Attachment JRP IR 4.1
AMEC Environment & Infrastructure A division of AMEC Americas Limited 3456 Opie Crescent, Prince George, BC Canada V2N 2P9 Tel +1 (250) 564-3243 Fax +1 (250) 562-7045 www.amec.com
LANDSLIDE-GENERATED WAVE HAZARD ANALYSIS KITIMAT ARM ENBRIDGE NORTHERN GATEWAY PROJECT
Submitted to:
Northern Gateway Pipelines Inc., Calgary, Alberta
Submitted by:
AMEC Environment & Infrastructure A division of AMEC Americas Limited Prince George, BC
EG0926008
20 September 2011
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
IMPORTANT NOTICE
This report was prepared exclusively for Northern Gateway Pipelines Inc., a subsidiary of Enbridge Pipelines Inc. (Enbridge) by AMEC Environment & Infrastructure1 Limited, a wholly owned subsidiary of AMEC. The quality of information, conclusions and estimates contained herein is consistent with the level of effort involved in AMEC services and based on: i) information available at the time of preparation, ii) data supplied by outside sources, and iii) the assumptions conditions and qualifications set forth in this report. This report is intended to be used by Northern Gateway Pipelines Inc., Enbridge only, subject to the terms and conditions of the contract between AMEC and Enbridge. Any other use of, or reliance on, this report by any third party, is at that party’s sole risk.
1 As of September 2011, AMEC Earth & Environmental changed its name to AMEC Environment & Infrastructure
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
EXECUTIVE SUMMARY
This report presents the results of a landslide-generated wave hazard assessment in support of a regulatory application for the Northern Gateway Project. The work was carried out by AMEC Environment & Infrastructure (AMEC) on behalf of Northern Gateway Pipelines Inc. The Northern Gateway Project (the Project) includes a planned crude oil and condensate terminal to be constructed on the west side of Kitimat Arm, about 6 km south of the Rio Tinto Alcan aluminum facility at Kitimat, BC.
The study involved an examination of the phenomenon and nature of the hazard related to landslide-generated waves and is presented as a qualitative analysis, consistent with the evaluation of other geohazards associated with the proposed project. The study area included the marine area from the north end of Kitimat Arm south to beyond Coste Island. The analysis has been supported by desktop studies, literature reviews and numerical modelling as well as a limited fieldwork program that included geophysical surveying in Kitimat Arm. Recently completed multibeam echo sounding (MBES) surveys conducted in 2009 and 2010 by the Canadian Hydrographic Service (CHS) provided relatively high resolution bathymetry throughout the study area.
The Kitimat Arm area had two reported landslide-generated waves in 1974 and 1975. A significant aspect of the study was an evaluation of these past events, including triggering mechanisms, in support of the evaluation of the future hazards to the Project. In addition to the experiences in Kitimat Arm, additional cases from other areas with similar hazards, such as Norway, Alaska and other locations in British Columbia were reviewed This review has shown that very large landslides moving rapidly in the marine environment are required to generate waves that are significant beyond typical marine surface conditions.
The results of the review of the past events in Kitimat Arm indicate that anthropogenic change (typically related to marine construction) could be a key contributing factor with respect to future landslide-generated waves if appropriate planning and engineering is not undertaken. The geomorphic record visible using the recent CHS MBES surveys shows that the two recent events are the only occurrences of large-scale, long-runout landslides in the marine environment in Kitimat Arm. The geomorphic record is interpreted to represent a period of thousands of years. This supports the conclusions of the work to date that suggest that the potential for a future naturally occurring submarine landslide capable of generating a wave is very low.
The proposed Kitimat Terminal and associated marine berth are located on the western side of Kitimat Arm along a steep rocky coastline both in terrestrial and subaqueous portions of the site. Steep, straight rocky coastlines are favourable locations as they minimize potential amplification of waves that typically occurs as a result of coastal shoaling or where they become trapped in enclosed bays. The lack of significant sediment accumulation at the locations of the proposed berth sites is also favourable from the perspective of avoiding construction activities in marine sediments. Together with additional work to further refine the hazard characterization, the study concludes that suitable planning and cooperation among all existing facility owners and potential Project sponsors for all foreshore and off-shore development in the Kitimat Arm should be adopted to mitigate the potential submarine landslide hazard on the proposed project facilities.
AMEC Project No. EG0926008 Executive Summary Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
TABLE OF CONTENTS PAGE #
1.0 INTRODUCTION ...... 1 1.1 GENERAL...... 1 1.2 NATURE OF THE HAZARD...... 1 1.3 SCOPE OF ANALYSIS ...... 1
2.0 DEFINITION OF A LANDSLIDE-GENERATED WAVE ...... 2
3.0 LITERATURE REVIEW AND SUMMARY OF EVENTS IN KITIMAT ARM ...... 2 3.1 LITERATURE REVIEW...... 2 3.2 SHORELINE CHANGES OBSERVED IN THE AIR PHOTO RECORD ...... 4 3.3 SUMMARY OF INFRASTRUCTURE DEVELOPMENT IN KITIMAT ARM ...... 4 3.4 REGIONAL GEOLOGY...... 6
4.0 HISTORIC LANDSLIDE-GENERATED WAVE EVENTS ...... 7 4.1 GENERAL TIMELINE ...... 7 4.2 ENVIRONMENTAL DATA ASSOCIATED WITH THE EVENTS ...... 8 4.2.1 Kitimat River Flow ...... 8 4.2.2 Precipitation ...... 9 4.2.3 Tides ...... 9 4.3 17 OCTOBER 1974 LANDSLIDE-GENERATED WAVES ...... 10 4.3.1 27 April 1975 Landslide-generated wave event ...... 11
5.0 CHS SUBMARINE MAPPING, PROJECT SURVEYS, AND SITE SPECIFIC DATA ..... 12 5.1 CANADIAN HYDROGRAPHIC SERVICE Data ...... 12 5.2 KITIMAT ARM GEOPHYSICAL SURVEY ...... 12 5.3 KITIMAT ARM POLYGON ATLAS ...... 13
6.0 KITIMAT ARM BASIN LANDSLIDE INVENTORY ...... 13
7.0 REQUIRED FACTORS FOR LANDSLIDES TO GENERATE A LARGE WAVE ...... 14 7.1 MARINE CLAY AND MARINE CLAY LEACHING ...... 15 7.2 REGIONAL GROUNDWATER PATTERNS ...... 16 7.3 Liquefaction ...... 16
8.0 TRIGGERING MECHANISMS ANALYSIS FOR THE 1974 AND 1975 EVENTS ...... 18 8.1 General ...... 18 8.2 Tides ...... 18 8.3 Geometric changes ...... 18 8.4 ANTHROPOGENIC changes ...... 19 8.5 EVENTS RELATED TO THE 17 OCTOBER 1974 SLIDE ...... 19 8.6 RE-VISITING THE EVENTS RELATED TO THE 27 APRIL 1975 SLIDE ...... 20
9.0 PRESENTATION OF NUMERICAL MODELING OF THE 1975 EVENT ...... 21
A M E C Project No. EG0926008 Table of Contents Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
10.0 SUMMARY OF FINDINGS ...... 22
11.0 QUALITATIVE HAZARD ANAL YSIS SUMMARY ...... 23
12.0 PATH FORWARD ...... 23
13.0 CLOSURE ...... 25
14.0 REFERENCE LIST ...... 26
LIST OF FIGURES
Figure 1 General Site Plan Figure 2 Air Photo Review, Northland Dock to Moon Bay Figure 3 Air Photo Review, Northland Dock to Moon Bay Figure 4 Air Photo Review, Kitimat Harbour Between Eurocan and Alcan Figure 5 Air Photo Review, MK Bay Marina to Kitimaat Village Figure 6 Chronological Summary of Water Line Position 1947 to Present Figure 7 Morrison (1984) Plan and Geological Section Figure 8 Timeline of Major Construction Activities and Tsunami Reports (included in text) Figure 9 1974 Initiation Zone Figure 10 Combined Tidal and River Flow Analysis Figure 11 1975 Initiation Zone
LIST OF APPENDICES
Appendix A Bornhold, B, 2011, Coastal and Ocean Resources Inc., “Submarine Failures and Associated Tsunamis, Norway - Literature Review”, Report to AMEC, 37 pp.
Appendix B Bornhold, B, 2010, Coastal and Ocean Resources Inc., “Initial Results, Kitimat Arm Geophysical Survey”, Report to AMEC, 18 pp.
Bornhold, B, Coastal and Ocean Resources Inc., Letter to AMEC, “Kitimat Harbour Geophysical Investigation Report” 30 August 2011, 1 page.
Appendix C Terra Remote Sensing Inc. Report to AMEC and Northern Gateway Pipelines, 10 November 2010, “Survey Operations Report”, 47 pp.
Terra Remote Sensing Inc., Maps 1 to 5 showing “Contoured Bathymetry and Survey Track; and, Isopach of Total Unconsolidated Sediment and Seabed Slope Magnitude”. 5 pp.
Appendix D Kitimat Arm Polygon Atlas, 67 pp.
Appendix E AMEC internal report, 2011, “Preliminary Assessment of Earthquake-Related Liquefaction and Associated Submarine Slope Instability Hazard”, 20 pp.
Appendix F Fine, I., and Bornhold, B., 2011. “Submarine Landslide-Tsunami Modeling, April 27, 1975”. Report to AMEC. 11 pp.
A M E C Project No. EG0926008 Table of Contents Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
1.0 INTRODUCTION
1.1 GENERAL
AMEC Environment & Infrastructure (AMEC), a division of AMEC Americas Limited, was retained by Northern Gateway Pipelines Inc. (Northern Gateway) to provide geotechnical engineering services in support of a hazard assessment related to the potential for landslide- generated waves in Kitimat Arm. The hazard could potentially affect the planned crude oil and condensate terminal to be constructed on the west side of Kitimat Arm, about 6 km south of the Rio Tinto Alcan aluminum facility at Kitimat, BC. The general location of the proposed terminal site and study area is shown on Figure 1.
The purpose of this report is to present the results of a hazard assessment related to landslide- generated waves in the Kitimat Arm. The study is consistent with the evaluation of other geohazards and is presented in qualitative terms. The potential zone of influence related to this hazard includes the marine areas surrounding the marine terminal as shown on Figure 1.
