Evaluating the status of Fraser River sockeye salmon and role of freshwater ecology in their decline

Full text

February 2011
technical report 3
Evaluating the Status of Fraser River Sockeye Salmon
and Role of Freshwater Ecology in their Decline
Marc Nelitz, Marc Porter, Eric Parkinson, Katherine Wieckowski, David Marmorek, Katherine Bryan,
Alexander Hall and Diana Abraham
The Cohen Commission of Inquiry
into the Decline of Sockeye Salmon
in the Fraser River

Technical Report 3
February 2011
Recommended citation for this report:
Nelitz, M., M. Porter, E. Parkinson, K. Wieckowski, D. Marmorek, K. Bryan, A. Hall and D. Abraham. 2011.
Evaluating the status of Fraser River sockeye salmon and role of freshwater ecology in their decline.
ESSA Technologies Ltd. Cohen Commission Tech. Rept. 3: 222p. Vancouver, B.C. www.cohencommission.ca
Evaluating the Status of
Fraser River Sockeye Salmon
and Role of Freshwater
Ecology in their Decline
Marc Nelitz, Marc Porter, Eric Parkinson, Katherine Wieckowski, David Marmorek, Katherine Bryan,
Alexander Hall and Diana Abraham
ESSA Technologies Ltd.
Suite 600 - 2695 Granville Street
Vancouver, BC V6H 3H4

Fraser River sockeye salmon are vitally important for Canadians. Aboriginal and non-Aboriginal
communities depend on sockeye for their food, social, and ceremonial purposes; recreational
pursuits; and livelihood needs. They are key components of freshwater and marine aquatic
ecosystems. Events over the past century have shown that the Fraser sockeye resource is fragile
and vulnerable to human impacts such as rock slides, industrial activities, climatic change,
fisheries policies and fishing. Fraser sockeye are also subject to natural environmental variations
and population cycles that strongly influence survival and production.
In 2009, the decline of sockeye salmon stocks in the Fraser River in British Columbia led to the
closure of the fishery for the third consecutive year, despite favourable pre-season estimates of
the number of sockeye salmon expected to return to the river. The 2009 return marked a steady
decline that could be traced back two decades. In November 2009, the Governor General in
Council appointed Justice Bruce Cohen as a Commissioner under Part I of the Inquiries Act to
investigate this decline of sockeye salmon in the Fraser River. Although the two-decade decline
in Fraser sockeye stocks has been steady and profound, in 2010 Fraser sockeye experienced an
extraordinary rebound, demonstrating their capacity to produce at historic levels. The extreme
year-to-year variability in Fraser sockeye returns bears directly on the scientific work of the
Commission.
The scientific research work of the inquiry will inform the Commissioner of the role of relevant
fisheries and ecosystem factors in the Fraser sockeye decline. Twelve scientific projects were
undertaken, including:
Project
1
Diseases and parasites
2
Effects of contaminants on Fraser River sockeye salmon
3
Fraser River freshwater ecology and status of sockeye Conservation Units
4
Marine ecology
5
Impacts of salmon farms on Fraser River sockeye salmon
6
Data synthesis and cumulative impact analysis
7
Fraser River sockeye fisheries harvesting and fisheries management
8
Effects of predators on Fraser River sockeye salmon
9
Effects of climate change on Fraser River sockeye salmon
10
Fraser River sockeye production dynamics
11
Fraser River sockeye salmon – status of DFO science and management
12
Sockeye habitat analysis in the Lower Fraser River and the Strait of Georgia
Experts were engaged to undertake the projects and to analyse the contribution of their topic area
to the decline in Fraser sockeye production. The researchers’ draft reports were peer-reviewed
and were finalized in early 2011. Reviewer comments are appended to the present report, one of
the reports in the Cohen Commission Technical Report Series.
Preface

i
Executive Summary
Although changes in marine conditions often play a key role in driving salmon population dynamics,
freshwater habitats are also important in how sockeye salmon express their resilience. Watershed
processes provide a high level of variability in conditions, which helps salmon express diverse life
history tactics, metapopulation structure, and genetic / phenotypic diversity. In Bristol Bay, Alaska
the diversity of sockeye salmon has been related to maintaining fish population stability across the
region and found to benefit ecosystems (by stabilizing inputs to terrestrial nutrient supplies and food
webs), and human communities (by stabilizing catch and reducing the number of fisheries closures).
Fraser River sockeye salmon and its component stocks demonstrate considerable life history
diversity. Stocks vary migration according to four adult run timing groups, demonstrate 4 year cycles
of abundance, and spend different lengths of time in freshwater / at sea. The abundance of Fraser
River sockeye salmon is also dominated by a few large stocks, which co-migrate with many smaller
stocks which are often less resilient to environmental stressors. Given this structure in abundance, it
is often difficult to maximize both harvest and population diversity. Weak stocks that are the target
of conservation are often harvested and become threatened when they co-migrate with the strong
stocks that are the target of the fishery. Thus, despite their inherent resilience this co-migration
illustrates how sockeye salmon are vulnerable.
This report is focused on evaluating changes in freshwater ecology and its role in recent sockeye
salmon declines for the Cohen Commission. This work includes examining the status of sockeye
salmon populations and habitats, as well as the impacts of human activities on freshwater habitats
(i.e., logging, hydroelectricity, urbanization, agriculture, and mining). Changes in freshwater ecology
due to natural and human forces are hypothesized as having three pathways of effects. These
pathways include effects on the: (1) quantity and quality of spawning habitats; (2) productivity of
nursery lakes for rearing; and/or (3) habitat conditions associated with migration of smolts / adults.
To assess the current status of Fraser River sockeye salmon populations, we have been charged with
three tasks: (1) summarizing existing delineations of population diversity into Conservation Units
(CUs); (2) evaluating Fisheries and Oceans Canada’s (DFO) methods for assessing conservation

ii
status; and (3) determining the status of Fraser River sockeye salmon CUs. Delineations of
Conservation Units were necessary to quantify habitat conditions, analyze landscape level
disturbances, and evaluate the relationship between changes in freshwater ecology and changes in
productivity. Strategy 1 of the Wild Salmon Policy includes a framework for delineating salmon
populations according to three major axes: ecology, life history, and molecular genetics. Using
DFO’s delineations, we identified 36 Conservation Units (30 lake and 6 river type CUs) within the
Fraser River basin. We use four criteria to evaluate alternative methods for assessing conservation
status of these CUs: (1) ecological criteria and indicators; (2) approach for setting benchmarks; (3)
data needs and availability; and (4) overall feasibility of implementation. No method is ideal across
these criteria; DFO’s method and two alternatives have different strengths and weaknesses. An
alternative to DFO’s method was used to summarize conservation status for 25 of 36 CUs; others
were not assessed due to insufficient data. Based on the results of the best available assessments, we
found that 17 of 36 Conservation Units have a poor population status and are distributed across all
timing groups (Early Stuart – Stuart, Takla / Trembleur; Early Summer – Nahatlatch, Anderson,
Francois, Taseko, Bowron, Shuswap Complex; Summer – Stuart, Takla / Trembleur; Late – Cultus,
Harrison u/s, Lillooet, Seton, Kamloops; River – Widgeon). The status of 11 CUs is unknown.
The majority of Fraser River sockeye salmon populations rear in large lakes for their first year of
life. Given our review of available data, measures of freshwater habitat condition are generally not
available across many CUs even though Strategy 2 of the Wild Salmon Policy is charged with
developing relevant habitat indicators. Given this gap, we developed direct and surrogate landscape
level indicators of the quantity and quality of migration, spawning, and rearing habitats for each
sockeye salmon lake-type CU using: (1) mapped habitat features we extracted or derived from
readily available GIS data, and (2) lake productivity datasets provided to us by DFO. These
indicators included: total spawn extent (m), ratio of lake influence to total spawning extent, nursery
lake area (ha), nursery lake productivity (estimated smolts / ha), migration distance (km), average
summer air temperature across adult migration (ºC), and average spring air temperature at the
nursery lake (ºC). Data were not available to describe basic habitat conditions for the river-type CUs.
Given a general lack of information that could be used to reliably define dynamic changes in
condition across sockeye salmon spawning, rearing, and migratory habitats we defined habitat
“status” as a combination of the: (1) intrinsic habitat vulnerability and (2) intensity of human

iii
stressors on those habitats. We used three independent and static indicators to define intrinsic habitat
vulnerability for each sockeye salmon freshwater life-stage. These independent indicators are: (1)
migration distance; (2) total area of nursery lakes; and (3) ratio of lake influence to total spawning
extent. The placement of an individual CU across these dimensions was used to illustrate its
vulnerability to watershed disturbances relative to other CUs in the Fraser River basin. The CUs with
the greatest relative habitat vulnerability include (i.e., have long migration distances, a low ratio of
lake influence to total spawning extent, and a small to moderate nursery lake area): Early Stuart –
Stuart, Takla / Trembleur; Early Summer – Bowron, Fraser; and Summer – Mckinley.
To understand the intensity of human stressors on habitats and assess the potential role of freshwater
stressors in recent declines of sockeye salmon we compiled and analyzed the best available data
describing six categories of human activities which have the potential to affect sockeye salmon:
forestry (e.g., forest harvesting activities, Mountain Pine Beetle disturbance, and log storage),
mining, hydroelectricity (large scale and run of river power projects), urbanization upstream of
Hope, agriculture, and water use. Next, we developed a spatial layer that represented “zones of
influence” on core habitats for migration, spawning, and rearing across each Conservation Unit using
DFO’s sockeye salmon habitat data (e.g., nursery lakes, spawning locations, monitoring sites, and
escapement data). We then intersected the stressor layers with our “zones of influence” layer to
summarize the intensity of human stresses on each Conservation Unit.
To assess the intensity, spatial distribution, and temporal patterns of forestry related stressors, we
examined the level of forest harvesting over time, density of roads and road-stream crossings, and
accumulated level of disturbance due to Mountain Pine Beetle (MPB) across sockeye salmon
watersheds. We also examined the best available site specific information to qualitatively assess the
impacts of log storage in the lower Fraser River. Our findings indicate that the level of forest
harvesting within the last 15 years is less than 10% of the area of sockeye salmon watersheds.
Drainage areas upstream of lake inlet spawning, tributary spawning, and nursery lakes tend to be
more heavily disturbed than the riparian zones adjacent to spawning downstream of lakes or along
migration corridors. There is considerable variation in road development across Conservation Units,
which tends to be concentrated in areas adjacent to spawning zones downstream of lakes and along
migration corridors. The level of MPB disturbance has increased dramatically since 2003, with the
level of disturbance being most dramatic in interior Fraser CUs as opposed to coastal CUs whose

iv
watersheds are largely absent of ponderosa and lodgepole pine. The intensity of Mountain Pine
Beetle disturbance has been very high; up to 90% of the area in some sockeye salmon watersheds.
Variation in the intensity of log storage appears to be larger across reaches than across seasons or
years within reaches of the lower Fraser River. Based on past studies, the historic intensity of log
storage has not appeared to have significant on juvenile salmon.
To assess the effects of mining, we examined the spatial distribution, number, and types of mines
occupying sockeye salmon watersheds in the Fraser River basin (e.g., placer mining, gravel mining,
industrial mineral production, metal mining, oil and gas production, coal mining, and exploration
related to these production activities). The occurrence of mining activity in the watersheds of
spawning streams varies substantially across sockeye salmon CUs. Placer mining is the dominant
mining activity and appears to have the highest potential to reduce early freshwater survival.
However, the data suggest the impacts of mining on sockeye salmon are likely small and difficult to
detect because the contrasts among stocks and strength of the effect relative to other factors is low.
To assess the effects of hydroelectricity, we reviewed scientific studies describing the effects of the
Bridge/Seton River power project and Alcan’s Kemano Project, as well as the spatial distribution of
small scale hydroelectric operations across sockeye salmon watersheds. The Bridge/Seton River
power project can affect migrations of smolts and adults on the Seton Rivers, but adverse effects
have been largely mitigated by changes in flow diversions and operations of the powerhouse.
Likewise, the Kemano Project affects water temperature on the lower Nechako River, but a
temperature compliance program has been implemented to ensure that water temperatures remain
suitable for adult passage. Our findings indicate that the history of interaction between IPPs and
sockeye salmon is very short and limited in number and spatial extent.
To assess the effects of urbanization upstream of Urban environments have a relatively small
footprint within watersheds and riparian zones that influence sockeye salmon, though urban
footprints have the most intense interaction with sockeye salmon migration corridors. The extent of
urban development along migration corridors is further illustrated by the human population data
which shows a similar pattern of concentration.

v
To assess the effects of agricultural activities (beyond impacts on water quality), we reviewed the
spatial distribution of agricultural lands. Compared to other land uses, agriculture has a relatively
small footprint within watersheds and riparian zones that influence sockeye salmon spawning and
rearing habitats. Agriculture does, however, have a greater interaction with migration corridors.
To assess the effects of water use, we calculated the total allocation of water, density of water
allocation restrictions, and distribution of water licenses across uses for all sockeye salmon water
sheds. Not surprisingly, high water demand is associated with the greatest concentrations of people
across the Fraser River basin. Migration corridors appear to have the greatest allocation of water
through licensing and the greatest density of water allocation restrictions, largely allocated to the
agricultural sector. The CUs of the Lower Mainland have the highest water allocations.
Given a lack of experimental design in the way population, habitat, and stressor data have been
collected, our ability to test for cause and effect relationships between the freshwater environment
and Fraser sockeye salmon declines was limited. As a result, we were only able to use a limited set
of quantitative techniques and data summaries to assess the role of freshwater influences.
We used three analytical approaches to gain insights into possible hypotheses about the role of
freshwater influences on Conservation Units. First, we developed a series of cumulative stressor
tables which: (1) aligned the hypothesized stressors to the relevant habitat types and Conservation
Units, (2) scored the relative intensity of and trend in disturbance, and (3) summarized the
cumulative level of stress on a Conservation Unit. Second, we plotted the measures of cumulative
stress against the indicators of habitat vulnerability to generate bivariate plots for each habitat type
and Conservation Unit (i.e., a summary of habitat status). Lastly, we developed a “dashboard”
summary of the all data available to describe population status, habitat vulnerability, and freshwater
stressors specific to each lake Conservation Units across the Fraser River basin.
We undertook three additional analyses to assess whether freshwater habitat conditions have
contributed to the recent declines in Fraser River sockeye salmon. First, we summarized key findings
from recent research examining alternative hypotheses for the declines in Fraser sockeye salmon.
This understanding was important for prioritizing our analytical efforts and developing testable
hypotheses that are consistent with these other studies. Second, we analyzed the habitat and stressor

vi
data to test whether they could explain declines in productivity. Lastly, for those habitat and stressor
variables for which we had time series data (i.e., forest harvesting, Mountain Pine Beetle
disturbance, summer air temperatures across adult migration, and spring air temperatures at nursery
lakes) we examined correlations with total salmon and juvenile productivity indices.
Due to our inability to rigorously test for cause effect relationships on survival at key life stages we
used a “weight of evidence” to reach a conclusion about significance of the role of freshwater
influences, drawing upon the data and analyses conducted through this effort. Using this approach
we believe that recent declines in Fraser River sockeye salmon are unlikely to be the result of
changes in the freshwater environment. An important piece of evidence in reaching this conclusion is
that juvenile survival has remained relatively stable across CUs where data are available, even
though there is substantial variation in stressor intensity across CUs.
Despite our belief that recent declines are not likely to be directly linked to deterioration in habitat
conditions, the protection of freshwater habitats remains important to the conservation of Fraser
River sockeye salmon because they contribute to their overall diversity and resilience. Given this
context, our recommendations include:
(1) To improve our understanding about survival at critical freshwater life stages, scientists
need better estimates of juvenile abundance, overwinter survival, and mortality during smolt
outmigration.
(2) To improve our understanding about population status across Conservation Units, scientists
need more information about the abundance and distribution of small lake and all river CUs.
(3) To improve our understanding about habitat status across Conservation Units, scientists
need information on habitats monitored in a consistent manner on a regular basis across a larger
number of rivers and nursery lakes.
(4) To improve our understanding about the population level effects of stressors on freshwater
habitats, scientists need more precise estimates of the biological consequences of disturbance as
a function of increasing stress.
(5) To improve transparency in the science and related decision making scientists, managers,
and the public need information that is more accessible and collected in a way that is more
integrated across federal and provincial agencies.

vii
Table of Contents
Executive Summary ............................................................................................................................. i
List of Figures ................................................................................................................................... viii
List of Tables ....................................................................................................................................... x
1.0 Context for assessing freshwater ecology of Fraser River sockeye salmon ............................. 1
2.0 Current status of Fraser River sockeye salmon ......................................................................... 6
2.1 Populations ................................................................................................................................... 6
2.1.1 Background on assessing status ............................................................................................ 6
2.1.2 Evaluation of status methodologies ...................................................................................... 8
2.2 Habitats ...................................................................................................................................... 12
2.2.1 Indicators of migratory habitat quantity / quality ............................................................... 15
2.2.2 Indicators of spawning habitat quantity / quality ................................................................ 16
2.2.3 Indicators of rearing habitat quantity / quality .................................................................... 17
2.2.4 Integrated summary of habitat vulnerability ....................................................................... 18
3.0 Freshwater stressors affecting Fraser River sockeye salmon ................................................. 21
3.1 Forestry ...................................................................................................................................... 21
3.1.1 Forest harvesting activities .................................................................................................. 21
3.1.2 Mountain Pine Beetle disturbance ...................................................................................... 24
3.1.3 Log storage / handling in the Fraser River estuary ............................................................. 27
3.2 Mining ........................................................................................................................................ 29
3.3 Hydroelectricity .......................................................................................................................... 33
3.3.1 Large scale .......................................................................................................................... 33
3.3.2 Small scale .......................................................................................................................... 39
3.4 Urbanization upstream of Hope ................................................................................................. 41
3.5 Agriculture ................................................................................................................................. 43
3.6 Water use .................................................................................................................................... 45
4.0 Freshwater influences on Fraser River sockeye salmon .......................................................... 49
4.1 Assessment within Conservation Units ...................................................................................... 49
4.2 Assessment across Conservation Units ...................................................................................... 52
4.3 Summary and conclusions.......................................................................................................... 56
5.0 State of the science ...................................................................................................................... 58
6.0 Recommendations ....................................................................................................................... 59
7.0 Figures .......................................................................................................................................... 62
8.0 Tables ........................................................................................................................................... 92
9.0 References .................................................................................................................................. 117
Appendix 1 – Statement of work ................................................................................................... 130
Appendix 2 – Reviewer evaluations and author responses ......................................................... 133
Appendix 3 – Dashboard summaries ............................................................................................ 158
Appendix 4 – Data sources ............................................................................................................. 220

viii
List of Figures
Figure 1. Four-year moving average of total adult returns per spawner across all Fraser River
sockeye salmon stocks divided by total spawners 4 years before.. ................................... 62
Figure 2. Conceptual model of the factors considered by this study (boxes with solid lines) and
relevant factors considered by other Cohen Commission studies (boxes with dashed
lines). ................................................................................................................................. 62
Figure 3. Overview of the Fraser River basin, watershed boundaries (in shades of grey), and
nursery lakes (in black) for all lake sockeye salmon Conservation Units. ....................... 63
Figure 4. Summary of the status for 36 sockeye salmon Conservation Units into four risk
categories: IV – status probably poor, but little information; III – status poor, high
confidence; II – status probably good, high uncertainty; and I – status good, high
confidence. ........................................................................................................................ 64
Figure 5. Modified conservation status for some CUs based on work of Grant et al. (2010). Blue
squares indicate conservation status that did not change as a result of Grant et al.’s work.64
Figure 6. Panel A: Timing of smolt outmigration as a function of latitude across multiple nursery
lakes in BC and Alaska. Panel B: Timing of lake ice breakup within a single nursery lake
in Alaska. Images from Burgner (1991). .......................................................................... 65
Figure 7. Summary of the habitat vulnerability for all lake sockeye salmon Conservation Units
using three independent indicators of habitat quantity / quality: migration distance (x-
axis), total area of nursery lakes (y-axis), and ratio of lake influence to total spawning
(size of circles). ................................................................................................................. 65
Figure 8. Area (in hectares) and volume (in 1,000s m3) of harvested forest in British Columbia
from 1975 to 2007 (data from Statistics Canada 2009). ................................................... 66
Figure 9. Frequency distribution of the level of forest harvesting within the “zones of influence” of
each habitat type across all Fraser River lake sockeye salmon Conservation Units. ........ 66
Figure 10. Spatial distribution of forest harvesting cutblocks relative to watershed boundaries (light
grey shading) for all lake sockeye salmon Conservation Units. ....................................... 67
Figure 11. Time series of the level of forest harvesting within “zones of influence” for each habitat
type across six Fraser River lake sockeye salmon Conservation Units. ........................... 68
Figure 12. Frequency distribution of the density of roads (km / km2) within the “zones of influence”
of each habitat type across all Fraser River lake sockeye salmon Conservation Units. ... 69
Figure 13. Frequency distribution of the density of road-stream crossings (# / km2) within the
“zones of influence” of each habitat type across all Fraser River lake sockeye salmon
Conservation Units. ........................................................................................................... 69
Figure 14. Frequency distribution of the accumulated level of Mountain Pine Beetle disturbance
from 1999 to 2008 within the “zones of influence” of each habitat type across all Fraser
River lake sockeye salmon Conservation Units. ............................................................... 70
Figure 15. Spatial distribution of the accumulated level of Mountain Pine Beetle disturbance from
1999 to 2008 (dark grey shading) relative to watershed boundaries (light grey shading)
for all lake sockeye salmon Conservation Units. .............................................................. 71
Figure 16. Time series of the accumulated level of Mountain Pine Beetle disturbance from 1999 to
2008 within “zones of influence” of each habitat type across six Fraser River lake
sockeye salmon Conservation Units. ................................................................................ 72
Figure 17. Aerial photo overview of the lower reaches of the Fraser River and estuary in 2009. ..... 73
Figure 18. Overview of the distribution of main categories of mines across the Fraser River basin. 74

ix
Figure 19. Schematic representation of the development of the Seton / Cayoosh diversion ............. 75
Figure 20. Portage Creek and Gates Creek sockeye salmon escapement from 1938 to 2006 ............ 75
Figure 21. Nechako River watershed and location of Kenny Dam (Map from NFCP 2005). ........... 76
Figure 22. Frequency distribution of the number of small scale hydroelectricity installations within
the “zones of influence” of each habitat type across all Fraser River lake sockeye salmon
Conservation Units. ........................................................................................................... 76
Figure 23. Spatial distribution of small scale hydroelectricity installations (squares with dots)
relative to watershed boundaries (grey shading) for all lake sockeye salmon Conservation
Units. Nursery lakes are in black. ..................................................................................... 77
Figure 24. Intake and headpond of the Douglas Creek generating station (a typical installation). .... 78
Figure 25. Human population size in the Fraser River basin by region from 1981 to 2006 .............. 78
Figure 26. Frequency distribution of the area of urban land within “zones of influence” of each
habitat type across all Fraser River lake sockeye salmon Conservation Units. ................ 79
Figure 27. Spatial distribution of urban areas (dark grey shading) relative to the watershed
boundaries (light grey shading) for all lake sockeye salmon Conservation Units. ........... 80
Figure 28. Time series of average human population density along migration corridors of all lake
sockeye salmon Conservation Units. ................................................................................ 81
Figure 29. Time series of average human population density adjacent to rearing and spawning
habitats for all lake sockeye salmon Conservation Units. ................................................. 81
Figure 30. Number of farms in British Columbia from 1881 to 2006 (diamonds, data from Statistics
Canada 2009) and total area of the province within the Agricultural Land Reserve from
1974 to 2007 (solid line, data from BC MOE 2008). ........................................................ 82
Figure 31. Number of livestock (cattle, pigs, chickens) per unit area (# / ha) in Abbotsford. Number
of livestock is represented as an animal unit equivalency ................................................ 82
Figure 32. Frequency distribution of the level of agricultural land within “zones of influence” of
each habitat type across all Fraser River lake sockeye salmon Conservation Units. ........ 83
Figure 33. Spatial distribution of agricultural areas (dark grey shading) relative to the watershed
boundaries (light grey shading) for all lake sockeye salmon Conservation Units. ........... 84
Figure 34. Frequency distribution of total water allocation (cubic metres per year per hectare) within
the “zones of influence” of each habitat type across all Fraser River lake sockeye salmon
Conservation Units. ........................................................................................................... 85
Figure 35. Frequency distribution of the density of water allocation restrictions (number per hectare)
within the “zones of influence” of each habitat type across all Fraser River lake sockeye
salmon Conservation Units. .............................................................................................. 85
Figure 36. Overlay of water licenses, water allocation restrictions, population density, and
distribution of all salmon species in the province. ............................................................ 86
Figure 37. Frequency distribution of the allocation of water by main uses within the “zones of
influence” of spawning and migratory habitats across all Fraser River lake sockeye
salmon Conservation Units. .............................................................................................. 87
Figure 38. Representation of the relative level of vulnerability of and stress on freshwater habitats
across all lake sockeye salmon Conservation Units. ......................................................... 88
Figure 39. First page of a “dashboard” summarizing population status and habitats for the Quesnel
Conservation Unit (L_6_10, Summer timing group). ....................................................... 89
Figure 40. Second page of a “dashboard” summarizing human stressors on the Quesnel
Conservation Unit (L_6_10, Summer timing group). ....................................................... 90
Figure 41. Overview of the mechanisms by which stressors in the freshwater environment can have
impacts on habitats, growth and survival across life stages, and ultimately a population
level effect on sockeye salmon in the Fraser River basin. ................................................ 91

x
List of Tables
Table 1. Status of 36 sockeye salmon Conservation Units (as reported by Pestal and Cass 2009),
alignment of these CUs with stocks for which there are productivity data from the Pacific
Salmon Commission (as analyzed by Peterman et al. 2010), and summary of evidence /
rationale for modifying status where appropriate (as part of this report’s evaluation). .... 92
Table 2. Comparison of alternative methods for evaluating status of sockeye salmon Conservation
Units according to their assessment criteria / indicators, feasibility of implementation,
approach for setting benchmarks, and data needs / availability. ....................................... 93
Table 3. Summary of indicator classes included in each assessment method. ................................ 97
Table 4. Description of indicators of habitat quantity and quality reflecting vulnerability across
different sockeye salmon life stages. ................................................................................ 97
Table 5. Indicators of habitat vulnerability for spawning, rearing, and migratory habitats across all
lake sockeye salmon Conservation Units. ......................................................................... 98
Table 6. Months used to represent historical average air temperature exposure of adult sockeye
salmon CUs along the migration corridor based on associated run timing group. ........... 99
Table 7. Summary of hypothesized links between freshwater stressors and sockeye salmon
habitats, and the indicators being generated to represent these stressors. ......................... 99
Table 8. Count and density (number/100km2) of various types of mining activity in watersheds
that support sockeye salmon spawning by Conservation Unit.. ...................................... 100
Table 9. Average number of days per year when the mean daily water temperature exceeds 20°C
in the Nechako River above the Stuart River, and in the Stuart River, July 20 to August
20, 1953 to 2000* (data from NFCP 2005). .................................................................... 101
Table 10. Number of days when the mean daily water temperature exceeds 20°C, and maximum
and minimum mean daily water temperatures in the Nechako River above the Stuart
River, July 20 to August 20, 2002 to 2009 (data from Triton Environmental Consultants
Ltd. 2003 through 2010 as part of NFCP water temperature and flow management
program). ......................................................................................................................... 101
Table 11. Relative ranking of Conservation Units based on the intensity and trend (where available)
of human stressors potentially interacting with spawning locations downstream of lakes
or on mainstem rivers. ..................................................................................................... 102
Table 12. Relative ranking of Conservation Units based on the intensity and trend (where available)
of human stressors potentially interacting with tributary or lake inlet spawning locations..103
Table 13. Relative ranking of Conservation Units based on the intensity and trend (where available)
of human stressors potentially interacting with nursery lake rearing.. ........................... 104
Table 14. Relative ranking of Conservation Units based on the intensity and trend (where available)
of human stressors potentially interacting with migration corridors.. ............................ 105
Table 15. Matrix of pairwise correlations among indicators of habitat vulnerability, habitat stressors
at the watershed scale (as quantified within our nursery lake “zone of influence”), and
stock productivity (trend in Ricker residuals from 1984-2004). ..................................... 106
Table 16. Predictor variables, AICc values, model rankings, and adjusted R-square values for all
candidate linear regression models relating sockeye salmon productivity (trend from 1984
to 2004 in Ricker residuals as the response variable) to indicators of habitat vulnerability
and stress. ........................................................................................................................ 107
Table 17. Summary of correlation coefficients between indicators of total productivity and two
habitat condition indicators related to adult migration (summer air temperatures along

xi
migration corridor) and smolt outmigration (spring time air temperatures at nursery
lakes).. ............................................................................................................................. 108
Table 18. Summary of population status, habitat vulnerability, and relative level of cumulative
stress for all sockeye salmon Conservation Units in the Fraser River basin. ................. 109
Table 19. Seven questions (from Stewart-Oaten 1996) and the related responses to our overall
assessment of the cumulative effect of freshwater stressors in contributing to the recent
declines of Fraser River sockeye salmon. ....................................................................... 110
Table 20. Seven questions (from Stewart-Oaten 1996) and the related responses to our overall
assessment of the effect of Forest harvesting, Mountain Pine Beetle, and roads in
contributing to the recent declines of Fraser River sockeye salmon. .............................. 111
Table 21. Seven questions (from Stewart-Oaten 1996) and the related responses to our overall
assessment of the effect of agriculture and urbanization, water use, and mines in
contributing to the recent declines of Fraser River sockeye salmon. .............................. 113
Table 22. Seven questions (from Stewart-Oaten 1996) and the related responses to our overall
assessment of the effect of small hydro, large hydro, and log storage in contributing to the
recent declines of Fraser River sockeye salmon. ............................................................ 115

1
1.0 Context for assessing freshwater ecology of Fraser River sockeye
salmon
Sockeye salmon are an icon in British Columbia and an important species for human, marine,
freshwater, and terrestrial communities. For human communities they are a cultural cornerstone
providing food, social, and ceremonial values to First Nations, while contributing $2.5 to $250
million annually1 in financial benefits to commercial fisheries depending on abundance of returns on
the Fraser (Nelson 2006). For freshwater and terrestrial communities they provide a means of
transferring marine nutrients to watersheds that support the production of salmon, other fish, riparian
forests, and wildlife (Naiman et. al 2002; Gende et al. 2002; Nelitz et al. 2006).
Sockeye salmon’s complex life history and physiology allows them to thrive in vastly different
conditions in marine and freshwater environments, and traverse the large distances in between. For
centuries sockeye salmon across the north Pacific have had highly variable, but sustained abundance,
while responding to changes in these environments (e.g., Gresh et al. 2000; Finney et al. 2000). The
ability of sockeye salmon to thrive across such a large range in environmental conditions and
stressors has led to their recognition as an inherently resilient species (Hilborn et al. 2003; Healey
2009). Seven attributes contribute to their resilience: multiple, independent reproducing populations,
high reproductive capacity, metapopulation structure, high genetic diversity, phenotypic plasticity,
variable life history tactics, and opportunistic use of habitat (Healey 2009).
Although changes in marine conditions often play a key role in driving salmon population dynamics,
freshwater habitats also play an important role in how sockeye salmon express their resilience at the
population level (Hilborn et al. 2003). Watershed processes provide a high level of variability in
environmental conditions (e.g., climate, vegetation, stream conditions), which help salmon express
diverse life history tactics, metapopulation structure, and genetic / phenotypic diversity that are
evident among populations (Bisson et al. 2009). For instance, sockeye salmon populations will vary
their life history tactics and physiology in response to differences in migration distances to natal
streams, water temperatures on spawning grounds, or rearing conditions in nursery lakes (reviewed
by Burgner 1991). In Bristol Bay, Alaska the diversity of sockeye salmon has been related to
maintaining fish population stability across the region (Schindler et al. 2010). By analyzing 50 years
1 Also see Fisheries and Oceans Canada. Commercial salmon landings in British Columbia 1951-1995. Available from:
http://www.pac.dfo-mpo.gc.ca/stats/comm/summ-somm/smon/chart-tab/index-eng.htm

2
of data, variability in sockeye salmon abundance was found to be 2.2 times lower than if the
population were defined by the simplest and most dominant life history strategy. Thus, population
diversity was found to benefit both ecosystems (by stabilizing inputs to terrestrial nutrient supplies
and food webs), and human communities (by stabilizing catch and reducing the number of fisheries
closures). This “portfolio effect” has been considered analogous to a financial investment strategy
(Schindler et al. 2010), in which the stability of a portfolio (Bristol Bay regional stock complex) is
enhanced by the diversity of its assets (river stocks and their individual spawning populations).
Fraser River sockeye salmon and its component stocks demonstrate considerable life history
diversity. For instance, stocks vary migration according to four adult run timing groups (early Stuart,
early summer, summer, and late), demonstrate 4 year cycles of abundance, and spend different
lengths of time in freshwater / at sea. Despite this underlying diversity, the abundance of Fraser
sockeye salmon is dominated by a few large stocks (e.g., Quesnel, Adams, and Chilko River runs),
which co-migrate with many smaller stocks which are often less resilient to environmental stressors.
Given this structure in abundance, it is often difficult to maximize both harvest and population
diversity. Weak stocks that are the target of conservation are often harvested when they co-migrate
with the strong stocks that are the target of the fishery. As a result, weak stocks can become
endangered (see Cultus Lake stock, COSEWIC 2003). Such differences in abundance have also led
to a disparity in our level of understanding about different Fraser sockeye salmon stocks and
individual populations. We tend to have the best information on the biggest and most economically
important stocks.
Despite their inherent resilience, the co-migration of strong and weak stocks on the Fraser illustrates
how sockeye salmon can be vulnerable. Since the late 1980s Fraser sockeye salmon have
demonstrated dramatic declines in productivity (adult recruits produced per spawner, see Figure 1).
In 2009 these declines were punctuated by a return of 1.5 million fish, well below the median pre-
season forecast of 10.6 million (DFO 2009) and below the level of productivity at which the
population can replace itself in the long term. Complicating our understanding of recent trends, in
2010 Fraser returns totalled approximately 28.6 million fish, well above the median pre-season
forecast of 11.4 million (DFO 2010) and representing the highest number of returns on the Fraser

3
since 19132. Both 2009 and 2010 returns were within the statistical distributions of forecasted returns
but at opposite ends of these distributions (i.e., < 0.10 probability of being at/or below 1.5 million in
2009; ~ 0.90 probability of being at/or below 28.6 million in 2010).
This report is focused on evaluating changes in freshwater ecology and its role in recent sockeye
salmon declines for the Cohen Commission. This specific work includes examining the status of
sockeye salmon populations and habitats, as well as the impacts of human activities on freshwater
habitats (i.e., logging, hydroelectricity, urbanization, agriculture, and mining). Fisheries and Oceans
Canada’s 36 Conservation Units form the basis for delineating and assessing the status of sockeye
salmon sub-populations in the Fraser River basin (Holtby and Ciruna 2007). Changes in habitats due
to natural and human forces are hypothesized as having three potential pathways of effects
(mirroring the organization of Section 4.6 in Peterman et al. 2010). These pathways include effects
on the: (1) quantity and quality of spawning habitats; (2) productivity of nursery lakes for rearing
juveniles; and/or (3) habitat conditions associated with smolt outmigration / adult migration3. An
integrative consideration of these pathways is valuable because variation in natural conditions (e.g.,
Burgner 1991; Quinn 2005; Bisson et al. 2009) and human-mediated disturbances (e.g., Meehan
1990; Miller et al. 1997) can interact to affect habitats and ultimately the survival and productivity of
salmon.
This work leverages and builds upon a recent and preliminary review of factors that might explain
both the low 2009 returns and longer term declines in Fraser River sockeye salmon roughly over the
last two decades (Peterman et al. 2010). The evidence in support of nine explanations was examined,
which included a consideration of changes in freshwater habitat conditions. For each explanation,
researchers attempted to relate changes in the causal mechanism to changes in productivity for 18
sockeye salmon populations. The report did not reach a definitive conclusion, though three key
findings are relevant to this evaluation. First, analyses of the indicators of sockeye salmon
productivity suggest that recent declines are likely due to mortality in the post-juvenile stage or that
a non-lethal stressor in the freshwater environment is causing mortality during a later life stage (see
Section 3.1.2 in Peterman et al. 2010). Given the timing and location of juvenile sampling for these
2 Note that pre-season forecasts of abundance are always associated with an estimate of the cumulative probability that
the actual number returning will be at or below the forecasted value. In 2009 and 2010, the reported forecasts were
associated with a 50% cumulative probability that the actual number would be at or below the forecast level.
3 Note Hinch and Martins (2011) are evaluating potential changes in freshwater habitat conditions during adult migration.

4
analyses (i.e., after several months in the lake or at the onset of smolt outmigration), it is possible
that the source of mortality could be occurring during smolt outmigration in the freshwater
environment. Next, the direction of recent trends and magnitude of declines in productivity varied
across stocks. Most of the 18 stocks showed declines (all but Harrison and Shuswap), and the
magnitude of decline was greater for those stocks that had the largest distances from the ocean to
their nursery lake (see Selbie et al. in Appendix C of Peterman et al. 2010). Differences in the
magnitude of decline across stocks was not likely explained by differences in enroute and prespawn
mortality during adult migration. Looking across all potential factors, the authors concluded it was
unlikely that a single mechanism could explain declines in productivity across stocks. Based on the
evidence it seems most likely that changes in the physical and biological conditions in the Strait of
Georgia have led to an increase in mortality during marine life stages. Specific mortality agents
include lack of food, freshwater and marine pathogens, harmful algal blooms, and other factors.
Given the level of rigour and thoroughness of Peterman et al. (2010), we believe that our conclusions
need to be compared with their findings, and if contradictory, need to be supported by a defensible
scientific rationale.
The statement of work for this project (see Appendix 1) asked us to examine and evaluate four topics
each of which form a focal objective for our work:
(1) the population and habitat status of 36 sockeye salmon Conservation Units (CUs) within the
Fraser;
(2) Fraser River sockeye salmon ecology and survival in freshwater environments;
(3) industrial and urban activities in the Fraser River and their potential effects; and
(4) the impacts of surface water and groundwater diversions on Fraser River sockeye salmon.
To address these objectives, the remaining discussion is divided into five sections:
Section 2.0 Current status of Fraser River sockeye salmon sets the context for understanding the role
and influence of human stressors in the freshwater environment by describing the current status of
populations and habitats across all sockeye salmon Conservation Units in the Fraser River basin
(addressing Objective 1).

5
Section 3.0 Freshwater stressors affecting Fraser River sockeye salmon describes the potential
mechanisms of impact and results from our examination of the evidence about the potential role of
each stressor in contributing to recent sockeye salmon declines (addressing Objectives 3 and 4).
Section 4.0 Freshwater influences on Fraser River sockeye salmon analyzes the findings from
Sections 2.0 and 3.0 to assess the significance of these issues (addressing Objective 2). We use
qualitative and quantitative analyses to explain the trends and status within each and across all
Conservation Units (to the extent possible). Due to limitations in data and time availability, we
focused our analyses on understanding landscape level changes in habitat conditions and human
stressors as opposed to the detailed cause-effect relationships influencing each Conservation Unit.
Section 5.0 State of the science provides a brief summary of the state of knowledge and data
available to describe populations, habitats, and freshwaters stressors in the Fraser River basin.
Section 6.0 Recommendations summarizes the main findings of this work and the implications for
Fraser sockeye salmon.
Figure 2 presents a conceptual model showing how human and natural factors can mitigate the
effects of potential stressors on freshwater habitats, and places this report in context with the work of
other experts working for the Cohen Commission. Section 2.2 of this report develops various
indicators to describe the pink box in Figure 2 (i.e., indicators of habitat quantity and quality, which
affects the vulnerability of watersheds to natural and human disturbances). We acknowledge that
contaminants (MacDonald et al. 2011), diseases and parasites (Kent 2011), habitat conditions in the
lower Fraser River and Strait of Georgia (Johannes et al. 2011), and changes to in-river conditions
leading to en route loss and pre-spawn mortality (Hinch and Martins 2011) might be acting
independently, cumulatively, or synergistically with the stressors considered herein. Consequently,
this report provides only a partial understanding of the influence of freshwater habitat conditions on
productivity of Fraser River sockeye salmon. A separate research project has been tasked with
investigating the full range of environmental conditions and stressors across both freshwater and
marine life stages to assess the cumulative and synergistic effect of environmental conditions and
human stressors on Fraser River sockeye salmon (see Marmorek et al. 2011).

6
2.0 Current status of Fraser River sockeye salmon
2.1 Populations
2.1.1 Background on assessing status
We have been charged with three tasks related to evaluating the current status of Fraser River
sockeye salmon populations:
(1) Summarize existing delineations of population diversity;
(2) Evaluate DFO’s methods for evaluating sockeye salmon conservation status; and
(3) Determine status of Fraser River sockeye salmon CUs.
Our first task is to summarize existing delineations of the population diversity inherent within Fraser
River sockeye salmon. These delineations are necessary to form the basis for summarizing habitat
conditions (Section 2.2), analyzing landscape level disturbances (Section 3.0), and evaluating the
relationship between changes in freshwater ecology and changes in biological production (Section
4.0). Strategy 1 of DFO’s Wild Salmon Policy (DFO 2005) deals with standardized monitoring of
wild salmon status, including a framework for delineating and assessing the conservation status of
salmon populations. DFO delineated individual salmon stocks as Conservation Units (CUs),
determined using three major axes: ecology, life history, and molecular genetics (see Holtby and
Ciruna 2007). We use DFO’s delineations of CUs to define the spatial boundaries for all sockeye
salmon stocks within the Fraser basin (see Figure 3 and Table 1). However, it should be noted that
the delineation of the CU boundaries put forward by DFO have not been subject to the peer review
process in the traditional sense of the term. It is our understanding that CU delineation will be
independently evaluated in the near future.
Our second task is to evaluate the method that DFO has developed for assessing CU status. Action
Step 1.2 of the Wild Salmon Policy requires DFO to develop criteria to assess CUs and identify
benchmarks to represent biological status. Two technical reports summarize the method that DFO
has developed (Holt 2009; Holt et al. 2009). There are a variety of published methods on how to
assess a species’ conservation status, (e.g., Musick 1999; Mace et al. 2002; Dulvy et al. 2004; Faber-
Langendoen et al. 2009), including applications to sockeye salmon (e.g., COSEWIC 2003; 2006;
Rand 2008; Pestal and Cass 2009). For this task we compare DFO’s approach (Holt 2009; Holt et al.
2009) to two alternative methodologies: one developed for Fraser River sockeye salmon (Pestal and

7
Cass 2009); and a generic approach (NatureServe) which can be tailored for different species (Faber-
Langendoen et al. 2009). We were asked by the Cohen Commission to review the qualitative method
proposed by Pestal and Cass (2009) because, similar to that of Holt 2009, the method explicitly takes
uncertainty into consideration. We chose to include the NatureServe method in our assessment
because of its popularity and ease of use. In particular, we were interested in knowing whether a
more generic approach that was less resource intensive (from a technical perspective) was
comparable to the two methods developed specifically for salmon. We did not consider additional
methods beyond the three listed because of budget and time constraints.
Each method will have different strengths and weaknesses (e.g., Porszt 2009). We use four
considerations to summarize the details underlying each approach (see Table 2): (1) ecological
criteria and indicators used for assessing conservation status (i.e., measures describing abundance,
trend, distribution, diversity, productivity, fishing mortality, and habitat condition); (2) approach for
setting benchmarks; (3) data needs and availability; and (4) feasibility of implementation. We then
summarize the overall strengths and weaknesses of each method.
Our third and final task is to determine the status of sockeye salmon CUs within the Fraser River
basin, or where infeasible due to information gaps provide a logical grouping of status with
representative CUs. DFO has not yet published a status assessment of Fraser sockeye salmon using
the method developed for the Wild Salmon Policy by Holt (2009) and Holt et al. (2009). To our
knowledge, the only assessment of Fraser sockeye salmon CUs was completed by Pestal and Cass
(2009), using two dimensions (severity of risk and uncertainty of information) to describe status.
They also used several quantitative measures (i.e., abundance, trend, and distribution relative to
benchmarks) to separate CUs into five risk categories: UNK – insufficient information; IV – status
probably poor, but little information; III – status poor, high confidence; II – status probably good,
high uncertainty; and I – status good, high confidence.
We use the work of Pestal and Cass (2009) to summarize existing status rankings for all 36 CUs (see
Table 1 and Figure 4). To assist in our overall evaluation of Fraser sockeye salmon, Table 1 also
relates the 36 CUs to the 18 stocks for which there are productivity data (as analyzed in Peterman et
al. 2010). In cases where we have evidence suggesting that these status rankings may be
inappropriate, we’ve summarized in the last column of Table 1 the direction and magnitude of the

8
change. Changes to the severity score for each CU, as determined by Pestal and Cass 2009, are based
on the work of Grant et al. (2010) (see Table 1).
Grant et al. (2010) builds on Holt et al. (2009) and is still under review. As a result, we do not
include it in our evaluations of CU status. However, the work of Grant et al. (2010) is useful for our
purposes because they determined status of all Fraser River sockeye salmon CUs, thus providing a
point of comparison with Pestal and Cass (2009)4. Key differences in these methods are that Grant et
al.’s assessment is only based on indicators of abundance and trends in abundance, whereas Pestal
and Cass (2009) also used fishing mortality and distribution in their assessment. As a result,
comparing these two approaches is best restricted to common indicators, namely abundance and
trends in abundance.
2.1.2 Evaluation of status methodologies
Before discussing the results of the status methodology evaluation, it is useful to make note of what
each method is intending to communicate (i.e., how they define status and hence what a particular
outcome means). Doing so provides an additional filter through which we can evaluate each method,
i.e., is the definition of status by which a given method evaluates conservation status appropriate for
Wild Salmon Policy purposes. The primary purpose of the NatureServe Conservation Status
Assessments is to evaluate the potential extinction or extirpation risk of elements of biodiversity,
including regional extinction or extirpation. Pestal and Cass (2009) evaluate conservation status
using a combination of status (abundance (production) and trends in abundance) and vulnerability
(productivity, diversity, fishing mortality, and distribution) risk factors. Holt et al. (2009) assess
biological status using four classes of indicators, abundance (i.e., production), trends in abundance,
distribution, and fishing mortality. The latter two methods are based more on population abundance
and trends as per their definition of status than the NatureServe method. However, Holt et al. (2009)
assessment of status is more heavily based on population biomass reference points than that of Pestal
and Cass (2009)
Ecological relevance: A comprehensive list of indicators and their respective metrics for each of the
three methods is provided in Table 2. We categorized indicators into seven indicator classes:
4 Holt et al. (2009) did not determine status for Fraser sockeye CU so it is not possible to directly compare their
assessment of status to that of Pestal and Cass (2009).

