concentrations between on-system and off-system waterbodies is weak for comparisons of individual lakes, and is based on the weight of evidence for the entire dataset.

It should also be noted that Area 4 is the only area within the ROI that has information on mercury concentrations in Lake Trout for pre-and post-CRD years. Derksen (1978b) reported a concentration of 0.20 ppm in trout from both Buckland Lake in 1970 and Kiask Lake sometime from 1970–1972 (also see Derksen 1979). For post-CRD concentrations in trout from Kiask Lake see below.



Table 5.5.5-1: Mean (range) Mercury Concentrations, Part-per-Million, for Northern Pike, Walleye, and Lake Whitefish from On-System and Off-System Waterbodies in Area 4 for Years Predating Churchill River Diversion (1970–1972)

Off-system Location n Mean Range On-system

Location n Mean Range

Northern Pike

Bissett Lake 1 0.27 - Churchill River* 8 0.42 0.12–0.89

Limestone Lake 3 0.51 0.31–0.69

Waskaiowaka L. 19 0.37 0.13–0.66

Walleye

Caldwell Lake 2 0.37 0.33–0.41 Billard Lake 1 0.33 -

Christie Lake 3 0.29 0.22–0.39 Churchill River* 11 0.23 0.09–0.28

Campbell Lake 1 0.29 - Northern Indian L. 2 0.361 0.28–0.43

Holmes Lake 3 0.14 0.12–0.16

Limestone Lake 8 0.34 0.24–0.42

Waskaiowaka L. 8 0.21 0.07–0.34

Lake Whitefish

Buckland Lake 1 0.07 - Billard Lake 1 0.08 -

Campbell Lake 1 0.11 - Churchill River* 2 0.08 0.04–0.12

Wood Lake 2 0.06 0.04–0.09 Fidler Lake 1 0.10 -

Northern Indian L. 1 0.16 -

Note: All data are for fish from composite samples of commercial catches; n indicates the number of samples. Data sources: Derksen (1978a,b, 1979); McGregor (1980); Rannie and Punter (1987). * Near Churchill (Derksen 1978b) or near Hudson Bay (Derksen 1979).

1. From Derksen (1978b, 1979); McGregor (1980) lists one sample each for 1970 and 1972 with concentrations of 0.28 ppm.

5.5.5.4 Changes in Mercury Concentrations over Time This section will provide a detailed description of the time series of fish mercury observed for each focal waterbody. Subsequent sections will compare mercury concentrations to standards and guidelines, between on-system and off-system waterbodies, and will discuss the evolution of fish mercury concentrations in Area 4 with respect to hydroelectric development.



5.5.5.4.1 Northern Indian Lake

Northern Indian Lake located on the Churchill River approximately 43 km downstream of the Missi Falls CS experienced a reduction in average water surface elevation of approximately 2.3 m as a result of CRD.

Mercury data based on individual pike and Walleye are available for seven years between 1978 and 2013. In addition to the sometimes large gaps in the period of record, the samples collected from 1978-1996 were often small and the relationship between fish length and mercury concentration was not significant, precluding the calculation of standard means (Figure 5.5.5-1).

Northern Indian Lake

Year1970

19751980

19851990

19952000

20052010

2015

Mus

cle

mer

cury

(pp

m)

0.0

0.2

0.4

0.6

0.8


n=5

*

*

**

Start of flow changesdue to CRD

n=5n=5

n=6n=5 *

*

*

n=6

n=5

Means without CL prior to 1980 are from commercial samples. An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was use; n represents sample size. Stippled lines indicate the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.5-1: Mean (95% Confidence Limits [CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From Northern Indian Lake for 1971–2013

Considering these shortcomings in the available record, there is no indication of a temporal trend or consistent changes in concentrations that could be associated with hydroelectric development. Although no data are available from NIL for years close to 2005, the recent (2010) relatively high mercury concentrations of 0.53 ppm in pike and Walleye are consistent with increases in concentrations observed



for other on-system lakes in the ROI in years following of 2005. The decrease in concentrations in pike and walleye from NIL between 2010 and 2013 was significant for both species.

5.5.5.4.2 Churchill River

The Churchill River runs approximately 450 km downstream from the Missi Falls CS to its mouth at the town of Churchill. Fish mercury data mainly exist for two locations located within the most downstream 15 km of the river, with some very recent (CAMP) data coming from the confluence with the Little Churchill River approximately 200 km upstream of the mouth. As described above, the river flow just downstream of Missi Falls CS has been reduced by almost 80% as a result of CRD. Farther downstream, it is estimated that the percentage of flow at Red Head Rapids originating from Missi Falls was 46% with CRD and would have been 83% without CRD.

CHURCHILL RIVER MOUTH

Mercury concentrations in pike and Walleye from the mouth of the Churchill River are available for the years 1970–1998, however, commercial data from 1973–1976 cannot be assigned to a particular year (see Section 5.5.5.3) and were not considered in Figure 5.5.5-2. Similar to NIL, most of the survey samples are associated with small numbers of fish and the relationship between fish length and mercury concentration was not significant, precluding the calculation of standard means (Figure 5.5.5-2). The time series of mercury concentration in fish from the mouth of the Churchill River is characterized by a high interannual variability without a consistent trend, and considerable synchrony in mean concentrations between pike and Walleye from 1982 onwards (Figure 5.5.5-2). Concentrations in the two piscivores range from 0.02 ppm (the lowest mean recorded for both species anywhere in the ROI) to 0.64 ppm and 0.34 ppm, respectively. The frequent occurrence of mercury concentrations below 0.1 ppm is very unusual for pike and Walleye in the ROI or anywhere in their natural range (Bodaly et al. 1993; Garcia and Carignan 2000; Schetagne et al 2003), particularly when considering that these were relatively large fish (mean fork length 575–614 mm for pike, 331–476 mm for Walleye). One possible explanation for these low concentrations is that pike and Walleye captured in late summer or fall at the mouth of the river near the Hudson Bay (Figure 5.5.5-1) have been feeding mainly and for a prolonged period of time on food that has sequestered mercury in the marine environment. The marine species Capelin (Mallotus villotus) was sampled together with several freshwater fish at the river mouth in 1970, and at 0.04 ppm had the lowest mercury concentration of all species (together with one sample for Lake Whitefish; Derksen 1978a, b, 1979). Marine fish generally have lower mercury concentrations than similar-sized freshwater species (e.g., Atwell et al. 1998; Stern and MacDonald 2005). Thus, pike and Walleye feeding mainly on Capelin or other marine species would biomagnify mercury at a lower rate than when feeding on more typical prey fish, resulting in lower mercury concentrations. The yearly synchronous fluctuations in concentrations for both predatory species may be the result of changes in the distribution of marine prey species and their overlap with pike and Walleye foraging habitat. In those years when marine prey fish and predator foraging habitat distributions don’t or only marginally overlap, pike and Walleye would feed mainly on freshwater species and biomagnify mercury more rapidly. However, because mercury concentrations in the relatively large fish analyzed for mercury change relatively slowly over time (see page below), the sometimes pronounced and significant differences in



mercury concentrations of pike and Walleye between consecutive years (e.g., 1983–1984) must have an additional or alternative explanation.

Churchill R., Mouth

Year1970

19751980

19851990

19952000

20052010

Mus

cle

mer

cury

(pp

m)

0.0

0.2

0.4

0.6

0.8

Pike Walleye

n=5

*

*

**


Completionof weir

n=5

n=5

n=5

n=6

n=8

1.11

n=5

*

*

*

* * *

*

*

n=6

n=5

n=5

n=5

n=5

n=5

*

**

n=5

n=6

n=6

n=6

**

An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used. A number denotes an upper CL not shown as a bar; n represents sample size. The stippled line indicate the 0.5 ppm Health Canada standard for retail fish. Means without CL prior to 1980 are from commercial samples.

Figure 5.5.5-2: Mean (95% Confidence Limits [CL]) Length Standardized Muscle Mercury Concentrations of Northern Pike and Walleye From the Churchill River Mouth for 1970–1998

CHURCHILL RIVER UPSTREAM OF THE CHURCHILL WEIR

Mercury concentrations in Lake Whitefish and Northern Pike from the Churchill River just upstream of the weir have been regularly measured since 1994 as part of the Lower Churchill River Water Level Enhancement Weir Project (Hertam et al. 2014). More limited data are available for Walleye in this area.

Concentrations in whitefish and pike for the relatively large samples from the weir location did not show fluctuations similar to those observed for pike and Walleye from the mouth of the river. The standard mean of pike for 1995 (the one common data year) was five times higher than the arithmetic mean of their conspecifics from the river mouth, indirectly supporting the hypothesis of a temporally/spatially varying diet of the piscivorous species sampled near the mouth (see above).



Pike mercury concentrations at the Churchill Weir were highest (0.34 ppm) in 1995, but statistically similar to the concentrations measured in the other two pre-weir years, 1994 and 1996 (Figure 5.5.5-3). Since 1995 concentrations have continuously declined, reaching a minimum of 0.22 ppm in 2005. Starting in 2001, standard means have been significantly lower than any of those in pre-weir years.

Concentrations have increased from 2005 to 2008 after which they decreased in 2013 (Figure 5.5.5-3). The increase to 2008 was just not significant and the most recent concentration is still significantly lower than any of those during pre-weir years. The two available samples from 2010 and 2013 for pike from the 3rd site on the Churchill River (near the confluence of the Little Churchill River) indirectly support a potential pattern of a post-2005 increase and subsequent decrease in pike mercury concentrations. At 0.33 ppm, the concentration in 2010 represents a further increase from the 0.27 ppm measured for the weir ite in 2008, and at 0.27 ppm the concentration for 2013 is quite similar to the 0.24 ppm measured at the weir site. The increase in standard mean mercury concentrations of pike from 2005 to 2008 (i.e., nine years after the completion of the weir) is unlikely a consequence of project-related flooding and other explanations have to be sought.

Mercury concentrations in whitefish have fluctuated less than those in pike, ranging from 0.08 ppm in 1999 to 0.12 ppm in 1994 and 2013, and have not shown a temporal trend similar to that of the piscivorous species. Most of the concentrations recorded in years after completion of the weir were statistically similar to those of pre-weir years. There was no immediate increase on concentrations following 2005, but the increase in concentration between 2008 (0.09 ppm) and 2013 (0.12 ppm) was significant.



Churchill R., Weir

Year1975

19801985

19901995

20002005

20102015

Mus

cle

mer

cury

(pp

m)

0.0

0.1

0.2

0.3

0.4

0.5


n=5 *

*

**


Completionof weir

An asterisk indicates that the relationship between fish length and mercury concentration was not significant and the arithmetic mean was used; n represents sample size. The upper limit of the mercury concentration scale is equivalent to the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.5-3: Mean (95% Confidence Limits) Length Standardized Muscle Mercury Concentrations of Northern Pike, Walleye, and Lake Whitefish From the Churchill River Upstream of the Weir for 1994–2013

Based on an estimate of approximately 19% for the extent of re-flooded area, the EIS for the Churchill Weir Project predicted that mercury concentrations in whitefish and pike would increase “slightly to moderately” after weir construction (Manitoba Hydro and Town of Churchill 1997). In contrast to the predictions, concentrations have remained similar in whitefish and have significantly decreased in pike for several years after the completion of the weir. This post-Project reduction in pike mercury concentrations is inconsistent with the generally observed increase in mercury availability and bioaccumulation after flooding of boreal reservoirs (Bodaly et al. 2007) or periods of increased river water levels (Balogh et al. 2006).

In previous studies, biological parameters such as fish age have been used to explain unexpected patterns in standard means (KHLP 2012). Mercury concentrations are often more strongly associated with fish age than with fish length (Chumchal and Hambright 2009; Jansen and Strange 2009; Evans et al. 2005), and length standardization may not adequately standardize mercury concentrations if the age at a certain length (i.e., growth rate) differs between samples. For the lower Churchill River, mean



age of whitefish was only slightly lower in post-weir years, and the mean age of pike was similar in pre-weir and post-weir years. Thus differences in fish age cannot contribute to an explanation of the unexpected post-Project decrease in standard means of Northern Pike.

One possible explanation for the discrepancy between actual and predicted mercury concentrations is that the EIS predictions included re-watered habitat (previously wetted habitat which had become exposed following CRD) in the calculation of flooded area. Nevertheless, approximately 167 ha (equivalent to 7.5% flooding) were newly inundated after the construction of the weir (Bernhardt and Posthumus 2003). Perhaps the re-watered and the newly flooded areas on the Churchill River had a lower methylation potential than the mainly fringing wetlands and bogs flooded by the reservoirs used in the study by Bodaly et al. (2007). Both the quantity and the quality (lability) of organic carbon stores in the flood zone affect the rate and duration of methylmercury production (Bodaly et al. 2004), and it is likely that Churchill River stores were relatively small and of lower quality. Furthermore, methylmercury bioaccumulation in the food chain may not be closely linked to rates of methylmercury production (Bodaly et al. 2004). For these reasons, the predictions of fish mercury concentrations in the Churchill River upstream of the weir may have been overestimated in the EIS. Nevertheless, it is surprising that concentrations did not increase at all.



Fish mercury concentrations in on-system waterbodies of Area 4 have only rarely exceeded the Health Canada standard of 0.5 ppm for the commercial sale of fish. Most of these exceedances were for relatively large fish from commercial (e.g., some of the samples from the Churchill River in 1970–1972; Table 5.5.5-1) or survey samples with small numbers of fish (i.e., Walleye from NIL in 1982 (Figure 5.5.5-1), Northern Pike from the Churchill River mouth in 1993 (Figure 5.5.5.-2), and pike (0.79 ppm) and Walleye (0.70 ppm) from Partrige Crop Lake for the only available sample in 1982). The only large samples of individual fish with standard means exceeding the Health Canada standard were for 36 pike and Walleye each from NIL in 2010, which both had a concentration of 0.53 ppm (Figure 5.5.5-1).

A concentration exceeding 0.5 ppm was also recorded for a commercial samples from off-system Limestone Lake in 1971/1972, Table 5.5.5-1). Otherwise, the results from several other off-system lakes in Area 4 indicate the Health Canada standard has not been exceeded pre-CRD (Table 5.5.5-1) or in more recent years (Table 5.5.5-2). The range of mercury concentrations in fish from off-system waterbodies in Area 4 is quite large. Considering data across all available years, concentrations in Walleye were as low as 0.14 ppm (Thomas Lake in 2005) and as high as 0.39 ppm in Waskaiowaka Lake in 1982 (Table 5.5.5-2). Furthermore, a significant difference in concentrations among years existed for Waskaiowaka Lake. However, the range in concentrations across lakes and years was not different compared to that observed for the four (Granville Lake excluded) CAMP off-system reference lakes in the ROI (see Table 5.5.4-2), indicating that natural variability in fish mercury among waterbodies and over time can be substantial.



Table 5.5.5-2: Standard Mean or Arithmetic Mean Mercury Concentrations with 95% Confidence Limits, Parts-per-Million, for Northern Pike, Walleye, and Lake Whitefish from Nine Off-System Lakes in Area 4 for Years 1978–2005

Lake 1978-81 1982 2005

Northern Pike

Christie - - 0.218 (0.194–0.244)

Cygnet - 0.184 (0.143–0.237) -

Holmes - 0.255 (0.198–0.312) 1 -

Kiask - - 0.205 (0.172–0.245)

Recluse - 0.240 (0.183–0.316) -

Settee - 0.250 (0.187–0.333) -

Thomas - - 0.134 (0.111–0.162) 2

Waskaiowaka 0.439 (0.367-0.511) 1 0.421 (0.352–0.490) 1 0.233 (0.187–0.279) 1,3

Wood 0.183 (0.126-0.264)* - -

Walleye

Caldwell - - 0.247 (0.218–0.280)

Christie - - 0.290 (0.258–0.327)

Pelletier - - 0.381 (0.348–0.417) 4

Recluse - 0.251 (0.188–0.337) 0.250 (0.206–0.304)

Thomas - - 0.140 (0.121–0.162) 2

Waskaiowaka 0.202 (0.182-0.224) 3 0.387 (0.311–0.481) 0.266 (0.227–0.311)

Wood 0.160 (0.125-0.205)* - -

Lake Whitefish

Caldwell - - 0.039 (0.033–0.047)

Christie - - 0.036 (0.031–0.043)

Kiask - - 0.056 (0.045–0.069)

Thomas - - 0.042 (0.036–0.048)

Waskaiowaka - - 0.036 (0.031–0.042)

Note: Means are based on a minimum of 13 fish. * Mean was calculated by combining data for at least two years within the indicated time period.1. Arithmetic mean (relationship between fish length and mercury concentration was not significant).2. Significantly lower than other means for the species in 2005.3. Significantly lower from means of other years for the lake.4. Significantly higher than other means for the species in 2005.



Together with the Nelson River in Area 2 (see Section 5.5.3.4.1), the Churchill River has the most extensive record on mercury concentrations in Lake Sturgeon within the ROI. The available data for individual fish all come from the CAMP sampling location near the confluence of the Little Churchill River and were collected from 2009–2014. They support the conclusion based on data from Area 2 that sturgeon generally have low mercury concentrations (in the range of 0.05–0.2 ppm), but that some of the larger (> 1,000 mm fork length) individuals can even exceed the 0.5 ppm Health Canada standard for retail fish (Figure 5.5.5-4).

Length (mm)0 200 400 600 800 1000 1200 1400

Mus

cle

Mer

cury

(pp

m)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Churchill RiverHayes River

Significant regression lines are shown. The stippled line indicates the 0.5 ppm Health Canada standard for retail fish.

Figure 5.5.5-4: Relationship Between Mercury Concentration and Fork Length for Lake Sturgeon From the Churchill River at the Little Churchill River for 2010–2014 and the Hayes River for 2009–2013

5.5.5.5.2 Evolution of Fish Mercury Concentrations in Relation to Hydroelectric Development

As discussed for each of the focal waterbodies (see Sections 5.5.5.4.1 and 5.5.5.4.2) there is no indication that the reduction and increased variability in Churchill River flows and water levels as a result of CRD increased mercury concentrations in fish from the river or a riverine lake in Area 4. These results are in line with findings for the Eastmain River in Québec for which a 90% flow reduction had no obvious



effect on mercury concentrations in Lake Whitefish and Northern Pike (Schetagne and Verdon 1999a). There was no evidence of a downstream effect on fish mercury concentrations in the Churchill River. Contrary to EIS predictions, mercury concentrations in Lake Whitefish from the lower Churchill River did not increase and concentrations in Northern Pike significantly decreased after the construction of the Churchill Weir and the associated minor flooding.



5.5.6 Effects of Changes in Hydroelectric Development on Fish Mercury Concentrations in the Region of Interest

• Information on muscle mercury concentrations in fish species from the ROI exists since 1969 andincludes almost 54,000 analyses of individuals from 23 species from more than 200 waterbodies. Thevast majority of the results are for Lake Whitefish (17.2%), Northern Pike (32.1%), and Walleye(29.8%), which together account for almost 80% of the individual data. In addition, more than 1,000analyses of composite samples of fish from commercial captures are available, mainly for years1969–1985.

• Fish mercury concentrations were impacted by hydroelectric development throughout the ROI,particularly by the inundation of soils and vegetation as a result of reservoir flooding. The magnitudeand temporal extent of increases in concentrations varied mainly depending on the intensity offlooding and fish trophic position. These findings are well established and remain unaffected by theexisting data limitations.

• Research within Area 3 of the ROI in the mid-1970s for the first time clearly attributed the increase infish mercury concentrations after flooding to a greater supply of inorganic mercury from the leachingof humic rich soils and an increase in mercury bioavailability due to bacterial mercury methylation.

• Fish mercury concentrations increased relatively rapidly after flooding to reach a transient maximum,after which concentrations declined over an extended period of time, relatively quickly at first, andmore gradually later. Maximum concentrations in piscivorous species (Northern Pike, Walleye,Sauger) from the more extensively flooded lakes reached approximately 2 ppm, whereas maxima inlarge bodied species of lower trophic levels generally did not exceed 0.6 ppm and remained below0.3 ppm in Lake Whitefish. Maximum concentrations were usually reached within 3–9 yearspost-flood in Northern Pike, Walleye, and Lake Whitefish, with the latter species often being at thelower end of the range. Maxima represented a 1.4–8.7 fold increase in mercury concentrationcompared to those from pre-flood conditions. The increases for Lake Whitefish were similar or evenhigher (e.g., Long Spruce forebay, SIL) compared to pike and Walleye from lakes with little ormoderate flooding (i.e., < 100%), whereas the increases were larger in the two piscivorous speciesfor the more extensively flooded lakes (e.g., Rat and Notigi).

• The time to reach minimum concentrations after peaking was quite variable between lakes andspecies, spanning a range from slightly less than 10 years to 30 years. In almost all cases minimawere either measured for the last year of sampling or occurred in 2005 and were followed by mostlysignificantly higher concentrations in subsequent years.

• Based on the entire data set for the ROI, mean mercury concentrations of piscivorous fish speciesfrom on-system waterbodies have regularly and often substantially exceeded the 0.5 ppm HealthCanada standard for the commercial sale of fish. These exceedances have typically been observedfor 5–25 years but in some cases for more than 35 years after flooding. Mean concentrations in fishoccupying lower trophic levels have rarely exceeded the standard for commercial sale.

• With few exceptions (e.g., Northern Pike and Walleye from Rat Lake) current (2010–2014) meanconcentrations in on-system waterbodies are all below the 0.5 ppm standard for the commercial sale



of fish, but individual (large) pike and Walleye fish may reach concentrations of 1 ppm. Mean concentrations in Northern Pike and Walleye for the large number of off-system lakes in the ROI have always been lower than 0.5 ppm, although variability among lakes and over time has been substantial, including standard means as high as 0.38 ppm. Standard means for whitefish from off-system lakes have been generally below 0.05 ppm.

• The lower Churchill River (Area 4) was affected by CRD mainly by decreased and variable flows,amounting to an almost 80% reduction at the Missi Falls CS. There is no indication that thesechanges in the hydrology increased mercury concentrations in fish from Northern Indian Lake and theChurchill River at and below the confluence with the little Churchill River.

• Both historic (1970s) and increasingly current (since 2002) data on mercury concentrations in LakeSturgeon indicate that this species generally has low mercury concentrations in the range of0.05-0.2 ppm, but that some of the larger (> 1,000 mm fork length) individuals can reachconcentrations of 0.6 ppm.


DECEMBER 2015 5.5-100

5.5.7 Bibliography

5.5.7.1 Literature Cited and Data Sources Abernathy, A. R., and Cumbie, P. M. 1977. Mercury accumulation by largemouth bass (Mictropterus

salmoides) in recently impounded reservoirs. Bulletin of Environmental Contamination and Toxicology 17: 595–602 pp.

Ahlgren, M. 1990a. Diet selection and contribution of detritus to the diet of the juvenile White Sucker (Catostomus commersoni). Canadian Journal of Fisheries and Aquatic Sciences 47: 41–48 pp.

Ahlgren, M. O. 1990b. Nutritional significance of facultative detritivory to the juvenile White Sucker (Catostomus commersoni). Canadian Journal of Fisheries and Aquatic Sciences 47: 49–54 pp.

Atwell, L., Hobson, K. A., and Welch, H. E. 1998. Biomagnification and bioaccumulation of mercury in an arctic marine foodweb: Insights from stable isotope analysis. Canadian Journal of Fisheries and Aquatic Sciences 55: 1114–1121 pp.

