1

Adjusted Route Assessment for

Boreal Woodland Caribou and Moose

Bipole III Transmission Project

2

Wabowden AFPR Segment

3

Methods - Evaluation of Wabowden Caribou • Habitat Modeling Analysis and Constraints;

• Range-Wide Calving Habitat Suitability Assessment

• Known Calving Locations

• Winter and Summer Core Use Areas

• Aboriginal Traditional Knowledge

• Caribou Cumulative Effects Analysis

4

Winter and Summer Core Habitats – Wabowden Evaluation Range

5

Summary of AFPR on Wabowden Caribou Aspect of Bipole III Transmission Project Potential Effect on Wabowden Caribou

Length of Transmission Line The AFPR eliminates the requirement of approximately 49 km of

new ROW through the Wabowden evaluation range.

Length of Transmission Line The AFPR follows existing ROWs and disturbed areas within the

evaluation range, resulting in no new additional fragmentation or

access in the area.

Length of Transmission Line The total length of ROW which intersects the Wabowden

evaluation range has been reduced by approximately 8.85 km.

Access The existing forestry buffers along the Provincial Road (PR) 373

provide additional mitigation to any additional effects from the

ROW on local caribou movements;

Disturbance Use of small core use areas by the Wabowden caribou between PR

373, Wabowden, and Bucko Lake is not expected to change or be

affected by this new routing due to the existing disturbance

regimes in the local area.

6

Wabowden AFPR - Conclusions

• The AFPR is similar to the original PPR recommend for BWC. – Reduces new ROW in the Wabowden BWC Range,

follows existing ROW’s and disturbed areas.

– Avoids winter core areas near PTH #6.

• Reduces scientific uncertainty regarding potential residual effects and conclusions for the FPR.

7

Moose Meadows AFPR Segment

8

Methods

• Moose Aerial Survey;

• Model Verification – Distance to Feature Analysis;

• Habitat Modeling;

• ATK and EACP;

• Enhanced Assessment.

9

Moose Meadows Survey

10

Moose Meadows Survey Results Results for Entire Survey Area

Area Elk Bulls Cows Calves Total Moose

GHA 12 2 3 5 2 10 GHA 13/13a 16 33 56 22 111

GHA 14 (including Moose Meadows)

36 28 34 24 86

Total 52 64 95 48 207

Moose Meadow Area Alone

n/a 6 12 8 26

Overall Calves:100 Cows for Survey Area = 51 Calves:100 Cows

11

High Quality Moose Habitat in Moose Meadows and Observed Concentrations

12

Summary of High Quality Habitat Intersected by the AFPR and FPR

FPR AFPR

Area of

Segment

Amount of

Modeled

Habitat within

3 mile buffer

(km2 and %)

Amount of

Modeled

Habitat within

66m ROW (km2

and %)

Area of

Segment

Amount of

Modeled

Habitat within

3 mile buffer

(km2 and %)

Amount of

Modeled

Habitat within

66m ROW (km2

and %)

138.834 6.589 (4.74%) 0.055 (0.83%) 159.237 35.006

(21.98%)

0.389 (1.11%)

13

14

AFPR – Moose Habitat

15

AFPR- Shrub 5

16

AFPR- Shrub 19

17

Main Moose Meadows (2&3)

18

Moose Meadows Main 12

19

Moose Meadows Main 3

20

Moose Meadows Main 14

21

Willow Areas (South of MM)

22

FRP- Moose Meadows 5

23

FRP- Moose Meadows 2

24

Moose Observations and Willow Habitat

25

Moose Observations Relative to FRI in Moose Meadows

26

Bellsite Habitat (5)

27

Bellsite- Shrub 8

28

Model Verification

Two distance to feature analyses were undertaken to characterize moose locations relative to high quality moose habitat and linear features:

1) Wilcoxon Rank Sum Tests;

2) Principal Component Analysis.