9
abundance, trends in abundance, distribution, diversity, productivity, fishing mortality, and habitat
condition. Holt et al. (2009) explicitly includes indicators from four of the seven classes, Pestal and
Cass (2009) includes indicators from all seven classes, while Faber-Langendoen et al. (2009)
includes indicators from six of the indicator classes (see Table 3).
The number of indicator classes within an assessment method, although important, does not
necessarily correlate with the strength/validity of a method. There is a trade-off between too many
and too few indicators. Using fewer indicators produces clearer recommendations, whereas use of
multiple indicators produce more consistent evaluations (Pestal and Cass 2009). In addition, it may
not be important to explicitly capture each indicator class. For example, Holt et al. (2009) do not
include metrics for all components of diversity (i.e., habitat, genetic, ecological), but they do have
distribution metrics that may be used as surrogate measures of that diversity (e.g., distribution of
spawners among habitat types, distribution in temporal trends). Given the logistic and financial
constraints around data collection it is important to consider the use of surrogate indicators where
feasible and appropriate.
The metrics associated with each indicator are very important; they determine a method’s ability to
appropriately assess conservation status in accordance with management priorities. Holt et al. (2009)
and Pestal and Cass (2009) developed metrics that reflect the unique life history of sockeye salmon,
while Faber-Langendoen et al. (2009) use more generic metrics that are applicable to a wide array of
animals. As a result Holt et al. (2009) and Pestal and Cass (2009) are better able to capture
conservation status of Fraser River sockeye salmon and CU specific vulnerabilities than Faber-
Langendoen et al. (2009). Both Holt et al. (2009) and Pestal and Cass (2009) have good
representation of the various indicator classes. One advantage of the approach taken by Pestal and
Cass (2009) is that it explicitly considers habitat condition. This allows the method to be more
proactive in its management applications (i.e., management actions related to poor habitat quality can
be implemented before the poor habitat condition translates into a population effect).
On a separate note, Grant et al. (2010) do not include distribution metrics in their assessment
method. In our opinion this is a substantial oversight because Fraser River sockeye salmon
conservation status and population viability within a CU is a product of spatial distribution, habitat
condition, and abundance, not population abundance in and of itself. Furthermore, there is a

10
possibility that assessments carried out using Grant et al.’s (2010) method in it current version may
become irrelevant or outdated following a COSEWIC assessment, which would weigh distribution
indicators heavily in its evaluation of status.
Approach for setting benchmarks: Faber-Langendoen et al. (2009) take a qualitative approach to
setting benchmarks which is not consistent across metrics or across evaluation units (in this case
CUs). The method uses points along a continuum of extinction risk, rather than break points and
thresholds. All metrics are on a scale of 0 to 5.5, with equal contribution to an aggregate score;
however different metrics may have different numbers of increments within the 0 to 5.5 scale.
Another inconsistency that can arise across CUs is that metrics within indicator classes are weighted
differently (see Table 8 in Faber-Langendoen et al. (2009)) as are the indicator classes themselves.
The method has no rules that can be applied across CUs to standardize the assessment process which
has the potential to result in inconsistent CU evaluations (i.e., each CU is evaluated using different
metrics, benchmarks, and weightings).
Pestal and Cass (2009) take a qualitative approach to setting metric specific benchmarks; however
they employ clear and consistent rules for benchmark setting and indicator roll-up across all CUs to
ensure that results are comparable. In addition, benchmarks are based in part on the magnitude of
uncertainty in observed spawner data. As a result the method for setting benchmarks is defensible
and transparent. A major advantage of the qualitative approach is that it is not as data intensive as the
quantitative approach taken by Holt et al. (2009). Consequently, benchmarks for each metric can be
set for all CUs irrespective of data availability. Despite these more modest information requirements,
11 of 36 Fraser sockeye salmon CUs still did not have enough information for risk characterization
(Figure 4).
Holt et al. (2009) take a quantitative approach to setting benchmarks for abundance and trends in
abundance metrics, but take a similar approach to Pestal and Cass (2009) for distribution metrics.
Concrete benchmarks for distribution metrics have yet to be finalized by DFO. Where the
quantitative approach is taken, uncertainty in the data is explicitly incorporated into benchmark using
Monte Carlo simulations, thus making them robust and defensible.

11
Data needs: A list of data required by each assessment method is provided in Table 2. The method
developed by Holt et al. (2009), and by extension Grant et al. (2010), is the most data intensive.
Consequently, it was not possible to evaluate a number of CUs with this method because of
insufficient data. Pestal and Cass (2009) and Faber-Langendoen et al. (2009) are not as data
intensive. It was possible to evaluate conservation status for a greater number of CUs using Pestal
and Cass’s method (CUs were not evaluated using Faber-Langendoen et al. (2009)). In general, there
are few data on spatial distribution of Fraser sockeye salmon and habitat condition within a CU; this
affects all assessment methods to some degree.
Feasibility: By feasibility we mean the ability to effectively implement a conservation assessment
method. Both Pestal and Cass (2009) and Holt et al. (2009) are robust methods that are scientifically
defensible. They are both geared specifically towards salmon. However, they differ substantially in
their feasibility of implementation given DFOs current operating budget and resources (i.e.,
personnel, historical data availability, monitoring capacity, etc.). We believe that in the short term
it’s more feasible for DFO to implement the method of Pestal and Cass (2009) than to use the
approach of Holt et al. (2009). This conclusion simply reflects the more qualitative nature and less
stringent data requirements of Pestal and Cass (2009). Should DFO acquire the necessary resources
and capacity to collect the information required by Holt et al.’s method, it would be logical to to
switch to this more quantitative and rigorous method. From a management perspective, it is better to
have qualitative assessments for many CUs (i.e., Pestal and Cass 2009), than to have very few CUs
assessed due to data limitations (i.e., Holt et al. 2009). It is possible that elements of Pestal and Cass
(2009) could be incorporated into Holt et al. (2009) to address data poor CUs.
Conservation Status: Of the 36 Fraser River sockeye salmon CUs, conservation status could not be
assessed by Pestal and Cass’ (2009) in 11 of 36 cases because of insufficient data, resulting in an
uncertainty score of 10 (maximum value) for each of these CUs (see Figure 4). The two sockeye
salmon management groups with the highest level of uncertainty in the data (are early summer (ES)
and river-type sockeye salmon (see Table 1). As previously mentioned, Holt et al. (2009) did not
assess conservation status for any CU; however, Grant et al. (2010) did a partial assessment of status
for 26 CUs using information on abundance and trends in abundance.

12
Combining the work of Pestal and Cass (2009) and Grant et al. (2010), we recommend modifying the
severity score (i.e., condition of the population) for 9 CUs (see Table 1). The difference between the
modified severity score and the original provided by Pestal and Cass (2009) is graphically illustrated
in Figure 5. All modifications are relatively minor and consist of moving the severity score up or
down by half a point. In no cases does the modification alter the severity category (i.e., quadrant) in
which the CU was classified. A more structured and analytical comparison of the two methods (i.e.,
Holt et al’s (2009) method to that of Pestal and Cass (2009)) would be very useful to DFO for WSP
implementation because it would allow them to see the extent to which the methods provide similar
results and they could then evaluate the tradeoffs between the two. However, this would only be
worthwhile if DFO is planning to move forward with Holt et al’s (2009) method. If Grant et al’s
(2010) modified method is favoured over that of Holt et al. (2009), the rigorous comparison should
take place between Grant’s and Pestal and Cass’ methods.
2.2 Habitats
The majority of Fraser River sockeye salmon populations rear in large lakes for their first year of
life. Because their rearing lakes are generally large, they are considered to be more buffered from
impacts compared to small streams and rivers that are used by other salmon species. However, most
of the total egg-adult mortality in sockeye salmon occurs in the freshwater stage of the life cycle.
Bradford (1995) estimated that 58% of the total mortality occurred in the natal spawning and rearing
areas, and there is additional mortality during migration from natal lakes to the ocean (Welch et al.
2009). Consequently, there is potential for impacts on sockeye salmon populations from
development activities (e.g., logging, road construction, dams, etc.) if they significantly affect the
quantity or quality of lake, stream, or river habitats used by different life stages.
As most sockeye salmon populations are associated with lakes this can lead to isolation, highly
specialized adaptations, and a stronger degree of population genetic differentiation than is found in
other Pacific salmon (Burgner 1991; Wood 1995). Although lake-type sockeye salmon populations
can be enormously productive these specialized habitats may also be vulnerable to environmental
changes, which could move them beyond the range to which local populations are especially well
adapted (Waples et al. 2009). Conversely, river-type sockeye salmon are more generalized in their
habitat requirements and only weakly differentiated by genetic markers (Wood 1995; Wood et al.
2008). Being highly specialized, there have been suggestions that lake-type sockeye salmon

13
populations could be considered evolutionary dead ends (Wood 1995; 2007; Wood et al. 2008).
Though relatively rare, river-type populations are more likely to stray from their natal stream to
spawn and colonize new habitats (but see Pavey et al. 2007). While this flexibility would indicate
that river-type sockeye salmon could be important in conferring some greater overall resilience to the
species as a whole, there is unfortunately little information available on the status of Fraser river-type
populations (i.e., status only available for Widgeon and Lower Fraser CUs in Pestal and Cass (2009))
or the condition of their river habitats (Birtwell et al. 1987; M. Bradford, Fisheries and Oceans
Canada, pers. comm.).
For lake rearing sockeye salmon, measures of freshwater habitat condition are often not currently
available or can be difficult to directly and reliably quantify (J. Hume and D. Selbie, Fisheries and
Oceans Canada, pers. comm.), even though Strategy 2 of the Wild Salmon Policy is charged with
developing relevant habitat metrics. For example, it may be possible to define the extent of potential
spawning reaches but it is more difficult to define actual quality of spawning substrates. To assist our
evaluation of freshwater ecology we developed direct and (where required) surrogate landscape level
indicators of the quantity and quality of migration, spawning, and rearing habitats for each sockeye
salmon lake-type CU using: (1) mapped habitat features we extracted or derived from readily
available GIS data, and (2) lake productivity datasets provided to us by Fisheries and Oceans
Canada. Data were not available that could be used to reliably describe basic habitat conditions for
the six river-type CUs.
Migration: We determined the migration route and distance for each sockeye salmon lake CU by
developing a connected hydrology network that allowed us to trace a path from the outlet of each
CU’s nursery lake to the mouth of the Fraser River. We also defined a 1 km buffer along each
migration corridor to represent the “zone of influence” within which we could assess the extent of
human stressors. While this 1 km distance is arbitrary, we selected this buffer width to ensure that we
capture the most important potential stressors within the riparian zones of both small and large rivers.
This buffer width is substantially larger than the distance (typically less than 30 m) used to protect
riparian zones of small streams from the effects of forest harvesting across the Pacific Northwest
(Young 2000; Stalberg et al. 2009). We used a significantly larger width to ensure that that we
captured the effect of activities that have a longer spatial extent than forest harvesting, and to ensure
that we captured the effects of riparian disturbance along large rivers, such as the Fraser.

14
Spawning: We defined sockeye salmon spawning areas within each CU using GIS depictions of
spawning extent that are available from the government-maintained Fisheries Information Summary
System (FISS). Expert opinion and knowledge of spawning areas then allowed us to assign reaches
according to whether they were lake influenced (lake outlet / large mainstem) or not (lake inlet /
tributary stream). We used GIS to delineate “zones of influence” for these two types of spawning
reaches. For lake inlet / tributary reaches we defined the extent of upstream influence by delineating
the total area of the local watershed upstream of the spawning reach. Such tributary stream spawning
habitats can be exposed to a variety of human land use impacts immediately adjacent to spawning
sites and/or from the more cumulative effects of activities upstream of the spawning zone occurring
within the local watershed (Harr and Fredriksen 1988; Hicks et al. 1991; Stednick and Kern 1994).
Impacts of land use on spawning habitat are reduced by the presence of a lake upstream of a
spawning area because lakes act as sediment traps, reduce flow variation, and reset temperatures in
their outlet streams (Arp et al. 2006; Myers et al. 2007; Jones 2010). By contrast, lake outlet /
mainstem spawning reaches were considered buffered from upstream effects and the “zone of
influence” for these was described simply as the area within a 1 km buffer along either bank of the
spawning reach.
Rearing: We identified all nursery lakes within each CU and calculated their combined total area
using GIS. We also defined the upstream “zone of influence” by delineating the area of all
watersheds upstream of each CU’s nursery lake. Sockeye salmon nursery lakes, with larger dilution
volumes and more varied habitat, can better buffer sockeye salmon from land use impacts during
rearing, relative to streams (Selbie et al. 2010). Nursery lake habitats will not be completely immune
to broader landuse impacts, however, and human activities upstream of the nursery lake(s) may have
cumulative effects on lake conditions (Anderson 1974; Klock 1985).
The resulting “zones of influence” on migration, spawning and rearing areas were used as the base
layers for calculating the specific indicators of habitat conditions (Table 4) and against which to
intersect the human stressor layers so as to understand the potential interactions between human
activities and these habitats (see Section 3.0). Table 5 presents data on these indicators for all lake
CUs. The sections that follow describe the biological relevance of the individual habitat indicators.

15
2.2.1 Indicators of migratory habitat quantity / quality
Migration distance: Lengthy migrations can increase stress and the exposure to pre-spawning
mortality factors of adult sockeye moving upstream (Crossin et al. 2004), or plausibly affect
mortality of smolts during downstream migration or fitness once they reach the ocean. The
relationship found by Selbie et al. (in Peterman et al. 2010) suggests that there may be an underlying
biological mechanism related to upstream or downstream migration that is differentially affecting
survival of sockeye salmon stocks across the Fraser basin (beyond changes in en-route mortality).
For instance, migration distance is related to migration timing (i.e., earlier timing required to cover
longer distances) and the length of time spent in freshwater (i.e., stocks with longer migrations will
have a larger chance of exposure to harmful stressors, including diseases, parasites, contaminants,
and high water temperatures). We used migration distance as a surrogate for these more direct
biological indicators given the relationship found by Selbie et al. (in Peterman et al. 2010) and
because we could measure migration distance consistently across all CUs.
Thermal profile of adult migration (°C): Adult spawning success may be affected by water
temperatures encountered during migration. High water temperatures can increase fish energy
expenditures, increase the progression of diseases and parasites, and decrease fecundity of eggs
(Crossin et al. 2008). Fisheries and Oceans Canada suggests that Fraser River water temperatures
above 18-20oC can degrade spawning success, while water temperatures above 24oC can be fatal5.
Although water temperature is largely driven by natural processes it can be influenced by activities
such as riparian clearing, water withdrawal and diversion, logging, and other changes to land cover
(Poole and Berman 2000), as well as global warming (Martins et al. 2011). Given the well known
and strong relationship between water and air temperatures (Stefan and Preud'homme1993; Poole
and Berman 2000; Morrison et al. 2002; Cooke et al. 2008; Voss et al. 2008) we used mean
summer/fall air temperature as a surrogate indicator of the quality of the migratory corridor during
the period of adult migration. We identified the migration dates for the run-timing group (see Table
6) associated with each CU and calculated the average monthly air temperatures that each CU would
have experienced during their upstream migrations (both for the full historical time series and also
for the more recent average temperature beginning in 1985).
5 Fisheries and Oceans Canada. Predicting the Temperature of the Fraser River. Available from: http://www.pac.dfo-
mpo.gc.ca/science/oceans/fleuve-Fraser-river/index-eng.htm

16
Thermal trigger for smolt outmigration (°C): Research has shown that the timing of smolt
outmigration can have an important influence on the survival of juvenile salmon to adulthood
(Scheuerell et al. 2009). The belief is that changes in timing of smoltification can lead to a mismatch
between the timing of arrival at the estuary and the timing of other conditions (e.g., food supply)
which ultimately affects survival. Like all life history transitions, sockeye salmon smolts are cued to
migrate towards the ocean in response to changing environmental conditions, which includes
responding to day length, lake springtime temperatures (related to the timing of ice break-up in
nursery lakes), and springtime peak flows (Foerster 1937; Burgner 1991; Quinn 2005). Not
surprisingly these variables are strongly associated with local climate conditions and geographic
location of the nursery lake, such that more northern and colder nursery lakes tend to have later
migration dates (see Figure 6A). Moreover, within a nursery lake, the timing of outmigration is
affected by year to year variations in climate conditions (see Figure 6B). Given this relationship, and
the known link between air temperatures and lake water temperatures (Sharma et al. 2008), we used
springtime air temperature as an indicator of the timing of ice break-up in nursery lakes (one of the
cues of smolt outmigration). We used this indicator as a surrogate of the potential for mismatch
between the timing of arrival at the estuary and timing of other conditions in the estuary, not a direct
measure of the magnitude of mismatch. Our hypothesis is that if lake ice breaks up significantly
earlier than experienced historically, there is potential for earlier smolt outmigration and greater
potential for a mismatch. A consideration of the changes in spring time temperatures at nursery lakes
seems justifiable given evidence that the timing of ice break up on lakes in Canada is becoming
earlier (IceWatch no date).
2.2.2 Indicators of spawning habitat quantity / quality
Total spawning extent (m) and ratio of lake influence to total spawning: The total length of stream
spawning habitat available to sockeye salmon will determine the scope of opportunities for
successful spawning (longer length is better), while the ratio of lake-buffered outlet spawning to
tributary/inlet spawning can indicate the degree of sensitivity to upstream impacts (higher ratio is
better). Lakes stabilize discharge by buffering flood effects, thereby reducing stream bank erosion
and bedload movement compared to streams with more variable discharge regimes (Montgomery et
al. 1996). Thus, spawning habitat quality and egg-to-fry survival should be less affected by
disturbance in spawning channels buffered by lake influences than in small, non-lake moderated
tributaries (Chapman 1988; Northcote and Larkin 1989; Montgomery et al. 1996). Sockeye salmon

17
also spawn along lake shores but mapping of the location and extent of these spawning zones is
fairly limited (Stalberg et al. 2009; Brad Mason, Fisheries and Oceans Canada, pers. comm.).
However, lake spawning is considered of relatively minor importance (except for Cultus Lake) with
most sockeye salmon populations in BC spawning within streams (Roberge et al. 2001).
2.2.3 Indicators of rearing habitat quantity / quality
Nursery lake area and productivity (measured and estimated): The Fraser River is the world’s largest
single producer of sockeye salmon, being surpassed only by the combined sockeye salmon
production from several river systems flowing into Bristol Bay in Alaska (Northcote and Larkin
1989). The Fraser system’s exceptionally high productivity is due to the presence of many large
lakes (66% of B.C.’s nursery lake area) that are accessible to anadromous fish (Hume et al. 1996;
Shortreed et al. 2000). Further, most of these lakes are sufficiently productive to sustain a
zooplankton community considered capable of supporting high juvenile sockeye salmon densities
(Stockner and Shortreed 1983). It is generally assumed that most Fraser sockeye salmon stocks are
recruitment limited (Hume et al. 1996; Shortreed et al. 2000) with freshwater rearing habitats often
capable of supporting juvenile sockeye salmon densities far higher than presently occur (i.e., a
greater number of spawning escapements would produce additional smolts)6. However, most BC
nursery lakes are naturally oligotrophic or even ultra-oligotrophic (Stockner and Shortreed 1994) and
strongly nutrient-limited (Shortreed et al. 2000). Consequently the continued long term reduction in
marine-derived nutrients from anadromous spawner carcasses has likely resulted in the further
oligotrophication of lakes and streams, with a corresponding reduction in their productive capacity
(Shortreed et al. 2001).
Juvenile sockeye salmon production from nursery lakes can be measured directly either from smolt
counts (i.e., from counting fences) or from counts of summer and fall fry (i.e., based on pelagic fish
surveys using hydroacoustics and midwater trawling, J. Hume, Fisheries and Oceans Canada, pers.
comm.). Long-term time series data exists only for Chilko Lake (smolts), and Quesnel and Shuswap
lakes (fry), with limited fry data existing for another 13 lakes (J. Hume, Fisheries and Oceans
Canada, pers. comm.). Sockeye salmon fry are almost exclusively linmetic plankitivores, however,
so they are strongly coupled to limnetic zooplankton production (Shortreed et al. 2000). Given their
6 Though we also acknowledge an opposing density dependent hypothesis which suggests that escapement may already
be too high in some rivers (see Section 4.7 in Peterman et al. 2010).

18
use of lake habitats, it is possible to estimate the quantity and quality of sockeye salmon rearing
habitat in BC from lake size and measures of lake productivity such as photosynthetic rate (PR)
(Hume et al. 1996; Shortreed et al. 2000). Lake area is also considered a reasonable surrogate of
habitat productivity since it is a primary driver in productivity relationships (Randall 2003). While
fry models provide a direct estimate of rearing capacity, many years of data are required to generate
a relationship for any lake. The PR model appears to be a useful predictor of rearing capacity, and
predictions can be made after only 1–2 years (Hume et al. 1996). PR-based estimates of productivity
have been proposed for habitat benchmark setting within DFO’s Wild Salmon Policy (Stalberg et al.
2009; D. Selbie, Fisheries and Oceans Canada, pers. comm.). Correlations between PR and juvenile
sockeye salmon abundance can be used to estimate the maximum capacity of a nursery lake to
produce smolts (biomass and numbers), as well as estimate optimum escapement (Shortreed et al.
2000). At least one year’s PR data now exists for 19 sockeye salmon nursery lakes in the Fraser
drainage (J. Hume, Fisheries and Oceans Canada, pers. comm.) representing about 80% coverage of
all Fraser nursery lakes. Smaller sockeye salmon nursery lakes (such as Taseko, Nahatlach, and
McKinley) have not been evaluated because of funding constraints and the prioritization of
evaluations towards lakes which produce more sockeye salmon (J. Hume, Fisheries and Oceans
Canada, pers. comm.). Given the broader availability of juvenile production estimates based on PR
vs. direct monitoring we have used DFO’s PR-based estimates as a comparative measure of juvenile
productivity across sockeye salmon CUs.
2.2.4 Integrated summary of habitat vulnerability
Given a general lack of information that could be used to reliably define dynamic changes in
condition across sockeye salmon spawning, rearing, and migratory habitats we defined relative
sockeye salmon CU habitat “status” as a combination of the: (1) intrinsic habitat vulnerability (see
below) and (2) intensity of human stresses on those habitats (Section 3.0). With our approach, a CU
that was considered highly vulnerable (relatively more sensitive) to potential habitat impacts, while
also exposed to relatively high human development pressures within its spawning, rearing and/or
migratory habitats, would be considered as having a relatively poor habitat status. Conversely, a CU
with limited vulnerability (relatively less sensitive) and relatively little human development pressure
would be considered as having a relatively good habitat status. We stress that these are only
relative indices. Even those CUs considered to have relatively poor habitat status by the above

19
procedure may not have any demonstrated actual negative impacts of human stressors on sockeye
salmon freshwater survival.
To determine intrinsic habitat vulnerability, we believe reporting out on the large number of sockeye
salmon habitat indicators presents a challenge in providing an overall and simplified assessment. A
possible approach would be to provide a single score using a ‘habitat index’ that integrates several
indicators. This can be easy to interpret, but information will be lost and there may be multiple
approaches to aggregating indicators without certainty about which is best. Should all indicators be
weighted equally? If not, how should they be weighted? Each approach to summarizing habitat has
strengths and weaknesses. For example, agency programs that currently monitor watershed condition
in the Canadian and U.S. Pacific Northwest (e.g. FREP, EMAP, AREMP, PIBO) use a variety of
methods to aggregate their habitat data (Pickard et al. 2008). No single approach has been widely
accepted.
To provide an integrated summary of habitat vulnerability we chose not to develop a single index
score. While also reporting out (to the extent possible across CUs) on our full suite of habitat
indicators, we instead focused on presenting three independent, static indicators that could be
calculated for each CU and that we felt could best define intrinsic habitat vulnerability for each
sockeye salmon freshwater lifestage. These independent habitat indicators are: (1) migration
distance; (2) total area of nursery lakes; and (3) ratio of lake influence to total spawning extent. First,
we chose migration distance as a surrogate indicator for the relative exposure of each CU to a suite
of cumulative (but difficult to directly quantify) natural and human stresses that sockeye salmon will
experience while migrating upstream as adults and downstream as smolts. Next, we chose lake area
to represent a simple, surrogate indicator that could identify the intrinsic capability of a CU’s rearing
habitats to produce large numbers of smolts. We recognize that incorporation of direct or modelled
estimates of juvenile productivity could provide a description of current lake habitat condition.
However such productivity estimates are only available for a subset of CUs (and generally lack time
series data) while nursery lake area is available for all CUs and is highly correlated with total
juvenile production. While annual lake-to-lake differences in productivity per unit area are
important, the extent of the rearing habitat available will more strongly dictate the potential total
smolt production from a CU. Lastly, we chose the ratio of lake-influenced spawning to total
spawning reaches as a surrogate indicator of the long term ability of a CU to consistently maintain

20
good quality spawning habitat. Spawning reaches below lake outlets will be buffered from natural
and human watershed impacts that can affect substrate quality, water quality, and flows.
The three independent measures of habitat vulnerability are represented by separate axes within 3-
dimensional figures (see Figure 7). The placement of an individual CU across these dimensions
illustrates its vulnerability to watershed disturbances relative to other CUs in the Fraser River basin.
Figure 7 illustrates the relationship among these variables across all CUs, while the dashboard
summaries (see Appendix 3) highlights the values of habitat vulnerability for each CU.

21
3.0 Freshwater stressors affecting Fraser River sockeye salmon
To understand the potential role of freshwater stressors in recent declines of sockeye salmon we
compile and analyze the best available data describing six categories of human activities which have
the potential to affect sockeye salmon: forestry (e.g., forest harvesting activities, Mountain Pine
Beetle disturbance, and log storage), mining, hydroelectricity (e.g., large scale and run of river power
projects), urbanization upstream of Hope7, agriculture, and water use. These activities have the
potential to affect sockeye salmon during freshwater life stages, in particular, by having effects on
the: (1) quantity and quality of spawning habitats; (2) productivity of nursery lakes for rearing
juveniles; and/or (3) habitat conditions associated with smolt outmigration / adult migration. To
examine the interaction between a stressor and sockeye salmon habitats, we intersect each spatial
layer representing a stressor (described below) with the spatial layers delineating “zones of
influence” on core habitats for migration, spawning, and rearing across each Conservation Unit
(described in Section 2.2). We recognize that not all stressors and pathways of effects on habitat are
plausible, however. Table 7 summarizes the linkages between each stressor and habitat that we
examined in our analyses and the indicators being generated to represent these stressors. The context
for a stressor, related mechanisms of effect, and results from our analysis are described in more detail
in the following sections. The role of these stressors in declines of Fraser sockeye salmon is
discussed in Section 4.0.
3.1 Forestry
In this section we investigate three topics related to forestry that have had the potential to affect
Fraser River sockeye salmon: (1) forest harvesting activities, (2) Mountain Pine Beetle disturbance,
and (3) log storage / handling on the Fraser River estuary.
3.1.1 Forest harvesting activities
In the Fraser River basin, 75% of the land area is covered by forests and as a result forestry is an
important contributor to rural economies (FBC 2009). Since the early 1970s the area and volume of
forested land that has been harvested annually in British Columbia has varied, but remained
relatively stable (Figure 8). The Southern Interior and Central Interior Ecoprovinces (which overlap
with the majority of sockeye salmon watersheds) have some of the highest cumulative concentrations
7 See Johannes et al. (2011) for an evaluation of the impacts of urbanization downstream of Hope.

22
of roads in the province. From 2000 to 2005, in the Southern Interior and Central Interior regions the
density of road-stream crossings increased by 21% and 10%, respectively, with a similar level of
increase in the density of roads – 18% and 10%, respectively (BC MOE 2008). Accompanying the
economic benefits of forest harvesting activities are known impacts on fish and habitats, including
effects on sockeye salmon (e.g., Salo and Cundy 1987; Meehan 1991; FPB 2009; Smerdon et al.
2009; Daigle 2010). The state of knowledge about the interaction between forests and fish is based
on years of rigourous research in watershed studies across British Columbia and elsewhere in the
U.S. Pacific Northwest (e.g., Carnation Creek (Hartman et al. 1996), Queen Charlotte Islands (Hogan
et al. 1998), Stuart-Takla (MacIsaac 2003), and Slim Creek (Brownlee et al. 1988)). Though sockeye
salmon have not always been the focal species in these kinds of studies, the mechanisms of impact
and implications of forestry development on sockeye salmon can be inferred with a relatively high
degree of confidence from studies investigating fish-forestry interactions.
Three core forest harvesting activities can have a potential impact on sockeye salmon habitats and
survival at different life stages (reviewed by Meehan 1991). Road construction interferes with the
natural patterns of water flow through a watershed as water drains across exposed road surfaces,
which can increase sediment inputs into streams. Sedimentation can cover spawning redds and
reduce oxygenation of incubating eggs. Road-stream crossings can also interfere with access to
habitats by adult spawners when these crossings pose obstructions to fish passage. Upslope
harvesting can alter the hydrology of a watershed which affects the delivery of water and gravels
throughout the stream network. These alterations can affect the amount and timing of water and
sediment available to streams and lakes thereby leading to impacts on spawners, eggs, and/or
juveniles. Activities in riparian areas can additionally affect water quality by disturbing stream bank
integrity, reducing watershed nputs of nutrients and woody debris, and increasing stream
temperatures through reduced streamside shading. These changes have the potential to affect the
growth and survival of eggs and juveniles.
Coastal and interior B.C. watersheds respond differently to the impacts of logging (Winkler et al.
2009). Along the coast, steeper topography and more intense storms can cause more severe
landslides than in interior watersheds, bringing large amounts of sediment and debris into streams.
However, the steep topography and flashy hydrology of coastal watersheds also moves sediment and
woody debris more quickly downstream, leading to a shorter recovery time from logging impacts

23
than for interior basins. The level of impacts on stream hydrology, sediment transport and fish
habitat can change significantly with the percent of the watershed that is clear cut (Grant et al. 1986;
Chamberlin et al. 1991; NCASI 2001; MOF 2001; McCaffery et al. 2007; Carson et al. 2008).
To examine these issues we quantitatively assess the geographic extent and history of forest
harvesting activities (primarily forest harvesting and road building) across all sockeye salmon
Conservation Units in the Fraser River basin. We compile three spatially explicit layers of
information to examine this interaction (see Appendix 4). The Reporting Silviculture Updates and
Land Status Tracking System (RESULTS) database provides an indication of the extent of forest
harvesting; the British Columbia Digital Road Atlas summarizes the extent of all roads (including
roads associated with other human activities, see sections below); and a road crossing database
identifies road-stream crossings of potential concern. Given the nature of these underlying data,
salvage harvesting and related activities associated with Mountain Pine Beetle are included here in
these spatial layers.
An examination of the correlation among the indicators derived from these data sources revealed
strong correlations between the magnitude of recent harvesting and Mountain Pine Beetle
disturbance (see Section 3.1.2) and the density of roads and road-stream crossings, while the
magnitude of forest harvesting and road development are largely independent of each other.
Figure 9 through Figure 11 illustrate the intensity, spatial extent, and temporal trends of forest
harvesting across all lake sockeye salmon Conservation Units. The level of forest harvesting within
the last 15 years is less than 10% of the “zones of influence” on habitat types, though varies widely
across CUs and habitat types (Figure 9). Drainage areas upstream of lake inlet spawning, tributary
spawning, and nursery lakes tend to be more heavily disturbed than the riparian zones adjacent to
spawning downstream of lakes or along migration corridors. The most heavily disturbed headwaters
include the following Conservation Units which have had more than 7.5% of their headwaters
disturbed within the last 15 years: Early Summer – Pitt, Nahatlatch, Fraser, Francois; Summer –
Fraser, Francois, Stuart; Late – Harrison (D/S). In contrast, the headwaters of the following CUs
have been mostly undisturbed within the last 15 years: Early Summer – Chilliwack, Chilko, Taseko
Summer – Chilko; Late – Cultus, Kawkawa. The distribution of the level of harvesting shows
indiscernible changes over time when viewed at the scale of the Fraser River basin (Figure 10),

24
although the pattern of disturbance over time has varied across CUs (Figure 11). In this illustration,
some Conservation Units have shown declines in the extent of harvesting (Early Summer – Pitt; Late
– Harrison (D/S), Shuswap Complex), while others (Early Stuart – Stuart; Summer – Quesnel) have
shown a recent and sharp increase, which is likely due to increases in salvage harvesting associated
with Mountain Pine Beetle disturbance (see Section 3.1.2).
Figure 12 and Figure 13 represent the intensity of road development as measured by density of roads
(across all development activities) and road-stream crossings. As mentioned these stressor indicators
are highly correlated with each other which suggests they represent the same relative level of
disturbance associated with roads across CUs. However, there is considerable variation in road
development across CUs and habitat types. In contrast to forest harvesting, road development tends
to be more concentrated in areas adjacent to spawning zones downstream of lakes and along
migration corridors. CUs with high road densities in lake influenced spawning zones include: Early
Summer – Kamloops; Summer – Francois, Fraser; Late – Kawkawa, Seton, Kamloops. CUs with the
highest road densities along migration corridors include the: Early Summer – Chilliwack, Pitt,
Shuswap Complex; Late – Kawkawa, Seton, Cultus.
3.1.2 Mountain Pine Beetle disturbance
The current Mountain Pine Beetle outbreak in the interior of British Columbia has expanded at an
unprecedented rate and is considered the largest in the province’s recorded history (reviewed by
McGarrity and Hoberg 2005). The interior pine forests of the Fraser River basin have been the focal
point for this disturbance agent. Two main factors have contributed to the outbreak. First, the supply
of mature lodgepole pine has increased in recent decades due, in part, to fire suppression activities.
Second, warmer winters have increased the survival of beetles, which has helped beetle populations
flourish. During the initial stages of the outbreak, the management response focused on containing
the infestation. As the scale, magnitude, and severity of disturbance increased it became increasingly
clear that containment was not possible. Consequently, management efforts shifted to salvage
logging to minimize economic losses from the outbreak; in some regions allowable annual cut was
increased 78% above pre-outbreak levels (McGarrity and Hoberg 2005). The outbreak has caused a
major disturbance to forest ecosystems and watersheds, and therefore has the potential to be a major
direct stressor on sockeye salmon of the Fraser River basin (Johannes et al. 2007). As well, the fish

25
habitat impacts sometimes associated with forest harvesting could be exacerbated with greater
fractions of clear-cut area under salvage logging.
Our understanding of the effects of Mountain Pine Beetle on watersheds is riddled with uncertainties
given the unprecedented magnitude / spatial extent of the outbreak, relatively short time frame within
which the outbreak has emerged, and relatively limited research on the topic to date. Key
uncertainties include a lack of understanding about how (Uunila et al. 2006):
 beetle-induced hydrologic changes vary across watersheds, climate conditions, and recover
through time;
 standing dead timber affects hydrologic processes compared to salvage logging;
 the magnitude of summer low flows will change (i.e., increase or decrease); and
 rates of hydrologic recovery differ between beetle-affected and salvage logged watersheds.
Despite these unknowns, our current understanding is grounded in some certainties (Uunila et al.
2006; EDI 2008; Redding et al. 2008). In general, the effects of beetle disturbance on hydrologic
processes are different than timber harvesting because affected forests retain standing timber and
understorey vegetation. Hydrologists generally agree that the resulting defoliation of pine forests
leads to a decrease in interception of precipitation and loss of transpiration, which increases the
amount of water in soils and in turn affects surface water and groundwater supplies. The loss of
forest canopy will also affect the accumulation of snow and rates of snowmelt. These changes are
expected to lead to an increase in total water yields and higher peak flows (e.g., FPB 2007). Most
recent research indicates that hydrological processes within beetle-affected stands are somewhere
between a mature forest and clearcut, with hydrologic recovery taking between 20 and 60 years.
Even-aged stands lacking understory tend to have greater impacts on hydrologic processes than
uneven-aged stands with understorey vegetation. More detailed changes in the magnitude and timing
of streamflow are difficult to predict, however, given the complexity of factors governing hydrology
(e.g., elevation, topography, type of vegetation cover, weather patterns). For instance, some
empirical evidence suggests that is difficult to detect beetle-related changes in hydrology and that the
direction of change can vary across watersheds (Stednick and Jensen 2007).

26
From the perspective of impacts on sockeye salmon watersheds, increased soil water and streamflow
can lead to decreased slope stability, increased flooding, and alterations in the quality and quantity of
freshwater habitats. In particular, the combined effects of beetles and salvage logging on watershed
hydrology will affect the delivery of water and gravels, which can affect the amount of water
available for spawning / rearing, sedimentation of streams and lakes, and consequently affect
spawners, eggs, and/or juveniles.
The above-described effects will be most evident in years with intense storms. For example, Schwab
(1998) estimated that three quarters of all the sediment delivered during the twentieth century in
watersheds of the Queen Charlotte Islands occurred during just four storms. While the Fraser River
basin generally has both more gently topography and less rainfall than the Queen Charlotte Islands,
the intensity of storms is expected to increase with global warming (Spittlehouse and Murdock
2010), which could exacerbate the effects of beetle kill and salvage logging.
Similar to our analysis of harvesting disturbance, we assess the geographic extent and recent
expansion of Mountain Pine Beetle as a disturbance agent across interior sockeye salmon
watersheds. Due to the nature of the available data, impacts associated with salvage harvesting are
described in the analyses of section 3.1.1. To examine the time series of disturbance for the recent
outbreak (since 1999), we compiled the Forest Health Network Archives Pest Data for British
Columbia from the Canadian Forest Service and Forest Health from the Aerial Overview Surveys
from the Ministry of Forests and Range (see Appendix 4).
Mountain Pine Beetle is a dominant disturbance agent across most CUs and all habitat types. Figure
14 through Figure 16 illustrate the intensity, spatial extent, and temporal trends of disturbance due to
Mountain Pine Beetle across all lake sockeye salmon Conservation Units. In comparison to the
intensity of forest harvesting (maximum disturbance of ~10% by area), the magnitude of Mountain
Pine Beetle disturbance across CUs and habitat types has been much greater (maximum disturbance
of ~90% by area, see Figure 14). Watersheds upstream of and adjacent to sockeye salmon spawning
(both lake influence and tributary spawning locations) and nursery lakes tend to be more heavily
disturbed than migration corridors, though some corridors still have a substantial level of disturbance
(up to 30%). As has been well documented, the expansion of Mountain Pine Beetle within the last 10
years has been dramatic and now affects land cover across most interior Fraser CUs (Figure 15).

27
Since 1999 the level of disturbance increased dramatically after 2003 (Figure 16), with the level of
disturbance being most dramatic in interior Fraser CUs (e.g., Early Stuart – Stuart; Early Summer –
Taseko; Summer – Quesnel) as opposed to coastal CUs whose watersheds are largely absent of
ponderosa and lodgepole pine, the beetle’s host tree species (e.g., Early Summer – Pitt; Late –
Harrison (D/S)).
3.1.3 Log storage / handling in the Fraser River estuary
The history of transportation and storage of logs along major waterways in the Pacific Northwest
began over a century ago. In the early days, log drives were the most cost-effective and efficient way
of transporting lumber to mills and played a significant role in forestry operations. Today, on the
intertidal and estuarine sections of the lower Fraser River, log storage is a key component of forestry
operations. Logs reach the lower Fraser by travelling down the coast in booms or by barge and are
stored for processing or shipment elsewhere. This area is valued because brackish waters protect logs
from wood borers and storage areas are located in proximity to many processing mills (Sedell et al.
1991). In the 1990s, the Fraser River estuary was estimated to support 40 processing operations
consuming 25% of B.C.’s coastal production, provide six weeks of inventory for these mills, and
store 6% of logs in transit elsewhere (FREMP 1991; FREMP 1994).
For many years industry and government have been working together to manage log storage (the
predominant use in the Fraser and Pitt Rivers) and reduce impacts on the estuary. Waterlots are used
to designate areas for a variety of purposes, including log storage, which are governed by temporary
permits (up to 1 year), licenses of occupation (10 years) and leases (30 years). Port Metro Vancouver
is responsible for log storage in the estuary and works with government, communities, and other
industry to balance competing uses (see http://www.portmetrovancouver.com/). The Fraser River
Estuary Management Program (FREMP) is an inter-governmental partnership that coordinates
planning, protection, and decision making among varied interests as related to the estuary (see
http://www.bieapfremp.org/main_fremp.html).
Our understanding of the impacts of log storage and handling on estuarine environments is based on
summaries of the science from across western North America (e.g., Sedell et al. 1991; Levy et al.
1996), though studies on the effects on salmon in the Fraser River estuary, in particular, are limited.
There are four phases of log handling which can cause physical and chemical impacts on estuarine

28
environments – dumping, booming, storage, transport. In general we know that logs can compact,
scour, and shade nearshore habitats which in turn can reduce plant cover and food availability for
juvenile salmon. As well, wood and bark debris can accumulate beneath storage areas which can
alter the composition of food sources, smother emergent vegetation, increase biological oxygen
demand, and increase concentrations of potentially toxic log leachates. The magnitude of these
disturbances is considered to be a function of the flushing characteristics of the river, the specific
methods of log handling / storage, and intensity of use in each area.
These disturbances in the estuary have the potential to affect Fraser sockeye salmon. The brackish
and freshwater channels of the lower river are known to support millions of outmigrating salmon
which occupy marine foreshore areas after smoltification, and prior to migrating out to sea. The
lower reaches of the river also act as a staging area for adults migrating upstream to their natal
streams (FREMP 2003). Despite the potential for disturbance and importance of the habitats, a
comparative study in the Fraser River estuary revealed that densities of juvenile salmon (chinook,
pink, and chum) and amphipods did not differ between a large log storage site and nearby marsh
areas (Levy et al. 1982; 1989). Little is known about impacts on adult salmon (Sedell and Duval
1985 as cited by Levy et al. 1996), though there is evidence that log storage at the outlet to sockeye
salmon nursery lakes can block upstream migration of adults (DFO 2002).
For waterlots where log storage occurs, temporary permits and leases are currently available through
Port Metro Vancouver. Across its area of oversight, the Port estimates there are approximately 48
different tenants distributed across 256 log storage agreements (193 leases and 63 permits), covering
a total area of 862 hectares within the estuary (693 hectares under lease and 169 hectares under
permits) (Nathan Nottingham, Port Metro Vancouver, pers. comm.8).
Data describing the year to year variation in log storage across the Fraser River estuary are not
available from the Port. Despite this gap, others have reported on the extent of log storage in the
estuary. Given variation in reporting for different spatial areas of the lower Fraser, these data are not
comparable across years but are provided here for some context. In the 1980s, about 1,485 hectares
of the entire lower Fraser River foreshore was reported as leased for log-boom storage (from Higham
8 The figures are the most recent estimates, which might include some errors due to the consolidation of databases
following the recent amalgamation of the Fraser River and North Fraser Harbour Port Authorities.

29
1983 as cited in Birtwell et al. 1988; also see FERIC 1980). In 1991, 555 of 970 hectares available
for storage were reported as being used (FREMP 1994). While in 1999, 636 waterlot leases, licenses,
and permits were being held by 334 tenants on the main reaches of the lower Fraser (i.e., excluding
the north arm, VFPA 2008).
Given the gap in data describing year to year variation, the next best available source of information
was a time series of aerial photos available from Google with date and year stamps (see Appendix 4).
We visually inspected these images to qualitatively examine spatial (across reaches) and temporal
changes (across seasons and years) in log storage from 2001 to 2009. Based on visual examinations
of air photos from the last decade, four areas were identified as having the highest relative
concentrations of log storage: at the mouth of the north arm, along the upper north arm, throughout
the north channel around Annacis Island, and within reaches of the Fraser main channel near the
confluence of the Pitt and Fraser Rivers (see boxes A-D respectively in Figure 17). The
concentration of activity in the north arm is well known, which includes what is considered one of
the largest log storage area in the world at its mouth (FREMP 1994; FREMP 2003). Air photo
examinations also revealed seasonal variation in log storage, with storage appearing to be lowest
during the winter. This observation is consistent with statements that the amount of log storage varies
across seasons in response to changes in logging and flows (FREMP 1994). Across years of
observation, however, we did not notice any significant changes in the magnitude or spatial coverage
of log storage across the estuary. Variation appeared to be larger across reaches than across seasons
or years within a reach, which suggests that specific locations are preferred for log storage in
response to seasonal and annual needs.
3.2 Mining
Mining development has a long history in the Fraser River basin. Today, the Ministry of Forests,
Mines, and Lands (formerly the Ministry of Energy Mines and Petroleum Resources) actively
encourages mining development as a generator of economic activity across the province (EMPR
2006). Mining is a contentious issue with respect to salmon conservation. Proponents argue that tight
regulation of activities and a small geographic footprint have minimized environmental impacts
relative to the potential economic benefits, while others (e.g., Kean 2010) suggest that mining
development can still have impacts on fish habitats.

30
Several processes associated with mining have the potential for impacts on sockeye salmon
spawning habitats. In some cases, permanent loss of habitat can occur when a mine site or tailing
pond is built directly on top of a lake or stream. Mining of gravel or placer minerals from the stream
bed itself leads to less obvious disruption of the stream bed (Kondolf 1997). Silt and sand from
roads, pits, and gravel washing can be transported to spawning areas, thereby reducing egg survival
(Meehan 1991). In addition, mines can produce acid drainage, heavy metals, and other contaminants9
that may have lethal or sublethal effects on all life history phases (Nelson et al. 1991). With the
exception of contaminants, the processes that link land use to migration and rearing are generally
much weaker that the land use to spawning habitat linkage. Sediment from mining activities can
increase lake turbidity, which can reduce light penetration and productivity (e.g. Lloyd et al. 1987) or
increase nutrients and productivity (Tilzer et al. 1976). High levels of inorganic sediment deposition
have also been associated with lower densities of benthic invertebrates (e.g. Edmonds and Ward
1979). Along migration routes, turbidity can reduce mortality of smolting juveniles (Gregory and
Levings 1998). However, the net of result of mining sediment on lake and migration habitat appears
to be minor in comparison to the potential impacts of mining sediment on stream spawning habitat.
We investigate mining activity as a potential stressor on Fraser sockeye salmon by classifying
activities into seven categories: (1) placer mining; (2) gravel mining; (3) industrial mineral
production; (4) metal mining; (5) oil and gas production; (6) coal mining; and (7) exploration related
to all of these production activities.
Placer Mining
Placer mining targets alluvial deposits in modern or ancient streambeds. Minerals, such as gold, that
are denser than sand tend to settle out and concentrate at the base of alluvial deposits. The impacts of
placer mining on sockeye salmon populations is potentially severe because many alluvial deposits
are closely associated with existing streams and water is often used to separate placer minerals from
the gravel matrix (Birtwell et al. 2005). In areas with poor environmental regulation, placer mining
and hydraulic mining can convert natural streams into barren, sediment-filled channels with
devastating impacts on fish populations (Nelson et al. 1991).
9 See MacDonald et al. 2011 for an evaluation of related impacts of water pollution.