Ayles, H., Brown, S., Machniak, K., and Sigurdson. J. 1974. The fisheries of the lower Churchill lakes, the Rat-Burntwood lakes and the upper Nelson lakes: Present conditions and the implications of hydroelectric development. Lake Winnipeg, Churchill and Nelson Rivers Study Board Technical Report, Appendix 5, Volume 2, Section I. 99 pp.

Baker, R. F. 1990a. A fisheries survey of the Limestone forebay, 1989 — year I. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 46 pp.

Baker, R. F. 1990b. A fisheries survey of the Nelson River estuary, August 1989. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 26 pp.

Baker, R. F. 1991. A fisheries survey of the Limestone forebay, 1990 — year II. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 43 pp.

Baker, R. F. 1992. A fisheries survey of the Limestone forebay, 1991 — year III. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 56 pp.

Baker, R. F., and Davies, S. 1991. Physical, chemical, and biological effects of Churchill River diversion and Lake Winnipeg regulation on aquatic ecosystems. Canadian Technical Report of Fisheries and Aquatic Sciences 1806: 53 pp.

Baker, R. F., MacDonell, D. S., and Davies, S. 1990. A fisheries survey of the Long Spruce forebay 1989. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 49 pp.

Balogh, S. J., Swain, E. B., and Nollet, Y. H. 2006. Elevated methylmercury concentrations and loadings during flooding in Minnesota rivers. Science of the Total Environment 368: 138–148 pp.


DECEMBER 2015 5.5-101

Bernhardt, W. J. 1995. Cross Lake outlet control weir post-project fish stock assessment 1994, Cross and Pipestone lakes. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 113 pp.

Bernhardt, W. J. 2000. Lower Churchill River water level enhancement weir project post-project monitoring: fish population responses to operation of the project, year I, 1999. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 65 pp.

Bernhardt, W. J. 2001. Lower Churchill River water level enhancement weir project post-project monitoring: Fish population responses to operation of the project — 2000, year II. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 66 pp.

Bernhardt, W. J. 2002. Lower Churchill River water level enhancement weir project post-project monitoring: fish population responses to operation of the project, year III, 2001. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 66 pp.

Bernhardt, W. J. 2003. Lower Churchill River water level enhancement weir project post-project monitoring: a synthesis of aquatic environmental monitoring 1999–2002. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 75 pp.

Bernhardt, W. J., and Caskey, R. 2009. Lower Churchill River water level enhancement weir project post-project monitoring: fish population responses in the lower Churchill River 2008. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 144 pp.

Bernhardt, W. J., and Holm, J. 2003. Lower Churchill River water level enhancement weir project post-project monitoring: Fish population responses to operation of the project — 2002 year IV. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 104 pp.

Bernhardt, W. J., and Holm, J. 2007. Lower Churchill River water level enhancement weir project post-project monitoring: a synthesis of aquatic environmental monitoring 1999–2006. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 107 pp.

Bernhardt, W. J., and Pisiak, D. J. 2006. Lower Churchill River water level enhancement weir project post-project monitoring: fish population responses in the lower Churchill River year VII. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 119 pp.

Bernhardt, W. J., and Posthumus, B. 2003. Lower Churchill River water level enhancement weir project post-project monitoring: fish habitat assessment of the Lower Churchill River Mainstem, 2001. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 28 pp.

Bernhardt, W. J., and Schneider-Vieira, F. 1994. Cross Lake outlet control weir post-project fish stock assessment 1993, Cross and Pipestone lakes. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 110 pp.

Bligh, E. G.1970. Mercury and the contamination of freshwater fish. Fisheries Research Board of Canada MS Report 1088: 27 pp.


DECEMBER 2015 5.5-102

Bligh, E. G. 1971. Mercury levels in Canadian fish. In Proceedings of the symposium on mercury in man’s environment, 15–16 February 1971. Edited by J. E. Watkin. Royal Society of Canada, Ottawa, ON. 73–90 pp.

Bodaly, R. A., and Hecky, R. E. 1979. Post-impoundment increases in fish mercury levels in the Southern Indian Lake reservoir, Manitoba. Canadian Fisheries and Marine Service MS Rep 1531: 15 pp.

Bodaly, R. A., and Rosenberg, D. M. 1990. Retrospective analysis of predictions and actual impacts for the Churchill-Nelson hydroelectric development, northern Manitoba. In Managing the effects of hydroelectric development. Edited by C.E. Delisle and M.A. Bouchard. A symposium sponsored by the Canadian Society of Environmental Biologists. Collections Environment et Geologie 9. Université de Montreal, Montreal QC. 221-242 pp.

Bodaly, R. A., Hecky, R. E., and Fudge, R. J. P. 1984a. Increase in fish mercury levels in lakes flooded by the Churchill River diversion, northern Manitoba. Canadian Journal of Fisheries and Aquatic Sciences 41: 682–691 pp.

Bodaly, R. A., Rosenberg, D. M., Gaboury, M. N., Hecky, R. E., Newbury, R. W., and Patalas, K. 1984b Ecological effects of hydroelectric development in northern Manitoba, Canada: the Churchill-Nelson River diversion. In Effects of pollutants at the ecosystem level. Edited by P. J. Sheehan, D. R. Miller, G. C. Butler, and Ph. Boudreau. John Wiley and Sons Ltd., Mississauga, ON. 273–309 pp.

Bodaly, R. A., Strange, N. E., Hecky, R. E., Fudge, R. J. P., and Anema, C. 1987. Mercury content of soil, lake sediment, vegetation, and forage fish in the area of the Churchill River diversion, Manitoba, 1981-82. Canadian Data Report of Fisheries and Aquatic Sciences 610: 33 pp.

Bodaly, R. A., Strange, N. E., and Fudge, R. J. P. 1988. Mercury content of fish in the Southern Indian Lake and Issett reservoirs, northern Manitoba, before and after Churchill River diversion. Canadian Data Report of Fisheries and Aquatic Sciences 706: 59 pp.

Bodaly, R. A., Rudd, J. W. M., Fudge, R. J. P., and C. A. Kelly. 1993. Mercury concentrations in fish related to size of remote Canadian shield lakes. Canadian Journal of Fisheries and Aquatic Sciences 50: 980-987 pp.

Bodaly, R. A., St. Louis, V. L., Paterson, M. J., Fudge, R. J. P., Hall, B. D., Rosenberg, D. M., and Rudd, J. W. M. 1997. Bioaccumulation of mercury in the aquatic food chain in newly flooded areas. In Metal ions in biological systems, Volume 34: mercury and its effects on environment and biology. Edited by A. Sigel and H. Sigel. Marcel Dekker Inc., New York, NY. 259-287 pp.

Bodaly, R. A., Beaty, K. G., Henzel, L. H., Majewski, A. R., Paterson, M. J., Rolfhus, K. R., Penn, A. F., St. Louis, V. L., Hall, B. D., Matthews, C. J. D., Cherewyk, K. A., Mailman, M., Hurley, J. P., Schiff, S. L., and Venkitesvaran, J. J. 2004. Experimenting with hydroelectric reservoirs. Environmental Science and Technology 38: 347A–352A pp.

http://dfo-mpo.gc.ca/libraries-bibliotheques/data-stat-eng.htm

http://dfo-mpo.gc.ca/libraries-bibliotheques/data-stat-eng.htm


DECEMBER 2015 5.5-103

Bodaly, R. A., Jansen, W. A., Majewski, A. R., Fudge, R. J. P., Strange, N. E., Derksen, A. J., and Green, D. J. 2007. Post-impoundment time course of increased mercury concentrations in fish in hydroelectric reservoirs of northern Manitoba, Canada. Archives of Environmental Contamination and Toxicology 53: 379–389 pp.

Bretecher, R. L., and Horne, B. D. 1997. Lower Nelson River forebay monitoring program 1996. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 62 pp.

Bretecher, R. L., and MacDonell, D. S. 1999. Lower Nelson River forebay monitoring program, 1998. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 47 pp.

Bretecher, R. L., and MacDonell, D. S. 2000. Lower Nelson River forebay monitoring program, 1999. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 101 pp.

Brinkmann, L., and Rasmussen, J. B. 2010. High levels of mercury in biota of a new prairie irrigation reservoir with a simplified food web in Southern Alberta, Canada. Hydrobiologia 641: 11-21 pp.

CAMP (Coordinated Aquatic Monitoring Program). 2014. Three year summary report (2008–2010). Report prepared for the Manitoba/Manitoba Hydro MOU Working Group by North/South Consultants Inc., Winnipeg, MB.

Canada and Manitoba. 1987. Canada-Manitoba agreement on the study and monitoring of mercury in the Churchill River diversion: summary report. Environment Canada, Ottawa, ON, and Manitoba Environment and Workplace Safety and Health, Winnipeg, MB. 77 pp.

CCME (Canadian Council of Ministers of the Environment). 1999. Canadian environmental quality guidelines. Canadian Council of Ministers of the Environment, Winnipeg, MB. Updated to 2015.

Chan, L., Receveur, O., Sharp, D., Schwartz, H., Ing, A., Fediuk, K., Black, A., and Tikhonov. C. 2010. First Nations food, nutrition and environment study (FNFNES): Results from Manitoba. University of Northern British Columbia, Prince George, BC. 177 pp.

Choy, E. S., Hodson, P. V., Campbell, L. M., Fowlie, A. R., and Ridal, J. 2008. Spatial and temporal trends of mercury concentrations in young-of-the-year spottail shiners (Notropis hudsonius) in the St. Lawrence River at Cornwall, ON. Archives of Environmental Contamination and Toxicology 54: 473-481 pp.

Chumchal, M. M., and Hambright, K. D. 2009. Ecological factors regulating mercury contamination of fish from Caddo lake, Texas, USA. Environmental Toxicology and Chemistry 28: 962-972 pp.

Clarkson, T. W. 2002. The three modern faces of mercury. Environmental Health Perspectives 110: 11-23 pp.

Davies, S., Baker, R. and Horne, B. 1998. The quality of Lake Whitefish, Walleye, and Northern Pike from Playgreen Lake, Little Playgreen Lake and Mossy Bay (Lake Winnipeg). North/South Consultants Inc., Winnipeg, MB. 36 pp.


DECEMBER 2015 5.5-104

Depew, D. C., Basu, N., Burgess, N. M., Campbell, L. M., Devlin, E. W., Drevnick, P. E., Hammerschmidt, C. R., Murphy C. A., Sandheinrich, M. B., and Wiener, J. G. 2012. Toxicity of dietary methylmercury to fish: derivations of ecologically meaningful threshold concentrations. Environmental Toxicology and Chemistry 31: 1536-1547 pp.

Depew, D. C., Burgess, N. M., Anderson, M. R., Baker, R., Bhavsar, S. P., Bodaly, R. A., Eckley, C. S., Evans, M. S., Gantner, N., Graydon, J. A., Jacobs, K., LeBlanc, K. E., St. Louis, V. I., and Campbell, L. M. 2013. An overview of mercury concentrations in freshwater fish species: a national fish mercury dataset for Canada. Canadian Journal of Fisheries and Aquatic Sciences 70: 1-16 pp.

Derksen, A. J. 1978a. A report on the preliminary testing of mercury contamination in fishes from the Hudson Bay drainage system connected with Manitoba. MS Report No. 78-70, Fisheries Management Branch, Fisheries Research Section, Manitoba Department of Northern Affairs, Renewable Resources and Transportation Services, Winnipeg, MB. 40 pp.

Derksen, A. J. 1978b. A review of possible natural sources of mercury contamination in Manitoba waters. MS Report No. 78-71, Fisheries Management Branch, Fisheries Research Section, Manitoba Department of Northern Affairs, Renewable Resources and Transportation Services, Winnipeg, MB. 40 pp.

Derksen, A. J. 1979. A summary report of mercury contamination in fishes from Manitoba waters to March, 1971. MS Report No. 79-55, Manitoba Department of Mines, Natural Resources and Environment, Winnipeg, MB. 43 pp.

Derksen, A. J., and Green, D. J. 1987. Total mercury concentrations in large fishes from lakes on the Churchill River diversion and Nelson River. Canada-Manitoba Agreement on the Study and Monitoring of Mercury in the Churchill River Diversion, Technical Appendix 17, Volume 4. 195 pp.

DFO (Department of Fisheries and Oceans). 1987. Mercury data from inspected commercial shipments of fish from the Churchill River diversion area in Manitoba. Canada-Manitoba Agreement on the Study and Monitoring of Mercury in the Churchill River Diversion, Technical Appendix 16, Volume 4. 18 pp.

EC and DFO (Environment Canada and Department of Fisheries and Oceans). 1992. Federal ecological monitoring program: final report, volume 2. Environment Canada and Department of Fisheries and Oceans, Winnipeg, MB. 120 pp.

EC. 2015. Historical hydrometric data [online]. Available from http://wateroffice.ec.gc.ca/mainmenu/historical_data_index_e.html [accessed November 2015].

Evans, M. S., Lockhart, W.L., Doetzel, L., Low, G., Muir, D., Kidd, K., Stephens, G., and Delaronde, J. 2005. Elevated mercury concentrations in fish in lakes in the Mackenzie River Basin: the role of physical, chemical, and biological factors. Science of the Total Environment 351-352: 479-500 pp.

http://www.ncbi.nlm.nih.gov/pubmed/?term=Lockhart%20WL%5BAuthor%5D&cauthor=true&cauthor_uid=16183101

http://www.ncbi.nlm.nih.gov/pubmed/?term=Doetzel%20L%5BAuthor%5D&cauthor=true&cauthor_uid=16183101

http://www.ncbi.nlm.nih.gov/pubmed/?term=Low%20G%5BAuthor%5D&cauthor=true&cauthor_uid=16183101

http://www.ncbi.nlm.nih.gov/pubmed/?term=Muir%20D%5BAuthor%5D&cauthor=true&cauthor_uid=16183101

http://www.ncbi.nlm.nih.gov/pubmed/?term=Kidd%20K%5BAuthor%5D&cauthor=true&cauthor_uid=16183101

http://www.ncbi.nlm.nih.gov/pubmed/?term=Stephens%20G%5BAuthor%5D&cauthor=true&cauthor_uid=16183101

http://www.ncbi.nlm.nih.gov/pubmed/?term=Delaronde%20J%5BAuthor%5D&cauthor=true&cauthor_uid=16183101


DECEMBER 2015 5.5-105

Fazakas, C. R., and Remnant, R. A. 1997. Fish utilization of the lower Churchill River mainstem, year II, 1996. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 93 pp.

Garcia, E., and Carignan. R. 2000. Mercury concentrations in Northern Pike (Esox lucius) from boreal lakes with logged, burned, or undisturbed catchments. Canadian Journal of Fisheries and Aquatic Sciences 57: 129-135 pp.

Goldstein, R. M., Brigham, M. E., and Stauffer, J. C. 1996. Comparison of mercury concentrations in liver, muscle, whole bodies, and composites of fish from the Red River of the North. Canadian Journal of Fisheries and Aquatic Sciences 53: 244-252 pp.

Göthberg, A. 1983. Intensive fishing — a way to reduce the mercury level in fish. Ambio 12: 259-261 pp.

Government of Alberta. 2009. Alberta Health and Wellness: common questions about methylmercury levels in Alberta fish. Government of Alberta, Edmonton, AB. 2 pp.

Green, D. J. 1986. Summary of fish mercury data collected from lakes on the Rat-Burntwood and Nelson River systems, 1983–1985. MS Report No. 86-06, Fisheries Branch, Manitoba Department of Natural Resources, Winnipeg, MB. 359 pp.

Green, D. J. 1990. Updated summary of fish mercury data collected from six lakes on the Rat-Burntwood and Nelson systems, 1983–1989. MS Report No. 90-10, Manitoba Department of Natural Resources, Winnipeg, MB. 277 pp.

Grieb, T. M.,Driscoll, C. T., Gloss, S. P., Schofield, C. L., Bowie, G. L., and Porcella, D. B. 1990. Factors affecting mercury accumulation in fish in the upper Michigan peninsula. Environmental Toxicology and Chemistry 9: 919–930 pp.

Hall, B. D., St. Louis, V. L., Rolfhus, K., Bodaly, R. A., Beaty, K. G., Paterson, M. J., and Cherrewyk, K. A. P. 2005. Impacts of reservoir creation on the biogeochemical cycling of methyl mercury and total mercury in boreal upland forests. Ecosystems 8: 248–266 pp.

Harris, R., and Hutchinson, D. 2009. Predicting mercury cycling and bioaccumulation in reservoirs: development and application of the RESMERC simulation model. A report prepared for Manitoba Hydro by Tetra Tech Inc., Winnipeg, MB.

Harris, R. C., and Snodgrass, W. J. 1993. Bioenergetic simulations of mercury uptake and retention in Walleye (Stizostedion vitreum) and Yellow Perch (Perca flavescens). Water Quality Research Journal of Canada 28: 217–236 pp.

Health Canada. 2007a. Human health risk assessment of mercury in fish and health benefits of fish consumption. Health Canada: Bureau of Chemical Safety, Food Directorate, Health Products and Food Branch, Ottawa, ON. 48 pp.

Health Canada. 2007b. Updating the existing risk management strategy for mercury in fish. Health Canada: Bureau of Chemical Safety, Food Directorate, Health Products and Food Branch, Ottawa, ON. 30 pp.


DECEMBER 2015 5.5-106

Health Canada. 2010. Toxicological reference values, estimated daily intakes or dietary reference values for trace elements. Last updated Sept. 2010. Developed by the Chemical Health Hazard Assessment Division, Bureau of Chemical Safety, Food and Health Products Branch, Health Canada, Ottawa, ON.

Hecky, R. E., Bodaly, R. A., Strange, N. E., Ramsey, D. J., Anema, C., and Fudge, R. J. P. 1987. Mercury bioaccumulation in Yellow Perch in limnocorrals simulating the effects of reservoir formation. Canadian Fisheries and Aquatic Sciences Data Report 628: 158 pp.

Hecky, R. E., Ramsey, D. J., Bodaly, R. A., and Strange, N. E. 1991. Increased methylmercury contamination in fish in newly formed freshwater reservoirs. In Advances in mercury toxicology. Edited by T. Suzuki, A. Imura and T. W. F. Clakson. Plenum Press, New York, NY. 33-52 pp.

Hendzel, M. R., and Jamieson, D. M. 1976. Determination of mercury in fish. Analytical Chemistry 48: 926–928 pp.

Holm, J., Cooley, M., and Shipley, E. 2007. Trace elements in fish from the Keeyask study area: fall 2004. Keeyask Project Environmental Studies Program Report # 04-11. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 50 pp.

Horne, B. D., and Baker, R. F. 1993. A fisheries survey of the Limestone forebay, 1992 — year IV. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 45 pp.

Horne, B. D., and MacDonell, D. S. 1995a. Lower Nelson River forebay monitoring program 1994, year VI. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 41 pp.

Horne, B. D., and MacDonell, D. S. 1995b. Lower Nelson River tributaries fish utilization study: Angling River 1994 — year II. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 66 pp.

Jansen, W. 2010a. Mercury in fish from six northern Manitoba lakes and reservoirs: results from 2007–2008 sampling and an update of time trends of monitoring data. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 45 pp.

Jansen, W. 2010b. Fish mercury concentrations from the Keeyask study area, 2009. Keeyask Project Environmental Studies Program Report # 09-05. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 32 pp.

Jansen, W. 2013. Fish mercury concentrations from the Aiken River, 2012. Keeyask Project Environmental Studies Program Report # 12-04. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 31 pp.

Jansen, W., and Strange, N. 2007a. Mercury in fish in northern Manitoba reservoirs: results from 1999-2005 sampling and a summary of all monitoring data for 1970–2005. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 102 pp.


DECEMBER 2015 5.5-107

Jansen, W., and Strange, N. E. 2007b. Mercury concentrations in fish from the Keeyask project study area for 1999–2005. Keeyask Project Environmental Studies Program Report # 05-04. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 152 pp.

Jansen, W., and Strange, N. 2009. Fish mercury concentrations from the Keeyask project study area for 2006. Keeyask Project Environmental Studies Program Report # 06-11. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 55 pp.

Jensen, S., and Jernelöv, A. 1967. Biological methylation of mercury in aquatic organisms. Nature 223: 753–754 pp.

Johnels, A. G., Westermark, T., Berg, W., Persson, P. I., and Sjöstrand, B. 1967. Pike (Esox lucius L.) and some other aquatic organisms in Sweden as indicators of mercury contamination in the environment. Oikos 18: 323-333 pp.

Johnson, M. W. 2010. Trace elements in fish from the Conawapa study area: fall 2009. Conawapa Project Environmental Studies Program Report # 09-03. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 82 pp.

Johnson, M. W., and MacDonell, D. S. 2004. Nelson River mainstem index gillnetting study, summer and fall, 2003. Lower Nelson River Aquatic Studies Report # 04-01. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 89 pp.

Johnson, M. W., MacDonell, D. S., and Maclean, B. 2004. Limestone and Long Spruce forebays index gillnetting studies, summer 2003. Lower Nelson River Aquatic Studies Report # 04-04. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 45 pp.

Johnston, T. A., Bodaly, R. A., and Mathias, J. A. 1991. Predicting fish mercury levels from physical characteristics of boreal reservoirs. Canadian Journal of Fisheries and Aquatic Sciences 48: 1468–1475 pp.

Johnston, T. A., Bodaly, R. A., Latif, M. A., Fudge, R. J. P., and Strange, N. E. 2001. Intra- and interpopulation variability in maternal transfer of mercury to eggs of Walleye (Stizostedion vitreum). Aquatic Toxicology 52: 73-85 pp.

Kamman, N. C., Burgess, N. L., Driscoll, C. T., Simonin, H. A., Goodale, W., Linehan, J., Estabrook, R., Hutchenson, M., Major, A., Scheuhammer, A., and Scruton, D. 2005. Mercury in freshwater fish of northeast North America — a geographic perspective based on fish tissue monitoring databases. Ecotoxicology 14:163–180 pp.

Keeyask Mercury and Human Health Technical Working Group. 2014. Keeyask Generation Project: Keeyask mercury and human health technical working group final report. [s.n.], [s.l.].

Kelly, C. A., Rudd, J. W. M., Bodaly, R. A., Roulet, N. P., St. Louis, V. L., Haynes, A., Moore, T. R., Schiff, S., Aravena, R., Scott, K. J., Dyck, B., Harris, R., Warner, B., and Edwards, G. 1997. Increases in fluxes of greenhouse gases and methyl mercury following flooding of an experimental reservoir. Environmental Science and Technology 31: 1334–1344 pp.


DECEMBER 2015 5.5-108

KHLP (Keeyask Hydropower Limited Partnership). 2012. Keeyask Generation Project: Environmental impact statement. Supporting volume — Aquatic Environment. Keeyask Hydropower Limited Partnership, Winnipeg, MB.

Koshinsky, G. D. 1973. The limnology-fisheries of the Outlet Lakes Area: present conditions and implications of hydro-electric development. Lake Winnipeg, Churchill and Nelson Rivers Study Board Technical Report, Appendix 5, Volume 2, Section A. 156 pp.

Kroeker, K., and Bernhardt, W. J. 1993. Cross Lake outlet control weir post-project fish stock assessment, 1992. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 83 pp.

Kroeker, K., and Horne, B. D. 1993. A fisheries survey of the Long Spruce forebay, 1992. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 42 pp.

Lockhart, W. L., Uthe, J. F., Kenney, A. R., and Mehrle, P. M. 1972. Methylmercury in Northern Pike (Esox lucius): distribution, elimination, and some biochemical characteristics of contaminated fish. Journal of the Fisheries Research Board of Canada 29: 1519–1523 pp.

Lodenius, M., Seppänen, A., and Herranen, M. 1983. Accumulation of mercury in fish and man from reservoirs in northern Finland. Water Air Soil and Pollution 19: 237-246 pp.