Note that this study focussed on submarine landslide events that might result in the production of landslide-generated waves. Landslide-generated waves have also been referenced in the scientific and engineering literature as landslide-generated tsunamis, or just tsunamis. In addition, a landslide that results in the generation of a wave field can also be termed a tsunamigenic-landslide.
A tsunami is a series of waves created as a result of an impulsive disturbance in the water column and could be the result of crustal displacement as a result of fault rupture, landslides, meteorite impacts, or volcanic eruptions. For the purposes of this report, the authors will make reference to landslide-generated waves, although for the purposes of referencing other work, the interchangeable landslide-generated tsunami and tsunamigenic-landslide terminology may also be used.
1.2 NATURE OF THE HAZARD
The hazards posed by landslide-generated waves could include:
• Direct wave impact on coastal infrastructure; • Impacts from debris entrained in the waves; • Scour around structures related to water movements during foreshore inundation and drain-back; • Impacts related to wave induced movements of a ship at berth; • Direct wave impact on a ship manoeuvring in Kitimat Arm or Douglas Channel.
1.3 SCOPE OF ANALYSIS
The scope of the review carried out to date with respect to the potential for submarine landslides and resultant landslide-generated waves in Kitimat Arm has included the following:
• Definition of a landslide-generated wave (Section 2.0); • Literature review and summary of available published information from various sources (Section 3.0); • Review of the documented landslide-generate wave events in Kitimat Arm (Section 4.0)
AMEC Project No. EG0926008 Page 1
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
• Review, collection and zonation of site-specific submarine data for Kitimat Arm including MBES and geophysical data (Section 5.0); • Preparation of a submarine landslide inventory in Kitimat Arm using the bathymetric data to catalogue the occurrence, distribution and character of submarine landslides in the study area including those documented in the available literature (Section 6.0); • Review of preconditioning factors required to generate submarine landslides (Section 7.0) • Review of potential submarine landslide triggering mechanisms and possible contributing factors related to past events(Section 8.0); • Back-analysis of the 27 April 1975 event including construction of a coupled dynamic submarine landslide and hydrodynamic model to examine the potential wave behaviour in areas not directly observed or discussed in the literature (Section 9.0);
• Summary of findings and basis for qualitative assessment (Section 10.0); • Assignment of a qualitative hazard category to the polygons within the study area with respect to the potential for a landslide-generated wave to originate from a landslide within the polygon (Section 11.0); • Review of future recommended work planned in the study area (Section 12.0).
2.0 DEFINITION OF A LANDSLIDE-GENERATED WAVE
Landslide-generated waves (wave train) are a series of characteristically long period and/or higher than normal waves resulting from an impulsive disturbance (specifically a landslide) in the marine environment that displaces water. The landslide can originate in the terrestrial environment and/or the marine environment, although to generate a wave, the displaced landslide mass must travel in the marine environment.
While all landslides into or within the marine environment cause displacement of water and thus a response in the water column, a key distinction is drawn in this report between displacements from all submarine landslide movements and the subset of those that are both sufficiently large and rapid to generate waves that are significant with respect to typical marine conditions. This is supported by the work of Clague (2001) who notes that a large, rapidly moving landslide is required to generate significant waves.
3.0 LITERATURE REVIEW AND SUMMARY OF EVENTS IN KITIMAT ARM
This section presents a summary of literature reviewed by the Project team to support this study. The summary includes the history of industrial and infrastructure development activities in the Kitimat Arm, river flows and tides, and reported accounts of landslide generated waves in Kitimat Arm that occurred on 17 October 1974, and 27 April 1975. The review also included research into occurrences elsewhere including similar geographic and geological environments in Norway.
3.1 LITERATURE REVIEW
The Project team carried out a review of the available literature with respect to submarine landsliding and landslide-generated waves focussing primarily on failures that resulted in wave generation. The literature reviewed fell into several general categories including historical accounts of developments in Kitimat Arm, occurrences of landslide-generated waves in Kitimat
AMEC Project No. EG0926008 Page 2
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
Arm and in other regions, the phenomenon and occurrence of landslide-generated waves, and landslide-generated wave modelling.
Hardy and Ripley (1954), Pugh (1954), and Morrison (1984) provide background on aspects of industrial development in the Kitimat Arm area. Morrison (1984) also references landslide- generated wave occurrences in the region.
A significant body of literature is available with respect to the 27 April 1975 landslide and associated landslide-generated wave. Initial research carried out on behalf of the Province of British Columbia was documented in a report by Campbell and Skermer (1975). Several later researchers reviewed various aspects of the 1975 event. Work by Murty (1979), discussed the correlation of landslide volume with wave height based on the 1975 event. Following the work of Murty (1979), researchers such as Jiang and LeBlond (1992) provided additional insight with respect to modelling wave heights that form the basis of more recent predictive tools. Several publications from 1982 through 1986 by Prior et al (1982), Bornhold (1983), Prior et al (1984), and Johns et al (1986) provided reviews of the submarine landslide morphology in Kitimat Arm, predicted sedimentation rates, characteristics of the landslide morphology, and detailed discussions, including geotechnical aspects of the submarine slope failure. Morrison (1984) provided accounts of the 1975 event, and results of geotechnical investigations that pre-dated the 1975 event. Additional papers by Swan (1978) and Lutenauuer and Swan (1978) provided interpretation of the 1975 event. Accounts reported in the local newspaper, the Kitimat Sentinel were also reviewed during the work, including the October 24, 1974 (Vol 21, No. 43), May 1, 1975 (Vol 22, No. 18), and the May 15, 1975 (Vol 22, No. 20) editions.
Jiang and LeBlond (1992 and 1994) discussed numerical methods that couple submarine landslide movement with the production of landslide generated waves. Rabinovich et al (2003) provided a numerical modelling example predicting wave train characteristics from hypothetical landslides in Georgia Strait. Skvortsov (2005) completed a back analysis of the wave from the April 1975 failure in Kitimat Arm.
Northern Gateway commissioned a literature review of the events related to landslide-generated waves in Norway as the geological conditions in that region are similar to those in the Kitimat area. The results of the review are outlined in Bornhold (2011) and included in Appendix A. In addition to the Norway experiences, the Project team reviewed publications relating to landslide- generated waves in Skagway, Alaska (Kulikov and Rabinovich, 1996), Nice, Italy, (Dan et al, 2007) case histories of coastal BC accounts of landslide-generated waves, and accounts of other such waves in publications such as Clague (2001) and Bornhold et al (2007).
As noted by Bornhold (2011), in reviewing submarine failures and associated landslide- generated waves related to occurrences documented in Norway, “[u]nlike subaerial slope failures the precise shapes and volumes of subaqueous failures have been, until recently, poorly known.” While this is taken from work relating to literature reviews of the Norwegian experiences, it reflects conditions in many other areas including submarine landslides along the BC coast. In most instances in Norway, slope failures and resultant waves can be related to anthropogenic activities, mainly coastal infrastructure development. Both in Norway and Alaska, there have also been cases where one tsunamigenic failure has led to seafloor erosion, triggering a second failure and tsunami with devastating consequences.
In support of the literature review, AMEC examined historical airphotos available for the region to examine the changes in the environment since the establishment of significant foreshore and offshore infrastructure in the region.
AMEC Project No. EG0926008 Page 3
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
3.2 SHORELINE CHANGES OBSERVED IN THE AIR PHOTO RECORD
Government air photos spanning a period between 1947 through 2003 were reviewed to examine changes near the shoreline in the region. The flight-line and photo number references are provided together with the images on the figures in this report. The review is provided as a graphical compilation of imagery from select regions in Kitimat Arm as described below. The air photo review was limited by the timing of available imagery, coverage area and scale of the photographs, and was suitable to review large-scale changes with respect to infrastructure development footprints on the previously undeveloped delta area as well as the inter-tidal zone.
Air photo compilations are provided on Figures 2 and 3 for the area between Northland Dock and Moon Bay, Figure 4 for the area between Eurocan and Alcan, and Figure 5 for the area between MK Bay Marina and Kitamaat Village. A compilation of the interpreted shoreline changes is provided on Figure 6. Changes in the western fjord wall or delta front visible in the air photo record include:
• Fill placement and dredging in the delta area and adjacent tidal flats between the present Eurocan and Alcan dock industrial facilities area as shown on Figure 4. • Marina construction, including a small breakwater, occurred at MK Bay Marina as shown on Figure 5. • Marina construction at Kitamaat Village 400 m south of the original dock structure as shown on Figure 5. • Changes to the shore profile at Moon Bay where a portion of the beach was lost in 1975 as shown on Figure 2. • Changes to river channels behind the delta front including significant lateral erosion and changes to the lower reaches of Moore and Anderson Creeks that previously ran across the tidal flats and now run through the Alcan facility as shown on Figure 6.
It is noted that small differences between shorelines visible in the air photo record, and as shown on Figure 6 are interpreted to be due to air photo distortion and/or tidal fluctuation. The air photo review was used additionally to develop a timeline for industrial development in the region as is detailed in the following section.
3.3 SUMMARY OF INFRASTRUCTURE DEVELOPMENT IN KITIMAT ARM
The Municipality of Kitimat was incorporated in 1953 to support construction and operation of an aluminum smelting facility for the Aluminum Company of Canada Ltd. (Alcan) located at the northwest corner of Kitimat Arm. Subsequent to the initial developments, additional industrial facilities have been constructed in the region, requiring infrastructure development on the foreshore and offshore portions of the north end of Kitimat Arm. The following provides a summary of the history of major development in the region to provide the reader with a timeline and general nature of the developments. The timeline is taken from available published literature and/or the available air photo record.
The initial construction of the Alcan smelter and related infrastructure including a marine terminal occurred between 1953 and approximately 1959; additional development continued sporadically beyond 1959. Development of the Alcan site included dredging and placement of large volumes of fill including hydraulic fill, granular fill from dredged materials and fill from on- land sources. The placement of hydraulic and granular fills appears to have been mostly complete by August 1954 based on air photo review. Construction of the marine terminal AMEC Project No. EG0926008 Page 4
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011 included, placement of large rock-filled concrete cribs along the delta front and dredging of delta material.
The Northland Navigation dock located between the Alcan site and Moon Bay was built in tandem with the Alcan site, and was in place by August 1954 based on air photo review. The dock configuration changed from its original footprint following the major submarine landslide in April 1975. The dock was completely removed sometime between 1978 and 1985. Several smaller docks were also located near this facility and appear to remain today. Note that the Northland Dock was the furthest to the south in the group of docks and was also the largest.