31
Actual placer activity is difficult to quantify, but active placer claims can be used as a relative index
of placer mining interest in a watershed. There were 2965 placer mining claims in the Fraser that
were active at some point between 2000 and 2009 (EMPR 2010a).
Gravel (Construction Aggregate) Mining
Gravel mining also has potential for severe impacts on sockeye salmon populations because it also
targets alluvial deposits. Alluvial deposits are desirable sources of aggregate because the action of
water eliminates weak materials by abrasion and attrition leaving durable, well-sorted gravels that
are ideal for producing concrete (Barksdale 1991). Recognizing the potential for damage to aquatic
habitats, the B.C. government has restricted the discharge of both water and sediment from mining
operations into natural waterbodies (EMPR 2002). Like placer mining, gravel mining is a widely
dispersed activity with 450 operations in the Fraser Drainage (EMPR 2000). Most gravel mining is
done close to where it is used, typically near major cities and large rural construction projects such as
dams or roads. Rural activity is difficult to track because of large year to year variations.
Industrial Mineral Extraction
Industrial minerals include a range of non-metalic minerals such as clay, diatomite, gemstones, slate,
gypsum, limestone, pumic, silica, volcanic ash and rare elements (EMPR 2010b). Most operations
are relatively small because of a limited market or a limited supply of raw material. Risks from these
operations are lower because, in contrast to placer and gravel mining, most of these minerals are not
linked to alluvial deposits and processing does not depend on large volumes of water.
Metal Mining
Metal mining activity is extremely concentrated with only five active metal mines in the Fraser
drainage (EMPR 2010c). With the exception of the Endako mine near Francois Lake none of these
mines are in close proximity to habitat occupied by juvenile sockeye salmon. A large number of
inactive mines are present but most of these have very small footprints. The main risks from
abandoned mines are the continuous release of acid drainage and heavy metal contamination9.
Coal Mining
Active coal mining does not occur in the Fraser River basin (EMPR 2010d). Proposals to mine the
Hat Creek deposit have been prepared in the past but have never been implemented. A coal deposit

32
has also been identified in the Horsefly drainage, but little work has been done to develop its
potential.
Oil and gas production
Exploration wells drilled in the Nechako Basin, Quesnel Trough, and Georgia Basin have identified
oil and gas potential in these areas (Hannigan et al. 1998), though no production has been initiated to
date.
Exploration Activities
Exploration activities in British Columbia dropped from $350 million in 1988 (all figures in 2006
dollars) to less than $50 million in 2001 before rising to over $400 million by 2007 (EMPR 2008).
Details of the locations and results for individual projects are reported regularly (EMPR 2010b), but
a database of exploration is not readily available. Roads are thought to be the major environmental
impact associated with mining exploration. The impacts associated with road building and
exploration is captured in Section 3.1.1. Other activities (e.g., trenches, drill holes, adits) have likely
had few impacts on sockeye salmon populations.
The impact of these processes on Fraser River sockeye salmon will vary with the amount of activity,
severity of effects, and overlap with zones of influence on sockeye salmon spawning. We use three
spatial layers from the Ministry of Energy, Mines, and Petroleum Resources to identify locations of
these mining activities and mineral / placer claims across the province (see Appendix 4). For each
mining category, we quantify the extent of each of these activities in watersheds utilized by various
stocks and Conservation Units of sockeye salmon.
The occurrence of mining activity in the watersheds of spawning streams varies substantially across
sockeye salmon Conservation Units (see Table 8 and Figure 18). These data suggest that the impacts
of mining on sockeye salmon population densities will be small and difficult to detect. The causal
mechanisms that link mining, particularly sediment deposition, to lower egg survival are well
documented, but the contrasts among stocks and the strength of the effect relative to other factors is
too low to be easily detected. The Shuswap Complex CUs appear to be the most heavily impacted
(Figure 18), but most of the activity is in the upper Shuswap River drainage and is not geographically
linked to areas that support the majority of spawning. The majority of CUs have little or no mining

33
activity in the watersheds of tributary spawning streams. Placer mining is the dominant mining
activity across sockeye salmon Conservation Units and appears to have the highest potential to
reduce early freshwater survival, although we expect that environmental regulations have
ameliorated some of the impacts of placer mining (EMPR 2002; 2009). There are also a variety of
inactive mine sites which may continue to impact watersheds, particularly through acid mine
drainage (Province of BC 2002). The highest density of these is also in the Shuswap Complex CUs.
Acid mine drainage also has the potential to impact migration corridors. EMPR (1998) guidelines
mandate the control of acid mine drainage, usually by permanent flooding and storage of mine tailing
in a tailing pond.
Gravel mining on sockeye salmon migration corridors, such as the lower Fraser River, is a
contentious issue that has received considerable public attention (which is outside the scope of this
report, see Johannes et al. 2011). Although the immediate physical effects of gravel mining are
obvious, biological impacts are difficult to detect because they are obscured by natural variation
among sampling sites (Rempel and Church 2009). The major impacts of these activities are thought
to be on species that spawn (pink salmon, chum salmon) or rear (Chinook salmon, steelhead trout) in
these areas rather than those, like sockeye salmon, that utilize these reaches as migration corridors
(Rosenau and Angelo 2000). River-type sockeye salmon CUs may rear in these reaches, but there is
very little data on either the ecology or status of these small CUs.
3.3 Hydroelectricity
In this section we investigate two topics related to generation of hydroelectricity that have the
potential to affect Fraser River sockeye salmon: (1) large scale hydroelectric projects and (2) small
scale hydroelectric projects (i.e., run-of-river independent power projects).
3.3.1 Large scale
Development of hydropower potential in the Fraser River basin began in the early 1900s but has
been limited to the tributary systems with the mainstream remaining undammed. The main flow of
the Fraser has never been dammed partly because high levels of sediment would shorten any dam’s
lifespan, but also due in part to strong support for salmon fisheries (Roos 1991; Ferguson and Healey
2009). Small hydro-electric dams have been in place for many years with many projects in the Lower

34
Fraser (Alouette, Bunzten, Coquitlam, Ruskin, Stave, and Wahleach) and Bridge River area (Lajoie
and Seton). The Bridge/Seton River power project and Alcan’s Kemano Project on the Nechako
River are the two large-scale hydro facilities in the basin that could potentially impact Fraser sockeye
salmon populations (Roos 1991).
Large hydro projects can create physical barriers that block or delay migration to spawning areas;
affect the quality, quantity, and accessibility of salmon habitats; create conditions that increase stress
on migrating salmon (making them more susceptible to disease and pre-spawning mortality);
increase susceptibility to predators, and cause direct mortality of migrating adults or smolts that pass
through hydro turbines or over spillways (Roos 1991; Marmulla 2001).
Bridge/Seton River Power Project
The Bridge/Seton Power Project is a hydroelectric power development located near Lillooet.
Commissioned in 1956 and later expanded it harnesses the power of the Bridge River, a tributary of
the Fraser, by diverting it through a mountainside to the separate drainage basin of Seton Lake,
utilizing a system of three dams, four powerhouses and a canal. From the lake's outlet, a specially-
built canal carries the diverted flow of the Bridge River to the last possible bit of head before the
Fraser River (i.e., elevation drops with the potential for generating hydro power). The power canal,
known as Seton Canal, is highly unusual in that it bridges both Seton and Cayoosh Creeks before
being briefly tunneled through a low rock bluff to the Seton Powerhouse, located on the Fraser River
just below the town of Lillooet.
There are two issues related to the construction and operation of the Bridge / Seton River power
project which have the potential to affect the Seton and Anderson sockeye salmon Conservation
Units. First, sockeye salmon smolts can migrate downstream of the Seton Dam to the Fraser River
via one of five exit routes: power canal/turbine, fish ladder, fish water release, siphon spillway, and
radial gate spillway. The entrainment rate is dictated by flow routing; smolts tend to concentrate in
the high flows of the power canal. Early studies indicated that over 90% of sockeye salmon smolts
were being entrained into the power canal, with the smolt mortality rate estimated as 17% when the
plant was fully operational (Groves and Higgins 1995). This estimate includes direct mortalities as
well as latent mortality from injuries, cumulative stresses, disease and predation. Based on the
number of spawners returning to Gates and Portage Creeks and the estimated number of smolts

35
produced per female spawner, the average number of smolts lost at the canal is estimated at
~200,000 (Levy and Sneep 2006). Assuming a 5% smolt:adult survival, this is equivalent to a loss of
~10,000 adult sockeye salmon annually from entrainment mortality (Levy and Sneep 2006). Actual
smolt:adult survival rates can vary tenfold from year to year.
There is a long history of fisheries investigation in the Seton River to determine ways to reduce
entrainment mortality of sockeye salmon smolts (Fretwell 1979; 1980; 1982; Fretwell and Hamilton
1983; Groves and Higgins 1995; R.L.& L. 1999; 2000). Many of the earlier solutions to mitigate
smolt mortality were largely ineffective (Levy and Sneep 2006). The Northern St’at’imc Fisheries
and BC Hydro have been working together since 2006 to devise practical ways for mitigating this
mortality at the Seton Generating Plant. Under a draft Settlement Agreement between BC Hydro and
St’at’imc Nation, a 5% entrainment mortality rate was selected as a target at the Seton power canal
(Levy and Sneep 2009). Recently it has been determined that if seasonal maintenance and nightly
shutdowns of the station powerhouse coincide with the peak smolt migration period, approximately
95% of emigrating smolts can be protected from entrainment. While implementing this mitigation
measure from 2006-2009, smolt mortality rates were estimated at 1.7%, 3.1%, 10.1%, and 1.8%,
respectively (Levy and Sneep 2006; 2009). By undertaking these measures BC Hydro believes it can
effectively meet the 5% (or lower) sockeye salmon smolt mortality target (Levy and Sneep 2009).
The second issue was first noted in the late 1960’s and early 1970’s when adult sockeye salmon
migration up Seton Creek was being delayed and fish were getting injured in the power house
tailrace while attempting to swim up the draft tube. Initially, homing problems were attributed to a
lack of continuous stream of Seton River water in the Fraser River. It was determined that delays
could be reduced if outflow were lowered and discharge in Seton Creek were increased, but this
option could not initially be incorporated into the operations of the facilities (Roos 1991). Field
telemetry and water preference studies in the late 1970’s and early 1980’s indicated that adult
sockeye salmon were able to discriminate between pure Seton water and water diluted by Cayoosh
Creek. Gates Creek sockeye salmon would move out of the tailrace area into Seton River without
delay if the concentration of Cayoosh Creek water in Seton River was 20% or lower. Portage Creek
sockeye salmon would not move into Seton River until the percentage of Cayoosh Creek water was
less than 10%. Given these targets, the problem was solved by diverting most of the Cayoosh Creek
flow through a tunnel, the Seton/Cayoosh diversion, into Seton Lake (see Figure 19, Fretwell 1989).

36
As a result, dilution guidelines have been in place since 1979 to reduce the proportion of Cayoosh
Creek water in Seton River during sockeye salmon migration so adults will not differentiate between
the tailrace and the river (target of 20% dilution for Gates Creek and 10% for Portage Creek).
Direct observations from 1979 to 1981 confirmed the effectiveness of these measures, while more
indirect evaluations in subsequent years (e.g., qualitative observations of delay in tailrace, fish
counter results, trends in abundance of Gates and Portage stocks) indicated that this action continues
to be effective (BC Hydro, unpublished). However, a 2007 assessment of attraction and delay at the
Seton powerhouse tailrace indicated that 13% of 27 sockeye salmon released into the tailrace did not
reach Seton Dam (Roscoe and Hinch 2008). This was unexpected since dilution levels were 2-6%,
much less than the 20% dilution target. It is uncertain whether these fish failed to enter Seton River,
or entered the river but fell back before reaching the dam. This study suggests that more research is
needed to better quantify delays and their implications on sockeye salmon, which includes a re-
evaluation of the ‘dilution level’ target, and examination of whether the tailrace is acting as a thermal
refuge from warmer Fraser River water (Roscoe and Hinch 2008). Gates and Portage Creek fish now
typically encounter Fraser water temperatures exceeding 18-19oC, temperatures that can be
extremely stressful to migrating sockeye salmon (Crossin et al. 2008).
In spite of the difficulties in providing safe passage for migrating smolts and adults, sockeye salmon
in the area have been at higher levels of abundance than they were prior to dam construction; though
these abundances have likely been aided by the construction of a spawning channel at Gates Creek in
1968 (see Figure 20, Roos 1991). However, the total productivity of Gates Creek sockeye salmon
(recruits per effective female spawner) has declined since the 1990’s, due to declines in post-juvenile
productivity (Figures D-J3, D-P3 and D-T5 on pages 146, 149 and 154 in Peterman et al. 2010. This
is a similar pattern to that observed for seven of the eight Fraser index stocks with monitoring of
juvenile abundance.
Kemano Power Project
The Nechako River drains the Nechako Plateau east of the Kitimat Ranges, flowing north toward
Fort Fraser, then east to Prince George where it enters the Fraser River (Figure 21). The Nechako is
one of the main tributaries of the Fraser River, although for many years most of its flow (up to 80%)
has been diverted through the Coast Mountains to the Kemano generating station for the purpose of

37
supplying power to the aluminum smelter in Kitimat. The main reservoir of the Nechako power
diversion is the Nechako Reservoir behind Kenney Dam. The Kemano Power Project originated in
1941 when the British Columbia government invited the Aluminum Company of Canada Limited
(now Alcan Inc.) to develop a hydroelectric power project and establish an aluminum industry on
Canada’s West Coast. In 1950, the provincial government entered into an agreement with Alcan
granting them a conditional water license for power generation (NFCP 2005).
River flow began being diverted in 1952 and the reservoir took four years to fill. Although the
project did not directly block migration to any existing sockeye salmon spawning grounds, the 1950
agreement and conditional water license allowed Alcan to reduce releases at the Skins Lake Spillway
during periods of below-average inflows to the Nechako Reservoir (Roos 1991). In 1980, Fisheries
and Oceans Canada expressed concern over the volume of water being released through the
Spillway. It was anticipated that sockeye salmon migrating through the Nechako River system would
be exposed to high summer water temperatures resulting from the low water flows. Most sockeye
salmon stocks moving through the river are only briefly exposed to warm thermal conditions, but the
concern was that increases in temperature can increase stresses on migrating salmon, making them
more susceptible to disease and pre-spawning mortality. Consequently, a river temperature control
program, the Summer Temperature Management Program (STMP), was developed in response to
these concerns (NFCP 2005).
A 1987 Settlement Agreement defined how Alcan was to manage water temperatures for the benefit
of migrating adult sockeye salmon. The Early Stuart – Stuart, Takla / Trembleur; Early Summer –
Nadina, Fraser, Francois; and Summer – Stuart, Takla / Trembleur, Fraser, Francois Conservation
Units use the river as a migration corridor, with adults spending two to four days in-river during
migration (though a small number of sockeye salmon spawn in the Nechako itself). Of these CUs,
those sockeye salmon migrating upstream of the confluence with the Stuart River face the greatest
stress as they are exposed to high temperatures for longer periods of time. The Agreement specified a
schedule of short term water releases, but did not specify the volume of water to be released to
protect migrating sockeye salmon. A water temperature model and accompanying protocols are used
for daily decisions on the volume of water to be spilled through the Spillway to meet temperature
targets in July and August. The need for additional cooling water is assessed daily using this
computer model which incorporates real-time data and forecasts of water temperature, water flow,

38
and meteorological conditions. The temperature control point on the Nechako River is located and
measured upstream of the confluence with the Stuart River at Finmoore. The long term goal of the
flow and temperature control program has been to maintain mean daily water temperatures at or
below 20°C during the period of adult sockeye salmon migration from July 20 to August 20 (IPSFC
1979; NFCP 2005).
Water temperatures have been monitored in both the Nechako and Stuart Rivers at their confluence
since 1953. Consequently, water temperatures can be evaluated for two periods: pre-STMP (1953 to
1979) and post-STMP (1983 onwards). To evaluate the effectiveness of the program, thermal
conditions of the Nechako River have been compared to water temperatures in the unregulated Stuart
River (deemed a control watershed), which shares the same hydrological basin and biogeoclimatic
influences (NFCP 2005)10. Recent reporting indicated that Nechako River summer water
temperatures have generally remained between 15°C and 21°C between 1953 and 2000, and have
infrequently exceeded the 20°C target (see Table 9). Mean daily water temperatures occasionally
exceeded 20°C in both the Nechako and Stuart Rivers, but did so more frequently in the Stuart River
(a control system whose temperatures are not influenced by the Kemano Power Project) than in the
Nechako River (NFCP 2005). Regular flow monitoring reporting subsequent to the NFCP 2005
summary has indicated similar adherence to temperature targets, with 5 of 8 of the reported years
between 2001 and 2009 having zero days in exceedance of the 20°C water temperature target.
Temperature targets were, however, exceeded in 2004 (13 days), 2006 (5 days) and 2009 (11 days),
with the maximum mean daily water temperature reaching as high as 21.7°C in 2006 (Table 10).
Since 1983 when the STMP was implemented in its current form, the program has managed the
release of water from the Nechako Reservoir to limit the frequency of days with mean daily
temperatures >20°C at the temperature control point. During this period Nechako River temperatures
have only rarely exceeded 20°C during the period of sockeye salmon migration, even though
analyses of meteorological conditions suggest some general warming of the area in recent years
(NFCP 2005).
10 We note that this comparison is limited because temperature data only exist for years in which the Nechako River was
regulated by flows (i.e., no pre-regulated data).

39
3.3.2 Small scale
In the last decade, the Provincial government has encouraged the development of Independent Power
Projects (IPPs) as an integral part of British Columbia’s long term energy plans (Province of British
Columbia 2010). IPPs are typically small installations (<50 MW) with diversion dams but little
storage capacity. In some cases, several small installations are combined into a single much larger
development11. Despite their appeal as sustainable sources of electricity, there are public concerns
over potential negative impacts of IPPs on the aquatic environment, including salmon (Douglas
2007; UBC 2010).
There are several plausible mechanisms by which IPPs could affect sockeye salmon survival, even
though most divert water from a relatively short stream channel that is often fishless. IPP operations
can affect Total Gas Pressure (TGP), gravel supply, and water temperature. Each of these effects can
be propagated to downstream reaches where they may have negative impacts on sockeye salmon
spawning habitat. Although water is not directly diverted from migration corridors, IPPs may affect
migration of sockeye salmon smolts or adults by changing TGPs and water temperatures in
downstream migration corridors. Assuming no interbasin transfer of water, there are no plausible
mechanisms by which IPPs could significantly affect sockeye salmon nursery lakes.
High TGP, usually in the form of nitrogen supersaturation, can occur when gas or air is entrained in
water and then subjected to high pressures. Elevated TGP is an issue for many hydro electric
facilities because it can produce gas bubble trauma in fish (Weitkamp and Katz 1980), including both
adult (Nebeker et al. 1976) and juvenile (Nebeker and Brett 1976) sockeye salmon. Elevated TGP
can persist for several kilometers (e.g., Scheibe and Richmond 2002) and therefore fish may be at
risk at substantial distance below an IPP installation. However, Douglas (2007) suggests that TGP
can be naturally high in turbulent headwater streams and that small hydro installations may actually
reduce TGP.
Dams can disrupt the gravel supply to downstream reaches if sediment is either trapped in a reservoir
or periodically removed from an intake structure. This disruption in gravel supply can have serious
negative effects on channel integrity and the quality of salmon habitat in reaches downstream of
dams (Kondolf 1997). On small diversion dams, such as those typical of many IPP installations, low
11 See http://www.plutonic.ca/s/Home.asp or http://www.purcellgreenpower.com

40
level outlets can be used to maintain a natural sediment regime if care is taken on the timing of
sediment release relative to peak flows and salmon spawning activity. Construction activities and
infrastructure may also result in increased sedimentation. The potential for these effects has been
recognized and regulations have been implemented in an attempt to reduce the risks from excessive
sedimentation (MoFR 2005).
Stream temperatures below reservoirs can be either higher or lower than natural thermal regimes.
Large reservoirs with deepwater outlets can depress downstream temperatures by more than 10 °C.
Surface outlets typically result in warmer downstream reaches. Even small diversion reservoirs may
raise downstream temperatures significantly, especially under low summer flows. Warmer
temperatures can stress sockeye salmon and result in delays or mortality during upstream migration
(Crossin et al. 2008).
To investigate the potential interaction between these issues and sockeye salmon, we gathered
geographic coordinates for all existing IPP locations in the Fraser River basin (see Appendix 4).
Next, we used GIS to intersect these IPP locations with sockeye salmon spawning areas / migratory
corridors for all CUs in the basin. Given this information we then qualitatively assessed the
significance of IPP installations that are directly upstream of a known spawning area or on a direct
tributary of a sockeye salmon migration route. We also reviewed the status of IPPs with respect to
interbasin transfers and diversions from stream reaches used for migration. The impact of roads
associated with IPP development is considered under forest harvesting activities (see Section 3.1.1).
The history of interaction between IPPs and sockeye salmon is very short and limited in number
(Figure 22) and spatial extent (Figure 23). Only one IPP has recently begun operating in a watershed
that supports sockeye salmon spawning (Harrison -downstream migrating-Late timing) in the lower
reaches (Figure 24). Another two installations are in final planning stages on Silver Creek (Late –
Harrison (D/S)) and on Sakwi-Weaver Creeks (Late – Harrison (U/S)). In each case, sockeye salmon
spawning is concentrated in the lower reaches but on Silver Creek the installation is more than 10 km
upstream. The other two installations are directly adjacent to sockeye salmon spawning areas.
Survival of sockeye salmon is intensively monitored at the Weaver Creek spawning channel, which
is directly below the proposed Sakwi Creek IPP. Historically, adult and egg survivals have been
consistently high at the Weaver Creek spawning channel (Essington et al. 2000). Recent increases in

41
mortality of Weaver Creek sockeye salmon adults appears to be the result of higher temperatures
associated with climate change and changes in migration timing (Mathes et al. 2010). Although 24 of
30 CUs have IPPs associated with migration corridors, the number of installations is small (see
Figure 22 and Figure 23). Three are on small tributaries to major rivers (two on the Fraser River and
one on the Quesnel River). The remaining installation is on a tributary to a large lake (Seton) through
which sockeye salmon migrate. Interbasin transfers and diversions from stream reaches used for
migration was not an issue of concern for Fraser River sockeye salmon. Given the available data and
these results, IPPs have not had significant impacts on sockeye salmon populations. This conclusion
is based on the small number in proximity to spawning grounds or migration corridors.
3.4 Urbanization upstream of Hope
More than two-thirds of British Columbians live in the Fraser River basin (FBC 2009), many of
which live in urban environments. Although the relative size of urban footprints is typically lower
than that of other human activities (e.g., agriculture or forestry), the intensity of disturbance is
generally regarded as higher, in part, due to the concentration of activities and irreversibility of
disturbance associated with the built environment (Paul and Meyer 2001). Due to the effects of
human activities on Pacific salmon (including urbanization), population growth has generally been
recognized as one of its greatest threats (Hartman et al. 2000). In the Fraser River basin, the potential
interaction between people and salmon is a valid consideration given the growth over recent decades
and strong overlap between human populations and salmon distribution (Nelitz et al. 2009). The pace
of growth from 1981 to 2006 has varied across the basin, being markedly higher (81%) in the lower
Fraser than areas upstream of Hope (2%, 5%, and 25% total growth for Nechako, upper Fraser, and
Thompson respectively). The distribution of people is similarly varied with the majority living in
urban environments of the lower Fraser River valley (Figure 25), though this area is outside the
scope of our evaluation (see Johannes et al. (2011) for an evaluation of impacts of urbanization
downstream of Hope).
Despite the smaller population upstream of Hope, urbanization and the related built environment
have the potential to affect freshwater habitats for Fraser sockeye salmon in three ways (Rosenau and
Angelo 2009). First, residential, business, and industrial development, as well as related road
construction can increase the amount of impervious surfaces in urban watersheds which affect rates
of interception, patterns of runoff, and in turn the magnitude and timing of instream flows (e.g., peak

42
and low flows). Direct extraction from groundwater and surface water supplies for municipal
purposes can affect water availability for sockeye salmon spawners and incubating eggs. Second,
construction of roads and buildings along stream channels and lake foreshore areas have the potential
to reduce riparian vegetation, channelize streams, and block access to habitats (e.g., Radomski et al.
2010). Such activities have been directly linked to alterations in sockeye salmon habitats in the lower
Fraser (e.g., Bocking and Gaboury 2003; COSEWIC 2003). Lastly, roads, stormwater runoff, as well
as municipal and industrial effluents have been known to alter water quality in watercourses across
the Fraser River basin by changing concentrations of sediments, nutrients, and contaminants
(Birtwell et al. 1988; Dorcey and Griggs 1991; B.C. Ministry of Environment, Lands and Parks and
Environment Canada 1994). Beyond a consideration of the extent of road development in urban
watersheds, the impacts of water pollution on Fraser sockeye salmon are outside the scope of this
report (see MacDonald et al. 2011 for an evaluation of related impacts of water pollution).
We examined four data sources to evaluate the significance of these issues on Fraser sockeye salmon
at a landscape level (see Appendix 4). We use the TANTALIS municipal boundary layer from the
provincial government to identify areas of overlap between urban environment and sockeye salmon
Conservation Units. Due to the nature of the underlying data, this layer cannot be used to assess
changes in the concentration of urban development or urban land cover over time. We use the B.C.
Digital Road Atlas to consider the density of roads to assess proximity of potential road impacts on
sockeye salmon habitats (see Section 3.1.1). We use the provincial water license layer to identify the
location and amount of water designated for domestic or waterworks purposes (see Section 3.6).
Lastly, we use the census boundaries and population estimates from the federal and provincial
governments to examine spatial and temporal trends in population growth and how these trends
might interact with sockeye salmon Conservation Units of the Fraser River. We consider human
population numbers as a surrogate indicator of the related impacts of urbanization (e.g., increase in
impervious area or destruction of riparian habitats due to construction of buildings and roads
adjacent to streams and lake foreshore areas).
Figure 26 through Figure 29 illustrate the results of our examination between these urban stressors
and all lake sockeye salmon Conservation Units. Relative to forest harvesting and Mountain Pine
Beetle disturbance, urban environments have a relatively small footprint within watersheds and
riparian zones that influence sockeye salmon spawning and rearing habitats (Figure 26). However, of

43
all other types of land cover (forest harvesting, Mountain Pine Beetle, and agriculture), urban
footprints have the largest interaction with migration corridors. The extent of this interaction across
many CUs is largely a function of the need for all sockeye salmon in the Fraser River basin to
migrate through the Lower Mainland (Figure 27). CUs with longest migration distances tend to have
the lowest proportion of urban development across their migration (e.g., ~ 16% for Bowron and
Nadina – Early Summer), while those with the shortest migrations have the greatest proportional
extent of urban development along their migration (e.g., > 88% for Pitt – Early Summer and Cultus –
Late). The extent of urban development along migration corridors is further illustrated by available
human population data which shows that the average density of people along migration corridors
across CUs (Figure 28) is significantly higher than the population density within areas influencing
sockeye salmon spawning and rearing areas (Figure 29). Conservation Units with the highest
population densities along their migration corridors and spawning / rearing areas again include: Early
Summer – Chilliwack and Pitt; Late – Cultus. Those CUs with the lowest population densities
include: Early Stuart – Stuart, Takla / Trembleur; Early Summer – Fraser, Francois, Nadina, Bowron;
Summer – Stuart, Takla / Trembleur, Fraser, Francois. For the past 25 years, the trend in population
density has been a steadily increasing and relatively parallel pattern across CUs.
3.5 Agriculture
The Fraser River basin supports 53% of the province’s farmland and more than 9,000 farms (FBC
2009). Agriculture can be a significant stressor on freshwater ecosystems because it often occurs
within valleys and riparian areas adjacent to larger mainstem rivers that provide high quality salmon
habitats. This concern is supported by empirical evidence which has related agricultural development
to declines in coho salmon in the basin (e.g., Bradford and Irvine 2000). Two trends are informative
for setting the context and understanding the potential impacts of agriculture. First, the extent of land
in the Agricultural Land Reserve (ALR) and number of farms has remained relatively stable in recent
decades at a provincial level, though trends differ across regions (Figure 30). For instance, from 1979
to 2000 the interior and south coast regions of the Fraser River basin have experienced net losses of
2,686 and 13,136 hectares, respectively, largely the result of population pressure and demands for
urbanization in these areas (BC MOE 2008). Second, evidence from British Columbia (Figure 31)
and across the country (Statistics Canada 2009) suggests that the agricultural intensity on these lands
(e.g., number of livestock) have increased in recent years.

44
There are three general pathways by which agriculture can have landscape level impacts on salmon
habitats (reviewed Platts 1991; Rosenau and Angelo 2009). First, livestock grazing and crop
production can lead to physical alterations of streams, riparian zones, and floodplains. Cattle crossing
through streams can increase sedimentation, destroy spawning redds, and destabilize stream banks /
widen the stream channel. Removal and continuous disturbance of vegetation in the riparian zone
can reduce stream shading and increase stream temperatures which affect spawners and eggs. Further
upslope, crop production and farm roads in the floodplain can compact soils leading to less
interception of precipitation and more surface water runoff. A second impact pathway is the direct
removal of water from groundwater and surface water supplies for irrigation and livestock purposes.
Extraction of surface waters can constrain access to habitats, while extraction of groundwater can
reduce the supply of cool summer baseflows to streams and rivers. Lastly, agricultural activities can
have significant impacts on water quality of streams and lakes by increasing biochemical oxygen
demand, introducing pathogens, and affecting concentrations of sediments, nutrients, and
contaminants through the introduction of manure, fertilizers, and pesticides into waterways (e.g.,
Schendel et al. 2004; Schindler et al. 2006; Smith et al. 2007; Jokinen et al. 2010). As mentioned
earlier, the impacts of water pollution, however, are outside the scope of this report (see MacDonald
et al. 2011 for an evaluation of the related impacts of water pollution).
We used two data sources to examine the potential interaction between agriculture and sockeye
salmon Conservation Units across the Fraser River basin (see Appendix 4). The Agricultural Land
Reserve layer is used to represent the spatial distribution of agriculture in recent decades and the
provincial water license layer is use to identify the location and amounts of water allocated (not
actual use) for stockwatering and irrigation purposes (see Section 3.6). These are the best data
sources available to describe the spatial extent of agricultural activities which at a broad scale does
not appear to be changing dramatically (as discussed above). Yet we acknowledge that agricultural
impacts will depend on the type of agriculture and intensity of use which have likely changed more
dramatically. For instance, given noted increases in livestock it would have been preferable to
examine changes in the type and intensity of pressure on lands and streams associated with livestock
production. These data, however, are generally lacking in an easily accessible and consistent format
across the basin.

45
Relative to forest harvesting, Mountain Pine Beetle disturbance, and urban development, agriculture
has a relatively small footprint within watersheds and riparian zones that influence sockeye salmon
spawning and rearing habitats (Figure 32). Agriculture has its greatest interaction with migration
corridors, the extent of which is less than that of urban development yet more than that of forest
harvesting and Mountain Pine Beetle disturbance. The concentration of agriculture along migration
corridors is consistent with the observation that agricultural activities tend to be located adjacent to
large rivers or within river valleys that have productive soils and are close to water for crop irrigation
and livestock watering. This interaction with migration corridors across many interior CUs is largely
due to a concentration of agricultural lands in the Cariboo-Chilcotin along the Fraser River mainstem
(Figure 33). Conservation Units with the smallest extent of agricultural lands along their migration
corridors include: Early Summer – Pitt, Nahatlatch, Anderson; Late – Harrison (D/S), Harrison
(U/S), Seton, Lillooet. CUs with the greatest extent of agricultural lands include: Early Stuart –
Stuart; Early Summer – Chilko, Fraser, Francois, Bowron; Summer – Chilko, Fraser, Francois,
Quesnel, Stuart.
3.6 Water use
Water has direct and significant influences on the economic, social, and environmental well being of
British Columbians. Water use conflicts arise when faced with water scarcity — either limited
supply or high demand — which leads to situations where there is not enough water for both human
and ecosystem needs. Across the province we already face challenges balancing human and fish
needs for water in some locations and on some years12. The Fraser River basin, in particular, shows
the strongest overlap among water licenses, water allocation restrictions, population density, and
salmon distribution in the province (Nelitz et al. 2009). This observation is not surprising given that
the region has one of the lowest water yields per person in the country (Statistics Canada 2010). The
provincial and federal governments inherently recognize the potential for conflict in the basin due to
the extent of natural flow sensitivities, number of heavily developed aquifers, and existing water
allocations restrictions (Rood and Hamilton 1995; BC MOE 2008; Government of British Columbia
2010). In recent years, groundwater observation wells have shown a large increase in the percentage
of wells with declining levels due to human use (BC MOE 2008). Priority areas of concern related to
groundwater use in the Fraser River basin include aquifers in the Lower Fraser Valley, near Merritt,
12 Ministry of Environment. September 18, 2009. Information Bulletin – Water use reduction order to protect fish
populations. See http://www.livingwatersmart.ca/news/docs/2009ENV0020-000367.pdf

46
lower Nechako River, and upper Shuswap (Government of British Columbia 2010). Moreover, large
portions of the Thompson basin, Cariboo plateau, and upper Nechako are considered naturally flow
sensitive, regardless of human use (Government of British Columbia 2010).
The potential for conflicts with sockeye salmon are driven by high water demands across a variety of
sectors. Industrial, commercial, municipal / domestic, and agricultural withdrawals constitute the top
consumptive uses in the province (BC MOE 2006), which is consistent with the top uses in the
country (NRTEE 2010). In B.C., per capita rates of water consumption are among the highest in the
country and the world, even though rates have generally been declining since the 1980s (BC MOE
2008). Agricultural production is also heavily reliant on consumptive water use, mostly for irrigation
and livestock. Across the province 33% of farms rely on irrigation with 13-17% of cultivated lands
dependent on irrigation – the highest dependency among provinces (Statistics Canada 2010; NRTEE
2010).
Potential impacts of water use on sockeye salmon habitats are related to alterations in water flows
and temperatures. Consumptive use of surface water at critical times of year can reduce instream
flows that constrain access to spawning habitats or in extreme cases dewater redds. Extraction of
groundwater for irrigation can reduce the amount entering streams which provides important source
of cold water and late summer / fall baseflows for migrating adults, rearing juveniles, and incubating
eggs (Douglas 2006; Smerdon and Redding 2007). While reductions in both surface water and
groundwater supplies can increase water temperatures which affect sockeye salmon adults and eggs.
To examine the significance of water use on sockeye salmon, we used the provincial water license
layer to identify the location and amounts of water allocated for consumptive purposes across
different sectors (i.e., licenses designated for municipal, domestic, residential, industrial,
commercial, irrigation, or stockwatering purposes). We did not examine the significance of water
licenses for non-consumptive purposes even though changes in the timing and magnitude of flow
releases can affect sockeye salmon habitats. These effects are being explored, in part, by our
examination of the operations of large and small scale hydroelectric facilities in the Fraser River
basin (see Section 3.3). An analysis of the effects of other non-consumptive licenses is not possible
given the need for a detailed understanding of site specific and yearly operating conditions for each
license. We also used the water allocation restriction layer to identify streams and watersheds where

47
additional water licenses are currently restricted and conflicts between water use and sockeye salmon
might be highest (see Appendix 4).
These data represent the best sources of information for water use, but we acknowledge some
significant limitations and weaknesses. Water use licenses represent the amount of water allocated in
a river at a single snapshot in time, not actual rates of consumption (i.e., monitoring of water use and
compliance with water license conditions does not occur). This gap is problematic because in many
streams it has been recognized that water allocations exceed the amount of water available.
Consequently for this study, there is no way to assess whether changes in water consumption over
the last 15 years are related to declines of Fraser River sockeye salmon. Moreover, groundwater use
is unlicensed and not monitored in a consistent way across the province or basin. DFO has
recognized the lack of data related to volume and locations of use when developing habitat indicators
under the Wild Salmon Policy (Stalberg et al 2009). This gap is problematic because actual water use
can be higher than estimates using licenses given known linkages between groundwater and surface
water supplies. Lastly, information describing water licenses (long term use) does not represent
water allocated through temporary water permits (short term use) which is a regulatory tool being
used in the oil and gas sector, an industry that requires an increasing abundance of water (Pembina
and Forest Ethics 2010).
Figure 34 and Figure 35 represent the intensity of water use across Conservation Units as measured
by total water allocation per year and per hectare, as well as the density of water allocation
restrictions. A comparison of the correlation between these indicators revealed that they are largely
independent measures of stress on water resources. A clear and perhaps obvious observation is that
high water demand is associated with greater concentrations of people across the Fraser River basin,
which also coincides with salmon habitats (Figure 36). By both measures migration corridors appear
to have the greatest allocation of water through licensing and the greatest density of water allocation
restrictions. The CUs of the Lower Mainland have the highest water allocations within their
migration corridors, mostly assigned to agricultural purposes (e.g., Early Summer – Pitt, Chilliwack;
Late – Cultus, Harrison (D/S), Harrison (U/S)). This observation is consistent with the earlier note
that agricultural activities are greatest in areas adjacent to migration corridors. In contrast, the
measure of density of water allocation restrictions identifies the Summer – Mckinley, Quesnel, and
Late – Shuswap Complex as having the highest pressures within their migration corridors. Both

48
measures of stress on water resources indicate concerns in some CUs that have spawning locations
downstream of lakes, and identify the same CUs as having the highest stress: Early Summer –
Kamloops; Late – Seton, Shuswap Complex, Kamloops). Figure 37 represents the kinds of water
uses and proportions of the total being allocated to that use. The agricultural sector tends to have the
greatest allocation across areas that affect spawning locations (lake influenced and tributary) and
migration corridors. Allocations to urban activities are second, while industrial allocations tend to be
relatively minor across habitat types.

49
4.0 Freshwater influences on Fraser River sockeye salmon
4.1 Assessment within Conservation Units
We used three approaches to summarize the abundance of data generated to describe habitats
(Section 2.2) and freshwater stressors (Section 3.0) so we can begin to discern patterns and gain
insights into possible hypotheses about freshwater influences on Conservation Units. First, we
developed a series of cumulative stressor tables which summarize the (1) alignment among
hypothesized stressors, habitat types, and Conservation Units, (2) relative intensity of and trend in
disturbance (from analyses discussed in Section 3.0), and (3) cumulative level of stress on a
Conservation Unit. Second, we plotted these cumulative stress results against the indicators of
habitat vulnerability (from Section 2.2.4) to generate bivariate plots of the combined stress and
vulnerability four each habitat type and Conservation Unit. Lastly, we developed a “dashboard”
summary of the data available to describe population status, habitat vulnerability, and freshwater
stressors specific to each lake Conservation Units across the Fraser River basin.
To summarize the cumulative level of stress affecting a Conservation Unit, we generated relative
intensity scores within each stressor category using a k-means cluster analysis of the data generated
for each landscape level stressor indicator. K-means clustering splits a set of values into a selected
number of groups by maximizing between-cluster variation relative to within-cluster variation. The
procedure allows the user to set the number of similar groups to be identified, which we set up to
identify three distinct groups of stressor intensity (i.e., low, moderate, or high values for a stressor
indicator).
Using the results from this technique, we then developed a matrix of derived scores for each of the
defined stressor indicators where relatively low intensity for a stressor was given a score of 1,
moderate intensity a score of 2, and high intensity a score of 3. Where time series data was available
for stressors (i.e., forest harvesting and Mountain Pine Beetle disturbance) we modified the scoring
based on whether the trend in the stressor appeared to be increasing (+1) or decreasing (– 1) over
recent decades. Where a stressor did not spatially overlap with a habitat we assigned a fourth default
stressor intensity category of “none”, with an associated stressor score of zero (0). For each
Conservation Unit, and within each habitat type, a cumulative stressor score was derived by
summing individual scores across all applicable stressor types. By default, each stressor was
weighted equally in this calculation.

50
We recognize the use of such a clustering method has limitations and should best be thought of as
descriptive methods for pattern analysis. Given our general goal of discerning broad variation in
stressor intensities across Conservation Units, we felt the method was well suited for our needs.
Moreover, as noted in Section 2.2.4, this clustering provides a measure of the relative intensity of
stressors across CU habitat types, and it should not be inferred that habitats experiencing a low
intensity of stress are somehow immune from detrimental effects. Conversely, habitats defined as
experienced high intensity stress or sockeye salmon utilizing those habitats may be sufficiently
resilient to withstand those impacts without serious consequences. Further work would need to be
undertaken to field validate relative differences across stressor categories. Table 11 to Table 14
provide a summary of the relative ranking of CUs according to the cumulative level of stress on lake
influenced / mainstem spawning, lake inlet / tributary spawning, nursery lake rearing, and migration
corridors, respectively.
Figure 38 illustrates our second approach to summarizing the complexity of information by
combining the stress on and vulnerability of habitats across all lake Conservation Units by habitat
type. These graphs are useful for clustering CUs into groups, where those having both high stress and
high vulnerability would be less resilient and more prone to disturbance from human activities. If
stressors in the freshwater environment are contributing to declines of salmon, then based on our
analyses the CUs with high stress and high vulnerability would be the most likely candidates where
the effects of freshwater influences could be detected. The ability to detect the effect of freshwater
stressors will also depend on the type of habitat disturbance that is most likely to contribute to the
declines. For instance, the list of CUs with both high stress on and high vulnerability of migration
corridors (Nadina, Francois, Stuart, Fraser, and Takla / Trembleur Conservation Units), is different
than those CUs with high stress on and high vulnerability of nursery lakes (Fraser, McKinley,
Kamloops, Nadina, and Cultus Conservation Units). Likewise, the relative level of stress /
vulnerability on lake influenced vs. tributary spawning is different across CUs.
Lastly, Figure 39 and Figure 40 illustrate our use of a “dashboard” summary to communicate the
abundance and complexity of information generated through our analyses in this report. Relative to
the summary figures (see Figure 5, Figure 7, and Figure 38) and tables (see Table 1, Table 5, Table
11, Table 12, Table 13, and Table 14) provided throughout this report, these dashboards provide the

51
greatest level of detail describing the conditions and stressors affecting each Conservation Unit.
These summaries are based upon Schindler et al.’s (2010) analogy of an investment portfolio as a
way to organize and communicate information. With this analogy, we relate the productivity of
Fraser sockeye salmon to the performance of a portfolio, which is based on the productivity of its
underlying Conservation Units (or performance of its assets). We extend this analogy further by
adding that the productivity of each Conservation Unit (i.e., asset or mutual fund) can be described
using a prospectus which summarizes the underlying global (population status or ocean conditions)
and local (habitat status or human stressors) factors that affect the performance of individual assets.
We believe this analogy is useful because it helps clarify the way in which the performance of
Conservation Units is nested within the performance of the Fraser aggregate, and helps clarify how
there is a complex arrangement of factors that can affect performance. Like an investment portfolio,
it is a daunting task to summarize the complexity of information that describes freshwater habitat
conditions in a way that helps a person understand the drivers of recent trends in Fraser River
sockeye salmon. The dashboards help in this regard by providing a snapshot of all the factors that
contribute to the performance of a Conservation Unit. Furthermore, as with investment portfolios,
past performance is not necessarily a predictor of future performance (particularly in consideration of
the effects of climate change, see Hinch and Martins 2011).
Walking through the results presented for the Quesnel summer timing Conservation Unit illustrates
how a user can quickly assess freshwater conditions for an individual Conservation Unit. Using
Figure 39, the biological data provides an orientation on the dramatic declines in productivity since
the late 1980s, in a period with years of relatively high escapement. The CU is set within a
geographic context to orient the user to the location of its nursery lake within the central interior of
the basin. Despite the declines, the population status has been rated as relatively good with high
confidence. From a habitat vulnerability point of view, the Quesnel appears reasonably resilient due
to the moderate migration distance, a relative large proportion of spawning locations being buffered
by lakes, and a relatively moderate to large amount of nursery lake capacity.
Using Figure 40 to examine the freshwater stressors on this CU, we can quickly assess key threats on
habitats. First, there is a low human population density across spawning and rearing habitat and a
much higher density across the migration corridor. Disturbance to lands influencing spawning and
nursery lakes is mostly due to Mountain Pine Beetle, while impacts on migration corridors are split

52
across a variety of disturbance types (agriculture, urbanization, Mountain Pine Beetle). Road density
associated with land use activities is moderate relative to other CUs. Small hydro is largely non-
existent, while placer mining claims are very high in areas affecting the nursery lake and migration
corridor. A relative abundance of other mines is also prevalent along the migration corridor.
Allocation of water is dominated by agricultural uses, though the number of licenses is dominated by
urban uses. Although the total allocation is at the lower end of water allocation across all CUs, the
density of water allocation restrictions is relatively high in areas influencing nursery lakes and
migration corridors of this CU. Appendix 3 provides the dashboard summaries for all lake
Conservation Units across the Fraser River basin.
4.2 Assessment across Conservation Units
We undertook three tasks to assess whether freshwater habitat conditions and stressors on habitats
have contributed to the recent declines in Fraser River sockeye salmon (summarized in detail below).
First, we summarize key results from the recent work of scientists examining factors that could
explain the declines (Selbie et al. in Appendix C of Peterman et al. 2010). This understanding is
important for prioritizing our own analytical efforts and developing testable hypotheses that are
comparable with the findings from other studies. Second, we analyzed the habitat and stressor data
(generated through this effort) to test whether they could explain declines in productivity. Lastly, for
those habitat and stressor variables for which we had time series data (i.e., forest harvesting,
Mountain Pine Beetle disturbance, summer air temperatures across adult migration, and spring air
temperatures at nursery lakes) we examined correlations with the total salmon and juvenile
productivity indices developed by Peterman et al. (2010).
Selbie et al. (in Appendix C of Peterman et al. 2010), tested four hypotheses to examine whether
changes in freshwater habitat conditions have contributed to the declines of Fraser River sockeye
salmon. As the foundation to their analyses, they generated two measures of sockeye salmon
productivity that are relevant to our study (see Peterman et al. 2010 for more details). A measure of
freshwater or “juvenile productivity” was calculated as the annual abundance of juveniles (fry or
smolts) per effective female spawner. This measure represents productivity across freshwater life
stages. A measure of “total productivity” was calculated as the annual abundance of adult recruits
per effective female spawner. This measure represents the productivity across all life stages of
sockeye salmon, from eggs to adult returns. Noteworthy in the way they calculated total productivity,

53
was the fact that the number of adult recruits included losses due to en-route mortality. To account
for the effect of density dependence and allow researchers to more easily detect the effect of other
factors on productivity, data from each stock and life stage were fit to a standard Ricker stock-
recruitment equation. Annual residuals from the best-fit function were then obtained for each stock
and used as indices of total productivity and juvenile productivity.
Using these indices of productivity, Selbie et al. first analyzed the trend in the decline of total
productivity across stocks and whether they could be explained by habitat indicators or measures of
landscape level stress on those stocks. Next, where data were available they examined whether
changes in the productive capacity of nursery lakes (using photosynthetic rate, zooplankton biomass,
and spawner production) could explain the declines. Given a limited data set, they also examined
patterns of growth and survival of juveniles in nursery lakes. Lastly, they examined smolt
outmigration timing to see if any discernable changes have occurred over time. Based on these
analyses, they were unable to find any quantitative evidence to support the general hypothesis that
declines in total productivity are related to changes in freshwater habitat conditions. An interesting
observation, however, was that they found a relationship between migration distance and the trend in
the decline across stocks (i.e., those with longer migration distances demonstrated more dramatic
declines).
Similar to the work of Selbie et al. (in Appendix C of Peterman et al. 2010), we examined whether
variation in the intensity of freshwater habitat stressors could be related to trends in total productivity
(using the Ricker model residuals as described above and derived by Peterman et al. (2010)). In other
words, we would expect the declines in total productivity to be more extreme in heavily impacted
CUs if conditions of freshwater habitat are contributing to the declines. We applied the same
analytical approach because of improvements in the data we generated to represent habitat
vulnerability and freshwater stressors with our work. Similar to Selbie et al., we calculated the slope
of the regression of Ricker residuals across brood years 1984-2004 for each of 17 sockeye salmon
CUs (those with time series of total productivity). We then evaluated the trend in Ricker residuals in
relation to: (1) our landscape level measures of stress on nursery lakes (i.e., derived using GIS as
described in Section 3.0), and (2) measures of habitat vulnerability (i.e., migration distance, nursery
lake area, nursery lake productivity, total spawning extent, and ratio of lake influenced spawning).
The stressors used in the analysis included: road density, stream crossing density, level of forest

54
harvesting (as a percent of area), accumulated level of Mountain Pine Beetle disturbance (as a
percent of area), agriculture land (as a percent of area), urban land (as a percent of area), total water
license allocations (volume/per year/ha), active mines (count), placer claims (count), and small scale
hydroelectric operations (count).
We initially developed a simple correlation matrix of our full suite of freshwater vulnerability and
stressor indicators in relation to the trend in Ricker residuals (Table 15). Migration distance (r = -
0.57) and percent Mountain Pine Beetle disturbance (r = -0.46) were the only variables to show even
a moderate inverse correlation with the trend, though migration distance and percent of Mountain
Pine Beetle disturbance were highly correlated with each other (r = 0.95). There were many other
strong correlations across the freshwater indicators (Table 15) which indicates a strong potential for
confounding. To address this concern we modified our analytical approach from that of the simple
linear regression model used by Selbie et al. (in Appendix C of Peterman et al. 2010) to an
information-theoretic framework (Burnham and Anderson 1998) using Akaike Information Criterion
(AIC) to identify regression models that could best explain the trend in Ricker residuals across
sockeye salmon CUs (see Table 16). AIC is a relative ranking statistic, with AIC values being
interpreted in terms of the magnitude of the differences among candidate models rather than the
magnitude of any particular value (Thompson and Lee 2002). The approach gives a formal
accounting of the relative plausibility of the estimated models and can be helpful in selecting the
most plausible models (i.e., those with the lowest AIC scores) when confounding occurs (Paulsen
and Fisher 2005).
We assessed 67 possible combinations of the freshwater vulnerability and stressor indicators to the
indices of total productivity for sockeye salmon (i.e., trend in Ricker residuals). Each of the three
highest ranked models (i.e., most plausible) had migration distance as a predictor variable, with the
highest ranked model having migration distance as the sole predictor. The second and third ranked
models had percent of Mountain Pine Beetle disturbance and road density as additional predictors,
respectively. The explanatory value of the top model (migration distance alone) was low, however,
with an adjusted R2 = 0.324. Inclusion of the highly correlated percent of Mountain Pine Beetle
disturbance in the second ranked model improved explanatory value only slightly (R2 = 0.341), while
the third ranked model had an even lower explanatory value (R2 = 0.296).

