re: hearing order oh4-2011 northern gateway pipelines inc ... · g.resume of diane whited h....

Delivered by E-file December 22, 2011 Secretary to the Joint Review Panel Enbridge Northern Gateway Project 444 Seventh Avenue S.W. Calgary, Alberta T2P 0X8 Attn: Ms. Louise George Dear Ms. George: Re: Hearing Order OH4-2011 Northern Gateway Pipelines Inc. Enbridge Northern

Gateway Project Application of 27 May 2010 – File OF-Fac-Oil N304-2010-01 – Northwest Institute Written Evidence

Further to the Joint Panel Review order re filing of evidence please find our written evidence and attachments enclosed. Yours truly, Pat Moss Executive Director Enc.

sinclairr

Typewritten Text

<personal information removed>

Northwest Institute Hearing Order OH-4-2011 File No. OF-Fac-Oil-N304-2010-01 01 File No. OF-Fac-Oil-N304-2010-0101

IN THE MATTER OF ENBRIDGE NORTHERN GATEWAY PROJECT JOINT REVIEW PANEL

WRITTEN EVIDENCE OF NORTHWEST INSTITUTE

December 2011


1.0 Introduction 1. Northwest Institute hereby submits the following documents as its written evidence in the

matter of the Enbridge Northern Gateway Project Joint Review Panel.

(a) the written evidence of David Bustard and Mike Miles;

(b) the written evidence of James Schwab;

(c) the written evidence of Jack Stanford and Diane Whited.

2. The following documents are submitted as attachments to these written submissions.

a. Resume of David Bustard

b. Resume of Mike Miles

c. “Potential Effects of an Oil Pipeline Rupture on Reach 2 of Morice River” report

d. Resume of James Schwab

e. “Hillslope and Fluvial Processes Along the Proposed Pipeline Corridor, Burns Lake to

Kitimat, West Central British Columbia” report

f. Resume of Jack Stanford

g. Resume of Diane Whited

h. “Analysis of Skeena River Tributaries Downstream from the Proposed Enbridge Pipeline” report

3. Northwest Institute proposes to present Mr. Bustard, Mr. Miles, Mr. Schwab, and Mr.

Stanford as a panel at the hearing if requested.

2.0 Written evidence of David Bustard and Mike Miles Please state your name and business address.


4. David Bustard Mike Miles

David Bustard and Associates Ltd. M. Miles and Associates Ltd. Box 2792 645 Island Road Smithers, BC V0J 1N0 Victoria, B.C. V8S 2T7

Please provide your background and work history.

5. We have included our resumes as Attachment “A” and “B” to this written submission

Have you previously testified before the National Energy Board (“NEB”)?

6. No.

Do you submit the contents of the report entitled “Potential Effects of an Oil Pipeline Rupture

on Reach 2 of Morice River” as your written evidence and was the report written by you?

7. Yes. The report is concurrently filed with this evidence and is incorporated here as

written evidence of David Bustard of David Bustard and Associates and Michael Miles of

M. Miles and Associates.

Please provide a brief summary of the report.

8. This submission focuses on a portion of the pipeline route that is located adjacent to

Morice River, 70 km south of Smithers. This 34 km long section of channel is referred to

as ‘Reach 2’. Within this area, Morice River has formed a wide floodplain that contains

numerous active secondary channels, log jams and wetlands that comprise the core

spawning and rearing habitat for Morice River fish populations. Schwab (2011) indicates

that slope instability in this area has the potential to rupture the proposed pipelines.

This report examines the implications of a pipeline rupture and subsequent clean-up

efforts to river processes, fish and fish habitat.


9. The submission relies on data presented in the Enbridge 2010 application including

impact pathways and spill volumes associated with a pipeline rupture in the Morice

Watershed. As well, it relies on a rich background of fish and geomorphology

information collected in the Morice Watershed during the past 40 years, observations

from recent oil spill events in North America, and years of personal field experience in

this watershed. On this basis, the potential consequences of a diluted bitumen spill into

Morice River have been evaluated.

10. Current fish escapements to the Morice River are strong, the watershed is productive

and habitats are intact. The Morice River supports the largest chinook salmon run in the

Skeena River comprising more than 30% of the total Skeena escapement. The Skeena

River is the second largest chinook river in BC. The Morice summer steelhead run is also

the largest in the Skeena River supporting more than 20% of the total Skeena

escapement. Morice River, particularly Reach 2, also supports a large recovering

population of coho salmon, pink salmon, and blue-listed bull trout, and is a corridor for

sockeye salmon adults and smolts moving to and from upstream spawning areas. Reach

2 in the Morice River provides critical spawning habitat and is the most productive

rearing area for millions of juvenile salmon and steelhead that are present year-round.