Logging and construction activities related to development of the Kitimat town site were ongoing during the initial Alcan construction.
Dock and marina facilities in Minette Bay were constructed after 1947 but prior to August 1963 based on air photo review. Minette Bay has a shallow water entrance as it is located north of the mouth of the Kitimat River. It is unknown whether dredging has occurred in this channel.
Most of the construction of the Eurocan pulp and paper mill was carried out, from about 1968 to 1971, including construction of dock facilities at the head of Kitimat Arm (east of Alcan). Construction at the Eurocan dock facilities included fill placement and dredging along the delta front as well as driving many wooden piles to support the main wharf structure.
Rivtow Straits’ planned tug dock development in Moon Bay is reported to have begun in 1974, but was not visible at an intermediate construction stage in the air photo record. The 1974 air photos show no road development to the area and no activity in the bay. The 1975 air photos were taken in June following the April 1975 submarine landslide that resulted in a loss of the partially constructed facility. The 1975 air photos show the road development to the area and an altered beach profile that was a result of the April 1975 landslide.
The site of MK Bay Marina appears to have been used for log boom storage prior to 1975 as evidenced by a lack of shore developments and the presence of log booms in the bay on early photos. Based on local newspaper reports, it appears that regulatory approval and financing for the marina was in place in about 1975 and it was constructed shortly afterwards. The 1978 air photo shows development of a small breakwater and other minor grading changes at the MK Bay marina site in the intertidal zone. The breakwater parallels the north side of Wathl Creek.
Two docks were observed at Kitamaat Village in the air photo review. The first dock is visible in a 1947 oblique photo series and existed until at least 1978. Sometime between 1978 and 1988, the Kitamaat Village fishing fleet moved to a new marina constructed in a cove located about 400 m to the south of the original site and the original dock was removed.
Further industrial development of the Kitimat Harbour waterfront continued into the early 1980s, with the construction of the Ocelot (subsequently Methanex) plant and dock facility in 1981. The Methanex wharf is located between Eurocan and Alcan wharves at the head of Kitimat Arm. The wharf facility was constructed on granular fill obtained by dredging west of the jetty. Piled dolphins were constructed adjacent to the wharf.
AMEC Project No. EG0926008 Page 5
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
3.4 REGIONAL GEOLOGY
Together with the literature review, history of development in the region, and the air photo review, the study included a review of the regional geology to understand the possible extent and contributing conditions relating to the potential for submarine landsliding in the region.
Kitimat Arm is a fjord, defined as a long narrow arm of the sea, in a valley that is U-shaped and steep walled, generally several hundred metres deep, with high rocky cliffs or slopes along a mountainous coast. The Kitimat River flows into the north end of the arm where a large delta has been deposited. Regional bedrock consists of Coast Plutonic quartz diorite igneous rocks of Late Cretaceous to Middle Jurassic age (100 million to 150 million years old). Quartz diorite is typically a very strong rock.
A regional surficial geology study was carried out by Clague (1984) covering an area extending north from Kitimat Arm. Clague stated that the terrain in the region has been subject to up to about 200 m of uplift relative to the current sea level due to unloading following deglaciation. The significance of the uplift is that the present Kitimat River valley and adjacent terrain was previously submerged to an elevation of about 200 m which resulted in marine deposition much farther to the north following deglaciation and before rebound of the area to current levels. The massive granular deposit (geologically termed the Onion Lake Delta) that extends for many kilometres in the Onion Lake area between Lakelse Lake and the northern extent of the Kitimat River in the main valley is an ancient delta deposited by glaciofluvial flows into the submerged inlet. The relative sea level drop in the Douglas Channel – Kitimat Arm fjord following deglaciation has resulted in a geologically rapid southward advance of the river delta. The present Kitimat River delta has been deposited within approximately the last 9,000 years.
The deltaic sedimentation process is a function of many factors such as river flows, river sediment loads, tidal currents, and distance from the fluvial-tidal interface. In general, a delta is comprised of layered granular materials such as gravel, sand and silt that are deposited across a wide area during lateral migration of the river. Marine muds (a term generally used to describe cohesive sediments, marine clay, or clayey silt) are often not deposited in the delta itself, instead they are suspended in the water column and gradually deposited through a relatively quiescent water column farther offshore. In the fjord environment, tidal and river currents can be quite strong and marine mud deposition may extend significant distances down the channel.
The combination of the deltaic sedimentation process and the shifting of the delta deposition location toward the south over time have been important factors in the development of the sediment profile in Kitimat Arm. During deglaciation, the delta was located much farther to the north and the current delta and Kitimat Arm study area was a zone of mud accumulation. As the delta shoreline moved southward, granular deltaic sediments were deposited over previous accumulations of marine mud. As the relative uplift of the terrain occurred, some of the previously deposited marine sediments were exposed, resulting in marine mud deposits extending up the walls of the fjord to approximately 200 m elevation.
Morrison (1984) had access to the results of drilling investigations carried out in the northwestern portion of Kitimat Arm and presented a geological interpretation based on those data. The geological sections from Morrison (1984) are presented on Figure 7. Section EE, perpendicular to the axis of the fjord through the Alcan site, shows a steeply-dipping bedrock profile draped with a layer of silt and clay several tens of metres thick, overlain by generally silty fine-grained sand, in turn overlain by a near horizontal surface layer of sand and gravel less
AMEC Project No. EG0926008 Page 6
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011 than about 20 m thick. This profile is consistent with the expected sedimentary sequence described above through the post-glacial period.
4.0 HISTORIC LANDSLIDE-GENERATED WAVE EVENTS
4.1 GENERAL TIMELINE
Historic reports of landslides and related tsunami waves in Kitimat Arm have been summarized by several authors as noted in Section 3.1. The available references document two specific tsunami events, occurring on 17 October 1974 and 27 April 1975. The 1975 event is directly linked to a submarine landslide in the literature, while the 1974 event is linked to observed debris paths revealed in recently acquired MBES data (Canadian Hydrographic Service, 2009) and inferred from eye-witness accounts of a tsunami at the time. The total entrained volumes of the submarine slides and volumes of the initiating events are difficult to establish due to the resolution and extent of pre-failure bathymetric surveys as well as the mixed nature of the runout zones on the fjord basin. It is likely that the individual initial slides were each well over 1 million m3. Estimates in the literature suggest that total entrained volumes of both 1974 and 1975 events are on the order of 10 to 25 million m3. Runout lengths along the floor of Kitimat arm are on the order of 5 km.
Several papers note that unusual water movements and changes in booming grounds were reported between the 1950’s and 1970’s, although no specific data are available. A short reference is made in Morrison (1984) to a 1971 investigation by Klohn Leonoff and C.F. Ripley of a shear failure in soft silt-clay soils related to embankment loading between the boat launch ramp and the Northland dock. This may be the same event referenced by Campbell and Skermer (1975) where they summarized a slide in marine clay at the Nechako Dock. The slide was reported to have dislocated piles that had been driven to support the dock and walkway. No specific reference to a tsunami was reported.
Morrison (1984), reported two waves in October 1974, one on the 17th and one on the 24th. This is different from all the other reports, including newspaper articles following the April 1975 event, which refer to a single wave event in October 1974. It is possible that confusion may have arisen over the fact that the wave occurred on 17 October, although it was not reported by local news media until the release of the next weekly edition on 24 October. Most accounts of the 17 October 1974 event indicate that it occurred at 11:05 PM, shortly after low tide, damaged and sank boats at Kitamaat Village, and caused damage to the Rivtow Straits facilities.
Figure 8, below, provides a graphical timeline showing the relationship between the major construction activities and landslide/tsunami events in Kitimat Arm. Based on available information, there have been no tsunami events since the 1975 event. In addition to the review with respect to general industrial activities, the Project team also examined environmental factors such as river flow, precipitation, and tidal stages at the times of the events.
AMEC Project No. EG0926008 Page 7
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
Figure 8: Timeline of Major Construction Activities and Tsunami Reports
Municipality of Kitimat Incorporated Period of Construction of Kitimat Flood Protection Works Eurocan Rivtow Straits Moon Bay Alcan W h ar f Construction Wharf/Breakwater Ocelot Wharf Construction Construction Construction
1953 1960 1970 1980 1990 2000 2009 7.6 m tsunami Several landslides in the on Apr 27/75 near-shore are noted in various reports but no 6.1 m tsunami; tsunamis were reported Oct 17/74
4.2 ENVIRONMENTAL DATA ASSOCIATED WITH THE EVENTS
Environmental factors are often important contributing factors with respect to landslide occurrence, and therefore several factors were examined with respect to the 1974 and 1975 landslide-generated waves. The available data from the time of the submarine failures includes river flows from the Kitimat River, precipitation records from Kitimat, and tidal predictions for the period. No tide gauge data is available for Kitimat Arm for the period in question.
The daily average flow data for the Kitimat River were taken from Environment Canada’s Archived Hydrometric Database2. The data used were collected between 1964 and 2008 from the Kitimat River monitoring station below Hirsch Creek located at 54°3'34" N and 128°40'29" W. The overall drainage area upstream of this monitoring station was 1990 km2. Rainfall data were taken from Environment Canada’s National Climate Data and Information Archive3. The data were collected from the monitoring station “Kitimat 2”, located at 54°00'35" N and 128°42'18" W.
4.2.1 Kitimat River Flow
The daily average flow of the Kitimat River record is available from 1968 to 2008 for a total of 15,616 data points. The highest average daily flow recorded during this period was 2820 m3/s on 29 September 1992. The following summary of the records is provided:
• The highest 1 percent of average daily flow records (highest 157 records) represent average daily flows up to 2820 m3/s. • By season, the highest 1 percent of the records show the relative distribution of occurrence (presented here through the calendar seasons) was: o 12 percent Spring; o 14 percent Summer;
2 Water Survey of Canada / Relevés Hydrologique Du Canada. Web. June 2010.
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
o 64 percent Fall; and, o 8 percent W inter. • 17 October 1974 was reported to have had an average daily flow of 351 m3/s, this follows an extended period of high flows with flows of 1650 m3s recorded 2 days prior. • In the week preceding 17 October 1974, the 7-day moving daily average flow rate for the Kitimat River was up to 625 m3/s; this was in the top 0.2 percent of flow records for a 7- day period. • 27 April 1975 was reported to have an average flow of 92 m3/s; this was in the lower 50 percent of the daily flow records. • In the week preceding the 1975 tsunami, the 7-day moving average flow for the Kitimat River was reported as 95 m3/s, also in the lower 50 percent of flows for a 7-day period • The reported 100-year return period flow for the Kitimat River is 2820 m3/s. • The projected 200-year return period flow for the Kitimat River is 3170 m 3/s. • Based on the recorded annual peak flow data, the 1:100 year maximum daily flow event has occurred once (in 1992) over the available period of record. This was preceded by a flood event in 1991 that had a return period slightly greater than the 1:50 year event.