55
Generally consistent with the earlier findings of Selbie et al., our analyses indicated that migration
distance was the only freshwater variable to show an appreciably strong (negative) relationship to the
trend in residuals (i.e., sockeye salmon CUs with greater migration distances seem to have done
more poorly in recent years). We have no direct explanation for why migration distance emerged as a
predictor of recent trends in total productivity. It may relate to differential exposure to a suite of
stresses along the migration route; stresses which we attempted to quantify with our index of
cumulative stress developed for migration corridor of each CU (see description in Section 4.1). Our
index of the cumulative stress along migration corridors was negatively correlated with total
productivity (r = -0.497), but this cumulative stressor indicator was also highly correlated with
migration distance (r = 0.678), so these 2 factors are confounded. None of our other indicators of
cumulative stress showed any correlation with total productivity (mainstem spawning, r = 0.163;
tributary spawning, r = -0.0047; nursery lakes, r = -0.113). As noted by Selbie et al. (2010) distance
from the ocean is also significantly correlated to other factors reflecting watershed position,
including elevation and latitude, so migration distance may also be capturing parallel influences on
total productivity that are unrelated to stresses associated with human activities.
For those habitat and stressor variables for which we had time series data (i.e., forest harvesting,
Mountain Pine Beetle disturbance, summer air temperatures across adult migration, and spring air
temperatures at nursery lakes) we examined correlations with total and juvenile productivity indices
developed by Peterman et al. (2010) as described above. For these analyses we aligned the years of
forest harvesting and Mountain Pine Beetle disturbance with the year of juvenile productivity that
represented the year of fry emergence (i.e., brood year + 1). This alignment of years was used to best
test for the potential influence of forest disturbances on egg-to-fry survival. For analyses of habitat
conditions, we aligned the years of summer air temperature along the migration corridor to brood
years, and years of spring time air temperature at the nursery lake to the year of ocean entry (i.e.,
brood year +2). Again this alignment was necessary to accurately test for the effect of available
habitat conditions on total productivity of sockeye salmon. Note that the effects of summer air
temperature on en-route mortality are already accounted for in estimates of recruitment and Ricker
residuals (as described above); this analysis explores any additional effects on total productivity.
Results of the analysis between total productivity and the two habitat indicators are presented in
Table 17. To account for the fact that we are conducting multiple comparisons, we used a Bonferonni

56
adjustment to the alpha level to detect significance (i.e., P value divided by the number of
comparisons or P = 0.005 in our case). This adjustment was made because by chance alone we would
have expected 1 in 20 stocks to have a significant, though spurious, correlation. When examining
correlations between total productivity and summer air temperature across adult migration, 16 stocks
had negative correlations (i.e., years with warm summer air temperature along the migration corridor
tended to be associated with years of lower total productivity), though only 1 was significant.
Similarly when examining correlations between total productivity and spring time air temperatures at
the nursery lake, 14 stocks had negative correlations (i.e., years with warm spring air temperatures at
the nursery lake tended to be associated with years of lower total productivity), though none were
significant. The plausibility of a mechanism underlying a relationship between air temperature and
total productivity is questionable given that the total productivity index already accounts for en-route
mortality (as described above). In contrast, however, the plausibility of a potential relationship
between nursery lake air temperatures and total productivity is more likely (see Section 2.2.1).
Despite the plausibility of the underlying mechanisms there are no significant correlations and a lack
of consistency in direction of the correlation coefficients across stocks, which suggest the
relationship is spurious or that some air temperature indicators are weakly correlated with other
factors that have an influence on total productivity. In examining correlations between juvenile
productivity, forest harvesting, and Mountain Pine Beetle disturbance, we found no significant
correlations across any of the 8 stocks for which there are juvenile productivity data.
4.3 Summary and conclusions
In this report we use the best available data to quantitatively describe the status of Fraser River
populations of sockeye salmon (Section 2.1), vulnerability of freshwater habitats that support
migration, spawning, and rearing life stages (Section 2.2), and human stressors interacting with those
habitats (Section 3.0). Table 18 provides a simple summary of the population status, habitat
vulnerability, and relative level of cumulative stress on habitats for each Conservation Unit. We then
summarize these data for each Conservation Unit to help gain insights into possible hypotheses about
freshwater influences on different Fraser River populations (Section 4.1 and Appendix 3). Finally,
we analyze these data to determine whether habitat vulnerability and freshwater stressors are related
to trends in productivity or current population status (Section 4.2).

57
For freshwater life stages of Fraser River sockeye salmon, there is a complex pathway of effects that
results from changes in human stressors to changes in habitat conditions to changes in abundance of
sockeye salmon populations. Many habitat conditions can be affected by multiple stressors and the
biological effects on survival and growth interact to produce an outcome at the population level (see
Figure 41). Unless data have been collected using experimental design principles, it is often difficult
or impossible to conduct a statistical test of cause and effect that will answer whether a particular
stressor or group of stressors has resulted in a particular habitat change or population level effect.
The lack of experimental design certainly relates to our inability to test for cause and effect
relationships between the freshwater environment and Fraser sockeye salmon declines. An
alternative approach is to evaluate the quantitative and qualitative evidence in a structured and
rigorous way, where the outcome of the exercise is an evaluation of the “weight of evidence” to
reach a conclusion about significance. Stewart-Oaten (1996) propose the use of a series of questions
to structure scientific evaluations of evidence for determining cause and effect. This series of
questions can be asked in order which provide a summary of the weight of evidence in favour or
against a particular cause and effect relationship.
Our assessment of the cumulative effect of freshwater stressors suggests that the recent declines in
Fraser River sockeye salmon are unlikely to be due to changes in freshwater habitats (see Table 19).
An important piece of evidence in reaching this conclusion is that juvenile survival has remained
relatively stable across CUs where data are available (see Peterman et al. 2010), even though there is
substantial variation in stressor intensity across CUs. In the literature, there is strong evidence that
the stressors examined here can lead to declines and extinctions of populations in a variety of
species, including sockeye salmon. A consideration of individual stressors (see Table 20, Table 21,
and Table 22) suggests that the highest levels of overall stress are generated by forest harvesting and
roads, while water use and large hydro also generate significant stress for individual CUs.
The effect of freshwater stressors on sockeye salmon population resilience is difficult to detect in this
type of analysis. Higher density independent survival can lower resilience without noticeable effects
on population numbers. High freshwater resilience allows populations to recover quickly from
transient stressors and to compensate for lower spawning escapements. Stressors that induce higher
density independent mortality may have no noticeable effects unless another factor creates additional
stress on the population.

58
5.0 State of the science
The state of the science related to freshwater ecology can best be described by understanding the
state of knowledge and state of data related to Fraser River sockeye salmon. In regards to the state of
knowledge, sockeye salmon are one of the most well studied fish species in the world and agencies
have historically emphasized monitoring of key Fraser River populations. Consequently, their basic
ecology is relatively well understood (Burgner 1991). This strong foundation has helped with
implementation of the Wild Salmon Policy (DFO 2005). Specific efforts have focused on delineating
sockeye salmon populations into distinct Conservation Units using three major axes: ecology, life
history, and molecular genetics (Holtby and Ciruna 2007). The cause-effect pathways of natural and
human stresses on stream and watershed processes leading to alterations of sockeye habitats are well
documented. Consequently, our general understanding of the interaction among freshwater life
stages, habitats, and human stressors has allowed for the identification of defensible indicators to
monitor habitat condition (Stalberg et al. 2009). However, our knowledge about the specific effect of
human stresses on sockeye salmon habitats is largely dependent on the frequency and intensity of
disrupting events and the vulnerability of affected habitats. The strength and form of the relationship
between a particular stressor (or the cumulative / synergistic effect of multiple stressors), changes in
freshwater habitat condition, and related changes in sockeye salmon survival / productivity remains
largely unknown. Moreover, we lack an integrated understanding of how freshwater and marine
conditions influence survival and productivity at different life stages and across the entire life cycle.
In regards to the state of data, we have reasonable data related to adult abundance (recruits and
spawners), extent of spawning habitats, and nursery lake conditions for strong stocks in the Fraser
River. For the human stressors considered in this report, we have reasonable data describing the
spatial distribution and intensity of disturbance related to hydroelectric development, forest
harvesting, road development, and Mountain Pine Beetle. We also have a reasonable understanding
of the spatial distribution of mining, urbanization, and agriculture. However, there are substantial
data gaps. Given the emphasis on monitoring strong stocks, we lack good information describing
abundance and survival across freshwater life stages of many weak stocks and in-river populations.
Similarly, there is a lack of data that consistently quantifies the quality of migratory, spawning, and
rearing habitats across all Conservation Units (i.e., both strong and weak stocks). We lack time series
data for almost all human stressors considered in this report and for a subset of stressors we lack data
on intensity of disturbance (e.g., water use, log storage, agriculture, mining, and urbanization).

59
6.0 Recommendations
As discussed in Section 1.0, freshwater habitats are known to contribute to the overall diversity and
resilience of sockeye salmon. Thus, protecting freshwater habitats is important to the conservation of
Fraser River sockeye salmon, even though recent declines are not likely to be directly linked to
deterioration in habitat conditions.
Section 5.0 highlights that there are significant data gaps which hinder our ability to effectively
manage sockeye salmon populations, habitats, and human activities. Long term and consistent
monitoring of a mix of sockeye salmon populations embedded within well planned geographically-
based experimental comparisons would help scientists and managers better understand the cause and
effect relationships between human activities and resultant sockeye salmon habitat and population
responses. However, identifying unique monitoring recommendations that will help improve the
state of knowledge and quality of data is a challenge because others working with the sole purpose of
identifying monitoring requirements have already reported on key information gaps and the reasons
to address them (e.g., Day 2007; Nelitz et al. 2008; Stalberg et al. 2009; Selbie et al. in Appendix C
of Peterman et al. 2010; also see Wild Salmon Policy advisories from Pacific Fisheries Resource
Conservation Council at www.fish.bc.ca). Despite our heightened awareness of the needs, below we
reiterate what we believe are some key recommendations to improve our ability to conduct scientific
inquiries into cause and effect, and improve decision making related to land use, water use, and
management of freshwater habitats and sockeye salmon populations.
To improve our understanding about survival at critical freshwater life stages scientists need
better estimates of juvenile abundance, overwinter survival, and mortality during smolt outmigration.
Some data are currently available though for only a few populations and with limited time series.
This lack of information means it is difficult to conclusively test for cause and effect between
freshwater habitat conditions, human stressors, and salmon productivity across many Conservation
Units. If survival in the freshwater life stage is found to be a more important contributor than
determined in this report, then management actions can be taken to mitigate impacts on survival.
To improve our understanding about population status across Conservation Units scientists
need more information about the abundance and distribution of small lake and all river CUs, though
we recognize that filling this gap may be impractical for river CUs. Existing programs for monitoring

60
fry and adults are essential for understanding status, but historically resources have been dedicated to
large lake Conservation Units. This emphasis is inconsistent with the Wild Salmon Policy which
places importance on protecting diversity of populations. Ensuring conservation of small CUs could
have dramatic effects on harvest policies and in-season management.
To improve our understanding about habitat status across Conservation Units scientists need
information on habitats monitored in a consistent manner on a regular basis across a larger number of
rivers and nursery lakes (i.e., expanded in-river monitoring and limnology programs). The current
approach to monitoring habitat condition and stressors is largely ad hoc, with monitoring
responsibilities distributed across many different government agencies. Habitat evaluations tend to
focus on a particular issue (i.e., linkage to a specific habitat variable or stressor activity) in a
particular location using a particular methodology. Without a consistent and repeatable methodology
much of the information on trends is lost and comparisons across Conservation Units are not
possible. In addition to monitoring habitat condition and stressors, it is equally important to track and
rigorously monitor the policies and practices taken to protect freshwater habitats and reduce the
adverse effect of stressors.
To improve our understanding about the population level effects of stressors on freshwater
habitats scientists need more precise estimates of the biological consequences of disturbance as a
function of increasing stress. For most human stressors the general mechanisms of effect are known,
but estimates of the population level significance of a given stressor level are crude, especially when
occurring in the presence of other types of stressors. Attempts to define such thresholds have had
limited success (e.g., determining Equivalent Clearcut Area thresholds), but their delineation is a key
requirement for more defensible decision making. Once available, this information could be used to
model the “environmental envelope” for persistence of sockeye salmon in freshwater habitats so that
future conflicts might be better anticipated and avoided. Given the importance and extent of
legislation and policies designed to govern land and water use, we believe this gap is critical to fill.
Without this information managers can not ensure that policies are achieving their intended
objectives of protecting freshwater habitats and reliant fish species like sockeye salmon.
To improve transparency in the science and related decision making scientists, managers, and
the public need information that is more accessible. The high level of public interest in the work of

61
the Cohen Commission highlights the large number and diversity of audiences interested in
understanding the complex ecology of Fraser River sockeye salmon. Similarly, the challenges of
independent scientists working for the Cohen Commission to acquire and compile the necessary data
in a useable format for analyses have revealed the lack of integration in data collection and sharing
across and within government agencies.
For improved access to information by stakeholders, better communication tools are needed to relay
the status of sockeye salmon and clarify expectations for returns in the face of large uncertainties.
Though very detailed and technical in nature, the dashboard summaries in this report could be used
as a model for condensing large quantities of information into a more digestible summary for the
informed public. Web-based platforms such as the Community Mapping Network
(http://www.cmnbc.ca/) or HectaresBC (http://www.hectaresbc.org/app/habc/HaBC.html) could be
expanded to consolidate and convey population, habitat, and stressor information for sockeye
salmon, or examples from elsewhere in the Pacific Northwest could be used to develop a new model
for summarizing and reporting out on fish and fish habitat information (e.g., Columbia Basin’s Fish
and Wildlife Authority’s (CBFWA) Monitoring Strategies and Status of the Resource reporting,
http://www.cbfwa.org/index.cfm).
For improved access to information by scientists, formal data sharing agreements, pooling of
resources for monitoring, and more integrated decision making are needed. The current lack of
consistency and integration of monitoring programs exist because many federal and provincial
agencies are responsible for collecting, summarizing, and reporting out on key variables of relevance
to Fraser River sockeye salmon (e.g., Fisheries and Oceans Canada, Environment Canada, Ministry
of Natural Resource Operations, Ministry of Forests, Mines, and Lands, and Ministry of
Environment). Others have commented in more detail on the need and ways to improve integration
(Day 2007; Nelitz et al. 2008), but at the core is a need to have a well resourced body of scientists (in
terms of staff and funding) to coordinate an integrated or harmonized fish and fish habitat monitoring
program. A useful working example is the Columbia Basin Fish and Wildlife Authority (CBFWA)
and its associated Members Advisory Group comprised of federal, state, and tribal entities in the
Columbia River basin (member’s charter available at
http://www.cbfwa.org/RegionalIssues/Correspondence/CBFWA/CBFWACharterAdopted_20April2
010_FINAL.pdf).

62
7.0 Figures
TotalFraserRiversockeyeproductivity
(Returns/Spawner),4‐yravg.
0
2
4
6
8
10
1952 1960 1968 1976 1984 1992 2000 2008
Returnyear
Returns/Spawner
(movingaverage)
Figure 1. Four-year moving average of total adult returns per spawner across all Fraser River sockeye salmon
stocks divided by total spawners 4 years before. Note this averaging reduces annual variation. The
horizontal dashed line indicates the productivity at which the population can replace itself. Data from
the Pacific Salmon Commission.
FW stressors examined in this
study (logging, beetles, mining,
hydro, urbanization, agriculture)
Level of regulation & enforcement (mitigate potential impact)
Quality of spawning, rearing or migratory habitats
egg-fry, fry-smolt, or downstream migration survival
Recruits / Spawner over whole life cycle
estuary &
ocean factors,
harvest
management
Freshwater
Factors
Other FW stressors
(e.g., temperature,
contaminants)
Ability of watersheds to naturally mitigate potential impacts
Figure 2. Conceptual model of the factors considered by this study (boxes with solid lines) and relevant factors
considered by other Cohen Commission studies (boxes with dashed lines).

63
Figure 3. Overview of the Fraser River basin, watershed boundaries (in shades of grey), and nursery lakes (in
black) for all lake sockeye salmon Conservation Units. Note that different shades of grey are used to
represent the upstream watershed boundaries for different CUs. In some cases several CUs overlap and
as a result their boundaries only appear once.

64
Figure 4. Summary of the status for 36 sockeye salmon Conservation Units into four risk categories: IV – status
probably poor, but little information; III – status poor, high confidence; II – status probably good, high
uncertainty; and I – status good, high confidence. Eleven other CUs are classified as having insufficient
information (Early Summer – Chilko, Fraser, Indian / Kruger, and Nadina; Summer – Francois and
McKinley; Late – Kawkawa; River: Fraser Canyon, Middle Fraser, Thompson, and Upper Fraser).
Image and data from Pestal and Cass 2009.
Figure 5. Modified conservation status for some CUs based on work of Grant et al. (2010). Blue squares indicate
conservation status that did not change as a result of Grant et al.’s work. Grey circles represent
conservation status as determined by Pestal and Cass (2009), and red diamonds represent modified CU
status based on input from Grant et al. (2010).
0
1
2
3
4
5
0 1 2 3 4 5
Severity
Uncertainty
Same
Grant et al. 2010
Pestal & Cass 2009 StuartKamploops (L)Taseko
Cultus
Shushwap
Complex (L)
ChilkoPitt
Fraser
Takla/Trambleu
r
(ES)
Stuart Lillooet
Quesnel Shushwap Complex (ES)
Harrison DS,
Kamploops (ES)
Chilliwack
L. Fraser
Widgeon
Takla/Trambleur (S)
Nahatlatch,
Bowron
Harrison US,
Seton, Anderson,
Francois
Uncertaint
y
Severity
II IV
I III

65
A
B
Figure 6. Panel A: Timing of smolt outmigration as a function of latitude across multiple nursery lakes in BC and
Alaska. Panel B: Timing of lake ice breakup within a single nursery lake in Alaska. Images from
Burgner (1991).
Figure 7. Summary of the habitat vulnerability for all lake sockeye salmon Conservation Units using three
independent indicators of habitat quantity / quality: migration distance (x-axis), total area of nursery
lakes (y-axis), and ratio of lake influence to total spawning (size of circles). The Conservation Units with
the most vulnerable habitats would appear as small bubbles in the bottom left corner of this graph,
while Conservation Units with the least vulnerable habitats would appear as large bubbles in the top
right corner. A summary of habitat vulnerability for the 6 river-type Conservation Units is unavailable
due to a lack of information on locations of habitat use.
Migration distance (km)
Area of nursery lakes (ha)
Ratio of lake influence to total spawning

66
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
400,000
1975 1980 1985 1990 1995 2000 2005 2010
Area harvested (ha)
0
10,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
90,000
100,000
Volume harvested (1000's m3)
Area harvested
Volume harvested
Figure 8. Area (in hectares) and volume (in 1,000s m3) of harvested forest in British Columbia from 1975 to 2007
(data from Statistics Canada 2009).
85.0
10.0 0.0 0.0 0.0 0.0 0.0 5.0 0.0 0.0
.00 .02 .04 .06 .08 .10
11.1
22.2
5.6 5.6 5.6 5.6
16.7 22.2
5.6 0.0
.00 .02 .04 .06 .08 .10
50.0
13.3 20.0
10.0 6.7
.00 .02 .04 .06 .08 .10
100.0
0.0 0.0 0.0 0.0
.00 .02 .04 .06 .08 .10
Figure 9. Frequency distribution of the level of forest harvesting within the “zones of influence” of each habitat
type across all Fraser River lake sockeye salmon Conservation Units. Numbers above bars represent
percentage of CUs in the respective bin.
Lake influence / mainstem spawning
Nursery lake rearing
Lake inlet / tributary spawning
Migration corridors
Level of forest harvesting across different
habitat areas (as proportion of area)
Percent of CUs

67
Figure 10. Spatial distribution of forest harvesting cutblocks relative to watershed boundaries (light grey shading)
for all lake sockeye salmon Conservation Units. Forest harvesting cutblocks (scattered points of dark
grey shading) represent the cumulative level of forest harvesting across four time periods. Nursery
lakes are in black.
1981-1995 1986-2000
1991-2005 1996-2010

68
0.00
0.02
0.04
0.06
0.08
0.10
1994 1999 2004 2009 0.00
0.02
0.04
0.06
0.08
0.10
1994 1999 2004 2009
0.00
0.02
0.04
0.06
0.08
0.10
1994 1999 2004 2009
0.00
0.02
0.04
0.06
0.08
0.10
1994 1999 2004 2009
Harrison (d/s) - Late
Pitt - Early Summer
Quesnel - Summer
Stuart - Early Stuart
Taseko - Early Summer
Shuswap Complex - Late
Figure 11. Time series of the level of forest harvesting within “zones of influence” for each habitat type across six Fraser River lake sockeye salmon
Conservation Units.
Nursery lake rearing Migration corridors
Lake influence / mainstem spawning Lake inlet / tributary spawning
Level of forest harvesting across different
habitat areas
(
as
p
ro
p
ortion of area
)

69
15.0
30.0
45.0
5.0 0.0 5.0
0123456
66.7
27.8
5.6 0.0 0.0 0.0
0123456
63.3
36.7
0.0 0.0 0.0 0.0
0123456
0.0 13.3
80.0
6.7 0.0 0.0
0123456
Figure 12. Frequency distribution of the density of roads (km / km2) within the “zones of influence” of each
habitat type across all Fraser River lake sockeye salmon Conservation Units. Numbers above bars
represent percentage of CUs in the respective bin.
15.0
45.0
15.0 15.0
5.0 5.0 0.0 0.0
0 1 2 3 4
16.7
66.7
11.1 0.0 0.0 0.0 0.0 5.6
0 1 2 3 4
33.3
60.0
3.3 3.3 0.0 0.0 0.0 0.0
0 1 2 3 4
0.0
20.0
6.7
66.7
6.7 0.0 0.0 0.0
0 1 2 3 4
Figure 13. Frequency distribution of the density of road-stream crossings (# / km2) within the “zones of influence”
of each habitat type across all Fraser River lake sockeye salmon Conservation Units. Numbers above
bars represent percentage of CUs in the respective bin.
Lake influence / mainstem spawning
Nursery lake rearing
Lake inlet / tributary spawning
Migration corridors
Density of roads (km / km2)
Lake influence / mainstem spawning
Nursery lake rearing
Lake inlet / tributary spawning
Migration corridors
Density of road-stream crossings (# / km2)
Percent of CUs Percent of CUs

70
35.0
5.0 0.0 5.0 0.0 5.0
20.0 20.0
10.0
0.0
020 40 60 80 100
22.2
11.111.1
5.60.00.05.65.6
11.1
0.00.05.60.00.00.00.05.65.6
11.1
0.0
020 40 60 80 100
26.7
13.3 16.7
3.3 0.0
16.7
6.7 10.0 6.7
0.0
020 40 60 80 100
46.7
10.0
43.3
0.0 0.0 0.0 0.0 0.0 0.0 0.0
020 40 60 80 100
Figure 14. Frequency distribution of the accumulated level of Mountain Pine Beetle disturbance from 1999 to 2008
within the “zones of influence” of each habitat type across all Fraser River lake sockeye salmon
Conservation Units. Numbers above bars represent percentage of CUs in the respective bin.
Lake influence / mainstem spawning
Nursery lake rearing
Lake inlet / tributary spawning
Migration corridors
Accumulated level of Mountain Pine Beetle disturbance
from 1999 to 2008 (as percent of habitat area)
Percent of CUs

71
Figure 15. Spatial distribution of the accumulated level of Mountain Pine Beetle disturbance from 1999 to 2008
(dark grey shading) relative to watershed boundaries (light grey shading) for all lake sockeye salmon
Conservation Units. Nursery lakes are in black.
1999 1999-2002
1999-2005 1999-2008

72
0
20
40
60
80
100
1999 2001 2003 2005 2007 0
20
40
60
80
100
1999 2001 2003 2005 2007
0
20
40
60
80
100
1999 2001 2003 2005 2007
0
20
40
60
80
100
1999 2001 2003 2005 2007
Harrison (d/s) - Late
Pitt - Early Summer
Quesnel - Summer
Stuart - Early Stuart
Taseko - Early Summer
Shuswap Complex - Late
Figure 16. Time series of the accumulated level of Mountain Pine Beetle disturbance from 1999 to 2008 within “zones of influence” of each habitat type
across six Fraser River lake sockeye salmon Conservation Units.
Nursery lake rearing Migration corridors
Lake influence / mainstem spawning Lake inlet / tributary spawning
Accumulated level of Mountain Pine Beetle
disturbance
(
as
p
ercent of habitat area
)

73
Figure 17. Aerial photo overview of the lower reaches of the Fraser River and estuary in 2009. Boxes A-D
delineate areas with the highest relative concentrations of log storage across all years examined.
A
BC
D

74
Figure 18. Overview of the distribution of main categories of mines across the Fraser River basin.

75
Figure 19. Schematic representation of the development of the Seton / Cayoosh diversion (from Roscoe and Hinch
2008).
-
10,000
20,000
30,000
40,000
50,000
1938 1949 1960 1971 1982 1993 2004
Escapement (Portage)
-
20,000
40,000
60,000
80,000
100,000
Escapement (Gates)
Portage Creek
Gates Creek / Channel
Figure 20. Portage Creek and Gates Creek sockeye salmon escapement from 1938 to 2006 (data from DFO
NuSEDS database).

76
Figure 21. Nechako River watershed and location of Kenny Dam (Map from NFCP 2005).
95.0
5.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0 2 4 6 8 10
88.9
5.6 5.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0 2 4 6 8 10
86.7
3.3 0.0 0.0 3.3 0.0 0.0 0.0 6.7 0.0
0 2 4 6 8 10
20.0
3.3
66.7
10.0 0.0 0.0 0.0 0.0 0.0 0.0
0 2 4 6 8 10
Figure 22. Frequency distribution of the number of small scale hydroelectricity installations within the “zones of
influence” of each habitat type across all Fraser River lake sockeye salmon Conservation Units.
Numbers above bars represent percentage of CUs in the respective bin.
Lake influence / mainstem spawning
Nursery lake rearing
Lake inlet / tributary spawning
Migration corridors
Percent of CUs
Number of small scale hydroelectric installations

77
Figure 23. Spatial distribution of small scale hydroelectricity installations (squares with dots) relative to watershed
boundaries (grey shading) for all lake sockeye salmon Conservation Units. Nursery lakes are in black.

78
Figure 24. Intake and headpond of the Douglas Creek generating station (a typical installation). The intake weir
creates a headpond and raises the water so a portion of stream flow can enter the penstock. The size of
the headpond is determined by the topography at the intake location. This installation is upstream of an
alluvial fan on which Harrison -downstream migrating-Late timing sockeye salmon spawn (image from
http://cloudworksenergy.com/projects/photo-tour/).
0
500,000
1,000,000
1,500,000
2,000,000
2,500,000
1981 1986 1991 1996 2001 2006
Population size
Upper Fraser
Thompson
Nechako
Lower Fraser
Figure 25. Human population size in the Fraser River basin by region from 1981 to 2006 (data from Statistics
Canada 2009).

79
80.0
5.0 5.0 0.0 0.0 0.0 0.0 0.0 0.0 5.0
0.2 .4 .6 .8 1
100.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.2 .4 .6 .8 1
96.7
0.0 0.0 0.0 3.3 0.0 0.0 0.0 0.0 0.0
0.2 .4 .6 .8 1
0.0
46.7
6.7
16.7
6.7 3.3 3.3 10.0 3.3 3.3
0.2 .4 .6 .8 1
Figure 26. Frequency distribution of the area of urban land within “zones of influence” of each habitat type across
all Fraser River lake sockeye salmon Conservation Units. Numbers above bars represent percentage of
CUs in the respective bin.
Lake influence / mainstem spawning
Nursery lake rearing
Lake inlet / tributary spawning
Migration corridors
Level of urban land (as proportion of habitat area)
Percent of CUs

80
120°W122°W124°W126°W
56°N
54°N
52°N
50°N
05010015020025 km
´
Figure 27. Spatial distribution of urban areas (dark grey shading) relative to the watershed boundaries (light grey
shading) for all lake sockeye salmon Conservation Units. Nursery lakes are in black.

81
10
100
1,000
1986 1991 1996 2001 2006
Population density (# / km2)
Figure 28. Time series of average human population density along migration corridors of all lake sockeye salmon
Conservation Units. Only CU labels for the four highest and four lowest population densities are
represented on this graph. Time series for each CU represented separately in Appendix 3.
0.1
1
10
100
1986 1991 1996 2001 2006
Population density (# / km2)
Figure 29. Time series of average human population density adjacent to rearing and spawning habitats for all lake
sockeye salmon Conservation Units. Only CU labels for the four highest and four lowest population
densities are represented on this graph. Time series for each CU represented separately in Appendix 3.
Pitt – Early Summer
Cultus – Late
Harrison (d/s) – Late
Harrison
(
u/s
)
–
Late
Nadina – Early Summer
Francois – Late
Bowron – Late
Indian / Kruger – Early
Stuart – Early Stuar
Stuart – Summer
Takla / Trembleur – Early Stuart
Takla / Trembleur – Summer
Pitt – Early Summer
Cultus – Late
Kawkawa – Late
Nahatlatch
–
Late

82
0
5
10
15
20
25
30
1870 1890 1910 1930 1950 1970 1990 2010
Number of farms (1,000s)
4.0
4.2
4.4
4.6
4.8
5.0
Area in Agricultural Land Reserve
(millions of hectares)
Figure 30. Number of farms in British Columbia from 1881 to 2006 (diamonds, data from Statistics Canada 2009)
and total area of the province within the Agricultural Land Reserve from 1974 to 2007 (solid line, data
from BC MOE 2008).
0
1
2
3
4
5
6
1986 1991 1996 2001
Animal unit equivalents per unit area (#/ha)
Figure 31. Number of livestock (cattle, pigs, chickens) per unit area (# / ha) in Abbotsford. Number of livestock is
represented as an animal unit equivalency (i.e., 1 cow = 3 pigs = 75 hens). Data from Statistics Canada
as reported in Smith et al. (2007).

83
80.0
5.0 5.0 5.0 0.0 0.0 5.0 0.0 0.0 0.0
0.2 .4 .6 .8 1
94.4
5.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.2 .4 .6 .8 1
86.7
13.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.2 .4 .6 .8 1
0.0 10.0
33.3
50.0
6.7 0.0 0.0 0.0 0.0 0.0
0.2 .4 .6 .8 1
Figure 32. Frequency distribution of the level of agricultural land within “zones of influence” of each habitat type
across all Fraser River lake sockeye salmon Conservation Units. Numbers above bars represent
percentage of CUs in the respective bin.
Lake influence / mainstem spawning
Nursery lake rearing
Lake inlet / tributary spawning
Migration corridors
Level of agricultural land (as proportion of habitat area)
Percent of CUs

84
120°W122°W124°W126°W
56°N
54°N
52°N
50°N
0 50 100 150 20025 km
´
Figure 33. Spatial distribution of agricultural areas (dark grey shading) relative to the watershed boundaries
(light grey shading) for all lake sockeye salmon Conservation Units. Nursery lakes are in black.

85
86.4
0.0 4.5 0.0 0.0 9.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
02500 5000 7500 10000
100.0
0.0 0.0 0.0 0. 0 0.0 0.0 0.0 0. 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
02500 5000 7500 10000
100.0
0.0 0.0 0.0 0. 0 0.0 0.0 0.0 0. 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
02500 5000 7500 10000
0.0
37.5
15.6
28.1
0.0 6.3 6.3 3.1 0.0 0.0 0.0 0.0 0.0 0.0 3.1 0.0
02500 5000 7500 10000
Figure 34. Frequency distribution of total water allocation (cubic metres per year per hectare) within the “zones
of influence” of each habitat type across all Fraser River lake sockeye salmon Conservation Units.
Numbers above bars represent percentage of CUs in the respective bin.
70.0
0.0 10.0 10.0 0.0 5.0 0.0 0.0 5.0 0.0
0.1 .2 .3 .4 .5
100.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.1 .2 .3 .4 .5
96.7
3.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.1 .2 .3 .4 .5
30.0
70.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.1 .2 .3 .4 .5
Figure 35. Frequency distribution of the density of water allocation restrictions (number per hectare) within the
“zones of influence” of each habitat type across all Fraser River lake sockeye salmon Conservation
Units. Numbers above bars represent percentage of CUs in the respective bin.
Lake influence / mainstem spawning
Nursery lake rearing
Lake inlet / tributary spawning
Migration corridors
Total water allocation (m3 per year / ha)
Lake influence / mainstem spawning
Nursery lake rearing
Lake inlet / tributary spawning
Migration corridors
Density of water allocation restrictions (# / ha)
Percent of CUs Percent of CUs

86
Figure 36. Overlay of water licenses, water allocation restrictions, population density, and distribution of all
salmon species in the province (map redrawn from Nelitz et al. 2009).

87
75.0
5.0 0.0 5.0 5.0 0.0 0.0 0.0 5.0 5.0
0.2 .4 .6 .8 1
55.6
0.0 0.0 0.0 11.1 0.0 0.0 0.0 11.1
22.2
0.2 .4 .6 .8 1
0.0 0.0 0.0 0.0 0.0
30.0
6.7 13.3
26.7 23.3
0.2 .4 .6 .8 1
70.0
5.0 0.0 0.0 0.0 0.0 5.0 0.0 5.0 0.0
0.2 .4 .6 .8 1
55.6
11.1 0.0 0.0 11.1 0.0 0.0 0.0 0.0 0.0
0.2 .4 .6 .8 1
63.3
6.7
30.0
0.0 0.0
0.2 .4 .6 .8 1
95.0
0.0 0.0 0.0 5.0 0.0 0.0 0.0 0.0 0.0
0.2 .4 .6 .8 1
94.4
5.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.2 .4 .6 .8 1
100.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.2 .4 .6 .8 1
Figure 37. Frequency distribution of the allocation of water by main uses within the “zones of influence” of spawning and migratory habitats across all
Fraser River lake sockeye salmon Conservation Units. Numbers above bars represent percentage of CUs in the respective bin.
Lake influence / mainstem spawning
Urban water allocation (as proportion of total allocation)
Lake inlet / tributary spawning
Agricultural water allocation (as proportion of total allocation)
Industrial water allocation (as proportion of total allocation)
Migration corridors
Percent of CUs

88
Figure 38. Representation of the relative level of vulnerability of and stress on freshwater habitats across all lake sockeye salmon Conservation Units.
Horizontal and vertical lines are used to highlight separations among Conservation Units into nine quadrants, where the top right quadrant
represent those CUs with high stress and high vulnerability and the bottom left represent those CUs with low stress and low vulnerability. Data to
represent vulnerability and stress are summarized in Table 5 and Table 11 through Table 14.
Lake influence / mainstem spawning Lake inlet / tributary spawning
Mi
g
ration corridorsNurser
y
lake rearin
g
Cumulative stress
Vulnerability

89
Figure 39. First page of a “dashboard” summarizing population status and habitats for the Quesnel Conservation
Unit (L_6_10, Summer timing group). See Appendix 3 for additional dashboards and a description to
help in the interpretation of the graphs and information contained therein.

90
Figure 40. Second page of a “dashboard” summarizing human stressors on the Quesnel Conservation Unit
(L_6_10, Summer timing group). See Appendix 3 for additional dashboards and a description to help in
the interpretation of the graphs and information contained therein.

91
Figure 41. Overview of the mechanisms by which stressors in the freshwater environment can have impacts on
habitats, growth and survival across life stages, and ultimately a population level effect on sockeye
salmon in the Fraser River basin.

92
8.0 Tables
Table 1. Status of 36 sockeye salmon Conservation Units (as reported by Pestal and Cass 2009), alignment of these CUs with stocks for which there are
productivity data from the Pacific Salmon Commission (as analyzed by Peterman et al. 2010), and summary of evidence / rationale for modifying
status where appropriate (as part of this report’s evaluation). Status is defined by severity (sev) and uncertainty (unc). Severity: 1 = low risk and
5 = high risk. Uncertainty: 1 = low uncertainty/high confidence, 5 = high uncertainty/low confidence, and 10 = insufficient information.
CU
Index Management
group CU
Type Conservation Unit
Freshwater
adaptive zone Status
category Status
scores
Stock name(s) for
productivity data Data avail
Status adjustment
based on
Grant et al. 2010
Sev
Unc
Total
Juv
L-6-12 Early Stuart Lake Stuart Middle_Fraser
IV
5
4
E.Stuart X
X
Not assessed by Grant et al.
L-6-14 Early Stuart Lake Takla/Trembleur
Middle_Fraser
III
4
2
Same
L-3-1 Early Summer Lake Chilliwack Lower_Fraser
II
1
3
Severity 1/2
L-3-5 Early Summer Lake Pitt Lower_Fraser
I
1
1
Pitt X
Same
L-5-2 Early Summer Lake Nahatlatch Fraser_Canyon
IV
3
3
Severity 4 (at risk by COSEWIC stds.)
L-6-1 Early Summer Lake
A
nderson Middle_Fraser
IV
3
3
Gates X
X
Unchanged
L-6-16 Early Summer Lake Taseko Middle_Fraser
IV
3
4
Severity 4/3 (at risk by COSEWIC stds.)
L-6-2 Early Summer Lake Chilko Middle_Fraser
UNK
10
Chilko X
X
Same
L-6-4 Early Summer Lake Francois Middle_Fraser
IV
3
3
Nadina X
X
Same
L-6-6 Early Summer Lake Fraser Middle_Fraser
UNK
10
Not assessed by Grant et al.
L-6-9 Early Summer Lake Nadina Middle_Fraser
UNK
10
Nadina X
X
Not assessed by Grant et al.
L-7-1 Early Summer Lake Bowron Upper_Fraser
IV
3
3
Bowron X
Severity 4 (at risk by COSEWIC stds)
L-7-2 Early Summer Lake Indian/Kruger Upper_Fraser
UNK
10
Not assessed by Grant et al.
L-9-2 Early Summer Lake ShuswapComplex
South_Thompson
III
3
2
Scotch and Seymour
X
Same
L-10-1 Early Summer Lake Kamloops North_Thompson
II
1
3
Fennel and Raft
X
Same
L-6-10 Summer Lake Quesnel Middle_Fraser
I
2
2
Quesnel X
X
Severity 2/3
L-6-13 Summer Lake Stuart Middle_Fraser
III
3
2
L. Stuart X
Same
L-6-15 Summer Lake Takla/Trembleur
Middle_Fraser
IV
4
3
L. Stuart X
Severity 3/4
L-6-3 Summer Lake Chilko Middle_Fraser
I
2
1
Chilko X
X
Same
L-6-5 Summer Lake Francois Middle_Fraser
UNK
10
Stellako X
X
Not assessed by Grant et al.
L-6-7 Summer Lake Fraser Middle_Fraser
I
1
1
Stellako X
X
Severity 1/2
L-6-8 Summer Lake Mckinley Middle_Fraser
UNK
10
Quesnel X
X
Lumped with Quesnel-S
L-3-2 Late Lake Cultus Lower
_
Fraser
III
5
1
Same
L-3-3 Late Lake Harrison(D/S) Lower_Fraser
II
1
3
Harrison X
Same
L-3-4 Late Lake Harrison(U/S) Lower_Fraser
IV
3
3
Weaver X
X
Same
L-4-1 Late Lake Lillooet Lillooet
III
3
2
Birkenhead X
Severity 2/3
L-5-1 Late Lake Kawkawa Fraser_Canyon
UNK
10
Not assessed by Grant et al.
L-6-11 Late Lake Seton Middle_Fraser
IV
3
3
Portage X
Same
L-9-1 Late Lake Kamloops South_Thompson
IV
4
4
Same (at risk by COSEWIC standards)
L-9-3 Late Lake ShuswapComplex
South_Thompson
III
3
1
L.
Shuswap X
X
Severity 2/3
R02 River River Widgeon Widgeon
IV
4
3
Same (at risk by COSEWIC standards)
R03 River River Lower_Fraser LFR
II
2
3
Same
R04 River River Fraser_Canyon
FRCany
UNK
10
Not assessed by Grant et al.
R05 River River Middle
_
Fraser MFR
UNK
10
Not assessed by Grant et al.
R06 River River Upper_Fraser UFR
UNK
10
Not assessed by Grant et al.
R07 River River Thompson_River
THOM
UNK
10
Not assessed by Grant et al.