Lucotte, M., Schetagne, R., Thérien, N., Langlois, C., and Tremblay, A. 1999. Mercury and biogeochemical cycle: natural environments and hydroelectric reservoirs of northern Québec (Canada). Springer-Verlag Verlin Heidelberg, New York, NY. 334 pp.

LWCNRSB (Lake Winnipeg, Churchill and Nelson Rivers Study Board). 1975. Lake Winnipeg, Churchill and Nelson Rivers Study Board: technical report and appendices. Lake Winnipeg, Churchill and Nelson Rivers Study Board, Winnipeg, MB.

MacDonell, D. S. 1991. Enumeration and biological description of migrating fish in Limestone River, Manitoba, 1990. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 81 pp.

MacDonell, D. S. 1992. Lower Nelson River tributary fish utilization studies: Angling and Kaiskwasotasine rivers, 1991. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 85 pp.

MacDonell, D. S., and Horne, B. D. 1994. Lower Nelson River forebay monitoring program 1993 — year V. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 110 pp.

MacKay, G. H., Davies, S., and Westdal, H. 1990. Post project assessment of Kelsey and Lake Winnipeg Regulation impacts on Wabowden. The Wabowden Study Team, Winnipeg, MB. 98 pp.

MacLaren Plansearch Inc. and Beak Consultants Ltd. 1988. Implementation of NFA article 19.4 1. Relevant background data. MacLaren Plansearch Inc., Winnipeg, MB, and Beak Consultants Ltd., Brampton, ON.


DECEMBER 2015 5.5-109

Maclean, B. D., and Pisiak, D. J. 2005. Results of fish community investigations conducted in the Aiken River, Manitoba, spring 2003: year II. Keeyask Project Environmental Studies Program Report # 03-13. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 94 pp.

Manitoba Hydro. 1998. Kettle Generating Station: Change in summer operations: Initial environmental evaluation. A report prepared for Manitoba Hydro by TetrES Consultants Inc. and North/South Consultants Inc. Winnipeg, MB. 63 pp

Manitoba Hydro and Nisichawayasihk Cree Nation. 2003. Wuskwatim Generation Project: environmental impact statement. Volume 5 plus Appendices. Support from North/South Consultants Inc., Winnipeg, MB.

Manitoba Hydro and the Town of Churchill. 1997. Lower Churchill River water level enhancement weir project-environmental Impact Statement including appendices. Manitoba Hydro, Winnipeg, MB, and the Town of Churchill, Churchill, MB.

McGregor, G. W. G. 1980. Summary of mercury levels in lakes on the Churchill-Rat-Burntwood and Nelson River systems from 1970 to 1979. Canadian Fisheries and Aquatic Sciences Data Report 195: 16 pp.

Mergler, D., Anderson, H. A., Chan, L. H. M., Mahaffey, K. R., Murray, M., Sakamoto, M., and Stern, A. H. 2007. Methylmercury exposure and health effects in humans: a worldwide concern. Ambio 36: 3-11 pp.

Mochnacz, N. J., Barth, C. C., and Holm, J. 2004. Results of fisheries investigations conducted in the Aiken River and at the mouth of the Ripple River, Manitoba, spring 2002. Keeyask Project Environmental Studies Program Report #02-08. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 106 pp.

Munn, M. D., and Short, T. M. 1997. Spatial hetrogeneity of mercury bioaccumulation by Walleye in Franklin D Roosevelt Lake and the upper Columbia River, Washington. Transactions of the American Fisheries Society 126: 477–487 pp.

MWS (Manitoba Water Stewardship). 2011. Manitoba water quality standards, objectives, and guidelines. MWS Report 2011-01:November 28, 2011, Water Science and Management Branch. 67 pp.

Newbury, R. W., McCullough, G. K., and Hecky, R. E. 1984. The Southern Indian Lake impoundment and Churchill River diversion. Canadian Journal of Fisheries and Aquatic Sciences 41(4): 548–557 pp.

North/South Consultants Inc. 2010. Lower Churchill River water level enhancement project: an evaluation of weir performance. A report prepared for the Town of Churchill and Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 239 pp.


DECEMBER 2015 5.5-110

North/South Consultants Inc. 2012. Limestone Generating Station: aquatic environment monitoring programs: a synthesis of results from 1985 to 2003. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 192 pp.

Porvari, P. 1998. Development of fish mercury concentrations in Finnish reservoirs from 1979 to 1994. Science of the Total Environment 213: 279-290 pp.

Potter, L., Kidd, D., and Standiford, D. 1975. Mercury levels in Lake Powell — bioamplification of mercury in man-made desert reservoir. Environmental Science and Technology 9: 41-46 pp.

Ralston, N. V. C., Azenkeng, A., Ralston, C. R., Blackwell III, L., and Raymond, L. J. 2014. Selenium-health benefit values as seafood safety criteria. In Seafood science: advances in chemistry, technology and applications. Edited by Se-Kwon Kim. CRC Press, Boca Raton, FL. 433–457 pp.

Ramsey, D. J. 1990. Measurements of methylation balance in Southern Indian Lake, Granville Lake, and Stephens Lake, Manitoba, 1989. Northern Flood Agreement Manitoba Ecological Report Series No. 90-3, Environment Canada and Department of Fisheries and Oceans, Winnipeg, MB. 89 pp.

Ramsey, D. J. 1991. Analysis of fish mercury data from lakes on the Rat-Burntwood and Nelson River systems, 1983–1989. MS Report No. 91-14, Manitoba Department of Natural Resources, Winnipeg MB. 84 pp.

Rannie, W. F., and Punter, E. 1987. Survey of existing data on levels and sources of mercury within the region covered by the Canada-Manitoba Mercury Agreement. Canada-Manitoba Agreement on the Study and Monitoring of Mercury in the Churchill River Diversion, Technical Appendix 2, Volume 1. 172 pp.

Remnant, R. A. 1995. A fisheries survey of the lower Churchill River mainstem 1994. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 105 pp.

Remnant, R. A., and Barth, C. C. 2003. Results of experimental gillnetting on the Nelson River between Birthday and Gull rapids, Manitoba, fall 1999. Keeyask Project Environmental Studies Program Report # 99-01. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 75 pp.

Remnant, R. A., and Bernhardt, W. J. 1994. An assessment of fish utilization of Goose Creek, near Churchill, Manitoba 1993. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 248 pp.

Remnant, R. A., and Kitch, I. 1996. Fish utilization and habitat classification of the lower Churchill River mainstem, 1995. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 114 pp.

Remnant, R. A., Graveline, P. G., and Bretecher, R. L. 1997. Range extension of the Rainbow Smelt, Osmerus mordax in the Hudson Bay drainage of Manitoba. Canadian Field-Naturalist 111(4): 660–662 pp.


DECEMBER 2015 5.5-111

Rempel, G., Schiefer, K., Wichenko, G., and Williams, D. 1989. Initial environmental evaluation for the Churchill River diversion augmented flow program: technical report. A report prepared for Manitoba Hydro Northern Flood Committee Inc. by The Augmented Flow Program Study Team, Winnipeg, MB. 68 pp.

Rosenberg, D. M., Bodaly, R. A., Hecky, R. E., and Newbury, R. W. 1997. Chapter 4: the environmental assessment of hydroelectric impoundments and diversions in Canada. In Canadian aquatic resources. Edited by M. C. Healey and R. R. Wallace. Department of Fisheries and Oceans, Ottawa, ON. 71-104 pp.

Sandheinrich, M. B., and Wiener, J. G. 2011. Methylmercury in freshwater fish — recent advances in assessing toxicity in environmentally relevant exposures. In Environmental contaminants in biota. Edited by W. N. Beyer and J. P. Meador. CRC Press, Boca Raton, FL. 169–190 pp.

Schetagne, R., and Verdon, R. 1999a. Post-impoundment evolution of fish mercury levels at the La Grande Complex, Québec, Canada (from 1978 to 1996). In Mercury and biogeochemical cycle: natural environments and hydroelectric reservoirs of northern Québec (Canada). Edited by M. Lucotte, R. Schetagne, N. Thérien, C. Langlois, and A. Tremblay. Springer-Verlag Verlin Heidelberg, New York, NY. 235–258 pp.

Schetagne, R., and Verdon, R. 1999b. Mercury in fish of natural lakes of Québec. In Mercury and biogeochemical cycle: natural environments and hydroelectric reservoirs of northern Québec (Canada). Edited by M. Lucotte, R. Schetagne, N. Thérien, C. Langlois, and A. Tremblay. Springer-Verlag Verlin Heidelberg, New York, NY. 115–130 pp.

Schetagne, R., Therrien, J., and Lalumiere, R. 2003. Environmental monitoring at the La Grande complex — evolution of fish mercury levels. Summary report 1978–2000. Direction Barrages et Environnement, Hydro-Québec Production and Groupe conseil GENIVAR Inc., Montreal, QC. 185 pp.

Smith, F. A., Sharma, R. P., Lynn, R. I., and Low, J. B. 1974. Mercury and selected pesticide levels in fish and wildlife of Utah: I. Levels of mercury, DDT, DDE, Dieldrin and PCB in fish. Bulletin of Environmental Contamination and Toxicology 12: 218-223 pp.

Sopuck, R.D. 1977. Manitoba inter-departmental memo regarding mercury contamination in Hill, Wapisu and Mynarski Lake. Government of Manitoba, [s.l.]. 1 p.

Split Lake Cree-Manitoba Hydro Joint Study Group. 1996. History and first order effects: Manitoba Hydro projects and related activities in the Split Lake Cree study area. Split Lake Cree Post Project Environmental Review, Volume 2. Support from InterGroup Consultants Ltd. and William Kennedy Consultants Ltd. 64 pp

St. Louis, V. L., Rudd, J. W. M., Kelly, C. A., Bodaly, R. A., Paterson, M. J., Beaty, K. G., Hesslein, R. H., Heyes, A., and Majewski, A. R. 2004. The rise and fall of mercury methylation in an experimental reservoir. Environmental Science and Technology 38: 1348–1358 pp.


DECEMBER 2015 5.5-112

Stern, G. A., and Macdonald, R. W. 2005. Biogeographic provinces of total and methyl mercury in zooplankton and fish from the Beaufort and Chukchi Seas: Results from the SHEBA drift. Environmental Science and Technology 39 (13): 4707-4713 pp.

Strange, N. E. 1985. The effect of the Churchill River Diversion on muscle mercury concentrations in fish from Southern Indian Lake and Issett Lake, Manitoba. North/South Consultants Inc. Winnipeg, MB. 38 pp.

Strange, N. E. 1993. Mercury in fish in northern Manitoba reservoirs and associated waterbodies: results from 1992 sampling. A report prepared for the program for monitoring mercury concentrations in fish in northern Manitoba reservoirs by North/South Consultants Inc., Winnipeg, MB.

Strange, N. E. 1995. Mercury in fish in northern Manitoba reservoirs and associated waterbodies: results from 1994 sampling. A report prepared for the program for monitoring mercury concentrations in fish in northern Manitoba reservoirs by North/South Consultants Inc., Winnipeg, MB. 46 pp.

Strange, N. E., and Bodaly, R. A. 1997. Mercury in fish in northern Manitoba reservoirs and associated water bodies: summary report for 1992, 1994 and 1996 sampling. A report prepared for the program for monitoring mercury concentrations in fish in northern Manitoba reservoirs by North/South Consultants Inc., Winnipeg, MB. 71 pp.

Strange, N. E., and Bodaly, R. A. 1999. Mercury in fish in northern Manitoba reservoirs and associated waterbodies: results from 1998 sampling. A report prepared for the program for monitoring mercury concentrations in fish in northern Manitoba reservoirs by North/South Consultants Inc. and DFO, Winnipeg, MB. 56 pp.

Strange, N. E., Bodaly, R. A., and Fudge, R. J. P. 1991. Mercury concentrations of fish in Southern Indian Lake and Issett Lake, Manitoba, 1975-88: the effect of lake impoundment and Churchill River diversion. Canadian Technical Report of Fisheries and Aquatic Sciences 1824: 29 pp.

Swanson, G. 1986. An interim report on the fisheries of the lower Nelson River and the impacts of hydro-electric development, 1985 data. MS Report No. 86-19, Fisheries Branch, Manitoba Department of Natural Resources, Winnipeg MB. 228 pp.

Swanson, G. M., and Kansas, K. R. 1987. A report on the fisheries resources of the lower Nelson River and the impacts of hydro-electric development, 1986 data. MS Report No. 87-30, Fisheries Branch, Manitoba Department of Natural Resources, Winnipeg MB. 240 pp.

Therrien, J., and Schetagne, R. 2008. Aménagement hydroélectrique de L'Eastmain-1. Suivi environnemental en phase d'exploitation (2007). Suivi du mercure dans la chair des poissons. Joint report by GENIVAR Income Fund and Hydro-Québec. 45 pp.

Therrien, J., and Schetagne, R. 2009. Réseau de suivi environnemental du complexe La Grande (2008) — Évolution du mercure dans la chair des poissons dans le secteur ouest. Joint report by Hydro-Québec and GENIVAR Income Fund. 50 pp.

https://www.researchgate.net/journal/0013-936X_Environmental_Science_and_Technology


DECEMBER 2015 5.5-113

Van Walleghem, J. L. A., Blanchfield, P. J., and Hintelmann, H. 2007. Elimination of mercury by Yellow Perch in the wild. Environmental Science and Technology 41: 5895–5901 pp.

Watras, C. J., Back, R. C., Halvorsen, S., Hudson, R. J. M., Morrison, K. A., and Wente, S. P. 1998. Bioaccumulation of mercury in pelagic freshwater food webs. Science of the Total Environment 219: 183–208 pp.

Wheatley, B. 1979. Methylmercury in Canada: exposure of Indian and Inuit residents to methylmercury in the Canadian environment. Health and Welfare Canada, Medical Service Branch, Ottawa, ON. 200 pp.

Wiener, J. G., and Spry, D. J. 1996. Toxicological significance of mercury in freshwater fish. In Environmental contaminants in wildlife: interpreting tissue concentrations. Edited by W. N. Beyer, G. H. Heinz, and A. W. Redmon-Norwood. Society of Environmental Toxicology and Chemistry, Lewis Publisher, Boca Raton, FL. 297–339 pp.

Wobeser, G., Nielsen, N. O., and Dunlop, R. H. 1970. Mercury concentrations in tissues of fish from the Saskatchewan River. Journal of the Fisheries Research Board of Canada 27: 830-834 pp.

5.5.7.2 Personal Communications Bodaly, Drew. 2015. Retired DFO Scientist. Telephone correspondence with Wolfgang Jansen,

North/South Consultants Inc., Winnipeg, MB, May 21, 2015.

Burgess, Neil. 2015. Wildlife Toxicologist. Environment Canada, Mount Pearl Newfoundland and Labrador. Telephone correspondence with Wolfgang Jansen, North/South Consultants Inc., Winnipeg, MB, May 7, 2015.

Burns-Flett, Erin. 2015. Quality Manager. National Aquatic Animal Health Program, DFO, Winnipeg Manitoba. Telephone correspondence with Wolfgang Jansen, North/South Consultants Inc., Winnipeg, MB, June 4, 2015.

Schetagne, Roger. 2011. Mercury Program Manager. Hydro Québec, Montreal Quebec. Conversation with Wolfgang Jansen, North/South Consultants Inc., Winnipeg, MB, July 26, 2011.

REGIONAL CUMULATIVE EFFECTS ASSESSMENT – PHASE II WATER – FISH QUALITY

DECEMBER 2015 5.6-1

5.6 Fish Quality

5.6.1 Introduction Fish quality was selected as a Regional Study Component (RSC), because reduced quality can affect domestic fish consumption and can reduce the market price of fish harvested by the commercial fisheries, which are key economic activities in many of the affected communities.

Fish quality, for the purpose of this document, is considered primarily in terms of: • the palatability of the fish (e.g., taste and texture); and

• the rate of infestation (RI) of the tapeworm Triaenophorus crassus, in Lake Whitefish (Coregonus clupeaformis).

PALATABILITY

Palatability can directly affect the acceptability of fish as a domestic food source. Over the years, most of the First Nations in the Regional Cumulative Effects Assessment (RCEA) Region of Interest (ROI) have expressed concerns, under the auspices of the Northern Flood Agreement, regarding the reduced quality of fish from impacted waterbodies, which they attribute to the effects of hydroelectric development. At the request of some of these communities, studies have been conducted to evaluate the quality of fish for human consumption when prepared using traditional methods. These studies will be discussed, by area, in the following sections.

LAKE WHITEFISH RATES OF T. CRASSUS INFESTATION

Triaenophorus crassus infects multiple hosts during its life cycle, but it is an intermediate phase that forms visible, yellowish cysts in the muscle of, among other species, Lake Whitefish and Cisco (Coregonus artedi) that is of primary concern. Although not harmful to humans (i.e., humans are not a viable host), the cysts are considered objectionable in appearance and undesirable, affecting the market price of commercial Lake Whitefish catches and, thus, the economic viability of many fisheries. The Freshwater Fish Marketing Corporation (FFMC) grades the quality of Lake Whitefish from lakes with commercial fisheries based on their RI (RI = number of cysts per 100 lbs [45.4 kg] of dressed [gutted and cleaned for market] fish) with T. crassus. As the number of cysts in the muscle of fish increases, the marketability of the catch decreases. Currently, the FFMC grades fish to two levels of marketability: “export” and “other” (“other” was formerly subdivided into “continental” and “cutter”). It is important to note that the grade or classification of a given commercial fishery is determined by the worst result during inspection. Even if there are abundant export quality fish in a lake, small numbers of lower quality fish will result in a grade of “other” for the entire fishery. Until 1988, RI values of 0–50 were graded as “export”, values of 51–80 were graded as “continental”, and values greater than 80 were graded as “cutter” by the FFMC (FFMC unpublished [unpubl.] data). The current FFMC approach grades all RI values of 0–50 as “export” and everything higher as “other”. Canada Department of Fisheries and Oceans (DFO) has used lake classifications for Lake Whitefish with an export range of 0-40 cysts/100 lbs, slightly different than the


DECEMBER 2015 5.6-2

range used by the FFMC. The grade of Lake Whitefish can greatly affect its marketability. For example, the FFMC prices to fishers during the 2013/2014 commercial fishing season for “export” Lake Whitefish were up to $0.75/kg ($0.34/lb) higher than for “other”.

OTHER FISH QUALITY ISSUES

Recently, with the arrival of the invasive Rainbow Smelt (Osmerus mordax) in the RCEA ROI, there have often been reports of “belly burn” in predatory fish such as Walleye (Sander vitreus) if the fish are not cleaned shortly after capture. This is caused by an enzyme in smelt that increases the decomposition rate of the predatory fish’s stomach following its death; therefore, “belly burn” is a function of the presence of smelt in the predatory fish’s diet and is not connected to hydroelectric development. As such, this aspect of fish quality is not included in the following discussion.

Mercury in fish also affects its acceptability for domestic consumption and commercial sale. Due to its relative importance and the quantity of available information, mercury in fish is discussed in detail in its own chapter (Mercury in Fish, Chapter 5.5).

5.6.1.1 Pathways of Effect

PALATABILITY

There are no known linkages between hydroelectric development in the RCEA ROI and the change in the palatability of fish reported by the First Nations. A linkage diagram has, therefore, not been provided for palatability. However, it should be noted that hydroelectric development can cause changes to fish diet, water quality, water temperature, algae, and growth rates, all of which can, in turn, affect the taste and texture of fish.

It should be noted, however, that cyanobacteria (blue-green algae) and actinomycetes bacteria produce compounds (geosmin and 2-methylisoborneol) that have been linked to objectionable or off-flavours (often described as muddy, earthy, or mouldy) in fish (Persson 1980; Yurkowski and Tabachek 1980; Schrader and Blevins 1993). High levels of these compounds have been found in nutrient rich waterbodies which contain cyanobacteria blooms including; aquaculture ponds (Lorio et al. 1992) and flooded lakes and reservoirs (Yurkowski and Tabachek 1980; Izaguirre et al. 1999). Yurkowski and Tabachek (1980) stated that the presence of this off-flavour in several species of fish from Cedar Lake, Manitoba (MB) in the mid-1970s was likely the result of input of additional organic matter and corresponding increases in cyanobacteria following construction of the Grand Rapids Generating Station (GS) and flooding of the lake. However, they were unable to directly link the muddy flavour to a source at the time it occurred and the problem has not been reported since. Information on cyanobacteria blooms in lakes in the RCEA ROI is limited, but anecdotal evidence suggests they are uncommon. As a result, it seems unlikely that fish palatability in the RCEA ROI has been affected by proliferation of blue green algae/cyanobacteria.


DECEMBER 2015 5.6-3


The life history of T. crassus has been well documented (Miller 1952; Lawler and Scott 1954; Lawler 1955; Watson and Dick 1979a). The parasite requires three hosts to complete its life cycle. The first larval stage is found in cyclopoid copepods (a type of zooplankton). The second host, in which the parasite forms cysts in the muscles, is typically a coregonine fish (e.g., Lake Whitefish and Cisco). The final host, in which the parasite matures to an adult in the intestine, is typically Northern Pike (Esox lucius). During Northern Pike spawning in the spring, the adult T. crassus detaches itself from the gut wall of the fish and, as it passes out of the fish and into the water, releases its eggs in littoral areas. The eggs hatch after one to two weeks and the free-swimming larvae are then consumed by copepods (usually Cyclops bicuspidatus in the RCEA ROI). When the copepod is, in turn, consumed by Lake Whitefish or Cisco, the parasite is released, penetrates the lining of the stomach, and becomes embedded in the tissue of the fish. Northern Pike feeding on these Lake Whitefish become infected and the cycle continues.

There are several effects that could increase the T. crassus RI in Lake Whitefish, including an increase in the abundance of cyclopoid copepods, increased proportions of copepods in Lake Whitefish diets (as opposed to benthic prey items), and increased abundance of Northern Pike (Figure 5.6.1-1). Pulkkinen and Valtonen (1999) have also suggested that there is an increase in the annual accumulation of cysts in fish hosts that is independent of dietary changes, but is associated with a threshold intensity of infestation above which the probability of acquiring further parasites increases. This is likely a result of a compromised immune system in heavily infested fish.

In some lakes where there are multiple, distinct populations of Lake Whitefish, cyst counts will vary between populations. Dietary changes associated with growth can also result in multiple RI, by size cohort, within the same population in a lake. For example, Lake Whitefish may switch from feeding on zooplankton as juveniles to consuming larger, benthic prey items as adults. Commercial fishers generally target the whitefish with the lower cyst counts as these are sold at higher prices. In some lakes, changes in water levels, habitat, and/or hydrology, which may all be effects of hydroelectric development, have caused fish to redistribute themselves and the Lake Whitefish with the lower cyst counts are no longer in the traditional fishing areas (Bodaly et al. 1984). Redistribution of Lake Whitefish populations within or between lakes may also expose them to increased rates of infestation in the copepods they consume.

Increases in copepod and Northern Pike abundance or redistribution of any of the three hosts, which may occur with flooding and reduction of velocities in impounded waterbodies, are the most likely linkages between hydroelectric development and increases in rates of T. crassus infestation in Lake Whitefish (Figure 5.6.1-1).

5.6.1.2 Indicators and Metrics

PALATABILITY

Indicators for fish palatability are: a) the acceptability of fish to community members; and b) the results of scientific tests conducted on fish palatability. As mentioned above, fish taste is a very subjective quality, influenced by a variety of factors. Because the results of the scientific studies generally differ from the


DECEMBER 2015 5.6-4

experiences expressed by the community, information is provided on both with no attempt to determine which is correct.