29

Model Verification - Wilcoxon Rank Sum Tests

Observed Moose

Distribution - Mean

Distance to Linear

Feature

Random Moose

Distribution - Mean

Distance to Linear

Feature

Wilcoxon Test

P-Value

Evaluation

Minor Roads 2710 3257 0.046 Closer

Forestry Roads 3783 4814 0.0057 Closer

Major Roads 4716 6656 0.0004 Closer

Transmission

Lines

5134 6920 0.0013 Closer

Rail Lines 4395 6604 0.0002 Closer

Cut Blocks 2866 3882 0.0012 Closer

Fire 1668 2377 0.0288 Closer

High Quality

Habitat

713 1706 <0.0001 Closer

30

GHA 19A and 14A AFPR Segment

31

GHA 19A AFPR Survey Area 91 moose observed

Note: No previous survey data available for GHA 19A

32

High Quality Moose Habitat in GHA 19A and 14A

33

Summary of High Quality Habitat Intersected by the AFPR and FPR

FPR AFPR

Area of

Segment

Amount of

Modeled Habitat

within 3mile

buffer (km2 and

%)

Amount of

Modeled

Habitat within

66m ROW (km2

and %)

Area of

Segment

Amount of

Modeled Habitat

within 3mile

buffer (km2 and

%)

Amount of

Modeled

Habitat within

66m ROW (km2

and %)

337.218 121.941 (36.16%) 1.871 (1.53%) 359.317 119.141 (33.16%) 1.96 (1.65%)

34

Enhanced Assessment

35

Introduction

• Provide a review of temporal, historical, and present day disturbance regimes with respect to moose populations for GHAs in western Manitoba in proximity to the Bipole III Project.

• Identify (if possible) landscape thresholds of disturbance that explain moose decline in the western region GHA’s.

36

Methods

• Natural Disturbance Assessment

• Comparisons of Historical Moose Density to Disturbance Metrics (Past & Present) – Regression Analysis

• Moose population modelling via Riskman

37

Study Area

STUDY AREA

38

Landscape Cover Data

Decade Available FRI or LCCEB

1960s GHA’s 12, 13, 14A

1970s GHA’s 18

1980s GHA’s 13, 14, 18, 19, 19A

Current LCCEB for all areas

39

Source Name Description Years

Manitoba Infrastructure and Transportation

Manitoba Official Government Road Map, Manitoba Official Road Map, Manitoba Official Highway Map

Digitized highways from the hard copy map

1945-1946, 1953, 1959, 1970, 1980-1981, 1990-1991, 2000-2001, 2012

Manitoba Mines Branch Drill Holes up to 2008 Drill holes in Manitoba up to 2008 2008

Manitoba Conservation and Water Stewardship (via Manitoba Land Initiative)

Fires up to 2011 Fires in Manitoba up to 2011 2012

Manitoba Conservation and Water Stewardship

FRI data 1960s, 1970’s, and 1980’s

Forest Resource Inventory for the 1960's, 1970’s, and 1980’s 1960, 1970, 1980

Tolko Industries Ltd Tolko Harvest 1968 - 2011 Historical Tolko Harvest Polygons (used 2001 - 2011) 2011

Louisiana Pacific Canada Ltd.

LP Harvest 1990 – 2011, LP Planned Harvest 2013 - 2022

Historical LP Harvest Polygons (used 2001 - 2011), Future LP Planned Harvest Polygons for 10 year plan

2011, 2012

Data Sources

40

Land cover classification

Habitat Classification Classes Cover Type Codes

LCCEB

Contiguous Mature Coniferous 210-213

Deciduous 220

Broadleaf 221-223

Mixedwood 230-233

Shrub Shrub Classes 50-52

Wetlands Wetland Classes 80-83

Land cover classification

Habitat Classification Sub Type Cutting Class

FRI

Contiguous Mature Productive Forests Stand Types 01-98, cutting Class 2-5

Shrub Productive Forests Stand Types 01-98, cutting Class 0-1

Willow Alder Classes 721-725

Wetlands Marsh/Muskeg Classes 831-839

Treed Wetland Classes 701-704

Creation of Common Land Cover

41

FRI Data

42

Contiguous Forest Patches

43

Landscape Metrics

Metric Definition

Edge density Amount of edge relative to the landscape area

Mean patch edge Average amount of edge per patch

Mean patch size Average patch size

Median patch size The middle patch size, or 50th percentile. Example: Median

Patch size of Conifer Patches

Number of patches Total number of patches in the landscape if ‘Analyze by

Landscape’ is selected, or Number of Patches for each individual class, if ‘Analyze by Class’ is selected