93
Table 2. Comparison of alternative methods for evaluating status of sockeye salmon Conservation Units according to their assessment criteria / indicators,
feasibility of implementation, approach for setting benchmarks, and data needs / availability.
Alternative methodologies for assessing conservation status
Indicators of Status and Benchmarks for CUs
(Holt 2009; Holt et al. 2009) Qualitative Risk Evaluation
(Pestal and Cass 2009) NatureServe
(Faber-Langendoen et al. 2009)
Definition of
Status - Biological status is defined using four
classes of indicators: abundance (i.e.,
production), trends in abundance,
distribution, and fishing mortality. Status is
largely driven by productivity.
- Overall status is defined using several
classes of indicators: abundance
(production), trends in abundance,
productivity, diversity, fishing mortality,
distribution, and habitat condition. Indicators
fall into 1 of 4 risk factors: status,
vulnerability, direct human impact, and
environmental condition. Status is largely
driven by abundance whereas vulnerability is
driven by productivity.
- Status is largely driven by the potential
extinction or extirpation risk of elements of
biodiversity, including regional extinction or
extirpation.
Evaluation criteria, indicators, and metrics
Abundance - spawner abundance in current year
- geometric mean spawner abundance (most
recent generation)
- ratio of geometric mean of current
generation to historical geometric mean
- ratio of geometric mean of current
generation to highest generational geometric
mean on record
- geometric mean of escapement over last 4
yrs
- recent abundance relative to current
capacity (% of observations in last 12 yrs
with abundance outside of ± 75% of
capacity)
- recent abundance relative to potential
capacity
- recent abundance relative to capacity
indicated by traditional ecological knowledge
- variability in abundance (CV in avg. escape)
- % of recent 4 year abundance in most
abundant cycle line
- population size
- number of occurrences
Trends - reduction in spawner abundance over 3
gens or 10 yrs
- probability that declines are ≥ 25% over 3
gens or 10 yrs
- change in escapement over last 3
generations (slope in 4 yr running geometric
mean of escapement over last 12 years)
- recent avg. escapement (last 4 yrs) / overall
average escapement (geometric mean)
- recent avg. escapement (last 4 yrs) / highest
10 yr running geometric mean
- largest observed decline by cycle line
(geometric mean of 2 most recent cycle
escapements divided by geometric mean of
escapement 3 and 4 cycles ago)
- long-term
- short-term
Distribution - change in areal extent of spawn / juv over
time
- distribn of abundance across populations in
CU (decline in abund ance criterion if most
- range extent
-area of occupancy

94
Alternative methodologies for assessing conservation status
Indicators of Status and Benchmarks for CUs
(Holt 2009; Holt et al. 2009) Qualitative Risk Evaluation
(Pestal and Cass 2009) NatureServe
(Faber-Langendoen et al. 2009)
- spatial extent (area of occupancy)
- # of spawning groups with abund > 1000
fish, & change in that value over last 3 gens
or 10 yrs
- min. # of spawning groups that comprise
80% of total abund when ranked from most
to least abund & change over last 3 gens or
10 yrs
- area under curve between rank of spawning
group (as % of total number of groups)
versus % contribution of that group to the
total abundance
- proportion of spawning groups with a
geometric mean abundance over most
recent generation with > 1000 fish
- proportion of spawning groups that have
rates of change in abun ≥ 30% over 3 gens
or 10 yrs
abundant population were lost (avg. over
last 4 yrs))
- % area with good viability/ecological integrity
Diversity - type of habitat used by spawners or
juveniles and changes over time
- life history (e.g., # of populations in CU)
-genetic (TBD)
Productivity - average recruits / spawner over 3
generations
Fishing
mortality - fishing mortality in current year
- mean fishing mortality over most recent gen
- overlap with CU that is of high harvest
potential
-average mortality rate over last generation
- catch
Habitat
condition - sensitivity of critical habitat - habitat stressors
Feasibility - Highly technical and complex method
- High effort
- Data limitations make it difficult to assess
status for half of CUs
- Holt et al. 2009 – incomplete
- Grant et al. 2010: 15 CUs not
assessed
- Indicator roll up has not been developed.
- Simple and straight forward method
- Medium level of effort
- Data limitations not as restrictive, but exist
- 11 CUs not assessed
- Simple guidelines for rolling up indicators
- Simple and straight forward method
- Medium level of effort
- Method may result in inconsistent evaluation
across CUs

95
Alternative methodologies for assessing conservation status
Indicators of Status and Benchmarks for CUs
(Holt 2009; Holt et al. 2009) Qualitative Risk Evaluation
(Pestal and Cass 2009) NatureServe
(Faber-Langendoen et al. 2009)
Benchmarks - Developed using quantitative approach for
abundance and fishing mortality metrics
(monte carlo simulation for some metrics)
- Some metrics use qualitative method for
classifying pop. around benchmarks (based
on magnitude of uncertainty in observed
spawner data)
- incorporates uncertainty from data into the
benchmark
- robust and defensible
- undefined benchmarks for distribn metrics
- use fishing mortality remains uncertain
- Metrics use qualitative method for classifying
pop. around benchmarks (based on
magnitude of uncertainty in observed
spawner data)
- clear and consistent rules for setting
benchmarks across CUs
- defensible
Does not use break points for ratings as
thresholds, rather points along a continuum of
extinction risk. All metrics are on a scale of 0 to
5.5 (each has equal contribution to aggregate
score). Metrics may have different number of
increments within the 0 to 5.5 scale, but always
evenly spaced. Rank of A is the lowest possible
value (i.e., most at risk), with the exception of # of
occurrences where it is the opposite. Factors
within a status factor category are weighted
differently (see Table 8, pg 14), as are the status
factor categories themselves.
Data needs and
availability - Ricker spawner-recruitment relationship
(estimates of a and b parameters; estimate
of spawners)
- freshwater capacity
- time series of spawner abundance by
spawning group and counting locations
- spawner distribution across habitats
- # of spawning groups
- # of extant spawning locations, spawner
surveys
- Fishing mortality
- Data availability varies across CUs.
Resolution of escapement by populations
within a CU is likely variable. High quality
escapement data available for 21 CUs,
reasonable data available for 6 CUs, and
scarce data available for 9 CUs.
- time series of spawner abundance by
population and recruitment
- spawner abundance by population
- freshwater capacity
- fishing mortality
- habitat sensitivity
- Traditional Ecological Knowledge
- Data availability varies across CUs.
Resolution of escapement by populations
within a CU is likely variable. High quality
escapement data available for 21 CUs,
reasonable data available for 6 CUs, and
scarce data available for 9 CUs.
- Extent to which TEK is available across CUs
is not known. The quality of information on
detailed habitat within a CU varies across
CUs, with the majority of CUs having poor
fine scale information.
- time series of spawner abundance
- area / range extent relative to current
spawning and rearing
- Data availability varies across CUs.
Resolution of escapement by populations
within a CU is likely variable. High quality
escapement data available for 21 CUs,
reasonable data available for 6 CUs, and
scarce data available for 9 CUs.
- time series of spawner distribution
- spawning habitat/ground surveys
- habitat use surveys
Not available from stock assessment data.
- lake capacity for sockeye salmon production
(based on photosynthetic rate)
- Lake productivity estimates readily available.
- habitat condition, extent, and related
stressors
Not available from stock assessment data.
- catch and estimate of biomass (stock size)
Data availability varies across CUs, though
reasonable estimates exist. Direct fishing
mortality is available, though no data for FSM.
- lake capacity Traditional Ecological
Knowledge, genetic diversity, habitat
inventory, and run timing (not available).
- life history characteristics related to specific
habitat
Unknown availability, stock dependent.

96
Alternative methodologies for assessing conservation status
Indicators of Status and Benchmarks for CUs
(Holt 2009; Holt et al. 2009) Qualitative Risk Evaluation
(Pestal and Cass 2009) NatureServe
(Faber-Langendoen et al. 2009)
- catch and estimate of biomass (stock size)
- High quality data on direct human impact.
Level of uncertainty in data is low. Other
non-targeted sources of mortality (incidental,
non-harvest, harvest induced) not well
estimated. But total mortality from these
sources is low.
- catch and estimate of biomass (stock size)
- High quality data on direct human impact.
Level of uncertainty in data is low. Other
non-targeted sources of mortality (incidental,
non-harvest, harvest induced) not well
estimated. But total mortality from these
sources is low.
Strengths - technically robust
- clear quantitative rules for assessing status
and uncertainty
- explicitly incorporates uncertainty
- identifies information gaps
- explicitly incorporates uncertainty
- straight forward and easy to apply
- provides rapid appraisal of Fraser River
sockeye salmon
- identifies information gaps and major threats
- good start at defining clear, consistent
qualitative rules for assessing current status
and uncertainty that can be applied across
CUs
-flexibility to include metrics based on TEK
- straight forward and easily applied
- general public can understand the method
Weaknesses - complex method requiring significant
resources that lower feasibility of
implementation
- time consuming
- apart from fishing mortality, does not take
into consideration stressors on the
population
- using this method, 11 of 36 CUs have
insufficient information to determine status
- qualitative approach has led to issues with
consistency, repeatability, and transparency
of status assessment. Need extensive
training and review to minimize problem.
Subjective assessments are influenced by
personal judgments, perceptions of risk, and
systemic biases. Need to create a well
defined framework to minimize user bias and
increase transparency.
- not clear on weighting of factors; not obvious
documentation. Likely combination of expert
judgment and scientific literature.
- factors are not salmon centric, and may miss
key elements such as productivity
- standard benchmarks across CUs may not
be appropriate because doesn't take into
account unique characteristics and
circumstances of CU.
- poor performance of trend criterion
compared to other metrics (Porszt 2009)

97
Table 3. Summary of indicator classes included in each assessment method.
Indicator class Holt et al. 2009 Pestal and Cass 2009 Faber-Langendoen et al. 2009
(NatureServe)
Abundance x x x
Trend in spawner abundance x x x
Distribution x x x
Diversity indirect x
Productivity Indirect x x
Fishing mortality x x x
Habitat condition x x
Table 4. Description of indicators of habitat quantity and quality reflecting vulnerability across different sockeye salmon life stages.
Life stage Habitat quantity Habitat quality
Migration (adults and smolts) Migration distance (km): Delineated the migration route for each
sockeye salmon CU by identifying the stream reach immediately
downstream of a CU’s nursery lake outlet point and following the
mainstem stream network back from this point to the mouth of the
Fraser. A 1 km wide buffer was delineated on each side of the
migratory corridors.
Summer air temperatures across adult migration (ºC): Determined the
average summer/fall air temperatures along the migration corridor for
each sockeye salmon CU using ClimateWNA data adjusted for run
timing (see Appendix 4).
Spring air temperatures at nursery lake (ºC): Summarized historical
spring air temperatures at each sockeye salmon nursery lake,
extracted from ClimateWNA.
Spawning (adults, eggs, alevins) Tributary / inlet spawning or total spawning extent (m): Delineated the
total length of spawning zone(s) within each CU, and separated these
into lake inlet / tributary or lake outlet spawning extents (i.e., lake
influenced).
Ratio of lake influence spawning extent to total spawning extent:
Compared the linear extent of spawning reaches that are lake
influenced to the total extent of spawning for each sockeye salmon CU
(i.e., upstream disturbance effects would be buffered by the lake).
Rearing (fry and smolts) Area of nursery lakes (ha): Summed the area of all nursery lakes
within each sockeye salmon CU.
Nursery lake productivity (estimated): Summarized a biological
measure of juvenile productivity for each CU based on photosynthetic
rate (e.g., smolt abundance/density). Data were provided by Fisheries
and Oceans Canada’s Cultus Lake Salmon Research Laboratory.
Productivity was expressed as the average over time for a nursery
lake or by an averaging across lakes if multiple nursery lakes were
present within a CU.

98
Table 5. Indicators of habitat vulnerability for spawning, rearing, and migratory habitats across all lake sockeye salmon Conservation Units.
CU
Index Manage-
ment
group
Conservation
Unit Spawning Rearing Migration
Total
spawn
extent
(m)
Tributary /
inlet
spawn
extent (m)
Ratio of lake
influence:
total spawn
Spawning note Nurs
lake
area
(ha)
Nursery lake
productivity
(estimated
smolts / ha)
Migratn
distance
(km)
Summer air
temp across
adult
migratn (C)
Spring air
temp at
nursery
lake (C)
L_06_12 Early Stuart Stuart 13,259 13,259 0.00 35,919 1,578 998 17.44 2.7
L_06_14 Early Stuart Takla / Trembleur 229,647 218,533 0.05 36,253 449 1069 17.23 1.8
L_03_01 Early Sum Chilliwack 26,174 26,174 0.00 1,182 1,176 156 17.82 5.3
L_03_05 Early Sum Pitt 13,945 13,945 0.00 5,348 734 57 17.85 8.3
L_05_02 Early Sum Nahatlatch 3,870 3,188 0.18 303 255 18.59 6.2
L_06_01 Early Sum Anderson 7,387 7,387 0.00 Gates spawning
channel
2,872 3,387 359 18.66 6.8
L_06_16 Early Sum Taseko 2,395 -
1.00 2,124 709 17.34 0.6
L_06_02 Early Sum Chilko 12,490 -
1.00 18,447 1,157 680 17.49 1.4
L_06_04 Early Sum Francois 1,278 -
1.00 Nadina spawning
channel
25,164 2,912 1024 17.07 2.5
L_06_06 Early Sum Fraser 21,702 21,702 0.00 5,385 5,696 989 17.18 2.8
L_06_09 Early Sum Nadina -
-
Glacier Creek
spawn not mapped
930 1182 16.62 1.6
L_07_01 Early Sum Bowron 16,450 16,450 0.00 1,021 2,165 1102 16.96 3.3
L_07_02 Early Sum Indian/Kruger 10,448 -
1.00 235 1094 17.00 2.6
L_09_02 Early Sum Shuswap Complex 115,549 111,224 0.04 55,491 957 487 18.92 7.3
L_10_01 Early Sum Kamloops 60,026 28,105 0.53 6,014 4,358 387 18.92 7.5
L_06_10 Summer Quesnel 82,401 36,931 0.55 32,863 2,137 754 17.58 4.8
L_06_13 Summer Stuart 77,055 13,259 0.83 35,919 1,578 998 16.89 2.7
L_06_15 Summer Takla / Trembleur 9,911 3,347 0.66 36,253 449 1069 16.72 1.8
L_06_03 Summer Chilko 12,490 -
1.00 18,447 1,157 680 17.27 1.4
L_06_05 Summer Francois 11,460 -
1.00 25,164 2,912 1024 16.80 2.5
L_06_07 Summer Fraser 11,460 -
1.00 5,385 5,696 989 16.91 2.8
L_06_08 Summer Mckinley 4,743 4,743 0.00 513 849 17.23 3.5
L_03_02 Late Cultus -
-
Foreshore
spawning
631 6,841 111 13.33 8.1
L_03_03 Late Harrison (D/S) 10,534 10,156 0.04 22,192 1,245 127 13.21 7.6
L_03_04 Late Harrison (U/S) 1,986 1,164 0.41 Weaver spawning
channel
22,192 1,245 127 13.21 7.6
L_04_01 Late Lillooet 31,642 31,642 0.00 3,220 2,762 252 12.67 7.5
L_05_01 Late Kawkawa 837 -
1.00 76 164 13.19 7.9
L_06_11 Late Seton 6,766 -
1.00 2,475 2,591 333 12.57 7.0
L_09_01 Late Kamloops 11,446 -
1.00 5,517 4,358 387 12.50 7.5
L_09_03 Late Shuswap Complex 134,871 85,881 0.36 53,265 957 487 12.17 7.3

99
Table 6. Months used to represent historical average air temperature exposure of adult sockeye salmon CUs along the migration corridor based on
associated run timing group.
Sockeye salmon
run-timing group Average date that 50% of a run-
timing group passes Months used in our analysis to represent
average air temperatures during run-timing
Early Stuart July 14 July
Early Summer August 7 July and August
Summer August 17 August
Late October 4 (before 1995)
September 2 (1995 onwards)
October
September
Table 7. Summary of hypothesized links between freshwater stressors and sockeye salmon habitats, and the indicators being generated to represent these
stressors.
Freshwater stressor Quantity / quality of
spawning habitats Productivity of
nursery lakes Conditions related to
smolt outmigration /
adult migration
Stressor indicators
Forestry
Forest harvesting
activities X X X ** cumulative proportion of habitat type as forest disturbance over
rolling 15 year window (%)
** density of roads (km / km2)
** density of road-stream crossings (# / km2)
Mountain Pine
Beetle disturbance X X X ** cumulative proportion of habitat type as MPB disturbance (%)
Log storage
X n/a – qualitative evaluation of site specific information / data
Mining
X ** number and type of mines across habitat types
Hydroelectricity
Large hydro
X n/a – qualitative evaluation of site specific information / data
Small hydro
X X ** count of IPPs across habitat types
Urbanization
upstream of Hope X X X ** proportion of habitat type as urban area (%)
** density of human population (# / km2)
** allocation of urban water use per unit area (m3 / ha)
Agriculture X X X ** proportion of habitat type as agricultural land (%)
** allocation of agricultural water use per unit area (m3 / ha)
Water use X X ** allocation of urban, agricultural, and industrial water uses per
unit area (m3 per year / ha)
** density of water allocation restrictions (# / km2)
** proportion of water licenses across uses (%)

100
Table 8. Count and density (number/100km2) of various types of mining activity in watersheds that support sockeye salmon spawning by Conservation
Unit. Mining sites where an intervening lake buffers the impact on downstream juvenile habitat are not included.
CU Index Conservation Unit Spawning
type Placer Claims
Gravel Pits
Industrial
Mineral
Quarries
Metal Mines
Major
Exploration
Projects
Inactive Mine
Sites Total
L_09_03 Shuswap Complex-Late Tributary
46 (0.69)
12 (0.18)
0 (0.18) 0 (0)
0 (0)
14 (0.21)
72 (1.26)
L_09_02 Shuswap Complex-Early Summer Tributary
9 (0.17)
6 (0.11)
1 (0.02) 0 (0)
0 (0)
3 (0.06)
19 (0.36)
L_06_06 Fraser-Early Summer Tributary
2 (0.1)
10 (0.5)
1 (0.5)
1 (0.05)
0 (0)
0 (0)
14 (1.15)
L_10_01 Kamloops-Early Summer Tributary
11 (0.52)
0 (0)
0 (0)
0 (0)
1 (0.05)
2 (0.1)
14 (0.67)
L_03_03 Harrison-downstream migrating-Late Tributary
10 (1.16)
1 (0.12)
0 (0.12) 0 (0)
0 (0)
0 (0)
11 (1.4)
L_06_14 Takla/Trembleur-Early Stuart Tributary
5 (0.1)
3 (0.06)
0 (0.06) 0 (0)
0 (0)
1 (0.02)
9 (0.24)
L_06_15 Takla/Trembleur-Summer Tributary
5 (0.57)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
5 (0.57)
L_04_01 Lillooet-Late Tributary
0 (0)
4 (0.57)
0 (0.57) 0 (0)
0 (0)
0 (0)
4 (1.14)
L
_
06_10 Quesnel-Summer Tributary
2 (0.06)
0 (0)
0 (0)
0 (0)
1 (0.03)
0 (0)
3 (0.09)
L_03_01 Chilliwack-Early Summer Tributary
0 (0)
1 (0.22)
0 (0.22) 0 (0)
0 (0)
1 (0.22)
2 (0.66)
L_06_12 Stuart-Early Stuart Tributary
1 (0.27)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
1 (0.27)
L_06_13 Stuart-Summer Tributary
1 (0.27)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
1 (0.27)
L_03_04 Harrison -upstream migrating-Late Tributary
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_03_05 Pitt-Early Summer Tributary
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_05_02 Nahatlatch-Early Summer Tributary
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_01
A
nderson-Early Summer Tributary
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_08 Mckinley-Summer Tributary
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_07_01 Bowron-Early Summer Tributary
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_11 Seton-Late Mainstem
4 (20.9)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
4 (20.9)
L_09_03 Shuswap Complex-Late Mainstem
0 (0)
3 (2.77)
0 (2.77) 0 (0)
0 (0)
0 (0)
3 (5.54)
L_10_01 Kamloops-Early Summer Mainstem
0 (0)
2 (2.7)
0 (2.7)
0 (0)
0 (0)
0 (0)
2 (5.4)
L_09_01 Kamloops-Late Mainstem
0 (0)
1 (3.45)
0 (3.45) 0 (0)
0 (0)
0 (0)
1 (6.9)
L_06_10 Quesnel-Summer Mainstem
1 (1.07)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
1 (1.07)
L_03_03 Harrison -downstream migrating-Late Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_03_04 Harrison -upstream migrating-Late Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_05_01 Kawkawa-Late Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_05_02 Nahatlatch-Early Summer Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_02 Chilko-Early Summer Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_03 Chilko-Summer Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_04 Francois-Early Summer Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_05 Francois-Late Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_07 Fraser-Summer Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_13 Stuart-Summer Mainstem
0 (0)
0 (0)
0 (0)
0
(0)
0 (0)
0 (0)
0 (0)
L_06_14 Takla/Trembleur-Early Stuart Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_15 Takla/Trembleur-Summer Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_06_16 Taseko-Early Summer Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_07_02 Indian/Kruger-Early Summer Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
L_09_02 Shuswap Complex-Early Summer Mainstem
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)

101
Table 9. Average number of days per year when the mean daily water temperature exceeds 20°C in the Nechako
River above the Stuart River, and in the Stuart River, July 20 to August 20, 1953 to 2000* (data from
NFCP 2005).
Time period Nechako River (> 20°C) Stuart River (> 20°C)
1953 - 1979 3.2 5.0
1983 - 2000 2.9 7.6
*1980, 1981 and 1982 excluded due to different reporting method
Table 10. Number of days when the mean daily water temperature exceeds 20°C, and maximum and minimum
mean daily water temperatures in the Nechako River above the Stuart River, July 20 to August 20,
2002 to 2009 (data from Triton Environmental Consultants Ltd. 2003 through 2010 as part of NFCP
water temperature and flow management program). Data from 2007 were not reported due to failure of
thermographs.
Year Number of days >20°C Maximum mean daily
water temperature Minimum mean daily
water temperature
2001 0 19.6 16.4
2002 0 19.9 14.4
2003 0 20.0 17.1
2004 13 21.2 18.7
2005 0 20.0 16.8
2006 5 21.7 17.8
2008 0 19.5 15.7
2009 11 21.4 17.3

102
Table 11. Relative ranking of Conservation Units based on the intensity and trend (where available) of human stressors potentially interacting with
spawning locations downstream of lakes or on mainstem rivers. Stressor summary represents a relative measure of the cumulative level of stress
on a Conservation Unit across all types of human stressors. Note the following notations are used to denote intensity (I) of disturbance (high ++,
moderate +, low -, or none 0) and trend (T) in disturbance (increasing +, decreasing -, or stable 0).
CU Index Mgmt group Conservation Unit Forest harvesting MPB disturbance Road
density
Urban
area
Agric.
area
Water
alloc’tn
Water
rest’cn
Small
hydro
Placer
Mines
Stressor
summary
T I T I I I I I I I I
L_06_11 Late Seton - - + - + + - ++ + 0 ++ 15
L_09_01 Late Kamloops 0 0 + + + - ++ ++ + 0 0 14
L_09_03 Late Shuswap Complex - + + - + 0 ++ ++ + 0 0 13
L_10_01 Early Sum Kamloops - + + + + + + + 0 0 0 12
L_06_10 Summer Quesnel - + + ++ + 0 + - - 0 - 12
L_05_01 Late Kawkawa 0 0 0 0 ++ ++ 0 - ++ 0 0 10
L_03_04 Late Harrison (U/S) + - 0 0 + 0 - - 0 ++ 0 9
L_06_14 Early Stuart Takla / Trembleur - ++ + ++ + 0 0 0 0 0 0 8
L_06_05 Summer Francois - - + ++ + 0 - 0 - 0 0 8
L_06_07 Summer Fraser - - + ++ + 0 - 0 - 0 0 8
L_03_03 Late Harrison (D/S) 0 0 0 0 - ++ 0 - ++ 0 0 8
L_05_02 Early Sum Nahatlatch + - + + + 0 0 0 0 0 0 7
L_06_04 Early Sum Francois - ++ + ++ - 0 0 0 0 0 0 7
L_07_02 Early Sum Indian/Kruger - ++ + ++ - 0 0 0 0 0 0 7
L_06_13 Summer Stuart - ++ + ++ - 0 0 0 0 0 0 7
L_06_15 Summer Takla / Trembleur - ++ + ++ - 0 0 0 0 0 0 7
L_06_16 Early Sum Taseko 0 0 + ++ - 0 0 0 0 0 0 5
L_06_02 Early Sum Chilko 0 0 + ++ - 0 0 0 0 0 0 5
L_06_03 Summer Chilko 0 0 + ++ - 0 0 0 0 0 0 5
L_09_02 Early Sum Shuswap Complex - + + - - 0 0 0 0 0 0 4
L_06_12 Early Stuart Stuart 0 0 0 0 + 0 0 0 0 0 0 2
L_03_01 Early Sum Chilliwack 0 0 0 0 + 0 0 0 0 0 0 2
L_03_05 Early Sum Pitt 0 0 0 0 + 0 0 0 0 0 0 2
L_06_01 Early Sum Anderson 0 0 0 0 + 0 0 0 0 0 0 2
L_06_06 Early Sum Fraser 0 0 0 0 + 0 0 0 0 0 0 2
L_06_09 Early Sum Nadina 0 0 0 0 + 0 0 0 0 0 0 2
L_07_01 Early Sum Bowron 0 0 0 0 + 0 0 0 0 0 0 2
L_06_08 Summer Mckinley 0 0 0 0 + 0 0 0 0 0 0 2
L_03_02 Late Cultus 0 0 0 0 + 0 0 0 0 0 0 2
L_04_01 Late Lillooet 0 0 0 0 + 0 0 0 0 0 0 2

103
Table 12. Relative ranking of Conservation Units based on the intensity and trend (where available) of human stressors potentially interacting with
tributary or lake inlet spawning locations. Stressor summary represents a relative measure of the cumulative level of stress on a Conservation
Unit across all types of human stressors. Note the following notations are used to denote intensity (I) of disturbance (high ++, moderate +, low -,
or none 0) and trend (T) in disturbance (increasing +, decreasing -, or stable 0).
CU Index Mgmt group Conservation Unit Forest harvesting MPB disturbance Road
density
Urban
area
Agric.
area
Water
alloc’tn
Water
rest’cn
Small
hydro
Placer
mines
Stressor
summary
T I T I I I I I I I I
L_09_03 Late Shuswap Complex - - + - + ++ + ++ + 0 ++ 17
L_09_02 Early Sum Shuswap Complex + - + - + ++ + - + 0 + 16
L_10_01 Early Sum Kamloops + ++ + + + - - - - 0 + 15
L_06_06 Early Sum Fraser - + + ++ 0 + + - + 0 - 13
L_09_01 Late Kamloops 0 ++ 0 + + - - - - 0 + 13
L_06_08 Summer Mckinley - ++ + ++ + 0 ++ - 0 0 0 12
L_06_10 Summer Quesnel + + + + + 0 0 - - 0 - 11
L_06_12 Early Stuart Stuart + + + ++ - 0 0 0 0 0 - 9
L_06_13 Summer Stuart + + + ++ - 0 0 0 0 0 - 9
L_03_04 Late Harrison (U/S) + + 0 0 ++ 0 0 + 0 - 0 9
L_06_01 Early Sum Anderson - - + - - 0 - - ++ 0 0 8
L_06_15 Summer Takla / Trembleur - - + ++ + 0 0 0 0 0 + 8
L_03_03 Late Harrison (D/S) - ++ + - - 0 0 0 0 - + 8
L_04_01 Late Lillooet - ++ + - - - - - 0 0 0 8
L_06_14 Early Stuart Takla / Trembleur - - + + - 0 0 - 0 0 + 7
L_03_01 Early Sum Chilliwack - - + - + 0 - - 0 0 0 6
L_03_05 Early Sum Pitt - ++ 0 0 - - 0 - 0 0 0 5
L_05_02 Early Sum Nahatlatch - ++ + - - 0 0 0 0 0 0 5
L_07_01 Early Sum Bowron 0 - + + 0 0 0 0 0 0 0 4
L_06_04 Early Sum Francois 0 0 0 0 + 0 0 0 0 0 0 2
L_06_16 Early Sum Taseko 0 0 0 0 0 0 0 0 0 0 0 0
L_06_02 Early Sum Chilko 0 0 0 0 0 0 0 0 0 0 0 0
L_06_09 Early Sum Nadina 0 0 0 0 0 0 0 0 0 0 0 0
L_07_02 Early Sum Indian/Kruger 0 0 0 0 0 0 0 0 0 0 0 0
L_06_03 Summer Chilko 0 0 0 0 0 0 0 0 0 0 0 0
L_06_05 Summer Francois 0 0 0 0 0 0 0 0 0 0 0 0
L_06_07 Summer Fraser 0 0 0 0 0 0 0 0 0 0 0 0
L_03_02 Late Cultus 0 0 0 0 0 0 0 0 0 0 0 0
L_05_01 Late Kawkawa 0 0 0 0 0 0 0 0 0 0 0 0
L_06_11 Late Seton 0 0 0 0 0 0 0 0 0 0 0 0

104
Table 13. Relative ranking of Conservation Units based on the intensity and trend (where available) of human stressors potentially interacting with nursery
lake rearing. Stressor summary represents a relative measure of the cumulative level of stress on a Conservation Unit across all types of human
stressors. Note the following notations are used to denote intensity (I) of disturbance (high ++, moderate +, low -, or none 0) and trend (T) in
disturbance (increasing +, decreasing -, or stable 0).
CU
Index
Mgmt group Conservation Unit Forest harvesting MPB disturbance Road
density
Urban
area
Agric.
area
Stressor
summary
T I T I I I I
L_06_04 Early Sum Francois - ++ + ++ + - ++ 12
L_10_01 Early Sum Kamloops + + + + + + + 12
L_06_05 Summer Francois - ++ + ++ + - ++ 12
L_09_01 Late Kamloops + + + + + + + 12
L_06_06 Early Sum Fraser - ++ + ++ + - + 11
L_06_07 Summer Fraser - ++ + ++ + - + 11
L_06_08 Summer Mckinley - ++ + ++ + 0 ++ 11
L_09_02 Early Sum Shuswap Complex - + + - + + + 9
L_06_10 Summer Quesnel + + + + + 0 - 9
L_09_03 Late Shuswap Complex - + + - + + + 9
L_06_09 Early Sum Nadina + + + ++ - 0 0 8
L_06_13 Summer Stuart - + + ++ - - - 8
L_03_02 Late Cultus - - 0 0 ++ + ++ 8
L_06_12 Early Stuart Stuart - + + ++ - - 7
L_07_01 Early Sum Bowron + - + + - 0 0 6
L_07_02 Early Sum Indian/Kruger - - + ++ + 0 0 6
L_03_03 Late Harrison (D/S) - - + - - + - 6
L_03_04 Late Harrison (U/S) - - + - - + - 6
L_04_01 Late Lillooet - - + - - + - 6
L_05_01 Late Kawkawa - - 0 0 ++ ++ 0 6
L_06_14 Early Stuart Takla / Trembleur - + + + - 0 0 5
L_06_02 Early Sum Chilko - - + + - 0 - 5
L_06_15 Summer Takla / Trembleur - + + + - 0 0 5
L_06_03 Summer Chilko - - + + - 0 - 5
L_06_11 Late Seton - - + - - - - 5
L_03_01 Early Sum Chilliwack - - + - - 0 - 4
L_06_01 Early Sum Anderson - - + - - 0 - 4
L_06_16 Early Sum Taseko - - + + - 0 0 4
L_05_02 Early Sum Nahatlatch - - + - - 0 0 3
L_03_05 Early Sum Pitt - - 0 0 - - 0 2

105
Table 14. Relative ranking of Conservation Units based on the intensity and trend (where available) of human stressors potentially interacting with
migration corridors. Stressor summary represents a relative measure of the cumulative level of stress on a Conservation Unit across all types of
human stressors. Note the following notations are used to denote intensity (I) of disturbance (high ++, moderate +, low -, or none 0) and trend (T)
in disturbance (increasing +, decreasing -, or stable 0).
CU Index Mgmt group Conservation Unit Forest
harvesting
MPB
disturbance
Road
density
Urban
area
Agric.
area
Water
alloc’tn
Water
rest’cn
Small
hydro
Large
hydro
Log
storag
Stressor
summary
T I T I I I I I I I I I
L_06_14 Early Stuart Takla / Trembleur + ++ + ++ - - ++ - + - ++ - 21
L_06_15 Summer Takla / Trembleur + ++ + ++ - - ++ - + - ++ - 21
L_06_12 Early Stuart Stuart + + + ++ - - ++ - + - ++ - 20
L_06_06 Early Sum Fraser - + + ++ + - ++ - ++ - ++ - 20
L_06_13 Summer Stuart + + + ++ - - ++ - + - ++ - 20
L_06_07 Summer Fraser - + + ++ + - ++ - ++ - ++ - 20
L_06_04 Early Sum Francois - + + ++ + - ++ - + - ++ - 19
L_06_05 Summer Francois - + + ++ + - ++ - + - ++ - 19
L_03_01 Early Sum Chilliwack + ++ + - ++ ++ + + - 0 0 - 18
L_06_09 Early Sum Nadina - + + ++ - - ++ - + - ++ - 18
L_05_02 Early Sum Nahatlatch + ++ + - + + - - - - 0 - 15
L_06_01 Early Sum Anderson - - + - + + - - + - ++ - 15
L_06_10 Summer Quesnel - - + + + - ++ - ++ - 0 - 15
L_05_01 Late Kawkawa + - + - + ++ + + - 0 0 - 15
L_06_11 Late Seton - - + - + + - - + - ++ - 15
L_07_01 Early Sum Bowron - + + + - - ++ - + - 0 - 14
L_07_02 Early Sum Indian/Kruger - + + + - - ++ - + - 0 - 14
L_09_02 Early Sum Shuswap Complex - - + - + + + - ++ - 0 - 14
L_06_08 Summer Mckinley - - + + - - ++ - ++ - 0 - 14
L_03_02 Late Cultus - ++ 0 0 ++ ++ + + - 0 0 - 14
L_09_03 Late Shuswap Complex - - + - + + + - ++ - 0 - 14
L_06_16 Early Sum Taseko - - + + - - ++ - + - 0 - 13
L_06_02 Early Sum Chilko - - + + - - ++ - + - 0 - 13
L_10_01 Early Sum Kamloops - - + - + + + - + - 0 - 13
L_06_03 Summer Chilko - - + + - - ++ - + - 0 - 13
L_09_01 Late Kamloops - - + - + + + - + - 0 - 13
L_03_05 Early Sum Pitt 0 0 0 0 ++ ++ - ++ 0 0 0 - 11
L_03_03 Late Harrison (D/S) - - 0 0 + ++ + + - 0 0 - 11
L_03_04 Late Harrison (U/S) - - 0 0 + ++ + + - 0 0 - 11
L_04_01 Late Lillooet - + + - - + - - - - 0 - 11

106
Table 15. Matrix of pairwise correlations among indicators of habitat vulnerability, habitat stressors at the
watershed scale (as quantified within our nursery lake “zone of influence”), and stock productivity
(trend in Ricker residuals from 1984-2004). All correlations of 0.4 or greater are bolded. (n =18). Note
use of the following definitions for variables in the correlation matrix: (1) Migration distance (km); (2)
Total spawning extent (m); (3) Ratio of lake influenced : total spawning extent; (4) Area of nursery
lakes (ha); (5) Nursery lake productivity; estimated – adj. smolt density (N/ha); (6) Total water license
allocations (m3/yr/ha); (7) Agricultural area (% - ALR 2010); (8) Urban area (%); (9) Forest harvest
(%); (10) Mountain pine beetle disturbed area (% cumulative 2008); (11) Road density (km / km2 – all
roads); (12) Stream Crossing density (# / km2); (13) Placer claims (total count); (14) Active mines (total
count); (15) IPPs (total count); and (16) Trend in Ricker residuals.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1
2 -0.03
3 0.15 -0.03
4 0.10 0.75 -0.02
5 0.02 -0.19 0.23 -0.60
6 -0.24 0.54 0.04 0.12 0.48
7 0.11 0.55 0.29 0.46 0.31 0.62
8 -0.62 0.48 -0.29 0.31 -0.01 0.63 0.35
9 0.38 0.43 0.25 0.3 0.32 0.46 0.89 0.07
10 0.95 -0.17 0.24 -0.01 0.17 -0.21 0.20 -0.61 0.50
11 -0.08 0.80 0.07 0.63 0.15 0.72 0.88 0.52 0.75 -0.07
12 -0.32 0.38 0.02 0.19 0.37 0.55 0.55 0.40 0.53 -0.24 0.72
13 -0.25 0.56 0.03 0.32 0.20 0.63 0.48 0.52 0.43 -0.25 0.62 0.39
14 -0.25 0.56 0.01 0.20 0.43 0.98 0.62 0.70 0.45 -0.21 0.73 0.54 0.64
15 -0.52 -0.33 -0.16 -0.09 -0.19 -0.22 -0.29 0.46 -0.41 -0.43 -0.29 -0.08 0.10 -0.14
16 -0.57 -0.13 -0.26 -0.07 -0.16 0.22 0.05 0.38 -0.06 -0.46 0.12 0.27 0.18 0.20 0.34

107
Table 16. Predictor variables, AICc values, model rankings, and adjusted R-square values for all candidate linear
regression models relating sockeye salmon productivity (trend from 1984 to 2004 in Ricker residuals as
the response variable) to indicators of habitat vulnerability and stress. The three most plausible models
(three highest ranked AICc models) are in bold. Note use of the following definitions to describe the
predictor variables in the alternative models: (M) Migration distance (km); (L) Lake influenced : total
spawning extent ratio; (N) Nursery lake(s) area (ha); (J) Juvenile productivity of nursery lakes;
estimated – adj. smolt density (N/ha); (W) Water license allocations (m3 / yr / ha); (U) Urban area (%);
(F) Forest harvested area (% - 15 yr cumulative 2010); (R) Road density (km / km2 – all roads); (A)
Agricultural area (% - ALR 2010); (P) Pine beetle disturbed area (% cumulative 2008); (AM) Active
Mines (total count); and (PC) Placer Claims (total count).
Predictor Variables AICc AICc
rank Adjusted
R2*
M -44.83 1 0.3239
L -38.44 13 0.0148
N -37.21 24 -0.0583
J -37.50 22 -0.0406
W -37.86 18 -0.0189
U -39.72 7 0.0870
F -37.13 25 -0.0635
R -37.30 23 -0.0530
A -37.10 26 -0.0654
P -41.54 4 0.1798
AM -37.74 20 -0.0260
PL -37.58 21 -0.0354
M, P -42.96 2 0.3413
M, U -41.36 6 0.2763
P, U -38.41 14 0.1393
P, W -38.45 11 0.1414
M, R -41.83 3 0.2960
P, R -38.44 12 0.1407
R, W -34.43 33 -0.0881
N, J -34.93 30 -0.0561
F, R -34.97 29 -0.0541
M, AM -41.51 5 0.2828
P, AM -38.40 15 0.1388
N, AM -34.51 32 -0.0826
AM, PL -34.32 35 -0.0947
M, P, U -38.89 10 0.2929
M, P, W -39.03 8 0.2988
P, W, U -34.39 34 0.0786
N, J, M -37.81 19 0.2466
L, R, A -31.07 39 -0.1200
R, F, M -38.17 16 0.2621
M, P, AM -38.94 9 0.2948
M, L, R, A -34.21 36 0.2458
M, P, U, W -34.09 37 0.2405
P, U, W, R -29.47 41 0.0035
R, N, F, J -37.87 17 0.3917
R, M, F, U -33.27 38 0.2028
M, J, N, P -36.12 27 0.3258
M, P, AM, W -34.57 31 0.2616
R, N, F, AM -26.00 44 -0.2223
M, J, N, P, W -35.07 28 0.4517
M, W, P, U, N -28.83 42 0.2090
W, P, U, N, J -26.86 43 0.1113
U, F, L, A, R -21.69 47 -0.2043

108
Predictor Variables AICc AICc
rank Adjusted
R2*
M, W, P, AM, N -30.21 40 0.2706
U, F, L, A, R, M -21.45 48 0.1387
J, F, L, A, R, M -22.46 46 0.1880
J, P, L, A, R, M -24.72 45 0.2892
J, P, L, A, R, AM -18.20 49 0.3480
U, F, L, A, R, M, J -13.60 51 0.1423
U, F, L, A, R, M, P -14.86 50 0.2034
W, F, L, A, R, M, P -12.99 53 0.1109
W, F, L, A, R, U, P -8.60 55 -0.1507
W, F, N, A, R, M, P -7.46 56 -0.2306
J, P, L, A, R, AM, M -13.44 52 0.1344
W, F, L, A, R, U, P, M -1.91 57 0.1038
W, F, L, A, R, U, P, N 4.31 59 -0.2921
W, F, N, A, R, U, P, M -0.08 58 0.0020
W, F, N, J, R, U, P, M, A -10.73 54 0.4666
W, F, N, J, R, U, P, M, A 7.40 61 0.3904
W, F, N, J, R, U, P, M, L 5.49 60 0.4552
W, F, N, J, R, U, P, A, L 13.55 62 0.1247
M, L, N, J, W, U, F, R, A, P 32.31 63 0.3782
M, L, N, J, W, U, F, R, A, AM 36.12 64 0.2221
P, L, N, J, W, U, F, R, A, AM 38.43 65 0.1088
M, L, N, J, W, U, F, R, A, P, AM 66.34 66 0.6163
M, L, N, J, W, U, F, R, A, P, AM, PL 154.24 67 0.5924
*Adjusted R2 measures the proportion of the variation in the dependent variable accounted for by the explanatory
variables. Unlike R2, adjusted R2 allows for the degrees of freedom associated with the sums of the squares and so is
generally considered to be a more accurate goodness-of-fit measure than simple R2.
Table 17. Summary of correlation coefficients between indicators of total productivity and two habitat condition
indicators related to adult migration (summer air temperatures along migration corridor) and smolt
outmigration (spring time air temperatures at nursery lakes). Note use of the following symbols to
denote the Bonferroni adjusted significance levels: * = P <0.005 and NS = Not significant.
CU Index Stock name Total productivity vs.
migration summer air
temperatures
N Sig Total productivity vs.
nursery lake spring
air temperatures
N Sig
L_04_01 Birkenhead -0.42 58 * -0.09 58 NS
L_07_01 Bowron -0.24 58 NS -0.11 58 NS
L_06_02 Chilko -0.15 58 NS 0.09 58 NS
L_06_12 E. Stuart -0.32 58 NS -0.33 58 NS
L_10_01 Fennel -0.30 42 NS 0.03 42 NS
L_06_01 Gates -0.22 38 NS -0.25 38 NS
L_03_03 Harrison 0.19 57 NS 0.20 57 NS
L_09_03 L. Shuswap -0.11 57 NS -0.35 57 NS
L_06_13 L. Stuart -0.21 57 NS -0.14 57 NS
L_06_09 Nadina -0.19 33 NS 0.04 33 NS
L_03_05 Pitt -0.09 58 NS -0.25 58 NS
L_06_11 Portage -0.37 52 NS -0.03 52 NS
L_06_10 Quesnel 0.07 58 NS -0.08 58 NS
L_10_01 Raft -0.13 58 NS -0.26 58 NS
L_09_02 Scotch -0.10 41 NS -0.27 41 NS
L_09_02 Seymour -0.07 58 NS -0.20 58 NS
L_06_05 Stellako -0.05 58 NS -0.03 58 NS
L_03_04 Weaver -0.15 40 NS -0.30 40 NS

109
Table 18. Summary of population status, habitat vulnerability, and relative level of cumulative stress for all sockeye salmon Conservation Units in the
Fraser River basin. This summary is based on more detailed data found in Table 1, Table 5, Table 11, Table 12, Table 13, and Table 14.
CU
Index Management
group Conservation Unit Population
status Indicators of habitat vulnerability Relative measures of cumulative stress on habitats
Migration
distance
(km)
Ratio of lake
infl: total
spawn
Area of
nursery
lakes (ha)
Stress on
migration Stress on
lake infl /
main spawn
Stress on
lake inlet /
trib spawn
Stress on
rearing
L_06_12 Early Stuart Stuart Poor 998
0.00 35,919 High None Moderate Moderate
L_06_14 Early Stuart Takla / Trembleur Poor 1,069
0.05 36,253 High Moderate Moderate Moderate
L_03_01 Early Summer Chilliwack Moderate 156
0.00 1,182 High None Moderate Moderate
L_03_05 Early Summer Pitt Good 57
0.00 5,348 Moderate None Low Low
L_05_02 Early Summer Nahatlatch Poor 255
0.18 303 High Moderate Low Low
L_06_01 Early Summer Anderson Poor 359
0.00 2,872 High None Moderate Moderate
L_06_02 Early Summer Chilko Unknown 680
1.00 18,447 Moderate Moderate None Moderate
L_06_04 Early Summer Francois Poor 1,024
1.00 25,164 High Moderate Low High
L_06_06 Early Summer Fraser Unknown 989
0.00 5,385 High None High High
L_06_09 Early Summer Nadina Unknown 1,182
930 High None None High
L_06_16 Early Summer Taseko Poor 709
1.00 2,124 Moderate Moderate None Moderate
L_07_01 Early Summer Bowron Poor 1,102
0.00 1,021 High None Low Moderate
L_07_02 Early Summer Indian/Kruger Unknown 1,094
1.00 235 High Moderate None Moderate
L_09_02 Early Summer Shuswap Complex Poor 487
0.04 55,491 High None High High
L_10_01 Early Summer Kamloops Moderate 387
0.53 6,014 Moderate High High High
L_06_03 Summer Chilko Good 680
1.00 18,447 Moderate Moderate None Moderate
L_06_05 Summer Francois Unknown 1,024
1.00 25,164 High Moderate None High
L_06_07 Summer Fraser Good 989
1.00 5,385 High Moderate None High
L_06_08 Summer Mckinley Unknown 849
0.00 513 High None High High
L_06_10 Summer Quesnel Good 754
0.55 32,863 High High Moderate High
L_06_13 Summer Stuart Poor 998
0.83 35,919 High Moderate Moderate High
L_06_15 Summer Takla / Trembleur Poor 1,069
0.66 36,253 High Moderate Moderate Moderate
L_03_02 Late Cultus Poor 111
631 High None None High
L_03_03 Late Harrison (D/S) Moderate 127
0.04 22,192 Moderate Moderate Moderate Moderate
L_03_04 Late Harrison (U/S) Poor 127
0.41 22,192 Moderate Moderate Moderate Moderate
L_04_01 Late Lillooet Poor 252
0.00 3,220 Moderate None Moderate Moderate
L_05_01 Late Kawkawa Unknown 164
1.00 76 High High None Moderate
L_06_11 Late Seton Poor 333
1.00 2,475 High High None Moderate
L_09_01 Late Kamloops Poor 387
1.00 5,517 Moderate High None High
L_09_03 Late Shuswap Complex Poor 487
0.36 53,265 High High High High
R02 River Widgeon Poor Unknown Unknown N/A Unknown Unknown Unknown Unknown
R03 River Lower_Fraser Moderate Unknown Unknown N/A Unknown Unknown Unknown Unknown
R04 River Fraser_Canyon Unknown Unknown Unknown N/A Unknown Unknown Unknown Unknown
R05 River Middle_Fraser Unknown Unknown Unknown N/A Unknown Unknown Unknown Unknown
R06 River Upper_Fraser Unknown Unknown Unknown N/A Unknown Unknown Unknown Unknown
R07 River Thompson_River Unknown Unknown Unknown N/A Unknown Unknown Unknown Unknown

110
Table 19. Seven questions (from Stewart-Oaten 1996) and the related responses to our overall assessment of the
cumulative effect of freshwater stressors in contributing to the recent declines of Fraser River sockeye
salmon.
Question Response
(1) How plausible is the hypothesized causal
mechanism? Based on known physical and
biological principles, is the proposed mechanism
realistic?
Freshwater habitat quality is clearly an important component of salmon
conservation. A very large scientific literature has demonstrated that
changes in factors such as sediment supply, channel structure,
temperature, stream hydrology, large woody debris supply, total gas
pressure, and migration barriers can have negative impacts on sockeye
salmon. These factors are affected by both human activities and natural
events.
(2) What is the strength of the estimated effect? The
stronger it is, the more likely we are to correctly
distinguish the mechanism causing an observed
response from background variation and
observation error, as well as from changes arising
from other simultaneously operating mechanisms.
Note that in such analyses, emphasis here is on
estimating the strength of some effect and
uncertainty in that estimate, rather than on formally
testing some null hypothesis about the mechanism.
The cumulative impact of freshwater stressors has the potential to be very
strong. The absence of salmon from many streams can be associated
with poor habitat quality due to natural or anthropogenic causes.
(3) Does the consistency of direction, magnitude, and
duration of observed effects across studies of
similar systems also lend credibility to a hypothesis
about a given mechanism causing those effects?
For instance, does empirical evidence show such a
mechanism working in the same way for other
species or stocks or situations?
Scientific support is available for the link between stressor, impact and
outcome (e.g. more roads  higher sediment load  lower egg survival)
for all of the stressors that were considered. In most cases, the chain of
evidence for cause and effect for a specific stressor is not available for
Fraser River sockeye salmon. However, evidence from other species and
stocks consistently supports the mechanisms that have been
hypothesized.
(4) Are life stages affected by the proposed
mechanism affected whereas others are not?
Species or life stages or stocks that should not be
affected by the mechanism do not show change,
whereas the stages that should be affected do
show a response.
Adult returns and lifetime survival of sockeye salmon have consistently
declined across many Fraser River CUs, but juvenile populations have
not shown the same consistent trend. This suggests that declines in adult
populations are not the direct result of cumulative changes in freshwater
habitat quality.
(5) Did the timing of observed changes coincide with
a change in the state variable of the proposed
causal mechanism? If there is a time lag in the
response, it should be on a realistic time scale
based on what is known about the processes
involved.
Upward temporal trends in some freshwater stressors (mountain pine
beetle, road density) coincide with recent declines in Fraser sockeye
salmon. We believe that this correlation is spurious because juvenile
population density remains high for most CUs.
(6) Is there a similarity or coherence of responses
across space, time, populations, species, and
indicators that strengthens the case for a particular
mechanism?
Both the amount and rate of change in freshwater stressors varies
substantially among watersheds that support Fraser River sockeye
salmon In contrast, the recent decline in adult sockeye salmon
populations is coherent across most of the CUs in the Fraser drainage,
which suggests that more general decline that is not driven by changes in
these freshwater stressors.
(7) Are there natural gradients or contrasting
conditions that result in outcomes that are
consistent with the proposed mechanism? These
are not human-manipulated experiments, but they
may create distinct enough contrasting situations to
learn about mechanisms causing observed
changes.
Variation in habitat quality is generally associated with variation in
salmonids survival across various species in many locations. However,
patterns of variation among Fraser River sockeye salmon CUs in both
stressors and freshwater survival do not support the hypothesis that the
recent decline in adult sockeye salmon populations is the result of
changes in freshwater habitat quality.