This river is the principal salmon spawning area within the Wet’suwet’en First Nations

territory, and these runs have been fished for at least six thousand years.

11. An oil pipeline rupture would spread hydrocarbons throughout Reach 2 and would

contaminate the log jams, side channels and shoreline areas that comprise key fish

habitats. The more volatile fractions of the oil would be immediately toxic to fish and

developing eggs located in this reach. The heavier bitumen components would slowly

release polycyclic aromatic hydrocarbons (PAHs) that would have chronic effects on

salmon egg development and juveniles rearing in these habitats for many years.

12. The volume of oil in the pipeline is sufficiently large that, even if the valves were closed

immediately at the time of rupture, a large volume of oil could drain into the

environment. Water velocities in Morice River exceed Enbridge’s criteria for using

conventional containment booms, absorbents and skimmers to collect hydrocarbons for


much of the year, and ice conditions would curtail clean-up activities during periods of

lower streamflow that occur in the winter.

13. The ability to promptly respond to a pipeline rupture would be hindered by the remoteness of the area, poor access along much of the river floodplain, and the complex network of debris and side channels in the river. The Morice River is covered in ice and snow during the winter and carries high sediment loads during spring run-off. These factors, along with the tendency for bitumen to sink and move into sediments on the river bed or banks, would make it impractical to effectively contain or recover spilled oil once it has entered the river.

14. Remedial actions that might be taken following a spill, such as collecting oil-covered debris and

sediments and removal to decontamination sites, or burning oiled debris on gravel bars, could

cause long-term habitat impacts. Observations on Pine River in north central BC indicate that

log jam removal and re-construction following an oil spill in 2000 resulted in dramatic increases

in channel instability. Log jam removal in Reach 2 of Morice River could lead to similar

mainstem channel destabilization, with a subsequent loss of critical habitats for fish.

15. It is our opinion that diluted bitumen attached to debris and accumulated in the spawning

gravels and shoreline sediments would persist and affect salmon and steelhead survival in

Morice River for an extended period. Habitat impacts could similarly persist for decades. There

do not appear to be any proven techniques for effectively mitigating these impacts.

3.0 Written Evidence of James Schwab

Please state your name and business address.

16. James Schwab Smithers, B.C.


sinclairr

Typewritten Text



17. I have included my resume as Attachment “D” to this written submission


18. No.

Do you submit the contents of the report entitled “Hillslope and Fluvial Processes Along the

Proposed Pipeline Corridor, Burns Lake to Kitimat, West Central British Columbia” as your

written evidence and was the report written by you?

19. Yes. The report is concurrently filed with this evidence and is incorporated here as

written evidence of James Schwab.


20. This paper provides an overview of the landscape, terrain, hillslope processes and fluvial

processes found within the general area of the proposed pipeline corridor across west

central B.C. The intent of this paper is to help formulate discussion, encourage more in-

depth study, direct more detailed on-the-ground investigation, and stimulate

investigation into possible safer alternative routes to the unstable terrain found in west

central B.C. This paper does not discuss environmental consequences and risk

associated with the proposed pipelines although the environmental consequences of an

oil pipeline break do differ considerably from a break sustained by a natural gas

pipeline.

21. The proposed corridor crosses three distinct physiographic units: the Nechako Plateau,

the Hazelton Mountains, and the Kitimat Ranges. These units are distinct

topographically as reflected in present day landforms, erosion, and landslides, and thus

present different hazards to a pipeline.

22. The Nechako Plateau appears relatively benign; however, large landslides have occurred

in volcanic rock overlying other older volcanic and sedimentary rock. Active bedrock


spread is occurring to the east of Parrott Creek, possibly foreshadowing further

movement along the northwest-southeast trending ridges running between Houston

and Francois Lake. Along the Morice River, advance-phase glaciolacustrine sediments

have historically experienced landslides. Road construction and wildfires have

reactivated these landslides. The proposed pipeline corridor crosses an historic earth

flow west of Owen Creek, glaciolacustrine sediment along Owen Creek, and probably

buried advance-phase glaciolacustrine sediments near Owen Creek, Fenton Creek and

Lamprey Creek.

23. The pipeline corridor follows the Crystal forest access road up Gosnell Creek. Shifting

channels on active alluvial fans pose road maintenance challenges along a 10 km section

of the road. Pipelines will likely present similar challenges crossing these fans. There is

considerable lateral bank instability at the proposed Crystal Creek and Gosnell Creek

crossing.

24. The proposed pipeline corridor dissects the floodplain (glaciofluvial and glaciolacustrine

terrain) located immediately upstream from the Clore Canyon. No development of any

sort has occurred to date upstream of the Clore Canyon so it is unclear how this terrain

will respond to pipeline development.