44

Summary of Analysis

Population Estimates summarized by year in each GHA, per Era of FRI

GHA 12 GHA 13 GHA 14 GHA 18 GHA 19

FRI - ERA Years Years Years Years Years

1960 - 1969 1965

1970 - 1979

1980 -1989 1983

1992

1990 - 1999 1991 1993

1998 1997 1998

2000 to Current 2007 2002 2007

2010 2011 2010

2012 2012

45

Example of linear disturbance over time

46

Example: 1980’s Shrubland habitat in GHA 18

47

Example: Current Shurbland habitat in GHA 18

48

Example: 1980 Fire in Porcupine Mts

49

Summary of Data

GHA

Productive

Moose

Habitat (km2)

Era of Landscape Data Years of

Survey

Years of

Major

Burn

Population

Density

(moose/km2)

Linear

Density

(km/km2)

12 2,426.93 1990 Estimate (metrics

current) 1991, 1998 2002 0.23 0.05

12 2,343.17 1960s 1965 1961 0.18 0.06

13 2,054.23 Current 2007, 2010,

2012 0.43 0.15

13 2,129.98 1990 Estimate (metrics

1980s) 1997 1980 0.52 0.15

14 4,497.94 Current 2002, 2011 0.07 0.10

14 4,526.31 1980s 1983, 1992 1989 0.45 0.11

18 4,269.67 Current 2007, 2010,

2012 0.38 0.16

18 4,321.10 1990s metrics from 1980s 1993, 1998 0.73 0.13

50

Methods – Regression Analysis

• Using all available historical and current FRI, LCCEB, LP, and MIT data, landscape and linear density metrics;

• Regression analysis was conducted using moose density to test for any significance among landscape and linear density metrics;

• IE: do any of these variables explain population density?

51

Regression Analysis – Example: Shrubs

Linear regression of relationship of shrubland to moose density per square kilometre. The regression slope is positive but the relationship is non-significant with a r2 = 0.36 and a P-value of 0.11

52

Regression Analysis – Example: Linear Density

Linear regression of relationship of linear feature/road density per square kilometre to moose density per square kilometre. The regression slope is positive but the relationship is non-significant with a r2 = 0.36 and a P-value of 0.11

53

Regression Analysis – Percent Wetland

Linear regression of percent wetland to moose density per square kilometre. Note: Significance

54

Wetlands – Moose Meadows

55

Summary: Regression Analysis

The results of single and multi-regression analyses did not yield any potential threshold value or significant correlations that explain moose density. Higher densities of moose are associated with more shrubland, more contiguous mature forest, higher densities of linear feature/roads, and higher total linear feature/road length, however, in all cases these relationships are not significant.

56

Summary of future disturbance in the Game Hunting Areas intersected by the Bipole III Project within the Project Study Area (20 Years)

Linear Density Thresholds Used/Reported in Canada (km/km2)

Greater Fundy Ecosystem Research Group (2005) 0.6

Manitoba Model Forest 0.58

Salmo et al. (2004) - Target Threshold 0.4

Salmo et al. (2004) - Critical Threshold 0.9

GHA

Current Linear Density

(km/km2)

Future Linear Density (including

Bipole III AFPR Future Drilling (km2)

Future % of Area Forest

Harvest (km/km2)

12 0.05 0.074 0.3 0.0047

14 0.1 0.143 0 0.0022

19A 0.31 0.367 0 0.00083*

57

Moose Population Modeling

58

Western Manitoba moose populations

Modelling Objectives

• Consider likely limiting factors • Examine their effects on population growth • Candidate limiting factors:

– diseases; – parasites; – predation; – licensed hunting; and – other

59

Diseases and parasites

• Chronic Wasting Disease (CWD) – Never detected in Manitoba in any species

• Brain worm (Parelaphostrongylus tenuis) – Single verified case in moose in Manitoba – near Cromer – Verified cases in deer in study area – Verified cases in moose across border in Saskatchewan – Likely present in study area at times, prevalence likely low

• Winter tick (Dermacentor albipictus) – Occasional outbreaks can yield mass mortality in moose – Anecdotal evidence suggests about 1/3 of moose in

western Manitoba may have died in 2002

60

Predation

• Predators in area include black bears and wolves

• Regional study (RMNP) shows wolves consume 3 times as much elk as moose

• From other NA studies, wolf and bear predation is highest on moose calves

• In western MB, surveys show more than 50 calves per 100 cows

61

Moose Population Density of Manitoba Game Hunting Areas, Saskatchewan Wildlife Management Zones, and Riding Mountain National Park