Newspaper articles reported significant flooding in 1966, which greatly impacted the municipality of Kitimat as a result of overbank flows from the Kitimat River that blocked roads and pushed debris into developed areas in the delta. Construction of flood protection works was carried out after the 1966 flood, and was reported as completed sometime before October 1974, and therefore provided protection to the town during the flood in mid-October 1974. The specific period of construction within that timeframe is unknown. Flood protection was by means of dykes that prevented over-bank flow in vulnerable areas. The flood protection would have resulted in a reduction of floodplain storage through the lower parts of the delta thus resulting in more concentrated and higher flow downstream of areas that previously provided overbank flow and flood volume storage. In addition to the flood protection measures, development on the delta area would also have contributed to a reduction in floodplain storage.
4.2.2 Precipitation
River flows are strongly dependent upon precipitation in the contributing drainage area. Peak flow events can occur in response to extreme rainfall events, spring snowmelt or a combination. The magnitude of spring snowmelt events is a function of accumulated winter snowpack, rate of snowmelt, and whether rainfall occurs over the snowmelt period. A detailed review of the precipitation was not carried out, with exception of the precipitation records for the month prior to the April 1975 event, as detailed in Section 8.6 of this report.
4.2.3 Tides
All historic tidal information was taken from the public website “Tide and Current Predictor”4, as Environment Canada’s archives did not include complete historic tidal data for the Kitimat Arm region. The Canadian Hydrographic Service was also contacted but only had data for 1977 and 1978. The values generated by the public website were compared with the few historical measurements in 1977 and 1978 and as these two sources matched, the public website information was deemed reliable. The tidal predictions are provided at 53.9833° N, 128.7000° W (near Moon Bay), between the years 1970 and 2010.
4 Historical Tide Information taken from http://tbone.biol.sc.edu/tide/tideshow.cgi?site=Kitimat%2C+British+Columbia+%282%29 AMEC Project No. EG0926008 Page 9
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
The 1970 through 2010 tidal data set had a total of 59,288 data points. For the area under discussion, the typical daily pattern has approximately two high and two low tides. By definition, half of the records (29,644) represented low tides. The predicted maximum tide was 6.53 m and the predicted maximum low was -0.07 m, resulting in a total range of 6.6 m over the period of the records, neglecting any secondary effects such as wind-driven water elevation differences. Based on the records the following basic summary is provided:
• The lowest 1 percent of the low tides (296 records) had values between -0.07 m and 0.13 m. • The lowest 1 percent of the tides occurred more than 300 times in the 40-year record, and occurred in each year in the record. • By season (calendar seasons represented), the lowest 1 percent of the records had the following distribution: o 51 percent Spring; o 22 percent Summer; o 22 percent Fall; and, o 11 percent W inter.
• The 1974 tsunami was reported to have occurred shortly after a low tide of 0.43 m; representing a value in the lowest 5.3 percent of the low tide records. • The 1975 tsunami was reported to have occurred shortly after a low tide of 0.15 m; representing a value in the lowest 1.4 percent of the low tide records.
4.3 17 OCTOBER 1974 LANDSLIDE-GENERATED WAVE S
The following points summarize the available accounts of the 17 October 1974 event that is presumed to have been the result of landslide-generated waves.
15 October 1974 – Peak and daily flows of the Kitimat River on October 15, 1974 were the highest recorded (1650 m3/s) for the previous ten years back to the start of records in 1964. In the week preceding 17 October 1974, the 7-day moving daily average flow for the Kitimat River was up to 625 m3/s, which is in the top 0.2 percent of 7-day records. The 1974 peak flow has subsequently been exceeded three times.
17 October 1974 – This date was reported to have had an average daily flow of 351 m3/s, following an extended period of high flows including a value of 1650 m3/s recorded 2 days prior. Shortly after a low tide of 0.43 m (representing a value in the lowest 5.3 percent of the low tide records) at about 11:15PM, a tsunami wave impacted the Kitamaat Village dock. The wave was reported to be up to 20 feet (6.1 m) high, as measured from low tide to the high water marks on piles where the wharf rose up. The initial occurrence at Kitamaat Village, as recounted in the newspaper by a man in his fishing boat who experienced the wave, was reported to be a wave trough. Several boats were damaged. Damage at Rivtow Straits was reported to have included office damage related to a broken dolphin or piling that then caved in the roof. Waves reported by Rivtow employees (assumed to be on boats in the northwest corner of Kitimat Arm, likely at Moon Bay), were between 2.4 and 4.6 m high. One article reported that “scientists in Victoria” suggested the wave was the result of a landslide. Information on the 1974 event was taken mainly from newspaper articles as noted above in Section 3.1.
AMEC Project No. EG0926008 Page 10
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
4.3.1 27 April 1975 Landslide-generated wave event
The following points summarize the available accounts from the event on 27 April 1975 that is reported to have been the result of a landslide-generated wave.
Summer 1974 – Minor grading was completed for a road to a planned breakwater at Moon Bay (no major earthworks were reported). In the fall of 1974, Rivtow began construction of a timber crib wharf for a tug tie-up facility including dredging in Moon Bay to provide suitable water depth and to obtain material for construction of a breakwater in the bay. Campbell and Skermer (1975) reported that construction included earthworks that extended about 600 feet (183 m) along the beach.
January 1975 – A timber crib wharf was completed by Rivtow Straits at the north end of Moon Bay, with a face 6 to 7.6 m (20-25 feet) high (Campbell and Skermer 1975). The authors did not include reference to any other structural dimensions such as the length of the crib, although it is reported by Campbell and Skermer (1975) that the construction was completed by filling out from the beach to provide a crib-contained loading platform. The start date of crib construction was not provided.
April 1975 – Rivtow Straits was engaged in the construction of a rockfill and dredged sand breakwater on Alcan water lots in Moon Bay “within and slightly seaward of the tidal zone” (Campbell & Skermer 1975). It is reported from news articles that these activities had begun in the fall of 1974.
27 April 1975 – The top of the breakwater (groin) remained below the high tide level. Employees of Rivtow were on a barge carrying out maintenance on a clam shell dredge north of breakwater.
• 10:05AM – 53 minutes after a low tide of 0.15 m (representing a value in the lowest 1.4 percent of the historic low tide records), the Rivtow employees noticed the breakwater moving seaward and their barge moving toward shore. A slump started at the breakwater and within 2 minutes propagated around the shoreline along the road to the wharf which was subsequently carried away (Campbell & Skermer 1975). • 10:15AM – Murty (1979) reported an eyewitness account from a Canadian Hydrographic Service employee as saying the bottom was visible. It is unclear what, or even where, this observation referred to, although it is suggested that as a CHS employee he was likely referring to an area that would otherwise not have been visible even at low tide, so likely this was a tsunami wave trough observation. • Following the slide: o Slumping was followed by a wave that struck Northland Navigation’s dock, the Eurocan dock and Kitamaat Village. This appears to have occurred almost immediately. A portion of the Northland dock was destroyed during the event, as reported by an eyewitness in the Northern Sentinel newspaper on 1 May 1975. o A witness launching a boat (assumed to be in the area of the Northland dock) noticed the harbour was dry and the water was far away from the boat ramp. Looking toward the Eurocan dock, the water appeared to be plunging into the earth. He observed a shoreward moving wave or swell of water beyond the Northland dock that appeared to be as high as the dock, about 8 m above the low tide level. He drove his car and boat trailer up the ramp, which was then “barely covered by the swell of water” (Morrison 1984). The wave receded and was followed by several smaller oscillations. AMEC Project No. EG0926008 Page 11
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
o Witnesses at the Eurocan dock noticed the wave, which crested at about 8 m and did not flood the terminal deck (Morrison 1984). o The oscillations from the initial wave lasted over 1 hour and the high water mark was observed at 8.2 m on piles across the inlet near Kitamaat Village – reported by Campbell & Skermer (1975) as 25 feet (7.6 m). As pointed out by Skvortsov (2005), it was unclear if these measurements were reported as relative to low tide elevation on the day of the event. o Murty (1979) quotes an unpublished reference from R.E. Brown (identified in Campbell and Skermer 1975 as an employee of the CHS and in a 1975 Northern Sentinel newspaper as a Federal Tidal Surveyor) that states that the water level disturbance lasted about 1 hour, and that at least one wave was observed in Clio Bay and at Bish Creek. The wave in Clio Bay (no documented location of measurement) was estimated by Brown at 6.7 m. Some damage (no further specific reference to the nature of damage) was reported near Bish Creek as well.
28 April 1975 – Two further slumps were reported near the Rivtow Straits wharf area in Moon Bay at the north end of the initial failure but no waves were reported.
5.0 CHS SUBMARINE MAPPING, PROJECT SURVEYS, AND SITE SPECIFIC DATA
The following sections provide data and graphical compilations of recent submarine surveys carried out in Kitimat Arm, and mapping of the combined datasets.
5.1 CANADIAN HYDROGRAPHIC SERVICE DATA
In 2009 and 2010 the Canadian Hydrographic Service (CHS) carried out deep-water and near- shore multibeam bathymetric surveys that included the Kitimat Arm study area. The survey data were obtained by the Project team and made available for the review of the landslide-generated waves reported in the literature. The data provided complete multibeam submarine bathymetry data below depths between 10 to 30 m.
The data are presented in combination with the Kitimat Arm geophysical data as described in Section 5.2 below.