25. The volcanic bedrock of the Hazelton Mountains is inherently unstable as evident in

many prehistoric landslides. Three documented large landslides within the Bulkley

Range of the Hazelton Mountains have severed the natural gas pipeline since its

construction in the early 1970s; large landslides have also impacted forest roads and

highways.

26. Deep-seated gravitational slope deformation is prevalent in the volcanic bedrock found

in the Kitnayakwa, Clore and Bernie watersheds. Sackungen or slope sagging, indicative

of slope deformation, indicates active slope movement that commonly foreshadows a

pending landslide. A thorough geotechnical investigation is required to determine the

stability of the bedrock and hillslope in areas of slope deformation. Avoidance of these

unstable hillslopes is generally the preferred engineering development option.


27. The proposed corridor crosses through a mountainside to the southeast of the Clore

Canyon. The highly fractured bedrock in the canyon is undergoing active mass erosion.

This visibly Bulkley Valley Centre for Natural Resources Research & Management v

unstable rock reaches up to about 1200 m above sea level (asl) and extends around the

mountain into an adjacent tributary valley. This bedrock along the north and west side

of the mountain is extensively gullied and contains many landslide scarps and an actively

moving landslide. Along the east side of the mountain, sackungen parallel the slope and

extend through old landslide scarps. The active instability of the mountain slope places

major constraints on development.

28. Steep narrow valleys characterize the Kitimat Ranges. Colluvial-fluvial fans are at the

base of most steep gully channels in the Hoult Creek and Upper Kitimat watershed.

These steep gully channels extend from the alpine onto the valley flat or directly into

Hoult Creek or the Kitimat River. Many of these high-energy systems experienced debris

flows during extreme rainstorms in the fall of 1978 and the fall of 1992. Debris flows

commonly occur under seemingly ―normal‖ storm events during summer convective

storms and fall frontal rainstorms. Debris flows are powerful landslides that can damage

or rupture pipelines.

29. Hunter Creek, a large active alluvial fan, has historically pushed the Kitimat River across

the valley. The most recent catastrophic channel avulsion occurred in 1992. This

avulsion was caused by road construction up the fan and the construction of a levée

above the bridge crossing on Hunter Creek. Channel changes will likely recur on the fan

during major flood events.

30. The Kitimat Trough, situated between Terrace and Kitimat, is an uplifted fiord. Sensitive

glaciomarine sediments occupy much of the valley floor. Deep deposits of glaciofluvial

sediments and postglacial materials (floodplain, alluvial fans and bogs) cover the

glaciomarine sediments. These glaciomarine sediments have experienced large

prehistoric and recent landslides. A high incidence of prehistoric landslides occur around


Mink Creek, the Nalbeelah wet land complex, the foreslope of the Onion Lake Flats (fan-

delta), Cecil Creek and Deception Creek.

31. Recent large flow slides occurred at Mink Creek (winter 1992-93) and Lakelse Lake in

May and June 1962. A large submarine flow slide occurred in sensitive marine muds at

the front of the fiord-head delta at Kitimat Arm in April 1975. These recent landslides

serve to show the continuing sensitivity of the glaciomarine sediments in the Kitimat

Trough and the marine sediments on the fan-delta at the fiord-head of Kitimat Arm.

Natural and human caused factors such as increases in surface load, removal of lateral

support by stream bank undercutting or excavation, vibration by heavy equipment,

earthquake shock, high water pressures and interruption of intertidal drainage can

trigger these landslides. Thus, the potential exists for landslides to occur during pipeline

construction and in the future.

32. Glaciomarine sediments in the vicinity of Cecil Creek, Deception Creek, Wedeene River,

Little Wedeene River, along the west side of Kitimat Arm and along Chist Creek will be

encountered during pipeline construction. The pipeline corridor crosses features

indicative of prehistoric flow slides near Cecil Creek through to the little Wedeene River.

The presence of prehistoric flow slides in the glaciomarine sediments suggest a high

probability that future landslides will occur. These failures commonly start with a small

landslide from bank erosion or loading that exposes a layer of sensitive material. Then,

they rapidly retrogress with a flow of material from the displacement basin. Pipelines

crossing glaciomarine sediments must therefore avoid areas that lie within potential

flow slide depletion zones as landslides will break or disrupt pipeline service.