111
Table 20. Seven questions (from Stewart-Oaten 1996) and the related responses to our overall assessment of the effect of Forest harvesting, Mountain Pine
Beetle, and roads in contributing to the recent declines of Fraser River sockeye salmon.
Question Forest Harvesting Mountain Pine Beetle Roads
(1) How plausible is
the hypothesized
causal
mechanism?
An extensive scientific literature documents a
variety of plausible mechanisms that link forest
harvesting to degradation in stream fish habitat.
These include changes in sediment supply, large
woody debris supply, channel structure,
temperature and hydrology.
Forestry impacts on lake and migration habitats
are less clear. Suspended sediment can reduce
light penetration but the effects of changes in
wood supply, stream temperature and hydrology
are unlikely to affect sockeye salmon survival or
growth in these habitats.
Loss of forest cover has been linked to
hydrological changes (timing of flows, low flows,
peak flow events) that can affect sockeye salmon
spawning habitat. Higher peak flow events can
destabilize stream channels, which would lead to
lower egg survival. Later, lower base flows could
impair migration and spawning of adults.
In some watersheds, salvage logging has
resulted in short-term increases in the intensity of
forest harvesting and road building.
Roads can have direct impacts but they can also
serve as a convenient indicator of other human
activities. Direct effects include blocking fish
passage and increasing sedimentation. Road
density is highly correlated with stressors such as
forest harvesting and urbanization.
(2) What is the
strength of the
estimated effect?
The impacts of forest harvesting on streams is
likely to be strong because of the continuing,
extensive nature of the impact combined with the
multiplicity of mechanisms. Forest harvesting
takes place in most Fraser River sockeye
salmon watersheds. Site specific factors such as
unstable soils, steep terrain and high variance in
flow can exacerbate logging impacts.
Impacts on lake habitat are thought to be much
less significant, particularly for pelagic species,
such as sockeye salmon.
In rare cases, sediment bars at the mouths of
small spawning streams may obstruct upstream
migration (e.g. early Talka CU).
The effects of mountain pine beetle are expected
to be relatively small. The area affected is far
larger than that of the other stressors that were
considered. However, the effects of the loss of
forest cover by itself are considered to be much
smaller than the combined effects of roads and
mechanical forest removal associated with forest
harvesting. Habitat impacts are limited mainly to
a single mechanism, hydrological change.
The impact of high road density is strong, but it
can be difficult to disentangle the effects of roads
from that of related activities, particularly forest
harvesting.
Culverts may prevent upstream migration,
particularly in smaller streams. This effect is
expected to be weak in sockeye salmon, which do
not rear in streams and tend to use larger rivers.
A Provincial project is in the process of identifying
impassible road culverts.
(3) How consistent is
the direction,
magnitude, and
duration of
observed effects
across studies of
similar systems?
The impacts of forestry are consistent in
direction across a wide range of ecosystem
types. However, the strength of the impact can
vary widely with site characteristics. The
biological effects of stream habitat impacts are
consistent across a variety of stream salmonids
species.
There is good evidence that the removal of forest
cover results in higher peak flows. However,
much of this information comes from forestry
related research where the effects of forest
removal are difficult to separate from the effects
of roads and mechanical disturbance.
Road density has been used as an indicator of
human impacts on many salmonids species
because of the consistent relationship between
road density population status. Higher harvest
rates associated with better access can be a
factor in many species, but not sockeye salmon.

112
Question Forest Harvesting Mountain Pine Beetle Roads
(4) Are life stages
affected by the
proposed
mechanism
affected whereas
others are not?
Most of the proposed mechanisms affect the
egg-fry in the case of sockeye salmon. Fall fry
densities have not declined to the same extent
as adult number. This is contrary to the expected
effect of lower egg-fry survival.
Peak flows are expected to have the highest
impact at the egg stage. While egg survival is
rarely monitored directly, fall fry densities have
not declined to the same extent as adult number.
This is contrary to the expected effect of lower
egg-fry survival.
Most of the proposed mechanisms affect the egg-
fry in the case of sockeye salmon. Fall fry
densities have not declined to the same extent as
adult number. This is contrary to the expected
effect of lower egg-fry survival.
Adult passage issues would be expected to affect
escapement numbers more than total recruits.
This is inconsistent with observations (but see
Section concerning migration mortality in
mainstems).
(5) Did the timing of
observed changes
coincide with a
change in the state
variable of the
proposed causal
mechanism?
The overall level of activity has increased
through time but has been relatively stable over
recent decades. Timing of this activity varies
among watersheds but changes in population
parameters for individual CUs do not reflect this.
Changes are large but too recent to have
affected adult returns prior to about 2006. In
addition, the effect on peak flow takes the form of
a reduced return time interval for a particular
sized event (e.g. a 20 year flow may now occur
every 10 years). As a result, stochasticity in
weather events is expected to increase the time
lag between deforestation and the process of
channel disruption.
The overall level of activity has increased steadily
through time. Timing of this activity varies among
watersheds but changes in population parameters
for individual CUs do not reflect this.
(6) Is there a similarity
or coherence of
responses across
space, time,
populations,
species, and
indicators that
strengthens the
case for a
particular
mechanism?
There is good support for the proposed
mechanisms in terms of both the habitat impacts
and the resulting biological impacts on a variety
of salmonids.
Similar levels of insect deforestation are very
rarely observed.
Road density is a well accepted indicator of
human impacts, including impacts on a variety of
salmon species.
(7) Are there natural
gradients or
contrasting
conditions that
result in outcomes
that are consistent
with the proposed
mechanism?
Contrasting conditions do exist among CUs but a
multiple regression analysis does not support the
hypothesis that forest harvesting has had a
significant impact on Fraser sockeye salmon
population parameters.
Contrasting conditions do exist among CUs but a
multiple regression analysis does not support the
hypothesis that deforestation as a result of
mountain pine beetle has had a significant impact
on sockeye salmon population parameters. The
impact of this factor may become more apparent
in future since there is the potential for significant
time lags between the occurrence of the stressor
and the effect on sockeye salmon populations.
Contrasting conditions do exist among CUs but a
multiple regression analysis does not support the
hypothesis that road density has had a significant
impact on Fraser sockeye salmon population
parameters.

113
Table 21. Seven questions (from Stewart-Oaten 1996) and the related responses to our overall assessment of the effect of agriculture and urbanization,
water use, and mines in contributing to the recent declines of Fraser River sockeye salmon.
Question Agriculture and Urbanization Water Use Mines
(1) How plausible is
the hypothesized
causal
mechanism?
Several plausible mechanisms link this stressor
to degradation of sockeye salmon stream
habitat. These include: increases in
sedimentation, increases in peak flows,
decreases in low flows and changes to riparian
conditions. Scientific support is available for all
of these mechanisms, however, the amount of
land use change that is required to induce
significant impact is not well defined.
The mechanisms for impacts on lake habitat and
migration corridors are less plausible but may
include degradation of water quality or
impediments to migration.
The hypothesized causal mechanism is
plausible. Adequate water flow is essential to
successful sockeye salmon migration and
reproduction. Low flows can also result in
unfavorably high temperatures and low
intergravel oxygen levels.
The hypothesized causal mechanism is
plausible. There is strong scientific evidence
that increased sediment load in spawning
streams can increase egg-fry mortality by
smothering eggs and destabilizing channel
structure. Mining for gravel or placer minerals
from stream beds has been shown to disrupt
channel structure. Mines can increase sediment
yields in watersheds by exposing mineral soil to
erosion. Complete destruction of habitat by poor
placement of mine sites is also possible but rare
under current legislation. Contaminants, such as
acid mine drainage, are excluded from this
analysis.
(2) What is the
strength of the
estimated effect?
The strength of the impact is expected to be
generally low for all habitat types. Although
agriculture and urbanization has the potential to
strongly affect spawning streams, none of the
watersheds have high levels of these land uses.
The impacts are also expected to be very site
specific because of mitigation measures that
have been implemented on some land holdings.
Migration corridors are bordered by extensive
urban and agricultural land use, but these
appear to have little impact on migration
activities.
There is potential for strong impacts. In some
key watersheds, licensed water use exceeds the
natural flow of the stream. Poor data quality
makes it difficult to directly assess the strength
of Water Use impacts.
The impact is expected to be generally weak
because of the low level of activity. Impact is
proportional to the increased sediment load and,
in extreme cases, can be very strong. The
sources of sediment in decreasing order of
severity are: 1. Extraction of material from
stream beds and riparian areas, 2. Roads 3.
Non-riparian pits and washing facilities.
There are no good data on mines as a source of
sediment in the Fraser Basin. However, the
effect of mining on Fraser River sockeye salmon
is expected to be weak because: (a) mines are
not prevalent in watersheds used for sockeye
salmon spawning and (b) the introduction of
sediment into fish habitat is prohibited under the
Fisheries Act.
(3) How consistent is
the direction,
magnitude, and
duration of
observed effects
across studies of
similar systems?
Historical studies have demonstrated a
consistent pattern of habitat degradation and
loss due to these stressors. Habitat restoration
efforts can be very effective in restoring
individual salmon runs, which suggests that
habitat degradation was a major factor in the
original extirpation.
Water Use conflicts are a serious management
issue for stream salmonids across much of
western North America. As a result, there is a
large body of literature that documents and
attempts to the quantify the effects of low flows
on a variety of salmonids species.
Studies on a variety of salmonids species
strongly support the idea that higher sediment
loads negatively impacts egg survival.
Effects are most severe where sediment both
settles out and interferes with hyporheic
exchange (groundwater – stream interactions).
Effects may be lower for very fine sediment (e.g.

114
Question Agriculture and Urbanization Water Use Mines
glacial flour, some clays), which may not settle
out, and very coarse sediment, which may not
reduce the porosity of the streambed.
(4) Are life stages
affected by the
proposed
mechanism
affected whereas
others are not?
Most of the proposed mechanisms affect the
egg-fry in the case of sockeye salmon. Fall fry
densities have not declined to the same extent
as adult number. This is contrary to the expected
effect of lower egg-fry survival.
For sockeye salmon populations, the effects of
low flow should be observable in the course of
routine monitoring of spawner numbers. At this
point, low flows in streams do not appear to have
resulted in consistent increases in prespawning
mortality or barriers to migration.
Sediment effects are specific to egg-fry in the
case of sockeye salmon. In severe cases, light
penetration and primary production might be
reduced in nursery lakes. Fall fry densities have
not declined to the same extent as adult number.
This is contrary to the expected effect of
sediment on egg-fry survival.
(5) Did the timing of
observed changes
coincide with a
change in the state
variable of the
proposed causal
mechanism?
The overall level of activity appears to have
increased steadily through time. However, there
is no data that documents differences in the
timing or rate of increase among CUs.
There is very little time series data on water
usage.
There are no time series data in the assessment
of mines.
(6) Is there a similarity
or coherence of
responses across
space, time,
populations,
species, and
indicators that
strengthens the
case for a
particular
mechanism?
Land use practices are one of the key
contributors to aquatic habitat degradation
worldwide. Salmonids appear to be particularly
vulnerable because of their dependence on high
water quality.
The most severe water use conflicts typically
occur during summer low flow periods. With the
exception of migrating adults, sockeye salmon
are not typically in streams during this period.
Water use does not appear to be an issue for
most stocks that spawn during summer low flow
(e.g. early Stuart).
There are insufficient data to conduct an
analysis of the coherence of responses.
(7) Are there natural
gradients or
contrasting
conditions that
result in outcomes
that are consistent
with the proposed
mechanism?
Contrasting conditions do exist among CUs but a
multiple regression analysis does not support the
hypothesis that intensive land use has had a
significant impact on Fraser sockeye salmon
population parameters.
Contrasting conditions do exist among CUs but a
multiple regression analysis does not support the
hypothesis that higher levels of water use have
had a significant impact on Fraser sockeye
salmon population parameters. Water use varies
substantially among CUs but declines in sockeye
salmon abundance have occurred in both high
and low water use areas.
Studies on a variety of salmonids species
strongly support the idea that increases in
sediment loads have negative impacts on egg
survival. There are no good data on egg survival
among CUs for Fraser River sockeye salmon.

115
Table 22. Seven questions (from Stewart-Oaten 1996) and the related responses to our overall assessment of the effect of small hydro, large hydro, and log
storage in contributing to the recent declines of Fraser River sockeye salmon.
Question Small hydro Large-hydro Log storage
(1) How plausible is
the hypothesized
causal
mechanism?
Plausible mechanisms include changes to
temperature, total gas pressure, gravel supply,
fish passage and water flow in spawning
streams. Site specific impacts are important
considerations. For example, larger headponds
are more likely to result in temperature effects
and to interrupt gravel supply, especially if gravel
is removed rather than passed into the
downstream channel. There are no plausible
mechanisms for impacts on lake habitat.
Since there are no IPPs on sockeye salmon
migration routes, there are no plausible
mechanisms for impacts on migration (but see
section 3.3.2, Large scale hydro projects).
Plausible mechanisms for Fraser projects
involve interference with upstream or
downstream migration. Water diversion for the
Nechacko project has resulted in higher than
optimal temperature for migrating adults from
several CUs. In the the Seaton River, sockeye
salmon have to pass a dam and a turbine
installation. The effects of dams and turbines in
delaying upstream migration, and killing or
injuring downstream migrating smolts, are well
documented in the scientific literature.
The hypothesized causal mechanism is
plausible. Log storage and associated activities
can damage habitat used by fish, and injure or
kill fish through increased biological oxygen
demands or exposure to toxic leachates. The
magnitude of these disturbances is considered
to be a function of the flushing characteristics of
the river, the specific methods of log handling /
storage, and intensity of use in each area.
Outmigrating smolts from all Fraser CUs could
potentially be exposed to effects of log handling
as they move through the Fraser estuary. Log
storage areas in the estuary may also be used
as staging areas by adults of all sockeye salmon
CUs before they migrate upstream.
(2) What is the
strength of the
estimated effect?
Currently, there are very few operational IPPs,
which means that the cumulative IPP impact is
also small. In most cases, the strength of each
effect at an individual IPP site will be small
because the site footprint is small and located
upstream of the sockeye salmon distribution.
However, the potential for a strong impact is
present, given the right site characteristics. An
IPP on a migration corridor poses a clear risk to
both upstream and downstream migrants. If the
stream channel reach between the intake
structure and the powerhouse is occupied by
fish, then water diversion is also an issue.
The effect is potentially strong. This stressor is
absent from most Fraser sockeye salmon CUs.
Large hydro projects do not exist in the Fraser
River mainstem and are present only in a limited
number of Fraser River tributaries. Only 2 of
these projects (Bridge-Seton and Nechako) are
considered to have the potential to cause
significant impacts on sockeye salmon. The
most serious effects are on CUs associated with
Nechako (Stuart, Takla/Trembleur, Nadina,
Fraser and Francois), where temperatures on
the migration route can reach lethal levels.
Sockeye salmon appear to be able to pass the
Seaton Project in both directions and the
strength of the estimated effect on the Seton and
Anderson CUs would appear to be weak.
The strength of the estimated effect of log
storage would appear to be weak, although
evidence in this regard is limited. While there
has been no direct study of effects on sockeye
salmon, past research in the Fraser estuary has
indicated that densities and growth rates of
resident salmon juveniles (Chinook, pink, and
chum) do not differ in log storage areas vs.
nearby marsh areas in the Fraser, nor do
densities of their invertebrate prey. Sockeye
salmon use of the Fraser estuary is limited in
both time and space, lessening their potential
exposure to log storage related contaminants.
(3) How consistent is
the direction,
magnitude, and
duration of
observed effects
across studies of
similar systems?
IPPs are a recent phenomenon and are
therefore unlikely to be linked to sockeye salmon
declines over past decades.
Studies on a variety of salmonids species
strongly support the idea that changes in
temperature, TGP, gravel supply, fish passage
and water flow in spawning streams can
Localized impacts of large hydro projects as well
as broader consequences for salmon population
are well documented. For example, salmon
populations that must run a gauntlet of large
dams in the Columbia River on their migrations
to and from spawning and nursery grounds can
experience significant direct and indirect
(delayed) mortality as a consequence of dam
Localized impacts of log storage activities on fish
and fish habitat are well documented across
western North America for a variety of fish
species. An assessment of the consistency of
broader impacts is impossible because
population-level effects of log handling on fish
have not been studied. Log storage, however, is
unlikely to interfere significantly and directly with

116
Question Small hydro Large-hydro Log storage
negatively affect survival. passage, with resultant population-level effects.
Large dams have been cited as a major
contributing factor to the near extirpation of
Snake River sockeye salmon.
fish outside the relatively small area where the
disturbances occur, and sockeye salmon should
generally be able to avoid such areas.
(4) Are life stages
affected by the
proposed
mechanism
affected whereas
others are not?
There are no data that is specific to Fraser River
sockeye salmon because of the recent history of
IPPs.
Specific impacts (e.g., egg mortality, migrant
mortality, gas bubble disease) could in theory be
observed in the future and linked to specific
sites. in practice, detecting such impacts would
require focused monitoring efforts.
Large dams existing in Fraser River tributaries
could have potentially serious impacts on
upstream migrating sockeye salmon adults and
downstream outmigrating smolts. Effects on
other sockeye salmon life stages should be
limited. As there are only two major hydro
projects of concern in the Fraser Basin, only a
subset of sockeye salmon CUs would be
exposed to large hydro impacts.
Log storage activities could effect outmigrating
sockeye salmon smolts or migrating adults, as
these would be the only life stages exposed to
potential impacts within the estuary. These
impacts would be observed as lower marine
survival. Given the weakness of the expected
response, declines in marine survival of sockeye
salmon are not likely to be the result of log
storage activity.
(5) Did the timing of
observed changes
coincide with a
change in the state
variable of the
proposed causal
mechanism?
There are no time series data in the assessment
of IPPs.
The Bridge-Seton and Nechako projects have
both been in operation since the 1950’s. Both
have had known historical impacts on migrating
sockeye salmon (direct mortality of smolts and
adults at Bridge-Seton, and thermal stress on
adults at Nechako). For both projects mitigation
measures have been enacted with apparent
success so survival should have improved in
recent years relative to historical conditions.
Based on our qualitative assessment, log
storage in the Fraser estuary appears to have
changed little in terms of overall extent,
distribution of storage sites, or seasonal intensity
over the last decade.
(6) Is there a similarity
or coherence of
responses across
space, time,
populations,
species, and
indicators that
strengthens the
case for a
particular
mechanism?
NA Effects of large hydro projects on salmon
survival can be serious and varied, with repeated
evidence of negative effects on salmon
populations.
While the effects of log handling on fish habitat is
relatively clear, the population level effects of
these impacts is not.
(7) Are there natural
gradients or
contrasting
conditions that
result in outcomes
that are consistent
with the proposed
mechanism?
NA Presence absence of a large hydro project
provides a good contrast among CUs that are
affected or unaffected by large hydro projects.
Status of the affected CUs is generally poor (or
unknown) but most other sockeye salmon CUs
are currently considered to have similar status.
Contrasts do not appear to exist because all
sockeye salmon CUs experience similar levels of
stress as they migrate through the lower River.

117
9.0 References
Anderson, H.W. 1974. Sediment deposition in reservoirs associated with rural roads, forest fires, and
catchment attributes. Pp. 13:87-95 In Proceedings symposium on man’s effect on erosion and
sedimentation, UNESCO, 9-12 September, Paris, IAHS
Arp, C.D., J.C. Schmidt, M.A. Baker, and A.K. Myers. 2006. Stream geomorphology in a mountain
lake district: hydraulic geometry, sediment sources and sinks, and downstream lake effects.
Earth Surf. Process. Landforms 32:525–543
Aukema, B. H., Carroll, A. L., Zhu, J., Raffa, K. F., Sickley, T. A. and Taylor, S. W. 2006.
Landscape level analysis of mountain pine beetle in British Columbia, Canada: spatiotemporal
developments and spatial synchrony within the present outbreak. Ecography. 29: 427–441.
B.C. Ministry of Environment, Lands and Parks and Environment Canada. 1994. State of the
environment report for British Columbia. 128 pp.
B.C. Ministry of Forests (MOF). 2001. Watershed assessment procedure guidebook. 2nd ed.,
Version 2.1, Forest Practices Branch, Ministry of Forests, Victoria, B.C. Forest Practices Code
Guidebook. Available from:
www.for.gov.bc.ca/tasb/legsregs/fpc/FPCGUIDE/wap/WAPGdbk-Web.pdf
Barksdale, R.D. (ed). 1991. The Aggregate Handbook. National Stone Association. Washington,
D.C.
Birtwell, I.K., C.D. Levings, J.S. MacDonald, and I.H. Rogers. 1988. A review of fish habitat issues
in the Fraser River system. Water Pollution Research Journal of Canada. 23:1-30. Available
from: http://digital.library.mcgill.ca/wqrj/pdfs/WQRJ_Vol_23_No_1_Art_01.pdf
Birtwell, I.K., M.D. Nassichuk, and H. Beune. 1987. Underyearling sockeye salmon (Oncorhynchus
nerka) in the Estuary of the Fraser River, p. 25-35. In H.D. Smith, L. Margolis, and C.C. Wood
(ed.) Sockeye Salmon (Oncorhynchus nerka) population biology and future management.
Canadian Special Publication Fisheries and Aquatic Sciences 96.
Birtwell, I.K., S.C. Samis, and N.Y. Khan. 2005. Commentary on the management of fish habitat in
northern Canada: information requirements and policy considerations regarding diamond, oil
sands and placer mining. Canadian Technical Report of Fisheries and Aquatic Sciences. 2606.
Bisson, P. A., J. B. Dunham, and G. H. Reeves. 2009. Freshwater ecosystems and resilience of
Pacific salmon: habitat management based on natural variability. Ecology and Society 14(1):
45. Available from: http://www.ecologyandsociety.org/vol14/iss1/art45/
Bocking, R.C., and M.N. Gaboury. 2003. Feasibility of reintroducing sockeye and other species of
Pacific salmon in the Coquitlam Reservoir, BC. Report prepared by LGL Limted for Bridge-
Coastal Fish and Wildlife Restoration Program. 105 pp. Available from:
http://www.bchydro.com/bcrp/reports/docs/coq_feasibility_mar2003_final.pdf
Bradford, M.J. 1995. Comparative analysis of Pacific salmon survival rates. Canadian Journal of
Fisheries and Aquatic Sciences 52: 1327-1338.
Bradford, M.J., and J.R. Irvine. 2000. Land use, fishing, climate change, and the decline of
Thompson River, British Columbia, coho salmon. Canadian Journal of Fisheries and Aquatic
Sciences 57: 13–16.
British Columbia Ministry of Environment (BC MOE). 2006. Surface Water Allocated for
Consumptive / Non consumptive Use by Sector. Available from:
http://www.env.gov.bc.ca/wsd/water_rights/surface_water/consumptive_nonconsumptive.pdf
British Columbia Ministry of Environment (BC MOE). 2008. Environmental Trends in British
Columbia: 2007. Available from:
http://www.env.gov.bc.ca/soe/et07/EnvironmentalTrendsBC_2007.pdf

118
Brownlee, M.J., B. Shepherd, and D. Bustard. 1988. Some effects of forest harvesting on water
quality in the Slim Creek watershed in the central interior of British Columbia. Canadian
Technical Report of Fisheries and Aquatic Sciences. 1613. Available from:
http://www.for.gov.bc.ca/hfd/library/ffip/Brownlee_1988.pdf
Budy, P., G.P. Thiede, N. Bouwes, C.E. Petrosky, H. Schaller. Evidence linking delayed mortality of
Snake River salmon to their earlier hydrosystem experience. North American Journal of
Fisheries Management. 22: 35–51.
Burgner, R.L. 1991. Life history of sockeye salmon (Oncorhynchus nerka). In C. Groot and L.
Margolis (eds). Pacific Salmon Life History. University of British Columbia Press. Vancouver,
BC. 564 pp.
Burnham, K.P. and D.R. Anderson. 1998. Model Selection and Inference: A Practical Information-
Theoretical Approach. New York: Springer-Verlag.
Carson, B., D. Maloney, S Chatwin, M. Carver, and P. Beaudry. 2008. Protocol for Evaluating the
Potential Impact of Forestry and Range Use on Water Quality (Water Quality Routine
Effectiveness Evaluation). Forest and Range Evaluation Program, B.C. Min. For. Range and
B.C. Min. Env., Victoria, BC.
Chamberlin, T.W., R.D. Harr, and F.H. Everest. 1991. Timber harvesting, silviculture, and watershed
processes. American Fisheries Society Special Publication 19: 83-138.
Chapman, D.W. 1988. Critical review of variables used to define effects of fines in redds of large
salmonids. Transactions of the American Fisheries Society 117(1): 1-21.
Cohen Commission of Inquiry into the Decline of Sockeye Salmon in the Fraser River (Cohen
Commission). 2010. Discussion Paper: Issues that the Commission intends to investigate.
Dated June 3, 2010. Available from:
http://www.cohencommission.ca/en/pdf/DiscussionPaper.pdf#zoom=100
Committee on the Status of Endangered Wildlife in Canada (COSEWIC). 2006. COSEWIC's
assessment process and criteria, Canada. Available from
http://www.cosewic.gc.ca/eng/sct0/assessment_process_e.cfm
Committee on the Status of Endangered Wildlife in Canada (COSEWIC). 2003. COSEWIC
assessment and status report on the sockeye salmon Oncorhynchus nerka (Cultus population) in
Canada. Available from:
http://www.sararegistry.gc.ca/virtual_sara/files/cosewic/sr_cultus_sockeye_salmon_e.pdf
Cooke, S.J., S.G. Hinch, A.P. Farrell, D.A. Patterson, K. Miller-Saunders, D.W. Welch, M.R.
Donaldson, K.C. Hanson, G.T. Crossin, M.T. Mathes, A.G. Lotto, K.A. Hruska, I.C. Olsson,
G.N. Wagner, R. Thomson, R. Hourston, K.K. English, S. Larsson, J.M. Shrimpton, G. Van der
Kraak. 2008. Developing a Mechanistic Understanding of Fish Migrations by Linking
Telemetry with Physiology, Behavior, Genomics and Experimental Biology: an
Interdisciplinary Case Study on Adult Fraser River Sockeye Salmon. Fisheries. 33 (7): 321-
338.
Crossin, G.T., S. G. Hinch, A. P. Farrell, D. A. Higgs, A. G. Lotto, J. D. Oakes, and M. C. Healey.
2004. Energetics and morphology of sockeye salmon: effects of upriver migratory distance and
elevation. Journal of Fish Biology 654: 788–810.
Crossin, G.T., S.G. Hinch, S.J. Cooke, D.W. Welch, A.G. Lotto, D.A. Patterson, S.R.M. Jones, R.A.
Leggatt, M.T. Mathes, J.M. Shrimpton, G. Van Der Kraak., and A.P. Farrell. 2008. Exposure to
high temperature influences the behaviour, physiology, and survival of sockeye salmon during
spawning migrations. Canadian Journal of Zoology 86: 127-140.
Daigle, P. 2010. A summary of the environmental impacts of roads, management responses, and
research gaps: A literature review. BC Journal of Ecosystems and Management. 10 (3): 65-89.
http://jem.forrex.org/index.php/jem/article/viewFile/38/9

119
Day, A. 2007. Harmonizing salmon habitat monitoring. Prepared for the Fraser Salmon Watershed
Program. Prepared by Redfish Bluefish Consulting. Available from:
http://www.tapsource.org/library~3/salmonhabitatmonitoring/day2007harmonizingsalmonhabit
atmonitoringfswppdf
Day, S.J. and W.K. Fletcher. 1991. Concentration of magnetite and gold at bar and reach scales in a
gravel-bed stream. Journal of Sedimentary Research. 61: 871-882.
Dorcey, A.H.J., and J.R. Griggs (eds). 1991. Water in Sustainable Development: Exploring our
Common Future in the Fraser River Basin. Research Program on Water in Sustainable
Development: Volume II. Westwater Research Centre, Faculty of Graduate Studies.
Douglas, T. 2006. Review of groundwater-salmon interactions in British Columbia. Prepared for
Watershed Watch Salmon Society and Walter and Duncan Gordon Foundation. Available
from: http://www.watershed-
watch.org/publications/files/Groundwater+Salmon++hi+res+print.pdf
Douglas, T. 2007. “Green” Hydro Power Understanding Impacts, Approvals, and Sustainability of
Run-of-River Independent Power Projects in British Columbia. Watershed Watch Salmon
Society. Available from: http://www.watershed-watch.org/publications/files/Run-of-River-
long.pdf
Dulvy, N.K., J.R. Ellis, N.B. Goodwin, A. Grant, J.D. Reynolds, and S. Jennings. 2004. Methods of
assessing extinction risk in marine fishes. Fish and Fisheries. 5: 255-276.
Edmonds, J.S., and J.V. Ward. 1979. Profundal benthos of a multibasin foothills reservoir in
Colorado, U.S.A. Hydrobiologia 63: 199-208.
Environmental Dynamics Inc (EDI). 2008. Mountain Pine Beetle Infestation: Hydrological Impacts.
Report prepared for Ministry of Environment Mountain Pine Beetle Action Team. Available
from: http://www.llbc.leg.bc.ca/public/pubdocs/bcdocs/445817/finishdownloaddocument.pdf
Essington, T.E., T.P. Quinn, and V.E. Ewert. 2000. Intra- and inter-specific competition and the
reproductive success of sympatric Pacific salmon. Canadian Journal of Fisheries and Aquatic
Sciencies. 57: 205–213.
Faber-Langendoen, D., L. Master, J. Nichols, K. Snow, A. Tomaino, R. Bittman, G. Hammerson, B.
Heidel, L. Ramsay, and B. Young. 2009. NatureServe Conservation Status Assessments:
Methodology for Assigning Ranks. NatureServe, Arlington, VA. Available from:
http://www.natureserve.org/publications/ConsStatusAssess_RankMethodology.pdf
Ferguson, J. W and M. Healey. 2009. Hydro power in the Fraser and Columbia Rivers: a contrast in
approaches to fisheries protection. Catch and Culture: Fisheries Research and Development in
the Mekong Delta 15(1): 4-11.
Finney, B., I. Gregory-Eaves, J. Sweetman, M. Douglas, and J.P. Smol. 2000. Impacts of climatic
change and fishing on Pacific salmon abundance over the past 300 years. Science 290 (5492):
795–799.
Fisheries and Oceans Canada (DFO). 2002. Sakinaw Sockeye Salmon. DFO Science Stock Status
Report D6-12 (2002). Available from: http://www.dfo-
mpo.gc.ca/csas/Csas/status/2002/SSR2002_D6-13e.pdf
Fisheries and Oceans Canada (DFO). 2005. Canada’s policy for conservation of wild Pacific salmon.
Vancouver, BC. Cat. No. Fs23–476/2005E. 57 pp. Available at: http://www.pac.dfo-
mpo.gc.ca/publications/pdfs/wsp-eng.pdf
Fisheries and Oceans Canada (DFO). 2009. Integrated fisheries management plan – Salmon –
Southern B.C. – June 1, 2009 TO May 31, 2010. Available from: http://www-ops2.pac.dfo-
mpo.gc.ca/xnet/content/MPLANS/plans09/09-10SCSalmonIFMP.pdf

120
Fisheries and Oceans Canada (DFO). 2010. Integrated fisheries management plan – Salmon –
Southern B.C. – June 1, 2010 TO May 31, 2011. Available from: http://www-ops2.pac.dfo-
mpo.gc.ca/xnet/content/MPLANS/plans10/salmon/SC-IFMP-2010.pdf
Foerster, R.E. 1937. The Relation of Temperature to the Seaward Migration of Young Sockeye
Salmon (Oncorhynchus nerka). Journal of the Biological Board of Canada. 3(5): 421-438.
Foreman, M.G.G., D.K. Lee, J. Morrison, S. Macdonald, D. Barnes, and I.V. Williams. 2001.
Simulations and retrospective analyses of Fraser watershed flows and temperatures.
Atmosphere-Ocean. 39 (2): 89-105.
Forest Engineering Research Institute of Canada (FERIC). 1980. Logging use of foreshore leases on
the coast of British Columbia. FERIC Tech. Note TN-38: 27 pp.
Forest Practices Board (FPB). 2007. The Effect of Mountain Pine Beetle Attack and Salvage
Harvesting On Streamflows: Special Investigation. Report Number FPB/SIR/16.
Forest Practices Board (FPB). 2009. Fish passage at stream crossings. Special Investigation.
FPB/SIR/25. Available from:
http://www.fpb.gov.bc.ca/WorkArea/linkit.aspx?LinkIdentifier=id&ItemID=4236
Fraser Basin Council (FBC). 2009. 2009 State of the Fraser Basin Report Sustainability Snapshot 4:
The Many Faces of Sustainability. Available from:
http://www.fraserbasin.bc.ca/publications/documents/2009_SOFB_Report_SS4.pdf
Fraser River Estuary Management Program (FREMP). 1991. Log management in the Fraser River
estuary.
Fraser River Estuary Management Program (FREMP). 1994. An Estuary Management Plan for the
Fraser River: A Living, Working River. Available from:
http://www.bieapfremp.org/fremp/pdf_files/f_emp_complete.pdf
Fraser River Estuary Management Program (FREMP). 2003. The Estuary Management Plan for the
Fraser River: A Living, Working River, Updated 2003. Available from:
http://www.bieapfremp.org/fremp/pdf_files/Revised%20EMP%202003%20August%20.pdf
Fretwell, M. 1979. Migration of sockeye smolts at the Seton Creek hydro installation in 1978.
International Pacific Salmon Fisheries Commission. Prepared for BC Hydro.
Fretwell, M. 1980. Juvenile sockeye salmon studies at Seton Creek hydroelectric installation in 1979.
International Pacific Salmon Fisheries Commission. Prepared for BC Hydro.
Fretwell, M. 1982. Migration of juvenile sockeye salmon at Seton Creek hydroelectric installation in
1981. International Pacific Salmon Fisheries Commission. Prepared for BC Hydro.
Fretwell, M., and S. Hamilton. 1983. Migration of juvenile sockeye salmon at Seton Creek
hydroelectric installation in 1982. International Pacific Salmon Fisheries Commission.
Prepared for BC Hydro.
Fretwell, M.R. 1989. Homing behaviour of adult sockeye salmon in response to a hydroelectric
diversion of homestream waters at Seton Creek. International Pacific Salmon Fisheries
Commission. Commission Bulletin XXV.
Gende, S.M., R.T. Edwards, M.F. Willson, and M.S. Wipfli. 2002. Pacific Salmon in Aquatic and
Terrestrial Ecosystems. BioScience. 52 (10): 917–928
Government of British Columbia. 2010. British Columbia’s Water Act Modernization. Technical
Background Report. Available from: http://livingwatersmart.ca/water-act/docs/wam_tbr.pdf
Grant, J.W.A., J. Englert, and B.F. Bietz. 1986. Application of a method for assessing the impact of
watershed practices: effects of logging on salmonid standing crops. North American Journal of
Fisheries Management. 6: 24–31.
Grant, S., B. MacDonald, T. Cone, C. Holt, A. Cass, E. Porszt, J. Hume, and L. Pon. 2010. Fraser
sockeye (Oncorhynchus nerka) Wild Salmon Policy evaluation of stock status: state and rate.
Draft working paper 2010/P14.

121
Gregory, R.S., and C.D. Levings. 1998. Turbidity reduces predation on migrating juvenile pacific
salmon. Trans. Am. Fish. Soc. 127: 275–285.
Gresh, T., J. Lichatowich, and P. Schoonmaker. 2000. Estimation of historic and current levels of
salmon production in the northeast Pacific ecosystem: Evidence of a nutrient deficit in the
freshwater systems of the Pacific Northwest. Fisheries. 25 (1): 15–21
Groves, K.L., and P.S. Higgins. 1995. Seton Project: Overview study of downstream fish passage.
BC Hydro Report No. H2768. WPR-259.
Hannigan, P., P.J. Lee, and K. Osadetz. 1998. Conventional Natural Gas and Petroleum Potential in
British Columbia. Available from:
http://www.empr.gov.bc.ca/OG/oilandgas/petroleumgeology/ConventionalOilAndGas/Docume
nts/cog200310.pdf
Harr, R.D., and R.L. Fredriksen. 1988. Water quality after logging small watersheds within the Bull
Run watershed. Journal of the American Water Resources Association 24(5):1103–1111.
Hartman, G.F., C. Groot, and T.G. Northcote. 2000. Science and management in sustainable
salmonid fisheries: The ball is not in our court. Pages 31–50 in E.E. Knudsen, C.R. Steward,
D.D. MacDonald, J.E. Williams, and D.W. Reiser (eds.). Sustainable fisheries management:
Pacific salmon. CRC Press, New York.
Hartman, G.F., J.C. Scrivener, and M.J. Miles. 1996. Impacts of logging in Carnation Creek, a high-
energy coastal stream in British Columbia, and their implication for restoring fish habitat.
Canadian Journal of Fisheries and Aquatic Sciences 53(Suppl. 1): 237–251.
Healey, M.C. 2009. Resilient Salmon, Resilient Fisheries for British Columbia, Canada. Ecology and
Society 14 (1): 2. Available from: http://www.ecologyandsociety.org/vol14/iss1/art2/
Hicks, B.J., R.L. Beschta, and R.D. Harr. 1991. Long-term changes in streamflow following logging
in western Oregon and associated fisheries implications. Water Resources Bulletin 27: 217-
226.
Hilborn, R., T.P. Quinn, D.E. Schindler, and D.E. Rogers. 2003. Biocomplexity and fisheries
sustainability. Proceedings of the National Academy of Sciences. 100 (11): 6564-6568.
Hinch, S.G. and E.G. Martins. 2011. A review of potential climate change effects on survival of
Fraser River sockeye salmon and an analysis of interannual trends in en route loss and pre-
spawn mortality. Cohen Commission Tech. Rep. 9. In prep. Vancouver, B.C.
www.cohencommission.ca
Hogan, D.L., P.J. Tschaplinski, and S. Chatwin (eds). 1998. Carnation Creek and Queen Charlotte
Island Fish/Forestry Workshop: Applying 20 Years of Coastal Research to Management
Solutions. Prepared by BC Ministry of Forests - Research Branch. Victoria, BC. Land
Management Handbook 41. 275 pp.
Holt, C.A. 2009. Evaluation of Benchmarks for Conservation Units in Canada's Wild Salmon Policy:
Technical Documentation. Canadian Science Advisory Secretariat Research Document
2009/059. Available from: http://www.dfo-mpo.gc.ca/CSAS/Csas/Publications/ResDocs-
DocRech/2009/2009_059_e.pdf
Holt, C.A., A. Cass, B. Holtby, and B. Riddell. 2009. Indicators of Status and Benchmarks for
Conservation Units in Canada's Wild Salmon Policy. Canadian Science Advisory Secretariat
Research Document 2009/058. Available from: http://www.dfo-
mpo.gc.ca/CSAS/Csas/Publications/ResDocs-DocRech/2009/2009_058_e.pdf
Holtby, L.B. and K.A. Ciruna. 2007. Conservation Units for Pacific Salmon under the Wild Salmon
Policy. Canadian Science Advisory Secretariat. Research Document 2007/070. Available from:
http://www.dfo-mpo.gc.ca/CSAS/Csas/DocREC/2007/RES2007_070_e.pdf
Houghton, J. 2008. The Story of the Coquitlam River Watershed: Past, Present and Future. Prepared
for The City of Coquitlam and Kwikwetlem First Nation. Available from:

122
http://www.coquitlam.ca/NR/rdonlyres/14C2E85D-7108-4EF2-B757-
B05505D3D65C/85521/CITYDOCS692852v1BCoquitlam_River_Stakeholder_Engag.pdf
Hume, J.M. B., K.S. Shortreed, and K.F. Morton. 1996. Juvenile sockeye rearing capacity of three
lakes in the Fraser River system. Canadian Journal of Fisheries and Aquatic Sciences 53:719–
733.
IceWatch. No date. Changes in Lake Ice Signal a Changing Climate. Available from:
http://www.naturewatch.ca/english/icewatch/changes_in_lake_ice_signal_a_changing_climate.
html
International Pacific Salmon Fisheries Commission (IPSFC). 1979. Sockeye salmon studies on the
Nechako River relative to the potential Kemano II power development. Volume 2. International
Pacific Salmon Fisheries Commission, New Westminster, B.C.
Johannes, M.R.S., A. Kenney, J. Pouliotte, and D. Steel. 2007. Mountain Pine Beetle Threats to
Salmon and Fisheries Resources in BC: Proceedings of the Pacific Salmon Foundation and
Fraser Basin Council Workshop (January 30-31, 2007, Prince George). Prepared by Golder
Associates Ltd., Vancouver, BC. 71 p. Available from:
http://www.thinksalmon.com/reports/FSWP0618_MPB_Workshop.pdf
Johannes, M.R.S., R. Hoogendoorn, R.E. Scott, and L. Nikl. 2011. Fraser River sockeye salmon
habitat analysis: Lower Fraser and Strait of Georgia. Golder Associates Ltd. Cohen
Commission Tech. Rep. 12. In prep. Vancouver, B.C. www.cohencommission.ca
Jokinen, C.C., H. Schreier, W. Mauro, E. Taboada, J.L. Isaac-Renton, E. Topp, T. Edge, J.E.
Thomas, and V.P.J. Gannon. 2010. The occurrence and sources of Campylobacter spp.,
Salmonella enterica and Escherichia coli O157:H7 in the Salmon River, British Columbia,
Canada. Journal of Water and Health. 374-386.
Jones, N. 2010. Incorporating lakes within the river discontinuum: longitudinal changes in ecological
characteristics in stream–lake networks. Canadian Journal of Fisheries and Aquatic Sciences.
67: 1350–1362.
Kean, S. 2010. Fishing for Gold in The Last Frontier State. Science. 327 (5963): 263-265.
Kent, M. 2011. Infectious diseases and potential impacts on survival of Fraser River sockeye salmon.
Cohen Commission Tech. Rep. 1. In prep. Vancouver, B.C. www.cohencommission.ca
Klock, G. O. 1985. Modelling the cumulative effects of forest practices on downstream aquatic
ecosystems. Journal of Soil and Water Conservation 40: 237-41.
Kondolf, G.M. 1997. Hungry water: Effects of dams and gravel mining on river channels.
Environmental Management. 21: 533-551.
Levy, D. and J. Sneep. 2006. Effectiveness of Seton Powerhouse shutdowns for reducing
entrainment mortality of Sockeye salmon smolts during 2006. Report for B.C. Hydro 6911
Southpoint Drive, Burnaby, B.C.
Levy, D. and J. Sneep. 2009. Effectiveness of Seton Powerhouse shutdowns for reducing
entrainment mortality of Sockeye salmon smolts during 2009. Report for B.C. Hydro 6911
Southpoint Drive, Burnaby, B.C.
Levy, D.A., L.U. Young, and L.W. Dwernychuk (eds). 1996. Strait of Georgia Fisheries
Sustainability Review. Prepared by Hatfield Consultants Ltd.
Levy, D.A., T.G. Northcote, and R.M. Barr. 1982. Effects of estuarine log storage on juvenile
salmon. Westwater Research Centre Technical Report 26.
Levy, D.A., T.G. Northcote, K.J. Hall, and I. Yesaki. 1989. Juvenile salmonid response to log
storage in littoral habitats of the Fraser River Estuary and Babine Lake. Canadian Special
Publication of Fisheries and Aquatic Sciences 105: 82-91.
Lloyd, D.S., J.P. Koenings, , and J.D. LaPerriere. 1987. Effects of turbidity in fresh waters of Alaska.
North American Journal of Fisheries Management. 7: 18-33.