5.2 KITIMAT ARM GEOPHYSICAL SURVEY
Following a review of the preliminary CHS multibeam bathymetric data, Northern Gateway commissioned a submarine geophysical survey to examine sub-bottom conditions in select areas. The survey was completed in September 2010 and was carried out as a joint effort between Terra Remote Sensing Inc., and Coastal and Ocean Resources Inc., both of Sydney. BC. The surveys were carried out using a narrow beam echosounder and towed seismic profiler and hydrophone array. The surveys provided estimates of thickness and interpreted substrate composition in the targeted areas. The survey and data interpretation report is available in Appendix B. Note the proposed berth sites are planned to be constructed in water depths of up to about 30 m, and are within Polygon 12 as defined in the attached report and in Section 5.3, below. Polygon 12 is described as having very steep slopes with no significant thickness of sediment. Polygon 13 is adjacent to and below Polygon 12 at the proposed berth sites and does include significant thicknesses of sediment, although thicknesses beyond 10 m typically lie at depths below 70 m to 90 m.
AMEC Project No. EG0926008 Page 12
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
Compilation maps of the CHS multibeam bathymetry together with interpreted sub-bottom sediment thicknesses are also provided in Appendix C.
5.3 KITIMAT ARM POLYGON ATLAS
For the purposes of studying the overall Kitimat Arm basin, the study area was divided into polygons. Within each polygon, conditions such as interpreted geology, morphology, and specific assigned attributes such as slope angle were summarized. The polygons are shown in the Kitimat Arm Polygon Atlas, presented in Appendix D.
The two large submarine landslides associated with the 1974 and 1975 waves are distinct features visible in the mapping. Of particular note is the relatively long distance (five kilometers) down Kitimat Arm with landslide debris deposits on slopes of less than 1 degree. Total slope heights in the failure zones are on the order of 100 to 150 m. This long-run out length is characteristic of a rapidly moving slide. A detailed description of the morphology is presented in Prior et al (1982).
6.0 KITIMAT ARM BASIN LANDSLIDE INVENTORY
The polygon atlas presented in Section 5.3 provides a description of the various attributes of each polygon, including the interpretation of whether landslide activity has occurred in the past. The interpretation was based on the 2009/2010 CHS dataset which used as the baseline for submarine mapping. The results of the review were consistent with the previously reviewed literature that indicated significant large-scale landslides have occurred in the northeast and northwest corners of Kitimat Arm interpreted as the 1974 and 1975 slides respectively. With the exception of a few small slides elsewhere in the study area, other submarine landslides are not visible.
A key aspect of the analysis of the landslide-generated wave hazard in Kitimat Arm was an analysis of the geomorphic record to identify presence and assign relative age to slides from which a recurrence interval could be estimated. The geomorphic record from the available mapping on the fjord sidewalls is interpreted to represent a record of major events that occurred over a period extending to at least the past several thousand years before present. If large scale landsliding occurred over this time period, evidence of the event would be expected to be preserved in the bottom morphology. In particular, if there were multiple slide occurrences throughout this extended period of time, this might indicate natural causes of instability, for example, long-duration high-intensity storms, significant earthquake events, or extreme water level fluctuations. A pattern of widespread failures was not visible; instead only two major failure areas adjacent to the fjord-head Kitimat River delta representing the 1974 and 1975 events are visible.
As indicated above, the fjord walls have been divided into polygons having approximately similar geomorphic conditions. The polygons are shown in Appendix D. A summary of the interpretation of key polygons follows:
• The combined area of Polygons 15, 16, and 17 represents the initiation and transport areas of the large debris flow landslide that occurred on 27 April 1975. A portion of the area in Polygon 2 represents the long runout deposition zone for this landslide. The long runout length suggests high velocities in the failure mass. It is noted that source zones for the 1975 event were from the fjord sidewall and from the Kitimat delta slope. The
AMEC Project No. EG0926008 Page 13
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
interpretation of the morphology of the slide deposit, using the recent CHS surveys, suggests that the sidewall failure occurred first, followed by failure of the delta slope. • The combined area of Polygons 19 and 20 represents the initiation and transport areas of the large debris flow landslide that occurred on 17 October 1974. A portion of the area in Polygon 2 represents the long runout deposition zone for this landslide. The long runout length for this slide also suggests high velocities in the failure mass; this failure mass was partly overrun by the failure mass from the April 75 event. • Polygon 14 includes three small to medium sized landslide scars. The failure scars are subdued with respect to the 1975 and 1974 areas discussed above, and are thus interpreted to be significantly older than the larger slides of the 1970s. A portion of Polygon 2 includes the deposition zone for these slides. It is noted that these relatively small slides appear to have significant deposition upslope of where the larger debris from the long runout slides is present. It is interpreted that these smaller Polygon 14 slides did not have a long run out distance and thus likely had much lower initial velocities than the larger 1970’s landslides. • Polygon 22 includes relatively small features that are interpreted to be landslide scars. The scars are contained within the polygon area and are thus not classified as a long run out failures. The initial slide mass velocities are interpreted to be relatively low.
In addition to the analysis of the submarine mapping the project team has also carried reviews of the terrestrial environment surrounding Kitimat Arm with respect to the potential for large- scale landslides to enter the marine environment. The reviews were carried out using helicopter field reconnaissance, aerial photo review and review of LiDAR data in select areas. While select areas of relatively small rockfall and shallow sliding are present, no areas of large scale instability were identified through this process and thus the landslide inventory is restricted to the submarine study area events noted above.
7.0 REQUIRED FACTORS FOR LANDSLIDES TO GENERATE A LARGE WAVE
The previous sections have described the past occurrence and environmental conditions associated with landslide-generated waves in Kitimat Arm, as well as the history of development in the region, and regional geology. To evaluate areas beyond the limits of the past events, a set of characteristic factors has been established to provide a baseline for assigning other areas as potential source zones for submarine landsliding that could result in landslide-generated waves. These factors are defined as pre-conditioning factors and are distinguished from triggering mechanisms that are discussed in Section 8.0.
The basic requirements for a slide to generate a wave, as noted in Section 2.0 are both large- volume and rapid failure. These factors are consistent with the observations of the two past events, both in the scale and velocity, provided that the long run-out nature of the slide is taken to indicate high velocity in the absence of any direct measurements. From a geotechnical perspective, rapid failure requires either a large strength loss or low strength along the failure surface and/or major geometric changes such as undercutting, thus allowing substantial acceleration of the slide mass. The capacity for large scale failure and rapid slide movement is both a function of the availability of material to form the slide mass as well as the ability for movement of a large volume to be triggered and to accelerate quickly as a coherent mass. The following presents a discussion of how the geological conditions in the area such as the presence of high pore pressures, weak layers in the sedimentary sequence and time-dependent material properties could lead to the potential for large rapidly moving submarine landslides to occur. The discussions herein focus on modification of the existing interpreted deposits of clays
AMEC Project No. EG0926008 Page 14
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011 along the fjord walls and granular deltaic sediments present at the head of the arm and the tributary channels.
7.1 MARINE CLAY AND MARINE CLAY LEACHING
Marine clays, or as also termed, marine muds, are present in thick blankets on the steep fjord sidewalls and likely underlie granular deltaic sediment accumulations in Kitimat Arm. These marine sediments, together with a suitable trigger as noted in Section 8.0, are considered a potential source area for large scale landsliding due to the occurrence of thick, laterally extensive deposits on the submarine slopes and the potential for significant and rapid strength loss.
Regional geological mapping (Clague, 1984), published literature (Geertsema and Torrance, 2005), and Northern Gateway studies (Kitimat Terminal geotechnical report) have all discussed glaciomarine clay deposits in the Kitsumkalum Trough that exhibit sensitive behaviour. As noted Section 0, the terminology for describing the nature of the cohesive sediment accumulations that are present throughout the marine basin, including deposits on the fjord walls and in the uplifted terrestrial environment includes interchangeable reference to marine clays or marine muds. Note that the distinction is related to the alternative definitions provided by either the geological grain size distribution, or the engineering behaviour-based descriptions of the material. Mud, or the combination of both silts and clays, references the fact that the marine sediments in the region are generally defined as 30% to 50% clay and 50% to 70% silt by weight. Consistency between the results obtained in terrestrial sediment sampling at the proposed Kitimat Terminal site and those reported by Geertsema and Torrance (2005), and Johns et al (1986) is observed.
With respect to behaviour-based classification of the cohesive marine sediments, Atterberg Limits (ASTM D43185) tests on glaciomarine clays from the plant area6 generally typically indicate low to intermediate plasticity clay, while the marine sampling reported in Johns et al (1984) suggests the marine muds behave as low to high plasticity silt, generally with higher liquid limits and plasticity index. Mineralogy of the marine sediments, particularly clay mineralogy reported by Johns et al (1986) is consistent with Geertsema and Torrance (2005), so it is possible that the differences in Atterberg Limits may be associated with porewater chemistry differences. Geotechnical engineering descriptions of fine grained materials are typically presented as behaviour-based, and thus the common use of marine clays and/or marine silts is used in the engineering context of the cohesive marine sedimentation that is predominantly silt with respect to grain size classification. The terminology of cohesive marine sediments, marine mud and marine clay, refers to similar materials in this report.
Sensitivity, as applied to clays implies, strength loss ratios (peak to residual strength) of up to four, whereas quick clays may have strength losses of up to about 30. During investigations that included both drilling and review of slide characteristics, no quick clays have been found south of the Onion Lake Delta. An occurrence of quick clay is documented in Geertsema et al (2006) at Mink Creek north of the Onion Lake Delta and other occurrences have been documented in this general area. Some of the Norwegian experience has also documented the formation of quick clays in marine fjords as a result of freshwater leaching glaciomarine deposits.
5 ASTM D4318-10, Standard Test Methods for Liquid Limit, Plastic Limit, and Plasticity Index of Soils. ASTM International. 6 Enbridge Northern Gateway Pipelines Section 52 Application, Volume 3 “Engineering, Construction and Operations”, Appendix E-3. AMEC Project No. EG0926008 Page 15
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
Sensitivity of marine clay deposits is a condition where the post-failure strength of the material is reduced over time a result of leaching and exchanging predominantly monovalent ions from the originally saline porewater with divalent species such as calcium and magnesium as freshwater flows through the deposit. This is typically a condition that may occur for uplifted marine muds in the terrestrial environment, although it can occur in the marine environment in the presence of artesian conditions that move fresh water through submarine clays.
Evaluation of the peak to post-peak strengths (sensitivity) of clay deposits in the existing Kitimat Arm marine environment has not been undertaken by Northern Gateway. Johns et al (1986) shows sensitivities of between 1.5 and 3.0 in core samples extending about 5 m below mudline. The high mobility of the two large submarine slides is an indication of significant strength loss and could be interpreted to be a result of strength loss in sensitive clay deposits.