33. Landslides travel long distances and damage linear infrastructure such as pipelines. Six

large rock slides occurred in west central B.C. since 1978, five of these since 1999, and

four since 2002. Three of the six rock slides severed the natural gas pipeline (Howson

landslides in 1978 and 1999, and Zymoetz landslide in 2002). Damage to linear

infrastructure commonly occurs in runout zones many kilometres from the initial

landslide. This has occurred with recent landslides in west central B.C.; the longest

traveled in excess of 4 km along a slope of 9°. Therefore, the potential for damage to


pipelines extends to unstable terrain and potential landslides that start well outside the

construction corridor.

34. Long periods of increasing precipitation and temperature are associated with most

dated, large landslides across northern B.C. The climate of northern B.C. appears to have

become warmer and wetter since the beginning of instrumental observations. There is

evidence to suggest that landslide rates have increased in west central B.C. Climate

change scenarios suggest a warmer and wetter climate for west central B.C. Therefore,

the rate of landslide occurrence will likely increase and thus the likelihood of landslide

impact to a pipeline will increase.

35. Recognition and avoidance of unstable terrain is the most efficient and cost effective

method for management in landslide prone terrain. This requires detailed terrain

stability mapping and geotechnical investigation to identify unstable slopes, runout

zones, and depletion zones. Avoidance of unstable terrain is a difficult management

strategy to adopt over many sections of the proposed pipeline corridor given the

topographic constraints. Therefore, the unstable mountainous terrain across west

central B.C. is not a safe location for pipelines. Eventually a landslide will sever a

pipeline. An alternative safer route through B.C. needs investigation.

4.0 Written evidence of Jack Stanford and Diane Whited

Please state your name and business address.

36. Jack Stanford Diane Whited

sinclairr

Typewritten Text

<personal information removed> <personal information removed>



37. We have included our resumes as Attachment “F” and “G” to this written submission


38. No.

Do you submit the contents of the report entitled “Analysis of Skeena River Tributaries Downstream from the Proposed Enbridge Pipeline” as your written evidence and was the report written by you? 39. Yes. The report is concurrently filed with this evidence and is incorporated here as

written evidence of Jack Stanford and Diane Whited.


40. The report uses our existing databases to produce a series of maps describing the

geomorphology of tributaries of the Skeena River, British Columbia that will be crossed

by the proposed Enbridge pipeline. The objective was to develop and map metrics that

delineate the areas of the Skeena River and its tributaries most vulnerable to potential

pipeline spills associated with the proposed Enbridge corridor to Kitimat, BC.

41. The issue of concern is that bitumen, condensate or other petrochemicals could enter

the tributaries at or near pipeline crossings should there be a pipeline breach (Swift et

al. 2011). In the case of a spill that enters a river channel, petrochemicals will be

transported downstream and contaminate river water and aquatic habitat in relation to:

a) the slope of the channel (which determines water velocity), b) the volume of water in

the channel, c) the geomorphology of the river and its alluvial aquifers and d) the

amount and characteristics of the petrochemicals. In this analysis, we examined the

geomorphic character of the river channels downstream of crossings. Our analysis was

based on the likely scenario that a spill would result in petrochemicals being carried

downstream and distributed in the channels in relation to the geomorphology.


42. Unless completely constrained by bedrock, most rivers are alluvial, meaning that porous

bed-sediments underlie the flowing water and compose the river flood plains. High

gradient headwater streams or river segments in bedrock constrained canyons may be

expected to transport petrochemicals very quickly downstream, whereas aggraded

flood plains retard water velocity and increase contact with the bed-sediments as well

as the alluvial aquifers contained within the bed-sediments of the channel and its flood

plains (see Figure 1). If the river is flooding during a spill, petrochemicals may spread

expansively across flood plains and directly contaminate riparian vegetation. In any spill

scenario, some portion of the river channel and the alluvial aquifers influenced by influx

of surface water will be contaminated.

43. The key point here is that contamination from petrochemical spills in alluvial rivers like

the Skeena will occur in 3 spatial dimensions: upstream to downstream, laterally across

the channel and its flood plains, and vertically into the alluvial aquifer (Figure 1). The

fourth dimension is time; if the spill is of short duration and small volume, the

contamination attenuates more rapidly than if the spill is catastrophic and of long

duration. Exchange of water and materials in the water in these 4 dimensions is

precisely what makes floodplain rivers like the Skeena great salmon producers. Habitat

and productivity maximizes in the expansive floodplain reaches, which is not to say that

constrained reaches and small tributary streams are not important in overall

productivity of the river. Although a spill of any magnitude will be toxic regardless of

channel morphology for some distance downstream from the spill, the aggraded

floodplain reaches are the most vulnerable because these are the areas where the

contaminants will attenuate by entraining in the floodplain riparian vegetation, the bed

sediments and the alluvial aquifers. The extent of longitudinal, lateral and vertical

contamination depends on volume and duration of the spill and the characteristics, such

as specific gravity, water solubility, and volatility, of the petrochemicals. But it is a safe

bet that a spill of any magnitude will contaminate some portion of the river in all three

dimensions over some time period. Hence, we equated vulnerability with floodplain

area in this analysis, recognizing, of course, that the most vulnerable reaches are those

closest to the spill, regardless of geomorphology.