123
MacDonald, D.D., J. Sinclair, M. Crawford, H. Prencipe and M. Meneghetti. 2011. Potential effects
of contaminants on Fraser River sockeye salmon. MacDonald Environmental Sciences Ltd.
Cohen Commission Tech. Rep. 2. In prep. Vancouver, B.C. www.cohencommission.ca
Mace, P.M., A.W. Bruckner, N.K. Daves, J.D. Field, J.R. Hunter, N.E. Kohler, R.G. Kope, S.S.
Lieberman, M. Miller, J.W. Orr, R.S. Otto, T. Smith, N.B. Thompson, J. Lyke, and A.G.
Blundell. 2002. NMFS / Interagency Working Group Evaluation of CITES Criteria and
Guidelines. U.S. Department of Commerce, NOAA Techical Memo, NMFS-F/SPO-58.
MacIsaac, E.A. (ed). 2003. Forestry Impacts on Fish Habitat in the Northern Interior of British
Columbia: A Compendium of Research from the Stuart-Takla Fish-Forestry Interaction Study.
Canadian Technical Report of Fisheries and Aquatic Sciences 2509. Available from:
http://www.dfo-mpo.gc.ca/Library/329257.pdf
Marmorek, D., D. Pickard, A. Hall, K. Bryan, L. Martell, C. Alexander, K. Wieckowski, L. Greig
and C. Schwarz. 2011. Fraser River sockeye salmon: data synthesis and cumulative impacts.
ESSA Technologies Ltd. Cohen Commission Tech. Rep. 6. In prep. Vancouver, B.C.
www.cohencommission.ca
Marmulla, G. (ed.). 2001. Dams, fish and fisheries: Opportunities, challenges and conflict resolution.
FAO Fisheries Technical Paper No. 419. Rome, FAO.
Martins, E.G., S.G. Hinch, D.A. Patterson, M.J. Hague, S.J. Cooke, K. Miller, M.F. Lapointe, K.K.
English, and A.P. Farrell. 2011. Effects of river temperature and climate warming on stock-
specific survival of adult migrating Fraser River sockeye salmon (Oncorhynchus nerka). Global
Change Biology. 17: 99–114.
Mathes, M.T., S.G. Hinch, S.J. Cooke, G.T. Crossin, D.A. Patterson, A.G. Lotto, and A.P. Farrell.
2010. Effect of water temperature, timing, physiological condition, and lake thermal refugia on
migrating adult Weaver Creek sockeye salmon (Oncorhynchus nerka). Canadian Journal of
Fisheries and Aquatic Sciences. 67(1): 70-84.
McCaffery M., T.A. Switalski, L. Eby. 2007. Effects of road decommissioning on stream habitat
characteristics in the South Fork Flathead River, Montana. Transactions of the American
Fisheries Society 136(3): 553-561.
McGarrity, K., and G. Hoberg. 2005. The Beetle Challenge: An Overview of the Mountain Pine
Beetle Epidemic and its Implications. Available from:
http://www.policy.forestry.ubc.ca/PDF/The%20Beetle%20Challenge%20-(3).pdf
Meehan, W.R. (ed). 1991. Influence of Forest and Rangeland Management on Salmonid Fishes and
Their Habitats. American Fisheries Society. Bethesda, Md. 751 pp.
Miller, L.B., D.J. McQueen, and L. Chapman. 1997. Impacts of forest harvesting on lake
ecosystems: A preliminary review of the literature. Prepared by the Ministry of Environment,
Lands, and Parks. Wildlife Bulletin No. B-84. Available from:
http://a100.gov.bc.ca/pub/eirs/viewDocumentDetail.do?fromStatic=true&repository=BDP&doc
umentId=3263
Ministry of Energy Mines and Petroleum Resources (EMPR). 1998. Policy for Metal Leaching and
Acid Rock Drainage at Minesites in British Columbia. Available from:
http://www.empr.gov.bc.ca/Mining/Permitting-Reclamation/ML-ARD/Pages/Policy.aspx
Ministry of Energy Mines and Petroleum Resources (EMPR). 2000. Operating Companies -
Construction Aggregates – Index. Available from:
http://www.empr.gov.bc.ca/Mining/MineralStatistics/MineralSectors/ConstructionAggregates/
CompaniesAndMaps/Pages/default.aspx
Ministry of Energy Mines and Petroleum Resources (EMPR). 2002. Aggregate Operators Best
Management Practices Handbook for British Columbia. Available from:

124
http://www.empr.gov.bc.ca/MINING/MINERALSTATISTICS/MINERALSECTORS/CONST
RUCTIONAGGREGATES/REPORTSANDPUBLICATIONS/Pages/AggregateOperators.aspx
Ministry of Energy Mines and Petroleum Resources (EMPR). 2006. British Columbia Mining Plan.
Available from: http://www.gov.bc.ca/empr/popt/reports/down/bc_mining_plan_02_2007.pdf
Ministry of Energy Mines and Petroleum Resources (EMPR). 2008. Mineral Exploration
Expenditure Time Series. Available from:
http://www.em.gov.bc.ca/Mining/MiningStats/61expendminexp.htm
Ministry of Energy Mines and Petroleum Resources (EMPR). 2009. Handbook for Mineral and Coal
Exploration in British Columbia. Available from:
http://www.empr.gov.bc.ca/Mining/Exploration/Documents/MXHandbook2008-09.pdf
Ministry of Energy Mines and Petroleum Resources (EMPR). 2010a. Mineral and Placer Claims
SVW. GeoBC Geographic Data Discovery Service. Available from:
https://apps.gov.bc.ca/pub/geometadata/metadataDetail.do?recordUID=49898&recordSet=ISO
19115
Ministry of Energy Mines and Petroleum Resources (EMPR). 2010b. MINFILE Mineral Inventory.
Available from:
http://www.empr.gov.bc.ca/MINING/GEOSCIENCE/MINFILE/Pages/default.aspx
Ministry of Energy Mines and Petroleum Resources (EMPR). 2010c. Metal Mine Location Maps.
Available from: http://webmap.em.gov.bc.ca/mapplace/minpot/metalEconomy.asp
Ministry of Energy Mines and Petroleum Resources (EMPR). 2010d. Coal Map. Available from:
http://webmap.em.gov.bc.ca/mapplace/minpot/coal.cfm
Ministry of Forests and Range (MoFR). 2005. MoFR Design criteria for works that may impact
forest roads or timber tenures. Available from:
http://www.for.gov.bc.ca/hth/engineering/documents/publications_guidebooks/publications_re
ports/MOF-IPP-Design-Criteria_July7-05_.pdf
Montgomery, D.R., J.M. Buffington, N.P. Peterson, D. Schuett-Hames and T.P. Quinn. 1996.
Stream-bed scour, egg burial depths, and the influence of salmonid spawning on bed surface
mobility and embryo survival. Canadian Journal of Aquatic Sciences 53(5): 1061-1070.
Morrison, J., M.C. Quick, and M.C.G. Foreman. 2002. Climate change in the Fraser River
watershed: Flow and temperature projections. Journal of Hydrology 263: 230-244.
Musick, J.A. 1999. Criteria to define extinction risk in marine fishes – the American Fisheries
Society initiative. Fisheries. 24: 6-14.
Myers, A. K., A.M. Marcarelli, C.D. Arp, M.A. Baker, and W.A. Wurtsbaugh. 2007. Disruptions of
stream sediment size and stability by lakes in mountain watersheds: potential effects on
periphyton biomass. North American Benthological Society. 26(3): 390–400.
Naiman, R.J., R.E. Bilby, D.E. Schindler, and J.M. Helfield. 2002. Pacific Salmon, Nutrients, and the
Dynamics of Freshwater and Riparian Ecosystems. Ecosystems 5 (4): 399–417.
National Council for Air and Stream Improvement (NCASI). 2001. Forest roads and aquatic
ecosystems: A review of causes, effects and management practices. Corvallis, Or.
National Round Table on the Environment and the Economy (NRTEE). 2010. Changing currents:
Water sustainability and the future of Canada’s natural resource sectors. Ottawa, ON. Available
from: http://www.nrtee-trnee.com/eng/publications/changing-currents/changing-currents-water-
report-eng.pdf
Nebeker, A.V., and J.R. Brett. 1976. Effects of Air-Supersaturated Water on Survival of Pacific
Salmon and Steelhead Smolts . Transactions of the American Fisheries Society. 105: 338-342.
Nebeker, A.V., D.G. Stevens, and R.K. Stroud. 1976. Effects of air-supersaturated water on adult
sockeye salmon (Oncorhynchus nerka). Journal of the Fisheries Research Board of Canada. 33:
2629-2633.

125
Nelitz, M., C. Murray, and K. Wieckowski. 2008. Returning Salmon: Integrated Planning and the
Wild Salmon Policy in B.C. Report prepared for David Suzuki Foundation, Vancouver, B.C. by
ESSA Technologies Ltd., Vancouver, B.C. Available from:
http://www.davidsuzuki.org/publications/downloads/2008/Returning_Salmon.pdf
Nelitz, M., C. Murray, M. Porter, and D.R. Marmorek. 2006. Managing Pacific salmon for
ecosystem values: Ecosystem indicators and the wild salmon policy. Prepared by ESSA
Technologies Ltd., Vancouver, B.C. for Pacific Fisheries Resource Conservation Council,
Vancouver, BC. Available from:
http://www.fish.bc.ca/files/EcosystemIndicators_2006_0_Complete%20for%20web.pdf
Nelitz, M., T. Douglas, and M. Rutherford. 2009. Freshwater for fish and people: Moving towards
“Living Water Smart”. Prepared for the Pacific Fisheries Resource Conservation Council.
Available from: http://www.fish.bc.ca/freshwater-fish-and-people-moving-towards-living-
water-smart
Nelson, R.L., M.L. Mchenry, and W.S. Platts. 1991. Mining. In W.R. Meehan (ed). Influence of
Forest and Rangeland Management on Salmonid Fishes and Their Habitats. American Fisheries
Society. Bethesda, Md. 751 pp.
Nelson, S. 2006. Fraser River Sockeye Salmon Benchmark Study: A Business Perspective on Fraser
Sockeye. Prepared for AAFC CAFI Program, Seafood Value Chain Roundtable.
www.bcseafoodalliance.com/documents/BenchmarkSockeye.pdf
Northcote, T.G., and Larkin, P.A. 1989. The Fraser River: a major salmonine production system. In
Proceedings of the International Large River Symposium. Edited by D.P. Dodge. Canadian
Special Publication of Fisheries and Aquaticc Sciences No. 106. pp. 172–204.
Paul, M.J., and J.L. Meyer. 2001. Streams in the urban landscape. Annual Review of Ecological
Systems. 32: 333-365.
Paulsen, C.M. and T.R. Fisher. 2005. Do habitat actions affect juvenile survival? An information-
theoretic approach applied to endangered Snake River Chinook salmon. Transactions of the
American Fisheries Society 134: 68-85.
Pavey, S.A., T.R. Hamon, and J.L. Nielsen. 2007. Revisiting evolutionary dead ends in sockeye
salmon (Oncorhynchus nerka) life history. Canadian Journal of Fisheries and Aquatic Sciences
64:1199-1208.
Pembina and Forest Ethics. 2010. Submission for B.C.’s Water Act Modernization. Available from:
http://pubs.pembina.org/reports/pembina-forestethicswamsubmission-final.pdf
Pestal, G., and A. Cass. 2009. Using Qualitative Risk Evaluations to Prioritize Resource Assessment
Activities for Fraser River Sockeye. Canadian Science Advisory Secretariat Research
Document 2009/071. Available from: http://www.dfo-
mpo.gc.ca/CSAS/Csas/Publications/ResDocs-DocRech/2009/2009_071_e.pdf
Peterman, R.M., D. Marmorek, B. Beckman, M. Bradford, M. Lapointe, N. Mantua, B. Riddell, M.
Scheuerell, M. Staley, K. Wieckowski, J. Winton, and C. Wood. 2010. Synthesis of evidence
from a workshop on the decline of Fraser River sockeye: June 15-17, 2010 Vancouver Island
Conference Centre Nanaimo, British Columbia. Prepared for the Pacific Salmon Commission,
Vancouver, B.C. Available from: http://www.watershed-watch.org/publications/files/PSC-
FraserSockeye-Aug2010.pdf
Pickard, D., D. Robinson, M. Porter, and K. Wieckowski. 2008. Fisheries Sensitive Watershed
(FSW) Monitoring Framework and Workplan. Report prepared by ESSA Technologies Ltd. for
BC Ministry of the Environment (MOE), Victoria, BC.
Platts, W.S. 1991. Livestock grazing. In W.R. Meehan (ed). Influence of Forest and Rangeland
Management on Salmonid Fishes and Their Habitats. American Fisheries Society. Bethesda,
Md. 751 pp.

126
Poole, G.C., and C.H. Berman. 2000. Pathways of Human Influence on Water Temperature
Dynamics in Stream Channels. U.S. Environmental Protection Agency, Region 10. Seattle,
WA.
Porszt, J. 2009. An evaluation of criteria for assessing conservation status of Fraser River sockeye
salmon Conservation Units. Master’s Research Project in the School of Resource and
Environmental Management, Simon Fraser University, Burnaby, B.C.
Province of British Columbia. 2002. Environmental Trends in British Columbia 2002. Metal
Leaching / Acid Rock Drainage Mine Sites Requiring Mitigation or Further Assessment.
http://www.env.gov.bc.ca/soe/et02/10_mitigation/minesites.html
Province of British Columbia. 2010. Independent Power Production in B.C.: An Inter-agency
Guidebook for Proponents. Available from:
http://www.env.gov.bc.ca/wsd/water_rights/waterpower/application.html
Quinn, T.P. 2005. The Behavior and Ecology of Pacific Salmon and Trout. American Fisheries
Society, Bethesda, MD, and University of Washington Press, Seattle, WA.
Radomski, P., L.A. Bergquist, M. Duval, and A. Williquett. 2010. Potential impacts of docks on
littoral habitats in Minnesota lakes. Fisheries. 35 (10): 489-495.
Rand, P.S. 2008. Oncorhynchus nerka. In: IUCN 2008. 2008 IUCN Red List of Threatened Species.
Available from: http://www.iucnredlist.org
Randall, R.G. 2003. Fish productivity and habitat productive capacity: definitions, indices, units of
field measurement, and a need for standardized terminology. Canadian Science Advisory
Secretariat Research Document 2003/061. Available from: http://www.dfo-
mpo.gc.ca/CSAS/Csas/DocREC/2003/RES2003_061_e.pdf
Redding, T., R. Winkler, D. Spittlehouse, R.D. Moore, A. Wei, and P. Teti. 2008. Mountain Pine
Beetle and Watershed Hydrology: A Synthesis focused on the Okanagan Basin. One Watershed
– One Water Conference Proceedings, October 2008. Available from:
http://www.forrex.org/program/water/PDFs/Workshops/owow/Redding_etal2008_MPB_CWR
A.pdf
Rempel, L. and M. Church. 2009. Physical and ecological response to disturbance by gravel mining
in a large alluvial river. Canadian Journal of Fisheries & Aquatic Sciences 66:52-71
RL&L Environmental Services Ltd. 1999. Seton Dam fish diversion project, 1999 investigations and
testing. Prepared for BC Hydro, Kamloops, BC.
RL&L Environmental Services Ltd. 2000. Seton Dam fish diversion project, 2000 investigations and
testing. Prepared for BC Hydro, Kamloops, BC.
Roberge, M.T., T. Slaney, and C.K. Minns. 2001. Life history characteristics of freshwater fishes
occurring in British Columbia, with major emphasis on lake habitat characteristics. Canadian
Manuscript Report of Fisheries and Aquatic Sciences 2574: 189 pp.
Rood, K.M., and R.E. Hamilton. 1995. Hydrology and water use for salmon streams in the Quesnel
Habitat Management Area, British Columbia. Canadian Manuscript Report of Fisheries and
Aquatic Sciences 2296. Available from: http://dsp-psd.pwgsc.gc.ca/collection_2007/dfo-
mpo/Fs97-4-2296E.pdf
Roos, J. F. 1991. Restoring Fraser River Salmon. Pacific Salmon Commission, New Westminster,
B.C., Canada.
Roos, J.F. 1991. Restoring Fraser River salmon: A history of the International Pacific Salmon
Fisheries Commission 1937-1985. Published by Pacific Salmon Commission. Vancouver,
Canada.
Roscoe, D.W. and S. Hinch. 2008. Fishway passage, water diversion and warming temperatures:
Factors limiting successful spawning migration of Seton-Anderson watershed sockeye salmon.
Final report for the Bridge Coastal Fish and Wildlife Restoration Program, Project 07.BRG01.

127
Rosenau, M. and M. Angelo. 2000. Sand and Gravel Management and Fish-Habitat Protection in
British Columbia Salmon and Steelhead Steams. Report prepared for the Pacific Fisheries
Resource Conservation Council.
Rosenau, M.L., and M. Angelo. 2009. Landscape-level impacts to salmon and steelhead stream
habitats in British Columbia. Prepared for Pacific Fisheries Resource Conservation Council,
Vancouver, B.C. Available from:
http://www.fish.bc.ca/files/LandscapeReport_2009_0_Complete.pdf
Salo, E.O., and T.W. Cundy (eds). 1987. Streamside Management: Forestry and Fishery Interactions.
University of Washington - Institute of Forest Resources. Seattle, Wa. 469 pp.
Scheibe, T.D., and M.C. Richmond. 2002. Fish individual-based numerical simulator (FINS): a
particle-based model of juvenile salmonid movement and dissolved gas exposure history in the
Columbia River basin. Ecological Modelling. 147: 233–252.
Schendel, E.K., H. Schreier, and L.M. Lavkulich. 2004. Linkages between phosphorus index
estimates and environmental quality indicators. Journal of soil and water conservation. 59 (6):
243-251.
Scheuerell, M.D., R.W. Zabel, and B.P. Sandford. 2009. Relating juvenile migration timing and
survival to adulthood in two species of threatened Pacific salmon (Oncorhynchus spp.). Journal
of Applied Ecology. 46: 983–990.
Schindler, D.E., R. Hilborn, B. Chasco, C.P. Boatright, T.P. Quinn, L.A. Rogers, and M.S. Webster.
2010. Population diversity and the portfolio effect in an exploited species. Nature 465: 609-
612.
Schindler, D.W., P.J. Dillon, and H. Schreier. 2006. A review of anthropogenic sources of nitrogen
and their effects on Canadian aquatic ecosystems. Biogeochemistry. 79: 25-44.
Schwab, J.W. 1998. Landslides on the Queen Charlotte Islands: Processes, rates, and climatic events.
In Proceedings of Carnation Creek and Queen Charlotte Islands fish / forestry workshop:
Applying 20 years of coastal research to management solutions. Land Management Handbook
41. BC Ministry of Forests. Victoria, BC.
Sedell, J.R., F.N. Leone, and W.S. Duval. 1991. Water transportation and storage of logs. In W.R.
Meehan (ed). Influence of Forest and Rangeland Management on Salmonid Fishes and Their
Habitats. American Fisheries Society. Bethesda, Md. 751 pp.
Selbie, D.T., M. J. Bradford, M. J. Hague, J.M.B. Hume, E.A. MacIsaac, D.A. Patterson. 2010. Are
freshwater habitat conditions in the Fraser River watershed an important contributor to the
Fraser sockeye salmon situation? Summary report prepared for the Pacific Salmon Commission
Workshop to Examine the Decline in Survival of Fraser River Sockeye. Vancouver, BC.
Sharma, S., S.C. Walker, and D.A. Jackson. 2008. Empirical modelling of lake water-temperature
relationships: a comparison of approaches. Freshwater Biology. 53: 897–911.
Shortreed, K.S., Hume, J.M.B., and J.G. Stockner. 2000. Using photosynthetic rates to estimate the
juvenile sockeye salmon rearing capacity of British Columbia Lakes. In E.E. Knudsen, C.R.
Steward, D.D. MacDonald, J.E. Williams, and D.W. Reiser (eds.), Sustainable Fisheries
Management: Pacific Salmon. pp. 505-521.
Shortreed, K.S., Morton, K.F., Malange, K., and J.M.B. Hume. 2001. Factors limiting juvenile
sockeye production and enhancement potential for selected B.C. nursery lakes. Canadian
Science Advisory Secretariat. Research Document 2001/098. 69pp.
Smerdon, B.D., and T.E. Redding. 2007. Groundwater: More than water below the ground.
Streamline 10(2): 1–6. Available from:
www.forrex.org/streamline/ISS35/Streamline_Vol10_No2_art1.pdf

128
Smerdon, B.D., T.E. Redding, and J. Beckers. 2009. An overview of the effects of forest
management on groundwater hydrology. BC Journal of Ecosystems and Management
10(1):22–44. Available from: http://www.forrex.org/jem/ISS50/vol10_no1_art4.pdf
Smith, I.M, K.J. Hall, L.M. Lavkulich, and H. Schreier. 2007. Trace metal concentrations in an
intensive agricultural watershed in British Columbia, Canada. Journal of the American Water
Resources Association. 43 (6): 1455-1467.
Spittlehouse, D. and T. Murdock. 2010 (draft). Choosing and applying climate change scenarios.
Prepared for Future Forest Ecosystem Scientific Council.
Stalberg, H.C., R.B. Lauzier, E.A. MacIsaac, M. Porter, and C. Murray. 2009. Canada’s policy for
conservation of wild pacific salmon: Stream, lake, and estuarine habitat indicators. Canadian
Manuscript of Fisheries and Aquatic Sciences. 2859: xiii + 135p.
Statistics Canada. 2009. Human Activity and the Environment: Annual Statistics 2009. Catalogue no.
16-201-X. Available from: http://dsp-psd.pwgsc.gc.ca/collection_2009/statcan/16-201-X/16-
201-x2009000-eng.pdf
Statistics Canada. 2010. Human Activity and the Environment: Freshwater supply and demand in
Canada. Catalogue No. 16-201-X. Available from: http://www.statcan.gc.ca/pub/16-201-x/16-
201-x2010000-eng.pdf
Stednick J.D. and T.J. Kern. 1994. Risk assessment for salmon from water quality changes following
timber harvesting. Environmental Monitoring and Assessment 32: 227-238.
Stednick, J.D. and R. Jensen. 2007. Effects of pine beetle infestations on water yield and water
quality at the watershed scale in Northern Colorado. USGS report for 2007CO153B. Available
from: http://water.usgs.gov/wrri/07grants/progress/2007CO153B.pdf
Stefan, H.G. and E.B. Preud'homme. 1993. Stream temperature estimation from air temperature.
Water Resources Bulletin 29(1):27-45.
Stewart-Oaten, A. 1996. Goals in environmental monitoring. Chapter 2, pp. 17-27, In: Schmitt, R.J.
and C.W. Osenberg (editors). Detecting Ecological Impacts: Concepts and Applications in
Coastal Habitats. Academic Press, San Diego, California
Stockner, J.G. and K. S. Shortreed. 1994. Autotrophic picoplankton community dynamics in a pre-
alpine lake in British Columbia, Canada. Hydrobiologia 274:133–142.
Stockner, J.G. and Shortreed, K.S. 1983. A comparative limnological survey of 19 sockeye salmon
(Oncorhynchus nerka) nursery lakes in the Fraser River system, British Columbia. Canadian
Techical Report of Fisheries and Aquatic Sciences No. 1190.
Technical Committee of the Nechako Fisheries Conservation Program (NFCP). 2005. Nechako
Fisheries Conservation Program technical data review: 1988 -2002.
Thompson, W.L. and D.C. Lee. 2002. A two-stage information-theoretic approach to modelling
landscape-level attributes and maximum recruitment of Chinook salmon in the Columbia River
Basin. Natural Resource Modeling. 15(2): 227-257.
Tilzer, M.M., C.R. Goldman, R.C. Richards, R.C. Wrigley. 1976. Influence of sediment inflow on
phytoplankton primary productivity in LakeTahoe (California-Nevada). Int. Rev. Gesamt.
Hydrobiol. 61: 169-181.
Triton Environmental Consultants Ltd. 2003. The 2001 summer water temperature and flow
management project. Nechako Fisheries Conservation Program Technical Report no. rm01-1.
Triton Environmental Consultants Ltd. 2004a. The 2002 summer water temperature and flow
management project. Nechako Fisheries Conservation Program Technical Report no. rm02-1.
Triton Environmental Consultants Ltd. 2004b. The 2003 summer water temperature and flow
management project. Nechako Fisheries Conservation Program Technical Report no. rm03-1.
Triton Environmental Consultants Ltd. 2006a. The 2004 summer water temperature and flow
management project. Nechako Fisheries Conservation Program Technical Report no. rm04-1.

129
Triton Environmental Consultants Ltd. 2006b. The 2005 summer water temperature and flow
management project. Nechako Fisheries Conservation Program Technical Report no. rm05-1.
Triton Environmental Consultants Ltd. 2006c. The 2006 summer water temperature and flow
management project. Nechako Fisheries Conservation Program Technical Report no. rm06-1.
Triton Environmental Consultants Ltd. 2008. The 2008 summer water temperature and flow
management project. Nechako Fisheries Conservation Program Technical Report no. rm08-2.
Triton Environmental Consultants Ltd. 2010. The 2009 summer water temperature and flow
management project. Nechako Fisheries Conservation Program Technical Report. rm09-1
University of British Columbia (UBC). 2010. Program on Water Governance, Factsheet on IPPs in
BC. Available from: http://www.watergovernance.ca/factsheets/pdf/FS_IPPs_in_BC.pdf
Uunila, L., B. Guy, and R. Pike. 2006. Hydrologic effects of Mountain Pine Beetle in interior pine
forests of British Columbia: Key questions and current knowledge. Streamline Watershed
Management Bulletin. Vol. 9, No. 2. Available from:
http://www.forrex.org/streamline/ISS31/streamline_vol9_no2_art1.pdf
Vancouver Fraser Port Authority (VFPA). 2008. Vancouver Fraser Port Authority Land Use Plan.
217 pp. Available from:
http://www.portmetrovancouver.com/Libraries/PROJECTS/2008_VFPA_Land_Use_Plan.sflb.
ashx
Voss, F.D., C. A. Curran, C.A., and M.C. Mastin. 2008, Modeling water temperature in the Yakima
River, Washington, from Roza Diversion Dam to Prosser dam, 2005–06: U.S. Geological
Survey Scientific Investigations Report 2008-5070, 42 p.
Waples, R.S., T. Beechie, and G.R. Pess 2009. Evolutionary history, habitat disturbance regimes, and
anthropogenic changes: What do these mean for resilience of Pacific salmon populations?
Ecology and Society 14(1): 3. [online] URL:
http://www.ecologyandsociety.org/vol14/iss1/art3/
Weitkamp, D.E., and M. Katz. 1980. A review of dissolved gas supersaturation literature.
Transaction of the American Fisheries Society. 109:659–702.
Welch, D.W., M.C. Melnychuk, E.R. Reichinsky, A.D. Porter, M.C. Jacobs, A. Ladouceur, R.S.
McKinley, and G.D. Jackson. 2009. Freshwater and marine migration and survival of
endangered Cultus Lake sockeye salmon (Oncorhynchus nerka) smolts using POST, a large-
scale acoustic telemetry array. Canadian Journal of Fisheries and Aquatic Sciences. 66: 736-
750.
Winkler, R.D., R.D. Moore, T.E. Redding, D.L. Spittlehouse, B.D. Smerdon, and D.E. Carlyle-
Moses. 2009 (draft). Chapter 7: The Effects of Forest Disturbance on Hydrologic Processes and
Watershed Response. In Compendium of Forest Hydrology and Geomorphology in British
Columbia. Available from:
http://www.forrex.org/program/water/PDFs/Compendium/Compendium_Chapter07.pdf
Wood, C.C. 1995. Life history variation and population structure in sockeye salmon. American
Fisheries Society Symposium 17:195-216.
Wood, C.C. 2007. Sockeye salmon ecotypes: origin, vulnerability to human impacts, and
conservation value. American Fisheries Society Symposium 54:1-4.
Wood, C.C., J.W. Bickham, R.J. Nelson, C.J. Foote, and J.C. Patton. 2008. Recurrent evolution of
life history ecotypes in sockeye salmon: implications for conservation and future evolution.
Evolutionary Applications 1:207-22.
Young, K.A. 2000. Riparian zone management in the Pacific Northwest: Who’s cutting what?
Environmental Management 26(2): 131-144.

130
Appendix 1 – Statement of work

131

132

133
Appendix 2 – Reviewer evaluations and author responses
The authors’ responses to each reviewer’s comments are provided in bold below.
Report Title: Evaluating the status of Fraser River sockeye salmon and the role of
freshwater ecology in their decline
Reviewer Name: John Reynolds
Date: 6 January 2011
1. Identify the strengths and weaknesses of this report.
This report is very thorough, clearly written, and thoughtful. I like the dashboard
summaries and the large amount of analyses that went into them. Most of the
weaknesses are due to lack of data availability rather than being the fault of the authors.
Many of the figures and their captions should be improved, as detailed in my report
below.
2. Evaluate the interpretation of the available data, and the validity of any derived
conclusions. Overall, does the report represent the best scientific interpretation
of the available data?
I have no major problems with the interpretation of the data and the conclusions.
3. Are there additional quantitative or qualitative ways to evaluate the subject
area not considered in this report? How could the analysis be improved?
There are very few statistical analyses, and I do have concerns (e.g. Table 15) as
explained in my detailed report. I have also suggested below the use of more stream
data from other sources.
4. Are the recommendations provided in this report supportable? Do you have
any further recommendations to add?
Yes, these seem straight-forward, if a little vague. I have suggested in my report that it
would be nice to flesh these out a little, e.g. monitoring, or at least mention other similar
calls to arms.
5. What information, if any, should be collected in the future to improve our
understanding of this subject area?
I agree with the suggestions in the report.
6. Please provide any specific comments for the authors.
See below.

134
Review of Freshwater Report – John D. Reynolds
Earth to Ocean Research Group, Department of Biological Sciences, Simon Fraser University
I found this report to be exceptionally readable and clear, with a well thought-out progression from
objectives to methods and results of the analyses. Each of the key steps in methods and
corresponding data are presented clearly in tables, making this a very accessible report.
The authors have faced a key short-coming: lack of time series of change for most of the habitat
stressors they consider. This is noted on p. 16: there is “a general lack of information that could be
used to reliably define dynamic changes in condition across sockeye spawning, rearing, and
migratory habitats…”. As a result, much of the report is based on relative differences among
sockeye conservation units in indicators of status. This matches the terms of reference for this study.
But that comment by the authors is very important when considering the terms of reference for the
Cohen Commission itself, in particular the investigation of declining productivity and the low returns
of 2009. The lack of temporal information weakens the ability of this report to help the Commission
understand the changing fortunes of Fraser Sockeye salmon. All is not lost, of course, as there is
some temporal information here, such as with forestry and the spread of mountain pine beetles.
Furthermore, much of the other information in this report will provide a foundation for
understanding spatial variation in status and vulnerability. But I do think we should understand the
limitation that’s summarized in a sentence on p. 16 right up front. I want to emphasize that this is
not a criticism of the report, but a lament about the data that were available in the tight timeline, and
I give the authors a lot of credit for stressing on p. 17 the fact that their vulnerability indicators for
each CU are relative to the other CUs.
What we have, then, is in my opinion the most comprehensive analysis that could be expected in the
time available, summarizing a huge amount of patchy information. This summary is aided by an
innovative dashboard representation in the Appendix of the status and trends (where available) for
each sockeye conservation unit.
I agree with the authors’ conclusion that changes in freshwater habitats are unlikely to be the main
cause of the decline in productivity of Fraser River sockeye salmon. While freshwater habitat status
never was very high on most peoples’ list of suspects, including the Peterman et al. Pacific Salmon
Commission report, we needed more to go on more than mere impressions. This report looks hard
for evidence to support the freshwater habitat theory, and does not find it.
Below I provide a set of comments that range from minor editorial to more substantive.
p. 3. Line 1. “Both 2009 and 2010 returns were within the statistical distributions of forecasted
returns but at opposite ends of these distributions.” This depends on how the statistical distributions
are defined. There is always some probability of stocks falling within a certain range, even if it is
very small. This sentence should be more precise in giving the probabilities for each of the years.
Response: Included probabilities for returns in 2009 and 2010.
p. 4, top. I agree with the value of comparing findings from this report with those of the Peterman et
al. 2010 report, which examines additional habitats, but note that the Peterman et al. report was
somewhat preliminary, and I understand that another report that uses more standardized methods of
analysis is currently being done.

135
Response: Added “preliminary review” to description of Peterman et al. report.
Fig. 1. The caption should state that the data are based on rolling averages, which smooth out annual
variation.
Response: Revised figure caption to include this point.
Fig. 3. It’s not clear from the caption what, exactly, this figure is meant to show. Only major
nursery lakes are shown, presumably the 18 for which good time series data are available, and these
are not labeled. The caption should probably mention that there are others. There are three shades of
grey, but only dark grey is referred to.
Response: Revised figure caption to clarify purpose of figure and different uses of shading.
Fig. 4. The tops of some of the lake names are cut off.
Response: Revised figure to ensure CU names are fully included.
p. 6. Bottom. As the authors note, there are many ways of determining status of fish populations. It
would help to have an explanation of why they chose to focus on the three that they did, and ignored
the others. If this was for pragmatic reasons given limited time, for example, they could state this.
Response: Inserted text explaining rationale for why alternative two methodologies were
chosen as well as why did not include additional methods in our review.
The tables do not indicate what, exactly, each of the three methods of determining status, are aimed
at. In other words, do each of the three methods use exactly the same definition of “status”? For
example, fisheries-type methods are usually focused on status in terms of productivity or reference
points for population biomass, whereas IUCN-type methods (which includes COSEWIC) focus on
extinction risk. Some discussion of the objectives of status methodology would help here.
Response: Inserted paragraph at the beginning of section 2.1.2 speaking to this point.
Described status as defined by each of the methods. Also added a row to Table 2 summarizing
status definition.
p. 8, line 9. I believe the reference should be to Table 2, not Table 1.
p. 10, line 18. Typo “is” should be “in”.
p. 19, line 22. Typo should be “handling” not “handing”
p. 90, Table 2, line 10 up from bottom. Should be “life” not “ife”
Response: All of these changes have been made.
p. 10, bottom. In assessing the pros and cons of Pestal and Cass vs Holt et al., it might be worth
using both methods to assess the same stocks where data permit, to ask how congruent their answers
are, and then if the answers are highly correlated, finish the more data-deficient stocks using Pestal
and Cass only (though there will still be a problem for the 11 of 6 CUs that even the Pestal and Cass
method cannot handle due to lack of information.)

136
Response: Valid point, however it is beyond the scope of our analyses to undertake such a
comparison. Holt et al’s (2009) has not been applied across all CUs and we do not have the
resources to carry out this analysis. Consequently, it is not possible for us to compare the
results of Holt et al. (2009) to those of Pestal and Cass (2009). Grant et al. 2010 (modified
version of Holt et al. 2009) has been applied across all CUs and we have done a rudimentary
comparison of the results of their work to that of Pestal and Cass (2009). Changes in Pestal and
Cass’ status assessment based on the work of Grant et al. are illustrated in Figure 5 and
summarized in Table 1. To address this point, we also included a suggestion that DFO should
undertake a more structured and quantitative comparison.
p. 12. Bottom. I see the logic of distinguishing between populations that spawn upstream or
downstream of lakes, but I question the assumption that those that are downstream of lakes are not
affected by what happens upstream in the watershed. Indeed, the authors acknowledge this in the
next section on nursery lakes. I feel that this assumption therefore warrants further justification.
Response: We continue to support this assumption. In response to the reviewers comment, we
added a statement with citations strengthening the justification in the report.
p. 13. In addition to using air temperatures as a surrogate for stream temperatures, why not use
actual stream temperatures where available, e.g. from the Fraser mainstem, which are readily
available from the Water Survey of Canada and other places, such as the Fraser River Environmental
Watch program headed up by Dave Patterson at DFO?
Response: We agree that it would have been better to use actual stream temperatures in our
analysis as opposed to / in addition to air temperature. However, inclusion of these data was
not possible / practical for a combination of reasons.
First, the tasks to assess the “extent of en-route mortality and pre-spawning mortality” and
“impacts to stock status and potential causes of premature migration of adult (Late Run)
sockeye into freshwater” were shifted from our scope of work to that of another project for the
Cohen Commission (see Hinch, S.G. and E.G. Martins. 2011. A review of potential climate
change effects on survival of Fraser River sockeye salmon and an analysis of interannual
trends in en route loss and pre-spawn mortality. Cohen Commission Tech. Rep. 9). We
acknowledge, however, that water temperature is an important consideration for
understanding survival across adult migration, so we wanted to capture this variable in some
way despite the reduced emphasis in our scope of work.
Second, a key motivation guiding our selection and development of habitat indicators was to
ensure that the indicator could be generated for all Conservation Units over a long time series,
and that the indicator was different than what others have already tested. Based on our
understanding, actual stream temperature data are not available across the entire extent of
large river migration for all 30 lake Conservation Units, while the air temperature indicator
that we selected was available at regular intervals across the full extent of migration corridors
for a long time series. As well, Selbie et al. (in Appendix C of Peterman et al. 2010) examined
the potential influence of stock specific Fraser mainstem (at Qualark) water temperatures in
declines of sockeye salmon and found no significant relationship, which motivated us to use a
different indicator for our analysis.

137
Lastly, data from the Fraser River Environmental Watch program were deemed not readily
available to represent water temperatures at multiple locations across the migration corridors
because Selbie et al. did not develop or use this more complicated measure even though Dave
Patterson was a contributor to their analysis (they chose to use the simpler measure of Qualark
water temperatures).
p. 14. Thermal trigger for smolt outmigration. The logic of using this indicator is that a temperature
cue for outmigration could produce a mismatch between outmigration timing and conditions such as
food that affect survival. I don’t understand how a measurement of springtime air temperature
captures this mismatch potential. Is a high temperature good or bad? This would depend on the
phenology of food and predators that the smolts will encounter when they migrate, but that
match/mismatch is not captured by this metric.
Response: We have clarified the text in the report to say that springtime air temperature at the
nursery lake is being used as an indicator of the timing of ice break-up (one of the cues of smolt
outmigration). Thus, this indicator is a surrogate of the potential for mismatch between the
timing of arrival at the estuary and timing of other conditions in the estuary, not a direct
surrogate of the magnitude of mismatch. If a relationship exists, the magnitude of mismatch
would depend on how springtime air temperatures have changed relative to historic conditions
to which local stocks have adapted.
As illustrated by the correlations in Table 16, our initial examination of the relationship
between springtime air temperatures and total productivity indices for some stocks suggested
that years with warmer springtime air temperatures (and presumably earlier ice-break up and
smolt outmigration) were associated with years of lower total productivity. However, upon
further examination this relationship was not significant across many stocks.
p. 17. Personally, I don’t find migration distance to be a very compelling habitat indicator, though I
acknowledge the relationships found by Selbie et al. in the Peterman et al. 2010 report. I am not
suggesting the authors should drop this, and it does hold up in the simple regression presented later,
but it would help to have more rationale for using it.
Response: We maintain our support for and use of this indicator as a measure of the relative
habitat vulnerability across Conservation Units for three reasons. First, as mentioned by the
reviewer, the relationship found by Selbie et al. (in Peterman et al. 2010) suggests that there
may be an underlying biological mechanism related to upstream or downstream migration that
is differentially affecting survival of sockeye salmon stocks across the Fraser basin. Second,
this mechanism is plausible as we intuitively expect sockeye salmon with longer migration
distances will spend relatively longer periods in freshwater during their migration, which in
turn would increase the chance and magnitude of exposure to harmful stressors, including
diseases, parasites, contaminants, and high water temperatures. Lastly, in our view migration
distance is the best indicator available, given a need to use only one indicator that represents
habitat vulnerability across both upstream and downstream migrations and a lack of other
options due to data limitations.
To strengthen our rationale for using this indicator we have included some of these points in
the report.

138
p. 19. It would be helpful to explain here why urbanization is only considered upstream of Hope, i.e.
because downstream of Hope is covered in another report.
Response: Included a footnote clarifying how urbanization downstream of Hope is captured by
Johannes et al. 2011.
Fig. 7 (p. 60). It would help to have labels on the circles to indicate which CUs they are.
Response: Given a lack of spacing between circles in this figure, we can not include labels for
all CUs. We also do not believe that such labeling is necessary given the purpose of this figure
and other options for viewing this information. The intent of this specific figure is to illustrate
whether there are relationships among the three variables of habitat vulnerability, not to
illustrate the value of these dimensions for a particular CU. The dashboard summaries in the
Appendices include this figure, which highlights values for all CUs individually. As well, Figure
38 illustrate the relationship between each vulnerability indicator and level of cumulative stress
for each CU (with labels).
The text in the report has been revised to clarify the purpose of this figure and other options
for viewing this information.
Fig. 9. This is the first of many figures using this format, and it would be helpful to take readers by
the hand through it. When I first encountered it I found it very hard to interpret. More explanation
in the caption, and labeling of the y-axes, would help. The word “cumulative” in the caption seems
misleading, and I think it would be better described as a frequency distribution (if I understand
correctly). That leads to the question of what the units are for the x-axis.
Response: We have ensured that both x and y axes are labeled on this and all similar graphs.
We have also clarified the captions of these figures to help with their interpretation and ensure
it is clear that they are indeed frequency distributions. The use of the word “cumulative” to
represent the forest harvesting data was a term used to represent the way the indicator was
generated (i.e., a summation of forest harvesting across a rolling 15 year window). Given the
confusion, we have removed the use of this term in the caption and the text of the report has
changed to clarify how the indicator was calculated.
Fig. 10. What is the scale indicated by the dots?
Response: Clarified caption to clarify that the “dots” are actually forest cutblock polygons,
which appear as dots at this scale.
Fig. 11. Better to just use CU names and not codes in the legend. Then the codes can also be
scrapped from the text here and elsewhere. If these are cumulative plots, why do the lines go down
sometimes? From text elsewhere, I think these must be some sort of rolling averages, not cumulative
plots.
Response: We have removed the use of CU codes in the legend. Again, the use of the word
“cumulative” to represent the forest harvesting data was a term used to represent the way the
indicator was generated (i.e., a summation of forest harvesting across a rolling 15 year

139
window). It is possible that the values go down over time. Given the confusion, we have
removed the use of this term in the caption and the text of the report has changed to clarify
how the indicator was calculated.
Fig. 14. I suggest this figure can be omitted as Fig. 16 makes the same point much better.
Response: Removed Figure 14.
p. 28. Placer mining and gravel mining. Isn’t there anything to say about these, other than these
cursory descriptions of the activities? Gravel extraction from the mainstem of the Fraser River is a
very controversial issue, especially (but not exclusively?) downstream of Hope, which is outside the
scope of this report. But are there no data on changes in gravel extraction over time above Hope?
This topic is in the spotlight.
Response: We have added a paragraph discussing the issue of gravel extraction in more detail.
In general, our analysis reveals that there is little overlap between areas of gravel / placer
mining and spawning habitats for sockeye salmon.
I didn’t notice any reference to the Coquitlam dam, though Bocking and Gavoury’s report is cited.
The damage was probably done well before the drop in aggregate Fraser productivity since the early
1990s, but the authors could consider mentioning such projects.
Response: We included Coquitlam dam in our list of hydroelectric facilities in the Lower
Fraser, but note that construction of this dam in the early 1900s pre-dates the recent declines
of sockeye salmon in the Fraser.
p. 35. The discussion of water temperatures in the Nechako in relation to the Kemano power project
refers to how often (or rarely) the temperature exceeds 20 C, but as stated elsewhere in the report,
sockeye in the Fraser start experiencing difficulties in the upper teens. This is a very detailed
discussion about the history of the operation, but what is the message that readers are to take from it?
I would reduce the historical detail and cut more quickly to the chase.
Response: Edited the historical preamble to remove some unnecessary details.
Fig. 28. I presume this is a work in progress because this map is very poor resolution and there is no
legend to interpret the shading.
Response: Imported a map with better resolution and changed the caption to clarify use of
shading.
Figs. 29 and 30. There is no legend for the colour coding so I have no idea what this means. It
would be better to simply use lines instead of dots.
Response: Changed figures so they are line graphs and labeled a sub-set of CUs on this images.
Time series of population density for individual CUs are represented in the dashboard
summaries.
Fig. 32. It would help if the figures could stand better on their own, which could include explaining

140
in the caption what an “animal unit equivalent” is, exactly, though I can guess.
Response: Revised caption to clarify the term “animal unit equivalent”.
Fig. 34. Same comment as for Fig. 28.
Response: Imported a map with better resolution and changed the caption to clarify use of
shading.
p. 42. Water use. The report does the best it can within the time available to provide a snapshot of
spatial variation in some kinds of water use in the Fraser watershed. The fact that BC does not
monitor actual rates of consumption, including groundwater, is very troubling, and the Commission
should be aware of this short-coming (no fault of the authors, of course). Another limitation, which
is not brought out explicitly in the report, is that there is no information presented on changes in
water use through time. Therefore it is not possible to assess whether increases in water use relative
to availability over the past 15 years have corresponded with the decline of salmon stocks in those
areas relative to areas where consumption has not increased to the same extent. It would be very
helpful to have such an analysis if there are any data to support it.
Response: Included these points in the weaknesses / limitations description of these data in the
report and clarified the need for time series of allocations and use in the recommendations. We
also note, however, that even with such data it remains difficult to attribute cause and effect
between water use and fish population dynamics given the high natural variability in flow and
uncertainty in defining thresholds for ecosystem needs for water.
p. 50 and Table 15. This is an important table, containing virtually the only statistical analyses in the
entire report (except for Table 16). I do not think that a simple step-wise regression does the
question justice, as there are much more sophisticated ways of handling multiple explanatory
variables (which are often correlated). I would prefer to see first a correlation matrix between the
explanatory variables, or at least have some indication of how correlated they were, followed by AIC
to compare the explanatory power of competing models. Stepwise regressions can give misleading
answers based simply on what other variables are in the equation. At the very least the authors
should test for this effect by trying different combinations of variables.
Response: The draft final version of the report did not fully describe the details of our original
analysis. Our original analyses looked at the correlation among different explanatory
variables, and used an AIC approach to compare the explanatory power of competing models.
It was our oversight to not include these details in the draft final, and have included more
detail in the final report (i.e., correlation matrix and AIC comparison of competing models).
I found the summary provided by Table 17 to be very helpful.
p. 55. Recommendations. These seem fairly straightforward. There have been many calls for
improved monitoring. The authors could draw from some of these to flesh out what an effective, and
fully costed-out, long-term monitoring scheme would look like for watersheds, including full
representation of small systems. I’m thinking of some of the projects funded by the Fraser Salmon
and Watersheds Program, such as the harmonized monitoring initiative. Also, what about changes to
federal or provincial legislation? Are salmon freshwater habitats protected adequately by existing

141
legislation?
Response: We agree that our recommendations are relatively straightforward, and that clear,
specific, and fully costed recommendations are always better. However, we highlight that our
scope of work was focused on an evaluation of freshwater factors in the decline of sockeye
salmon. We were not tasked with a review of existing legislation or monitoring initiatives
across different agencies and whether they are sufficient to protect freshwater habitats or
detect cause and effect relationships between human stressors and declines of sockeye salmon.
Having said this, we have added more details to our recommendations to include more
examples where other reports have commented on the need for improved monitoring and
integration across agencies in BC and elsewhere across the Pacific Northwest.