7.2 REGIONAL GROUNDWATER PATTERNS
Regional groundwater flow can contribute to the potential for rapid failure in two ways: i) through the freshwater leaching of marine clays as noted above, and ii) by elevating pore pressures above hydrostatic conditions.
Two distinct regional groundwater patterns have been defined as follows:
1. Groundwater flow parallel to the axis of the fjord flowing within sediments underlying the Kitimat Valley basin to a down-gradient exit through the Kitimat River Delta. While the flows would predominantly follow permeable materials along the valley, it is possible that the flows could induce upward gradients through less permeable materials draped over valley sediments at the north end of Kitimat Arm. Similar mechanisms may occur at some of the major valleys with significant alluvial sediments including Bish and Emsley Creeks. 2. Groundwater flows normal to the axis of the fjord may also occur within the valley walls on both sides. Groundwater would be expected to flow from the upslope portions of the fjord walls to a terrestrial or submarine down-gradient exit at or through the lower fjord sidewalls. This component of groundwater flow would be from recharge areas on the upper slopes of the fjord to discharge areas on the lower fjord walls. Flows could potentially be through the bedrock or along permeable deposits such as sand layers. Where deposits of low permeability materials blanket the lower fjord slopes, there m a y be artesian pressures at or above sea level.
Artesian pressures have been noted in the literature (Morrison, 1984) in the area of the Alcan smelter, Moon Bay, Bish Cove, and Emsley Cove. These areas of artesian pressures appear to be consistent with the presence of significant thicknesses and extents of continuous low permeability materials. Areas of significant sediment blankets typically exist near the head of Kitimat Arm; however continuous blankets are rarely encountered along the much steeper fjord slopes to the south as evidenced by the frequent rock outcrops along both the terrestrial and submarine slopes.
7.3 LIQUEFACTION
Liquefaction is a soil behaviour phenomenon in which a saturated soil loses a substantial amount of strength due to high excess pore-water pressure generated by, and accumulated during, strong earthquake ground shaking. The potential for excess pore-water pressure generation and strength loss associated with volume change tendency is highly dependent on AMEC Project No. EG0926008 Page 16
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011 the density of the soil, with greater potential in looser soils. Permanent ground displacements due to lateral spreads or flow slides and differential settlement are commonly considered to be significant potential hazards associated with liquefaction. Typically, soils considered to be particularly susceptible to liquefaction include loose, relatively clean sand and silty sand deposits, as well as uniformly-graded, non-plastic silt, although loose gravels also are susceptible to liquefaction. Silty and clayey soils having plasticity tend to be less susceptible than cohesionless soils to liquefaction-type behaviours; some sensitive clays, however, have exhibited liquefaction-type strength losses.
To provide an understanding of the potential for liquefaction and the likely mechanisms of subsequent ground motions during seismic events, Northern Gateway commissioned AMEC Geomatrix to conduct a review and analysis of the available data. The AMEC Geomatrix review was supported primarily by the work in this report, published subject matter literature, and the results of the probabilistic seismic hazard assessment carried out for Northern Gateway (Atkinson, 2009). This section of the report summarizes the results of the AMEC Geomatrix work. A copy of the report is provided in Appendix E.
This report concludes that liquefaction of granular sediments on the delta fronts is not a significant hazard with respect to wave generation due to the limited mobility unless major geometric changes such as construction of large fills or undercutting by erosion or dredging of the toe areas result in significant steepening.
The liquefaction potential study suggests that based on geologic and depositional considerations alone, the deltaic deposits of upper Kitimat Arm may be considered to have high to very high susceptibility to earthquake-induced liquefaction. This characterization appears to be supported by the typically loose to medium dense sand, silty sand, gravelly sand and sandy gravel, and occasional silt conditions indicated on the limited number of available logs of borings drilled in and adjacent to the northern reaches of the Kitimat River delta and reported in the references. With the application of the probabilistic seismic hazard assessment results with respect to the peak horizontal ground accelerations, the current work established that the combined probability of a seismic event resulting in a liquefaction failure was estimated to be in the range of about 1x10 -4 to 2x10 -4 per annum or a return period of about 5,000 to 10,000 years
Permanent ground displacements associated with liquefaction are direct products of the soil behaviour phenomena (i.e., high pore water pressure and significant strength reduction) produced by the liquefaction process. Displacements that do not extend significant distances generally include lateral spreads and differential settlement. These displacements are contrasted by flow slides in liquefied materials that are typically associated with steeper slopes and may involve ground movements of hundreds of meters. The potential for flow slides is the focus of liquefaction related slope failures in the context of evaluating the hazard of landslide- generated waves.
As noted in the AMEC Geomatrix Report, “[b]ased on considerations of ground failures and permanent ground displacement mechanisms such as described, as well as stability and ground deformation analyses of other submarine slopes involving liquefiable soils, we are of the opinion that, if these surficial sandy deltaic deposits were to experience earthquake-triggered liquefaction and associated strength loss, they still would exhibit a non-zero undrained residual shear strength (e.g., Seed and Harder, 1990; Olson and Stark, 2002) that would inhibit downslope movements associated with slope instability. As such, given the relatively gentle slopes (i.e., 10° to 14° or flatter, except the 20° upper near-shore slopes) and the sediments’
AMEC Project No. EG0926008 Page 17
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011 undrained residual shear strength, we are of the opinion that the sediments would experience a limited displacement behaviour, rather than flow slide or turbidity current characteristics.”
Note that for static liquefaction events, the mobility would be similar to the mobility of a seismically triggered event.
8.0 TRIGGERING MECHANISMS ANALYSIS FOR THE 1974 AND 1975 EVENTS
8.1 GENERAL
In addition to defining the factors that may contribute to the potential occurrence of a tsunamigenic slide at a particular location, an analysis of the potential mechanisms for triggering a large scale submarine landslide was also undertaken to evaluate the hazard and the potential recurrence period. In general, landslides are triggered by one or a combination of the following: geometric changes in the slope, strength changes in materials over time, increased pore pressures, or external loading.
8.2 TIDES
Many researchers have observed that submarine landslides and associated landslide-generated waves commonly occur with extreme low tides. Extreme low tides create the equivalent to an increase in pore pressures and/or a decrease in supporting pressure. Since extreme low tides in the lowest one percentile occur regularly at least once per year and low tides in the range of the lowest 1 percent to 5 percent are known to have occurred just prior to the October 1974 and April 1975 events, a significant number of slides should be visible in the submarine geomorphic record if tides are a dominant and direct causal factor in submarine landsliding in the area. As was discussed in Section 6.0, there is little evidence to support a conclusion that significant or widespread submarine landsliding occurs in fine grained sediments in the area.
8.3 GEOMETRIC CHANGES
Significant geometric change was noted by Morrison (1984) with respect to a November 1969 bathymetric survey that indicated a significant build-up of sediments near the crest of the delta near the industrial developments resulting in an overall slope angle up to 27 degrees. This was in contrast to a survey of the same area in 1953 which showed a consistent slope of around 14 degrees, as well as the approximately 14 degree slope measured in the same area in 1977 following the April 1975 failure. The source of rapid sedimentation is attributed to discharge from the Kitimat River by Morrison (1984), although a steep build-up of sediments is not apparent in the more recent 2009 surveys, a period about twice as long as between the 1953 to 1969 survey. It is suggested that the steep build-up of sediments shown on the 1969 survey may be the result of deposition from development activities in the adjacent harbour and foreshore areas.
Scour of the submarine slopes during periods of high river flows with associated high sediment loads and resultant density currents has the potential to produce significant erosion and rapid geometric changes at the river deltas around the basin. Since high flows occur frequently each year, but not specifically on the reported dates of the two failures reviewed in 1974 and 1975, a dominant and direct causal factor relationship for the average daily flow of the Kitimat River on single days is not supported. As can be seen in the polygon atlas and/or in the overall
AMEC Project No. EG0926008 Page 18
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011 submarine contour mapping, deep scour paths are visible only at the east side of the Kitimat River delta and not on the other smaller fan deltas in the area.
The highest flows in the records reviewed for the Kitimat River did not occur during or prior to the reported events. Since the geomorphic evidence suggests that erosion at the outlet of the river did occur, sustained multi-day high flows were investigated as a potential cause of rapid erosion and undercutting of the sediments that formed the 1974 failure.
8.4 ANTHROPOGENIC CHANGES
Anthropogenic changes appear to be a direct causal factor in the case of the 27 April 1975 failure. Anthropogenic changes do not appear to be a direct cause of the 17 October 1974 event; however, upstream dyking is thought to have been a factor contributing to the 1974 event. The anthropogenic factors for the two slides are summarized below:
1. Construction activities were undertaken in the period leading up to the 1975 event and the failure appears to have been initiated in the area of construction filling. The failure occurred at extreme low tide. River discharge was low at the time, and no significant channels discharge into the Moon Bay area so scour was not a factor. 2. High river discharge is thought to have caused the fjord sidewall scour visible in the bathymetric data adjacent to the backscarp of the 1974 slide combined with low tides that occurred in the period prior to the 1974 slide. Anthropogenic changes to the delta area relating to channelization would have contributed to reductions in overbank flood storage and corresponding higher than natural flows at the discharge point potentially increasing the amount of scour relative to what had occurred prior to the dyking.
The 1974 and 1975 events are revisited in the following sections to examine the combination of events that are interpreted to have led to the failures.
8.5 EVENTS RELATED TO THE 17 OCTOBER 1974 SLIDE
It is thought that the 1974 slide was triggered by down-cutting erosion along the east side of the delta adjacent to, and perhaps coincident with, the steep fjord sidewall deposits. A summary plan of the 1974 slide area is included as Figure 9.
The fjord sidewalls are partially covered by variable depths of cohesive materials that are expected to be approximately normally consolidated, while the delta deposits are granular. The evidence to support the hypothesis of downcutting includes:
• The presence of a submarine trough that separates the east side of the delta deposits from the fjord sidewalls. This is shown in Figure 9. • The locations of scarps and the blocky nature of landslide debris in the northeast/central portion of Kitimat Arm that indicate the potential origin of the 1974 slide and cohesive nature of the failed materials. • Sub-bottom surveys that indicate the presence of cohesive soils east of the delta and the assumption (interpreted based on lack of signal penetration) of the presence of granular deposits in the delta as noted in Appendix B. The cohesive soils are in the location where they would have been undercut by the scour. • The environmental conditions at the time of the event as noted above (high flows carrying sediment, low tides) leading to a high potential for scour.