44. We produced physical metrics that were summarized in 4 maps: a map of the Skeena

River catchment showing the proposed pipeline corridor; maps of the Babine, Morice-

Bulkley and Clore- Zymoetz corridors. These are the river corridors most likely to be

directly affected by a pipeline breach. However, a catastrophic spill at or near any of the

crossing points could extend through the entire Skeena River system and its estuary.

The maps highlight the locations and areas of flood plains where contamination from

spills will likely be most pervasive because flood plains and their alluvial aquifers are

where river water and materials entrained in the river water is circulated most

expansively. Furthermore, these aquatic and riparian habitats determine river

biodiversity and productivity and are crucial to successful salmon spawning and rearing.

CONCLUSIONS

45. The portion of the Skeena catchment along the proposed pipeline route that was

examined in this analysis has an abundance of complex, productive floodplain habitat

that is very important for the persistence and production of resident and anadromous

fishes. Unfortunately, the characteristics of floodplain habitat that make it productive

for fisheries also make it highly susceptible to impacts from a pipeline breach.

Therefore, our conclusions are that a pipeline breach could have severely negative

impacts on resident and anadromous fishes.

46. Additionally, all the anadromous fish in the Babine, Bulkley, and Zymoetz systems have

already been impacted by 130 years of relatively high exploitation rates due to coastal

mixed stock fisheries. As well, all anadromous and freshwater resident fish have had

varying degrees of habitat modification due to development activities, including linear

perturbations such as railroad, highways, transmission and pipeline corridors,

agriculture, urbanization, forestry, and mining, with particular impacts to the productive

floodplain habitats. An oil spill or rupture from the proposed pipeline would have

significant environmental, effects within and beyond the Skeena River system.

ANALYSIS OF SKEENA RIVER TRIBUTARIES DOWNSTREAM FROM THE PROPOSED ENBRIDGE PIPELINE

December 20, 2011

By Jack A. Stanford1, and Diane C. Whited1 1Flathead Lake Biological Station, The University of Montana, Polson, Montana 59860 USA

INTRODUCTION

We were contracted by the Northwest Institute for Bioregional Research (Smithers, BC) to use our existing databases to produce a series of maps describing the geomorphology of tributaries of the Skeena River, British Columbia that will be crossed by the proposed Enbridge pipeline. The objective was to develop and map metrics that delineate the areas of the Skeena River and its tributaries most vulnerable to potential pipeline spills associated with the proposed Enbridge corridor to Kitimat, BC.

The issue of concern is that bitumen, condensate or other petrochemicals could enter the tributaries at or near pipeline crossings should there be a pipeline breach (Swift et al. 2011). In the case of a spill that enters a river channel, petrochemicals will be transported downstream and contaminate river water and aquatic habitat in relation to: a) the slope of the channel (which determines water velocity), b) the volume of water in the channel, c) the geomorphology of the river and its alluvial aquifers and d) the amount and characteristics of the petrochemicals. In this analysis, we examined the geomorphic character of the river channels downstream of crossings. Our analysis was based on the likely scenario that a spill would result in petrochemicals being carried downstream and distributed in the channels in relation to the geomorphology.

Unless completely constrained by bedrock, most rivers are alluvial, meaning that porous bed-sediments underlie the flowing water and compose the river flood plains. High gradient headwater streams or river segments in bedrock constrained canyons may be expected to transport petrochemicals very quickly downstream, whereas aggraded flood plains retard water velocity and increase contact with the bed-sediments as well as the alluvial aquifers contained within the bed-sediments of the channel and its flood plains (see Figure 1). If the river is flooding during a spill, petrochemicals may spread expansively across flood plains and directly contaminate riparian vegetation. In any spill scenario, some portion of the river channel and the alluvial aquifers influenced by influx of surface water will be contaminated.