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

2.00

1960 1970 1980 1990 2000 2010 2020

De

nsi

ty (

km

2)

RMNP

Saskatchewan56

Saskatchewan57

Saskatchewan59

18,18A,18B,18C

13,13A

14

18,18A

14,14A

12

62

1992-1998 Western Moose Densities 2007-2012 Western Moose Densities

63

Model Input Parameters

• Winter survey data from western Manitoba: – Calf:cow ratio = 0.56 (n=12) – Bull:cow ratio = 0.66 (n=10) – I.e., for every 100 cow moose observed there

were 56 calves and 66 bulls observed.

• Literature: – annual adult female survival of 0.88 to 0.92; – 50:50 sex ratio at birth.

64

Age in May Age Class Female Survival Rate

(SE)

Male Survival Rate†

(SE)

0 Calves 0.55 (0.05) 0.55 (0.05)

1 Yearlings 0.91 (0.01) 0.85 (0.01)

2-11 Two year olds plus

adults 0.91 (0.01) 0.85 (0.01)

12-17‡ Older adults 0.81 (0.02) 0.75 (0.02)

Annual survival rates

Model Input Parameters

†The yearling and adult male survival rates were selected to yield the mean bull:cow ratio observed in the surveys conducted after 1990 (66:100) ‡ The survival rate was set to 0.0 for males age 15 and for females age 18

65

Age specific parturition and twinning rates

Model Input Parameters

Age in May Age Class Parturition Rate* (SE) Twinning Rate† (SE)

1 Yearlings 0.00 0.00

2 Two year olds 0.30 (0.1) 0.00

3-11 Adults 0.91 (0.1) 0.30 (0.1)

12-17‡ Older adults 0.70 (0.1) 0.30 (0.1)

† Twinning rate is expressed as a proportion of parturient animals. ‡ The parturition rate was set to 0.0 for females age 18.

66

Moose - general Model Parameters

Age of Maturity 2.5 years 30% at 1.5 years 70% at 2.5 years

Pregnancy rate Up to 97% Yearlings 30% 2 - 11years old 91% ≥ 11 year old 70%

Twinning rate Up to 80%, Typically 25-50%

Yearlings 0% ≥ 2 years 30%

Fecundity rate (calves born per female)

Up to 1.48 1.01

Annual Recruitment rate (calves surviving per female)

Up to 1.00, Highly variable year to year Commonly 0.30-0.60

0.56

Adult female survival rate (without hunting)

88 to 98%, Typically about 90%

Weighted: 88%

Potential population growth rate (Lambda)

Up to 1.40 Commonly 1.10 – 1.20

1.13

67

Model Input Parameters - check

• Winter survey data from western Manitoba: – Calf:cow ratio = 0.56

– Bull:cow ratio = 0.66

• Literature: – annual adult female survival of 0.88 to 0.92;

– 50:50 sex ratio at birth.

68

Model Results: GHA 14 Model Type Time Span

MCWS Initial

Pop Estimate*

MCWS Final

Pop Estimate

Annual Licensed

Harvest (%)

Other Annual

Mortality

Modelled

Population Result

No Harvests

1983-1992 1,560 2,480 0.00 0.00 4,442

Licenced Harvest

1983-1992 1,560 2,480 0.08 0.00 2,011

No Harvests

1992-2002 2,480 494 0.00 0.00 7,490

Licenced Harvest

1992-2002 2,480 494 0.12 0.00 5,924

Other Mortality

1992-2002 2,480 494 0.12 0.20 514

No Harvests

2002-2011 494(329) 148 0.00 0.00 922

Licenced Harvest

2002-2011 494(329) 148 0.01 0.00 889

Other Mortality

2002-2011 494(329) 148 0.01 0.20 149

69

Model Results: Other Mortality Model Type Time Span

MCWS Initial

Pop Estimate*

MCWS Final

Pop Estimate

Annual Licensed

Harvest (%)