AMEC Project No. EG0926008 Page 19
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
A review of river flow magnitiude and duration and tidal stage was carried out to characterize the conditions that preceded the 1974 slide. The review included first grouping the top 1 percent of the 7-day average flow records into distinct events (note that several records separated by 1 or 2 days were considered to be the same event). The maxima of the individual events were then scaled from 0 to 1. For each of the event maxima, the lowest daily tidal stage was tabulated and scaled from 0 to 1 within the full low tide range (3.17 m). The product of the scaled event maxima and associated re-scaled lowest daily tide was computed and represents a measure of the conditions resulting from high river flows and low tides acting in combination. The peak daily 7-day average flows and tidal stage are shown below on Plot 1 of Figure 10. The products of the scaled values are shown on Plot 2 of Figure 10.
Note that the environmental conditions represented by the scaled factor of 0.28 for 1974 have been repeated at higher magnitudes a few times since 1974, specifically 0.58 on 6 November 1978; 0.40 on 15 October 1991; and 0.58 on 29 September 1992. The relatively high magnitude occurrences, including the 17 October 1974 event, are all in, or close to, the month of October, which reflects the high flows at this time of the year.
The occurrence of higher magnitude events beyond the triggering threshold defined by the 1974 slide is significant since the higher threshold events did not trigger further large-scale sliding. The sub-bottom surveys conducted in 2010 show remnant clay deposits on the order of single- metre thicknesses only in the northeast corner of Kitimat Arm, and thus the lack of sliding in this case is attributed to insufficient remaining fine-grained material to form a slide due to depletion of the source zone.
In addition to the analysis of the events post-1974, a review of the reports from flooding and construction of flood protection along the Kitimat River just upstream of the confluence are of particular relevance. The records, as noted in Section 4.2.1 above, show that a significant flood event occurred in Kitimat in 1966 and that flood protection measures were installed to reduce the impacts of flooding in the community of Kitimat. River training and industrial occupation of the delta through the use of dykes and fill would have resulted in a reduction of overbank flood storage in the delta area. The effects, while difficult to quantify, would have led to the delivery of higher flood flows further downstream in the system. It is believed that this effect was a contributing factor to the 1974 failure since without anthropogenic activities in the delta, similar combined low tide and high-intensity long-duration river flow events would have occurred regularly throughout geological history. It is possible that the slide terrain is overprinted with multiple events that culminated in the final depletion of cohesive sediments during the event in 1974.
8.6 RE-VISITING THE EVENTS RELATED TO THE 27 APRIL 1975 SLIDE
The 1975 failure and associated landslide-generated waves are attributed to anthropogenic activities associated with the constriction activities in Moon Bay at the time of the event. A summary plan of the 1975 slide area is included as Figure 11.
It is thought that the 1975 slide was a failure triggered by increased pore water pressures in foundation soils and increased loading from shoreline and marine earthworks activities on cohesive fjord sidewall sediments. A low tide an hour before the slide likely contributed additional excess porewater pressures as a result of a rapid draw-down. The precipitation for April 1975 totalled 53 mm, with rainfall occurring on 13 days from April 1 to 27; daily totals ranged from a minimum of 0.3 mm to a maximum of 10.7 mm. The maximum daily precipitation
AMEC Project No. EG0926008 Page 20
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011 is well below a typical storm level event for the region that can often reach up to 100 mm daily, although it did occur on the day prior to the slide.
For the purposes of comparison, the 1975 event is plotted on Figure 10; no peak flow events are associated with the slide event.
Environmental factors were not reviewed beyond those presented above (relatively low river flow) as the northwest corner of Kitimat Arm does not have a direct tributary stream. From the literature reviews it is known that no seismic activity occurred in the timeframe leading up to or during the 1975 slide.
It is judged that the geometric changes, increased loading and associated elevated porewater pressures as a result of dredging and fill placement near Moon Bay coupled with the low tide on April 27, 1975 reduced the stability conditions to a point that failure occurred in cohesive marine sediments. Without construction, a low tide just above the lowest 5th percentile would likely not have triggered failure as tides of this magnitude occurred frequently throughout the year and at the same relative frequency throughout geological history.
9.0 PRESENTATION OF NUMERICAL MODELING OF THE 1975 EVENT
Numerical modelling of the 27 April 1975 event was carried out to evaluate the characteristics of a wave field from the 1975 event at the proposed terminal site. The numerical model required several key inputs including the surface profile prior to the failure, definition of the slide surface, estimation of geotechnical properties of the soil, estimation of the failure velocity and overall basin geometry to suitable limits and depths to reduce model boundary effects. The model provided estimates of wave heights at known observation points in order to confirm results and calibrate the model.
Several modelling tools were used throughout the work in two main categories, namely geotechnical models for the landslide and a hydrodynamic model to examine the generation and propagation of the associated waves.
The geotechnical tools included finite element modelling to establish the potential depth of failure associated with the initial slide using LSDYNA7, a large strain finite element program that was run in 2d. DAN-W8 software was utilized to examine failure velocities and calibrate the viscosity used in the viscous flow model portion of the coupled landslide-hydrodynamic model. The coupled landslide-hydrodynamic program LIQUID9 was used to simulate the wave generation. The wave modelling is presented in a report by Fine and Bornhold (2011) in Appendix F.
The results of the modelling show general agreement with the limited observations recorded from the 27 April 1975 event and show that the significant waves reported in the accounts of the event at Kitimaat Village and at Moon Bay would also have propagated throughout the Kitimat Arm area. Maximum wave heights (measured trough to crest) at the proposed terminal location associated with the 1975 event would have been on the order of 1.2 m to 3.6 m, and the wave period would have been on the order of about 1 minute.
7 LSDYNA, Livermore Software Technology Corp. 8 O. Hungr Geotechnical Research, Inc. 9 Proprietary program, see references included in Fine and Bornhold (2011) AMEC Project No. EG0926008 Page 21
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
10.0 DISCUSSION AND SUMMARY OF FINDINGS
The findings of this study have confirmed that Kitimat Arm is exposed to the potential for submarine landsliding and associated landslide generated waves. The potential source zones for the waves are judged to be in locations where fjord sidewall slope deposits are of significant thickness, where the sidewall materials have the capability to move quickly after failure, and where triggering mechanisms of sufficient magnitude are present. The past large-scale rapid failures occurred in the head of Kitimat Arm where the combination of these conditions existed at the time of failure, For the purposes of evaluating the overall hazard, this study identified several areas throughout Kitimat Arm with significant thicknesses of fjord sidewall sediment accumulation, however, in many of the areas, such as adjacent to the terminal in Polygon 13, a natural triggering mechanism is absent. It is either the absence of material in areas with defined natural triggers, or the absence of natural triggers in areas with appropriate materials that lead to the opinion that the potential for landslide generated waves in Kitimat Arm is very low. Anthropogenic activities appear to represent the most significant potential triggering mechanism. Additional comments with respect to hazard mitigation are provided below in Section 11.0.
The following provides a summary of the findings of this study:
1. Sub-bottom profiling at the proposed Kitimat Terminal berth site shows very steep bedrock dominated slopes with no significant thickness of overlying sediment and therefore there is no potential for a tsunamigenic failure in marine muds at the berth site. 2. The proposed berth structures will consist of piled foundations set into the steep rocky slopes. The construction will not require construction within any significant amount of unconsolidated marine sediments. 3. Based on the available records, two very large (greater than 1 Mm3) rapidly moving landslides occurred in Kitimat Arm in 1974 and 1975 at the sidewalls of the fjord immediately adjacent to the fjord-head delta slope. 4. The 1974 event appears to have been triggered by undercutting erosion related to large flood flows in the Kitimat River, although low tide and flood protection measures likely contributed to the failure. The failure was in fjord-wall clay deposits and resulted in a long run-out landslide that ran several kilometres down the submarine basin. 5. The 1975 event appears to have been triggered by construction activities in Moon Bay, although low tide was a contributing factor. The initial failure was in fjord-wall glaciomarine clay deposits. The initial slide from the fjord side wall undercut the delta slope and entrainment of additional slide volume occurred from an oversteepened portion of the fjord-head Kitimat River delta slope. The 1975 event was a rapid failure and resulted in a significant wave field throughout Kitimat Arm including at the proposed marine terminal location. 6. Sub-bottom profiling data shows that the 1974 submarine landslide resulted in significant removal of glaciomarine clays from the fjord walls and very little material remains. On the basis that submarine landslides require significant volumes to generate waves, the potential for another event at this location is very low. Further investigation of the pattern of density current flow and potential erosion through and below the trough identified in the area should be carried out. 7. Submarine morphology at the locations of other marine deltas in Kitimat Arm do not show significant submarine erosion patterns consistent with observations of deep scour at the Kitimat River Delta leading to the opinion that this mechanism of triggering a slide is limited to the large flows of the Kitimat River.
AMEC Project No. EG0926008 Page 22
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
8. Morrison (1984) indicates the presence of artesian pressures at depth may occur in the region. This could suggest that leaching of marine clays could result in increased sensitivity over time in these materials, although widespread occurrence of sliding is not visible, thus suggesting that leaching is not a primary causal factor. If leaching and increased sensitivity is present, it appears that a significant triggering mechanism is required. However, based on the substantial rock exposures at many locations along Kitimat Arm, it may be that continuous confining glaciomarine clay layers only occur near tributary streams and that submarine leaching is not widespread. Additional work is recommended with respect to establishing the condition of the salinity in the glaciomarine clays and porewater pressures at depth in future phases of the work. 9. The geomorphic record along the sidewalls of Kitimat arm represents at least several thousand years of record that would indicate major landslide occurrences within this period. As discussed, no other large slides other than the 1974 and 1975 events are evident. It is therefore concluded that the return period for naturally occurring events is very long, considering that the source zone for the 1974 slide has been depleted. 10. Liquefaction of deltaic sediment in response to seismic events is predicted to result in liquefaction related failures with a return period of about 5,000 to 10,000 years. In the absence of substantial geometric changes, liquefaction related failure is unlikely to develop into a long-run out flow slide on deltaic slopes angles typically measured at the head of Kitimat Arm. 11. Sub-bottom profiling throughout Kitimat Arm identified areas with significant thicknesses of cohesive sediments that could theoretically result in large scale landsliding, although widespread sliding has not occurred as discussed above. Anthropogenic changes such as cuts and fills are the most likely influence on the potential for failure at these locations where significant thicknesses of sediment exist. 12. Overall, the most significant potential triggering mechanism for future large scale submarine landslides in Kitimat Arm appears to be anthropogenic activities, primarily uncontrolled and unmitigated construction.