The indicator used for Lake Whitefish quality regarding rate of T. crassus infestation is the RI or number of cysts per 100 lbs (45.4 kg) of dressed (gutted and cleaned for market) commercial whitefish.

5.6.1.3 Data Limitations The absence of any scientific assessments on fish palatability before hydroelectric development does not allow for quantitative assessments of change over time. Assessments can only be made between on-system and off-system waterbodies.

With the exception of Southern Indian Lake (SIL), there are few, robust T. crassus RI datasets for waterbodies either before or after hydroelectric development, which limits the ability to determine if changes have occurred in many of these locations.


DECEMBER 2015 5.6-5

Arrow width is not an indication of magnitude.

Figure 5.6.1-1: Linkage Diagram Showing Potential Pathways of Effect of Hydroelectric Development and Other Factors on Fish Quality


DECEMBER 2015 5.6-6

5.6.2 Area 1: Lake Winnipeg Outlet to Kelsey Generating Station

Area 1 comprises the reach of the Nelson River from the outlet of Lake Winnipeg to the Kelsey (GS) (Map 5.6.2-1). The upper portion of Area 1 consists of the Outlet Lakes area, including Playgreen Lake. The river flows via the east and west channels of the Nelson River to Cross Lake. Downstream of Cross Lake, the river passes over a series of rapids to Sipiwesk Lake and then, prior to the construction of the Kelsey GS, through several more rapids to Kelsey Rapids, which is the site of the Kelsey GS. Several hydroelectric developments have been constructed in the area. Kelsey GS at Kelsey Rapids on the Nelson River caused some flooding upstream as far as Sipiwesk Lake. Lake Winnipeg Regulation (LWR). via three diversion channels in the Outlet Lakes area, increased Lake Winnipeg outflow capacity by up to 50%, affecting several downstream waterbodies. The Jenpeg GS, constructed on the west channel of the Nelson River where it entered Cross Lake, flooded the river channel immediately upstream and regulates outflows from Lake Winnipeg along the west channel of the Nelson River. LWR can also affect seasonal pattern of flows along the west channel of the Nelson River and downstream, with peak flows now occurring during the winter months and lower flows during the open water season in low to average flow years. Flow regulation at Jenpeg also affects water levels and flows in Cross Lake and the river reach between Cross Lake and Sipiwesk Lake. Detailed information regarding project descriptions for hydroelectric developments in Area 2 and their effects on the local water regime can be found in Hydroelectric Development Project Description in the Region of Interest, Part II Appendices 2A, 2D, and 2E and in Water Regime, Section 4.3.2.

5.6.2.1 Approach and Methods

PALATABILITY

Canada Department of Fisheries and Oceans was subcontracted by North/South Consultants Inc. to conduct a taste-test study in 1993–1994 (Davies et al. 1998a). The Analytical Services Section of DFO had a specific protocol for analyzing fish quality by qualitatively assessing fish texture, colour, interior and exterior appearance, smell, and taste. Samples were also tested for a wide range of metals and pesticides. The protocol is outlined in DFO's "Fish Products Inspection Procedure Manual". Samples of Lake Whitefish, Northern Pike, and Walleye were obtained from Playgreen and Little Playgreen lakes and Mossy Bay in Lake Winnipeg and were prepared using traditional methods (boiling or pan-frying). The fish evaluated for quality were classified as either “satisfactory” (i.e., they passed the qualitative assessment of parameters listed above) or “unsatisfactory” (i.e., they failed one or more of the qualitative assessments) by an individual trained to detect substances contaminating fish flesh.


Within Manitoba, FFMC currently grades Lake Whitefish marketability from a given waterbody as either “export” or “other” based on T. crassus RI. To arrive at this grade, a set number of fish are inspected periodically. For each inspected fish, the total weight is recorded (± 25 g) and, if necessary, converted to


DECEMBER 2015 5.6-7

pounds. The fish are filleted and thin slices (no thicker than 6.35 mm [¼”]) are made at right angles to the length of each fillet. The number of T. crassus cysts in each slice are counted and the RI is calculated according to the following formula:

RI = (number of cysts/lb of fish samples) × 100

It is important to note that cyst counts describing “export” and “other” quality Lake Whitefish have historically differed between surveys. Some surveys (e.g., SIL studies from the 1970s and 80s) have used federal government (i.e., DFO) guidelines, for which a range of 0-40 cysts/100 lbs (dressed weight) is considered export and others have used the FFMC range of 0-50 cysts/100 lbs to describe export quality fish. Ranges used for each data set will be provided when possible.

5.6.2.2 Key Published Information

PALATABILITY

In response to concerns of a decline in the quality of fish (in terms of taste, texture, and size) expressed by Norway House, North/South Consultants Inc. initiated a taste-test study on fish from Little Playgreen Lake (including the Jack and Gunisao rivers), Playgreen Lake, and Mossy Bay in Lake Winnipeg (Davies et al. 1998a).


Rates of infestation by T. crassus cysts in Lake Whitefish have been recorded during scientific investigations in many waterbodies in Area 1 since the 1950s, including Sipiwesk Lake (Schlick 1966a; Sunde 1965a), Walker Lake (Sunde 1964, 1965a), Drunken Lake (Nelson River Group [NRG] 1986) and Cross, Duck, Kiskitto, Pipestone, and Playgreen lakes (Sunde 1965a). Mackay et al. (unpubl. data) summarized cyst count data from the Sipiwesk Lake commercial fishery between 1974 and 1988. Sowe (1986) examined the factors influencing T. crassus abundance in Lake Whitefish from several commercially fished lakes, including Playgreen, Sipiwesk, and Walker lakes in Area 1.

5.6.2.3 New Information and/or Re-analysis of Existing Information The FFMC routinely samples Lake Whitefish from Manitoba lakes with commercial fisheries (including at least 10 lakes in Area 1) and counts cysts from dressed fish in their laboratory in Winnipeg. Fish are then graded based on their RI. Current (2015) Lake Whitefish grading results (unpubl. data) were obtained for comparison with earlier data.

PimicikamakCross Lake (NAC)

Norway House Cree Nation Norway House (NAC)

Thicket Portage(NAC)

Pikwitonei (NAC)

Herb Lake Landing(NAC)

Wabowden (NAC)

Snow Lake

Thompson

6

6

39

373

Kelsey G.S.

Jenpeg G.S.

WuskwatimG.S.

KiskittoLake

Nelso

n

PlaygreenLake

KiskittogisuLake

LAKEWINNIPEG

PlaygreenLake

SipiweskLake

Cross LakeWalkerLake

DuckLake

River

River

PipestoneLake

DrunkenLake

Nelso

n

File

Loc

atio

n: J

:\MY

P\M

H_R

CE

A\F

ish_

Qua

lity\

PH

AS

E_2

\Mxd

\201

5120

5_R

CE

A_P

hase

II_A

E_F

Q_E

E_F

ishQ

ualit

yAss

essm

entA

reaL

akeW

inni

pegO

utle

tToK

else

yGS

RC

EA

Are

a1_c

p.m

xd

1.0

27-JUL-15


Cre

ated

By:

cpa

rker

- B

Siz

e La

ndsc

ape

STD

- M

AR

201

5

Sca

le: 1

:1,2

27,0

00

05-DEC-15

Government of Canada, Province of Manitoba, Manitoba Hydro

Fish Quality Assessment AreaLake Winnipeg Outlet to

Kelsey Generating Station RCEA Area 1

LegendRCEA Region of Interest

First Nation Reserve

Hudson Bay

Thompson

Churchill


DATA SOURCE:


CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:


0 10 20 Miles

0 10 20 Kilometres

Note: Northern Affairs Community (NAC)

RCEAArea 1

InfrastructureGenerating Station (Existing)

Transmission Line (Existing)

Rail

Highway

Map 5.6.2-1


DECEMBER 2015 5.6-9

5.6.2.4 Changes in Indicators over Time

PALATABILITY

Members of the Northern Flood Committee (including the communities of Norway House and Cross Lake) filed a claim in 1984, under the auspices of the Northern Flood Agreement, alleging that the modification of the water regime by Manitoba Hydro had adversely affected domestic fisheries, including several in Area 1 waterbodies. The primary impact identified by Norway House members with regards to the domestic fishery was a decline in the taste, texture, and size of fish in the domestic harvest (see People, Section 3.5.10). Lake Whitefish, Northern Pike, and Walleye from Little Playgreen (including the Jack and Gunisao rivers) and Playgreen lakes and Mossy Bay in Lake Winnipeg were captured and forwarded to DFO in 1993 and 1994 for analysis of colour, texture, smell and, in 1994 only, for taste (Davies et al. 1998a). Results of those tests identified all samples as “satisfactory”, as compared to “unsatisfactory”. Further complicating the issue, studies have shown extensive movements of fish between waterbodies in Area 1 (e.g., fish movement between Playgreen Lake and Lake Winnipeg; Davies et al. 1998b).

Cross Lake members also have stated that the taste and texture of fish in Cross Lake has declined since hydroelectric development. The domestic fishing program harvests fish from Cross Lake during the open water period and other lakes (see People, Section 3.5.5) during the winter period. Large quantities of fish have been provided to and consumed by the community. However, some Cross Lake members have stated that they prefer fish from off-system areas and some members have stated that they have purchased fish from those waterbodies (S. Davies pers. comm. 2015).

LAKE WHITEFISH T. CRASSUS RATE OF INFESTATION

Generally, Lake Whitefish from most Area 1 waterbodies sampled between 1960 and 1990 were graded as export quality (Table 5.6.2-1). Only Walker Lake (Sunde 1964, 1965a), an off-system lake, was graded as the lower value “other” classification (identified as continental using the older grading classification). In 1980, the Walker Lake fishery was downgraded from continental to cutter, which meant that harvesting of Lake Whitefish had become uneconomical in the lake (NRG 1986). In addition, since 1990, five Area 1 lakes (Drunken, Duck, Kiskitto, Pipestone, and Sipiwesk) previously identified as export have been downgraded to other (Table 5.6.2-1). However, historical data for those lakes were relatively limited (in survey duration and/or sample size) for all but Sipiwesk (Schlick 1966a; Sunde 1965a; NRG 1986; Sowe 1986; Mackay et al. unpubl. data), which may have affected the original classification.



Table 5.6.2-1: Historical and Current Grades for Lake Whitefish in Area 1 Waterbodies Based on Triaenophorus crassus Cyst Counts

Waterbody Pre-1980 1,2 1980–1990 3 2006–2011 4 2015 4

Cross Lake Export - Export Export

Drunken Lake - Export - Other

Duck Lake Export - Other Other

Kiskitto Lake Export - Other Other

Kiskittogisu Lake - - Export Export

Nelson River - - Other Other

Pipestone Lake Export - Other Other

Playgreen Lake Export Export Export Export

Sipiwesk Lake Export Export Other Other

Walker Lake Other Other Other Other

Lake Winnipeg Export Export Export Export

1. Export = 0–50 cysts/100 lbs of fish (dressed weight); Other (includes Continental and Cutter grades) = > 50 cysts/100 lbs of fish (dressed weight).

2. Data sources: Sunde (1964, 1965a); Schlick (1966a); Sowe (1986); Mackay et al. (1990). 3. Data sources: Sowe (1986); Mackay et al. (1990). 4. Data sources: FFMC (unpubl. data).

5.6.2.5 Cumulative Effects of Hydroelectric Development on Fish Quality in Area 1

PALATABILITY

There is no scientific evidence to suggest that the taste, texture or palatability of fish has been affected by hydroelectric development in Area 1. However, the people who consume the fish in the communities have clearly identified this as an issue that is related to hydroelectric development.

LAKE WHITEFISH T. CRASSUS INFESTATION RATES

As illustrated in the pathways of effects diagram (Figure 5.6.1-1) the RI for T. crassus in the muscle of Lake Whitefish can be linked to hydroelectric development. Some of these links have been established for waterbodies in other areas (e.g., SIL; see Section 5.6.4). However, there have been few detailed T. crassus studies in Area 1 waterbodies and no definitive link has been established. For example, although LWR was linked to a decline in the Lake Whitefish stocks in Cross Lake, the grade of the fishery remained export quality. There does not appear to be any known change to Northern Pike or zooplankton



abundance that may have affected T. crassus counts in Area 1 waterbodies, though more detailed studies would be required to completely rule out any linkage.



5.6.3 Area 2: Split Lake to Nelson River Estuary Area 2 comprises the reach of the Nelson River from the Kelsey GS to the Nelson River Estuary. Major waterbodies within this area include Split Lake, Gull Lake, Stephens Lake, Long Spruce reservoir, Limestone reservoir, and the lower Nelson River (Map 5.6.3-1). Several hydroelectric developments have been constructed in the area. Kelsey GS at Kelsey Rapids on the Nelson River caused some upstream flooding (Area 1), but there were negligible effects to downstream waterbodies. Kettle GS at Kettle Rapids on the Nelson River affected the water regime and fish habitat in the reach of the river extending to Gull Rapids, causing some flooding between Gull and Kettle rapids. Beginning in mid-1976, LWR affected the seasonal distribution of flow as it entered Split Lake and Churchill River Diversion (CRD) increased the total inflow to Split Lake. Long Spruce GS and Limestone GS on the Nelson River each caused small amounts of upstream flooding and, depending on local conditions, dewater large areas of the un-impounded downstream riverbed on a daily or weekly basis. Detailed information regarding project descriptions for hydroelectric developments in Area 2 and their effects on the local water regime can be found in Hydroelectric Development Project Description in the Region of Interest, Part II Appendices 2A, 2B, 2D, 2F, 2G, 2H, and 2L and in Water Regime, Sections 4.3.2 and 4.3.4.


PALATABILITY

Prior to construction of the Keeyask GS, and to address community concerns regarding a perceived reduction in fish quality in some Area 2 waterbodies, members of Tataskweyak Cree Nation (TCN) at Split Lake, York Factory First Nation (YFFN) at York Landing, and Fox Lake Cree Nation (FLCN) at Bird were asked to judge the palatability of Lake Whitefish, Northern Pike, and Walleye from waterbodies of their choice in the vicinity of each community (Ryland and Watts 2002a, 2004a, 2004b). The acceptability of fish from lakes that had previously been affected by hydroelectric development was compared to fish from at least one unaffected nearby lake.

Waterbodies sampled for fish used in the palatability tests included the Aiken River, Split Lake, Assean Lake, Gull Lake, and Stephens Lake and from three of the Adverse Effects Agreements (AEA) offsetting lakes: Atkinson Lake, War Lake, and Waskaiowaka Lake and the Limestone reservoir. The AEA offsetting lakes were used as regional reference lakes that will not experience direct physical effects of the Keeyask Generation Project and by specific request of the three communities.

Fish Quality Assessment Area Kelsey Generating Station to

Nelson River Estuary RCEA Area 2

Government of Canada, Province of Manitoba, North/South Consultants andManitoba Hydro.


1.0

27-AUG-15

LegendRCEA Region of Interest Generating Station (Existing)

Generating Station (Under Construction)


Transmission Line (Existing)

Highway

Rail

First Nation Reserve07-DEC-15

ButnauLake

HUDSONBAY

AtkinsonLake

Split Lake

Nelson River

StephensLake

Assean Lake

Angling Lake

Little Limestone Lake

Moose NoseLakeKelsey

G.S.

Long Spruce G.S.

Limestone G.S.KettleG.S.

KeeyaskG.S.

ConawapaG.S.

Gillam Fox Lake Cree NationA Kwis Ki Mahka Reserve

York Factory First NationYork Landing

(Kawechiwasik)

Fox Lake Cree NationFox Lake (Bird)

War Lake First NationIlford (NAC)

Tataskweyak Cree NationTataskweyak (Split Lake)

ShamattawaFirst Nation

WaskaiowakaLake

WarLake

RiverLimestone

Nelson RiverClarkLake

290280

Cre

ated

By:

cpa

rker

- B

Siz

e La

ndsc

ape

BTB

- M

AR

201

5

Sca

le: 1

:1,1

80,1

14



0 10 20 Miles

0 10 20 Kilometres

File

Loc

atio

n: J

:\MY

P\M

H_R

CE

A\F

ish_

Qua

lity\

PH

AS

E_2

\Mxd

\201

5120

5_R

CE

A_P

hase

II_A

E_F

Q_E

E_F

ishQ

ualit

yAss

essm

entA

reaK

else

yGS

toN

elso

nRE

stua

ryR

CE

AA

rea2

_cp.

mxd

DATA SOURCE:

DATE CREATED:

CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:

COORDINATE SYSTEM:

RCEAArea 2

Hudson Bay

Thompson

Churchill

Map 5.6.3-1



Species-specific size ranges of fish (Lake Whitefish: 1,400–2,000 g; Northern Pike: 2,700–3,000 g; Walleye: 700–1,000 g) were collected with gill nets set three to five weeks prior to taste testing. Fillets (Northern Pike and Walleye) or scaled and eviscerated whole fish (Lake Whitefish) were prepared on-site and stored frozen until taste sample preparation. The samples consisted of small (less than 10 g) pieces of fried (Northern Pike and Walleye) or boiled and deboned flesh (Lake Whitefish) prepared by local cooks, that were presented to panellists three to four at a time, with each piece representing a coded sample from each lake. Fish species were evaluated on separate days. Between 23 and 48 panellists, stratified by age (16–35, 36–55, greater than 55 years) and consumption habit (eat fish every day, three to four times per week, one to two times per week, three times per month) tasted fish in each community. Panellists expressed their acceptability of the fish based on a seven-point category scale from “like very much” (7) to “dislike very much” (1) (for a full list of categories see Ryland and Watts 2002a; 2004a, b).

Analysis of variance followed by, if applicable, Duncan’s test for means comparisons was conducted on the numerical categories of fish acceptability using PROC GLM (SAS, PC Version 8.2) to determine if there were significant differences among the acceptability of fish of the same species from different locations or among panellist age categories.


Methods used by the FFMC to grade commercial Lake Whitefish based on infestation rates is described in detail in Section 5.6.2.1.

In addition to the routine commercial samples assessed by the FFMC, 87 Lake Whitefish were captured for T. crassus analysis as part of the Keeyask environmental studies from Split, Gull, Stephens and Waskaiowaka lakes during summer/fall of 2003–2006. Sampled fish were gutted and stored at -18°C in freezers for subsequent delivery to the FFMC (Winnipeg, MB) where RI was calculated using established FFMC methods.


PALATABILITY

To describe existing conditions in the study area, fish palatability studies were conducted in the communities of Split Lake (TCN) on fish from Split, Gull, Assean, and Waskaiowaka lakes from 20–22 October, 2003 (Ryland and Watts 2004a), York Landing (YFFN) on fish from Split Lake, War Lake, and the Aiken River from 29 April to 6 May, 2002 (Ryland and Watts 2002a), and Bird (FLCN) on fish from Gull, Stephens, and Atkinson Lakes, and the Limestone reservoir from 28–30 October, 2003 (Ryland and Watts 2004b).

The effects of previous hydroelectric development in northern Manitoba, including the effects on fish palatability, were assessed in the Split Lake Resource Management Area as part of the Split Lake Cree Post Project Environmental Review (Split Lake Cree - Manitoba Hydro Joint Study Group 1996a, b, c).



LAKE WHITEFISH T. CRASSUS RATES OF INFESTATION

Extensive historical information on Lake Whitefish infestation rates with T. crassus cysts were recorded in the 1950s and 1960s (Schlick 1967; Sunde 1965a, b). These surveys included Split Lake in Area 2, which was sampled in 1959 and from 1962–1966. Four of the AEA offsetting lakes for the Keeyask Generation Project (Atkinson, War, Moose Nose, and Waskaiowaka lakes) were also sampled historically by DFO and/or by the Fisheries Branch of the Manitoba Department of Mines and Natural Resources (Sunde 1965a; Schlick 1967, 1979; Moshenko 1968). Historical information on rates of infestation of Lake Whitefish by T. crassus is also available from commercial monitoring in Split Lake by the FFMC (FFMC unpubl. data).

The effects of previous hydroelectric development in northern Manitoba, including the effects to rates of infestation of Lake Whitefish by T. crassus, were assessed in the Split Lake Resource Management Area as part of the Split Lake Cree Post Project Environmental Review (Split Lake Cree - Manitoba Hydro Joint Study Group 1996a, b, c).

Lake Whitefish were also captured for T. crassus analysis as part of the Keeyask environmental studies from Split, Gull, and Stephens lakes in Area 2 and from Waskaiowaka Lake during summer/fall of 2003-2006.

5.6.3.3 New Information and/or Re-analysis of Existing Information The FFMC routinely samples Lake Whitefish from Manitoba lakes with commercial fisheries and counts cysts from dressed fish in their laboratory in Winnipeg. Fish are then graded based on their RI. Current (2015) Lake Whitefish grading results (unpubl. data) were obtained from Angling, Assean, Butnau, Little Limestone, Split, and Stephens lakes for comparison with earlier data.


PALATABILITY

Members of TCN, YFFN, and FLCN all expressed concerns about the decline in fish quality (as determined by taste and texture) in area waterbodies affected or potentially affected by hydroelectric development. To address these concerns, fish palatability tests using Lake Whitefish, Walleye, and Northern Pike were conducted in each community (Ryland and Watts 2002a, 2004a, b). The results of those studies identified ratings for all samples as “like slightly” or “like moderately”. Panellists from all three communities showed some preference for fish from lakes that have not been previously affected by hydroelectric development. Generally, TCN and FLCN panellists rated fish, particularly Walleye, from Gull Lake as less acceptable than fish from Waskaiowaka Lake, War Lake, Atkinson Lake, and Assean Lake.

There was no significant difference in the mean acceptability of fish species sampled from different waterbodies by YFFN panellists. However, it should be noted that the results of the palatability study at York Landing do not reflect the views of most community members on the taste of fish species used in the study. In particular, there was a concern expressed by some members that no fish were tested that were captured when water temperatures were warmer and taste differences would be more noticeable.




Historical RI data from Area 2 were restricted to Split Lake. More recently, data have been collected from an increasing number of Area 2 waterbodies (Table 5.6.3-1). Of all the sampled lakes, only Split Lake has been graded as export quality and it has consistently been graded as such since the 1960s (Sunde 1965a, Schlick 1967, FFMC unpubl. data). All other waterbodies have been graded as either cutter or continental (i.e., other). There have been no noticeable changes to the RI of T. crassus in Area 2 lakes and no obvious links to hydroelectric development in the region.


Waterbody Pre-1980 1,2 1980–1990 3 1991–2005 3, 4 2006–2011 3, 4 2015 3

Angling Lake - - - - Other

Assean Lake - - - - Other

Atkinson Lake 5 Export - - - -

Butnau Lake - - - - Other

Clark Lake - - - - -

Limestone River - - - Other Other

Little Limestone Lake - - - - Other

Moose Nose Lake 5 Other - - - -

Nelson River - - - Other Other

Split Lake Export Export Export Export Export

Stephens Lake - - - - Other

War Lake 5 Export - - - -

Waskaiowaka 5 Other - Other - -

1. Export = 0–40 cysts/100 lbs of fish (DFO) or 0–50 cysts/100 lbs of fish (FFMC); Other (includes Continental and Cutter grades) = > 40 or > 50 cysts/100 lbs of fish (dressed weight).

2. Data sources: Sunde (1965a); Schlick (1967); Moshenko (1968); Schlick (1979). 3. Data source: FFMC (unpubl. data). 4. Data source: Keeyask Hydropower Limited Partnership (2012). 5. Adverse Effects Agreements (AEA) offsetting lakes; though not technically in Area 2, these lakes have often been used as

supposedly unaffected comparisons with Area 2 lakes that may be affected by hydroelectric development.