The key point here is that contamination from petrochemical spills in alluvial rivers like the Skeena will occur in 3 spatial dimensions: upstream to downstream, laterally across the channel and its flood plains, and vertically into the alluvial aquifer (Figure 1). The fourth dimension is time; if the spill is of short duration and small volume, the contamination attenuates more rapidly than if the spill is catastrophic and of long duration. Exchange of water and materials in the water in these 4 dimensions is precisely what makes floodplain rivers like the Skeena great salmon producers. Habitat and productivity maximizes in the expansive floodplain reaches, which is not to say that constrained reaches and small tributary streams are not

sinclairr

Typewritten Text


Analysis of Skeena River Tributaries

2

important in overall productivity of the river. Although a spill of any magnitude will be toxic regardless of channel morphology for some distance downstream from the spill, the aggraded floodplain reaches are the most vulnerable because these are the areas where the contaminants will attenuate by entraining in the floodplain riparian vegetation, the bed sediments and the alluvial aquifers. The extent of longitudinal, lateral and vertical contamination depends on volume and duration of the spill and the characteristics, such as specific gravity, water solubility, and volatility, of the petrochemicals. But it is a safe bet that a spill of any magnitude will contaminate some portion of the river in all three dimensions over some time period. Hence, we equated vulnerability with floodplain area in this analysis, recognizing, of course, that the most vulnerable reaches are those closest to the spill, regardless of geomorphology.

Headwaters

Headwater Canyon

Piedmont

Valley

Flood Plain

Piedmont Canyon

Montane Canyon

Montane

Flood Plain

Coastal

Flood Plain

Delta-Estuary

a b

Figure 1. Idealized view of (a) the longitudinal distribution of flood plains and canyons (“beads on a string”) within a river ecosystem from headwaters to the ocean and (b) the 3-D structure of alluvial the flood plains (beads), emphasizing dynamic longitudinal, lateral and vertical dimensions and recruitment of wood debris. The groups of arrows in (a) indicate the expected strength of ground- and surfacewater exchange (vertical), channel and flood plain (lateral) interactions and upstream to downstream or longitudinal (horizontal) connectivity in the context of (b). The floodplain landscape contains a suite of structures produced by the legacy of cut and fill alluviation as influenced by position within the natural-cultural setting of the catchment. The hyporheic zone is defined by penetration of river water into the alluvium and may mix with phreatic ground water from hillslope or other aquifers not directly recharged by the river. Alluvial aquifers usually have complex bed sediments with interstitial zones of preferential groundwater flow (paleochannels). From Stanford et al. (2005).

We produced physical metrics that were summarized in 4 maps: a map of the Skeena River catchment showing the proposed pipeline corridor; maps of the Babine, Morice-Bulkley and Clore- Zymoetz corridors. These are the river corridors most likely to be directly affected by


3

a pipeline breach. However, a catastrophic spill at or near any of the crossing points could extend through the entire Skeena River system and its estuary. The maps highlight the locations and areas of flood plains where contamination from spills will likely be most pervasive because flood plains and their alluvial aquifers are where river water and materials entrained in the river water is circulated most expansively. Furthermore, these aquatic and riparian habitats determine river biodiversity and productivity and are crucial to successful salmon spawning and rearing.

METHODS

Data sources

The majority of the data used in the analysis were compiled by the Flathead Lake Biological Station within the Riverscape Analysis Project (RAP). RAP data and analysis can be viewed at http://rap.ntsg.umt.edu. See Luck et al. (2010) for a detailed description of the database.

We used supplemental data from GeoBase Canada (http://www.geobase.ca/geobase/ en/index.html) and Johanna Pfalz at Eclipse GIS to provide detailed base layers for map generation. In addition we downloaded and compiled a 25-m resolution Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) DEM from the National Aeronautics and Space Administration (NASA) Jet Propulsion Laboratory (http://asterweb.jpl.nasa.gov /content/03_data/01_Data_Products/release_DEM_relative.htm) to improve our floodplain delineation within the Skeena catchment. Floodplain delineation

Flood plains were identified from the Aster 25 m digital elevation model (DEM) using a modified Arc/INFO and Arc Macro Language (AML)/C code program developed by Scott Basset at the University of Nevada, Reno. DEM-derived stream order and elevation information was used to identify flood plains and estimate floodplain areal extent based on lateral distances and maximum elevation thresholds perpendicular to and along the DEM-derived river flow path. For each stream order, buffer distances and maximum elevation thresholds were established to define the corresponding floodplain spatial extent. Buffer distances and maximum elevation thresholds (Table 1) were increased for larger stream order categories to account for larger floodplain areas consistent with larger rivers.

Stream elevation profiles

We resampled the ASTER 25 m DEM to a coarser 100 m DEM to minimize the elevation variability inherent in the ASTER DEM. We generated stream elevation profiles from this resampled DEM to represent the potential flow paths of petrochemicals and highlighted stream reaches with flood plains and lakes as zones of retention as described above. Although the resampling smoothed out some of the variability within the DEM, some elevation variability remained within the profiles; this variability is expressed as “roughness” of the profiles. However, the steep versus aggraded (i.e., flat) reaches of the stream profiles were clearly evident.