11.0 QUALITATIVE HAZARD ANALYSIS SUMMARY
There is a very low likelihood of occurrence of a submarine landslide-generated wave in Kitimat Arm over the lifetime of the project. This qualitative assessment relies on the assumption that any and all regional developments in the foreshore and offshore environment will consider the potential for submarine slope instability both directly at a proposed development site and around the basin. Further development within the Kitimat Arm needs to be planned cooperatively among existing facility owners and potential project sponsors and be executed with an appropriate level of due diligence, meeting appropriate engineering standards and mitigation techniques.
12.0 PATH FORWARD
It is recommended that the following additional geotechnical studies be conducted during detailed engineering to confirm the findings of this report. :
1. Drilling investigations in selected polygons to examine the nature of the clay deposits visible in the sub-bottom geophysical surveys and to quantify the range of values of strength, sensitivity, porewater and pressure conditions and the potential for time- dependent strength change;
AMEC Project No. EG0926008 Page 23
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
2. Drilling investigations at selected polygons to examine the conditions at depth in the Kitimat River Delta and the sidewall fan deltas; 3. Additional modelling, as necessary, based on the findings of the geotechnical investigations; 4. Consequence analysis based on wave train characteristics derived from updated hydrodynamic models, as required.
AMEC Project No. EG0926008 Page 24
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
13.0 CLOSURE
Recommendations and evaluations presented herein are based on preliminary data and are considered preliminary. The site specific investigations carried out, as presented in this report, have been of a preliminary nature only. It is expected that further investigations will be undertaken for the areas discussed in this report during detailed engineering for design and construction.
This report has been prepared for the exclusive use of Northern Gateway Pipelines Inc., for specific application to the area within this report. Any use which a third party makes of this report, or reliance on or decisions made based on it, are the responsibility of such third parties. AMEC Environment & Infrastructure accepts no responsibility for damages, if any, suffered by any third party as a result of decisions made or actions based on this report. It has been prepared in accordance with generally accepted geotechnical engineering practices. No other warranty, expressed or implied, is made.
Respectfully submitted,
AMEC Environment & Infrastructure, A division of AMEC Americas Limited Reviewed by:
Shane Kelly, M.Eng., P.Eng. (AB&BC) P. Barlow, M.Sc., P.Eng. (AB&BC) Associate Geotechnical Engineer Principal Engineer
Original paper copies signed and sealed by D.S. Cavers, M.Eng., P.Eng., P.Geo.
D.S. Cavers, M.Eng., P.Eng.(AB&BC), P.Geo (BC) Principal Engineer
AMEC Project No. EG0926008 Page 25
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
14.0 REFERENCE LIST
Bornhold, B.D. 1983. Sedimentation in Douglas Channel and Kitimat Arm. Proceedings of a Workshop on the Kitimat Marine Environment, Canadian Technical Report of Hydrography and Ocean Sciences No. 18, Department of Fisheries and Oceans.
Bornhold, B.D., J.R. Harper, D. McLaren and R.E. Thomson. 2007. Destruction of the First Nations Village of Kwalate by a Rock Avalanche-generated Tsunami. Atmosphere- Ocean Vol. 45 No. 2.
Bornhold, B.D., 2011. Submarine Failures and Associated Tsunamis, Norway, Literature Review. Report submitted to AMEC by Coastal and Ocean Resources Inc.
Campbell, D.B. and N.A. Skermer. 1975. Report to B.C. Water Resources Service on Investigation of Seawave at Kitimat, B.C. Golder Associates.
Clague, J. 1984. Quaternary Geology and Geomorphology, Smither-Terrace-Prince Rupert Area, British Columbia. Geological Survey of Canada, Memoir 413.
Clague, J.J. 2001. Tsunamis in A Synthesis of Geological Hazards in Canada. Geological Survey of Canada, Bulletin 548, p. 27-42.
Dan, G., N. Sultan and B. Savoye. 2007. The 1979 Nice Harbour Catastrophe Revisited: Trigger Mechanism Inferred from Geotechnical Measurements and Numerical Modelling. Marine Geology, Vol. 245, p. 40-64. Geertsema, M., Cruden, D.M., and Schwab, J.W., 2006, A large rapid landslide in sensitive glaciomarine sediments at Mink Creek, northwestern British Columbia, Canada: Engineering Geology, v. 83, p. 36-63. Geertsema, M., and Torrance, J.K. 2005. Quick clay from the Mink Creek landslide near Terrace, British Columbia: Geotechnical properties, mineralogy, and geochemistry. Canadaian Geotechnical Journal, v. 42, p. 907-918. Fine, I., and Bornhold, B.D. 2011. Submarine Landlside-Tsunami Modelling, April 27, 1975. Report submitted to AMEC by Coastal and Ocean Resources Inc.
Hardy, R.M. and C.F. Ripley. 1954. Foundation Investigation for the Kitimat Smelter and Design and Construction of the Kitimat Smelter. The Engineering Journal, p. 1460-1479.
Morrison, K.I. 1984. Case History of Very Large Submarine Landslide, Kitimat, British Columbia. Unknown source, author was Principal Consultant – Geotechnical for Klohn Leonoff Ltd.
Jiang, L. and P.H. LeBlond. 1992. The Coupling of a Submarine Slide and the Surface Waves Which It Generates. Journal of Geophysical Research, Vol. 97, No. C8.
Jiang, L. and P.H. LeBlond. 1994. Three-Dimensional Modeling of Tsunami Generation Due to a Submarine Mudslide. Journal of Physical Oceanography, Vol. 24.
Johns, M.W., D.B. Prior, B.D. Bornhold, J.M. Coleman and W.R. Bryant. 1986. Geotechnical Aspects of a Submarine Slope Failure, Kitimat Fjord, British Columbia. Marine Geotechnology, Vol. 6, No. 3.
AMEC Project No. EG0926008 Page 26
Attachment JRP IR 4.1 Northern Gateway Pipelines Inc. Kitimat Arm Landslide-Generated Wave Hazard Analysis 20 September, 2011
Kulikov, E.A. and A.B. Rabinovich. 1996. The Landslide Tsunami of November 3, 1994, Skagway Harbor, Alaska. Journal of Geophysical Research, Vol. 101, No. C3.
Luternauer, J. and D. Swan. 1978. Kitimat Submarine Slump Deposit(s): A Preliminary Report. Current Research, Part A, GSC, Paper 78-1A. p. 327-332.
Murty, T.S. 1979. Submarine Slide-Generated Water Waves in Kitimat Inlet, British Columbia. Journal of Geophysical Research, Vol. 84, No. C12.
Prior, D.B., B.D. Bornhold, J.M. Coleman and W.R. Bryant. 1982. Morphology of a Submarine Slide, Kitimat Arm, British Columbia. Geology, Vol. 10, p. 588-592.
Prior, D.B., B.D. Bornhold and M.W. Johns. 1984. Depositional Characteristics of a Submarine Debris Flow. Journal of Geology, Vol. 92, p. 707-727.
Pugh, W.L. (Alcan) 1954. The Kitimat Harbour. The Engineering Journal, p. 1450-1459.
Rabinovich, A.B. R.E. Thomson, B.D. Bornhold, I.V. Fine and E.A. Kulikov. 2003. Numerical Modelling of Tsunamis Generated by Hypothetical Landslides in the Strait of Georgia, British Columbia. Pure and Applied Geophysics, Vol. 160.
Skvortsov, A. 2005. Numerical Simulation of Landslide-Generated Tsunamis, Kitimat 1975 Failure Scenario. Masters Thesis, University of Victoria.
Swan, D. 1978. Acoustic Imaging of the Sea Bed in Northern Kitimat Arm, B.C. Bachelor of Science Thesis, The University of British Columbia.
AMEC Project No. EG0926008 Page 27
Attachment JRP IR 4.1
LIST OF FIGURES
Figure 1 General Site Plan Figure 2 Air Photo Review, Northland Dock to Moon Bay Figure 3 Air Photo Review, Northland Dock to Moon Bay Figure 4 Air Photo Review, Kitimat Harbour Between Eurocan and Alcan Figure 5 Air Photo Review, MK Bay Marina to Kitimaat Village Figure 6 Chronological Summary of Water Line Position 1947 to Present Figure 7 Morrison (1984) Plan and Geological Section Figure 8 Timeline of Major Construction Activities and Tsunami Reports Figure 9 1974 Initiation Zone Figure 10 Combined Tidal and River Flow Analysis Figure 11 1975 Initiation Zone
Attachment JRP IR 4.1
Appendix A
Bornhold, B, 2011, Coastal and Ocean Resources Inc., “Submarine Failures and Associated Tsunamis, Nor wa y - Literature Review”, Report to AMEC, 37 pp.
Attachment JRP IR 4.1
Appendix B
Bornhold, B, 2010, Coastal and Ocean Resources Inc., “Initial Results, Kitimat Arm Geophysical Survey”, Report to AMEC, 18 pp.
Bornhold, B, Coastal and Ocean Resources Inc., Letter to AMEC, “Kitimat Harbour Geophysical Investigation Report” 30 August 2011, 1 page.
Attachment JRP IR 4.1
Appendix C
Terra Remote Sensing Inc. Report to AMEC and Northern Gateway Pipelines, 10 November 2010, “Survey Operations Report”, 47 pp.
Terra Remote Sensing Inc., Maps 1 to 5 showing “Contoured Bathymetry and Survey Track; and, Isopach of Total Unconsolidated Sediment and Seabed Slope Magnitude”. 5 pp.
Attachment JRP IR 4.1
Appendix D
Kitimat Arm Polygon Atlas, 67 pp.
Attachment JRP IR 4.1
Appendix E
AMEC internal report, 2011, “Preliminary Assessment of Earthquake-Related Liquefaction and Associated Submarine Slope Instability Hazard”, 20 pp.
Attachment JRP IR 4.1
Appendix F
Fine, I., and Bornhold, B., 2011. “Submarine Landslide-Tsunami Modeling, April 27, 1975”. Report to AMEC. 11 pp.