PALATABILITY

The people who consume fish in Area 2 communities have clearly identified this as an issue that is directly related to hydroelectric development. Based on the palatability studies conducted from 2002-2004, fish from Gull Lake, which is affected by hydroelectric development, currently receive the



lowest acceptability scores by members of TCN and FLCN compared to other study area lakes and AEA offsetting lakes. However, the results of these studies were not statistically significant.


As illustrated in the pathways of effects diagram (Figure 5.6.1-1) the infestation rate for T. crassus in the muscle of Lake Whitefish can be linked to hydroelectric development. Some of these links have been established for waterbodies in other areas (e.g., SIL in Area 3; Section 5.6.4). Long-term data exist for Split Lake, which has consistently been an export quality fishery. However, there are insufficient T. crassus data from other Area 2 waterbodies to identify trends. There is currently no definitive link between increased RI in Lake Whitefish and hydroelectric development for any Area 2 waterbodies.



5.6.4 Area 3: Southern Indian Lake to Split Lake Inlet The upper portion of Area 3 is bounded on the upstream end by Leaf Rapids on the Churchill River and extends through Opachuanau Lake and all of SIL with its outlets at the Missi Falls Control Structure (CS) and the South Bay Diversion Channel (Map 5.6.4-1). Hydroelectric developments in the area include the CRD and Wuskwatim GS. The CRD Route extends from the inlet of the South Bay Diversion Channel downstream to First Rapids on the Burntwood River and includes the Missi Falls CS and Notigi CS. The CRD caused major changes to the water regimes of several waterbodies in Area 3, in particular Southern Indian and Notigi lakes. Wuskwatim GS caused some flooding of Wuskwatim Lake. Detailed information regarding project descriptions for hydroelectric developments in Area 2 and their effects on the local water regime can be found in Hydroelectric Development Project Description in the Region of Interest, Part II Appendix 2F and Water Regime, Section 4.3.3.


PALATABILITY

The University of Manitoba was hired by North/South Consultants Inc. to conduct a taste-test study in 2002 (Ryland and Watts 2002a) at Nelson House. The researchers conducted a double blind taste test using traditionally prepared fillets of Walleye, Northern Pike, and Lake Whitefish from Wuskwatim and Footprint lakes that were affected by CRD and Leftrook and Baldock lakes that were not affected by CRD. The taste test was conducted with 49 panellists from Nelson House. Samples of 30–40 g of freshly fried (Walleye and Northern Pike) or boiled (Lake Whitefish) pieces of fish were assigned unique, three digit random numbers and were evaluated by panellists for acceptability using a seven point category scale (1 = dislike very much, 2 = dislike moderately, 3 = dislike slightly, 4 = neither like nor dislike, 5 = like slightly, 6 = like moderately, 7 = like very much). Statistical analyses were conducted on the results.


Methods used by the FFMC to grade commercial Lake Whitefish based on infestation rates is described in detail in Section 5.6.2.1.

Additional cyst count information was collected from SIL during recent fish community investigations (Michaluk and Remnant 2012; Aiken and Remnant 2013; Aiken 2014). During these studies, Lake Whitefish fillets were cut into approximately 1.0 cm transverse strips to expose all T. crassus cysts, which were then enumerated. The sampling protocol for these studies was developed in conjunction with the SIL Fishermen’s Association and Manitoba Conservation and Water Stewardship.

Leaf Rapids

Thompson


Nisichawayasihk Cree NationNelson House (NAC)

KelseyG.S.

Wuskwatim G.S.

493

280

6

391

BirchTreeLake

Burntwood

RiverWuskwatim

L.

NotigiLake

IssettLake

OpachuanauLake

Lake

KinosaskawLake

KarsakuwigamakLake

MynarskiLakes

Southern

Indian

Osik L.

MacheewinLake

PemichigamauLake

MysteryL.

ThreepointLake

WapisuLake

Rat Lake

Apussigamasi

L.

Lake

NorthKnife

First Rapids

NotigiControl Structure

Missi FallsControl Structure

1.0

22-OCT-15



05-DEC-15

Hudson Bay

Thompson

Churchill

Fish Quality Assessment AreaSouthern Indian Lake to First Rapids

RCEA Area 3Regional Cumulative Effects Assessment


0 10 20 Kilometres

0 10 20 Miles

DATA SOURCE:

DATE CREATED:

CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:

COORDINATE SYSTEM:

RCEAArea 3

LegendRCEA Region of Interest Generating Station (Existing)

Control StructureTransmission Line (Existing)

HighwayRailFirst Nation Reserve

NOTE: (NAC) Northern Affairs Community

Created By: cparker - B Size Portrait BTB - MAR 2015 Scale: 1:1,057,000File Location: J:\MYP\MH_RCEA\Fish_Quality\PHASE_2\Mxd\20151205_RCEA_PhaseII_AE_FQ_EE_FishQualityAssessmentAreaSouthernIndianLakeToFirstRapidsRCEAArea3_cp.mxd

Map 5.6.4-1



5.6.4.2 Key Published Information Residents of both South Indian Lake and Nelson House have expressed concerns regarding the taste and texture of the fish harvested from impacted lakes after CRD, the Nisichawayasihk Cree Nation (NCN) requested that the palatability of the fish be tested by an external agency. The University of Manitoba was subsequently hired to conduct double blind taste tests (Ryland and Watts 2002b). The results of this study are also discussed as part of the Wuskwatim Generation Project Environmental Impact Statement (Manitoba Hydro and NCN 2003).

Lake Whitefish infestation rates with T. crassus cysts have been recorded during scientific investigations in many northern Manitoba waterbodies since the 1950s, including some in Area 3 (e.g., McTavish 1952; Sunde 1963a, b; Schlick 1966b) and records from the early 1960s have been summarized by Sunde (1965a). Extensive information was also collected from SIL during the 1970s and 1980s from multiple areas within the lake, including comparisons of infestation rates in dark and light coloured Lake Whitefish (Watson and Dick 1979a, b; Bodaly et al. 1984; Johnston 1984; Peristy 1989). Much of the work conducted in SIL tied increased cyst counts in Lake Whitefish with the downgraded quality of the Lake Whitefish fishery (Bodaly et al. 1984). More recently, cyst counts were collected from small samples of Lake Whitefish in Wuskwatim Lake as part of studies in support of the Wuskwatim Generation Project Environmental Impact Statement (Manitoba Hydro and NCN 2003).

The FFMC routinely samples Lake Whitefish from Manitoba lakes with commercial fisheries (including at least 12 lakes in Area 3) and counts cysts from dressed fish in their laboratory in Winnipeg. Fish are then graded based on their RI.

5.6.4.3 New Information and/or Re-analysis of Existing Information The status and health of Lake Whitefish stocks in SIL were investigated from 2011-2013 as part of Manitoba and Manitoba Hydro’s Coordinated Aquatic Monitoring Program and the SIL Environmental Steering Committee studies. During these studies, cysts counts were collected from multiple areas in the lake (Michaluk and Remnant 2012; Aiken and Remnant 2013; Aiken 2014) and an attempt was made to distinguish between light and dark colour whitefish (Michaluk and Remnant 2012).

The FFMC routinely samples Lake Whitefish from Manitoba lakes with commercial fisheries and counts cysts from dressed fish in their laboratory in Winnipeg. Fish are then graded based on their RI. Current (2015) Lake Whitefish grading results (unpubl. data) were obtained from 19 waterbodies in Area 3 for comparison with earlier data, where applicable.


PALATABILITY

Both O-Pipon-Na-Piwin Cree Nation and NCN have stated that the taste and texture of the fish harvested for domestic consumption declined after CRD. The decline in the quality was the subject of a domestic fishing claim in South Indian Lake, which was settled with Manitoba Hydro in 1992. The reduced



palatability of the fish harvested for domestic purposes also was raised by NCN during the Wuskwatim Generation Project environmental assessment. Due to concerns regarding the future palatability of fish in Wuskwatim Lake following the construction of the Wuskwatim GS and the changes that NCN felt had occurred as a result of CRD, a scientific study was requested by the First Nation to examine this.

The University of Manitoba was hired to conduct the study in 2002 (Ryland and Watts 2002b). Results from this study, which was conducted in a scientific manner with all appropriate controls, found no significant difference in the acceptability of fish among lakes for any species. However, following the receipt of the report, some members of NCN stated that some of the panellists were not known to consume large quantities of fish and that several other factors may have affected the results of the test (S. Davies pers. comm.). The study did not resolve the issue of fish palatability for NCN.


The vast majority of historical information on T. crassus infestations in Area 3 waterbodies is from SIL where there has been a significant commercial fishery since 1941 (Bodaly et al. 1984). Due to its size and varying physical characteristics and uses between different regions of the lake, SIL has traditionally been subdivided into seven distinct areas (Map 5.6.4-2). For the purposes of this document and to avoid confusion when referring to the overall Area 3, which includes multiple lakes and rivers, areas within SIL will be referred to as Historical Geographic Areas and identified as SIL Area 4, SIL Area 5 etc. Scientific studies on the lake have generally sampled most of these areas, but the commercial fishery was historically, before CRD, concentrated (approximately 99%) in SIL Area 4.

Prior to lake impoundment and CRD in 1976, the mean annual Lake Whitefish catch was 333,500 kg (735,242 lbs) (see Fish Community, Section 5.3.6), most of which (greater than 99%) was graded as light-coloured export quality fish that were only slightly parasitized by T. crassus (Figure 5.6.4-1). There were differences in RI between areas of SIL, with Lake Whitefish captured in SIL Area 5, a relatively shallow, inshore area of the lake populated primarily by dark variants of whitefish, typically more heavily infested than fish from other areas of the lake. Fish from SIL Area 5 were rarely graded as export quality and were avoided by the commercial fishery. Pre-1976, RI values from SIL Area 4 scientific studies and commercial catches consistently graded the Lake Whitefish as export quality. Following impoundment, commercial catches decreased to 115,000 kg (253,532 lbs) by 1982 largely due to a reduction in catch-per-unit-effort (CPUE). Reduced catch-per-unit-effort was attributed to a redistribution of fish from SIL Area 4 to other areas of SIL and to nearby waterbodies (Bodaly et al. 1984). Because some of these areas (e.g., SIL Area 5) traditionally had lower grade Lake Whitefish, there was an overall increase in the RI of T. crassus to more than 50 cysts/100 lbs dressed weight, downgrading the quality of the fishery from export to other (identified as continental at the time). This increased RI persisted until at least the late 1980s (Figure 5.6.4-1). To maintain the economic viability of the fishery during this period, Manitoba Hydro compensated for the difference in market value between continental and export grade Lake Whitefish to the commercial fishers. In recent years, cyst counts from experimental studies (Michaluk and Remnant 2012; Aiken and Remnant 2013; Aiken 2014) and commercial catches (FFMC unpubl. data) have decreased and the FFMC has since returned the fishery to export status, though the productivity of the fishery remains lower than historic values due to slow growth rates and other factors.



Recent FFMC grades for Lake Whitefish in other Area 3 lakes are presented in Table 5.6.4-1. Detailed historical data for many of these lakes are not available for comparison, though little change in RI is expected for those not directly affected by CRD unless there is a redistribution of fish from other areas that have different RI values. For those with pre-CRD infestation data, Karsakuwigamak, Mynarski, Notigi, and Opachuanau lakes have all been downgraded from export to other quality. All four of these lakes are located between the Notigi and Missi Falls CSs and are affected, to varying degrees, by CRD. Though no direct links have been identified, due to a general lack of scientific data from these lakes, it is possible that effects from flooding (changes in copepod or Northern Pike abundance or redistribution of Lake Whitefish) have altered the infestation rates. Lake Whitefish grades in Wapisu and Wuskwatim lakes have not changed.

Missi FallsControl Structure


R

OpachuanauLake

LemayIsland

Indian

SouthBay

Long Point

Lake

Southern

StrawberryIsland

SILArea 1

SILArea 4

SILArea 6

SILArea 0

SILArea 7

SILArea 5

SILArea 2

SILArea 3

493

1.0

22-OCT-15



05-DEC-15

Hudson Bay

Thompson

Churchill

Southern Indian Lake Historical Geographic Areas



0 5 10 Kilometers

0 4 8 Miles

DATA SOURCE:

DATE CREATED:

CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:

COORDINATE SYSTEM:

LegendSouthern Indian Lake (SIL) Historical Geographic AreaControl StructureHighwayFirst Nation Reserve

Created By: cparker - B Size Portrait BTB - MAR 2015 Scale: 1:423,640File Location: J:\MYP\MH_RCEA\Fish_Quality\PHASE_2\Mxd\20151205_RCEA_PhaseII_AE_FQ_EE_SouthernIndianLakeHistoricalGeographicAreas_cp.mxd

Map 5.6.4-2



Thick gray line represents current FFMC export/other grade cut-off (50 cysts/100 lbs dressed weight).

Figure 5.6.4-1: The Rate of Infestation of Triaenophorus crassus Cysts in Lake Whitefish from Commercial Fishery and Experimental (SIL Areas 4 and 5 only) Catches in Southern Indian Lake, 1952–2013




Waterbody Pre-CRD 1,2 1976–1990 3 2006–2013 4,5 2015 5

Apussigamasi Lake - - - Other

Baldock Lake - - Other Other

Birch Tree Lake - - - Other

Burntwood River - - Export Export

Issett Lake - - Export Export

Karsakuwigamak Lake Export - Other Other

Kinosaskaw Lake - - - Other

Leftrook Lake - - - Other

Macheewin Lake - - Other Other

Mynarski Lake Export - - Other

Mystery Lake - - - Other

Notigi Lake Export - Other Other

Opachuanau Lake Export - Other Other

Osik Lake - - Other Other

Pemichigamau Lake - - - Other

Rat Lake - - Export Export

Southern Indian Lake (all) Export Other Export Export

Southern Indian Lake (north basin) Export Other Export Other

Threepoint Lake - - Export Export

Wapisu Lake Other - Other Other

Wuskwatim Lake Export - Export Export

1. Export = 0–40 cysts/100 lbs of fish (DFO) or 0–50 cysts/100 lbs of fish (FFMC); Other (includes Continental and Cutter grades) = > 40 or > 50 cysts/100 lbs of fish (dressed weight).

2. Data sources: Bodaly et al. (1984); McTavish (1952); Schlick (1966b); Sunde (1963a, 1963b, 1965a). 3. Data sources: Bodaly et al. (1984); Peristy 1989. 4. Data sources: Michaluk and Remnant (2012); Aiken and Remnant (2013); Aiken (2014). 5. Data source: FFMC (unpubl. data).




PALATABILITY

There is no scientific evidence to suggest that the taste, texture or palatability has been affected by hydroelectric development in Area 3. However, the people who consume the fish in the communities have clearly identified this as an issue that is directly related to hydroelectric development. The issue of palatability remains unresolved.


As illustrated in the pathways of effects diagram (Figure 5.6.1-1) the infestation rate for T. crassus in the muscle of Lake Whitefish can be linked to hydroelectric development and there is evidence linking increased RI in SIL with the CRD. Specifically, the post-CRD redistribution of SIL fish from areas with low RI (e.g., SIL Area 4) to areas with higher RI (e.g., SIL Area 5, other nearby lakes) increased the overall RI in the commercial catch and resulted in the downgrading of the fishery. Watson and Dick (1979a) predicted that an increase in copepods would result in an increase in T. crassus in Lake Whitefish in SIL. However, there was a decrease in water temperature and an increase in turbidity associated with impoundment and diversion, both of which were implicated in reduced zooplankton (including cyclopoid copepods) abundance (Patalas and Salki 1984). Despite this decrease in abundance of copepods, opportunities for infestation may have increased due to the redistribution of many Lake Whitefish to shallower, inshore areas and possibly other waterbodies (Bodaly et al. 1984). Watson and Lawler (1965) demonstrated that the most heavily infected C. bicuspidatus in Heming Lake, Manitoba were collected from inshore areas, presumably near Northern Pike spawning habitat.

Although redistribution of fish appeared to be the primary linkage between increased cyst counts in the SIL commercial fishery in the 1980s, an increase in Northern Pike, which often occurs with flooding caused by hydroelectric development, may have also been a contributing factor. Increases in Northern Pike were noted for young year classes immediately following impoundment in SIL (Bodaly and Lesack 1984). Increases in Northern Pike in shallow, flooded areas would increase the availability of early T. crassus life cycle stages to copepods. There are no specific examples of this linkage in Area 3 waterbodies.

A small increase in zooplankton abundance may be expected in other waterbodies that experience flooding from hydroelectric development in the absence of excessive turbidity or decreases in water temperature. For example, it was predicted that an expected increase in retention time, volume and slight nutrient enrichment of water in tributary waterbodies of Wuskwatim Lake (e.g., Sesep Lake, Wuskwatim Lake south, Wuskwatim Brook) may result in a slight increase in zooplankton abundance (Manitoba Hydro and NCN 2003).

The return of SIL to a mainly export grade fishery following several years of lower grades could be due to several factors. One potential cause could be an increase in the proportion of Northern Pike captured in the commercial fishery as Lake Whitefish catch-per-unit-effort decreased post-impoundment. Peristy (1989) indicated that Northern Pike proportions in the commercial catch had increased relative to pre-



diversion, particularly in SIL Area 5, which consistently had the highest infestation rates in the lake. Most of these Northern Pike were also captured in the spring, before spawning, which could be reducing the number of mature T. crassus that are contributing to the parasite life cycle. Impacts of fishing pressures on RI have been observed in other fish communities; for example, Amundsen and Kristoffersen (1990) noted a dramatic reduction in T. crassus infestation rates in Powan (Coregonus lavaretus) following the experimental reduction in population density of Northern Pike in a Norwegian lake. In addition, if cyclopoid copepod abundance remained as low as observed shortly after impoundment (Patalas and Salki 1984), this would further limit the rate of transmission between hosts. Unfortunately, recent zooplankton data from SIL are not available for comparison. Commercial fishers may have also, over time, adapted to target only the areas within SIL and other waterbodies where RI remained low, avoiding the more heavily infested Lake Whitefish.



5.6.5 Area 4: Missi Falls Control Structure to Churchill River Estuary

Area 4 comprises the reach of the Churchill River from the Missi Falls CS at SIL to the Churchill River Estuary. Major waterbodies within this area include Northern Indian Lake, Fidler Lake, Billard Lake, and the lower Churchill River (Map 5.6.5-1). Hydroelectric developments in the area include CRD and the Lower Churchill River Water Level Enhancement Weir Project (Churchill Weir). The CRD, via the Missi Falls CS, reduced flows in downstream waterbodies in the Churchill River system, particularly during construction and staging periods. Construction of the Churchill Weir in 1998 increased upstream water levels, mitigating some of the effects of CRD. Detailed information regarding project descriptions for hydroelectric developments in Area 2 and their effects on the local water regime can be found in Hydroelectric Development Project Description in the Region of Interest, Part II Appendix 2F and Water Regime, Section 4.3.3.


PALATABILITY

There have been no palatability concerns raised in Area 4 waterbodies by local resource users and, therefore, no associated studies.


Methods used by the FFMC to grade commercial Lake Whitefish based on infestation rates is described in detail in Section 5.6.2.1. Non-commercial sampling for T. crassus infestations has not been conducted in Area 4.


PALATABILITY

There is no published information describing the palatability of fish in Area 4 waterbodies either before or after CRD or construction of the Churchill Weir.


There is historical information (primarily from the 1960s) on Lake Whitefish infestation rates with T. crassus cysts from the Churchill River, Fidler Lake, Gauer Lake and, in particular, Northern Indian Lake (Sunde 1965a; Anthony 1966).

Churchill River Weir

FidlerLake Billard

Lake

PartridgeBreast Lake

Churchill

River

HUDSONBAY

NorthernIndian Lake

GauerLake


280

Churchill

1.0

06-OCT-15


07-DEC-15

Province of Manitoba, Government of Canada, Manitoba Hydro.

Fish Quality Assessment AreaMissi Falls Control Structure to

Churchill River Estuary RCEA Area 4

LegendRCEA Region of InterestFirst Nation Reserve

Hudson Bay

Thompson

Churchill


DATA SOURCE:


CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:


0 10 20 Miles

0 10 20 Kilometres

Cre

ated

By:

cpa

rker

- B

Siz

e La

ndsc

ape

STD

- M

AR

201

5

Sca

le: 1

:1,2

47,0

00

RCEAArea 4

InfrastructureControl StructureTransmission Line (Existing)RailHighway

Map 5.6.5-1



5.6.5.3 New Information and/or Re-analysis of Existing Information The FFMC routinely samples Lake Whitefish from Manitoba lakes with commercial fisheries and counts cysts from dressed fish in their laboratory in Winnipeg. Fish are then graded based on their RI. Lake Whitefish grading results (unpubl. data from 2011 and 2015) were obtained from the FFMC for the Churchill River, Fidler Lake, Gauer Lake, Northern Indian Lake, Billard Lake, and Partridge Breast Lake for comparison with earlier data, where possible (FFMC unpubl. data).


PALATABILITY

There has been no known information provided by resource users regarding decreased taste, texture, and palatability specifically for Area 4. There is no published information on changes to palatability and there is no scientific evidence to suggest that the taste, texture or palatability of fish has been affected by hydroelectric development.


The majority of historical RI data were from Northern Indian Lake, though Fidler Lake, Gauer Lake, and the Churchill River were also surveyed (Sunde 1965a; Anthony 1966). More recently, data have been collected from at least six Area 4 waterbodies (FFMC unpubl. data; Table 5.6.5-1). Of all the sampled waterbodies, only Partridge Breast Lake is currently graded as export quality, but there are no historical data for comparison (FFMC unpubl. data). The Churchill River and Fidler and Northern Indian lakes were graded as export quality during surveys in the 1960s, but have recently been downgraded to continental or cutter (i.e., other) by FFMC. There is insufficient information to confirm any potential linkages to hydroelectric development in Area 4. Gauer Lake has had consistently high cyst counts and has always been graded as “other” (specifically cutter).




Waterbody Pre-1980 1,2 1980–1990 2006–2011 3 2015 3

Billard Lake - - - Other

Churchill River Export - - Other

Fidler Lake Export - Other Other

Gauer Lake Other - Other Other

Northern Indian Lake Export - - Other

Partridge Breast Lake - - - Export

1. Export = 0–50 cysts/100 lbs of fish (dressed weight); Other = > 50 cysts/100 lbs of fish (dressed weight). 2. Data from Sunde (1965a); Anthony (1966); and Topolinski (1972). 3. Data from FFMC (unpubl.).


PALATABILITY

As noted above, there has been no known information provided by resource users regarding decreased taste, texture, and palatability specifically for Area 4. There is no published information on changes in palatability and there is no scientific evidence to suggest that the taste, texture or palatability of fish has been affected by hydroelectric development.


As illustrated in the pathways of effects diagram (Figure 5.6.1-1) the infestation rate for T. crassus in the muscle of Lake Whitefish can be linked to hydroelectric development. Some of these links have been established for waterbodies in other areas (e.g., SIL in Area 3; Section 5.6.4). Long-term data exist for Northern Indian Lake, which was consistently an export quality fishery until recently when it was downgraded to continental (i.e., other). However, there are insufficient T. crassus data from other Area 4 waterbodies to identify trends. There is currently no definitive link between increased RI in Lake Whitefish and hydroelectric development for any Area 4 waterbodies.

.