142
Freshwater Ecology in their Decline
Reviewer Name: Ken Ashley
Date: January 6, 2011
1. Identify the strengths and weaknesses of this report.
The strength of the report is that it used a comprehensive GIS analysis based approach
to examine the status of and threats to the freshwater environment for each of the
identified 36 Fraser River sockeye Conservation Units (30 lake and 6 river type CU’s),
and generated a ‘Dashboard Summary’ for each Conservation Unit.
The weakness of the report is that the ‘Dashboard Summaries’ are somewhat
complicated and not particularly intuitive, and the report was mainly GIS based and
used existing information, hence did not explore in any detail the possibility of climate
change altering the underlying ecological processes in Fraser Basin sockeye nursery
lakes that could reduce their capacity to produce healthy fry and smolts.
Response: In Section 4.1 we have clarified the purpose of dashboards as providing the
greatest level of detail describing the population status, habitat vulnerability, and habitat
stressors for each Conservation Unit. This level of detail is being presented in the report
while also providing readers with other summaries that simplify the level of detail in the
dashboards into more digestible formats (see Figures 5, 7, and 38). We believe it best to
provide a variety of ways to summarize the data given the abundance and complexity of
information that is available.
An explicit and detailed consideration of the effects of climate change on Fraser River
sockeye was covered by another report for the Cohen Commission (see Hinch and Martins
2011), and was not the primary purpose of our work.
2. Evaluate the interpretation of the available data, and the validity of any derived
conclusions. Overall, does the report represent the best scientific interpretation
of the available data?
The interpretation of the available data and the validity of the derived conclusions are
sound.
The report represents a scientifically defensible interpretation of the available data.
3. Are there additional quantitative or qualitative ways to evaluate the subject
area not considered in this report? How could the analysis be improved?
The analysis could be improved by exploring the possibility that climate change is
altering the underlying ecological processes in Fraser Basin sockeye nursery lakes that
could reduce their capacity to produce healthy fry and smolts. The reality is that most of
this data is not currently available. However, in a few lakes there may be adequate time
series of duration of thermal stratification and concentration of limiting nutrients in the

143
epilimnion to conduct a preliminary analysis.
Ideally, data on size fractionated primary production and the biomass and species
composition of phytoplankton and zooplankton are required to examine the trend over
the past 50 years to determine if there has been a change in the length of thermal
stratification, the concentration and ratio of limiting nutrients, and the amount and quality
of juvenile sockeye food supply in these nursery lakes.
Response: See response under Section 5.
4. Are the recommendations provided in this report supportable? Do you have
any further recommendations to add?
The recommendations provided in this report are supportable.
I recommend that a limnology program is initiated on a sub-set of sockeye nursery lakes
to examine the hypothesis that the quantity and quality of planktonic food required to
maintain historical sockeye productivity may be declining as a result of climate change.
Response: We agree that limnology data are currently limited to a few nursery lakes for a
limited time series, and that better data could be collected to improve our understanding of
change in nursery lake conditions and juvenile productivity. We have explicitly added the
need for improved monitoring of nursery lakes in the recommendations.
5. What information, if any, should be collected in the future to improve our
understanding of this subject area?
It has been known for some time that one of the fundamental differences between
coastal and interior sockeye nursery lakes is the length of their respective pelagic food
webs, and the existence and role of the ‘microbial loop’.
In general, interior sockeye nursery lakes have shorter food webs; hence intrinsically
produce more sockeye per unit limnetic area or volume than coastal lakes (see Figure
1).

144
Figure 1. Schematic of pelagic food webs of interior and coastal BC sockeye lakes.
The reason for this intrinsic difference in productive capacity is that most of the carbon
fixed in interior lakes flows through a shorter nanoplankton and microplankton food web
to sockeye juveniles, whereas in coastal lakes a greater fraction of the carbon flows
through longer picoplankton and microbial pathway and less to sockeye (see Figures 2
and 3).

145
Figure 2. Relative magnitude of energy and carbon flow in coastal and interior sockeye
lakes.
One of the known effects of climate change on lakes is an earlier melting of ice, and/or
a longer period of thermal stratification. If climate change is increasing the duration of
thermal stratification in Fraser Basin sockeye lakes, which are naturally oligotrophic
(i.e., nutrient poor), in theory, this could cause a shift towards a less productive food
web as the lakes would stay stratified longer at the end of the summer when there was
less watershed nutrient loading, which would trend the lake towards becoming less
productive, a process known as ‘oligotrophication’.
Characteristics of microbial loop dominated lakes are increased C production by pico-
nanoplankton, nutrient regenerative recycling systems, long food chains, microbially
dominated long food webs, trophically inefficient, microzooplankton dominated and little
or no benthic-pelagic coupling and low pelagic or demersal fish production.

146
Figure 3. Schematic of the microbial food web relative to the ‘classic food web.
In addition, climate driven changes in watershed nutrient loading may also alter the
concentration and chemistry of limiting nutrients in the epilimnion. This was observed to
be occurring in Okanagan Lake during the 1990s and 2000s, where the ratio of
dissolved inorganic nitrogen (DIN) to soluble reactive phosphorus (SRP) declined,
which produced phytoplankton and zooplankton which contained lower concentrations
of essential fatty acids for juvenile kokanee (same species as sockeye – O. nerka).
It was hypothesized that this subtle change in watershed nutrient loading and resultant
N:P ratio was responsible for the decline of kokanee in Okanagan Lake, because the
quality of juvenile kokanee food had been reduced. This was experimentally verified by
a series of enclosure and lab experiments conducted by Dr. Mike Brett and Dr. Joe
Ravet of the University of Washington in Seattle (see Figures 4 and 5).

147
Figure 4. Daphnia growth and biochemical composition responses to different food
types. Note: Cyano = Blue-green algae, Chlor = green algae, Crypto = Cryptomonad
algae.
Figure 5. Nutrient addition experiment conducted in Okanagan Lake in July, 2003
demonstrating that N and P and micronutrients were required to produce the highest
concentration of essential fatty acids in phytoplankton, necessary for the production of
high quality planktonic food for juvenile kokanee.
In summary, it is possible that subtle climate driven changes in the ecology of the
nursery lakes may be producing less food, or food of lower quality than historically,
which is creating higher mortalities in juvenile sockeye during their migration to the open
ocean as they are nutritionally deficient in energy reserves.
This may be the explanation to statements given at the Nov 30-Dec 1/10 workshop and
in various Cohen Commission science reports that “…This observation indicates either

148
that the primary mortality agents in the sockeye occurred in the post-juvenile stage, or
that certain stressors that were non-lethal in freshwater caused mortality later in the
sockeye’s life history” (Peterman et al., 2010).
Nutritionally deficient juvenile sockeye may be more susceptible to variations in food
quantity and quality in Georgia Strait, to ward off microbial pathogens and parasitic sea
lice from open net pen fish farms, and variable ocean productivity on their early ocean
migrations in Georgia Strait and on the continental shelf.
In other words, some juvenile sockeye may have left home on an empty stomach, or a
diet of junk food, and were poorly equipped to deal with the rigors of smoltification,
migration and predator/disease avoidance.
Response: We acknowledge the plausibility of this hypothesis. We are also aware, however,
that Selbie et al. (in Appendix C of Peterman et al. 2010) examined the data for Quesnel,
Shuswap, and Chilko Lakes to investigate whether changes in growth and primary /
secondary productivity have occurred and found no detectable changes over time. Given
the evidence above that interior and coastal lakes have different food webs, it is not clear
how or whether this hypothesis is consistent with the observation that Harrison (a coastal
nursery lake) and Shuswap Complex (an interior nursery lake) CUs have not seen declines
to the same extent as other CUs (which include a mix of coastal and interior nursery lakes).
We also note that the effects of climate change are being considered by another report (see
Hinch and Martins 2011) and was not the primary purpose of our work. For these reasons
we do not believe it would be feasible or practical for us to explore the available data to test
this hypothesis.
We do, however, believe this hypothesis is worth testing in the future with better research
and monitoring of a strategically selected set of inland lakes, and would be informed by our
recommendation to improve monitoring of smolt condition and timing of outmigration.
6. Please provide any specific comments for the authors.
There are no specific comments for the authors.
Literature cited:
Peterman et al. 2010. Synthesis of evidence from a workshop on the decline of Fraser
River sockeye. June 15-17, 2010. A report to the Pacific Salmon Commission,
Vancouver, B.C., 123 pp. + 35 pp. of appendices.
Stockner, J.G. 1991. Autotrophic picoplankton in freshwater ecosystems: the view from
the summit. Int. Revue ges. Hydrobiol. 76:483-492.
Weiss, T. and J.G. Stockner. 192. Eutrophication: the role of the microbial food web.
Memorie dell’Istituto Italinao di Idrobiologia 52:133-150.
Stockner, J.G., E. Rydin and P. Hyenstrand. 2000. Cultural oligotrophication: causes
and consequences for fisheries resources. Fisheries 25:7-14.

149
Report Title: Freshwater Ecology
Reviewer Name: Eric B. Taylor
Date: January 2011
1. Identify the strengths and weaknesses of this report.
The report is well-written and accomplishes the rather daunting task of both reviewing
the general life history of sockeye salmon in freshwater as well as potential and actual
land uses that may impact sockeye salmon.
1. The k-means clustering method (page 46) is a useful approach to identify groups
of similarly-impacted sockeye salmon CUs. An alternative would be to employ
the IUCN Threats Calculator which is basically an excel-spreadsheet-based
calculator that generates overall threat levels to populations from qualitative input
on cumulative threats. This would allow an alternative objective ranking of each
CU based on an internationally-recognized threats assessment system. I am a bit
concerned that there are different threat-based assessments going around and
how they will be integrated. For instance, COSEWIC is reviewing the status of
sockeye salmon CUs and may use the IUCN calculator and how it will mesh with
what is in this report may be an issue.
Response: Early on we decided to use the current assessment approach for our evaluation
and received support from the Cohen Commission for doing so. It might have been possible
to use an alternative approach if directed to do so earlier on, but is not possible to go back
at this time to use an alternative assessment method.
Regardless, we stand by the scientific defensibility of our approach because it uses the best
available data to quantify the magnitude, spatial extent, and where possible, temporal
changes in stressors across sockeye habitats. It is an approach that relies on detailed
information about the specific location of stressors and vulnerability of specific habitats
across CUs. Although internationally recognized, the IUCN alternative is a more generic
approach that does not explicitly account for the level of detail considered herein.
Moreover, qualitative interpretations of the cumulative level of stress would be required as
inputs, which would ultimately be based on the kind of data generated in this report. If not,
the inputs would rely on expert based interpretations of the cumulative level of stress which
could be difficult to justify with existing evidence. For these reasons we believe the
quantitative approach used here is more objective, transparent, and defensible than a
generic assessment tool that uses qualitative inputs.
As well, we are not as concerned about the use of alternative methods for assessing threats.
Regardless of whether there are differences or similarities in results, these findings would
be important to document and understand. Differences in results would highlight CUs
where we are least certain about the cumulative level of stress. CUs with similar relative
rankings of the levels of stress would highlight areas where we are the most certain. If the
models provide completely different results, it would be important to understand the
dynamics driving each model, underlying assumptions, and why differences exist.
Comparative techniques and methods for explicitly considering uncertainties are available.
The resulting insights would be important to understand if COSEWIC were developing

150
conclusions about threats.
2. As in at least one other report, I am struck by the lack of comparative analysis to
other population aggregates in British Columbia. For instance, this report makes
no mention of trends in other important areas like the Skeena River or Barkley
Sound? Surely some information could be obtained that might support or refute
some of the conclusions of this report. For instance, have any of these other
areas been assessed for habitat changes in freshwater or habitat vulnerability
and population status? If these aggregates have shown less fluctuation than
Fraser populations and yet experience similar changes in freshwater parameters
relevant to productivity would this not be useful information and support the
authors’ overall conclusion that freshwater conditions are unlikely to be the
primary driver of fluctuations in Fraser River sockeye salmon adult abundance?
Response: We agree that insights from a comparative analysis would be useful for
providing support for / against our conclusions. Moreover, such a consideration is
consistent with the questions from Stewart-Oaten (1996) that we use to guide our
assessment of the role of different freshwater factors in declines of sockeye salmon.
However, a comparison to other systems was not within our scope of work, and not possible
with the time available. In doing our work, we were also aware that a comparative analysis
of patterns of productivity for sockeye salmon across the North Pacific was being
completed by another Cohen Commission project (see Peterman, R.M. and B. Dorner.
2011. Fraser River sockeye production dynamics. Cohen Commission Tech. Rep. 10). Our
understanding is that the results from this other study would help assess whether the
patterns of decline are unique to the Fraser or more broad-scale, which can help support or
refute some of our conclusions. Moreover, the Cohen Commission project on cumulative
effects (see Marmorek et al. 2011. Fraser River sockeye salmon: data synthesis and
cumulative impacts. Cohen Commission Tech. Rep. 6) is tasked with integrating the
findings of individual projects to identify consistencies / inconsistencies and assess the role
of factors influencing all life stages in declines of sockeye.
3. Section 5 (The State of the Science) is the weak point of the report. It is quite
vague and does not really summarize the “state of the science”. What, for
instance, has been the major progress made in understanding sockeye salmon
ecology and persistence in terms of freshwater ecology? What are the remaining
uncertainties? What are examples of minor populations for which information is
lacking? Who collects watershed-level data, how can the Province of BC
contribute. Does DFO have the capacity or willpower to initiate these critical
studies? What are the specific research questions that remain unanswered?
Response: We have strengthened the “State of the science” section to highlight
uncertainties about our state of existing knowledge (e.g., understanding of the population
level effects of freshwater stressors) and state of existing data (e.g., gaps in space or time
for populations, life history stages, or specific stressors). This summary and clarification is
intended to help justify the recommendations that follow. It was not our purpose with this
section or the report to summarize the state of and capacity for monitoring within
provincial and federal agencies (and we were not tasked to do so). Also, the brevity /

151
vagueness of this section was purposeful because we were constrained to writing a
summary of the “State of the science” within a 1 page limit.
4. The statistical analyses (Table 15 and 16) seem, perhaps, a bit simplistic and
dated. Just wondering if any alternative Bayesian type model construction and
assessments have been considered. There are methods such as “Bayesian
Belief Networks” that assess species (or CUs within species) under uncertainty
that might be useful. See Environmental Modelling & Software 25 (2010) 15–23
Response: We agree that the statistical analyses presented in this report are relatively
simple, but also acknowledge that our original description of these analyses was limited. In
the final version of the report we have added a matrix of correlations among different
explanatory variables, and described our use of an AIC approach to compare the
explanatory power of the competing models evaluated in our analyses.
Our ability to develop a more complex model (using a Bayesian Belief Network, for
instance) was constrained by a number of factors. First, we were limited in our ability to
account for other explanatory factors that would likely have important effects on survival
across the entire life cycle and freshwater life stages. These factors were being quantified
by other Cohen studies (e.g., effect of marine conditions, contaminants in freshwater, in
river conditions and enroute losses, or stressors in the lower Fraser River). Second, we
were faced with severe constraints in the availability of measurements of productivity
across juvenile life stages (i.e., few CUs and years of productivity across freshwater life
stages).
Due to limitations in availability of the response and explanatory variables, we would have
been constrained in our ability to define the probability distributions needed in a Bayesian
Belief Network. We were also aware that another study by the Cohen Commission (see
Marmorek et al. 2011. Fraser River sockeye salmon: data synthesis and cumulative
impacts. Cohen Commission Tech. Rep. 6) was tasked with investigating the effect of
explanatory variables across freshwater and marine conditions and would be using more
sophisticated methods than could have been applied here.
5. Other more explicit suggestions are given in section 6 below.
2. Evaluate the interpretation of the available data, and the validity of any derived
conclusions. Overall, does the report represent the best scientific interpretation
of the available data?
The report does a good job of trying to handle a vast and unwieldy dataset (information
collected from a diversity of sources and methods). I have made suggestions that might
help tighten-up some of the statistical procedures (see below) that are basically fine.
With the caveat that disease/pathogen factors in freshwater have not being addressed, I
think the overall conclusion of the authors is correct and is what I have suspected.
Although I admit to being predisposed to this conclusion, I do believe that the authors
have done a good job at testing the underlying hypotheses as best as is possible.
3. Are there additional quantitative or qualitative ways to evaluate the subject

152
area not considered in this report? How could the analysis be improved?
See comments above about IUCN Threats Calculator, Bayesian Belief Networks, etc to
help better quantify threats to individual CUs.
4. Are the recommendations provided in this report supportable? Do you have
any further recommendations to add?
Yes, the recommendations are supportable, but I find them vaguely-worded. Who
exactly are the “scientists” or “government agencies” the recommendations refer to??
Without naming species groups of scientists or agencies, the recommendations appear
too vague to be useful. Similarly, I’d like to see more specificity about the
“communication tools” recommendation. I fully believe that transparency and sharing of
data for a public resource like salmon is critical. It would be very informative for
“independent scientists” like the authors of this report to suggest a specific
model/structure of how this could happen. For instance, perhaps DFO, in the interests
of transparency should allocate some of its funds (or obtain more funds specifically for
this purpose) to set-up an arms-length assessment and monitoring “board” made up of
DFO and independent scientists that plan such programs. It would be useful for the
authors to suggest a specific model rather than just make vague suggestions for greater
transparency. What can be learned from other jurisdictions (i.e., outside BC and
Canada)?
Response: We agree that our recommendations are relatively vague, and that clear,
specific, and fully costed recommendations are always better. However, we highlight that
our scope of work was focused on an evaluation of freshwater factors in the decline of
sockeye salmon. We were not tasked with a review of existing legislation, agency capacity,
or monitoring initiatives to assess whether they are sufficient to detect cause and effect
relationships between human stressors and declines of sockeye salmon.
Having said this, we have added more details to our recommendations to include more
examples where other reports (specifically charged with addressing some of the issues
raised here) have commented on the need for improved monitoring and integration across
agencies in BC and elsewhere across the Pacific Northwest.
5. What information, if any, should be collected in the future to improve our
understanding of this subject area?
The factors all discussed in this report will be to varying extents subject to changes from
climate shifts and human demographic changes. Some modelling of the “environmental
envelope” for persistence of sockeye salmon in freshwater habitats under human and
climate change should be undertaken so that future conflicts might be anticipated.
Response: We fully agree that the factors considered in this report are influenced by
human activities and climate shifts, and that these factors can ultimately affect survival of
sockeye salmon. Thus, some modeling of the “environmental envelope” of acceptable
changes in human stressors and freshwater habitat conditions would be informative to
improve our understanding of the persistence of sockeye salmon in freshwater habitats. We

153
also agree that this information would be valuable to fisheries and habitat managers “so
that future conflicts might be anticipated.”
However, such a modeling approach is no small undertaking and was not within the scope
of what we were tasked to do. Undertaking this task would require developing a model that
explicitly accounts for the interaction among all factors affecting survival across marine
and freshwater life stages, because these factors do not interact in isolation of each other.
We believe such a modeling approach is necessary, but we were unable to do so without a
significant investment of additional resources.
This comment is consistent with our recommendation “To improve our understanding
about the population level effects of stressors on freshwater habitats…”. As a result we
have elaborated on this recommendation to include this suggestion.
6. Please provide any specific comments for the authors.
This report will be related closely to at least two others in the biological sense; those on
diseases and parasites and cumulative effects (in freshwater). I think some comments
on these factors and how they might relate to the issues in the current report would be
appropriate.
Response: We acknowledge that contaminants (MacDonald et al. 2011), diseases and
parasites (Kent 2011), habitat conditions in the lower Fraser River and Strait of Georgia
(Johannes et al. 2011), and changes to in-river conditions (Hinch and Martins 2011) might
be acting independently, cumulatively, or synergistically with the stressors considered in
this report. Consequently, we have edited the introduction to clarify the links between this
report and these other studies. We have also clarified the link to the study that has been
tasked with integrating the findings from all Cohen studies and investigating the role of
environmental conditions and stressors across both freshwater and marine life stages on
Fraser River sockeye salmon (see Marmorek et al. 2011).
Page 1, Line 3. What is the evidence that sockeye salmon are a “keystone” species??
This has a rather precise ecological definition and I am aware of no studies that
demonstrate such a status of sockeye salmon. Suggest substituting “important” for
“keystone”.
Response: Replaced “keystone” with “important”.
Page 2, Line 17. The Sakinaw AND Cultus lakes’ populations are actually assessed
under COSEWIC as “Endangered”, not “threatened” as implied here (although the
Minister of Fisheries decided not to list them as Endangered under SARA).
Response: Replaced “threatened” with “endangered”.
Page 2, Lines 24-26. It would be helpful to include the upper and lower confidence
intervals of these estimates rather than just the median. It is critical that in documents
such as these that, I assume, the public will eventually have access to that some explicit
presentation of the variability around these median estimates be given, preferably in

154
graphical form. The statement included about “…contained with the statistical
distributions..” will not be generally understandable and a graphic will make the point
more forcefully.
Response: Graphics of the statistical distributions are neither readily available nor can be
easily produced from readily available data. To address this comment as best as possible
we included statements of forecasted probabilities associated with returns in 2009 and
2010.
Page 4, line 9. “…increase in MARINE mortality…”??
Response: Added the following text: “increase in mortality during marine life stages.”
Section 2. While for the purposes of the Cohen Commission reports the justification for
36 conservation units as defined by DFO can be accepted, it should be known that the
delineation of these units has not be subject to peer review in the normal sense of the
term and will be evaluated in the near future in an independent analysis.
Response: Included a note about this point in the first paragraph of 2.1.1.
Page 8, lines 8-11. This is an important point. Distributional criteria (number of
locations, index of area of occupancy, extent of occurrence) are key variables in
assessing conservation status both under COSEWIC and IUCN criteria. The DFO-
based assessments, therefore, may well be come irrelevant after the COSEWIC
assessments that are underway.
Response: All three methods reviewed include distribution indicators so none of them
would be irrelevant after a COSEWIC assessment. Grant et al. 2010 on the other hand
does not include distribution criteria. These points are noted in the report. We added some
language around the possibility of the assessment outcomes becoming irrelevant / outdated
following a COSEWIC assessment.
Page 12, line 24-25. Agreed (re: arbitrary), but surely there is some literature to support
this statement.
Response: We included some citations that summarize the buffer widths that are being
used to protect streams elsewhere and added to our discussion of the rationale for using
this distance.
Throughout: inconsistent use of “sockeye” and “sockeye salmon” in text. The full
common name of “sockeye salmon” should be used.
Response: Ensured all references to “sockeye” use the full common name “sockeye
salmon”.
Page 17, line 29. Is there not a basic limnological concept or citation that could be used
to support this measure? Otherwise it all seems rather arbitrary and ad hoc.

155
Response: Added the following citation to support use of lake area – Randall, R.G. 2003.
Fish productivity and habitat productive capacity: definitions, indices, units of field
measurement, and a need for standardized terminology. Canadian Science Advisory
Secretariat Research Document 2003/061. Available from: http://www.dfo-
mpo.gc.ca/CSAS/Csas/DocREC/2003/RES2003_061_e.pdf
Page 19, line 6. How about urbanization d/s of Hope. Changes to the lower Fraser
Valley and estuary could impact sockeye smolts, no?
Response: Included a footnote clarifying how urbanization downstream of Hope is
captured for the Cohen Commission by Johannes et al. 2011.
Page 21, line 26 and throughout. I have no idea what the annotation in parentheses
means, e.g., “…Harrison (D/S) (L_03_03), Pitt (L_03_05), Nahatlatch (L_05_02), Fraser
and Francois (L_06_04;27 L_06_05; L_06_06; L_06_07), and Stuart (L_06_13)…”
It is very cumbersome and distracting.
Response: We have removed the use of the CU index labels and replaced them with the
Conservation Unit name and timing group throughout the text of the report. The use of the
CU index labels remains in the tables of the report.
Page 22, line 9. They are INTERcorrelated or correlated “…with each other…”
Response: Added “…with each other…” to this sentence.
Page 27, lines 17-20. This seems rather qualitative and should be replaced with some
quantitative analysis (e.g., area covered from digitized maps)
Response: Yes, understanding the spatial and temporal variation in log storage was based
on a qualitative interpretation of air photos. We pursued this approach because we were
unable to locate digitized maps of log storage from federal (DFO), provincial (MOE,
ILMB), or private (Vancouver Port Authority) agencies. Though we agree that a
quantitative analysis would have been preferred (as consistent with most other stressors we
examined) it was not possible within the scope of this project to digitize log storage using
available air photos. We also do not believe this more accurate data would have changed
our conclusions about the role of log storage in the declines of sockeye salmon.
Page 30, lines 28-29. As the Prosperity Mine has been shelved by the feds perhaps this
should be modified.
Response: Removed reference to Prosperity Mine.
Page 50, lines 25-30. Could migration distance not be related to migration time (i.e.,
time spent in the freshwater migration) or timing (i.e., is early, summer, late, fall run
timing associated with migration distance) and the real driver of the poor performance of
the father-migrating populations?

156
Response: We agree that the underlying biological mechanism is likely related to the length
of time spent in freshwater (and thus time of exposure to a stressor) or timing of migration
(i.e., earlier timing required to cover longer distances). We are using migration distance as
a surrogate for these more direct indicators, because we can measure migration distance
consistently across all CUs, which we can not do for time spent during freshwater
migration and migration timing. As well, others (Selbie et al. in Appendix C of Peterman et
al. 2010) tested for the relationship between the declining trends and timing of migration
and found no relationship, yet did find a relationship with migration distance.
We added details in the report to clarify that migration distance is related to migration
time and migration timing.
Page 50. What percentage of the variation in productivity did migration distance account
for?? It may be significant statistically, but still account for only a small amount of
variation in productivity which might ease the interpretation here.
Response: Adjusted R2 values associated with different models and explanatory variables
have been included in a new table in the report.
Table 15. What is the number of populations used in this analysis (i.e., mention in
caption to Table)?
Response: The number of populations used in the analysis has been included in the report.
Page 50. Is the Ricker model the best one to use? Are there alternatives that might be
appropriate? At the least, briefly explain what “Ricker model residuals” are and why they
are used here.
Response: We used the “Ricker model residuals” in our analyses to allow for comparisons
to the work of Peterman et al. 2010 (work commissioned by the Pacific Salmon
Commission). An additional study conducted for the Cohen Commission (see Peterman
and Dorner 2011) compared the Ricker model to the Larkin model, which confirmed that
the Ricker model was the best model for almost all of the Fraser River stocks. In Section
4.2 we have included a description of what the “Ricker model residuals” are based on how
they were calculated by Peterman et al. 2010.
Page 51, line 13. You need to define “productivity” and “total productivity” in biological
terms.
Response: At the beginning of the report we clarify that “productivity” of sockeye salmon
is referring to the number of adult recruits produced per spawner. In Section 4.2 we have
included a definition of “total productivity” in terms of how it was calculated by Peterman
et al. 2010.
Page 51, line 22. I think this is the first mention of “en route” mortality and you need to
more explicitly explain how it is accounted for in the measure of “total productivity” to
allow the reader to better understand your logic for this impt. conclusion.

157
Response: In Section 4.2 we describe how indices of “total productivity” were calculated by
Peterman et al. 2010, which includes a description of how enroute mortality is accounted
for in this calculation.
Page 51, line 29. Define “juvenile productivity”
Response: In Section 4.2 we have included a definition of “juvenile productivity” in terms
of how it was calculated by Peterman et al. 2010.
Table 16 and associated analyses. There should be some accommodation made for the
multiple testing issue here (multiple correlations tested simultaneously). Sequential
Bonferroni adjustments to the alpha level or similar false discovery rate controls need to
be implemented. This will undoubtedly lower the number of “significant” associations
here.
Response: We adjusted our analyses to account for the multiple testing issue by making a
Bonferroni adjustment to the alpha level.

158
Appendix 3 – Dashboard summaries
Biological Data
Productivity and escapement data reported by brood year for the stock representing the CU:
 Total productivity index – where available, Ricker residuals of adult recruits/effective females vs.
effective females (data from Peterman et al. 2010). The stock used to represent the CU is
indicated in the footnote.
 Total annual escapement – measured in number of individual fish, thousands of fish, or millions
of fish, depending on the scale of escapement for each CU.
Location
A map of the area of upstream influence for the CU and the CU’s location within the Fraser
watershed. Nursery lakes are indicated in black. The migration route between the mouth of the
Fraser and the most downstream point of the CU is indicated by a thicker, darker river.
Population Status
The overall status of each CU and the level of uncertainty embodied in that assessment are plotted on
this figure. The results of two independent CU status assessments are recorded (Pestal and Cass
2009; Grant et al. 2010), identifying where each agree or disagree on the status of each CU. The
present CU is identified on this figure with a bold circle or oval. This figure represents all 36 lake-
and river-type CUs, even though dashboards have only been developed for the 30 lake-type CUs.
Habitat Status
Metrics representing habitat status are presented by life-history stage: spawning, rearing, and
migration. Histograms describe the distribution of values across all CUs. The value of the present
CU is identified by a vertical dashed bar. The temperature time series are specific to the present CU.
Spawning
 Total spawning extent (km) – total linear length of all spawning areas
 Ratio of lake influence spawning to total spawning – ratio of the total extent of spawning areas
buffered by lake influence (km) to the total extent of all spawning areas (km).
Rearing
 Area of nursery lakes (1000 ha) – total area of all nursery lakes within the CU
 Nursery lake productivity (estimated) (100 smolts/ha) – the number of smolts produced in
nursery lakes scaled to the total area (of links for which smolts production data is available), as
an estimated measure of nursery lake productivity within the CU.
Migration
 Migration distance (km) – total distance of migration, measured as the distance between the
mouth of the Fraser and the most downstream entrance to the nursery lake(s) of each CU.
 Average spring air temperature at nursery lake (°C) – average spring (March – May) air
temperature at the nursery lake(s) of each CU, averaged over the period of 1901-2009.
 Spring air temperature at nursery lake (°C) – average spring (March – May) air temperature at the
nursery lake(s) of each CU, from 1901-2009.
 Average air temperature across adult migration (°C) – average air temperature during adult
migration (based on seasonal timing of CU’s run timing group), averaged across entire migration
corridor of each CU over the period of 1986-2009. Table 6 indicates how run timing groups are
aligned with monthly temperature data.
 Air temperature across adult migration (°C) – average air temperature during adult migration
(based on seasonal timing of CU’s run timing group), averaged across entire migration corridor
of each CU, from 1901-2009.

159
Integrated Summary of Habitat Vulnerability
This figure compares all CUs according to three independent measures of habitat that contribute to
habitat vulnerability (i.e., migration distance, nursery lake area, and the ratio of lake influenced
spawning extent to total spawning extent). Each circle represents a CU that is plotted according to its
total area of nursery lakes and its migration distance. The ratio of lake influenced spawning extent to
total spawning extent determines the size of the circle (the smallest being 0.0 and largest is 1.0).
Freshwater Stressors
Human Population
 Population density (persons/km2) – population density over time within the nursery lake rearing
and migration corridor or habitat areas for each CU. Note that population is plotted on a log
scale.
Land Use
 Forest harvesting (% of habitat area) – total area of Forest harvesting over a 15 year trailing
period within each habitat area, expressed as a percentage of the total area of that habitat type.
 Mountain pine beetle disturbance (accumulated % of habitat area) – the accumulated area of
mountain behind beetle disturbance within each habitat area, expressed as a percentage of the
total area of that habitat type.
 Land use type (% of habitat area) – the proportional distribution of land use types within each
habitat area. Land use is classified as urban area, agricultural area, forest harvesting disturbance,
mountain pine beetle disturbance, and other remaining land uses. If a habitat type is shown to be
100% “other”, this indicates that the specific habitat type is not applicable for the CU – this only
occurs for mainstem and tributary spawning habitats.
Resource Development
All three resource development figures are plotted on a log scale to display the large range of
variation among all CUs. Each figure is overlaid with “error” bars that represent the range between
the minimum and maximum value across all CUs for that particular metric within each habitat type.
For the graphs of small scale hydro and placer mining claims, the occasional absence of these “error”
bars for mainstem or tributary spawning habitat types indicates that the specific habitat type is not
applicable for the CU.
 Small scale hydro – the number of independent power producers located within each habitat area.
 Placer mining claims – the total number of placer mining claims located within each habitat
area.
 Other mines – the total number of mines and mining claims (other than placer mining claims)
located within each habitat area of each CU. These "other mines" have been categorized as active
mines, developed prospects, inactive mines, and major exploration projects.
Water Use
 Total allocation by use – the proportional distribution of water allocation among urban,
agricultural, and industrial uses, by habitat type. Each use, for each habitat type is labeled with
the actual allocation value (measured in m3/ha).
 Total allocation (m3/ha) – the total water allocation (across all uses) by habitat type
 Water Licenses – the proportional distribution of water licenses among urban, agricultural, and
industrial uses, by habitat type.
 Water Restrictions (#/km2) – the total number of water restrictions within each habitat type.
Road Development
 Road density (km/km2) – the density of all roads (highways, urban streets, and resource roads) within each
habitat type.

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
0 5 102.5 km
Anderson — Early Summer — L-6-1
† Representative stock for productivity: Gates ‡ Temperature calculated according to run group timing.
†† Spawning note: Gates spawning channel.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
0 6 123km
Bowron — Early Summer — L-7-1
† Representative stock for productivity: Bowron ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
010 205km
Chilko — Early Summer — L-6-2
† Representative stock for productivity: Chilko ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
010 205km
Chilko — Summer — L-6-3
† Representative stock for productivity: Chilko ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
0 2.5 51.25 km
Chilliwack — Early Summer — L-3-1
† Representative stock for productivity: None ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
0 1 20.5 km
Cultus — Late — L-3-2
† Representative stock for productivity: None ‡ Temperature calculated according to run group timing.
†† Spawning note: Foreshore spawning. ‡‡ Cultus CU cannot be represented on this figure (spawning ratio = n/a)
‡‡

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
020 4010 km
Francois — Early Summer — L-6-4
† Representative stock for productivity: Nadina ‡ Temperature calculated according to run group timing.
†† Spawning note: Nadina spawning channel.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
020 4010 km
Francois — Summer — L-6-5
† Representative stock for productivity: Stellako ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
025 5012.5 km
Fraser — Early Summer — L-6-6
† Representative stock for productivity: None ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
025 5012.5 km
Fraser — Summer — L-6-7
† Representative stock for productivity: Stellako ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
025 5012.5 km
Harrison (downstream) — Late — L-3-3
† Representative stock for productivity: Harrison ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
025 5012.5 km
Harrison (upstream) — Late — L-3-4
† Representative stock for productivity: Weaver ‡ Temperature calculated according to run group timing.
†† Spawning note: Weaver spawning channel.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
0 5 102.5 km
Indian/Kruger — Early Summer — L-7-2
† Representative stock for productivity: None ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
060 12030 km
Kamloops — Early Summer — L-10-1
† Representative stock for productivity: Raft ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
060 12030 km
Kamloops — Late — L-9-1
† Representative stock for productivity: None ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
0 0.6 1.20.3 km
Kawkawa — Late — L-5-1
† Representative stock for productivity: None ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
020 4010 km
Lillooet — Late — L-4-1
† Representative stock for productivity: Birkenhead ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
0 6 123km
McKinley — Summer — L-6-8
† Representative stock for productivity: Quesnel ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
0 4 82 km
Nadina — Early Summer — L-6-9
† Representative stock for productivity: Nadina ‡ Temperature calculated according to run group timing.
†† Spawning note: Glacier Creek spawning not mapped. ‡‡ Nadina CU cannot be represented on this figure (spawning ratio = n/a)
‡‡

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
0 8 164km
Nahatlatch — Early Summer — L-5-2
† Representative stock for productivity: None ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
010 205km
Pitt — Early Summer — L-3-5
† Representative stock for productivity: Pitt ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
0280 560
km
Quesnel — Summer — L-6-10
† Representative stock for productivity: Quesnel ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
010 205km
Seton — Late — L-6-11
† Representative stock for productivity: Portage ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
040 8020 km
Shuswap Complex — Early Summer — L-9-2
† Representative stock for productivity: Seymour ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
040 8020 km
Shuswap Complex — Late — L-9-3
† Representative stock for productivity: Late Shuswap ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
040 8020 km
Stuart — Early Stuart — L-6-12
† Representative stock for productivity: Early Stuart ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
040 8020 km
Stuart — Summer — L-6-13
† Representative stock for productivity: Late Stuart ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
030 6015 km
Takla/Trembleur — Early Stuart — L-6-14
† Representative stock for productivity: None ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
030 6015 km
Takla/Trembleur — Summer — L-6-15
† Representative stock for productivity: Late Stuart ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

Spawning††
Total spawning extent (km)
Ratio of lake influence
spawning to total spawning
Rearing
Area of nursery lakes
(1000 ha)
Nursery lake productivity
(estimated) (100 smolts/ha)
Migration
Migration distance (km)
Average spring air temper-
ature at nursery lake (°C)
Spring air
temperature at
nursery lake (°C)
Average air temperature
across adult migration (°C)‡
Air temperature
across adult
migration (°C)‡
POPULATION STATUS
High
Low
High
III
IV
I
II
Low
1901 1921 1961 1941 1981 2001
1901 1921 1961 1941 1981 2001
Ratio of lake influence spawning to total spawning
Area of nursery lakes (ha)
Migration distance (km)
BIOLOGICAL DATA †
HABITAT STATUS
LOCATION
CONSERVATION UNIT
January 31, 2011
Insufficient
Information
Uncertainty
Severity
CU Status
Pestal & Cass 2009
Grant et al. 2010
Same evaluation
Integrated Summary of Habitat Vulnerability
010 205km
Taseko — Early Summer — L-6-16
† Representative stock for productivity: None ‡ Temperature calculated according to run group timing.
†† Spawning note: None.

FRESHWATER STRESSORS
Road Development
Road
density
(km/km2)
ms t
nl mc
Abbreviations:
Urb urban
Agr agricultural
Harv harvesting disturbance
MPB mountain pine beetle
disturbance
Oth other land use
Ind industrial
Act active mines
Dvlp developed prospects
Inac inactive mines
Expl major explorations
Water Use
Total
allo-
cation
by use
Total
allocation
(m3/ha)
Water
Licenses
Water
Restrictions
(#/km2)
ms t
nl mc
ms t
nl mc
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
nl
mc
ms
t
Habitat types: Main stem spawning (ms) Tributary spawning (t) N ursery lake rearing (nl) Migration corridor (mc)
Resource Development
Small scale
hydro
Placer mining
claims
Other
mines
Human Population
Population
density
(persons/
km2)
Land Use
Forest
harvesting
(% of habitat
area)
Mountain
pine beetle
disturbance
(cumulative
% of habitat
area)
Land
use
type
(% of
habitat
area)
no data
no data

220
Appendix 4 – Data sources
Data label Data description Source Data provider Report
section
FISS Sockeye
Salmon Points
1:50K Sockeye
salmon distribution
points as recorded in
the BC Fisheries
Information Summary
System.
http://www.canbcdw.pac.dfo-
mpo.gc.ca/ows/metadata/sk_bc_pt.html
Fisheries and
Oceans Canada
2.2
FISS Sockeye
Salmon
Presence
1:50K Watershed
Atlas streams with
line features coded
for presence of
sockeye salmon.
http://www.canbcdw.pac.dfo-
mpo.gc.ca/ows/metadata/FISSSalmon_bc.htm
Fisheries and
Oceans Canada
2.2
FISS Sockeye
Salmon
Waterbody
Points
1:50K Points
representing streams
where sockeye
salmon distribution is
recorded in the BC
Fisheries Information
Summary System.
http://www.canbcdw.pac.dfo-
mpo.gc.ca/ows/metadata/fisswb_sk_blpt.html
Fisheries and
Oceans Canada
2.2
FISS Sockeye
Salmon
Waterbody
Polygons
1:50K Waterbody
polygons
representing sockeye
salmon distribution
as recorded in the
BC Fisheries
Information Summary
System.
http://www.canbcdw.pac.dfo-
mpo.gc.ca/ows/metadata/fisswb_sk.html
Fisheries and
Oceans Canada
2.2
Historical
Climate Data
ClimateWNA
generates historical
climate data for
Western North
America.
http://www.genetics.forestry.ubc.ca/cfcg/ClimateW
NA/ClimateWNA.html
UBC Forestry and
Genetics
2.2
BC Watershed
Groups (1:50K)
BC Watershed Atlas
watershed group
polygons (1:50K).
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=43753&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
2.2
Freshwater
Atlas
Assessment
Watersheds
1:20K mesoscale
aquatic units
designed to replace
the 3rd order 1:50K
watersheds.
http://apps.gov.bc.ca/pub/geometadata/metadataD
etail.do?recordUID=57079&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
2.2
Freshwater
Atlas Lakes
All lake polygons for
the province (1:20K).
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=50640&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
2.2
Freshwater
Atlas Stream
Network
1:20K flow network
arcs, directionalized
and connected.
Contains hierarchical
key and route
identifier.
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=50648&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
2.2
Freshwater
Atlas Watershed
Groups
1:20K polygons
delimiting the
watershed group
boundary.
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=50651&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
2.2

221
Data label Data description Source Data provider Report
section
Third Order and
Greater
Watersheds
BC Watershed Atlas
third order and
greater watershed
polygons (1:50K).
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=43756&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
2.2
Watershed Atlas
Lakes
1:50K Lake polygons
for the province.
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=43693&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
2.2
Watershed Atlas
Stream
Centreline
Network
Stream centerline
network (1:50K).
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=43752&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
2.2
Logging History
(RESULT –
Openings)
Administration
boundary that has
been harvested with
silviculture
obligations or natural
disturbance with
intended forest
management
activities on Crown
Land.
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=52583&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
3.1.1
BC Road
Crossings layer
Point locations of
stream crossings
within BC. Crossing
locations have been
determined based on
a GIS intersection of
the province’s
Freshwater Atlas
1:20K stream
hydrology with roads
that are delineated in
the province’s Digital
Road Atlas.
Not currently available to the public; must be
accessed through Ministry of Environment
Richard Thompson,
Ecosystems
Protection and
Assurance Branch,
BC Ministry of
Environment
3.1.1
Digital Road
Atlas – Master
Partially
Attributed
Partially attributed
road data for the
named roads from
the Digital Road
Atlas.
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=45674&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
3.1.1
Mountain Pine
Beetle
disturbance
(1992-1996)
Forest Health
Network Archives
Pest Data for British
Columbia.
http://www.pfc.cfs.nrcan.gc.ca/entomology/pests/bc
/mpb_e.html
Canadian Forest
Service
3.1.2
Mountain Pine
Beetle
disturbance
(1999-2009)
Forest Health –
Aerial Overview
Survey.
http://www.for.gov.bc.ca/hfp/health/overview/overvi
ew.htm
Ministry of Forests
and Range
3.1.2
Aerial photos of
the Fraser River
estuary
Time series (2001-
2009) of aerial
photos of the
Vancouver area
accessed using
Google Earth
software.
http://maps.google.com/maps?t=k&hl=en&ie=UT
F8&ll=49.143089,-
123.071594&spn=0.352161,0.455246&z=11
Google 3.1.3

222
Data label Data description Source Data provider Report
section
Gravel mining
activities in BC
Locations of gravel
pits in BC.
http://www.empr.gov.bc.ca/Mining/Geoscience/Surf
icialGeologyandHazards/AggregateProject/Pages/
Downloads.aspx
Vic Levson, Ministry
of Energy Mines
and Petroleum
Resources
3.2
Mineral and
Placer Claims in
BC
Shapefile with
polygons of mining
claims with valid from
and to dates.
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=49898&recordSet=ISO19115
Laurel Nash,
Ministry of Energy
Mines and
Petroleum
Resources
3.2
Mining activities
in BC
Locations, type of
activity, local
geology, production
history for exploration
and mining activities.
http://www.empr.gov.bc.ca/MINING/GEOSCIENCE
/MINFILE/Pages/default.aspx
Sarah Meredith-
Jones, Ministry of
Energy Mines and
Petroleum
Resources
3.2
Drainage Points
for Independent
Power
Producers in the
Fraser
Clean Energy
Projects with BC
Hydro Electricity
Purchase
Agreements located
on rivers that drain
into the Fraser River.
n/a David Ingleson, BC
Hydro (16 Sept
2010)
3.3.2
Census
boundaries
2006 census division
boundaries.
http://apps.gov.bc.ca/pub/geometadata/metadataD
etail.do?recordUID=56799&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
3.4
Population
Estimate by
Census Division
Population estimates
from 1986 to 2009.
http://www.bcstats.gov.bc.ca/data/pop/pop/dynamic
/PopulationStatistics/Query.asp?category=Census
&type=DR&topic=Estimates
BC Statistics 3.4
TANTALIS -
Municipalities
Representation of all
municipalities in BC,
a subset of Admin
Areas
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=50339&recordSet=ISO19115
Scott MacPhail,
LRDW/Integrated
Land Mgmt Bureau
3.4
Agricultural
Land Reserve
Polygons
Spatial
representation for
agricultural reserve
land.
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=3553&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
3.5
Points of
Diversion with
Water License
Information
Province-wide spatial
layer displaying water
license points of
diversion joined with
license information.
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=47674&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
3.6
Water Allocation
Restrictions
Province-wide layer
showing streams
having a water
allocation restriction
https://apps.gov.bc.ca/pub/geometadata/metadata
Detail.do?recordUID=34251&recordSet=ISO19115
LRDW/Integrated
Land Management
Bureau
3.6