Table 1. Buffers around stream channels were used to set the maximum extent of floodplains during extraction from DEMs.


4

Stream Order Elevation (m) Buffer Distance (m) 1 1 300 2 1 600 3 1.5 1200 4 2 1500 5 2 1750 6 3 2000

>6 4 2500

RESULTS AND DISCUSSION

The four maps that we produced are appended to this report, along with a summary table comparing the areas of flood plains in the primary tributaries. Maps should be printed in large format and in color to be most useful.

The map of the entire Skeena catchment provides perspective on the distribution of floodplain reaches throughout the system. In general, the headwaters of the system have substantial areas that are low gradient wetlands and small flood plains. The river is very constrained in canyons above and below the Babine confluence, except for lower reaches of the Sustut. The mainstem is mainly confined in canyon reaches until below the Zymoetz confluence near Terrace, BC. From there to the expansive estuary, the mainstem river has broad flood plains with complex channel networks.

The Sutherland-Babine, Morice-Bulkley and Clore-Zymoetz corridors were mapped in detail because the pipeline crosses them in the mountainous headwater areas and therefore the points of direct spills into the system would be distributed from these rivers and through their associated flood plains (Table 2). Note that the pipeline is proposed very close to the Gosnell Creek/Morice River for over 60 km. Note also that the Gosnell Creek/Morice River is an expansive floodplain river throughout most of its course and therefore is very vulnerable to spills that could flow into the river anywhere along this 60 km reach. A spill on the Sutherland River could contaminate its floodplain reaches and deliver petrochemicals into Babine Lake. A spill along the tributaries to Helene and Taltapin Lake would greatly impact these lakes, as well as Pinkut Creek (a large sockeye spawning tributary) that drains directly into Babine Lake. A spill on the high gradient Clore system could rapidly move into the flood plains of the Zymoetz and on into the expansive flood plains of the mainstem Skeena.


5

Table 2. Comparison of floodplain area and normalized floodplain area per river km for the three potentially affected river reaches.

Potentially affected River Reaches Floodplain Area (Ha)

Normalized Floodplain area (Ha) per river km

Babine Corridor Sutherland River to Babine Lake 1589.08 34.73 Above Helene Lake to Babine Lake 263.32 13.19 Above Taltapin Lake to Babine Lake 514.29 12.48 Below Babine Lake to Confluence with Skeena 1561.25 16.10 Bulkley Corridor Buck Creek Above Confluence 483.56 20.49 Bulkley River Below Morice River Confluence 8244.98 50.35 Maxan Creek/Bulkley River Above Morice River Confluence

4101.30 46.06

Gosnell Creek/Morice River Above Bulkley River Confluence

6783.94 65.66

Clore/Zymoetz Corridor Clore River Above Zymoetz River Confluence 416.14 9.52 Zymoetz River Below Clore River Confluence 343.73 9.46

Compared with adjacent river systems, the Skeena has a fairly large ratio of flood plains to total catchment area (Figure 2). The Nass and Fraser rivers are very constrained by narrow canyons throughout the catchments. The Stikine has very expansive flood plains in its lower reach; and, even though it also has dramatic canyons, the ratio is quite high (Figure 2). Nonetheless, all four river systems have very high physical complexity values (Table 3). The intent of this comparison is simply to underscore the point that the Skeena’s abundant and expansive flood plains operate with the on-channel lakes, notably Babine Lake, to provide a complex array of habitats that drive system biodiversity and productivity, especially for salmon, steelhead, and resident fishes and their food webs.

Table 3. Summary of RAP metrics that describe physical complexity of the Skeena River in comparison to adjacent river systems. The term, nodes, refers to channel junctions or places where channels separate or converge.

Number of Main Channel

Nodes Nodes per FP

river km Nodes per river km

Number of Tributary

Nodes

Tributary Nodes per river km

Skeena 855 1.283 0.096 640 0.072 Fraser 1283 1.275 0.031 2791 0.067 Nass 502 1.251 0.146 262 0.076 Stikine 2224 1.791 0.258 663 0.077


6

Petrochemical spills are problematic for any river system but the impact may be expected to intensify if the rivers are broadly three-dimensional, as is the Skeena. The tributaries crossed by the proposed pipeline are especially vulnerable in this regard. All are well known to function as primary spawning and rearing areas for Skeena fisheries.