5.6.6 Summary of Effects of Hydroelectric Development in the Region of Interest on Fish Quality

PALATABILITY

There are no known scientific linkages or pathways of effect between palatability and hydroelectric development in the RCEA ROI. However, resource users in local communities continue to express concerns about a decline in the quality of fish (in terms of taste, texture, and size) in waterbodies affected by hydroelectric development, preferring the taste of fish from unaffected areas.

However, it should be noted that hydroelectric development can cause changes to fish diet, water quality, water temperature, algae, and growth rates, all of which can, in turn, affect the taste and texture of fish. These potential indirect linkages have not been subject to scientific studies in the RCEA ROI.


It should be noted that there are lakes affected by hydroelectric development to varying degrees (e.g., Playgreen Lake, Split Lake, Wuskwatim Lake, to name a few) that have maintained a consistent grade of commercial Lake Whitefish catches since before hydroelectric development, indicating that the linkages may be weak in some waterbodies and that effects do not always occur. If a particular hydroelectric development has only negligible effects to each of the three T. crassus life cycle hosts, then effects on RI in Lake Whitefish should be similarly minimal.

Although some waterbodies appear to have remained unaffected, increased RI in Lake Whitefish has been observed in several other RCEA ROI waterbodies affected by hydroelectric development with most of the detailed data coming from studies conducted in SIL following CRD. The specific pathway(s) of effect are not precisely known for most of the affected waterbodies and likely vary among them. For example, redistribution of Lake Whitefish populations in SIL is thought to be at least one of the factors that resulted in a decreased grade of the commercial catch in that lake following CRD. Changes to copepod and/or Northern Pike host abundances and distributions that may be associated with flooding or erosion during hydroelectric development may have also affected infestation rates in other waterbodies.



5.6.7 Bibliography

5.6.7.1 Literature Cited and Data Sources Aiken, J.K. 2014. Results of fish community investigations conducted in Southern Indian Lake, Manitoba,

2013. A report prepared for the Southern Indian Lake Commercial Fishermen’s Association by North/South Consultants Inc., Winnipeg, MB. 84 pp.

Aiken, J.K. and Remnant, R.A. 2013. Results of fish community investigations conducted in Southern Indian Lake, Manitoba, 2012. A report prepared for the Southern Indian Lake Commercial Fishermen’s Association by North/South Consultants Inc., Winnipeg, MB. 128 pp.

Amundsen, P.A. and Kristoffersen, R. 1990. Infection of whitefish (Coregonus lavaretus L. s.l.) by Triaenophorus crassus Forel (Cestoda: Pseudophyllidea): a case study in parasite control. Canadian Journal of Zoology 68: 1187-1192.

Anthony, L. 1966. Northern Indian Lake Whitefish Infestation Test. Manitoba Department of Mines and Natural Resources, Fisheries Branch. 4 pp.

Bodaly, R.A. and Lesack, L.F.W. 1984. Response of a boreal northern pike (Esox lucius) population to lake impoundment: Wupaw Bay, Southern Indian Lake, Manitoba. Canadian Journal of Fisheries and Aquatic Sciences 41: 706-714.

Bodaly, R.A., Johnson, T.W.D., Fudge, R.J.P., and Clayton, J.W. 1984. Collapse of the lake whitefish (Coregonus clupeaformis) fishery in Southern Indian Lake, Manitoba, following lake irnpoundment and river diversion. Canadian Journal of Fisheries and Aquatic Sciences 41: 692-700.

Davies, S., Baker, R., and Horne, B. 1998a. The quality of Lake Whitefish, Walleye, and Northern Pike from Playgreen Lake, Little Playgreen Lake, and Mossy Bay (Lake Winnipeg). A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 44 pp.

Davies, S., Baker, R., and Horne, B. 1998b. Fish movements, species composition, and catch data for Playgreen Lake, Little Playgreen Lake, and Mossy Bay (Lake Winnipeg). A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 36 pp.

Izaguirre, G., Taylor, W.D. and Pasek, J. 1999. Off-flavor problems in two reservoirs, associated with planktonic Pseudanabaena species. Water Science and Technology 40: 85-90.

Johnston, C.H. 1984. The genetic and environmental basis for external colouration in lake whitefish (Coregonus clupeaformis (Mitchill)) from Southern Indian Lake, Manitoba. M. Sc. thesis, Faculty of Graduate Studies, The University of British Columbia, Vancouver, BC. 132 pp.

Keeyask Hydropower Limited Partnership. 2012. Keeyask Generation Project Environmental Impact Statement: Aquatic Environment Supporting Volume, Winnipeg, Manitoba. June 2012.



Lawler, G.H. 1955. Life history studies of Triaenophorus at Heming Lake, Manitoba. Part I. The life of T. crassus and T. nodulosus in the final host, Esox lucius with special reference to changes in infection rates from year to year. Fisheries Research Board of Canada, MS Report 604. 17 pp.

Lawler, G.H. and Scott, W.B. 1954. Notes on the geographical distribution and the hosts of the cestode genus Triaenophorus in North America. Journal of the Fisheries Research Board of Canada 11: 884-893.

Lorio, W.J., Perschbacher, P.W., and Johnsen, P.B. 1992. Relationship between water quality, phytoplankton community and off-flavors in channel catfish (Ictalurus punctatus) production ponds. Aquaculture 106: 285-292.

Mackay, G.H., Davies, S., and Westdal, H. 1990. Post project assessment of Kelsey and Lake Winnipeg Regulation impacts on Wabowden. The Wabowden Study Team, Winnipeg, MB. 98 pp.

Manitoba Hydro and NCN (Nisichawayasihk Cree Nation). 2003. Wuskwatim Generation Project Environmental Impact Statement. Volume 5 Aquatic Environment.

McTavish, W.B. 1952. A biological survey of Southern Indian Lake. Report for the Manitoba Department of Mines and Natural Resources.

Michaluk, Y. and Remnant, R.A. 2012. Results of fish community investigations conducted in Southern Indian Lake, Manitoba, 2011. A report prepared for the Southern Indian Lake Commercial Fishermen’s Association by North/South Consultants Inc., Winnipeg, MB. 105 pp.

Miller, R. B. 1952. A review of the Triaenophorus problem in Canadian lakes. Bulletin of the Fisheries Research Board of Canada No. 95.

Moshenko, W.R. 1968. Whitefish infestation in Moose Nose Lake, Manitoba, in 1968. Department of Mines and Natural Resources, Fisheries Branch. 3 pp.

NRG (Nelson River Group). 1986. Cross Lake environmental impact assessment study: Volume 3: environmental impact statement. Nelson River Group, Winnipeg, MB. 54 pp.

Patalas, K. and Salki, A. 1984. Effects of impoundment and diversion on the crustacean plankton of Southern Indian Lake. Canadian Journal of Fisheries and Aquatic Sciences 41: 613-637.

Peristy, D. 1989. A biological assessment of the post-impoundment commercial fishery at Southern Indian Lake, Manitoba, with historical comparisons. M.N.R.M. practicum, Natural Resources Institute, The University of Manitoba, Winnipeg, MB. 141 pp.

Persson, P.-E. 1980. Sensory properties and analysis of two muddy odour compounds, geosmin and 2-methylisoborneol, in water and fish. Water Research 14: 1113-1118.

Pulkkinen, K., and Valtonen, E.T. 1999. Accumulation of plerocercoids of Triaenophorus crassus in the second intermediate host Coregonus lavaretus and their effect on growth of the host. Journal of Fish Biology 55: 115-126.



Ryland, D., and Watts, B. 2002a. Fish taste studies for York Factory First Nation. A report prepared for Manitoba Hydro by the Department of Human Nutritional Sciences, University of Manitoba, Winnipeg, MB. 31 pp.

Ryland, D. and Watts, B. 2002b. Fish taste studies for Nisichawayasihk Cree Nation (NCN). Final Report (Reference File /70.01.412.2), Department of Human Nutritional Sciences, University of Manitoba, 28 pp.

Ryland, D., and Watts, B. 2004a. Fish taste studies for Tataskweyak Cree Nation. A report prepared for Manitoba Hydro by the Department of Human Nutritional Sciences, University of Manitoba, Winnipeg, MB. 39 pp.

Ryland, D., and Watts, B. 2004b. Fish taste studies for Fox Lake Cree Nation. A report prepared for Manitoba Hydro by the Department of Human Nutritional Sciences, University of Manitoba, Winnipeg, MB. 37 pp.

Schlick, R.O. 1966a. Whitefish infestation test, Sipiwesk Lake, 1966. Manitoba Department of Mines and Natural Resources, Fisheries Branch. 5 pp.

Schlick, R.O. 1966b. Whitefish infestation test, Wuskwatim Lake, August 1966. Manitoba Department of Mines and Natural Resources, Fisheries Branch. 6 pp.

Schlick, R.O. 1967. Whitefish infestation Split Lake, 1966. Manitoba Department of Mines and Natural Resources, Fisheries Branch. 3 pp.

Schlick, R.O. 1979. A fisheries survey of Waskaiowaka Lake, 1972. MS Report No. 79-61, Manitoba Department of Mines, Natural Resources and Environment, Research Branch. 29 pp.

Schrader, K.K. and Blevins, W.T. 1993. Geosmin-producing species of Streptomyces and Lyngbya from aquaculture ponds. Canadian Journal of Microbiology 39: 834-840.

Sowe, M.S. 1986. The coregonid and pike fishery in Manitoba: factors influencing abundance of Triaenophorus crassus Forel in Lake Whitefish (Coregonus clupeaformis Mitchill) in commercially fished lakes. . M. Sc. thesis, Department of Zoology, The University of Manitoba, Winnipeg, MB. 243 pp.

Split Lake Cree – Manitoba Hydro Joint Study Group. 1996a. Analysis of change: Split Lake Cree Post Project Environmental Review. Split Lake Cree – Manitoba Hydro Joint Study Group; vol. 1 of 5.

Split Lake Cree – Manitoba Hydro Joint Study Group. 1996b. History and first order effects: Manitoba Hydro projects - Split Lake Cree Post Project Environmental Review. Split Lake Cree – Manitoba Hydro Joint Study Group; vol. 2 of 5.

Split Lake Cree – Manitoba Hydro Joint Study Group. 1996c. Environmental matrices: Summary of Manitoba Hydro impacts - Split Lake Cree Post Project Environmental Review. Support from William Kennedy Consultants Ltd. & InterGroup Consultants Ltd. Split Lake Cree - Manitoba Hydro Joint Study Group; vol. 3 of 5.



Sunde, L.A. 1963a. South Indian Lake whitefish infestation test, March 5 to 7, 1963. Manitoba Department of Mines and Natural Resources, Fisheries Branch MS Report.

Sunde, L.A. 1963b. Summary of whitefish infestation tests conducted in Northern Manitoba in 1963. Manitoba Department of Mines and Natural Resources, Fisheries Branch. 2 pp.

Sunde, L.A. 1964. Walker Lake whitefish infestation test. Manitoba Department of Mines and Natural Resources, Fisheries Branch MS Report. 12 pp.

Sunde, L.A. 1965a. Summary of department of fisheries whitefish infestation tests for northern Manitoba August 1962 to February 1965. Manitoba Department of Mines and Natural Resources, Fisheries Branch. 4 pp.

Sunde, L.A. 1965b. Summary of whitefish infestation tests conducted in northern Manitoba in 1964. Manitoba Department of Mines and Natural Resources, Fisheries Branch. 2 pp.

Topolinski, D. 1972. The Ilford fisherman’s cooperative: A case study in the commercial fisheries of northern Manitoba. M.N.R.M. practicum, Natural Resource Institute, University of Manitoba, Winnipeg, MB. 67 pp.

Watson, N.H.F. and Lawler, G.H. 1965. Natural infections of cyclopid copepods with procercoids of Triaenophorus spp. Journal of the Fisheries Research Board of Canada 22: 1335-1343.

Watson R.A. and Dick. T.A. 1979a. Metazoan parasites of whitefish Coregonus clupeaformis (Mitchill) and cisco C. artedi Lesueur from Southern Indian Lake, Manitoba. Journal of Fish Biology 15: 579-587.

Watson R.A. and Dick. T.A. 1979b. Metazoan parasites of pike, Esox lucius Linnaeus, from Southern Indian Lake, Manitoba, Canada. Journal of Fish Biology 17: 255-261.

Yurkowski, M. and Tabachek, J.-A.L. 1980. Geosmin and 2-methylisoborneol implicated as a cause of muddy odor and flavor in commercial fish from Cedar Lake, Manitoba. Canadian Journal of Fisheries and Aquatic Sciences 37: 1449-1450.

5.6.7.2 Personal Communications Davies, Stuart. 2015. Senior Biologist. North/South Consultants Inc. Email correspondence with Michael

Johnson, North/South Consultants Inc., Winnipeg MB, October 20, 2015.

REGIONAL CUMULATIVE EFFECTS ASSESSMENT – PHASE II WATER – SEALS

DECEMBER 2015 5.7-1

5.7 Seals

5.7.1 Introduction Both harbour (Phoca vitulina) and bearded (Erignathus barbatus) seals are year-round residents of Hudson Bay, with patchy distributions in circumpolar areas as far north as 85°N (Burns 1981; Sergeant 1986). The ringed seal is the most common seal species in Hudson Bay (Smith 1975; Sergeant 1986); however, this species is heavily dependent on sea ice year-round and is uncommon in both the Nelson and Churchill River areas. Consequently, ringed seals were not considered further. While harbour seals are more common to the Churchill River and estuary (Remnant 1997; Bernhardt 1999, 2000, 2001, 2003, 2006), bearded seals predominate in the Nelson River area (Baker 1989, 1990; Bernhardt 2014).

The global bearded seal population is estimated to be more than 500,000 animals (Riedman 1990) and is listed as “Least Concern” on the International Union for Conservation of Nature (IUCN) and Natural Resources Red List of Threatened Species (IUCN 2014). Bearded seals occur in low densities throughout the Canadian Arctic (Mansfield 1967; Burns 1981; Cleator 1996; Lunn et al. 1997) and while discrete populations have not been delineated (Cleator 1996; Kovacs 2002), geographical variation in vocalizations suggests some population substructure (Risch et al. 2007). The number of bearded seals in Canadian waters is unknown (Cleator 1996; Kovacs 2002); however, Cleator (1996) estimated a minimum of 190,000 animals. The population is considered stable, as hunters in the Arctic have not reported declines in bearded seal numbers while maintaining consistent hunting levels. Canadian populations are not listed under the Species at Risk Act and are listed as “Data Deficient” by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC) (2015). Small numbers of bearded seals occupy the Churchill and Nelson River mouths during the open-water season (e.g., Remnant 1997; Bernhardt 1999, 2000, 2001, 2006).

Global estimates for harbour seals vary. Fisheries and Oceans Canada (DFO) (2015) estimated the global population to be 5–6 million individuals, while a previous study conducted by IUCN (2008) estimated the total number to be closer to 350,000–500,000. Harbour seals are not listed under the Convention on International Trade in Endangered Species of Wild Fauna and Flora, and the IUCN has designated the species as “Least Concern” due to its large and either stable or increasing global population. Despite this stability, IUCN (2008) recommended that local populations required further assessment due to small and/or declining numbers in some areas. Estimates of harbour seal numbers in Canadian waters are not available (e.g., Baird 2001; Remnant 1997). Harbour seals prefer ice-free areas (Burns 2002) in shallower waters less than 50 m in depth (Tollit et al. 1997; Lesage et al. 2004; Bajzak et al. 2012). This species commonly occurs in the nearshore freshwater areas of the Churchill River estuary (Remnant 1997; Bernhardt 1999, 2000).

5.7.1.1 Pathways of Effect Due to the small amount of data relating to seals in the Churchill and Nelson River areas, pathways of effect were developed using information from both the study area and adjacent regions. Habitat


DECEMBER 2015 5.7-2

alterations may affect seals directly in the form of physiological and behavioural changes, or indirectly through ecological processes (e.g., trophic linkages; Figure 5.7.1-1). Water temperature, salinity, depth, flow and sedimentation rates are influenced by discharge and when combined with natural processes (e.g., tidal cycle), can alter the suitability of estuarine habitats (Lawrence et al. 1992). For example, increased river flow during the flood tide can decrease the availability of haul-out sites used by seals.

Hauling out is the most commonly observed seal behaviour in the Churchill and Nelson River areas; however, the purpose of this behaviour is poorly understood. Seals may haul out for a number of reasons, including: pupping (Mansfield 1967); skin cell maintenance (Feltz and Fay 1966), sleep (Schneider et al. 1980; Remnant 1997), predator avoidance (Terhune 1985), mate selection (Renouf and Lawson 1986), thermoregulatory advantage (Watts 1992; Remnant 1997; Bernhardt 2006); and social interaction (Remnant 1997). Consequently, alterations in the quality or quantity of haul-out areas may have implications to seals.

Indirect impacts of hydroelectric development focus primarily on trophic linkages. Bearded and harbour seals are largely opportunistic feeders, ingesting a variety of (mostly benthic) fish and invertebrates (Antonelis et al. 1994; Bjørge et al. 1995; Tollit et al. 1997; Lesage et al. 1999; Gjertz et al. 2001; Hastings et al. 2004; Bajzak et al. 2012). Evidence suggests that estuaries may be important feeding areas for seals (e.g., Lowry et al. 1980; Finley and Evans 1983; Antonelis et al. 1994; Remnant 1997; Hjelset et al. 1999; Dehn et al. 2007); however, harbour seals near Churchill appear to depend less on these areas for foraging (Remnant 1997; Bajzak et al. 2012). Although mercury-related studies linked to dietary preferences have been conducted for seals in Hudson Bay (e.g., Young et al. 2009), research directly pertaining to seals in the Churchill and Nelson rivers were unavailable.

Noise and other types of disturbance may cause a variety of behavioural responses in seals, including: increased vigilance; changes in foraging behaviour; increased energy demand; avoidance; changes in habitat use (including haul-out behaviour); decreased reproductive success (e.g., changes in lair attendance, nursing bouts, reproductive output); and individual survival (Allen et al. 1984; Kovacs and Innes 1990; Lidgard 1996; Barton et al. 1998; Shaughnessy et al. 1999; Henry and Hammill 2001; Edrén et al. 2004; Kucey 2005; Orsini et al. 2006). Responses appear to vary with species, reproductive state, and distance or type of noise produced. In most cases, the closer the disturbance, the more likely seals are to enter the water (e.g., Allen et al. 1984; Brown and Prior 1998; Suryan and Harvey 1999; Henry and Hammill 2001; Johnson and Acevedo 2007; Fox 2008; Jansen et al. 2010). Henry and Hamill (2001) found that reproductive and moulting states influenced behaviour when exposed to disturbance, with seals less likely to enter the water during moulting, and more readily entering the water during the pupping season.


DECEMBER 2015 5.7-3

Arrow width is not an indication of magnitude.

Figure 5.7.1-1: Linkage Diagram Showing Potential Pathways of Effect of Hydroelectric Development and Other Factors on Seals


DECEMBER 2015 5.7-4

Other drivers, both historic and current, not related to hydroelectric development have the potential to influence seal use of the Churchill and Nelson River estuaries. Key drivers include (Figure 5.7.1-1): • harvesting;

• Port of Churchill (construction and operation); • shipping; • military activities (e.g., air traffic, training exercises); and • research (e.g., rocket launching facilities out of Churchill).

The existence of multiple pathways makes distinguishing hydroelectric-related effects from other drivers a difficult task. For example, the Port of Churchill, completed in 1931, may have affected haul-out locations along the eastern shoreline of the Churchill River estuary. Ongoing maintenance activities also may affect seal use of the estuary.

Due to the lack of physiological and ecological information for seals within the RCEA Region of Interest (ROI), determining the magnitude of potential impacts from hydro-related and other anthropogenic activities is not possible.

5.7.1.2 Indicators and Metrics Indicators and metrics used to assess potential effects to seals were constrained by the amount of existing data specific to the study areas. For the purposes of the Phase II report, species abundance and distribution were combined into one indicator used to describe seal communities and to discuss potential impacts of hydroelectric development.

5.7.1.3 Approach and Methods The approach used to evaluate the potential effects of hydroelectric development on seals involved collecting and presenting all available information within the given area for pre- and post-hydroelectric development time periods.

5.7.1.4 Data Limitations The following describes the data limitations for seals in areas 2 and 4 that affect the ability to conduct a more comprehensive review. These limitations include:

• A lack of quantitative pre- and post-hydroelectric development data, which precludes a determination of hydroelectric development effects on seal populations in the RCEA ROI.

• Information on behaviour and habitat use in the context of climate change is limited. Long-term data sets examining the influence of climate change on seals, particularly as it relates to hydroelectric development, are lacking.


DECEMBER 2015 5.7-5

5.7.2 Area 2: Kelsey Generating Station to the Nelson River Estuary

For the purposes of this section, the Nelson River estuary is defined as the extent of the Nelson River between Gillam Island (the upstream extent of tidal influence) up to, and including, the freshwater-saltwater mixing zone (or “mixing zone”), the area where freshwater from the Nelson and Hayes rivers mixes with marine waters of Hudson Bay. The location and size of the mixing zone is dependent on several factors (Nelson River flow, tide state, wind and wave action, oceanographic currents); however, at its maximum, the zone begins approximately 5 km downstream of Port Nelson and may extend more than 20 km into Hudson Bay (Map 5.7.2-1).

In this assessment, 1976 (operation of the Lake Winnipeg Regulation [LWR] and Churchill River Diversion [CRD]) was selected as the beginning of effects to seals as a result of hydroelectric development due to the potential for alterations in shoreline habitat and effects to seal use of the river and estuary.

5.7.2.1 Key Published Information Information on seal use of the lower Nelson River and estuary prior to hydroelectric development is limited to opportunistic observations from the 1970s (Didiuk 1975). Post-hydroelectric development studies include multi-year pre- and post-Limestone Generating Station (GS) environmental monitoring programs conducted by Manitoba Hydro (Baker 1989, 1990; North/South Consultants Inc. 2012) and studies conducted in support of preparation of an Environmental Impact Statement (EIS) for the potential Conawapa Generation Project (GP) (Ambrose and Berger 2012a, b; Bernhardt 2014). Additional information included broader investigations of Hudson Bay populations (e.g., Lunn et al. 1997; Bajzak et al. 2012), and research focused on seals in the context of climate change (e.g., Stewart and Lockhart 2005).

5.7.2.2 New Information and/or Re-analysis of Existing Information No additional analyses were conducted for Area 2 seals.

5.7.2.3 Changes in Seals over Time The following discussion was based on qualitative and quantitative data presented in key published literature.

Port Nelson

York Factory

GillamIsland

Hayes River

River

Nelson

HUDSONBAY

Rupert CreekFi

le L

ocat

ion:

J:\M

YP

\MH

_RC

EA

\Mar

ine_

Mam

mal

s\P

HA

SE

_2\M

XD

s\20

1512

01_R

CE

A_P

hase

II_A

E_M

arin

eMam

mal

s_N

elso

nHay

esR

iver

sFre

shw

ater

Sal

twat

erM

ixin

gZon

e_cp

.mxd

1.0

22-OCT-15


Cre

ated

By:

cpa

rker

- B

Siz

e La

ndsc

ape

STD

- M

AR

201

5

Sca

le: 1

:343

,074

01-DEC-15

Province of Manitoba, Government of Canada, North/South Consultants, RSW-Environnement Illimité (2013), Manitoba Hydro.