Other Annual

Mortality

Modelled

Population Result

GHA 12 1991-1998 650 454 0.12 0.06 470

GHA 13/13A 1997-2007 1,118(745) 731 0.07 0.05 732

GHA 13/13A 2010-2012 1,122 817 0.03 0.23 829

GHA 14/14A 1992-2002 2,480 494 0.12 0.20 514

GHA 14/14A 2002-2011 494(329) 148 0.01 0.20 149

GHA 18-18C 1993-1998 3289 3066 0.08 0.06 3044

GHA 18-18C 1998-2007 3066(2044) 2008 0.08 0.04 2054

GHA 18-18C 2007-2010 2008 1349 0.04 0.20 1350

70

Survey Data & Model Discussion

• Moose recruitment rate is high (MCWS surveys)

• High recruitment rates are associated with low predation rates (literature)

• Given normal adult survival, populations should be growing (model)

• Observed population decline requires high and persistent additional source (s) of mortality (model)

71

• Survey data are inconsistent with high predation rates (i.e., recruitment is high)

• No evidence for disease related mortality

• Models accounted for winter tick mortality

• Models accounted for maximum effect of licensed hunting (non-selective harvest model)

• Unknown “other” mortality rate is high

• By process of elimination non-licensed hunting is best working hypothesis

• Plausible additive effect of access related mortality from Bipole III (hunting or predation) is insignificant by comparison.

72

Moose Response

• The contribution of predation facilitated by increased access resulting in direct moose population decline is limited. several examples of positive moose population response in highly fragmented and accessible habitats are found in Manitoba, Saskatchewan and Ontario.

• Crichton (2004) documented a positive increase in the Happy Lake area moose population following intensive access development and forest harvest through access management and hunting closures.

• A similar but less dramatic increase was also observed in the Beaver Creek area of eastern Manitoba where access was managed and hunting remained open (MCWS ).

73

Moose Response

• Saskatchewan monitored effectiveness of road management and wildlife refuges in forest harvest areas to protect moose from over hunting and documented increases in local moose populations after both large area and road corridor game preserves were established (pers. comm. E. Kowal, 2013).

• Rempel et al., 1997 illustrated that moose population rate of increase was positive in unmodified clear-cut areas and remained constant in modified clearcuts, which was thought to be a consequence of hunting.

• In all examples, hunting of moose was considered to be the main variable in explaining moose response with no documented effects or concern regarding increased predation by wolves as a result of increased access and fragmentation.

74

Conclusions

Based on the analysis conducted, no “threshold” of linear density or habitat metric could be determined to explain moose decline.

Increase of linear density in GHA’s 12, 14 or 19A as a result of Bipole III are minimal and well below linear density thresholds found in the literature for ecosystem maintenance and moose management.

I.E. Salmo et al. (2004) - Critical Threshold = 0.9 compared to cumulative effect calculations of 0.14 (GHA 14) and 0.36 (GHA 14A-19A).

75

Conclusions – Moose Meadows

•The GHA 14 (Moose Meadows) segment of the AFPR has 35.0 km2 (21.98%) of high quality moose habitat within the Local Study Area (LSA) 3 mile corridor; 0.389 km2 (1.11%) of this moose habitat within the LSA is within the 66 m ROW.

•Compared to the FPR (6.59 km2 within the LSA) the AFPR contains considerably more (28 km2) of high quality moose habitat.

•Based on the results of the aerial survey conducted between December 4 and 6, 2012, the AFPR compared to the FPR will intersect or come in proximity to additional areas of high moose density which are in proximity to existing access.

76

Conclusions – GHA 19A and 14A

• The GHA 19A and 14A segment of the AFPR has 119.14 km2 of high quality moose habitat within the LSA and 1.96 km2 within the 66 m ROW. The amount of high quality habitat within the FPR LSA is comparable to the AFPR (121.94 km2 high quality moose habitat).

•Based on the results of the aerial survey conducted between February 4th, 2013, the AFPR compared to the FPR intersects less areas of observed moose.

• Additional specific mitigation measures for 8 kilometres of the FPR have been identified by Manitoba Hydro to further reduce potential impacts to moose in GHA 19A and 14A.

77

Conclusions

Based on the results of the enhanced analysis and proposed (and enhanced) mitigation, the conclusions in the EIS remain consistent with those predictions in the original Bipole III EIS, Bipole III Mammal Technical Report, and the Route Adjustment Supplemental Report for the Bipole III Transmission Line Project.

“The residual effects on moose resulting from the project (AFPR / FPR) are not significant” Page 4-24.