Indeed, within the Babine catchment, five stream reaches and lake bodies are vulnerable and would be affected by a pipeline spill or rupture. The Babine drainage is the largest tributary to Skeena River and supports diverse stocks of chinook, pink, sockeye, coho, and steelhead salmon. The freshwater fish community is composed of rainbow and cutthroat trout, Dolly Varden, bull trout, and lake char, kokanee, lake and mountain whitefish, lamprey, burbot, sculpins, suckers, and shiners (Gottesfeld and Rabnett 2008). A potential spill from Kilometer Post (KP) 851 to 869 would flow into the Sutherland River and from KP 881 to 912.5 into the Pinkut Creek system. The Sutherland River supports sockeye salmon and kokanee spawning, as well as spawning, rearing, and holding habitat for coho, steelhead, and rainbow trout. Sutherland River rainbow trout are a singular race of large late-maturing trout, contributing 66% of the rainbow trout found in Babine Lake (Bustard 1990). Pinkut Creek system supports sockeye, coho, and pink salmon spawning, as well as burbot, kokanee, lake trout, lake whitefish, and rainbow trout spawning and rearing. Pinkut Creek is the second most productive salmon tributary to the Babine basin, and over the last 20 years, along with the adjacent spawning channel, has seen an annual average return of 211,165 sockeye spawners (Gottesfeld and Rabnett 2008). Fish values in the Bulkley basin are rated very high due to anadromous coho, sockeye, pink, and chinook salmon, as well as steelhead and Pacific lamprey. Dolly Varden, rainbow trout, and mountain whitefish are present in most fish bearing waters, and bull trout, lake trout, burbot, and a coarse fish community utilize various habitat types in the Bulkley upstream of Morice as well as the Morice system (Gottesfeld and Rabnett 2008). The proposed pipeline crosses through 119 km of the Bulkley drainage, which includes 34 km paralleling the gravel bed Reach 2 of Morice River, which is of high fisheries value (Bustard and Schell 2002). The pipeline is proposed to cross 12 km within the Zymoetz (Copper) River upstream of the Clore Canyon. This generally high elevation area supports bull trout, Dolly Varden, and rainbow trout (Bustard 1996). Downstream of Clore Canyon, the Copper supports coho, chinook, sockeye, pink, chum, steelhead, and a suite of freshwater residents (Gottesfeld and Rabnett 2008).

CONCLUSIONS The portion of the Skeena catchment along the proposed pipeline route that was examined in this analysis has an abundance of complex, productive floodplain habitat that is very important for the persistence and production of resident and anadromous fishes. Unfortunately, the characteristics of floodplain habitat that make it productive for fisheries also makes it highly susceptible to impacts from a pipeline breach. Therefore, our conclusions are that a pipeline breach could have severely negative impacts on resident and anadromous fishes.

Additionally, all the anadromous fish in the Babine, Bulkley, and Zymoetz systems have already been impacted by 130 years of relatively high exploitation rates due to coastal mixed-


7

stock fisheries. As well, all anadromous and freshwater resident fish have had varying degrees of habitat modification due to development activities, including linear perturbations such as railroad, highways, transmission and pipeline corridors, agriculture, urbanization, forestry, and mining, with particular impacts to the productive floodplain habitats. An oil spill or rupture from the proposed pipeline would have significant environmental effects within and beyond the Skeena River system.

LITERATURE CITED

Bustard, D. 1990. Sutherland River rainbow trout radio telemetry studies. 1989. Prepared by Dave Bustard and Associates for British Columbia Ministry of Environment, Smithers, BC.

Bustard, D. 1996. Fisheries assessment of the lower Clore River and tributaries – preliminary report. Prepared for Skeena cellulose Inc. Terrace, BC.

Bustard, D. and C. Schell. 2002. Conserving Morice Watershed fish populations and their habitat. Prepared for CFDC Nadina.

Gottesfeld, A.S. and K.A. Rabnett. 2008. Skeena River fish and their habitat. Skeena Fisheries Commission. Hazelton, BC.

Luck, M., N. Maumenee, D. Whited, J. Lucotch, S. Chilcote, M. Lorang, D. Goodman, K. McDonald, J. Kimball, and J. Stanford. 2010. Remote sensing analysis of physical complexity of North Pacific Rim rivers to assist wild salmon conservation. Earth Surface Processes and Landforms 35:1330–1343.

Stanford, J. A., M. S. Lorang, and F. R. Hauer. 2005. The shifting habitat mosaic of river ecosystems. Verh. Internat. Verein. Limnol. 29:123–136.

Swift, A., N. Lemphers, S. Casey-Lefkowitz, K. Terhune, and D. Droitsch. 2011. Pipeline and Tanker Trouble. The Impact to British Columbia’s Communities, Rivers, and Pacific Coastline from Tar Sands Oil Transport. A Joint Report of the Natural Resources Defense Council, The Pembina Institute, and The Living Oceans Society, New York, 29 pp.

re: hearing order oh4-2011 northern gateway pipelines inc ... · g.resume of diane whited h....

Documents