Nelson/Hayes River Freshwater-Saltwater Mixing Zone

LegendMinimum extent of the mixing zone

Maximum extent of the mixing zone

Hudson Bay

Thompson

Churchill


DATA SOURCE:


CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:


0 3 6 Miles

0 4.5 9 Kilometres

Map 5.7.2-1


DECEMBER 2015 5.7-7

5.7.2.3.1 Pre-Hydroelectric Development

Historical information related to seals in the Nelson River and estuary is limited to anecdotal observations from the early 1900s noting the presence of seals in the estuary (Comeau 1915), and opportunistic observations from a 1974/75 Provincial wildlife assessment (Didiuk 1975). Didiuk (1975) observed bearded, harbour and ringed seals in the area spanning Kettle GS to the estuary. Although one seal (species unknown) was reported on Seal Island (Map 5.7.2-2), no further information was provided.

Domestic seal harvests were relatively infrequent in the York region during the 1600s and 1700s (Reeves and Mitchell 1989). Seal meat was primarily used as dog food (Fast and Saunders 1996) and oil was sold to the Hudson’s Bay Company (Lytwyn 2002).

5.7.2.3.2 Post-Hydroelectric Development

In 1988, Manitoba Hydro initiated a multidisciplinary study of the Nelson River estuary to collect baseline data prior to completion of the Limestone GS. As part of that study, Baker (1989, 1990) found that bearded seals were often the only seal species present in the estuary and were most abundant during late summer and fall. In that study, bearded seals occurred in groups of up to six individuals as far upstream as Ship Island.

Bernhardt (2014) also found that, with the exception of a few harbour seals, all seals observed in the Nelson River estuary between 2003 and 2009 were bearded seals. Within the lower Nelson River, seals concentrated their activity in the immediate vicinities of Gillam Island and Port Nelson, hauling out on the shores of islands, small reefs and shoals, and on exposed tidal flats around the piers of Port Nelson. Time of year was an important determinant of seal haul-out activity during those years (Figure 5.7.2-1), while other factors such as time of day, weather conditions (e.g., air temperature, wind speed), river discharge, and tide state also played a role. No direct linkages were found between tide state and seal numbers between Gillam Island and Port Nelson; however, tide state appeared to influence the number of seals hauled out at Port Nelson. Few seals occurred upstream of Deer Island and only one seal was observed near the potential Conawapa GP site (Map 5.7.2-2; Bernhardt 2014).

Seals are occasionally reported along the entire reach of the lower Nelson River up to the Limestone GS tailrace (Bernhardt 2014). Group size in haul-out areas was larger in 2006 and 2007 than reported previously (i.e., Baker 1989, 1990); however, considerably fewer seals were observed during the 2009 surveys conducted during similar time periods (Bernhardt 2014). Although lower seal numbers in 2009 may have been associated with higher Nelson River flow, Bernhardt (2014) concluded that factors other than flow likely contributed as higher seal numbers were observed under similar conditions in 2007. Based on the small number of seals observed upstream of Gillam Island, and given published dietary information (e.g., Smith 1981), Bernhardt (2014) suggested that most foraging likely occurs in the estuary and coastal offshore areas rather than in the river itself.

Seals are rarely harvested in the Nelson River area due to the reduced numbers of people in the York Factory area (Lower Nelson Overview Study Team 1981) and the decline of commercial sealing in Hudson Bay (Stewart and Lockhart 2005). A resource use study conducted in 1995 reported that hunters


DECEMBER 2015 5.7-8

accessing the Nelson River by way of the Conawapa Access Road harvested two seals (MacDonell 1996). However, use of the access road for the purposes of seal harvesting is rare (MacDonell 1996).

Determining the impacts of hydroelectric development on seals in the Nelson River area is not possible without the appropriate historical data. However, Bernhardt (2014) reasoned that although meteorological factors may affect seal haul-out activity over short time periods, hydrological factors may have longer-term impacts. Bernhardt’s (2014) results suggest that Nelson River discharge, in addition to date, air temperature, and wind speed, may affect the number of seals in the lower-most portion of the Nelson River (i.e., between Gillam Island and Port Nelson). Increased Nelson River discharge and subsequent changes to water level may negatively affect seal use of this reach by altering the availability of suitable haul-out locations.

Determination of cumulative impacts related to hydroelectric development are further complicated by the lack of long-term climate change data associated with seal behaviour and habitat use both within, and outside of, the RCEA ROI.

Source: Bernhardt 2014.

Figure 5.7.2-1: Seasonal Differences in the Number of Seals Observed during Systematic Surveys in 2006 and 2007 between the Upstream End of Gillam Island to Port Nelson in Relation to Tide State, Compared with Opportunistic Observations Conducted at Low Tide in 2009

Distribution of Seal Haul-out Sites

Nelson and Hayes Rivers 2006 and 2007

Government of Manitoba; Government of Canada; Manitoba HydroNorth/South Consultants


3.0

DD-MMM-YY

LegendSeal Count

1 - 3

4 - 9

10 - 19

20 - 35

Generating Station (Existing)


Highway

Rail01-DEC-15

290

Seal Creek Creek

DeerIsland

Hayes River

Shilling Creek

Ten

Anglin

g

Nelson River

River

Weir River

SpenceLake

River

CooperCreek

Frenc

h

Hudson Bay

North

Roblin

Limestone G.S.

ConawapaG.S.

Port Nelson

York Factory



0 3 6 Miles

0 3.5 7 Kilometres

File

Loc

atio

n: J

:\MY

P\M

H_R

CE

A\M

arin

e_M

amm

als\

PH

AS

E_2

\MX

Ds\

2015

1201

_RC

EA

_Pha

seII_

AE

_Mar

ineM

amm

als_

Dis

tribu

tionO

fSea

lHau

lout

Site

sNel

sonA

ndH

ayes

Riv

ers2

006a

nd20

07_c

p.m

xd

DATA SOURCE:

DATE CREATED:

CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:

COORDINATE SYSTEM:

Hudson Bay

Thompson

ChurchillC

reat

ed B

y: c

park

er -

B S

ize

Land

scap

e B

TB -

MA

R 2

015

S

cale

: 1:3

12,9

96

Gillam Island

Seal Island

0 0.5 1 Kilometre

0 0.5 1 Mile

ShipIsland

0 250 500 Metres

0 1,000 2,000 Feet

Map 5.7.2-2



5.7.2.4 Cumulative Effects of Hydroelectric Development on Seals in Area 2

5.7.2.4.1 Summary • Bearded seals are the predominant seal species in the Nelson River and estuary. The impacts of past

hydroelectric development to seals in this region cannot be determined due to lack of both current and historical data. Based on behaviour and habitat use, hydroelectric impacts to seals may have been related to displacement of haul-out sites in the lower portion of the river in areas of increased discharge above the zone of tidal influence.

• Although group size in haul-out areas appeared to decrease during periods of higher Nelson River flow (e.g., 2009), factors other than flow (time of year/day, weather conditions, tide state) also affect seal use of the river and estuary.



5.7.3 Area 4: Missi Falls Control Structure to the Churchill River Estuary

For the purposes of this section, the Churchill River estuary is defined as the area from Mosquito Point to the mouth of the Churchill River between Cape Merry and Fort Prince of Wales (Map 5.7.3-1). As in Area 2, 1976 (operation of LWR and CRD) was selected as the beginning of effects to seals in this region as a result of hydroelectric development.

Activities (both past and present) not related to hydroelectric development also may influence seal use of the Churchill River and estuary. In addition to harvesting, these include military and research activities, construction and operation of the Port of Churchill (including dredging), and associated shipping activities. In 1942, the United States Air Force began construction of a military base approximately 8 km southeast of the Town of Churchill (Brandson 2011). The base was completed in 1945, and became the Joint Services Experimental Station in 1946, facilitating operations of the Royal Canadian Air Force, United States Air Force, Defence Research Northern Laboratory, Department of Transport, and Royal Canadian Navy. In 1964, the United States forces discontinued their presence at Fort Churchill and control of the base by the Canadian military ended. However, operation of the base continued under the Department of Public Works and Canadian military exercises (including aerial, boat- and land-based activities) continued until 1980 (Brandson 2011).

In 1957, a rocket launching facility was established in Churchill by the United States Special Committee for the International Geophysical year of 1957/58 (Brandson 2011). The Fort Churchill Rocket Facility (later re-named the Churchill Research Range) utilized a payload recovery area that spanned approximately 700,000 km2 of land and water (Brandson 2011). The last rocket was launched at the facility in 1988 (The Manitoba Historical Society 2015a).

Construction for the Port of Churchill began in the 1920s and was completed in 1931. The port maintains a 140,000-tonne elevator and four berths for the loading and unloading of grain, general cargo, and tanker vessels (The Manitoba Historical Society 2015b). Maintenance activities for port infrastructure (e.g., stabilization) and accessibility have been ongoing since the 1930s (Pratte 1976; Idle 1989), although dredging activities were last performed in the mid- to late 1990s (Omnitrax 2001a, cited in Stewart and Howland 2009). Shipping to and from the port continues throughout the open-water season.

The degree to which anthropogenic activities other than hydroelectric development may have influenced seals in the Churchill River area is unknown as environmental assessments were not available.

5.7.3.1 Key Published Information Historical information pertaining to seal use of the lower Churchill River and estuary is limited to a 1954 study comparing seal diets throughout the eastern Canadian Arctic (McLaren 1958a), and estimates of bearded seal populations in the Canadian Arctic through observations and harvest data (McLaren 1958b).



Information specific to the study area after 1976 included:

• surveys conducted by Manitoba Hydro in support of EIS preparation for the Lower Churchill River Water Level Enhancement Weir Project (Remnant 1997; Manitoba Hydro and Town of Churchill 1997);

• post-weir environmental monitoring (Bernhardt 1999, 2000, 2001, 2003, 2006; Bernhardt and Holm 2007), and

• current information regarding harbour seal abundance and distribution in the Churchill River estuary in 2013 and 2014 (Florko 2015).

Additional resources include status reports (e.g., Baird 2001; COSEWIC 2007) and population assessments (e.g., Lunn et al. 1997; DFO 2011), as well as biological and ecological investigations in the context of climate change (Stewart and Lockhart 2005; Bajzak et al. 2012).

5.7.3.2 New Information and/or Re-analysis of Existing Information No additional analyses were conducted for Area 4 seals.

5.7.3.3 Changes in Seals over Time The following discussion was based on qualitative and quantitative data presented in key published literature.

5.7.3.3.1 Pre-Hydroelectric Development

Seal use of the Churchill River and estuary prior to CRD and other hydroelectric developments is poorly documented. McLaren (1958b) observed small numbers of ringed and bearded seals near Churchill, MB in the early 1950s, while McLaren (1958a) reported benthic invertebrates such as decapods (e.g., Hippolytidae, Eualus sp., Lebbeus sp.), amphipods (e.g., Lysianassidae), and small fish (e.g., Cottidae) from the stomachs of two ringed seals captured in the Churchill area in 1954 (McLaren 1958a). In the years leading up to CRD, Churchill residents harvested a small number of seals for bait and dog food, primarily in the fall (Manitoba Department of Mines and Natural Resources [MDMNR] 1955, 1956; McRae and Remnant 1997).

5.7.3.3.2 Post-Hydroelectric Development

No pre-CRD quantitative data are available to make pre- and post-CRD comparisons of seal distribution and abundance in the lower Churchill River and estuary. However, of the 20 long-term Churchill residents interviewed in Edye-Rowntree (2006), one (Mr. Chris Campbell) noted that “There are more seals around the river than before…” suggesting that seals may have been more abundant along the lower Churchill River than in pre-CRD years in response to decreasing water levels. In support of the Lower Churchill River Water Level Enhancement Weir Project, Manitoba Hydro and the Town of Churchill (1997) conducted detailed seal surveys in 1996 and 1997. Based on the 1996/97 data, and incidental observations collected in 1994 and 1995, the authors reported that seals were typically present in the estuary by mid-June, with increased numbers in early summer and fall (late August-September). The



largest number of seals observed at one time in the haul-out area, located near the west bank of the river between Drachm and Mosquito points (Map 5.7.3-1), was 32. Harbour seals were the most commonly observed species, although juvenile bearded seals were occasionally present. Remnant (1997) reported that climatic factors could influence seal use of the estuary, reporting that strong northerly winds, in combination with rain and low air temperatures, appeared to reduce seal use of the haul-out site. High air temperatures (greater than 25°C) also appeared to reduce numbers at the haul out area (Remnant 1997; Bernhardt 2001; Watts 1992).

Potential uses for the main haul-out area included social interaction, thermoregulatory advantage and rest (Manitoba Hydro and the Town of Churchill 1997; Remnant 1997), while an area 750 m upstream was likely used opportunistically for feeding (Remnant 1997). Although pupping was not observed in the vicinity of the weir axis, the presence of seal pups in fall suggested that pupping occurred in the region. Pupping was documented in the area during 2013 and 2014 (Florko 2015). Most seals appear to spend the majority of their time resting at the haul-out area (Bernhardt 2006).

Seal responses to construction and operation of the Lower Churchill River Water Level Enhancement Weir Project were assessed by Bernhardt (1999, 2000, 2001, 2006) and summarized in Bernhardt and Holm (2007). In all years (i.e., before and after weir construction), seal numbers in the lower Churchill River fluctuated daily and, to a lesser extent, seasonally. Seals were observed from the Port of Churchill to as far as 22 km upstream of the weir axis prior to construction, with more concentrated numbers along the main haul-out site (Bernhardt 1999; Map 5.7.3-1). Although seal abundance and distribution did not change during construction, the animals were ultimately displaced from the main haul-out area in response to elevated water levels upstream of the weir (Manitoba Hydro and the Town of Churchill 1997). As water levels increased, seals began hauling out farther upstream (approximately 500 m), before eventually settling in downstream locations between the weir and Mosquito Point (Figure 5.7.3-1; Bernhardt 1999).

No seasonal or daily trends in seal numbers were observed at post-weir haul-out sites (Bernhardt 2006). Meteorological factors (wind speed, air temperature and cloud cover) also did not appear to affect seal numbers, but river discharge and water elevation downstream of the weir had significant negative effects on seal abundance during extreme high flows. Although not significant, tide state also appeared to affect seal numbers in some portions of the haul-out area. Fewer harbour seals were present in 2005 when Churchill River discharge was exceptionally high (2,500 cms) (Table 5.7.3-1; Bernhardt 2006). During that period, seals were restricted to hauling out on the few exposed boulders occurring along the river bank due to the inundation of haul-out sites (Bernhardt 2006). Seals were occasionally observed upstream of the weir, confirming that the structure did not limit movements along the river (Bernhardt 2001, 2003, 2006). However, the number of seals observed upstream of the weir was lower in post-project years.

More recent seal surveys suggest that the number of harbour seals using the Churchill River has increased dramatically since 2005 (Florko 2015). As many as 110 seals were observed at one time during June 2014 (Florko 2015). Continued use of the lower Churchill River by seals suggests that this area remains a suitable, and likely preferable, haul-out location for harbour seals (Bernhardt 2006). It is unclear whether the voluntary moratorium placed on seal harvests in the late 1990s (discussed below), or the discontinuation of seal harvests for the purposes of bear baiting contributed to this increase.

Churchill River

WarkworthLake

Mosquito Point

Drachm Point

HudsonBay

GooseCreek

Churc

hill Ri

ver

Churchill

LegendPre-Weir Haul-Out SitePost-Weir Haul-Out SiteChurchill River Weir Axis

2.0

27-JUL-15


Created By: cparker - B Size Portrait BTB - MAR 2015 Scale: 1:85,000

Government of Canada, North/South Consultants, Manitoba Hydro

01-DEC-15

File Location: J:\MYP\MH_RCEA\Marine_Mammals\PHASE_2\MXDs\20151201_RCEA_PhaseII_AE_MarineMammals_DistributionOfSealHauloutSitesLowerChurchillRiver_cp.mxd

Hudson Bay

Thompson

Churchill

Distribution of Seal Haul-out Sites Lower Churchill River



0 1 2 Kilometres

0 0.75 1.5 Miles

DATA SOURCE:

DATE CREATED:

CREATED BY:

VERSION NO:

REVISION DATE:

QA/QC:

COORDINATE SYSTEM:

Weir AxisPrimary Haul-out Site

Cape Merry

Prince of Wales Fort

Map 5.7.3-1



Source: Bernhardt 1999. Zone 1 includes the area from the downstream face of the weir extending downstream into and including the Churchill River. Zone 2 includes the area between Drachm Point downstream to the weir. Zone 3 includes the lower Churchill River upstream of Drachm Point.

Figure 5.7.3-1: Distribution of Seals Observed during Aerial Surveys of the Lower Churchill River, 1998

The small number of seals observed upstream, combined with seal movements north into the estuary, indicate that seals primarily forage downstream of the weir and/or in nearshore marine areas (Bernhardt 2001, 2003, 2006). Bajzak et al. (2012) used satellite-linked radio transmitters to track movements and dive behaviour of Churchill area harbour seals (Figure 5.7.3-2), reporting that seals conducted repeated trips between nearshore Hudson Bay and the Churchill River haul-out area in spring and fall, with frequent offshore trips. Offshore trips were presumed to be for feeding (Bernhardt 2006), which is consistent with harbour seal foraging behaviour observed in other locations (e.g., Lesage 1999; Lowry et al. 2001; Dietz et al. 2003). Harbour seals from the Churchill River showed a high degree of seasonal fidelity to the area, and to specific offshore areas (Bajzak et al. 2012).

Manitoba Hydro and the Town of Churchill (1997) predicted that increased harvest pressure as a result of weir construction would be the largest impact to seals. A voluntary moratorium on seal harvests in the vicinity of the weir was put into place and few seals were harvested in the area in post-weir years (Bernhardt 2003). The shallow water, high water velocity and an abundance of large boulders downstream of the weir prevented whale watching and other boat traffic from approaching hauled-out seals. Bernhardt (2003) reported that few people were observed near the current haul-out areas.



Table 5.7.3-1: Comparison of the Annual and Seasonal Mean Number of Seals Observed along the Lower Churchill River, 1996–2005

Year1 Season Maximum Number of Seals

Count Range Mean2 SE

1996 Seasons Combined 57 2–18 7.6 0.6

Summer 31 2–8 4.4 0.3 Fall 12 4–12 8.7 0.7 1999 Seasons Combined 80 0–20 10.2 0.7

Spring 34 0–17 7.8 0.9

Summer 36 1–20 14 0.8 Fall 10 0–13 4.6 1.9 2000 Seasons Combined 93 0–29 12.9 0.7

Spring 38 2–29 12.8 1

Summer 32 0–20 11.9 1.3 Fall 23 7–22 14.4 0.9 20053 Seasons Combined 74 0–15 5.2 0.5

Spring 13 0–8 5.1 0.7

Summer 29 0–15 5.2 0.9 Fall 32 1–11 5.3 0.6 1. Maximum hourly counts from systematic surveys were pooled for each year, and all data were included, regardless of the

number of hourly observations on any given day. 2. Significant differences in the mean annual number of seals were observed (Kruskal-Wallis H 0.05, 3 = 17.297, p < 0.001);

entire data set used in this comparison. 3. Non-significant differences between years are observed if 2005 data are excluded (Kruskal-Wallis H 0.05, 2 = 1.814,

p < 0.404).



Source: Bernhardt (2006). Dashed red line indicates the 50 m isobath.

Figure 5.7.3-2: Average Daily Locations for 17 Juvenile and Adult Harbour Seals Satellite Tagged in the Churchill River Estuary, Western Hudson Bay, during Fall, 2001 and 2002

5.7.3.4 Cumulative Effects of Hydroelectric Development on Seals in Area 4

5.7.3.4.1 Summary • There is insufficient information available to assess the effects of CRD on the abundance and

distribution of harbour seals within the lower Churchill River and estuary. However, local knowledge suggests that seal numbers may have increased in the lower Churchill River, which may have been due to decreased flow/water levels and increased availability of haul-out locations.

• Construction of the Lower Churchill River Water Level Enhancement Weir Project resulted in a minor shift (less than 1 km) in haul-out locations used by seals. Post-weir seal numbers in the river and estuary were comparable to those observed in pre-weir years (Bernhardt 2000, 2001, 2003, 2006); however, current information indicates that these numbers have increased considerably over the last decade (Florko 2015). These increases suggest that despite the minor displacement of seals to haul-out areas downstream, the weir did not negatively impact seal abundance in the Churchill River and estuary.

• It is unknown how other changes in the area may have influenced seal abundance, including a voluntary moratorium on seal harvest at the weir and anthropogenic impact at haul-out areas through access limitations due to low water.

a) b)



5.7.4 Summary of the Effects of Hydroelectric Development in the Region of Interest on Seals

The effects of hydroelectric development on seals relate primarily to the displacement of haul-out sites within the Nelson and Churchill River estuaries. Increasing seal numbers in post-hydroelectric development years and their use of alternate areas for hauling out suggest that despite minor displacements, seal abundance in the RCEA ROI does not appear to have been materially affected.



5.7.5 Bibliography

5.7.5.1 Literature Cited and Data Sources Allen, S.G., Ainley, D.G., Page, G.W., and Ribic, C.A. 1984. The effect of disturbance on harbour seal

haul-out patterns at Balinas Lagoon, California. Fisheries Bulletin 82: 493-500.

Ambrose, A., and Berger, R. 2012a. Results of mammal investigations in the Conawapa study area, 2005-2006. Conawapa Project Environmental Studies Program Report # 06-12. A report prepared for Manitoba Hydro by Wildlife Resource Consulting Services MB Inc., Winnipeg, MB. 465 pp.

Ambrose, A., and Berger, R. 2012b. Results of mammal investigations in the Conawapa study area, 2006-2007. Conawapa Project Environmental Studies Program Report # 07-06. A report prepared for Manitoba Hydro by Wildlife Resource Consulting Services MB Inc., Winnipeg, MB. 449 pp.

Antonelis, G.A., Melin, S.R., and Bukhtiyarov, Y.A. 1994. Early spring feeding habits of bearded seals (Erignathus barbatus) in the central Bering Sea, 1981. Arctic 47(1): 74-79.

Baird, R.W. 2001. Status of harbour seals, Phoca vitulina, in Canada. Canadian Field-Naturalist 115(4): 663-675.

Bajzak, C.E., Bernhardt, W., Mosnier, A., Hammill, M.O., and Stirling, I. 2012. Habitat use by harbour seals (Phoca vitulina) in seasonally ice-covered region, the western Hudson Bay. Polar Biology 36(4): 477-491.

Baker, R.F. 1989. An environmental assessment and biological investigation of the Nelson River estuary. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 160 pp.

Baker, R.F. 1990. A fisheries survey of the Nelson River estuary, August 1989. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 26 pp.

Barton, K, Booth, K., Simmons, D.G., and Fairweather, J.R. 1998. Tourist and New Zealand Fur Seal interactions along the Kaikoura Coast. Education Centre Report No. 9. Lincoln University, Lincoln, New Zealand.

Bernhardt, W.J. 1999. Lower Churchill River water level enhancement weir project, an assessment of seal responses to 1998 construction activities. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 22 pp.

Bernhardt, W.J. 2000. Lower Churchill River water level enhancement weir project post-project monitoring: An assessment of seal responses to operation of the project, Year I, 1999. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 18 pp.



Bernhardt, W.J. 2001. Lower Churchill River water level enhancement weir project post-project monitoring: An assessment of seal responses to operation of the project, Year II, 2000. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 37 pp.

Bernhardt, W.J. 2003. Lower Churchill River water level enhancement weir project post-project monitoring: A synthesis of aquatic environmental monitoring 1999–2002. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 75 pp.

Bernhardt, W.J. 2006. Lower Churchill River Water Level Enhancement Weir Project Post-Project Monitoring: An assessment of seal responses to operation of the Project, a synthesis of monitoring from 1996 to 2005. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 110 pp.

Bernhardt, W.J. 2014. Seal abundance and distribution in the Nelson River estuary, 2005–2009. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB.

Bernhardt, W.J., and Holm, J. 2007. Lower Churchill River water level enhancement weir project post-project monitoring: A synthesis of aquatic environmental 1999–2006. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 107 pp.

Bjørge, A., Thompson, D., Hammond, P., Fedak, M.A., Bryant, E., Aarefjord, H., Roen, R., and Olsen, M. 1995. Habitat use and diving behaviour of harbour seals in a coastal archipelago in Norway. In Whales, seals, fish and man. Edited by A.S. Blix, L. Walløe, and Ø. Ulltang.Elsevier Science B.V., Amsterdam. 211-223 pp.

Brandson, L.E. 2011. Churchill Hudson Bay. A guide to natural and cultural heritage. Churchill Eskimo Museum Inc., Churchill, MB. 384 pp.

Brown, E.G. and Prior, A. 1998. Recreational disturbance to breeding seabirds and seals on Mousa, SSSI. Report to Scottish Natural Heritage, Contract no: HT/97/98/33.

Burns, J.J. 1981. Bearded seal (Erignathus barbatus, Erxleben, 1777). In Handbook of Marine Mammals, Volume 2: Seals. Edited by S. H. Ridgway and R. J. Harrison. Academic Press. London, UK. 145-170 pp.

Burns, J.J. 2002. Harbour and spotted seals. In Encyclopedia of marine mammals, 2nd Edition. Edited by W. F. Perrin, B. Würsig, and J. G. M. Thewissen. Academic Press, San Diego. 553-560 pp.

Cleator, H.J. 1996. Status of the bearded seal (Erignathus barbatus) in Canada. Canadian Field-Naturalist 110: 501–510.

Comeau, N.A. 1915. Report on the fisheries expedition to Hudson Bay in the auxiliary schooner "Burleigh" 1914. In Reports on fisheries investigations in Hudson and James bays and tributary waters in 1914. Edited by C.D. Melville, A.R.M. Lower, and N.A. Comeau. King's Printer, Ottawa, ON. 10 pp.



COSEWIC (Committee on the Status of Endangered Wildlife in Canada). 2007. Assessment and update status report on the harbour seal Phoca vitulina. Atlantic and Eastern Arctic subspecies (Phoca vitulina concolor) and Lacs des Loups Marins subspecies (Phoca vitulina mellonae) in Canada. http://publications.gc.ca/site/archivee-archived.html?url=http://publications.gc.ca/collections/collection_2008/ec/CW69-14-78-2008E.pdf. (Accessed June 2, 2015).

COSEWIC. 2015. Candidate Species List Database. Available at: http://www.cosewic.gc.ca. (Accessed 24 March 2015).

Dehn, L.A., Sheffield, G.G., Follmann, E.H., Duffy, L.K., Thomas, D.L., and O’Hara, T.M. 2007. Feeding ecology of phocid seals and some walrus in the Alaskan and Canadian Arctic as determined by stomach contents and stable isotope analysis. Polar Biology 30: 167-181 pp.

DFO (Department of Fisheries and Oceans Canada). 2011. Review of aerial survey estimates for Ringed Seals (Pusa hispida) in western Hudson Bay, 2009 and 2010. DFO Canadian Science Advisory Secretariat, Science Advisory Report 2011/024. 5 pp.

DFO. 2015. Seals and Sealing in Canada: Seal species: Harbour Seals. http://www.dfo-mpo.gc.ca/fm-gp/seal-phoque/facts-faits/facts-faitsa-eng.htm#harbour. (Accessed May 29, 2015).

Didiuk, A.B. 1975. Fish and wildlife resources impact assessment, lower Nelson River: An interim report to the Lower Nelson River Advisory Board. Manitoba Department of Mines, Resources and Environmental Management, Winnipeg, MB. 216 pp.

Dietz, R., Teilmann, J., Henriksen, O.D., and Laidre, K. 2003. Movements of seals from Rodsand seal sanctuary monitored by satellite telemetry - Relative importance of the Nysted Offshore Wind Farm area to the seals. National Environmental Research Institute, Ministry of the Environment, Denmark. NERI Technical Report No. 429.

Edrén, S.M.C., Teilmann, J., Dietz, R., and Carstensen, J. 2004. Effect of the construction of Nysted Offshore Wind Farm on seals in Rødsand seal sanctuary based on remote video monitoring. Technical report to Energi E2 A/S for the Ministry of the Environment, Denmark. 31 pp.

Edye-Rowntree, J. 2006. Residents’ perspectives on the Churchill River. Aboriginal Issues Press, University of Manitoba. 109 pp.

Fast, H.B. and D. Saunders. 1996. Ch. 4.1. Subsistence over time: The York Factory Cree of Northern Manitoba. In From Kischewaskahekan to York Landing: A land use history of York Factory First Nation. Edited by H. Fast and D. Saunders. 86-133 pp.

Feltz, E.T., and Fay, F.H. 1966. Thermal requirements in vitro of epidermal cells from seals. Cryobiology 3: 261-264.

Finley, K.J. and Evans, C.R. 1983. Summer diet of the bearded seal (Erignathus barbatus) in the Canadian High Arctic. Arctic 36: 82-89 pp.

http://publications.gc.ca/site/archivee-archived.html?url=http://publications.gc.ca/collections/collection_2008/ec/CW69-14-78-2008E.pdf



http://www.dfo-mpo.gc.ca/fm-gp/seal-phoque/facts-faits/facts-faitsa-eng.htm#harbour

http://www.dfo-mpo.gc.ca/fm-gp/seal-phoque/facts-faits/facts-faitsa-eng.htm#harbour



Florko, K.R. 2015. Abundance estimates, haul out patterns, and natural history of harbour seals (Phoca vitulina) in the Churchill River estuary. Honours Thesis (BIOL-4111/6), Department of Biology, University of Winnipeg. 44 pp.

Fox, K.S. 2008. Harbor seal behavioural response to boaters at Bair Island refuge. M.Sc. theses, Department of Environmental Studies, San Jose State University, paper 3591. 87 pp.

Gjertz, I., Lydersen, C., and Wiig, Ø. 2001. Distribution and diving of harbour seals (Phoca vitulina) in Svalbard. Polar Biology 24: 209-214.

Hastings, K.K., Frost, K.J., Simpkins, M.A., Pendleton, G.W., Swain, U.G., and Small, R.J. 2004. Regional differences in diving behavior of harbor seals in the Gulf of Alaska. Canadian Journal of Zoology 82: 1755-1773.

Henry, E. and Hammill, M.O. 2001. Impact of small boats on the haulout activity of harbour seals (Phoca vitulina) in Métis Bay, Saint Lawrence Estuary, Québec, Canada. Aquatic Mammals 27(2): 140-148.

Hjelset, A.M., Andersen, M., Gjertz, I., Lydersen, C., and Gulliksen, B. 1999. Feeding habits of bearded seals (Erignathus barbatus) from the Svalbard area, Norway. Polar Biology 21: 186-193 pp.

Idle, P.D. 1989. Temporal aspects of beluga, Delphinapterus leucas, behaviour in the Churchill River estuary. M.Sc. thesis. Laurentian University, Sudbury, ON. 121 pp.

IUCN (International Union for Conservation of Nature). 2008. The IUCN Red List of Threatened Species. Harbour Seals, Phoca vitulina. http://www.iucnredlist.org/details/17013/0. (Accessed May 29, 2015).

IUCN. 2014. The IUCN Red List of Threatened Species. Version 2014.3. http://www.iucnredlist.org. [Accessed 24 March 2015].

Jansen, J.K., Boveng, P.L., Dahle, S.P., and Bengston, J.L. 2010. Reaction of harbour seals to cruise ships. Journal of Wildlife Management 74(6): 1186-1194.

Johnson, A., and Acevedo-Gutiérrez, A. 2007. Regulation compliance by vessels and disturbance of harbour seals (Phoca vitulina). Canadian Journal of Zoology 85: 290-294.

Kovacs, K.M. 2002. Bearded seal Erignathus barbatus. In Encyclopedia of Marine Mammals. Edited by W. F. Perrin, B. Wursig, and J. G. M. Thewissen. Academic Press, San Diego, California. 84-87 pp.

Kovacs, K.M. and Innes, S. 1990. The impact of tourism on harp seals (Phoca groenlandica) in the Gulf of St. Lawrence, Canada. Applied Animal Behaviour Science 26: 15-26.

Kucey, L. 2005. Human disturbance and the hauling out behaviour of stellar sea lions (Eumetopias jubatus). M.Sc. Thesis, Department of Zoology, University of British Columbia. 75 p. http://www.marinemammal.org.pdfs/kucey_2005.pdf. (Accessed March 24, 2014).

http://www.iucnredlist.org/details/17013/0

http://www.marinemammal.org.pdfs/kucey_2005.pdf



Lawrence, M.J., Paterson, M., Baker, R.F., and Schmidt, R. 1992. Report on the workshop examining the potential effects of hydroelectric development on beluga of the Nelson River estuary, Winnipeg, Manitoba, November 6 and 7, 1990. Canadian Technical Report of Fisheries and Aquatic Science 1838: 43 pp.

Lesage, V. 1999. Trophic relationships, seasonal diving activity and movements of harbour seals, Phoca vitulina concolor, in the St Lawrence River Estuary Canada. Ph.D. thesis, University of Waterloo, Waterloo, ON.

Lesage, V., Hammill, M.O., Kovacs, K.M. 1999. Functional classification of harbor seal (Phoca vitulina) dives using depth profiles, swimming velocity. and an index of foraging success. Canadian Journal of Zoology 77: 74- 87.

Lesage, V., Hammill, M.O., and Kovacs, K.M. 2004. Long-distance movements of harbour seals (Phoca vitlililla) from a seasonally ice-covered area, the St. Lawrence River estuary, Canada. Canadian Journal of Zoology 82: 1070-1081.

Lidgard, D.C. 1996. The effects of human disturbance on the maternal behaviour and performance of grey seals (Halichoerus gryptus) at Donna Brook, Loncolnshire, England. Preliminary Report to the British Ecological Society, UK.

Lower Nelson Overview Study Team. 1981. The lower Nelson overview study and appendices. A report prepared for Manitoba Hydro by the Lower Nelson Overview Study Team, Winnipeg, MB. 251 pp.

Lowry, L.F., Frost, K.J., and Burns, J.J. 1980. Feeding of bearded seals in the Bering and Chukchi seas and trophic interactions with Pacific walruses. Arctic 33: 330-342.

Lowry, L.F., Frost, K.J., Ver-Hoef, J.M., and DeLong, R.A. 2001. Movements of satellite-tagged subadult and adult harbour seals in Prince William Sound, Alaska. Marine Mammal Science 17: 835-861.

Lunn, N.J., Stirling, I., and Nowicki, S.N. 1997. The distribution and abundance of ringed (Phoca hispida) and bearded seals (Erignathus barbatus) in western Hudson Bay. Canadian Journal of Fisheries and Aquatic Sciences 54: 914-921 pp.

MacDonell, D.S. 1996. Conawapa access road resource harvest study. North/South Consultants Inc., Winnipeg, MB. Unpublished memo to Manitoba Hydro.

Manitoba Hydro and the Town of Churchill. 1997. Lower Churchill River water level enhancement weir project. Manitoba Hydro, Winnipeg, MB, and the Town of Churchill, Churchill, MB.

Mansfield, A.W. 1967. Seals of Arctic and eastern Canada. Bulletin of the Fisheries Research Board of Canada 137: 30 p.

McRae, S.M., and Remnant, R.A. 1997. Results of questionnaire on domestic harvesting in the lower Churchill River area, 1996. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 82 pp.



McLaren, I.A. 1958a. The biology of the ringed seal (Phoca hispida Schreber) in the eastern Canadian Arctic. Fisheries Research Board of Canada Bulletin 118: 97 pp.

McLaren, I.A. 1958b. The economics of seals in the eastern Canadian arctic. Fisheries Research Board of Canada, Arctic Biological Station Circular I. 94 pp.

MDMNR (Manitoba Department of Mines and Natural Resources), Game and Fisheries Branch. 1955. Registered Traplines Conference of Conservation Officers. The Pas, Manitoba. July 4 to 7, 1955. Churchill section by W.R. Burns. P. 45-56 of 243 pp.

MDMNR. 1956. Registered Traplines Conference of Conservation Officers. The Pas, Manitoba. July 16 to 19, 1956. Churchill section by W.R. Burns. p. 87-97 of 227 pp.

North/South Consultants Inc. 2012. Limestone Generating Station: Aquatic environment monitoring programs: A synthesis of results from 1985 to 2003. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 192 pp.

Orsini, J.P., Shaughnessy, P.D., and Newsome, D. 2006. Impacts of human visitors on Australian Sea Lions (Neophoca cinerea) at Carnac Island, Western Australia: implications for tourism management. Tourism in Marine Environments 2: 23-37.

Pratte, B.D. 1976. Churchill River salt-water tidal model. Coastal Engineering Proceedings No. 15. 15 pp.

Reeves, R.R., and Mitchell, E. 1989. History of exploitation, distribution and stock identity of white whales in western Hudson Bay and Foxe Basin. Okapi Wildlife Associates, Hudson, Quebec for Department of Fisheries and Oceans, Ste. Anne-de-Bellevue, QC. v + 85 pp.

Remnant, R.A. 1997. Abundance of seals in lower Churchill River, 1996 and 1997. A report prepared for Manitoba Hydro by North/South Consultants Inc., Winnipeg, MB. 20 pp.

Renouf, D., and Lawson, J.W. 1986. Harbour seal vigilance: watching for predators or mates? Biology of Behaviour 11: 44-49.

Riedman, M. 1990. The Pinnipeds: Seals, Sea Lions, and Walruses. University of California Press, Berkeley, California. 439 pp.

Risch, D., Clarke, C.W., Cockeron, P.J., Elepfandt, A., Kovacs, K.M., Lydersen, C., Stirling, I. and Van Parijs, S.M. 2007. Vocalizations of male bearded seals, Erignathus barbatus: classification and geographical variation. Animal Behavior 73: 747-762 pp.

RSW-Environnement Illimité. 2013. Conawapa Generating Station – Stage IV Studies. Physical Environment of the Nelson River Estuary. WBS: 1.4.3.3.2.11.2280 Manitoba Hydro File : 00192-11100-0060. A report prepared for Manitoba Hydro by RSW – Environnement Illimité Inc.

Schneider, D.C., Rosenfeld, M.B., Twichell, D.C., and Kesselheim, C. 1980. Studies of the harbour seal (Phoca vitulina concolor) at a winter haulout site in Massachusetts. Report of the U.S. Department of Commerce, National Marine Fisheries Service, Washington, D.C.



Schneider-Vieira, F., Baker, R. and Lawrence, M.J. 1993. The estuaries of Hudson Bay: a case study of the physical and biological characteristics of selected sites. North/South Consultants Inc., Winnipeg, MB. 35 pp.

Sergeant, D.E. 1986. Sea mammals. In Canadian inland seas. Edited by I. P. Martini. Elsevier Science Publishers B. V., Amsterdam, The Netherlands. 327-340 pp.

Shaughnessy, P.D., Nichols, A.O., and Briggs, S.V. 1999. Interactions between tourists and wildlife at Montague Island: fur seals, little penguins and crested terns. Report to the South Australian National Parks and Wildlife Service. CSIRO Wildlife and Ecology, Canberra, Australia.

Smith, T.G. 1975. Ringed seals in James Bay and Hudson Bay: population estimates and catch statistics. Arctic 28(3): 170-182.

Smith, T.G. 1981. Notes on the bearded seal, Erignathus barbatus, in the Canadian Arctic. Canadian Technical Report of Fisheries and Aquatic Sciences 1042. 49 pp.

Stewart, D.B. and Howland, K.L. 2009. An ecological and oceanographical assessment of the alternate ballast water exchange zone in the Hudson Strait region. DFO Canadian Science Advisory Secretariat Research Document. 2009/008. vi + 96 p.

Stewart, D.B., and Lockhart, W.L. 2005. An overview of the Hudson Bay marine ecosystem. Canadian Technical Report of Fisheries and Aquatic Sciences 2586: 487 pp.

Suryan, R.M., and Harvey, J.T. 1999. Variability in reactions of Pacific harbour seals, Phoca vitulina richardsi, to disturbance. Fisheries Bulletin 97: 332-339.

Terhune, J.M. 1985. Scanning behaviour of harbour seals on haulout sites. Journal of Mammalogy 66: 392-395.

The Manitoba Historical Society. 2015a. Manitoba History: Exploring Northern Skies: The Churchill Research Range. http://www.mhs.mb.ca/docs/mb_history/44/exploringnorthernskies.shtml (Accessed October 1, 2015).

The Manitoba Historical Society. 2015b. Historic Sites of Manitoba: Churchill Grain Terminal and Port Facilities (Churchill). http://www.mhs.mb.ca/docs/sites/churchillgrainterminal.shtml (Accessed October 1, 2015).

Tollit, D.J., Greenstreet, S.P.R., and Thompson, P.M. 1997. Prey selection by harbour seals, Phoca vitulina, in relation to variations in prey abundance. Canadian Journal of Zoology 75: 1508-1518.

Watts, P. 1992. Thermal constraints on hauling out by harbour seals (Phoca vitulina). Canadian Journal of Zoology 70: 553-560.

http://www.mhs.mb.ca/docs/mb_history/44/exploringnorthernskies.shtml

http://www.mhs.mb.ca/docs/sites/churchillgrainterminal.shtml

REGIONAL CUMULATIVE EFFECTS ASSESSMENT – PHASE II WATER – BELUGA

DECEMBER 2015 5.8-1

5.8 Beluga

5.8.1 Introduction Beluga whales (Delphinapterus leucas) are circumpolar, ranging from 82°N in the Arctic to 47°N in the St. Lawrence Estuary. Over 20 stocks or subpopulations have been tentatively identified using a combination of behavioural, genetic and contaminant data (Martin and Reeves 2000; Reeves et al. 2014). Combined, these subpopulations comprise a minimum of 150,000 beluga (International Whaling Commission 2000; Reeves et al. 2014), of which approximately 115,000 inhabit Canadian waters. In summer, most stocks maintain distinct geographic distributions; however, mixing can occur during spring and fall migrations and in overwintering habitat. Seven beluga stocks have been identified in Canadian waters: St. Lawrence Estuary; Ungava Bay; Eastern Hudson Bay; Western Hudson Bay (WHB); Eastern High Arctic-Baffin Bay; Cumberland Sound; and Eastern Beaufort Sea stocks (Committee on the Status of Endangered Wildlife in Canada [COSEWIC] 2004) (Map 5.8.1-1).

Belugas inhabiting the Churchill and Nelson River estuaries belong to the WHB stock. This stock is the largest in the world with an estimated 57,000 whales (95% confidence level [CL] = 37,700–87,100) (Richard 2005). Large aggregations occur in both estuaries during the summer, with the largest proportion inhabiting the Nelson River area. Belugas maintain strong fidelity to summering areas, and although they return to the same sites every year (Caron and Smith 1990; COSEWIC 2004), they are known to travel large distances between areas (Richard and Orr 2003). It is unclear why belugas congregate in estuaries and shallow nearshore areas. Reasons for this behaviour may include a combination of factors such as thermoregulatory advantage (Fraker et al. 1979; Hansen et al. 1988); moulting and growth (Watts et al. 1991); feeding (Kleinenburg et al. 1964); calving (Sergeant 1973; Sergeant and Brodie 1975); social functions; and avoidance of predators such as orcas (Orcinus orca) (Lawrence et al. 1992). Killer whales were not known to inhabit Hudson Bay prior to the mid-1900s (Degerbol and Freuchen 1935, cited in Ferguson et al. 2010; Reeves and Mitchell 1988) and until relatively recently, their presence in Western Hudson Bay was considered to be rare. Current evidence suggests that killer whale occurrence in Hudson Bay is increasing exponentially, likely in response to decreasing sea ice in Hudson Strait (Ferguson et al. 2010).

Beluga move offshore into deeper waters in mid-August and spend several weeks diving to the seafloor to (presumably) feed on fish and invertebrates prior to the fall migration (Smith and Martin 1994; Martin et al. 2001; Richard et al. 2001). Mating occurs in overwintering areas, which are shared by multiple stocks (COSEWIC 2004). The WHB population is believed to overwinter in Hudson Strait (Finley et al. 1982; Richard et al. 1990; Richard 1993; Richard and Orr 2003; Richard et al. 2005; Smith 2007) and as far east as the coast of Labrador (Brice-Bennett 1978 in COSEWIC 2004). Spring migration to summering areas occurs during late April/May (Sergeant 1973). Although migration routes have not been fully documented, the majority of WHB belugas are believed to follow the eastern coast of Hudson Bay south to the Belcher Islands, and then across through the pack ice to arrive along the Manitoba coast in late May and early June. A much smaller number of animals move westward towards

REGIONAL CUMULATIVE EFFECTS ASSESSMENT – PHASE II WATER – BELUGA

DECEMBER 2015 5.8-2

Southampton Island. Spring and fall movements of Hudson Bay beluga are outlined in Stewart and Lockhart (2005).

Beluga can reach ages of 25–30 years (Brodie 1969). Females reach sexual maturity at a mean age of five years and typically give birth to a single calf (Sergeant 1973). Gestation time is 14.5 months, and the calving interval is approximately three years (Sergeant 1973). Richard (1993) proposed an estimated rate of increase of 2–3% per year for the global beluga population. No other estimates have been published for beluga.

In Canada, belugas are managed by the Department of Fisheries and Oceans Canada (DFO) under the Fisheries Act and Marine Mammal Regulations. This species has been commercially and domestically harvested in Hudson Bay, although commercial harvests ceased in the late 1960s (Reeves and Mitchell 1989). Commercial harvesting of beluga for oil dates back to the 1600s. Commercial harvests were relatively sporadic through time, but were intensive during some periods (e.g., 4,500 between 1948 and 1958; Reeves and Mitchell 1989). Belugas have been domestically harvested throughout Hudson Bay for oil, food, skin and dog food (Reeves and Mitchell 1989). Although domestic harvests continue to hold cultural and social significance in the Canadian Arctic, few, if any, belugas are currently harvested in the Churchill and Nelson River areas (Manitoba Hydro and the Town of Churchill 1997; Lower Nelson Overview Study Team 1981). Between 1967 and 1992, 68 belugas were live-captured in the Churchill River estuary for zoos and aquariums around the world (DFO 2008). Live captures are regulated both internationally through the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) and non-convention members, and nationally under the 1993 Marine Mammal Regulations which are overseen by DFO (Richard 1993). Between 1975 and 1992/93, 53 live-captured belugas were traded to other countries under CITES permits (Richard 1993). This practice was banned by the Government of Canada in 1992 (DFO 2008).

Belugas have been red-listed by the International Union for Conservation of Nature (IUCN) as “Near Threatened”, with some stocks described as in serious trouble and others reduced in size but viable (Jefferson et al. 2012). The Committee on the Status of Endangered Wildlife in Canada categorizes the WHB beluga stock as “Special Concern” due to hunting pressure along migration routes, unknown effects of freshwater flow regulation on estuarine summer habitat resulting from hydroelectric development, and disturbance from potential increases in shipping (COSEWIC 2004). Natural limiting factors for beluga include ice entrapment (Mitchell and Reeves 1981; Brodie 1982), and predation by polar bears and orcas (Smith 1985; Lowry et al. 1987a, b).

Waterbodies with Information on Mercury in Fish...discuss the evolution of fish mercury...

Documents

Transcript of Waterbodies with Information on Mercury in Fish...discuss the evolution of fish mercury...